Skip to main content

Full text of "Hydro-electric power stations"

See other formats 


This  is  a  digital  copy  of  a  book  that  was  preserved  for  generations  on  library  shelves  before  it  was  carefully  scanned  by  Google  as  part  of  a  project 
to  make  the  world's  books  discoverable  online. 

It  has  survived  long  enough  for  the  copyright  to  expire  and  the  book  to  enter  the  public  domain.  A  public  domain  book  is  one  that  was  never  subject 
to  copyright  or  whose  legal  copyright  term  has  expired.  Whether  a  book  is  in  the  public  domain  may  vary  country  to  country.  Public  domain  books 
are  our  gateways  to  the  past,  representing  a  wealth  of  history,  culture  and  knowledge  that's  often  difficult  to  discover. 

Marks,  notations  and  other  marginalia  present  in  the  original  volume  will  appear  in  this  file  -  a  reminder  of  this  book's  long  journey  from  the 
publisher  to  a  library  and  finally  to  you. 

Usage  guidelines 

Google  is  proud  to  partner  with  libraries  to  digitize  public  domain  materials  and  make  them  widely  accessible.  Public  domain  books  belong  to  the 
public  and  we  are  merely  their  custodians.  Nevertheless,  this  work  is  expensive,  so  in  order  to  keep  providing  this  resource,  we  have  taken  steps  to 
prevent  abuse  by  commercial  parties,  including  placing  technical  restrictions  on  automated  querying. 

We  also  ask  that  you: 

+  Make  non-commercial  use  of  the  files  We  designed  Google  Book  Search  for  use  by  individuals,  and  we  request  that  you  use  these  files  for 
personal,  non-commercial  purposes. 

+  Refrain  from  automated  querying  Do  not  send  automated  queries  of  any  sort  to  Google's  system:  If  you  are  conducting  research  on  machine 
translation,  optical  character  recognition  or  other  areas  where  access  to  a  large  amount  of  text  is  helpful,  please  contact  us.  We  encourage  the 
use  of  public  domain  materials  for  these  purposes  and  may  be  able  to  help. 

+  Maintain  attribution  The  Google  "watermark"  you  see  on  each  file  is  essential  for  informing  people  about  this  project  and  helping  them  find 
additional  materials  through  Google  Book  Search.  Please  do  not  remove  it. 

+  Keep  it  legal  Whatever  your  use,  remember  that  you  are  responsible  for  ensuring  that  what  you  are  doing  is  legal.  Do  not  assume  that  just 
because  we  believe  a  book  is  in  the  public  domain  for  users  in  the  United  States,  that  the  work  is  also  in  the  public  domain  for  users  in  other 
countries.  Whether  a  book  is  still  in  copyright  varies  from  country  to  country,  and  we  can't  offer  guidance  on  whether  any  specific  use  of 
any  specific  book  is  allowed.  Please  do  not  assume  that  a  book's  appearance  in  Google  Book  Search  means  it  can  be  used  in  any  manner 
anywhere  in  the  world.  Copyright  infringement  liability  can  be  quite  severe. 

About  Google  Book  Search 

Google's  mission  is  to  organize  the  world's  information  and  to  make  it  universally  accessible  and  useful.  Google  Book  Search  helps  readers 
discover  the  world's  books  while  helping  authors  and  publishers  reach  new  audiences.  You  can  search  through  the  full  text  of  this  book  on  the  web 


at|http  :  //books  .  google  .  com/ 


.mt  11^  ■< 


Digitized  by 


Google 


Digitized  by 


Google 


Digitized  by 


Google 


Digitized  by 


Google 


Digitized  by 


Google 


'5 


s. 

J 

0) 

s 


0) 


o 


I 


H 


Digitized  by 


Google 


HYDROELECTRIC 
POWER  STATIONS 


BY 

ERIC  A.  JjOF 

t  • 
AND 

DAVID  B.  RUSHMORE 


FIHST  EDITION 


NEW  YORK: 
JOHN  WILEY  &  SONS,  Inc. 

London:  CHAPMAN  A  HALL,  Limitbd 
1917 


Digitized  by 


Google 


Copyright,  1917 

BY 

ERIC  A.  LOF    AND     DAVID  B.  RUSHMORE 


P<?CSB  Of 

BRAUNWORTH    &    CO. 

Bi»OK    MANUFACTUREH8 

BROOKLYN,   N.  Y. 


Digitized  by 


Google 


PREFACE 


Increased  activity  in  the  development  of  our  water-power 
resources  is  certain  to  take  place  in  the  near  future,  because 
of  the  rapid  and  general  increase  in  power  demands  on  the  central 
station  systems,  and  because  of  the  increased  cost  and  shortage  of 
fuel.  A  book,  therefore,  dealing  with  the  many  phases  of  this 
subject  from  a  practical  and  up-to-date  engineering  standpoint, 
will  be  of  great  benefit,  not  only  to  those  who  have  been  actively 
engaged  in  such  development  work,  but  also  to  those  who  may 
desire  to  enter  it  in  the  future. 

The  work  of  planning,  building,  operating  a  hydro-electric 
power  development  requires  a  full  understanding  of  the  economic 
factors  which  enter  into  the  problem,  and  a  thorough  knowledge 
of  both  the  hydraulic  and  the  electrical  engineering  sides  of  the 
subject.  Any  book  to  be  complete,  must,  of  necessity,  cover  all 
these  branches.  Limited  space,  however,  makes  it  impossible  to 
deal  with  minor  details,  and  the  book  is  not  intended  as  a  treatise 
on  the  design  of  individual  structures,  machinery  and  apparatus 
which  go  into  the  makeup  of  a  power  station.  Many  books  have 
been  written  dealing  with  such  detailed  desig^ns,  and  manufac-- 
turers  should  be  freely  consulted.  This  book  deals  with  and 
explains  the  problems  which  must  be  solved  in  connection  with 
the  construction  and  management  of  a  hydro-electric  power' 
station,  so  that  the  manager  or  engineer  may  select  power  equip- 
ment and  fully  understand  the  economic  factors  which  enter  into 
each  individual  situation.  The  authors  have  endeavored  to 
describe  the  most  recent  engineering  practice  and  they  have 
included  a  considerable  amount  of  information  not  available 
hitherto. 

This  is  an  educational  treatise  for  the  student  and  operating 
man  and  the  manuscript  has  been  submitted  to  experts  in  the 
various  branches  of  the  subjects  treated. 

It  is  believed  that  a  study  of  this  volume  will  result  in  improving 
the  service  and  operating  eflBciency  of  many  systems. 


Digitized  by 


Google 


vi  PREFACE 

The  authors  also  wish  to  take  this  opportunity  to  express 
their  appreciation  and  thanks  to  those  who  have  so  kindly  and 
wiUingly  assisted  in  the  preparation  of  this  work  with  their 
suggestions  and  advice.  Among  these  may  especially  be  men- 
tioned, Mr.  Lewis  F.  Moody  of  the  I.  P.  Morris  Company;  Mr. 
Chester  W.  Larner  of  the  Wellman-Seaver-Morgan  Company; 
Mr.  W.  A.  Doble  of  the  Pelton  Water  Wheel  Company;  Mr.  A. 
V.  Garratt  of  the  Lombard  Governor  Company;  Mr.  J.  H.  Man- 
ning of  the  Stone  &  Webster  Company;  and  Mr.  A.  S.  Crane  of 
the  J.  G.  White  Company. 

Eric  A.  Lof 

David  B.  Rushmore 
Schenectady, 
October,  1917. 


Digitized  by  VjOOQ IC 


CONTENTS 


PAOB 

I.  General  Introduction 1 

History  of  Water  Power  and  Electrical  Developments. 
Conservation  of  Natural  Fuel  Resources.  Available  and 
Developed  Water  Power  in  United  States.  Power  from 
Inland  Waterways.  Primary  Power  and  Its  Uses.  CJonv- 
mercial  Possibilities  for  Hydro-Electric  Power. 

II.  Htdboloqt 39 

1.  Properties  of  Water 39 

Weight.  Volume.  Critical  Temperatures.  Latent  Heat. 
Specific  Heat.  Effect  of  Atmospheric  Pressure.  Measure- 
ments. 

2.  Rainfall 44 

Source  of  Water  Supply.  Variation  in  Rainfall.  Rainfall 
Records. 

3.  Disposal  of  Rainfall 47 

Evaporation.    Absorption.    Run-off. 

4.  Stream-flow 63 

Definition  of  Terms.  Variation  in  Stream-flow.  Factors 
Affecting  Stream-flow.  Measurements  of  Stream-flow. 
Government  Records. 

6.  Energy  of  Flowing  Water .  ." 66  > 

Potential  Energy.  Kinetic  ,Energy.  Head.  Velocity. 
Quantity.    Horse-power. 

6.  Convenient  Equivalents 68 

Second-feet  per  square  mile  vs.  run-off  in  inches.  Second- 
feet  vs.  run-off  in  acre-feet.    Miner's  inch,  etc. 

III.  CXiASSmCATIGN  OP  DEVELOPlfENTS 70 

Low-head  Developments.  Medium  and  High-head  De- 
velopments. 

IV.  Dams  and  Headworks 74 

1.  Dams 74 

Classification.  Location.  Timber  Crib  Dams.  Earth-fiU 
Dams.  Rock-fill  Dams.  Masonry  Dams — Gravity — But- 
tressed— ^Arched.    Rules  Governing  Design. 

2.  Flashboards 91    / 

Stationary  Boards.    Sliding  Gates.    Tilting  Gates.    Tain- 
ter  Gates.    Rolling  Gates, 
vii 


Digitized  by 


Google 


viii  CONTENTS 

CHAPTER  PAGE 

3.  Fishways 99 

4.  Intakes 100 

Trash  Racks.  Low-head  Installations.  High-head  Instal- 
lations.    Influence  of  Ice. 

V.  Water  Conductors  and  Accessories 1C4 

1.  Water  Conductors iv/4 

Classification.  Canals.  Flumes.  Tunnels.  Pipe  Lines — 
Head — Loss  of  Head — Hydraulic  Gradient — Size  of  Pipe 
Line — Steel  Pipe — Wooden  Stave  Pipe — Concrete  Pipe. 

2.  Water  Hammer  and  Surge  Tanks 138 

Water  Hammer.     Surge  Tanks — Simple — Differential. 

3.  Gates  and  Valves 144 

Requirements.      Sluice    Gates.    Tainter    Gates.    Gate 
Valves.    Operation  and  Control.     Pivot  Valve.     Johnson's 
Hydraulic  Valve.    Air  Valves. 

VI.  Storage  Reservoirs 169 

Storage  and  Pondage.  Limitation  to  Storage.  Location  of 
Reservoir.    Intakes.    Seepage  and  Evaporation. 

VII.  Power-house  Design 165 

1.  Building 165 

General  Design.  Basement43.  Foundations.  Floors. 
Walls.  Roof.  Windows.  Doors.  Traveling  Crane.  Venti- 
lation.   Illumination.    Heating.    Miscellaneous. 

2.  Arrangement  of  Apparatus 175 

General  Considerations.  Turbines.  Governors.  Gen- 
erators. Exciters.  Transformers.  Current  Limiting  Reac- 
tors. Switchboards.  Oil  Circuit  Breakers.  Lighting  Ar- 
resters.    Outdoor  Apparatus. 

3.  Transportation  and  Erection 193 

Transportation.  Unloading.  Apparatus  Storage.  Sched- 
ule of  Erection.  Crane  Service.  Protective  Features.  Co- 
operation. 

4.  Starting  Up 198 

General  Precautions.     Drying-out.     Insulation  Resistaiice. 

VIII.  Hydraulic  Equipment 202 

1.  Turbines 202 

Reaction  Turbines.  Impulse  Turbines.  Selection  of  Tur- 
binet:.  Specific  Speed.  Actual  Speed.  Characteristic 
Curves.  Speed  Regulation.  Overspeed.  Mechanical  De- 
signs. Reaction  Type — Horizontal — Vertical — Runners — 
Gate  Mechanism — Speed  Rings — Casings — Draft  Tubes — 
Bearings.  Impulse  Type — Horizontal  and  Vertical — Runners 
— Arrangement  of  Runners — Nozzles — Housings. 


Digitized  by 


Google 


CONTENTS  ix 

CBAPTKB  PAOB 

2.  Governors 246 

Factors  affecting  Speed  Regulation.  Action  of  Governor. 
Pipe-line  Pressure  caused  by  Governor  Action.  Energy 
Output.  Arrangement  and  Operation.  Methods  of  Con- 
trol.   Typical  Designs. 

3.  Pressure  Regulators  or  Relief  Valves 258 

Governor-operated.    Pressure-operated. 

4.  Water-flow  Meters 262 

Venturi  Meter — ^Register — Manometer. 

5.  Water  Stage  Registers 266 

Printing — ^Recording. 
IX.  Electrical  EgtriPMENT '. 270 

1.  General  Considerations 270 

Voltage.     Frequency. 

2.  Ssmchronous  Generators 280 

General  Description.  Induced  E.M.F.  Effect  of  Power 
Factor  on  Operation.  Field  Excitation.  Regulation.  Short- 
circuit  Current.  Armature  Connections.  Wave  Shape. 
Grounding  of  Generator  Neutral.  Rating.  Efficiency. 
Speed.  Voltage.  Parallel  Operation.  Mechanical  Design. 
Lubrication.     Ventilation.     Brakes. 

3.  Induction  Generators 348 

Output  and  Excitation.  Comparative  Capacity  of  In- 
duction and  Synchronous  Generators.  Operation.  Places 
of  Utilization.     General  Construction. 

4.  Exciters 350 

Separate  Excitation.  Capacity  and  Rating.  Voltage. 
Characteristics.  Shunt  vs.  Compound  Wound.  Speed. 
Method  of  Drive.  Mechanical  Design.  Arrangement  and 
Connections.    Exciter  Batteries. 

5.  Voltage  Regulation 363 

Hand  Regulation.  T.A.  Regulator — Method  of  Opera- 
tion— Cycle  of  Operation — Regulator  Arrangements.  Line- 
drop  Compensation.  KR-System  of  Regulation.  High 
Voltage — High  Current  Relays.  Synchronous  Condenser 
Regulation. 

6.  Transformers 373 

Fundamental  Principles.  Induced  E.M.F.  Ratio.  Mag- 
netizing Current.  Reactance.  Regulation.  Core  and  Shell 
Type.  Method  of  Cooling.  Single  and  Polyphase.  Rating. 
Efficiency.  Voltage.  Taps.  Number  and  Size  of  Units. 
Connections.  Parallel  Operation.  Mechanical  Design.  Oil. 
Drying  Transformers.  Drying  Oil.  Testing  Oil.  Operation. 
Oil  Supply  System.    Cooling  Water  System. 

7.  Current  Limiting  Reactors 458 

Purpose  of  Reactors.  Rating.  Rating  as  Affected  by  Fre- 
quency, Voltage  and  Current.     Effect  of  Reactance  on  Power 


Digitized  by 


Google 


X  CONTENTS 

CBAPTSB  PAOa 

Factor  and  Regulation.  Lones.  Inductance.  Location. 
Number  of  Reactors.  Size  of  Reactor  and  Determination 
of  Three-phase  or  Balanced  Short-circuit  Currents.  Single- 
phase  Short-circuit  Currents.    Mechanical  Design. 

8.  Switching  Equipment 485 

System  of  Connection  and  Relay  Protection.  Oil  Circuit 
Breakers.  Relays.  Switchboards.  Instrument  Equipment. 
Current  and  Potential  Transformers.  Exciter  and  Field 
Control.  Voltmeter  and  Synchronizing  Receptacles.  Am- 
meter Transfer  Switches.  Throw-over  Switches.  Calibrat- 
ing Terminals.  Control  Switches.  Mimic  Buses.  Bus  and 
Switch  Structures.  Disconnecting  Switches.  Signal  Systems. 
Multi-recorder.   Oil  Circuit  Breaker  Batteries. 

9.  Over-voltage    Protection 503 

Classification    of    Over-voltages.     Lightning    Arresters. 
Arcing  Ground  Suppressor.    Short  Circuit  Suppressor.    Pro- 
tection of  Telephone  Lines. 
10.  Station  Wiring 625 

Insulation.  Open  Wiring.  Cables  in  Ducts  or  Conduits. 
Single  vs.  Multiple  Conductors.  General  Practice.  Size  of 
Cables.  Corona  limit  of  Voltage.  Economical  Considera- 
tions.   Voltage  Drop.    Resistance  and  Reactance  Tables. 

X.  Economical  Aspects 644    / 

Preliminary  Considerations.  Guide  for  Preparing  Water 
Power  Reports.  Amount  of  Energy  Available.  Power  De- 
mand. Load  and  Diversity  Factor.  Primary  and  Second- 
ary Power.  Water  Storage.  Auxiliary  Stations.  Intercon- 
nection of  Systems.  Investigating  an  Enterprise.  Cost  of 
Hydro-Electric  Power  Plants.    Cost  of  Power. 

XI.  Organization  and  Operation 746 

Management.      Operating   Force.      Operating    Records. 
Operating  and  Maintenance  Instructions. 

Appendix: 

I.  References  to  Descriptions  of  Plan)» 757 

II.  Principal  Data  on  Transmission  Systems  Operating  at  70,000 

volts  and  above 783 

■  III.  Turbine  Testing  Code 788 

Index 803 


Digitized  by 


Google 


HYDEO-ELECTKIC  POWEfi  STATIONS 


CHAPTER  I 

GENERAL  INTRODUCTION 

HISTORY  OF  WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS 

The  use  of  water  power  for  industrial  purposes  dates  back  to 
very  ancient  times.  The  crude  current  wheels  were  familiar 
to  the  Chinese  on  the  Yellow  River  and  the  Hamites  on  the  Nile 
and  Euphrates  fully  three  thousand  years  ago.  These  wheels 
operated  entirely  by  the  kinetic  energy  of  the  moving  water,  and 
the  power  thus  obtained  was  utilized  for  raising  the  water  of  the 
rivers  for  irrigating  the  arid  land  and  also  for  grinding  of  corn 
and  other  simple  applications.  Similar  current  wheels,  although 
necessarily  of  improved  design,  have  been  most  widely  utiUzed 
and,  while  very  inefficient,  they  are  still  used  for  minor  irrigation 
and  other  purposes  in  many  countries. 

The  first  radical  change  in  the  art  was  the  use  of  channels, 
by  which  the  water  could  be  conducted  and  directly  applied  to 
undershot  wheels.  This  improvement  resulted  in  the  utilization 
of  some  30  per  cent  of  the  theoretical  water  power,  and  the  system 
maintained  its  prominence  imtil  almost  the  middle  of  the  eigh- 
teenth century,  when  the  overshot  wheel  was  invented  by  John 
Smeaton,  who  showed  that  if  the  bucket  wheel  was  changed  into 
an  overshot  form,  its  useful  efficiency  would  be  increased  to  over 
6Q  per  cent.  In  this  type  of  wheel  the  energy  of  the  water  was 
applied  directly  through  its  weight  by  the  action  of  gravity  and 
3delded  a  very  high  efficiency.  Overshot  wheels  were  formerly 
built  of  great  size.  One  at  Laxey,  Isle  of  Man,  constructed  about 
1865  and  is  said  to  be  still  in  operation,  is  72  feet  6  inches  in 
diameter  and  develops  150  horse-power.    A  number  of  overshot 


Digitized  by 


Google 


2  GENERAL  INTRODUCTION 

wheels  are  also  in  use  at  old  mills  in  the  Catskill  Mountains 
in  New  York  State. 

The  breast  wheel,  which  followed  the  overshot  wheel,  was  de- 
veloped in  England  during  the  latter  part  of  the  eighteenth 
century  and  was  used  for  a  long  number  of  years.  It  consisted 
of  a  circular  drum,  having  on  its  periphery  a  series  of  buckets, 
the  sheathing  of  the  drum  forming  their  bottom.  They  were 
operated  partly  by  gravity  and  partly  by  kinetic  energy,  and 
the  water  was  appUed  through  a  fliune  and  controlled  by  gates. 
Below  these  was  located  the  "  breast "  which  consisted  of  a  con- 
cave cylindrical  surface  of  planking  concentric  with  the  wheel. 
The  clearance  was  very  small,  thus  preventing  the  water 'from 
spilling  out  of  the  buckets  until  it  had  reached  the  lower 
level.  This  type  of  wheel  gave  an  eflSciency  of  about  70  per 
cent. 

The  wheel  types  described  above  have,  nowever,  now  been 
almost  entirely  superseded  by  the  turbine,  and  are  therefore 
so  nearly  obsolete  that  they  may  be  considered  as  of  historical 
interest  only.  While  the  fundamental  principles  of  the  turbine 
may  be  distinguished  in  wheels  used  in  the  sixteenth  century, 
the  principal  developments  were  made  during  the  last  century. 
In  the  turbine  the  water  acts  mainly  by  impulse  or  reaction  or 
both,  and  the  velocity  has  a  definite  relation  to  the  head. 

In  1823  M.  Fourneyron  began  his  experiments  on  the  radial 
outward-flow  turbine,  the  first  of  which  was  installed  at  Pont 
Sur  rOgnon  in  France  in  1827.  Its  principle  consisted  in  an  out- 
ward discharge  from  a  pipe  to  a  wheel  with  curved  buckets  placed 
outside  of  the  apertures  of  discharge.  The  buckets,  revolving 
from  the  action  of  the  water,  finally  discharged  it  at  the  circum- 
ference with  its  force  exhausted.  The  tube  which  supplied  the 
water  was  closed  at  the  bottom  by  a  concave  cone  surrounding  the 
wheel  shaft,  which  passed  up  through  it  in  a  pipe,  so  as  not  to  be 
exposed  to  the  water.  This  cone  was  surrounded  by  a  number  of 
guide  plates,  which  directed  the  water  to  the  buckets  in  the  proper 
tangential  direction. 

The  axial  discharge  turbine  was  first  built  by  Henschel  &  Son  in 
Germany  in  1837.  There  has  always  been  doubt  as  to  whether 
this  turbine  should  be  attributed  to  Jonval  or  to  Henschel.  Jon- 
val  thoroughly  described  the  basic  idea  in  a  patent  dated  1841  and 
it  is  quite  possible  that  he  was  working  on  the  proposition  as  early 


Digitized  by 


Google 


WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS    3 

as  Henschel.    It  proved  to  be  far  superior  to  the  outward  dis- 
charge type  in  that  it  ahnost  entirely  eliminated  the  latter. 

The  inward-flow  wheel,  in  which  the  action  of  the  Foumeyron 
turbine  is  reversed,  was  patented  by  S.  B.  Howd,  of  Geneva,  N.  Y., 
in  1836,  and  seems  to  have  been  the  origin  of  the  American  type  of 
turbine.  Very  great  improvements  were,  however,  made  in  the 
construction  by  James  B.  Francis  about  1847,  and  many  regard 
him  as  the  originator.  The  Francis  turbine  of  to-day  has  dis- 
placed all  other  types  of  reaction  turbines,  and  with  its  rapid 
development,  radical  departures  have  been  made  from  the  strictly 
radial  inward-flow,  so  that  the  Francis  turbine  of  to-day  is  of  a 
combined  radial  or  diagonal  inward  discharge  type. 

The  impulse  wheels  were  among  the  earliest  forms  used.  Thus 
the  rovet  volant  or  flutter  wheels  were  used  for  centuries  in 
India,  Egypt,  Syria  and  Southern  France.  They  consisted  of  flat, 
vertical  vanes  projecting  radially  from  a  vertical  wooden  shaft, 
the  water  jet  from  the  feeding  spout  striking  the  vanes  tan- 
gentially  near  their  ends.  It  was  not,  however,  until  1853  that 
this  type  of  wheel  was  given  a  scientific  consideration  in  this 
country  by  Jearum  Atkins,  while  its  practical  development  must  be 
credited  to  Lester  A.  Pelton,  who,  in  1882,  and  following  years, 
made  radical  improvements  in  its  design.  This  type  of  wheel  is 
now  extensively  used  in  the  West,  where  the  high  heads  made  such 
a  wheel  necessary.  '^ 

The  first  great  water  power  developments  were  made  in  the 
New  England  States.  The  textile  industry  was  destined  to 
expand  rapidly  and  the  water  power  of  the  streams  was  its  sup- 
porting ally.  Under  this  influence  the  first  great  water  power 
was  developed  on  the  Merrimac  River,  in  1822,  where  subse- 
quently the  City  of  Lowell  became  a  great  cotton  manufacturing 
center.  Near  Lowell  there  were  soon  developed  the  equally 
prominent  water  powers  on  the  Merrimac  River  at  Manchester, 
in  New  Hampshire,  and  Lawrence,  Mass.  These  developments 
had  each  capacities  of  10,000  to  12,000  horse-power,* and  each 
was  chiefly  devoted  to  the  manufacture  of  cotton  goods,  as  were 
the  water  powers  of  Cohoes  (1828)  in  New  York,  and  Lewiston 
(1849)  in  Maine.  The  Connecticut  River  water  power  at  Hol- 
yoke  (1848)  was  largely  devoted  to  the  manufacture  of  paper,  as, 
later,  were  the  Fox  River  powers  in  Wisconsin.  The  water  pow- 
ers on  the  Genesee  River  at  Rochester,  N.  Y.  (1856),  and  on  the 


Digitized  by 


Google 


4  GENERAL  INTRODUCTION 

Mississippi  River  at  Minneapolis  (1857),  were  largely  devoted  to 
the  manufacture  of  flour. 

In  1861  the  development  of  the  mighty  power  of  Niagara  Falls 
was  begun,  a  canal  being  built  through  the  town  to  a  power- 
house at  the  edge  of  the  gorge  below  the  falls.  The  Niagara  Falls 
Hydraulic  Power  &  Manufacturing  Company  was  formed  in 
1872,  and  during  the  first  years  its  operation  consisted  in  fur- 
nishing water  to  numerous  water  wheels  of  different  manufactur- 
ing enterprises.  The  inefficiency  of  this  method,  however,  soon 
became  apparent,  and  a  central  power-house  was  built  in  1881 » 
the  energy  being  transmitted  to  the  factories  along  the  edge  of  the 
cliff  by  means  of  ropes,  belts  and  shafts. 

Different  opinions  exist  as  to  the  time  at  which  the  first  trans- 
mission of  electricity  took  place.  Its  possibility  was  pointed 
out  as  early  as  1850  and  possibly  earlier,  and  it  is  claimed  that  in 
1858  electricity  was,  for  the  first  time,  utilized  for  driving  a  com- 
mercial machine.  This  was  in  the  artillery  works  of  St.  Thomas 
d'Aquin,  France,  where  a  dividing  machine  was  driven  by  an 
electric  motor,  which  derived  its  current  from  an  adjacent  bat- 
tery. Though  the  electric  motor  existed  long  before  the  dynamo, 
it  attained  no  prominence  until  after  the  practical  demonstration 
of  the  latter.  As  long  as  the  galvanic  battery  constituted  the 
source  of  power,  the  application  was  naturally  restricted.  Another 
reason  was  the  defective  construction  of  the  earlier  motors,  their 
counter  E.M.F.  being  comparatively  weak,  and  hence  the  work 
which  could  be  obtained  from  them  was  small  in  comparison  with 
the  power  expended  and  their  size. 

While  the  principle  of  the  reversibility  of  the  electric  motor 
seems  to  have  been  known  as  early  as  1850,  it  was  the  practical 
experiments  carried  out  by  Gramme  at  the  Vienna  Exposition  in 
1873  that  clearly  demonstrated  the  practical  importance  of  this 
property.  Gramme  is,  therefore,  generally  given  the  credit  as 
being  the  one  who  first  practically  demonstrated  the  possibility 
of  employing  the  electric  current  for  transmitting  energy  from  one 
place  to  another.  His  experiments  at  this  Exposition  consisted 
in  transmitting  current  from  a  machine  working  as  a  generator 
to  a  second  machine  about  550  yards  distant,  working  as  a 
motor  driving  a  pump. 

In  1878  a  motor  was  installed  at  the  sugar  works  at  Sermaize, 
France.     It  was  used  to  operate  a  hoist  and  derived  its  current 


Digitized  by 


Google 


WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS        5 

from  a  steam-driven  Gramme  generator.  The  application  of 
water  power  for  driving  dynamos  followed  shortly,  and  the  same 
year  a  water-wheel-driven  generator  was  installed  at  the  Shaw 
Chemical  Works,  Eng.,  and  power  supplied  to  a  motor  150  yards 
distant  for  driving  miscellaneous  tools.  In  1882  the  first  com- 
mercial central  stations  for  lighting  began  operation  in  London 
and  New  York,  and  the  same  year  marked  the  building  of  the  first 
hydro-electric  central  station  in  the  United  States  at  Appleton, 
Wis.,  Rgs.  1  and  2. 

In  the  above  systems,  and  several  others,  the  electric  current 
was  transmitted  for  very  short  distances  only  but  in  1882  Marcel 
Deprez  bxiilt  the  first  long-distance  transmission  line  from  Mies- 
bach  to  Munich,  a  distance  of  37  miles.  It  was  built  purely  for 
experimental  and  demonstration  piurposes,  2400-volt  direct- 
current  being  used.  The  results  proved  to  be  very  encouraging 
and  financial  support  was  obtained  for  a  larger  project.  Thus, 
in  1884,  Deprez  began  preparations  for  the  Criel-Paris  trans- 
mission, which  was  completed  in  1886.  In  this  20  amperes  direct- 
ciurent  was  transmitted  the  25-mile  distance  at  a  potential  of 
7500  volts,  the  transmission  efficiency  obtained  being  about  32 
per  cent. 

The  first  A.C.  transmission  system  was  the  one  at  Cerchi, 
Italy,  made  in  1886  and  known  as  the  "  Cerchi  Tivoli-Rome 
Plant."  The  equipment  of  this  station  consisted  of  two  150  H.P. 
steam-driven,  single-phase  Ganz  generators  designed  to  operate  at 
112  volts.  Transformers  having  a  ratio  of  1  :  18  were  used  to 
step  from  this  voltage  up  to  2000,  at  which  voltage  energy  was 
transmitted  to  Rome,  a  distance  of  17  miles.  In  1889  the  capac- 
ity of  this  steam  plant  was  increased  to  2700  H.P. 

In  1887  Tesla,  Ferraris  and  Bradley  pointed  out  the  advan- 
tages of  the  three-phase  over  the  single-phase  system,  but  it 
was  not  until  1891  that  the  first  commercial  three-phase  trans- 
mission line  was  put  into  operation.  This  was  the  112-mile 
Laufifen-Frankfort  Une  supplying  a  Ughting  load  to  the  City  of 
Frankfort.  The  power-house  installation  consisted  of  one  225 
Kw.  three-phase  generator,  direct  connected  to  a  water  wheel 
operating  under  a  head  of  10  feet.    The  line  voltage  was  12,000. 

In  the  United  States  the  first  A.C.  hydro-electric  installation 
was  the  one  at  Oregon  City  by  the  Willamette  Falls  Electric 
Conipany,  now  owned  by  the  Portland  Railway,  Light  and  Power 


Digitized  by 


Google 


6 


GENERAL  INTRODUCTION 


Fio.  1. — Exterior  View  of  First  Hydro-Electric  Central  Station  in  United 
States  at  Appleton,  Wisconsin.    Installed  in  1882.    Capacity  250  lights. 


|ii^^ 


Fig.  2. — Interior  View  of  First  Hydro-Electric  Central  Station  in  United 
States  at  Appleton,  Wisconsin. 


Digitized  by 


Google 


WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS    7 

Company.  This  installation  took  place  in  1889  and  consisted  of 
two  300-H,P.  Victor  wheels  belted  to  4000-volt  single-phase  gen- 
erators, the  power  being  transmitted  to  Portland,  13  miles  distant. 
In  1890,  shortly  after  the  Willamette  Falls  Electric  Company 
had  completed  their  installation,  the  Telluride  Power  Company 
installed  at  Ames,  Col.,  two  150-Kw.  single-phase  generators 
directly  connected  to  Pelton  water  wheels  operating  under  a  head 


Fig.  3. — Power  House,  Mississippi  River  Power  Company,  Keokuk,  Iowa. 

of  500  feet.     Power  was  transmitted  to  Telluride^  a  distance  of 
5  miles,  at  3000  volts. 

In  1892  another  single-phase  transmission  plant  was  installed 
in  CaUfomia  and  deUvered  power  to  Pomona,  approximately  13 
miles  distant,  and  about  29  miles  to  San  Bernardino.  The  voltage 
at  the  beginning  of  operation  was  5000,  which  was  higher  than 
any  previously  used  commercially,  but  on  February  16,  1893, 
this  was  raised  to  10,000  volts,  and  on  May  2,  1893,  by  connecting 
their  transmission  Unes  all  in  series,  120  kw.  was  carried  42  miles 
with  a  transmission  efficiency  of  60  per  cent,  at  that  time  a  great 


Digitized  by 


Google 


8 


GENERAL  INTRODUCTION 


achievement  and  an  indication  of  the  possibilities  of  electric  trans- 
mission of  power. 

To  Southern  Califomia  also  belongs  the  distinction  of  having 
the  first  commercial  pol3^hase  transmission  system  installed  and 
operated  in  the  United  States.    In  1893  a  generating  station  was 


— 

— 

— 

— 

— 

— 

^ 

— 

•~ 

n 

■" 

ISO^OOO 

140,000 

180,000 

10,000 

110,000  - 

8 

100,000 1 

90,000^ 

80,000  3 

70,000  3 

__ 

80,000  1 

a 

Mazi 

40,000 

90,000 

SQlQOO 

10,000 

Yeat> 
Fig.  4. — Commercial  Transmission  Voltages. 

built  by  the  Redlands  Electric  Light  &  Power  Company  (now  the 
Southern  Califomia  Edison  Company),  at  the  mouth  of  Mill 
Creek  Canyon.  The  plant  consisted  originally  of  two  250-Kw., 
2400-volt,  three-phase,  Y-connected  generators,  running  at 
600  R.P.M.,  and  driven  by  Pelton  water  wheels  under  a  head  of 
295  feet.    The  power  was  transmitted  for  a  distance  of  7§  miles 


Digitized  by  VjOOQ IC 


WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS        9 

to  Redlands  and  there  used  for  lighting  and  industrial  motor 
applications. 

Before  the  end  of  the  same  year  another  polyphase  plant  was 
installed  at  Hartford,  Conn.,  where  400  Kw.  was  transmitted 
11  miles  at  5000  volts.  This  plant  replaced  a  single-phase  in- 
stallation which  had  been  delivering  power  for  lighting  over  the 
same  line  since  1891. 

With  these  plants  began  the  era  of  hydro-electric  power  trana- 
mission  in  the  country,  and  statistics  show  that  nearly  three 
hundred  plants  were  in  actual  operation  about  1896.  It  would  be 
almost  impossible  to  tabulate  all  the  thousands  of  plants  and 
systems  now  in  operation.  Single  plants  with  capacities  of  one-, 
quarter  million  horse-power  have  been  built,  Fig.  3,  and  power 
is  now  being  commercially  transmitted  for  distances  of  nearly 
250  miles  at  potentials  of  150,000  volts,  Fig.  4.  As  yet  the  limit 
is  not  in  sight,  but  one  thing  is  certain,  that  the  introduction  of 
the  electric  system  and  the  evolution  in  the  design  of  apparatus 
have  made  possible  the  concentration  of  such  enormous  amounts 
of  power  which  are  now  generated  in  modem  stations  and  its 
transmission  for  long  distances  to  centers  where  an  economical 
market  can  be  found. 

HISTORICAL  REVIEW  OF  WATER-WHEEL  DEVELOPMENT 

7 

1740  Barker's  Mill,  the  simplest  t3rpe  of  tangential  outflow  turbines,  was 
invented.    It  had  radial  arms  and  operated  purely  by  reaction. 

1823  M.  Foumeyron  began  his  experiments  with  the  radial  outward-flow 
turbine. 

1826  A  radial  inward-flow  turbine  waa  proposed  by  Ponoelet. 

1827  The  first  Fourneyron  turbine  was  erected  at  Pont  Sur  TOgnon,  France. 

1836  Samuel  B.  Howd  of  Geneva,  N.  Y,,  obtained  a  patent  on  an  inward- 

flow  turbine. 

1837  Foumeyron  erected  a  turbine  at  St.  Blaise,  Switzerland,  which  oper- 

ated under  a  head  of  354  feet. 
1837    O.  Henschel,  of  Cassel,  Germany,  invented  the  downward  axial-dis- 
charge turbine,  later  known  by  the  name  of  Jonval  or  Koechlin. 

1841  The  first  axial-discharge  wheel  was  introduced  into  practice  by  the 

French  engineer,  Jonval. 

1842  James  Whitelaw,  of  Paisley,  developed  an  improved  type  of  Barker's 

Mill  which  was  erected  on  Chard  Canal.    This  wheel  had  spiral 
tapering  arms  so  curved  that  the  water  flowed  radially  when  the  wheel 
was  running  at  proper  speed. 
1844    A  Foumeyron  turbine,  constmcted  by  Uriah  A.  Boyden,  was  erected 
at  Appleton  Company's  cotton  mills  in  Lowell,  Mass. 


Digitized  by 


Google 


10  GENERAL  INTRODUCTION 

1847  James  B.  Francis  made  radical  improvements  in  the  inward-flow  tur- 
bine. 

1850  About  this  time  the  Jonval  turbine  was  introduced  in  America  by 
Elwood  and  Emile  Geyelin,  of  Philadelphia. 

1853  Jearum  Atkins  was  the  first  scientifically  to  consider  the  impulse  wheel 
in  this  country. 

1859    The  "  American  "  or  mixed-flow  turbine  was  designed. 

1882.    Lester  A.  Pelton  made  radical  improvements  in  this  type  of  wheel. 

HISTORICAL  REVIEW  OF  THE  PROGRESS  OF  ELECTRIC  POWER 

TRANSMISSION 

1820  A.  M.  Ampere  announced  his  discovery  of  the  dynamical  action  between 

conductors  conveying  electric  currents;  currents  flowing  in  the  same 
directions  attracting  and  in  opposite  directions  repelling. 

1821  Michael  Faraday  discovered  the  electro-magnetic  rotation  in  causing 

a  wire  conveying  a  voltaic  current  to  rotate  continuously  around  the 
pole  of  a  permanent  magnet. 

1831  Faraday  discovered  the  principles  of  electro-magnetic  induction  and 

laid  the  foundation  for  all  subsequent  inventions  which  finally  led 
to  the  production  of  electro-magnetic  or  d3mamo-electric  machines. 

1832  H.  Pixii  built  a  magneto-electric  machine  consisting  of  a  fixed  horse- 

shoe armature,  wound  over  with  insulated  copper  wire,  in  front  of 
which  revolved  about  a  vertical  axis  a  horse-shoe  magnet. 
1832    H.  Pixii  invented  the  spfit-tube  commutator  for  converting  the  alter- 
nating current  into  continuous  current. 

1840  Henry  Pinkus  proposed  and  patented  the  principle  of  transmitting 

electric  energy  through  wires  to  an  electric  motor  on  a  railway 
car. 

1841  Prof.  Francois  Nolle t,  Brussels,  proposed  the  electrical  utilization  of 

water  and  wind  power  for  driving  dynamos. 

1850  Jacobi  claimed  that  an  electro-magnetic  machine  could  also  be  worked 

as  a  magneto-electric  machine  and  vice  versa. 

1851  Dr.  Sinsteden  suggested  the  use  of  currents  produced  by  magneto- 

electric  machines  for  driving  electric  motors. 
1855    Bessolo,  Italy,  suggested  and  patented  a  scheme  for  the  electrical 
utilization  of  natural  forces,  and  long-distance  transmission  of  elec- 
trical energy  for  power  purposes. 

1857  E.  W.  Siemens  invented  the  drum-wound  armature  and  improvement 

in  the  shape  of  field  magnets. 

1858  Beams  of  intense  electric  light  obtained  from  the  volataic  arc  by 

Faraday. 

1858  Eugene  Regnault  worked  a  Froment  electrical  motor  by  the  current 
from  a  Clarke  magneto-electric  machine,  driven  itself  by  a  mechan- 
ical motor. 

1858  A  dividing  machine  was  driven  by  an  electric  motor  at  the  Artillery 
Works  of  St.  Thomas  d'Aquin,  France.  Current  was  obtained  from 
a  battery. 


Digitized  by 


Google 


WATER  POWER  AND  ELECTRICAL  DEVELOPMENTS      11 

1864  Cazel  obtains  a  French  patent  on  an  electric  railway  system  in  which 
one  or  more  magneto-electric  machines  are  to  be  driven  by  hydraulic 
or  wind  motors,  and  the  current  generated  conveyed  to  a  rotary 
car  motor  by  wires  and  the  track  rails. 

1866  Dynamos  according  to  Siemens'  principle  began  to  be  built  commerci- 
ally and  were  employed  for  producing  electric  light. 

1866  Felice  Marco,  Italy,  was  granted  an  Italian  patent  for  the  electrical 

utilization  of  water  power. 

1867  Prof.  Pfaundler,  of  Innsbruck,  experimented  with  a  Kravogl  electric 

motor  exhibited  at  the  Paris  Exposition,  and  found  %at  it  could 

also  be  used  for  generating  electric  currents. 
1870    The  Gramme  ring  dynamo  was  invented. 
1870    Jacobi  works  an  electric  motor  by  means  of  a  secondary  battery. 
1873    Gramme  and  Fountaine  discovered  the  reversible  action  of  the  dynamo 

and  made  the  first  public  demonstration  of  power  transmission  at 

the  Vienna  Exposition.     Current  was  transmitted  from  a    machine  . 

working  as  a  generator  to  a  second  machine  550  yards  distant, 

working  as  a  motor  and  driving  a  pump. 

1875  Alcide  Girin  was  granted  a  French  patent  for  the  combination  of  elec- 

tro-magnetic inductive  apparatmi  and  a  certain  number  of  induction 
coils  in  order  to  obtain  in  the  secondary  circuits  a  lower  tension  and 
a  higher  intensity  than  in  the  primary  circiUts. 

1876  Jablochkofif's  arc  lamp  invented. 

1876    WaUaoe-Farmer  dynamo  at  the  Philadelphia  Centennial  Exposition. 

1878  A  motor  was  installed  in  the  sugar  works  at  Sermaize,  France,  for  oper- 

ating a  hoist.  Current  was  obtained  from  a  steam-driven  Gramme 
generator. 

1879  First  commercial  arc  lamp  system  (Brush)  installed  in  Cleveland. 
1879    Edison  incandescent   lamp  invented   and  first   complete  system  of 

incandescent  lighting  installed  at  Menlo  Park. 
1879    Siemens  and  Halske  install  the  first  electric  railway  in  which  current 
was  generated  by  dynamos.     It  was  at  the  Berlin  Exposition  that 
a  line  of  550  yards  was  laid  down  upon  which  a  small  locomotive 
drew  passenger  cars  merely  as  a  novelty. 

1881  Carpentier  and  Deprez  were  granted  a  patent  for  a  system  of  trans- 

porting electricity  to  a  distance  and  transforming  it. 

1882  Gaulard  and  Gibbs  suggested  the  transformer  for  practical  operation. 
1882    Marcel  Deprez  built  the  first  long-distance  experimental  line  from 

Miesbach  to  the  Exposition  in  Munich,  a  distance  of  about  37 
miles.  He  transmitted  one-half  horse-power  direct  current  at  a 
pressure  of  2400  volts. 

1882  First  hydro-electric  central  station  installed  at  Appleton,  Wis.  Ca- 
pacity 250  lights. 

1882  First  'commercial  central  station  for  incandescent  lighting  began  oper- 
ation in  London. 

1882  Pearl  Street  Station  of  Edison  Electric  Illuminating  Company  began 

operation  in  New  York. 

1883  "  Feeder  and  Main  "  system  first  used. 


Digitized  by 


Google 


12  GENERAL  INTRODUCTION 

1883  "  Thiee-wire  "  system  first  used. 

1884  Dr.  J.  Hopkinson  clearly  established  the  fact  that  similar  alternators 

could  be  run  as  generators  and  motors.    First  practically  demon- 
strated in  1889  by  Mordey. 

1884  American  Institute  of  Electrical  Engineers  was  organized. 

1885  First  transformer  built  in  this  country  by  Wm.  Stanley  at  Great 

Barrington,  Mass. 
1885-88    Nicola  Tesla  and  Galileo  Ferraris  invented  independently  the  poly- 
phase induction  motor  and  pointed  out  the  advantages  of  the  three- 
ph|pe  system. 

1885  National  Electric  Light  Association  was  organized  in  Chicago. 

1886  First  regular  133-cycle,  single-phase  lighting  plant  was   installed  in 

Buffalo. 

1886  Criel-Paris  transmission  was  completed.  Twenty  amperes  direct- 
current  was  transmitted  for  a  distance  of  25  miles  at  a  potential  of 
7600  volts. 

1886  Sprague  installed  the  first  electric  street  railway  in  this  country  at 
Richmond,  Va. 

1886  The  first  A.C.  transmission  system  was  installed  at  Cerchi,  Italy, 

150  H.P  being  transmitted  for  17  miles  at  2000  volts  single  phase. 

1887  Tesla,  Ferraris  and  Bradley  pointed  out  the  advantages  of  the  three- 

phase  system. 

1888  Rotating  field  principle  of  alternating-current  generators  was  invented. 

1889  The  first  A.C    hydro-electric  installation  in  the  United  States  was 

installed  by  the  Willamette  Falls  Electric  Co.,  300  H.P.  being  trans- 
mitted for  13  miles  at  4000  volts  single  phase. 
1891    Lauffen-Frankfort   Transmission.     110   H.P.   was   transmitted   from 
Lauffen  to  the  Exposition  at  Frankfort,  a  distance  of  112  miles  at 
12,000  volta,  three  phase. 

1891  Sixty  cycles  introduced  in  United  States. 

1892  The  first  long-distance  transmission  in  United  States  at  San  Antonio, 

Cal.    800  H.P.  was  transmitted  28  miles  at  10,000  volts  single  phase. 

1893  Twenty-five  cycles  introduced. 

1893  The  first  three-phase  hydro-electric  plant  in  United  States  was  in- 
stalled at  Redlands,  Cal. 

1895  The  first  5000-H.P.  generators  were  installed  at  the  Niagara  Falls 

Power  Company. 

1896  25,000-volt  S3rstem  of  the  Pioneer  Electric  Power  Company,  Utah. 
1903    60,000-volt  system  of  the  Guanajuato  Power  and  Electric  Co.,  Mexico. 
1908    110,000-volt  system  of  Au  Sable  Electric  Company,  Grand  Rapids, 

Mich. 
1913    150,000-volt  system  of  the  Pacific  Light  and  Power  Co.,  Los  Angeles, 
Cal, 


Digitized  by 


Google 


CONSERVATION  OF  NATURAL  FUEL  RESOURCES 


13 


WATER  POWERS  OF  THE  WORLD 

The  following  Table  is  based  on  the  area  of  the  different 
continents  and  on  the  assumption  that  the  water  power  per 
square  mile  is  approximately  14  H.P.  This  value  has  been 
found  to  be  the  average  of  a  number  of  investigations  in  Euro- 
pean countries.  For  Australia,  however,  this  value  is  entirely 
too  high,  and  3  H.P.  per  square  mile  has  been  assumed. 


TABLE  I 
Watbb  Powbbs  of  the  World 


Continent. 


Area  in 
Square  Miles. 


Horse-power. 


Africa 

America,  North . 
America,  South . 

Ada 

Australia. . . 

Europe 


11,613,679 
8,037,714 
6,g61,306 

17,057,666 
3,466,290 
3,764,282 


Total. 


161,190,116 

112,627,996 

96,918,284 

238,807,324 

10,368,870 

62,669,948 

671,372,638 


It  is  thus  seen  that  the  total  water  powers  of  the  world  rep- 
resent about  700  miUion  horse-power.  This  vast  amount  can, 
however,  not  be  economically  developed  at  the  present  time,  but 
the  tabulation  merely  shows  the  possibilities  that  may,  in  the 
future,  be  derived  from  this  natural  source. 

CONSERVATION  OF  NATURAL  FUEL  RESOURCES 

One  of  the  most  important  questions  of  the  present  time  is  the* 
one  relating  to  the  conservation  of  our  natural  fuel  resources. 
While  in  1880  the  yearly  coal  consumption  in  this  country  was, 
only  approximately  70  million  tons,  in  1913  it  amounted  to  about 
575  million  tons,  Fig.  5.  The  output  of  our  oil  fields  has,  during 
the  same  time,  also  increased  at  the  same  astonishing  rate,  while 
the  growth  of  our  population  during  this  period  was  only  about 
85  per  cent,  or  about  one-seventh  the  rate  at  which  the  fuel 
consumption  increased.  It  is  easily  realized  what  a  tremendous 
drain  this  consumption  has  been  on  our  natural  fuel  resources, 


Digitized  by 


Google 


14 


GENERAL  INTRODUCTION 


TABLE  II 
Land  and  Water,  and  Population  op  the  States  op  the  United  States 


Square  Miles. 


State  or  Territory. 


Alabama 

Arizona 

Arkansas 

California 

Colorado 

Connecticut 

Delaware 

District  of  Columbia. .  .  . 

Florida 

Georgia 

Idaho 

lUinois 

Indiana 

Iowa 

Kansas 

Kentucky 

Louisiana 

Maine 

Maryland 

Massachusetts 

Michigan 

Minnesota 

Mississippi 

Missouri 

Montana 

Nebraska 

Nevada 

New  Hampshire 

New  Jersey 

New  Mexico 

New  York 

North  Carolina 

North  Dakota 

Ohio 

Oklahoma 

Oregon 

Pennsylvania 

Rhode  Island 

South  Carolina 

South  Dakota 

Tennessee 

Texas 

Utah 

Vermont 

Virginia 

Washington 

West  Virginia 

Wisconsin 

Wyoming 

Totals  and  averages 


Gross  Area. 


52,250 
113,020 

53.850 
158,360 
103.925 

4,990 

2.050 

70 

58.680 

69.475 

84.800 
56.650 
36,350 
56,025 
82.080 

40,400 
48.720 
33.040 
12.210 
8.315 

58.915 
83.365 
46.810 
69.415 
146.080 

77.510 

110.700 

9,305 

7,815 

122,580 

49,170 
52.250 
70.795 
41.060 
70,430 

96.030 
45.215 
1.250 
30,570 
77.650 

42.050 

265,780 

84,970 

9,565 

42.450 

69,180 
24,780 
56.040 
97.890 


3.025.880 


Water. 


710 
100 
805 
2.380 
280 

145 

90 

10 

4.440 

495 

510 
650 
440 
550 
380 

400 
3,300 
3.145 
2.350 

275 

1.485 

4.160 

470 

680 

770 

670 
960 
300 
290 
120 

1.550 

3.670 

600 

300 

600 

1.470 
230 
197 
400 
800 

300 
3,490 
2.780 

430 
2.325 

2.300 
135 

1,590 
315 


54,842 


Land. 


61.540 
112.920 

53.045 
155.980 
103.645 

4.845 

1.960 

60 

54.240 

58.980 

84.290 
56.000 
35.910 
55.475 
81,700 

40.000 

45,420 

29,895 

9,860 

9.040 

67.430 
79.205 
46.340 
68.735 
145.310 

76.840 

109.740 

9.005 

7,525 

122.460 

47.620 
48.580 
70.195 
40,760 
69.830 

94,560 
44,985 
1.053 
30,170 
76,850 

41,750 

262,290 

82,190 

9,135 

40.125 

66.880 
24.645 
54.450 
97.575 


2.971.0.38 


Population 
1915. 


2.301,277 

247.299 

1,713.102 

2,848.275 

935.799 

1.223.583 
211.598 
358.679 
870.802 

2.816.289 

411.996 
6,069.519 
2.798.142 
2.221.038 
1.807.221 

2.365.185 
1,801.306 
767,638 
1.351.941 
3.662.339 

3.015.442 
2.246.761 
1,926.778 
3.391.789 
446.054 

1.258.624 
102.730 
440.584 

2.881.840 
396.917 

10,086.568 

2.371.095 

713.083 

5.088.627 

2.114.307 

809.490 
8.383.992 

602.765 
1.607.745 

680.046 

2.271.379 

4.343.710 

424.300 

362,452 

2.171,014 

1,471.048 

1,359,474 

2.473,533 

174.148 


100.399.318 


Digitized  by 


Google 


CONSERVATION  OF  NATURAL  FUEL  RESOURCES    15 


Oil  ProducUo 
»         S         ^        §         (^ 

n  in  Barrels  (42  (ral.) 

g    H    I     g     g    §     i 

__„n^ .    L  . . 

^   -ixxxii^Eiii:::: 

igTB-pX -J   -"t^   I 

fri"I"i —   "          ^^± 

^     —  i^    :^  ^,      IE 

«"*-l""i —  -          ^-^":: 

±_:::x  :-i-  ::t:  -uX 

in-          I          \                                           :        i 

±____: L-cri -J 

i8j*  -  jr- jr-^  ~              ^-T  " 

J i          ^u                    _^  -^ 

*<«*  —  h-_fH—     ■  ■  ■    -■    ^  ■  ^ 

jsTO--\-5:--  -                -^ — p 

IT ~T"99Z~       5 

.i>«      ^ki   -        ■     :         '-        *■ 

X    ■+""■     1  '  _ ^B  :„     5  '  - 

isn^-^^--                        ^-^ 

-I+'-ilte::     5x:: 

-,„  "f"  -^  '     ^±'" 

ia»       j-i--^ 

.  T    ^..^iii-i::     riT 

1HA1     "   r  V  ' 

-XxX      '  d^s          n     X 

^**^::±f^  !_,..  ,,  ,    

"     "'      it!      =1              r      1 

,1^  --    ~7Cl^   :     •     '     '        ^ 

_L                                 1                 i                   ■                   '     ^      '                    ,            "^-^                      i      - 

.«*_L    2^"'    V'  ■-    ■'    

'  "t       \    ^       '^^  \  ^^^     \^ 

**'^""    "     ^T  .  "^  . 

.««.     i^"_uX-i  ■  -L  ■  " 

.   ....  ..^^  ^  ^ ^^^ :2z 

^^""d'"* — ^^  .  J"  "  ..^  ^ 

■  ■     ■  -  i:  -3,^9-^ 

i«K       J"         '   JL    ■■-+-—    r 

•■**'^":  5    o-T:  X  :x^-  ^. 

_  L  J       _+      I^a^      ^^t^ .. 

^  "^  Jiirx       ^s,      -^t^'t,  .  , 

isftB 113"   "t —    i"^^ 

XX±      X         ^^^    J      -hS      '    - 

-l"!-^            :            ^l^-     *         r^ry    :\_ 

!**♦  T _L_t-T-J-                I 

I   TX       T     If  -K-     ^ol'T. 

t.ic.1  ^      J.                      ^                           ■        - 

■xXXiuj^t       *-. 

itKA  —  r:  ^.       u 

_| ,       ^      _J_         ...  ^  a^        '                   n  J     ■     '     r 

X_u            ^6-3  -t-        -'J-u^ 

ISBU               ^1            \          \ 

jlX  ;       -        1s*-     ^       pit 

;  si  -,  V   ■:,:.;-.: 

::  ^      _^^^--     g-u^ 

*" r2i=-i-±±-- 

-  •  -  --     -  Se--)-       ■;,. 

"■^""it      "    fe    ""]"" 

-    ;  ■  1  iy 

^    -„j_       ,    .,|- 

j^ ^.i...„     ---r                    ; 

-.    i   r-h-   -^    i        ^        ^  '  '  '  ' 

.    ,                  I 

1«8                         {     '             ,}■                   I 

.«i      ^  -^j\"        :  1     '^     'r' 

i           1                    ;                                          ,                               1          1     '     ' 

_l_                 '          J 

^<»                   ^"t      ±X."          J- 

X                                                      '    ' 

jjjgj _            rj_     _„     J-               ^ 

-\      \      -         -      - 

-■      — r                      .  :::"r,  .. 

i&K  —  ..,..,>           ;  I  "~  ' 

^     "     ■     J        ■    '       M±    ^-  ■  ' 

l^J .    ,     .  L              .^            ^^          ^ 

,«*.  —       '  /  ,  i_  i^^qr 

T 

^**         ^  .    I    u^i  '    1-^^.  ■ 

■   ■   ■   •  ^:.I^  _:x::±Fr 

,^  "'                X             ^    -r^t 

1190 "         -^     hT r    I 

'                  ""             1      1       '                          '   \ 

\'     ^'-j       ■ ^--         ^ 

^          _!  4-       J-  -                i 

,dm  ^it^     .  '"^   ":  iNt  ■     ^  "  -     ■ 

■    '  -^X      ^               ,    .      X 

1W2 -^      *-    1  .               \>j.    ■ 

.    >.^    X       X.--                     :      .  -  ' 

i«B::::~::":::-r-.^  L-. 

X   E=-±i  Xx±S 

im --   _ "ir^  . 

,  Ss     -J- ^  4-4-^ 

■    ._„X-*:^        TiX 

[    :    [V     --1  i— ^ 

<  :    1        ^^i             '    i  ,|  ' 

^!^J ^X!!^_^    -             1   1    ' 

\   T\  ,        J  > -^      ■    '    !  — 

rt«u      '^               •       I'    T                       ■"       ; 

^      -         "^t^            -taF'"^      ■         1         .     1    4              ' 

.  i  >,  ,   .'*=^.  ..   ■   :      . :_ 

IV.                       n  ,     ,            :            ,           n            r 

1910 '                             

^            "T    >J     !     7   !     !•    '^1^ 

_j                     |_                                    .                                    L                   . 

XX    -  -    1     '  \    \     I  \         ^^ 

wii —  ^-     -H-      ,  "_^j;i 

"    t     X    :    ,    A       >  J  -     .  ;-4^ 

II^                   V        ST 

x^'ix-iEii^^i 

iflia '        '      "^      -"  , 

Tj --   i  ^<;      ^X 

iti  "    "    X       X  '  ^i>i 

-:tit±±vj±5^Jj 

g     1      i     i     a 

Goal  ProductloD  i 

§    ^    §    1    §    §    § 

oSbort  Tons  (200011)1.) 

FiQ.  5. — Yearly  Coal  and  Oil  Production  in  United  States. 


Digitized  by  CjOOQ IC 


16  GENERAL  INTRODUCTION 

and  in  justice  to  the  welfare  of  the  nation  and  of  coming  genera- 
tions every  practicable  means  should  be  employed  for  reduc- 
ing it. 

A  material  saving  has  been  effected  by  the  introduction  of  more 
efficient  apparatus  and  improved  systems  of  operation.  In  the 
modern  central  station,  very  great  economies  have  been  the  result 
from  the  substitution  of  a  few  large  and  highly  efficient  boilers 
and  steam  turbines  for  a  large  number  of  relatively  small  and 
uneconomical  units,  and  from  the  introduction  of  plant  economics 
and  skill  not  attainable  in  the  smaller  plants.  The  fuel  economy 
of  the  gas  and  oil  engine  is  well  appreciated.  While  their  devel- 
opment has  been  slow,  a  number  of  large  gas-engine  plants  have 
been  built  during  the  last  few  years,  and  it  is  quite  possible  that 
the  gas  or  oil  engine  will  in  the  future  be  used  to  a  great  extent 
for  the  production  of  power.  The  application  of  the  power  directly 
to  the  work  through  electric  motors  instead  of  indirectly  through 
inefficient  countershafting  and  belting  has  also  resulted  in  a  very 
material  increase  in  economy. 

Beyond  the  above  gains,  which  may  be  considered  well  within 
the  limits  of  possible  attainment  by  our  present  knowledge,  it  is 
reasonable  to  assume  that  the  efficiency  of  our  fuel  engines  will 
not  be  increased  very  materially  in  the  near  future,  and  the  only 
safe  course  of  accomplishing  a  reduction  in  the  consiunption  of 
our  natural  fuel  resources  is  to  utihze  the  enormous  energy  of  the 
numerous  water  powers  which  is  now  going  to  waste. 

Based  on  the  Census  Report  the  developed  water  powers  of 
this  country  may  be  taken  as  approximately  6  million  horse-power. 
Assuming  that  one  hydraulic  horse-power  corresponds  to  an 
annual  coal  consumption  of  8  tons,  it  follows  that  the  utilization 
of  this  water  power  means  a  yearly  saving  in  the  coal  consumption 
of  48  million  tons. 

In  the  recent  Report  of  the  Bureau  of  Corporations  the  min- 
imum water  power  in  this  country  which  can  be  readily  developed 
is  placed  at  31  million  horse-power.  This  enormous  power, 
which  is  now  entirely  going  to  waste,  could,  if  developed,  effect 
a  yearly  saving  of  250  million  tons  of  coal,  besides  releasing  about 
750,000  men  for  other  work,  and  in  addition  dispense  with  the 
tremendous  railroad  equipment  required  for  its  transportation. 


Digitized  by 


Google 


AVAILABLE  AND  DEVELOPED  WATER  POWERS 


17 


AVAILABLE  AND  DEVELOPED  WATER  POWERS  IN  UNITED  STATES 

The  surveys  and  examinations  necessary  to  a  thorough  and 
ccurate  report  of  the  water-power  resources  of  the  United  States 
have  never  been  completed.  While  in  certain  parts  of  the  coun- 
try they  are  fairly  well  known,  in  other  parts,  however,  the  infor- 
mation is  very  fragmentary,  and,  therefore,  an  estimate  of  the 
available  water  powers,  such  as  given  in  Table  III,  must  neces- 
sarily be  considered  approximate. 


TABLE  III 
Estimated  Available  Water  Power  in  UNrrED  States 


Drainage 
Area  in 
Square 
MUes. 

Flow  per 

Annum  in 

Billion 

Cu.ft. 

Horse 

-POWBB. 

Principal  Drainaeres. 

Minimum. 

Assumed 
Maximum 
Develop- 
ment. 

North  AtUntlc  to  Cape  Henry.  Va.  .  . 

Southern  Atlantic  to  Cape  Sable,  Fla . 

Eastern  Qulf  of  Mexico  to  MissiBsippi 

River 

159.879 
123.920 

142.220 

433.700 
333,600 

905.200 
299.720 
225.000 

70.700 
290.400 
223,000 

62.150 

8.942 
5,560 

6,867 

2.232 
12.360 

9.580 

8.583 

521 

2.193 

15.220 

614 

1.761,000 
1,050.000 

466,000 

362.000 
2.180,000 

3,300,000 

5.570,000 

2,425.000 

2.680.000 

10.750,000 

433,000 

63.000 

3,481.000 
1,630,000 

803,000 

Western  Gulf  of  Mexico  west  of  Ver- 
million River 

686.000 

Miasiasippi  River  (tributaries  from  east) 
Mississippi     River     (tributaries    from 
west,  including  Vermillion  River).  . . 
St.  Lawrence  River  to  Canadian  line. . 
Colorado  River  above  Yuma,  Ariz .... 
Southern  Pacific  to  Point  BoniU.  Calif. 
Northern  Pacific 

4,450.000 

5.900,000 
6.740.000 
4.610.000 
6.500.000 
20,500,000 

Great  Basin 

670.000 

Hudson  Bay 

175,000 

Total 

3,260,490 

72,672 

31.040,000 

56,146.000 

These  values  are  based  on  estimates  prepared  by  the  United 
States  Geolc^cal  Survey  for  the  National  Conservation  Com- 
mission, 1908.  With  some  revision,  owing  to  lack  of  data  avail- 
able at  that  time,  these  estimates  would  place  the  minimum  water 
power  of  the  country  at  approximately  31  milUon  horse-power 
and  the  maximum  at  56  millions. 

In  arriving  at  this  minimum  horse-power,  the  minimum  flow  for 
the  two  lowest  seven-day  periods  in  each  year  for  seven  years  was 


Digitized  by 


Google 


18  GENERAL  INTRODUCTION 

determined  and  the  mean  of  these  values  for  the  period  of  record 
was  taken  as  the  minimum  flow.  It  is  obvious  that  this  is  some- 
what higher  than  the  absolute  minimum,  but  the  latter  is  usually 
of  so  short  duration  that  it  would  not  be  practicable  or  profit- 
able to  develop  a  site  on  this  basis.  The  efficiency  of  the  hydro- 
electric equipment  has  been  assumed  to  be  75  per  cent. 

The  assumed  maximum  power  has  been  based  upon  the  con- 
tinuous power  indicated  by  the  flow  of  a  stream  for  the  six  months 
of  the  year  showing  the  highest  flow.  The  average  flow  for  the 
lowest  week  of  the  lowest  month  of  these  six  highest  months  was 
then  taken  as  the  assumed  maximum  for  the  year.  The  yearly 
averages  thus  obtained  were  then  themselves  averaged  for  a 
series  of  years.  It  is,  however,  common  practice  to  estimate  on 
the  continuous  power  for  nine  months  instead  of  six,  which  would, 
of  course,  reduce  the  amount  of  maximum  power  available. 

The  above  estimates  do  not  include  any  storage  possibilities 
and  a  commercial  development  of  the  maximum  power  would 
have  to  be  based  on  the  assumption  that  it  would  be  profitable  to 
install  auxiliary  fuel  plants  to  supplement  the  deficiencies  during 
the  remaining  six  months  of  the  year. 

An  endeavor  has  been  made  to  determine  the  maximum  power 
that  might  be  produced  if  all  the  practicable  storage  facilities  on 
the  drainage  areas  were  utilized.  Surveys  on  many  of  the  basins 
make  possible  a  fairly  close  estimate,  but  inasmuch  as  fully  three- 
fourths  of  the  country  has  not  been  surveyed  in  a  manner  suitable 
for  this  purpose,  only  rough  estimates  can  be  given  for  the  entire 
area.  It  may,  however,  be  assumed  with  confidence  with  all 
practicable  storage  sites  utilized  and  the  water  properly  applied, 
there  might  be  established  eventually  in  the  country  a  total  water- 
power  installation  of  at  least  ICO  million  horse-power  and  possibly 
more.  It  should,  however,  not  be  assumed  that  all  this  power  is 
economically  available  to-day.  Much  of  it,  indeed,  would  be  too 
costly  in  development  to  render  it  of  commercial  importance  under 
the  present  condition  of  the  market  and  the  price  of  fuel  power.  It 
represents,  on  the  other  hand,  the  maximum  possibilities  in  the 
day  when  our  fuel  shall  have  become  so  exhausted  that  the  price 
thereof  for  production  of  power  is  prohibitive,  and  the  people  of 
the  country  shall  be  driven  to  the  use  of  all  the  water  power  that 
can  reasonably  be  produced  by  the  streams. 

The  total  developed  water  power  of  the  United  States,  exclud- 


Digitized  by 


Google 


AVAILABLE  AND  DEVELOPED  WATER  POWERS  19 

ing  developments  of  less  than  1000  H.P.  each,  as  computed  by  the 
Bureau  of  Corporations  and  given  in  Table  IV,  is  4,016,127  H.P. 
Of  this,  2,961,549  H.P.  is  classed  as  "  commercial "  power,  and 
1,054,578  H.P.  as  "manufacturing"  power.  Adding  2,000,000 
H.P.  to  represent  the  power  of  developments  of  less  than  1000 
H.P.  each,  gives  a  grand  total,  in  roimd  numbers,  of  at  least 
6,000,000  H.P.  and  possibly  6,500,000  as  the  total  water  power  of 
the  United  States  developed  and  under  construction. 

There  is  a  marked  geographical  concentration  of  developed 
water  power.  Thus,  nearly  50  per  cent  of  the  developed  "  com- 
mercial "  water  power  of  the  coimtry  is  located  in  five  States  as 
follows: 

Per  Cent 

California 14 

New  York 13 

Washington 10 

Pennsylvania 6 

South  Carolina 6 

Total 48 

An  even  more  marked  concentration  of  developed  water  power 
employed  in  manufacturing  is  shown  by  the  following  summary. 

Per  Cent 

New  York 30 

New  England  States 36 

Minnesota  and  Wisconsin 17 

South  Carolina 6 

Total 88 

The  accompanying  map,  Fig.  6,  shows  the  location  of  water- 
power  developments  and  power  sections  of  streams  in  the  United 
States. 


Digitized  by 


Google 


20 


GENERAL  INTRODUCTION 


TABLE  IV 
Developed  Water  Power  in  the  United  States  of  Concerns  Havino 
1000  H.P.  or  Over  (Including  Undeveloped  Power),  by  States 

(Compiled  by  Bureau  of  Corporations.  1912) 


State. 


Developed  and 
under  constbuction. 


Commer- 
cial. 


Manufac- 
turing. 


Undevel- 
oped.« 


Total 


United  States 

North  Atlantic  States: 

Maine 

New  Hampshire 

Vermont 

Massachusetts 

Connecticut 

New  York 

New  Jersey 

Pennsylvania 

South  Atlantic  States: 

Virginia 

West  Virginia 

North  Carolina 

South  Carolina 

Georgia 

Florida 

North  Central  States: 

Ohio 

Indiana 

Illinois 

Michigan 

Wisconsin 

Minnesota 

Iowa 

South  Dakota 

Kansas 

South  Central  States: 

Tennessee 

Alabama 

Western  States: 

Montana 

Idaho 

Colorado 

Arizona 

Utah 

Nevada 

Washington 

Oregon 

California 

Other  States,  not  enumerated  ^ 
Summary. 

North  Atlantic  States 

South  Atlantic  States 

North  Central  States 

South  Central  States 

Western  States 

Other  States,  not  enumerated  «. 


H.P. 
2.061.549 


H.P». 
1.064.678 


H.P. 
2.638.528 


H.P. 
6.664.656 


66.360 
16.450 
63.648 
76.697 
32,000 

398.058 
7.200 

169,632 

33.700 

5.260 

82.960 

135.040 

126.927 

6.000 

4.026 

10.425 

38.460 

102.682 

96.790 

95.815 

161.400 

6.000 

6.800 

62.000 
6.000 

139.260 
52,100 
69.690 
16.200 
52.700 
14.300 

300.510 
95.777 

429.467 
4.317 

819.045 
388.877 
511.406 
68.000 
,169.904 
4.317 


168,338 

103.658 

40.197 

53.922 

16,519 

316.313 


17.620 
16,160 
14.050 
47,467 
12.360 


4.260 

12,751 

30.420 

106.153 

72,200 


10.450 


6.000 
7.780 

696.947 

107.627 

225.774 

10.450 

6,000 

7.780 


100.000 

13,600 

44,460 

14,620 

4,000 

103.093 

i3!i43 

44,800 

1.260 

61,426 

96,686 

286.360 


6,676 

1,000 

62,100 

117,660 

91,400 

101.600 

161,000 

8,167 

200 

3.862 


105.700 
42,300 
59.000 

2,600 

24,000 

116,700 

143.600 

732.749 

2.000 

382.815 
489.410 
634.792 

3.862 
1.225.649 

2,000 


333.608 
133.608 
138.305 
146.239 
61,619 
906,464 
7,200 
182.774 

96.120 

22.660 

168,436 

278.082 

425.627 

6.000 

10.700 

15.675 

113.311 

260.762 

294.352 

269,616 

302,400 

8.167 

7.000 

65.862 
16,460 

244.960 
94.400 

128,690 
16.200 
66.300 
38,200 

416,210 

230.377 

1,168.216 

14.097 

1,808.807 

086,014 

1.271,072 

82.313 
2.401.663 

14.007 


>  Ownership  of  less  than  1000  H.P.  excluded.  States  omitted  from  this  table  had 
no  concerns  reporting  developed  water  powers  of  1000  H.P.  or  over,  except  as  Indicated 
in  note  1.  p.  63. 

9  Embracing  one  concern  in  Missouri  and  four  each  in  Maryland  and  Rhode  Island 

s  The  Census  Report  for  1910  gives  the  total  water  power  used  in  manufacturing 
•8  1.822.593  H.P. 

<  Sites  on  which  expenditures  have  been  made. 


Digitized  by 


Google 


AVAILABLE  AND  DEVELOPED  WATER  POWERS 


21 


ai 

P 

a 


S 


Digitized  by 


Google 


22  GENERAL  INTRODUCTION 


POWER  FROM  INLAND  WATERWAYS 

There  are  great  possibilities  of  hydro-electric  power  develop- 
ments in  connection  with  inland  waterways,  and  this  subject 
should  be  given  careful  consideration  when  improvements  or 
new  projects  are  contemplated.  The  advantages  to  conmiuni- 
ties  through  the  development  of  such  water  powers  would  be, 
besides  the  benefit  of  cheap  electric  power,  the  prevention  of  floods 
and  increased  efficiency  of  river  navigation. 

The  low  water  in  many  rivers  during  the  dry  season  would 
absolutely  prevent  navigation  unless  dams  with  locks  were 
provided  for  raising  the  water  level,  while  on  the  other  hand 
there  are  a  very  large  number  of  streams  that  are  not  now  navi- 
gable at  all,  but  which  could  easily  J^e  converted  into  streams  of 
great  commercial  value. 

When  a  dam  is  to  be  built  for  improving  the  navigation  of  a 
river,  consideration  should,  therefore,  always  be  given  to  the  fact 
that  every  dam  not  used  for  the  development  of  electrical  energy 
means  just  so  much  loss  of  income.  Such  dams  should,  therefore, 
be  built  of  adequate  height  for  possible  hydro-electric  develop- 
ment. 

The  prevention  of  floods  is  also  of  the  utmost  importance. 
In  the  United  States  alone  the  yearly  flood  loss  has  for  a  number 
of  years  exceeded  several  hundred  milUon  dollars. 

Storage  and  levee  systems  appear  to  be  the  only  practical 
.solution  for  flood  prevention.  Storage  of  flood  waters  is  effected 
by  forests  and  similar  surface  vegetation  and  by  artificial  reser- 
voirs. The  amount  stored  by  forests  is  and  probably  will  be  for 
a  long  time  to  come  indeterminate,  since  the  forest  is  merely  an 
agent  in  assisting  the  ground  to  absorb  the  water.  This  is,  there- 
fore, essentially  a  ground  storage,  and  the  ability  of  the  forest  to 
enhance  this  is  dependent  absolutely  on  the  soil  beneath  the  forest. 

The  extent  to  which  flood  waters  could  be  stored  by  reservoirs 
depends  on  the  available  reservoir  capacity  in  the  several  river 
basins.  As  a  rule,  the  more  diversified  the  character  of  the 
basins,  especially  in  contour,  the  greater  facilities  they  afford  for 
reservoir  storage.  Large  portions  of  many  rivers  are  not  subject 
to  correction  by  reservoirs,  as  in  the  Mississippi  Vallej'^  for  example. 
It  is,  therefore,  probable  that  streams  draining  one-third  of  the 
area  of  the  United  States  must  forever  be  subject  to  floods,  and  the 


Digitized  by 


Google 


POWER  FROM  INLAND  WATERWAYS  23 

only  treatment  that  now  appears  feasible  for  these  streams  is  the 
construction  of  levee  systems.  For  the  remaining  two-thirds  of 
the  United  States,  investigations  made  indicate  that  from  55  to 
60  per  cent  of  the  flood  waters  can  be  saved  by  the  utiUzation  of 
maximum  storage  capacity.  Although  the  cost  of  such  construc- 
tion would  be  enormous  in  the  aggregate,  it  is  apparent  that  the 
saving  that  would  accrue  for  relief  from  flood  damages  alone  would 
soon  return  the  entire  investment. 

In  addition,  the  construction  of  storage  reservoirs  will  natur- 
ally have  a  very  great  bearing  on  the  possibiUties  of  power  devel- 
opments. The  stream  can  be  regulated  and  the  flows  equaUzed 
by  storing  the  water  dming  the  wet  season  and  using  the  same  to 
increase  the  volume  of  the  stream  through  the  dry  season.  This 
means  a  consequent  increase  in  the  power  value  of  the  stream  due 
to  augmenting  the  low-water  flow.  It  is  thus  estimated  that  in 
this  manner  the  economical  water-power  possibilities  of  the  United 
States  would  be  increased  to  about  60  million  horse-power. 

Striking  examples  of  what  may  be  accomplished  by  an  eflScient 
regulation  of  navigable  rivers  is  shown  at  Keokuk  on  the  Mississippi 
River,  and  at  Hale's  Bar  on  the  Tennessee  River.  In  both  cases 
Federal  grants  were  given  to  private  companies  for  constructing 
a  dam  across  a  large  navigable  river,  the  result  being  a  com- 
bined river  improvement  and  a  power  development  of  immense 
size. 

The  Sanitary  District's  Canal,  at  Chicago,  with  its  50,000  H.P. 
power  development  at  Lockport,  111.,  clearly  illustrates  the  great 
possibilities  in  connection  with  canals.  This  subject  has  also  been 
given  careful  consideration  in  connection  with  the  Barge  Canal  in 
the  State  of  New  York,  and  the  principal  water  powers  created 
by  this  canal  are  given  in  Table  V.  From  this  it  is  seen  that  the 
increased  power  possibilities  attributable  to  it  will  amount  to 
about  40,000  H.P. 

The  possibilities  of  power  developments  in  connection  with 
water  supply  systems  is,  on  the  other  hand,  illustrated  by  the 
Lob  Angeles  Aqueduct.  This  has  a  length  of  about  250  miles  and 
a  capacity  of  258  million  gallons  of  water  every  twenty-four  hours. 
The  flow  of  this  water  will  be  utiUzed  for  generating  a  total  of 
90,000  Kw.  of  electric  energy  at  a  number  of  power  stations  along 
the  route,  from  where  it  will  be  transmitted  to  Los  Angeles.  It  is 
estimated  that  the  sale  of  this  energy  will  take  care  of  all  the  bonds 


Digitized  by 


Google 


24 


GENERAL  INTRODUCTION 


TABLE  V 
Summary  of  Principal  Water  Powers  Created  by  Barge  Canal  * 


Location. 


DiSTBXBUTBD  HtDBAUUC  HoBSE-POWKB 
WITH    ECONOMICAI.   DeVBLOPMXNT 


Before  Canal.        With  Barge  Canal. 


Lockport 

Rochester 

Baldwinsville . . 
Oswego  River. . 
Vischers  Ferry . 

Crescent 

Waterford 


1,700 


2,452 
33,960 


4,530 
9,732 
2,640 
41,640 
6,530 
6,980 
6,506 


>  From  Sixth  Annual  Report  of  N.  Y.  State  Water  Supply  Commission. 

and  interest  charges  upon  both   the  aqueduct  system  and  the 
entire  hydro-electric  installation. 

PRIMARY  POWER  AND  ITS  USES 

Statistics  have  never  been  compiled  giving  accurately  the 
total  mechanical  horse-power  used  in  the  United  States.  The 
following  estimate  may,  however,  be  considered  to  be  fairly  close 
to  the  actual  conditions,  and  it  is  safe  to  place  the  present  value 
at  approximately  180  milUon  horse-power,  or  nearly  two  horse- 
power per  capita  for  the  entire  population. 

TABLE  VI 
Primary  Power  in  UNrrED  States 

H.P. 

Manufacturers 26,000,000 

Central,  stations 8,500,000 

Isolated  plants 4,500,000 

Street  and  electric  railways 4,000,000 

Steam  railroads 50,000,000 

Steam  and  naval  vessels 5,000,000 

Mines  and  quarries 6,000,000 

Flour,  grist  and  saw  mills 1,500,000 

Irrigation 500,000 

Automobiles 50,000,000 

Horses  and  mules 25,000.000 

Total 180,000,000 


Digitized  by 


Google 


PRIMARY  POWER  AND  ITS  USES 


25. 


The  rapid  growth  in  central  electric  Ught  and  power  stations, 
as  taken  from  the  latest  Census  Report,  is  shown  in  Table  VII. 

TABLE  VII 
Central  Electric  Light  and  Power  Stations 


Per  Cent 

• 

1912 

1907 

1902 

of 
increase: 
1902-1912 

Number  of  stations  ^ 

5.221 

4,714 

3,620 

44.2 

Commercial 

3.659 
1.562 

3,462 
1,252 

2.805 
815 

30.4 

Municipal 

91.7 

Total  income 

$302,115,599 

$175,642,338 

$85,700,605 

252.5 

Light,  heat,  and  power. 

including  free  service . 

$286,980,858 

$169,614,691 

$84,186,605 

240.9 

All  other  sources 

$15,134,741 

$6,027,647 

$1,514,000 

899.7 

Total     expenses,     including 

salaries  and  wages 

$234,419,478 

$134,196,911 

$68,081,375 

244.3 

Total  number  of  persons  em- 

ployed   

79,335 

47,632 

30.326 

161.6 

Total  horse-power 

7,528.648 

4.098,188 

1,845,048 

308.0 

Steam      engines      and 

steam  turbines: 

Number 

7,844 
4.946.532 

8.054 
2,693,273 

6,295 
1,394.395 

24.6 

Horse-power 

254.6 

Water  wheels 

Number 

2.933 
2,471,081 

2.481 
1,349,087 

1.390 
438.472 

111.0 

Horse-power 

463.6 

Gas  and  oil  engines: 

Number 

1,116 
111,036 

463 
55.828 

165 
12.181 

576.4 

Horse-power 

811.5 

Kw.  capacity  of  dynamos. . . 

5,134.689 

2,709,225 

1,212.235 

323.6 

Kw.  capacity  per  station .  .  . 

983 

574 

334 

194.3 

Cost    of    construction    and 

equipment 

$2,175,678,266 

$1,096,913,622 

$504,740,352 

331.4 

Cost  per  kilowatt  capacity. . 

$425 

$404 

$416 

Output  of  stations,  kw.-hrs . 

11,602,963,006 

5,«62,276,737 

2,507,051.115 

358.8 

Estimated  number  of  lamps 

for  service: 

Arc 

505,395 

562,795 

385.698 

31.0 

Incandescent  and  other 

varieties 

76,507,142 

41,876,332 

18.194.044 

320.5 

Stationary  motors  served : 

Number 

435,473 

167,184 

101.064 

330.9 

Horse-power  capacity 

4.130,619 

1.649.026 

438,005 

843.1 

>  The  term  "  station  "  as  here  used  may  represent  a  single  electric  station  or  a 
number  of  stations  operated  under  the  same  ownership. 

The  statistics  represent  all  central  stations  which  furnish  elec- 
trical energy  for  Ught,  power  and  heat;  for  manufacturing,  mining 


Digitized  by 


Google 


•26  GENERAL  INTRODUCTION 

and  other  commercial  enterprises;  for  private  dwellings;  and  for 
public  uses  such  as  lighting  streets,  parks,  etc.  They  do  not 
include  electric  plants  operated  by  factories,  hotels,  etc.,  which 
consume  the  current  generated;  those  operated  by  the  Federal 
Government  and  State  institutions;  or  plants  that  were  idle  or 
in  course  of  construction. 

Aside  from  the  growth  in  the  number  of  stations  the  striking 
features  of  the  above  table  are  the  relatively  larger  increase  in 
the  kilowatt  capacity  per  station,  while  the  cost  of  construction 
and  equipment  remains  practically  the  same.  That  this  cost 
has  not  been  materially  reduced  is  no  doubt  due  to  the  increased 
cost  of  the  distributing  and  transmission  lines,  which  form  an 
important  part  of  the  total  cost  of  the  system. 

It  is  also  of  interest  to  note  that  the  percentage  increase  in 
the  use  of  water  power  for  the  period  of  1902  to  1912  was  463  per 
cent,  as  compared  to  264  per  cent  for  steam  power.  On  the 
other  hand,  gas  power  increased  811  per  cent,  but  this  is  not 
of  any  great  importance,  as  the  horse-power  capacity  of  the 
gas  engines  installed  at  the  beginning  of  above  period  was  very 
small. 

Water  power  was  used  more  extensively  than  steam  in  the 
manufacturing  industry  prior  to  1870.  Since  that  time,  however, 
it  declined  steadily,  while  the  use  of  steam  power  increased, 
reaching  a  maximum  of  about  87  per  cent  in  1900.  There  has 
since  been  a  marked  falling  off  in  the  percentage  of  directly  applied 
steam  power  and  this  has  been  due  to  the  rapid  introduction  of 
electric  power.  The  increased  use  of  the  electric  motor  for  driving 
industrial  machinery  has  been  phenomenal  and  this  is  again  best 
illustrated  by  a  reference  to  the  Census  Report. 

Table  VIII  shows  for  all  industries  combined  the  horse- 
power of  engines  and  motors  employed  by  manufacturing  con- 
cerns for  the  period  from  1870  to  1909.  The  figures  for  the  total 
primary  power  exclude  duplication  and  represent  the  primary 
power  of  engines,  water  wheels,  etc.,  owned  by  the  manufacturing 
establishments  themselves  plus  the  electric  and  other  power  pur- 
chased from  outside  concerns.  Especially  striking  is  the  increased 
use  of  electric  motor  applications  during  this  period.  While  the 
primary  power  increased  about  85  per  cent,  the  application  of 
electric  motors  for  manufacturing  industries  alone  increased  close 
to  900  per  cent. 


Digitized  by 


Google 


PRIMARY  POWER  AND  ITS  USES 


27 


TABLE  VIII 
Power  Used  in  Manufacturing  Industries 


Primary  i>ower,  total. 

Owned,  total 

Steam 

Gan 

Water 

Other 

Rented,  total 

Electric 

Other 

Electric  motors,  total. 
Run  by  own  power  . . 
Run  by  rented  power. 


1870 


2,346.142 


1,215,711 
1.130.431 


1880 


3.410,83; 


2.185,468 
1,225.370 


1890 


5.939,086 
5,850.516 
4.681,596 

8.930 
1.255,206 

4,784 
88.571 


88.571 
15.569 


1899 


10,097, 

9.778, 

8.139, 

134, 

1.464, 

49, 

319, 

182, 

136, 

492, 

310, 

182, 


,893 
,418 
,579 
,742 
,112 
,986 
.475 
,662 
,913 
,936 
,374 
,562 


1904 


,487.707 
,854,806 
,825,348 
289.423 
.647,880 
92.154 
632.902 
441.589 
191.313 
.592.475 
.150.886 
441,689 


1909 


18.680,776 

16.808.106 

14.202,137 

754.083 

1,822.593 

29.293 

1,872,670 

1.749.031 

123.639 

4.817,140 

3.068.109 

1.749.031 


i 

, 

^^ 

Ste 

am 

Poi 

or 

— 

— 



— 

... 

^ 

1    ' 

1 

- 

— 

-- 

I 

— 

-- 

-- 

^ 

1 

' 

Vati 

rP 

>we 

r 

— 

— " 

~ 

U-* 

• 

'_ 

Oas 

Power 

^__ 

^_ 

^_ 

— 

-^ 

— 

*— ' 

L— J 

.^m-J 

^^ 

1.— 1 

L— 1 

1— — 1 

L— 1 

L— J 

1— ^ 

L_J 

1      1 

^—1 

L_J 

1      1 

1900 


1904 


Year 


1906 


1912 


100 
00 
80 
70 
60 
30 
40 
30 
20 
10 


a 

s 

£ 


FiQ.  7.— Relation  of  Steam,  Water  and  Gas  Power. 


The  curves  in  Fig.  7  show  the  approximate  percentage  relation 
that  steam,  water  and  gas  power  bear  to  the  total  in  the  three 
principal  industries — Central  Stations,  Electric  Railways  and 
Manufacturing. 


Digitized  by 


Google 


28 


GENERAL  INTRODUCTION 


COMMERCIAL  OPPORTUNITIES  FOR  HYDRO-ELECTRIC  POWER 

During  recent  years  there  has  been  a  very  large  increase  in  the 
number  and  variety  of  electric  power  applications,  and  this  has  a 
very  important  bearing  in  stimulating  the  development  of  water 
powers.  Among  the  more  important  industries  affected  may  be 
mentioned:  Agricultural  work,  including  irrigation,  textile  mills, 
mining,  electrochemical  work,  railroad  electrifications,  etc. 

Agricultural  Work.  The  possibiUties  of  the  use  of  hydro- 
electric power  in  connection  with  farming  and  agricultural  work 


Fig.  8.— Operating  Thresher  at  Night  with  Portable  Motor  Outfit. 

are  many  and  offer  one  of  the  most  promising  fields  of  the  future. 
The  unqualified  success  that  the  appUcation  of  electric  power 
has  had  in  this  line  of  work  indicates  that  it  has  become  a  factor 
of  such  importance  that  it  must  now  be  seriously  considered  as 
affecting  both  the  cost  and  quality  of  the  products  of  the  modem 
farm.  Compared  to  other  forms  of  applied  power,  the  chief 
advantages  of  electricity  are  reliability,  safety,  cleanliness  and 
flexibility  in  application.  Power  can  be  readily  and  economically 
distributed  to  the  scattered  location  of  the  various  buildings  where 
the  cost  of  providing  separate  engines  would  be  practically  pro- 


Digitized  by 


Google 


COMMERCIAL  OPPORTUNITIES 


29 


hibitive.  Fire  risk  is  reduced  to  a  minimum,  which  is  of  greatest 
importance  on  isolated  farms,  where  fire-fighting  appliances  are 
limited,  Fig.  8.  With  a  number  of  motors  installed  for  the  various 
classes  of  service,  the  operating  periods  can  be  so  arranged  as  to 
secure  a  very  good  load-factor,  thus  minimizing  the  cost  of  power. 

The  power  supply  may  be  obtained  from  the  extensive  net- 
works of  high-tension  transmission  lines  which  are  now  being 
erected  in  so  many  sections  of  the  country,  and  which  are  con- 
tinuously being  extended  at  a  very  rapid  rate.  While  this  supply 
without  doubt  oflFers  the  simplest  and  cheapest  source  of  power, 
there  are  thousands  of  small  streams  whose  wasted  energy  might 
readily  be  transformed  and  appUed  to  useful  work  on  farms  by 
the  installation  of  small  and  inexpensive  water-power  plants. 

The  following  tables  show  some  of  the  more  important  appli- 
catioDS  of  electric  drive  for  farm  machinery  and  power  required. 

TABLE  IX 
Motors  for  Farm  Machinert 


Machines. 


Horse-power  of  Motor. 


Minimum. 


Maximum. 


Size  Most 

Commonly 

Used  on  Aver- 

Kge  Farms. 


Feed  grinders  (small) 

Feed  grinders  (large) 

Ensilage  cutters 

Shredders  and  buskers 

Threshers,  19-inch  cylinder . 
Threshers,  32-inch  cylinder  . 
Com  sheilers,  single  hole . .  . 

Power  shellers 

Fanning  wnilla 

Grain  graders 

Grain  elevators 

Concrete  mixers 

Groomer,  vacuum  system. . . 
Groomer,  revolving  system . 

Hay  hoists 

Root  cutters 

Cord  wood  saws 

Wood  splitters 

Hay  balers 

Oat  crushers 


3 
10 
10 
10 
12 
30 

i 

10 


10 
30 
25 
20 
18 
50 

IJ 
15 


li 

2 

1 

1 

3 

1 

3 

1 

3 

2 


5 
10 

3 

2 
15 

5 
10 

4 
10 
10 


5 
15 

15-20 
15 
15 
40 

1 
15 
i 
i 

3 

5 

2 

1 

5 

2 

5 

2 

7i 

5 


Digitized  by 


Google 


30 


GENERAL  INTRODUCTION 


TABLE  X 
Power  Required  to  Thresh  a  Bushel  op  Grain 


Kind  of 
Grain. 

No.  of 
Tests 
Made. 

Yield  Per  Acre. 

Kw.-hr. 

to  Thresh 

1  Ton. 

Kw.-hr. 
to  Thresh 
1  Bushel. 

Cost  or  Powek 
AT  5  Cents  Per 

KW.-HB. 

Tons  of 

Grain 

and 

Straw. 

Bushels 
of  Grain. 

Per  Ton. 

Per 
Bushel. 

Oats 

Barley.  .  . 
Wheat.  .. 

31 

5 

10 

1.99 
2.27 
1.97 

73.6 
49.9 
27.9 

2.62 
2.36 
2.27 

0.070 
0.108 
0.160 

$0.13 
0.128 
0.113 

$0.0035 
0.005 
0.008 

TABLE  XI 
Power  Required  for  Grinding 


Operation. 


Capacity  of 
Machine 
per  Hour 

in  Bushels. 


H.P.  of 
Motor 

Kw.-hr. 
Required 

Required. 

per 
Bushel. 

20 

0.411 

3 

0.37 

2 

0.045 

15 

0.272 

7.5 

0.086 

Power  Cost 
per  Bushel 
with  Electric- 
ity Costing 
5  Cents  per 
Kw.-hr. 


Grinding  corn  on  the  cob . . 

Grinding  oats 

Crushing  oats 

Grinding  shelled  com 

Cracking  corn 


41 
5.7 
50 
41.5 
65.8 


$0.0205 
0.0185 
0.0022 
0.0136 
0.0043 


Inigation.  Water  is  a  necessity  for  the  growth  of  every  crop. 
In  the  Western  States,  the  rainfall  is,  as  a  rule,  insufficient  to  sup- 
port even  a  scant  growth  of  vegetation,  but  in  the  Central  and 
Eastern  States  the  average  rainfall  during  the  growing  season  is 
ordinarily  considered  sufficient.  However,  in  the  latter  sections 
of  the  country  hardly  a  year  passes  without  some  particular 
section  being  badly  in  need  of  rain. 

As  rains,  to  be  beneficial,  must  come  at  such  times  and  in  such 
amounts  as  will  properly  moisten  the  soil  and  produce  growth,  a 
check  in  this  supply  of  soil  moisture  at  any  stage  of  the  growth 
affects  both  the  quality  and  quantity  of  the  yield  and  may  greatly 
reduce  the  profits  of  the  grower.  The  real  test  of  the  necessity  of 
irrigation  is  not  the  total  annual  rainfall,  but  the  monthly,  and, 


Digitized  by 


Google 


COMMERCIAL  OPPORTUNITIES 


31 


in  the  case  of  most  crops,  the  weekly  amount  of  precipitation 
throughout  the  growing  season.  Under  average  conditions,  it  is 
safe  to  say  that  a  drought  occurs  whenever  the  rainfall  totals  less 
than  one  inch  in  any  fifteen-day  period  and  crops  will  usually 
suffer  if  they  do  not  receive  more  than  this  amount  of  rain,  espe- 
cially during  the  spring  and  early  summer  months. 

Prof.  F.  H.  King  in  his  book  on  irrigation  and  drainage  fur- 
nishes the  following  data  as  to  the  highest  probable  duty  of  water 
per  acre  for  different  yields  of  different  crops: 

TABLE  XII 
Duty  op  Water  por  Different  Crops 


Bushels  per  acre. 


15 


20 


30 


40 


50 


60 


Least  Number  of  Acre-inches  of  Water. 


Wheat 

Barley 

Oats 

Maize  (com). 
Potatoes 


4.5 
3.2 
2.3 
2.5 


Tons  per  Acre. 


6.0 
4.3 
3.1 
3.3 
0.4 


9.0. 
6.4 
6.7 
5.0 
0.6 


12.0 
8.5 
6.3 
6.7 
0.8 


15.0 

10.7 

7.8 

8.4 

1.0 


6 


18.0 
12.8 

9.4 
10.0 

1.2 


8 


Clover  hay . . 
Com  (green). 


4.4 
2.1 


8.8 
4.2 


13.3 
6.2 


17.7 
8.3 


26.5 
12.5 


35.0 
16.6 


Some  artificial  means  of  supplying  water  to  the  land  is  there- 
fore a  necessity  in  the  western  section  of  the  coimtry,  and  would 
be  excellent  insurance  to  the  central  and  eastern  parts  as  well. 

Two  general  methods  of  supplying  this  water  are  now  in  use: 
The  ordinary  gravity  flow,  such  as  that  of  taking  water  from  a 
reservoir  or  ditch;  and  the  mechanical  lift,  such  as  pumping  water 
from  a  well,  pond,  river  or  lake.  Of  the  two,  the  development 
of  the  mechanical  lift  has  been  far  more  rapid.  There  are  two 
reasons  for  this:  First,  because  the  land  which  can  be  econom- 
ically irrigated  by  the  gravity  method  has  been  practically  all 
taken  up;  and,  second,  because  the  farmer  can  pump  water  to 
almost  any  elevation,  and  in  this  way  he  is  enabled  to  irrigate 
land  which  is  above  his  source  of  water  supply.  This  is  impos- 
sible when  the  gravity  system  is  used. 


Digitized  by 


Google 


32  GENERAL  INTRODUCTION 

Irrigation  pumping,  from,  the  farmer's  point  of  view,  has 
many  advantages,  in  that  a  pmnping  plant  will  give  him  water 
just  at  the  time  he  wants  it,  and  this  is  a  more  important  factor 
to  him  than  the  saving  of  the  money  effected.  It  is  exceptional 
to  be  able  to  get  water  just  at  the  time  when  it  is  wanted,  when 
irrigating  from  a  ditch,  as  ditch  riders  and  water  superintendents 
must  serve  all  alike.  Not  only  this,  but  when  water  is  turned 
into  a  ditch,  it  must  run  in  quantities  in  order  to  secure  economy, 
and  it  is  not  possible  that  every  man  along  a  ditch  will  be  similarly 
situated  with  regard  to  the  progress  of  his  work  so  that  all  will 
require  water  at  any  one  time. 

If  water  is  to  be  pumped,  some  kind  of  power  is  necessary  to 
operate  the  pump.  Among  the  more  important  sources  of  power 
are  the  gasolene  engine,  steam  engine,  and  electric  motor.  The 
latter,  however,  is  rapidly  displacing  the  other  two  wherever 
electric  power  is  available,  just  as  it  has  already  done  in  the  city. 
The  principal  advantage  of  the  electric  motor  is  that  its  power  is 
instantaneously  available  and  it  will  always  run  when  wanted. 
Furthermore  it  can  be  rim  for  months  at  a  time  without  shut- 
ting down  the  plant,  and  there  are  thousands  of  electric  pump- 
ing installations  in  the  Far  West  which  run  twenty-four  hours  a 
day  for  six  months  at  a  time;  this  being  entirely  feasible  as  the 
only  attendance  that  is  required  for  electrical  equipment  is  an 
occasional  oiling  of  the  motor  bearings.  The  steam  engine,  on 
the  other  hand,  requires  the  constant  attendance  of  a  Ucensed 
engineer,  while  the  gasolene  engine  has  a  large  number  of  moving 
parts,  which  must  necessarily  be  adjusted  from  time  to  time.  It  is 
practically  impossible  to  operate  a  gasolene  engine  for  six  months 
at  a  time  without  extensive  repairs  at  the  end  of  the  period.  Being 
able  to  run  the  electric  motor  all  the  time  is,  therefore,  a  distinct 
advantage,  in  that  a  small  reservoir  can  be  used  to  store  the  water 
pumped  during  the  night,  and  in  this  way  a  much  smaller  equip- 
ment can  be  used  than  would  otherwise  be  required.  The  electric 
motor  has  the  added  advantage  of  remote  control,  the  fanner 
being  able  to  stop  and  start  it  even  if  he  is  several  miles  away. 

The  advantages  of  electric  power  for  irrigation  purposes  have 
been  clearly  demonstrated  by  the  excellent  work  which  is  being 
done  by  the  United  States  Reclamation  Service,  the  United  States 
Indian  Service,  and  numerous  cooperative  and  individual  enter- 
prises.   The  Salt  River  project  in  Arizona,  when  completed,  will 


Digitized  by  VjOOQ IC 


COMMERCIAL  OPPORTUNITIES  33 

furnish  irrigation  to  over  one-quarter  million  acres  of  arid  lands 
in  the  Southwest,  and  the  Minidoka  project  in  Southern  Idaho 
will  be  capable  of  irrigating  approximately  fifty  thousand  acres. 
In  connection  with  these  projects,  electricity  plays  an  important 
part.  Hydro-electric  power  is  generated  on  the  nearest  available 
river  and  the  energy  is  transmitted  over  high-tension  trans- 
mission lines  to  piunping  stations  scattered  over  the  territory 
to  be  irrigated.  Besides  these,  there  are  numerous  other  projects 
where  hydro-electric  power  is  similarly  used  for  irrigating  the  land. 

Mining.  The  advantage  of  using  electric  power  for  mining 
operations  is  now  fully  recognized,  almost  all  new  mines  being 
equipped  for  electric  drive,  and  a  very  large  number  of  old  ones 
changing  over  to  this  system.  Not  only  does  this  reduce  the  cost 
of  working,  but  it  also  offers  a  much  safer  and  more  reUable  oper- 
ation. The  economy  of  electric-p)ower  distribution  to  the  various 
points  in  a  mine  surpasses  all  other  methods.  The  electric 
system  eliminates  long  and  expensive  steam  and  air  lines,  with 
which  the  danger  of  breakdown  and  the  difficulty  of  keeping  up 
the  necessary  working  pressure  increase  with  every  extension  to 
the  service.  Electric  distribution,  on  the  other  hand,  is  most 
simple  and  flexible.  Very  large  districts  can  be  efliciently  sup- 
plied and  additions  or  alterations  can  at  all  times  be  made  without 
the  least  difliculty. 

A  most  eflScient  appUcation  of  motors  to  the  many  forms  of 
mining  machines  is  readily  accomplished.  They  can  be  direct 
connected,  or  geared  to  the  driving  shafts,  thus  reducing  the  fric- 
tion losses  and  repair  charges  to  a  considerable  extent,  while, 
on  the  other  hand,  the  cost  of  belting  and  coimtershafts  is  entirely 
eliminated.  Individual  motors  can  be  substituted  for  driving 
conveyors,  scrapers  and  other  machinery  in  breakers  and  tipples, 
which  formerly  were  equipped  for  group  operation  by  means  of 
inefficient  engines.  In  motor-driven  breakers,  the  saving  in  belt- 
ing alone  is  considerable. 

Operation  with  the  electric  system  is  very  simple,  and  results 
in  a  materially  increased  output  of  a  mine.  Perfect  control  is  at 
all  times  possible.  Simple,  automatic,  safety  devices  can  be 
installed,  and  indicating  or  recording  meters  can  be  provided  in 
the  several  circuits  as  desired,  and  the  performance  of  every 
individual  machine  ascertained.  This  is  a  very  important  point, 
as  it  is  possible  to  maintain  the  machinery  in  the  best  possible 


Digitized  by 


Google 


34  GENERAL  INTRODUCTION 

condition.  Any  excess  consumption  of  power  can  at  once  be 
detected  and  the  defect  remedied,  while  also  an  accurate  record 
can  be  kept  of  the  cost  of  the  different  operations. 

Power  may  be  purchased  from  nearby  existing  hydro-electric 
transmission  companies,  or  available  water  powers  may  be  devel- 
oped and  the  energy  transmitted  to  the  mines.  That  water 
powers  may,  in  some  instances,  compete  with  very  cheap  steam 
power  is  also  illustrated  by  the  system  of  the  Appalachian  Power 
Company,  which  furnishes  a  considerable  amoimt  of  power  from 
its  hydro-electric  plants  on  the  New  River  in  Virginia  to  the 
Pocahontas  coal  fields,  a  distance  of  about  50  miles. 

Electro-chemical  Industries.  The  industrial  processes  foimded 
upon  electro-chemistry  have  a  large  and  important  part  in  the 
manufacture  of  a  very  wide  range  of  commercial  products,  such  as 
fertilizers,  explosives,  paper,  wood  pulp  and  numerous  electro- 
chemicals  among  which  may  be  mentioned:  aluminimi,  carbo- 
rundum, alimdum,  silicon,  graphite,  calcium  carbide,  cyanamid, 
ferroHsilicon,  ferro-chromium,  ferro-manganese,  caustic  soda, 
sodium,  chlorine,  chlorate,  chloroform,  carbon  tetrachloride,  etc. 

Table  XIII,  taken  from  the  Report  of  the  Bureau  of  Census, 
gives  comparative  statistics  for  1914  and  1909  of  the  production 
of  chemicals  and  allied  commodities  by  means  of  electricity. 
I  The  question  of  cheap  water  power  is  vital  in  connection  with 
electro-chemical  industries,  but,  on  the  other  hand,  the  location 
of  raw  materials  and  the  transportation  facilities  of  the  product 
to  the  market  centers  is  also  of  the  greatest  importance,  and  this 
latter  point  has  to  a  great  extent  been  detrimental  to  a  much 
greater  development  of  our  western  water  powers  for  electro- 
chemical products.  Niagara  Falls,  on  the  other  hand,  forms  an 
ideal  example  of  what  cheap  water  power  has  done  for  this  industry. 
At  this  point  are  now  situated  the  greatest  electro-chemical  indus- 
tries in  the  world,  not  one  of  which  was  in  existence  when  the 
Niagara  Falls  Power  Company  began  to  take  water  from  the 
Niagara  River  to  generate  electricity.  In  the  treaty  of  1910 
between  the  United  States  and  Great  Britain  it  was  stipulated  that 
the  volume  to  be  diverted  on  the  American  side  should  be  limited 
to  20,000  cubic  feet  per  second  and  on  the  Canadian  side  to  36,000 
cubic  feet  per  second.  The  volume  of  the  water  that  can  be 
diverted  at  the  Falls  is  thus  limited  to  56,000  cubic  feet  per  second 
until  such  time  as  the  two  Governments  may  determine  to  increase 


Digitized  by 


Google 


COMMERCIAL  OPPORTUNITIES 


35 


TABLE  XIII 
Chemicals  Made  by  Electricitt 


1914 


1909 


Number  of  establishments 

Products — 

Total  value 

Chlorates: 

Number  of  establishments 

Tons 

Value 

Hypochlorites: 

Number  of  establishments 

Tons 

Value 

Caustic  soda,  caustic  potash,  and  lye: 

Number  of  establishments 

Tons 

Value 

Ferro  and  other  alloys: 

Number  of  establishments 

Value 

Oxygen  and  hydrogen: 

Number  of  establishments 

Value 

All  other,  names  in  order  of  value — aluminum, 
calcium  carbide,  abrasives,  electrodes, 
sodium  and  sodium  peroxide,  phosphorus, 
silicon  chlorine,  carbon  bisulphide,  and 
muriatic  acid: 

Number  of  establishments 

Value 


36 

129,661,649 

5 

8,304 

11,131,316 

4 

73,197 

$1,714,837 

51 
48,663 
12,309,511 

7 
12,859,482 

5 
$68,441 


34 

$18,451,461 

5 

6,785 

$904,550 

5 

45,970 

$1,506,831 


17 
$21,578,062 


$16,040,080 


the  amount.  This  is  approximately  25  per  cent  of  the  total  flow^f 
the  river,  as  computed  by  Government  engineers.  It  is,  however, 
not  sufficient  to  allow  for  any  further  expansion  of  these  electro- 
chemical industries,  some  of  the  largest  and  most  important  of 
which  are  now  compelled  to  go  to  Norway  or  other  fields  where 
abimdant  water  power  can  be  had  cheaply;  this,  in  spite  of  the 
fact  that  at  least  one  million  horse-power  additionally  could  be 
developed  at  Niagara  Falls  without  seriously  interfering  with  the 
scenic  beauty  of  the  Falls.  On  the  other  hand,  the  shortage  of 
power  was  responsible  for  the  recent  installation  of  the  mammoth 
steam  plant  almost  within  the  shadow  of  the  Falls. 


Digitized  by 


Google 


36 


GENERAL  INTRODUCTION 


Another  great  need  for  the  unmediate  development  of  addi- 
tional water  power  is  the  imperative  necessity  of  increasing  our 
nitrate  supply  and  making  it  independent  of  foreign  deposits. 
Fixed  nitrogen  is  the  most  important  constituent  of  plant  food  and 
absolutely  indispensable  in  the  manufacture  of  explosives.  Europe 


Fig.  9.— Rjukan  II  Power  and  Furnace  House  for  Nitrogen  Fixation  in  Nor- 
way.   Capacity  120,000  horse-power. 

uses  per  acre  (rf  cultivated  land  200  pounds  of  fertilizer;  the 
United  States  28  pounds.  Germany,  in  twenty  years,  by  the  use 
of  fertilizers,  has  increased  the  average  yield  of  all  crops  grown 
three  and  one-half  times  as  much  per  acre  as  America,  the  yield 
per  acre  in  bushels  for  various  crops  being  as  follows: 


Digitized  by 


Google 


COMMERCIAL  OPPORTUNITIES 


.37 


TABLE  XIV 
Crop  Yields 


Wheat. 


Oat3. 


Barley. 


Rye. 


Potatoes. 


Europe 

United  States. 


32 
15 


47 
29 


38 
25 


30 
16 


158 
96 


As  a  measure  of  preparedness  our  reserve  stock  of  nitrates  is 
insif^nificant  and  our  nation  would  be  powerless  if  our  navy  were 
not  strong  enough  to  protect  our  import  from  Chile.  Fortunately 
enough,  nitrates  can  readily  be  extracted  from  the  atmosphere  and 
fixed  as  a  compound  by  the  utilization  of  electric  energy.  The 
possibilities  of  this  have  never  been  more  clearly  demonstrated 
than  during  the  European  war,  when  Germany's  entire  supply  was 
obtained  in  this  way.  In  Norway,  with  its  cheap  wat<er  powers, 
the  industry  has  long  been  established,  about  350,000  horse- 
power being  at  present  utilized  by  one  company  alone  for  the 
fixation  of  nitrogen  by  the  arc  process.  Fig.  9  shows  one  of  its 
power-houses  and  factories,  with  a  capacity  of  120,000  horse- 
power. 

The  power  requirements  vary  widely  for  the  different  electro- 
chemical products  as  seen  from  Table  XV,  and  in  many  instances 
it  is  a  large  item  in  the  cost  sheet  of  the  product. 


TABLE  XV 

Tower  Consumption  op  Electro-chemical  Frocessbs  Per  Ton 
OP  2000  Pounds 

Kw.-hra. 

Refining  of  lead 120 

Refining  of  copper 300 

Refining  of  steel 600-1,000 

Refining  of  nickel 3,000 

Refining  of  zinc 3,500 

Reduction  of  calcium  carbide 4,000 

Reduction  of  ferro-alloys 4,000-12,000 

Reduction  of  abrasives 7,500 

Reduction  of  aluminum 30,000 

Pig  iron  from  ore 2,000-3,000 

Brass  melting 220-280 

Nitrogen  (fixed) 15,000-60,000 


Digitized  by 


Google 


38 


GENERAL  INTRODUCTION 


Railroad  Electrification.  Hydro-electric  power  will  undoubt- 
edly play  an  important  part  in  connection  with  future  railroad 
electrifications,  especially  in  the  western  mountainous  States. 
440  miles  of  the  main  line  of  the  Chicago,  Milwaukee  &  St. 
Paul  Railroad  haVe  now  been  equipped  for  operation  by  elec- 
tricity, power  being  supplied'  by  nearby  hydro-electric  develop- 
ments, Fig.  10.    In  view  of  the  economical  success  of  this  elec- 


FiG.  10. — Electric  Trains  at  the  Entrance  to  Silver  Bow  Canyon  on  the 
Chicago,  Milwaukee  &  St.  Paul  and  the  Butte,  Anaconda  &  Pacific 
Railways. 

trification,  it  is  almost  certain  that  within  the  next  ten  years  a 
majority  of  the  railroads  operating  through  the  mountainous 
country  of  the  Far  West,  where  hydro-electric  power  can  be 
developed  cheaply,  will  adopt  electricity  as  a  motive  power.  It 
is  estimated  that  five  million  horse-power  would  be  required  to 
electrify  the  50,000  miles  of  railroad  in  the  western  States,  or 
one-ninth  of  the  total  hydro-electric  power  possible  to  develop  in 
the  territory  traversed  by  these  railroads. 


Digitized  by 


Google 


CHAPTER  n 

HYDROLOGY 

1.  PROPERTIES  OF  WATER 

Weight  The  weight  and  specific  gravity  of  water  vary 
somewhat,  depending  on  its  temperature  and  on  the  variou 
impurities  which  it  contains  in  solution  or  carries  in  suspension. 
For  pure  water  the  weight  may  be  considered  practically  constant, 
as  the  maximum  variation  has  been  found  to  be  so  inconsiderable, 
being  only  about  0.05  of  1  per  cent.  Its  weight  is  now  generally 
assumed  to  be  62.355  pounds  per  cubic  foot  at  a  temperature  of 
62°  F.,  although  authorities  differ  somewhat  about  the  exact 
figure.  Water  of  lakes  and  rivers  will,  under  •ordinary  circum- 
stances vary  between  62.3  and  62.5  pounds,  depending  on  the 
impurities.  Table  XVI,  however,  shows  that  a  considerable 
variation  may  be  expected  under  unusual  conditions,  as  for  exam- 
ple, the  Great  Salt  Lake,  where  the  water,  due  to  the  large  amount 
which  it  contains,  weighs  nearly  73  pounds  per  cubic  foot. 


TABLE  XVI 
Weights  and  Specific  Gravity  op  Water 


Weight  per 

Cubic  Foot 

62*»  Fah. 


Specific  Gravity. 


Pure  water 

Atlantic  Ocean 

Lake  Michigan 

Great  Salt  Lake,  Utah 

Mono  Lake,  Cal 

Mississippi  River 

Delaware  River 


62.355 
64.043 
62.336 
72.925 
65.134 
62.333 
62.333 


1.00000 

1.0275 

1.0011 

1.17 

1.045 

1.00006 

1.00006 


While  sometimes  invisible,  all  natural  waters  always  contain 
in  solution  more  or  less  of  the  substances  which  they  have  come  in 

39 


Digitized  by 


Google 


40  HYDROLOGY 

contact  with  in  their  course.  These  substances  may  be  either 
solidSy  liquids  or  gases.  The  quantity  of  a  solid  which  may  be 
dissolved  by  a  liquid  is  fixed  and  limited,  and  is  always  the  same 
for  the  same  temperature,  the  solubility  however,  generally 
increasing  with  the  temperature.  The  same  quantity  of  gas  will 
also  be  dissolved  by  a  liquid  if  the  temperature  and  the  pressure 
remains  the  same,  the  volume  of  gas  dissolved  being  proportional 
to  the  atmospheric  pressure.  Rain  water  always  contains  in 
solution  a  certain  amoimt  of  the  natural  gases  of  the  atmosphere. 
These  are,  however,  not  dissolved  in  proportion  to  their  occur- 
rence in  the  atmosphere,  but  more  nearly  to  the  solubility  of  the 
gases.  Deep  waters  and  waters  of  springs  which  have  been  under 
pressure  carry  in  solution  larger  percentages  of  carbonic  acid 
gas  than  natural  waters. 

There  is  a  distinct  difference  between  substances  in  solution 
and  in  suspension.  When  in  suspension  the  substance  still  retains 
its  ph3rsical  identity,  although  it  may  be  held  in  an  exceedingly 
finely  divided  state  and  thus  be  carried  in  suspension  for  indefinite 
periods.  When  <he  water  is  at  rest  the  heavier  suspended  par- 
ticles are  soon  deposited. 

Volume.  For  all  practical  purposes  water  may  be  considered 
non-compressible.  The  coefficient  of  compressibility  ranges  from 
0.00004  to  0.00005  per  atmosphere  at  ordinary  temperature  the 
coefficient  decreasing  as  the  temperature  increases. 

Table  XVII  gives  the  relative  volume  and  weight  of  pure 
water  at  various  temperatures,  as  compared  with  its  volume  at 
39.2°  F. 

Critical  Temperatures.  There  are  four  temperatures  of  water 
which  are  often  used  in  physical  calculations  and  which  should  be 
kept  in  mind,  viz.:  32®  F.  or  0®  C,  at  which  pure  water  freezes  at 
one  atmosphere  pressure  (sea  level).  The  weight  of  ice  is  57.5 
pounds  per  cubic  foot,  and  when  floating  in  pure  water  92  per  cent 
of  its  mass  is  submerged,  while  in  sea  water  about  89  per  cent. 

39.2**  F.  or  4**  C,  which  is  the  approximate  point  of  maximum 
density  of  pure  water. 

62"*  F.  or  le.e?"*  C,  which  is  the  British  Standard  tempera- 
ture, and  which  is  used  as  a  basis  in  calculating  the  specific  gravity 
of  bodies  in  England  and  United  States. 

212®  F.  or  100°  C.  is  the  boiUng  point  of  piure  water  at  atmos- 
pheric pressure. 


Digitized  by 


Google 


PROPERTIES  OF  WATER  41 

TABLE  XVII 

VOLUMB  AND  WbIQHT  OF  PURB  WaTER  AT  VaBIOUS  TeMFESATUBBB 

(From  Marks  and  Davis) 


Temperature 
in  deg.  Fah. 

Relative  Volume. 

Weight  per  Cu.ft. 
in  Pounds. 

32 

1.000176 

62.42 

39.2 

1.000000 

62.43 

40 

1.000004 

62.43 

50 

1.00027 

62.42 

60 

1.00096 

62.37 

70 

1.00201 

62.30 

80 

1.00338 

62.22 

90 

1.00504 

62.11 

100 

1.00698 

62.00 

110 

1.00915 

61.86 

120 

1.01157 

61.71 

130 

1.01420 

61.55 

140 

1.01705 

61.38 

150 

1.02011 

61.20 

160 

1.02337 

61.00 

170 

1.02682 

60.80 

180 

1.03047 

60.58 

190 

1.03431 

60.36 

200 

1.03835 

60.12 

210 

1.04256 

59.88 

212 

1.04343 

59.83 

Latent  Heat  This  is  the  heat  which  apparently  disappears 
in  producing  some  change  in  the  conditions  of  a  body  without 
increasing  its  temperature.  To  transform  ice  water  and  vapor 
or  steam  from  one  state  to  the  other,  it  is  only  necessary  to  supply 
a  certain  quantity  of  heat  energy,  —460®  F.  being  the  absolute 
zero  of  temperature. 

Thus  in  melting  1  poimd  of  ice  into  water  at  32®  F.,  about  142 
heat-units  are  absorbed  and  become  latent,  while  in  freezing  one 
pound  of  water  into  ice  a  like  quantity  of  heat  is  given  out  to  the 
surrounding  medium. 

Latent  heat  is  not  lost,  but  reappears  whenever  the  substances 
pass  through  a  reverse  cycle,  from  a  gaseous  to  a  liquid,  or  from  a 
liquid  to  a  soUd  state.  It  may,  therefore,  be  considered  as  the 
heat  which  apparently  disappears,  or  is  lost  to  the  thermometric 


Digitized  by 


Google 


42 


HYDROLOGY 


measurement,  when  the  molecular  constitution  of  a  body  is  being 
changed. 

Specific  Heat  The  specific  heat  of  water  is  greater  than  all 
known  substances  with  the  exception  of  bromine  and  hydrogen, 
and  it  is  the  basis  for  measurement  of  the  capacity  of  heat  absorp- 
tion of  all  other  substances.  Its  value  varies  with  the  tempera- 
ture of  the  water,  being  lowest  near  40®  C,  after  which  it  increases 
up  to  and  beyond  the  boiling-point.  The  generally  accepted 
values  as  determined  by  Peabody  are  given  in  Table  XVIII. 

TABLE  XVIII 
Specific  Heat  of  Water  at  Various  Temperatures 


Temperaturb. 

Specific  Heat. 

1 

Deg.  C. 

Deg.  F. 

0 

32 

1.0094 

5 

41 

1.0053 

10 

50 

1.0023 

15 

59 

1.0003 

16.11 

61 

1.000 

20 

68 

0.9990 

25 

77 

0.9981 

30 

86 

0.9976 

35 

95 

0.9974 

40 

104 

0.9974 

45 

113 

0.9976 

50 

122 

0.9980 

55 

131 

0.9985 

60 

140 

0.9994 

65 

149 

1.0004 

70 

158 

1.0015 

75 

167 

1.0028 

80 

176 

1.0042 

85 

185 

1.0056 

90 

194 

1.0071 

95 

203 

1.0086 

100 

212 

1.0101 

Effect  of  Atmospheric  Pressure.  At  sea  level  the  average 
atmospheric  pressure  is  14.72  pounds  per  square  inch,  but  it 
decreases  as  the  height  above  sea  level  increases.  With  water 
weighing  62.4  pounds  per  cubic  foot,  the  weight  of  a  column  having 


Digitized  by 


Google 


PROPERTIES  OF  WATER 


43 


a  cross-section  of  1  square  inch  and  a  height  of  1  foot  will  equal 

62  4  .  • 

-T^  or  0.43  poimd,  so  that  at  sea  level  water  will  rise  to  an  aver- 
144 

14  72 
age  height  of     '      or  33.9  feet  in  vacuum. 

The  barometric  pressure  in  inches  is  equal  to  the  pressiure  per 
square  inch  divided  by  0.4908. 

In  Table  XIX  are  given  the  relations  of  altitude  to  barometer 
and  atmospheric  pressure. 

TABLE  XIX 
Relations  of  Elevation  to  Barometer  and  Atmospheric  Pressure 


Height  aboye 

Average  Height 
Barometer  in 

Average  Pressure 
in  Pounds  per 

Average  Height  to 
which  Water  WUl 

Sea  Level. 

Inches  of  Mercury. 

Square  Inch. 

Rise  in  an  Ex- 
hausted Tube. 

0 

30.00 

14.72 

33.96 

100 

29.89 

14.67 

33.84 

200 

29.78  . 

14.62 

33.72 

300 

29.68 

14.57 

33.60 

400 

29.57 

14.51 

33.48 

£00 

29.47 

14.46 

33.35 

600 

29.36 

14.41 

33.23 

700 

29.25 

14.36 

33.11 

800 

29.15 

14.30 

32.99 

900 

29.04 

14.25 

32.87 

1,000 

28.94 

14.20 

32.76 

1,250 

28.67 

14.07 

32.47 

1,500 

28.42 

13.95 

32.19 

2,000 

27.92 

13.70 

31.61 

2,500 

,27.40 

13.45 

31.04 

3,000 

26.93 

13.21 

30.49 

3,500 

26.43 

12.98 

29.94 

4,000 

25.98 

12.74 

29.41 

4,500 

25.51 

12.51 

28.89 

5,000 

25.06 

12.29 

28.37 

6,000 

24.18 

11.85 

27.37 

7,000 

23.32 

11.43 

*   26.40 

8,000 

22.50 

11.04 

25.47 

9,000 

21.70 

10.65 

24.57 

10,000 

20.93 

10.28 

23.70 

Measurements.    Conversion  Table  XX  gives  the  most  com- 
mon units  in  which  water  is  measured. 


Digitized  by 


Google 


44 


HYDROLOGY 


TABLE  XX 
Equiyalent  Mbasubbs  akd  Weights  of  Water  at  4*  Centiqbade,  39.2'' 

Fahrenheit 


U.  B. 
GaUoDB. 

Liters. 

Cubic 
Metera. 

PoundB. 

Cubic 
Feet. 

Cubic 
Inches. 

1. 

3.7853 

.0037853 

8.34112 

.13368 

231. 

1.20017 

4.54303 

.004543 

10.0108 

.160439 

277.274 

.264179 

1. 

.001 

2.20355 

.035316 

61.0254 

264.179 

1000. 

1. 

2203.55 

35.31563 

61025.4 

.119888 

.453813 

.0004538 

1. 

.0160266 

27.694 

7.48055 

28.3161 

.0283161 

62.3961 

1. 

1728. 

.004329 

.0163866 

.0000164 

.0361089 

.0005787 

1. 

.0408 

.1544306 

.0001544 

.340008 

.005454 

9.4224 

2.  RAINFALL 

Source  of  Water  Supply.  The  ultimate  source  of  our  water 
supply  is  the  precipitation  in  the  form  of  rain  or  snow  which 
reaches  the  earth.  For  the  United  States  the  chief  source  of  this 
is  the  evaporation  from  the  Pacific  Ocean  which  by  westerly  winds 
is  carried  eastward  in  diminishing  quantities.  In  the  Mississippi 
Valley  the  small  supply  of  moisture  still  remaining  is  augmented 
by  a  generous  contribution  from  the  Gulf  of  Mexico,  from  where 
it  is  carried  inland  by  southerly  and  southwesterly  winds.  East 
of  the  Appalachian  Moimtains  the  precipitation  is  mainly  derived 
from  the  Atlantic  Ocean. 

Rain  is  formed  whenever  the  air  is  cooled  below  the  point  of 
satm-ation.  This  cooling  may  be  caused  by  the  air  currents 
being  forced  upward,  as  when  they  strike  mountain  ranges,  or 
they  may  be  intermingled  with  other  colder  air  currents,  or  come 
into  contact  with  a  cold  land. 

Variation  in  Rainfall.  The  rainfall  varies  greatly  in  different 
parts  of  the  country  and  is  governed  quite  largely  by  the  geo- 
graphic or  topographic  relations.  It  is  usually  given  in  either 
total  inches  of  rain  per  year  or  per  month,  while  the  daily  maps  of 
the  Weather  Bureau  show  the  variations  from  day  to  day.  A 
map  of  the  United  States  giving  the  average  annual  rainfall  in 
inches  for  the  different  sections  is  shown  in  Fig.  11,  the  mean 
annual  precipitation  for  the  whole  coimtry  being  29.4  inches. 
Table  XXI  also  gives  some  typical  values  of  rainfall  in  different 
parts  of  the  coimtry. 


Digitized  by 


Google 


RAINFALL 


45 


I 


I 

i 


I 

s 

OQ 

1 
•a 

S- 


2 


Digitized  by 


Google 


46 


HYDROLOGY 


TABLE  XXI 
Typical  Averages  of  Rainfall 


Inchee  Annual 
RalnfaU. 


Approximate 

Mean  Annual 

Run-off.  Inchei. 


North  Atlantic  States. 

Gulf  States 

Lake  Region 

Mississippi  Valley .... 

Mountain  Region 

Plains 

Pacific  Coast,  north. . . 
Pacific  Coast,  south. . . 


40-60 
50-60 
30-40 
30-60 
10-20 
0-10 
40-60 
10-^ 


Over  20 

Over  20 

10-20 

10-20 

2-  5 

0-  2 

10— Over  20 

2-10 


State. 


Spring. 

Summer. 

Autumn. 

winter. 

11.6 

11.4 

11.9 

11.7 

12.4 

15.6 

10.7 

12.7 

7.9 

9.7 

9.2 

7.0 

10.0 

12.0 

9.1 

6.5 

4.2 

5.5 

2.8 

2.3 

2.3 

0.8 

1.3 

3.2 

9.8 

2.7 

10.5 

21.0 

6.2 

0.3 

3.5 

11.9 

Total. 


Massachusetts . 

Georgia 

Michigan 

Missouri 

Colorado 

Nevada 

Oregon 

California.  .  . . . 


46.6 
51.4 
33.8 
38.0 
14.8 
7.6 
44.0 
21.9 


The  annual  as  well  as  the  monthly  rainfall  varies  irregularly 
from  year  to  year,  and  the  amount  of  these  variations  is  greater 
in  some  localities  than  other.  While  they  may  remain  within 
certain  Umits,  the  totals  are  made  up  of  still  greater  variations  in 
individual  storms. 

The  rainfalls  to  be  considered  for  practical  purposes  are  the 
average  monthly  and  the  montiily  of  the  driest  year,  both  of 
which  affect  the  supply,  while  a  knowledge  of  the  maximum  rainfall 
is  essential  for  determining  the  discharge. 

Rainfall  Record.  The  United  States  Weather  Bureau  main- 
tains several  thousand  stations  for  recording  the  rainfall  of  the 
coimtry  and  the  number  of  points  at  which  such  observations 
is  increasing  from  year  to  year.  There  are  some  places  where 
observations  have  extended  for  over  fifty  j-ears  and  suflScient 
information  can  therefore  usually  be  obtained  from  the  bulletins 
of  the  Weather  Bureau.    Where  small  watersheds  are  under  in- 


Digitized  by 


Google 


DISPOSAL  OF  RAINFALL 


vestigation  it  may,  however,  often 
be  found  necessary  to  make  indi- 
vidual rainfall  measurements. 

Diagrams,  Figs.  12  and  13,  repre- 
sent a  75-year  rainfall  record  at 
St.  Paul,  as  reported  by  the  Minne- 
sota Board  of  Water  Commis- 
sioners. 

3.  DISPOSAL  OF  RAINFALL 

Of  the  rainfall  a  portion  evap- 
orates, a  portion  enters  the  soil  and 
is  either  absorbed  by  plant  growth 
or  by  ground  flow  reaches  the  rivers 
or  lakes,  while  the  third  portion 
finds  its  way  into  streams  as  surface 
flow  or  run-off. 

Evaporation.  Of  the  tremendous 
losses  due  to  evaporation  from  the 
ground  surface  comparatively  little 
is  known.  It  is  impossible  to  arrive 
at  such  losses  by  taking  the  differ- 
ence between  rainfall  and  run-off, 
as  in  this  there  would  also  be  in- 
cluded the  losses  due  to  absorption 
by  the  soil  and  by  vegetation,  and 
again  the  rate  of  run-off  does  not 
altogether  depend  upon  the  rainfall. 

The  rate  of  evaporation  or  the 
proportion  of  the  rainfall  to  the  air 
varies  greatly  under  different  con- 
ditions and  is  affected  by  atmos- 
pheric conditions  as  well  as  by  the 
character  of  the  soil.  The  capacity 
of  the  atmosphere  to  take  up  and 
dissipate  the  moisture  depends  in 
turn  on  the  temperature,  the  wind, 
ahd  how  saturated  it  already  is. 
Wind  increases  the  evaporation  to 
a  great  extent,  especially  from  ex- 


FiG.  12. — Annual  Precipitation  at 
St.  Paul,  Minn.,  1837-1912. 


Digitized  by  CjOOQ IC 


48 


HYDROLOGY 


posed  water  surfaces,  as  the  saturated  air  in  contact  with  such 
surfaces  is  rapidly  removed  and  continually  replaced  by  fresh. 
In  cool  climates  with  light  breezes  the  evaporation  is  considerably 
lower  than  in  warm  climates  with  strong  winds. 

The  natiu*e  of  the  earth's  surface,  on  the  other  hand,  deter- 
mines the  rate  at  which  moisture  is  supplied.  Thus,  a  ver}* 
large  evaporation  takes  place  from  exposed  water  surfaces  such 
as  lakes,  swamp  lands,  etc.,  and  the  amount  may,  in  certain 
instances,  equal  the  actual  rainfall  itself.     They  tend,  however, 


8 


n  u  h  II  11  i§  i§  II  II  li  11  t 


n 

n     n 

% 

"1 

_- 

^       ^ 

^ 

^     > 

^ 

^       ^ 

V                    ^                    ^                     5 

Jan.    Feb.   Mu.  Ajydl   May   Jane  July  Aug.    Sept.    Oct    Not.    Dec 

FiQ.  13. — Monthly  Variation  in  Precipitation  at  St.  Paul.  Minn.     From 
Records  1837-1912. 

as  a  storage  of  flood  waters  and  add,  therefore,  materially  to  the 
regulation  of  the  stream  flow. 

The  depth  to  the  water  in  the  soil  and  its  capillary  action  in 
bringing  the  water  to  the  surface  also  naturally  affect  the  evapo- 
ration. A  light  rain  falling  on  an  impervious  rock  surface  may 
simply  wet  the  surface  and  quickly  disappear  as  vapor,  while 
saturated  surface  layers  of  the  soil,  such  as  after  heavy  rains,  will 
also  cause  considerable  evaporation. 

A  large  amount  of  water  is  necessarily  taken  up  by  the  vege- 
tation and  evaporated,  while  the  effects  of  forests  are  to  greatly 
reduce  the  evaporation  as  compared  to  open  fields. 

A  more  complete  study  has  been  made  of  the  evaporation  from 
the  water  surface  of  lakes  and  rivers,  the  greatest  use  of  such 
studies  being  in  the  investigation  of  storage  and  the  losses  which 


Digitized  by 


Google 


DISPOSAL  OF  RAINFALL  49 

are  likely  to  occur  on  such  reservoirs  through  evaporation.  That 
the  losses  on  lake  areas  are  very  great,  and  often  of  greater  extent 
than  precipitation,  is  well  known. 

The  map  in  Fig.  14  shows  the  mean  average  evaporation  in  the 
United  States  from  open  waters.  It  is  compiled  from  observa- 
tions of  the  United  States  Weather  Bureau  in  1887  and  1888.^ 

Absorption.  A  considerable  part  of  the  rain  which  falls  on 
the  earth  is  absorbed  by  the  groimd.  >  The  amoimt  varies,  how- 
ever, greatly,  depending  on  the  rate  of  precipitation,  texture  of 
soil,  slope  of  drainage  surface,  temperature  and  vegetation. 

A  light  shower  will  usually  be  quickly  evaporated,  while  a 
heavier  rain  may  be  absorbed,  and  if  lasting  for  some  time  there 
will  be  an  excess  amount  of  water  which  will  run  off  to  the  nearby 
stream.  On  the  other  hand,  less  may  be  absorbed  during  a  heavy 
rain  than  during  a  light,  gentle  rain,  because  each  type  of  soil  has 
a  certain  rate  of  absorption  due  to  its  porosity,  and  if  the  water  is 
supplied  more  rapidly  than  it  can  be  taken  up,  the  excess  runs  off. 
A  deep,  porous,  sandy  soil  naturally  will  absorb  and  hold  water 
more  than  a  compact,  shallow  one,  such  as  a  clayey  soil. 

If  the  slope  of  the  watershed  is  very  steep,  the  water  may 
drain  off  before  any  can  be  absorbed  by  the  soil,  and  if  the  slopes 
are  rocky  practically  no  water  is  absorbed. 

Temperature  necessarily  also  affects  absorption.  A  high  tem- 
perature increases  it  while  the  opposite  is  the  case  at  low  tem- 
peratures as  when  the  ground  is  frozen. 

On  slopes,  vegetation  and  forest  are  of  the  greatest  importance 
in  that  they  retard  part  of  the  drainage  water  during  heavy  rains, 
which  gives  the  soil  time  to  absorb  the  same.  They  are,  there- 
fore, of  great  value  in  reducing  the  intensity  of  floods  after  severe 
storms.  The  absorbed  water  seeps  into  the  ground,  which  it  sat- 
urates, and  some  of  it  percolates  still  further  into  the  pores  and 
fissures  and  trickles  slowly  toward  the  stream. 

These  groimd  waters  have  a  most  important  bearing  on  the 
stream  flow.  Areas  of  Uttle  or  no  underground  flow  are  subject 
to  violent  floods  and  extreme  droughts,  while  areas  with  a  large 
proportion  of  underground  storage  are  comparatively  free  from 
floods.  The  greater  part  of  the  low-water  of  streams  having  no 
lakes  or  swamps  in  their  watershed  is  also  supplied  by  this  under- 
ground flow. 

1  Monthly  Weather  Review,  September,  1888. 


Digitized  by 


Google 


50 


HYDROLOGY 


i=-^>^ic — ^t^^^^i^^-  ■■■■""■        'r-- 1^^^  ^-1 


^/"^^ 


-.■■x: 


n      ■  /■■  ' .h-  - 


d 
X 

T3 


5. 

> 


C 

a 
< 

G 


C 
X 

o. 

OS 
I 


o 


'iOi 


Lilll 


Digitized  by 


Google 


DISPOSAL  OF  RAINFALL  5.1 

A  detennination  of  the  exact  quantity  of  underground  waters 
is  a  very  diflScult  problem.  Numerous  papers  have  been  pre- 
pared on  the  subject  by  different  authors.  Water  Supply  and 
Irrigation  Paper  No.  163  of  the  United  States  Geological  Survey 
contains  a  bibliographic  review  and  index  of  underground-water 
literature  published  in  the  United  States  up  to  and  including  the 
year  1905. 

Run-off.  The  run-off  is  that  part  of  the  rainfall  which  drains 
off  the  surface  of  the  watershed  in  visible  streams.  It  is  that  part 
of  the  rainfall  which  remains  after  nature's  need  of  moistiu*e  has 
been  supplied  in  the  form  of  evaporation  and  absorption. 

The  close  relation  between  these  three  subdivisions  of  rain- 
fall has  been  referred  to  in  the  above,  and  it  follows  that  the  run- 
off is  affected,  both  directly  and  indirectly,  by  the  same  factors 
that  govern  the  rate  of  evaporation  and  absorption. 

It  is  often  important  to  know  the  relation  between  rainfall 
and  run-off,  as  this  may  in  many  instances  be  the  only  way  to 
ascertain  the  flow  of  a  stream.  Rainfall  observations  have  been 
made  for  many  years  and  it  may  be  possible  by  knowing  the  ratio 
between  run-off  and  rainfall  for  a  certain  <irainage  area,  to  apply 
this  value  to  a  watershed  in  another  place.  It  is,  of  course,  of 
the  greatest  importance  in  such  comparisons  that  the  areas  from 
which  the  deductions  are  made  must  be  of  similar  character. 
Also  that  they  are  of  approximately  the  same  size,  because  smaller 
drainage  areas  usually  have  a  wider  variation  between  maximiun 
and  minimum  run-off  than  large  ones.' 

It  is  apparent  that  there  can  be  no  constant  relation  between 
the  rainfall  and  the  run-off  for  the  whole  country,  although  in  this 
respect  the  ratio  for  the  Eastern  States  is  much  more  constant  than 
for  the  Western  States.  There  are  also  great  variations  in  the 
yearly  as  well  as  the  monthly  and  daily  run-off,  and  it  is  very 
diflScult  to  make  accurate  estimates  as  to  what  the  two  latter  may 
be  expected  to  be;  the  daily  being,  of  course,  almost  impossible  to 
foretell.  The  yearly  run-off,  however,  bears  a  more  nearly  uni- 
form ratio  to  the  rainfall,  so  that  with  a  good  knowledge  of  the 
presence  of  forests,  character  of  soil,  climate,  etc.,  a  fairly  accurate 
estimate  of  the  yearly  nm-off  may  be  made,  based  on  known 
values  under  similar  conditions. 

As  for  rainfall,  run-off  is  also  usually  expressed  in  inches,  and 
the  map  in  Fig.  15  shows  approximately  the  mean  annual  run-off 


Digitized  by 


Google 


52 


HYDROLOGY 


1 I 


Digitized  by 


Google 


STREAM-FLOW  53 

for  the  country.  By  comparing  this  map  with  that  of  rainfall 
in  Fig.  11,  a  fairly  good  idea  of  the  relation  between  rainfall  and 
nin-oflf  may  be  had.  Table  XXII  furthermore  gives  the  run-oflf 
for  various  watersheds  in  the  United  States. 

4.  STREAM-FLOW 

Definition  of  Tenns.  The  voliune  of  water  flowing  in  a  river 
IS  generally  defined  as  "  stream-flow  "  and  is  expressed  in  various 
terms  depending  upon  the  particular  class  of  work  for  which  it  is 
to  be  used.  The  term  used  in  the  reports  of  the  U.  S.  Geological 
Survey  are:  Second-feet,  second-feet  per  square  mile,  acre-feet 
and  depth  in  inches.  Of  these  the  first  two  represent  the  rate  of 
flow  only,  while  the  two  latter  represent  the  actual  quantity  of 
water.    They  are  defined  in  the  Survey  Reports  as  follows: 

"  Second-foot "  is  an  abbreviation  for  cubic  foot  per  second 
and  is  the  imit  for  the  rate  of  discharge  of  water  flowing  in  a  stream 
1  foot  wide,  1  foot  deep,  at  a  rate  of  1  foot  a  second.  It  is  gen- 
erally used  as  a  fundamental  imit  from  which  others  are  com- 
puted by  the  use  of  the  factors  given  in  the  following  table  of 
equivalents. 

"  Second-feet  per  square  mile  "  is  the  average  number  of  cubic 
feet  of  water  flowing  per  second  from  each  square  mile  of  area 
drained,  on  the  assumption  that  the  run-off  is  distributed  xmi- 
formly  both  as  regards  time  and  area. 

"  Depth  in  inches  "  is  the  depth  to  which  the  drainage  area 
would  be  covered  if  all  the  water  flowing  from  it  in  a  given  period 
were  conserved  and  imiformly  distributed  on  the  surface.  It  is 
used  for  comparing  run-off  with  rainfall,  which  is  usually  expressed 
in  depth  in  inches. 

An  "  acre-foot "  is  equivalent  to  43,560  cubic  feet,  and  is  the 
quantity  required  to  cover  an  acre  to  the  depth  of  1  foot.  The 
term  is  commonly  used  in  connection  with  storage  for  irrigation. 

The  direct  course  of  stream-flow  is  the  visible  run-off  from  the 
watershed  and  that  part  of  the  rain-fall  which  was  absorbed  by 
the  soil  and  which  slowly  finds  its  way  to  the  stream  bed  in  the 
form  of  an  underground  flow. 

Variation  in  Stream  Flow.  There  is  a  very  considerable  vari- 
ation in  the  flow  of  rivers  not  only  dm-ing  the  various  months  of 
the  year,  but  from  year  to  year  as  well,  and  the  variation  is  greater 


Digitized  by 


Google 


54 


HYDROLOGY 


TABLE  XXII 
Mean  Amkual  Run-off  for  Various  Watersheds  in  Unttbd  States  * 


River. 


Point  of  Measurement. 


Drainage 

Area 
Square 
MUes. 


Period. 


Run-off  in 
Depth  in 
Inches  on 
Draina^ 
Area. 


Kern 

San  Joaquin .  . 

Kings 

Sacramento. .  . 
UmatiUa 


Willamette.  . 


Boise. 
Green. 


Laramie 

Red 

Rio  Grande . . 

Animas 

South  Platte . 

Green. 

Logan. ., 

Carson 

Truckee 

Humboldt . .  . 
Colorado .... 


St.  Croix .  . . 
Menominee. 


liilnois . .  . 
Maumee . 
Scioto. .  . . 
Duck .  .  . . 


Tennessee .... 
Tombigbee.  .  . 
Black  Warrior. 

Alabama 

Savannah .... 

Catawba 

Tar 

Roanoke 

Potomac 

Oswego 

Delaware 

Susquehanna. . 

Hudson 

Mohawk 


Bakersfleld.  Cal. 
Herndon,  Cal.  .  . 
Sanger,  Cal .... 
Red  Bluff,  Cal. . . 
Umatilla,  Ore. .  . 


Albany,  Ore. 


Boise,  Idaho 

Green  River,  Wyo.  . 


Uva,  Wyo 

Grand  Forks,  N.  Dak. 
Rio  Grande,  N.  Mex. 


Durango,  Cal.. 
Denver.  Col. .  , 


Greenriver,  Utah . 

Logan,  Utah 

Empire.  Nev 

Vista.  Nev 


Orleans.  Nev. 


Yuma,  Ariz . 


St.  Croix  Falls   Wis. . 
Iron  Mountain,  Mich. 


Peoria,  111. 


Waterville,  Ohio. 
Columbus,  Ohio . 


Columbia,  Tenn. 


Chattanooga.  Tenn .  .  . 

Columbus,  Miss 

Cordova,  Ala 

Selma,  Ala 

Augusta,  Ga 

Rock  Hill,  S.  C 

Tarboro,  N.  C 

Randolph,  Va 

Pt.  of  Rocks,  Va 

Oswego.  N.  Y 

Port  Jarvis.  N.  Y 

Binghamton,  N.  Y. .  .  . 
Mechanicsville,  N.  Y. . 
Dunsbarh  Ferry,  N.  Y. 


2.340 
1,640 
1.740 
4,300 
2.130 

4.860 

2,610 
7.460 

3.180 

25.100 

14.000 

812 

3.840 

38.200 

218 

988 

1.620 

13.800 

225,000 

6,370 
2,420 

13,200 

6,110 

1,050 

1.260 

21,400 
4.440 
1,000 

15,400 
7,300 
2,990 
2,290 
3,080 
9,650 
5,000 
3.250 
2.400 
4.500 
3.440 


1896-1905 
1896-1901 
1897-1906 
1902-1906 
Nov.  1,  1900,  to 

Dec.  31.  1900 
Jan.  1,  1899,  to 

Dec.  31,  1908 
1895-1904 
May  1,  1896,  to 

Oct.  31,  1906 

May.  1895,  to 
Oct..  1903 

Sept.,  1902,  to 

Sept..  1908 
Jan.  1,  1896.  to 

Dec.  31,  1906 

July,  1895,  to 

Dec.,  1905 
Jan.  1,  1896,  to 

Nov.  30,  1906 

Jan..  1895.  to 
Dec.,  1908 
1896-1900 
1904-1906 

Nov..  1900.  to 
Dec..  1906 

Sept..  1899.  to 
Dec.,  1906 

Jan..  1897,  to 
Dec.,  1906 

Jan..  1902,  to 
Dec.,  1906 
1902-1904 

Sept.,  1902.  to 
Sept.,  1906 

Apr.  1,  1903,  to 

Jan.  30, 1906 

Dec.,  1898,  to 
Jan.,  1902 

1899  to 

July.  1906 

Nov.  1,  1904.  to 

Dec.  31,  1908 
1899-1908 
1905-1908 
1900-1908 
1900-1908 
1899-1908 
1895-1903 
1896-1900 
1901-1905 
1895-1906 
1897-1901 
1904-1908 
1901-1906 
1891-1900 
1898-1907 


4.36 
20.47 
20.38 
24.06 

3.94 


46.62 

16.60 

4.81 

1.10 

2.08 

1.46 

14.86 
1.44 
3.17 

21.18 

6.26 

0.18 

0.25 

1.16 

10.60 
18.92 

14.11 

13.61 

10.43 

18.87 

23.63 
15.48 
19.37 
24.01 
22.29 
25.21 
13.89 
18.86 
14.40 
11.60 
22.20 
28.88 
22.06 
23.28 


1  Prepared  by  Newell  and  Murphy  from  U.  S.  Geological  Survey  Records. 


Digitized  by 


Google 


STREAM-FLOW 


55 


in  some  regions  than  in  others.  In  Fig.  16  are  shown  some  t3rpical 
hydrograph  records  of  New  York  streams,  which  clearly  illustrate 
what  may  be  expected  in  the  way  of  variations  in  stream  flows. 


OSWEGO  RIVER 
BATTLE  ISLAND 
UNUSUALLY  STEADY  STREAM 


.1 


GENESEE  RIVER 

MOUNT  MORRIS 

UNUSUALLY  FLASHY  STREAM 


Fiu.  16. — ^Hydrographs  Showing  Natural  Fluctuations  of  Flow  of  New  York 
York  State  Streams. 

While  of  entirely  different  characteristics  it  can  be  seen  that 
there  are  certain  common  features  in  that  the  flows  are  heaviest 
during  the  spring  and  early  summer  and  lowest  in  autunm. 

This  irregularity  of  flow  is  a  very  important  factor  in  any  water- 
power  development  and  one  that  compels  the  reckoning  with  the 


Digitized  by 


Google 


56  HYDROLOGY 

minimiim  flow  and  the  possibilities  of  storage  for  increasing  the 
same  in  order  to  safely  develop  the  enterprise. 

Factors  A£fecting  Stream  Flow.  It  was  previously  shown  how 
absorption  and  the  natural  storage  of  underground  waters  had  a 
very  important  bearing  on  the  regularity  of  the  stream  flow,  these 
waters  being  the  main  source  of  supply  during  the  dry  season. 
It  was  also  shown  how  vegetation  and  heavy  forests  will  inter- 
pose an  appreciable  time  element  in  the  run-off.  In  addition 
there  are  several  other  factors  which  may  delay  the  same.  So 
for  example,  where  snow  and  ice  form  to  considerable  depths,  a 
large  part  of  the  precipitation  may  be  stored  for  weeks  or  months. 
On  the  other  hand,  the  effect  of  an  abnormally  dry  or  wet  season 
may  extend  beyond  a  single  year;  since  it  somewhat  affects  the 
conditions  of  the  ground  during  the  next  year,  so  that  a  succession 
of  dry  or  wet  years  may  disturb  the  expected  relations  of  run-off  to 
rainfall  producing  imexpected  drought  or  flood. 

Most  watersheds  have  some  natural  storage  features  tending 
to  equalize  the  stream-flow  as  compared  with  the  rainfall.  In  the 
northern  part  of  the  United  States  most  watersheds  have  distinct 
periods  in  the  water  year  as  distinguished  from  the  calendar  year. 
These  are  usually  classified  into  istoring,  growing  and  replenishing. 
Beginning  about  the  first  of  December  water  begins  to  accumulate 
in  the  form  of  snow,  ice,  or  in  the  soil,  and  for  months  there  is  an 
increasing  storage.  With  the  beginning  of  spring  the  storage 
period  terminates,  and  the  growing  period  begins,  during  which 
moisture  is  absorbed.  By  harvest  time  vegetation  has  ceased 
to  absorb  moisture  and  it  usually  tends  to  replenish  the  ground 
until  the  end  of  the  fall.  That  these  periods  have  great  effects  on 
run-off  can  readily  be  appreciated  and  how  great  the  effects  may 
be  can  well  be  judged  from  the  typical  figures  in  table  XXIII. 

The  curves  in  Fig.  17  indicate  graphically  the  approximate 
relations  for  this  area,  and  will  show  that  for  the  same  watershed 
the  percentage  run-off  increases  with  increasing  rainfall. 

Lakes,  ponds  and  swamps  are,  of  course,  of  great  value  in 
regulating  the  stream-flow,  and  very  frequently  broad  rivers  have 
storage  possibilities  not  readily  appreciated  at  first.  In  localities 
where  there  is  a  pronounced  dry  season  extending  over  several 
months'  time,  water-power  plants  have  been  built  in  which  it  is 
regularly  proposed  to  store  water  for  six  months  at  a  time,  thus 
enabling  the  average  daily  output  of  the  plant  to  be  increased 


Digitized  by 


Google 


STREAM-FLOW 


57 


TABLE  XXin 

Hudson  Riveb,  1888-1901 

Catchment  Area,  4500  Square  Miles 

Mean  Values 


Period. 

Rainfall 
in  Inches. 

Run-off 
in  Inches. 

Evaporation 
in  Inches. 

Per  Cent  Run- 
off to  Rainfall. 

Storage 

20.6 
12.7 
10.9 

44.2 

16.1 
3.5 
3.7 

23.3 

4.5 
9.2 
7.2 

20.9 

78  2 

Growingr .....,.,.. 

27  6 

Replenishing 

34.0 
62.7 

80 

/ 

/ 

/ 

/ 

/ 

i 

/ 

J 

/ 

1 

7 

J 

M 

r 

y 

^ 

y 

y 

'/ 

r 

y 

/ 

^ 

^ 

y 

10 

/ 

r 

€\ 

Z 

A 

y 

^ 

y 

y 

^ 

^r 

y 

^ 

y^ 

y^ 

y 

^- 

^ 

;^ 

^ 

», 

) 

^ 

•^ 

1 

0 

'A 

0 

80 

Inches  Precipitation 

FiQ.  17. — Curves  Showing  Mean  Rainfall  and  Run-off  on  Upper 
Hudson  River. 

several  fold.  This  occurs  usually  in  high-head  plants  where  the 
quantity  of  water  is  relatively  small  and  the  rough  character  of 
the  country  permits  the  construction  of  deep  reservoirs,  but  there 
are  some  low  head  plants  with  short  periods  of  low  water  where 
the  storage  of  some  important  tributary  stream  will,  at  reasonable 
expense,  greatly  increase  the  minimum  average  daily  output. 


Digitized  by 


Google 


58  HYDROLOGY 

The  diagrams  S  shown  in  Fig.  18,  represent  the  ideal  regula- 
tion of  the  Hudson  River,  and  was  based  on  a  proposed  extensive 
reservoir  system  and  the  stream-flows  for  the  years  1908-09. 
Other  stream-flow  records  would,  of  course,  modify  the  result, 
while,  on  the  other  hand,  such  ideal  flow  can  seldom  be  obtained 
at  a  cost  which  would  be  commercially  possible. 

From  the  above  it  can  readily  be  seen  that  usually  very  careful 
measiu^ments  of  stream-flow  extending  over  many  years'  time  are 
necessary  to  enable  good  .estimates  of  available  power  to  be  made, 
particularly  where  the  contemplated  development  has  no  storage 
facilities. 

Measurements  of  Stream-flow.  The  methods  by  which  the 
records  of  stream  discharge  are  made  differ  according  to  the  nature 
and  importance  of  the  work.  The  simplest  and  most  accurate 
method  for  a  small  stream  is  by  means  of  a  weir.  This  consists 
of  a  dam  extending  across  and  at  right  angles  to  the  stream,  and 
having  a  rectangular  notch  cut  in  the  top  plank,  with  both  side 
edges  and  bottom  sharply  beveled  toward  the  intake,  as  shown  in 
Fig.  19.  The  bottom  of  the  notch,  which  is  called  the  "  crest  " 
of  the  weir,  should  be  perfectly  level  and  the  sides  vertical. 

There  are  certain  proportions  which  must  be  observed  in  the 
dimensions  of  this  notch.  Its  length,  or  width,  should  be  between 
four  and  eight  times  the  depth  of  water  flowing  over  the  crest  of 
the  weir.  The  pond  back  of  the  weir  should  be  at  least  50 
per  cent  wider  than  the  notch  and  of  suflScient  width  and 
depth  that  the  velocity  of  flow  or  approach  be  not  over  1  foot  per 
second. 

On  the  up-stream  side  in  the  pond  a  stake  is  then  driven  down 
in  the  bottom  near  the  bank,  so  that  its  top  is  level  with  the  bot- 
tom edge  of  the  notch,  this  level  being  easily  found  when  the  water 
is  beginning  to  spill  over  the  crest.  The  stake  should  be  placed 
several  feet  from  the  board  and  at  least  not  nearer  than  the  length 
of  the  notch. 

By  means  of  a  rule,  as  shown  in  the  illustration,  the  depth  of 
water  over  the  top  of  the  submerged  stake  is  measured,  allow- 
ance being  made  for  the  capillary  attraction  of  the  water  against 
the  sides  of  the  weir.  Having  ascertained  this  depth,  the 
amoimt  of  water  flowing  the  weir  may  be  readily  found  from 
Table  XXIV. 

»  D.  W.  Mead,  "  Flow  of  Streams." 


Digitized  by 


Google 


STREAM-FLOW 


59 


IndiottM  Wauc  8t(md  darinff  the  Flood  SeMoa  added  to  the  Row  la  the  Dry  fi 


lllllllll  Indicates IVatonl  Ilactiiatioiia  of  Flow  ndiuive  oC  Flood  Waten. 


3  Indioatc*  Flood  Waten  which  coold  he  stored  ia  PropoMd  ReeerrtilEW' 
IHI^I  Indicates  Hood  Waters  UaaYoidabljr  wasted. 

Fig.    18. — Diagrams   Illustrating   Typical   Regulating   Effect   of   Proposed 
Reservoirs  on  the  Flow  of  the  Hudson  River. 


Digitized  by 


Google 


60 


HYDROLOGY 


Fig.  19. — ^Weir  for  Measuring  Flow  of  Water. 

TABLE  XXIV 
Table  for  Weir  Measureicent 
Giving  cubic  feet  of  water  per  minute,  that  will  flow  over  a  weir  1  inch  long 
and  from  |  to  20J  inches  deep. 


Depth, 
Inches. 

i 

i 

f 

i 

f 

} 

{ 

0 

.00 

.01 

.05 

.09 

.14 

.19 

.26 

.32 

1 

.40 

.47 

.55 

.64 

.73 

.82 

.92 

1.02 

2 

1.13 

1.23 

1.35 

1.46 

1.58 

1.70 

1.82 

1.95 

3 

2.07 

2.21 

2.34 

2.48 

2.61 

2.76 

2.90 

3.05 

4 

3.20 

3.35 

3.50 

3.66 

3.81 

3.97 

4.14 

4.30 

5 

4.47 

4.64 

4.81 

4.98 

5.15 

5.33 

5.51 

5.69 

6 

5.87 

6.06 

6.25 

6.44 

6.62 

6.82 

7.01 

7.21 

7 

7.40 

7.60 

7.80 

8.01 

8.21 

8.42 

8.63 

8.83 

8 

9.05 

9.26 

9.47 

9.69 

9.91 

10.13 

10.35 

10.57 

9 

10.80 

11.02 

11.25 

11.48 

11.71 

11.94 

12.17 

12.41 

10 

12.64 

12.88 

13.12 

13.36 

13.60 

13.85 

14.09 

14.34 

11 

14.59 

14.84 

15.09 

15.34 

15.59 

15.85 

16.11 

16.36 

12 

16.62 

16.88 

17.15 

17.41 

17.67 

17.94 

18.21 

18.47 

13 

18.74 

19.01 

19.29 

19.56 

19.84 

20.11 

20.39 

20.67 

14 

20.95 

21.23 

21.51 

21.80 

22.08 

22.37 

22.65 

22.94 

15 

23.23 

23.52 

23.82 

24.11 

24.40 

24.70 

25.00 

25.30 

16 

25.60 

25.90 

26.20 

26.50 

26.80 

27.11 

27.42 

27.72 

17 

28.03 

28.34 

28.65 

28.97 

29.28 

29.59 

29.91 

30.22 

18 

30.54 

30.86 

31.18 

31.50 

31.82 

32.15 

32.47 

32.80 

19 

33.12 

33.45 

33.78 

34.11 

34.44 

34.77 

35.10 

35.44 

20 

35.77 

36.11 

36.45 

36.78 

37.12 

37.46 

37.80 

38.15 

Digitized  by 


Google 


STREAM-FLOW 


61 


For  example:  Suppose  the  weir  to  be  72  inches  long^  and  the 
depth  of  water  over  the  stake  to  be  1  If  inches.  Follow  down  the 
left-hand  column  of  the  figures  in  the  table  until  you  come  to 
11  inches.  Then  run  across  the  table  on  a  line  with  the  11,  until 
under  f  on  top  line,  you  will  find  15.85.  This  multiplied  by  72, 
the  length  of  weir,  gives  1141.2,  the 
number  of  cubic  feet  of  water  passing 
per  minute. 

The  above  table  will  give  results 
sufficiently  close  for  all  practical  pur- 
poses, but  if  extreme  accuracy  is 
essential  the  following  formula  ^ 
might  be  used,  in  connection  with 
measurements  obtained  from  the 
method  previously  described: 

Q  =  3.33(L-.2H)«'/*. 

In  the  above  L= length  of  weir 
in  feet,  H=hesd  or  depth  of  flow  in 
feet  over  weir,  as  measured  on  the 
stake;  Q= cubic  feet  of  water  per 
second. 

The  Gurley  Hook  Gauge,  Fig.  20, 
is  a  very  useful  device  for  measuring 
the  depth  of  the  water  passing  over  a 
weir.  Its  arrangement  is  such  that 
the  readings  can  be  taken  by  the 
observer  with  the  greatest  possible 
convenience  and  at  some  distance 
from  the  surface  of  the  stream  being 
measured. 

This  gauge  is  used  in  a  box  attached  to  a  flume  at  any  con- 
venient point  near  the  weir,  the  water  from  the  flume  being  con- 
veyed to  the  box  by  rubber  or  lead  pipes,  thus  indicating  the  pre- 
cise level  of  the  water  in  the  flume,  the  surface  of  the  water  in  the 
box  being  at  rest.  The  exact  level  of  the  crest  of  the  weir  should 
be  taken  by  a  leveling  instrument  and  rod,  and  marked  by  a  line 
drawn  in  the  still  water  box  at  the  surface  of  the  water.  The 
1  Pelton  Water  Wheel  Ck). 


Fig.  20.— Gurley  Hook  Gauge. 


Digitized  by 


Google 


62 


HYDROLOGY 


scale  of  the  gauge  being  previously  set  at  zero  wth  the  vernier, 
the  base  ia  fastened  to  the  box  above  the  water  in  a  vertical  posi- 
tion and  at  such  a  height  that  the  point  of  the  hook  is  at  the  same 
level  as  the  crest  of  the  weir,  the  precise  point  being  secured 
by  moving  the  hook  in  the  tube.  The  point  of  the  hook 
will,  of  course,  be  under  water  and  level  with  the  crest  of  the 
weir. 

The  depth  of  water  flowing  over  the  weir  is  the  distance  between 
the  point  of  the  hook  in  the  position  named  and  the  exact  surface 
of  the  water.  To  ascertain  this,  the  hook  is  raised  by  turning  the 
milled  head  nut  imtil  the  point  of  the  hook,  appearing  a  Uttle 


Fig.  21. — ^Typical  Gauging  Station  with  Automatic  Gauge. 

above  the  surface,  causes  a  distortion  in  the  reflection  of  the  light 
from  the  surface  of  the  water.  A  slight  movement  of  the  hook  in 
the  opposite  direction  will  cause  the  distortion  to  disappear,  and 
will  indicate  the  surface  with  precision.  The  reading  of  the  scale 
will  then  give  the  depth  of  water  passing  over  the  weir,  in  thou- 
sandths of  a  foot. 

Where  measurement  by  weir  is  impracticable  the  amount  of 
water  can  be  calculated  by  ascertaining  the  average  velocity  of 
the  water  and  the  cross-section  of  the  stream,  the  quantity  being 
the  product  of  these  two  factors.  The  mean  velocity  is  the  func- 
tion of  the  cross-section,  surface  slope,  wetted  perimeter,  and 
roughness  of  the  bed,  while  the  cross-sectional  area  dep)ends  on 


Digitized  by 


Google 


STREAM-FLOW 


63 


the  permanency  of  the  bed  and  the  fluctuations  of  the  surface^ 
which  govern  the  depth. 

Gauging  stations  should  be  located  at  places  where  the  record 
of  flow  is  to  be  made.  Bridge  locations  are  preferable,  as  from 
them  the  measurements  can  be  easily  made,  with  the  least  expense. 
If  the  channel  conditions  are  not  satisfactory  at  such  points  it  is 
necessary  to  use  boats  or  erect  a  cable  station.  Fig.  21  showing  a 


Fig.  22.— Typical  Gauging  Station  for  Bridge  Measurement. 

typical  station  used  by  the  U.  S.  Geological  Survey.  The  location 
should  also  be  preferable  where  the  channel  is  straight  and  without 
cross-currents,  both  above  and  below  the  station,  and  the  bed 
should  be  as  free  from  obstructions  as  possible. 

The  methods  by  which  the  measxu^ments  are  made  are  in 
general  those  in  common  use  by  the  U.  S.  Survey.  An  arbitrary 
number  of  points  are  laid  off  perpendicular  to  the  thread  of  the 
stream,  Fig.  22.  They  are  known  as  measuring  points  and  divide 
the  gauging  section  into  strips.    The  area  for  each  strip  is  cal- 


Digitized  by 


Google 


64 


HYDROLOGY 


culated  from  careful  soundings  and  the  mean  velocity  ascertained 
by  making  measurements  at  different  depths.  By  multiplying 
the  area  and  the  velocity  for  each^strip,  its  discharge  value  ia 
determined  independently  of  the  other,  and  by  adding  them 
together  the  total  is  arrived  at  in  the  most  accurate  manner. 


Fig.  23. — Price  Electric  Current  Meter  with  Telephone  Sounder, 
factured  by  W.  &  L.  E.  Gurley,  Troy,  N.  Y.) 


(Manu- 


The  greatest  error  in  these  estimates  is  generally  due  to  inac- 
curate determination  of  the  mean  daily  gauge  heights,  ordinarily 
secured  from  a  few  observations  during  the  day  or  even  more 
infrequently.  This  has  led  to  the  introduction  of  automatic 
water  stage  registers  (see  page  265),  by  which  the  varying  height 
of  water  may  be  accurately  gauged  and  a  dependable,  continuous 
record  obtained. 


Digitized  by 


Google 


STREAM-FLOW 


65 


For  measuring  the  velocity  the  current  meter  is  jiow  most 
generally  used.  This  meter  is  primarily  an  instrument  for 
measuring  the  velocity  of  moving  water,  and  consists  essentially 
of  a  wheel  with  vanes,  which  may  be  shaped  like  those  of  a  wind- 
mill or  of  a  screw,  or  with  caps  like  those  of  an  anemometer,  the 
necessary  qualification  being  that  the  moving  water  shall  easily 
cause  the  wheel  of  the  meter  to  revolve.  The  velocity  of  the 
water  is  then  determined  from  the  revolutions  of  the  meter  in 
unit  time.    The  meter  which  has  been  adapted  by  the  U.  S. 


D  UOOO IMOO 16000 18000  20000  S9000  SiOOO  9 
Dtscharge  tn  Second  Feel 

Fig.  24.— Discharge,  Mean  Velocity,  and  Area  Curves  for  James  River  at 

Cartersville,  Va. 

Geological  Survey  after  years  of  experience  and  improvements  is 
the  Price  Current  Meter,  which  is  manufactured  by  W.  &  L.  R 
Gurley.    It  is  illustrated  in  Fig.  23. 

The  curves  in  Fig.  24  show  a  method  of  plotting  the  values 
of  discharge,  mean  velocity  and  area  in  relation  to  the  gauge 
height. 

Where  a  current  meter  is  not  available  or  its  expense  not  jus- 
tified as  in  minor  preliminary  investigations,  the  float  method  may 
be  used  for  approximately  determining  the  velocity.  This  may 
be  done  by  .laying  off  100  feet  of  the  bank  and  throw  a  float  into 


Digitized  by 


Google 


66  HYDROLOGY 

the  middle  of  the  stream,  noting  the  tune  it  takes  for  the  same  to 
pass  over  this  100-foot  stretch.  This  is  repeated  a  number  of 
times  and  the  average  taken.  As  the  stream-flow  at  the  swf ace  is 
greater  than  at  the  bottom,  the  average  must  be  taken  which  is 
about  83  per  cent  of  the  surface  velocity.  It  is,  therefore,  con- 
venient to  lay  o£f  the  distance  as  120  feet  and  reckon  it  as  100  feet, 
using  the  surface  velocity. 

Government  Records.  The  Water  Supply  and  Irrigation 
papers  of  the  United  States  Geological  Survey  furnishes  the  chief 
source  of  information  relating  to  stream-flow  measurements,  and 
a  complete  list  on  these  may  be  had  by  applying  to  the  Director, 
U.  S.  Geological  Survey,  Washington,  D.  C. 

The  U.  S.  Weather  Bureau  also  issues  annual  reports  on  the 
flow  of  the  principal  rivers  of  the  country,  while  the  War  Depart- 
ment from  time  to  time  issues  reports  dealing  with  special  investi- 
gations undertaken  by  ^the  engineers  for  determining  the  navi- 
gation faciUties  of  certain  rivers  and  the  possibilities  of  their 
improvement. 

In  addition  to  the  above  Federal  Reports,  numerous  investi- 
gations are  also  made  every  year  by  different  States  and  these  can, 
as  a  rule,  be  obtained  from  the  Geological  Survey  Departments 
of  these  States. 

It  is  thus  seen  that  there  is  an  abundant  amount  of  data  on 
stream-flows  in  the  different  sections  of  the  country.  These 
records  are,  however,  scattered  around  in  so  many  different  publi- 
cations, that  it  is  a  difficult  matter  to  find  the  desired  information. 
An  excellent  system  of  indexing  such  data  on  stream-flow  and 
rainfall  has  been  devised  and  is  used  by  H.  M.  Byllesby  A  Co., 
Chicago.  It  is  described  in  Engineering  Record  for  January  31, 
1914. 

6.  ENERGY  OF  FLOWmO  WATER 

The  energy  of  flowing  water  is  entirely  due  to  its  position,  or 
head.  It  follows  in  general  the  same  laws  as  falling  bodies  so 
that,  assuming  a  100  per  cent  efficiency,  its  potential  energy 
depending  on  the  position  must  be  equal  to  its  kinetic  energy 
depending  on  the  velocity.    That  is 


Digitized  by 


Google 


where 


thus 
and 


ENERGY  OF  FLOWING  WATER  67 


w 
m=iiia8s=-, 
9 

(/= gravity  acceleration  «  32. 16, 
A^heady 
»= velocity, 
w = weight  of  water = 62.4  lb.  per  cu.  ft. ; 


v^y/2gh. 
The  quantity  of  flowing  water  expressed  by  the  formula: 

where 

9= quantity; 

»= velocity; 

a = area  of  stream. 

From  the  above  the  following  formula  for  calculating  the  gross 
horse-power  of  a  stream  or  body  of  flowing  water  may  be  computed : 

^p_QXgX62.4. 
^•^'  550        ' 

in  which 

H.P. = gross  horse-power ; 

Q= discharge  of  water  in  cubic  feet  per  sec; 

£f= gross  head  in  feet. 

The  above  values  are,  however,  only  theoretical  and  never  realised 
in  practice.  This  is  caused  by  the  loss  in  head  due  to  friction  in 
the  water  conductors,  the  nature  and  value  of  which  will  be 
dealt  with  under  the  section  on  Water  Conductors. 


Digitized  by 


Google 


68 


HYDROLOGY 


6.  CONVENIENT  EQUIVALENTS 

The  following  is  a  list  of  convenient  equivalents  for  use  in 
hydraulic  computations: 

TABLE  XXV 

Table  for  oonyertmg  discharge  in  second-feet  per  square  mile  into  run-o£F 
in  depth  in  inches  over  the  area. 


Discharge  in 
Becond-feet  per 

Run-off 

(Depth  in 

Inches). 

Square  Mile. 

1  Day. 

28  Days. 

20  Days. 

30  Days. 

31  Days. 

1 

0.03719 

1.041 

1.079 

1,116 

1.153 

2 

.07438 

2.083 

2.167 

2.231 

2.306 

3 

.11167 

3.124 

3.236 

3.347 

3.459 

4 

.14876 

4.165 

4.314 

4.463 

4.612 

5 

.18696 

6.207 

5.393 

5.687 

5.764 

6 

,22314 

6.248 

6.471 

6.694 

6.917 

7 

.26033 

7.289 

7.660 

7.810 

8.070 

8 

.29752 

8.331 

8.628 

8.926 

9.223 

9 

.33471 

9.372 

9.707 

10.041 

10.376 

Note — For  partial  month  multiply  the  values  for  one  day  by  the  number  of  dasrs. 

TABLE  XXVI 
Table  for  converting  discharge  in  second-feet  into  run  ofif  in  acre-feet. 


Run-off  in  Acre- 

FEET. 

Discharge  in 

Second-feet. 

IDay. 

28  Days. 

29  Days. 

30  Days. 

31  Days. 

1 

1.983 

65.64 

67.60 

59.60 

61.49 

2 

3.967 

111.1 

116.0 

119.0 

123.0 

3 

6.960 

166.6 

172.6 

178.5 

184.6 

4 

7.934 

222.1 

230.1 

238.0 

246.0 

5 

9.917 

277.7 

287.6 

297.5 

307.4 

6 

11.90 

333.2 

345.1 

367.0 

368.9 

7 

13.88 

388.8 

402.6 

416.  ^ 

430.4 

8 

16.87 

444.3 

460.2 

476.  J 

491.9 

9 

17.86 

499.8 

617.7 

635.6 

663.4 

Note. — For  partial  month  multiply  the  values  for  one  day  by  the  number  of  days. 


Digitized  by 


Google 


CONVENIENT  EQUIVALENTS  69 

1  Beoond-foot  equals  40  California  miner's  inches  (Law  March  23, 1001). 

1  second-foot  equals  38.4  Colorado  miner's  inches. 

1  second-foot  equals  40  Arizona  miner's  inches. 

1  second-foot  equals  7.48  United  States  gallons  per  second;   equals  448.8 

gallons  per  minute;  equals  646,317  gallons  for  one  day. 
1  second-foot  for  one  year  covers  1  square  mile  1.131  feet  or  13.572  inches 

deep. 
1  second-foot  for  one  year  equals  31,536,000  cubic  feet. 
1  second-foot  for  one  day  equals  86,400  cubic  feet. . 
1  second-foot  equals  about  1  acre-inch  per  hour. 
1,000,000,000  (1  United  States  billion)  cubic  feet  equals  11,570  second-feet 

for  one  day. 

1,000,000,000  cubic  feet  equals  414  second-feet  for  one  28-day  month. 

1,000,000,000  cubic  feet  equals  399  second-feet  for  one  29-day  month. 

1,000,000,000  cubic  feet  equals  386  second-feet  for  one  30Kiay  month. 

1,000,000,000  cubic  feet  equals  373  second-feet  for  one  31-day  month. 

100  California  miner's  inches  equals  18.7  United  States  gallons  per  second. 

100  California  miner's  inches  for  one  day  equals  4.96  acre-feet. 

100  Colorado  miner's  inches  equals  2.60  second-feet. 

100  Colorado  miner's  inches  equals  19.5  United  States  gallons  per  second. 

100  Colorado  miner's  inches  for  one  day  equals  5.17  acre-feet. 

100  United  States  gallons  per  minute  equids  0.223  second-foot. 

100  United  States  gallons  per  minute  for  one  day  equals  0.442  acre-foot. 

1,000,000  United  States  gallons  per  day  equals  1.55  second-feet. 

1,000,000  United  States  gallons  equals  3.07  acre-feet. 

1,000,000  cubic  feet  equals  22.95  acre-feet. 

1  acre-foot  equals  325,850  gallons. 

1  inch  deep  on  1  square  mile  equals  2,323,200  cubic  feet. 

1  inch  deep  on  1  square  mile  equals  0.0737  second-ioot  per  year. 

1  foot  equals  0.3048  meter. 

1  mile  equals  1.60935  kilometers. 

1  mile  equals  5,280  feet. 

1  acre  equals  0.4047  hectare. 

1  acre  equals  43,560  square  feet. 

1  acre  equals  209  feet  square,  nearly. 

1  square  nule  equals  2.59  square  kilometers. 

1  cubic  foot  equals  0.0283  cubic  meter. 

1  cubic  foot  of  water  weighs  62.4  pounds  approx. 

1  cubic  meter  per  minute  equals  0.5886  second-foot. 

1  horse-power  equals  550  foot-pounds  per  second. 

1  horse-power  equals  76  kilogram-meters  per  second. 

1  horse-power  equals  746  watts. 

1  horse-power  equals  1  second-foot  falling  8.80  feet. 

IJ  horse-power  equals  about  1  kilowatt.  ..,,.► 

^       ,    ,  ^        X  .  ,,       sec.-ft.Xfallin  feet        ^  . 

To  calculate  water  power  qmckly:   — =iiet  horse-power  on 

water  wheel  realizing  80  per  cent  of  theoretical  power, 


Digitized  by 


Google 


CHAPTER  III 

CLASSIFICATION  OF  DEVELOPMENTS 

LOW-HEAD  DEVELOPMENTS 

Water-poweb  developments  may  be  divided  in  two  broad 
classes:  First,  low-head  and  second,  inedimn  and  high-head. 

To  the  former  class  belong  those  plants  which  consist  of  a  dam 
which  creates  pondage  at  the  point  where  the  water  is  to  be  utilized, 
so  that  the  water  passages  to  the  turbine  luits  will  be  compara- 
tively short  while  the  quantity  is  large.    The  chief  items  com- 


Fio.  26. — Map  Showing  General  Lay-out  of  Pennsylvania  Water  and  Power 
Companies,  Development  at  Holtwood,  Pa. 

prising  the  headworks  of  such  a  development  are:  The  dam  with 
its  spillway,  the  forebay,  the  intake  and  the  taUrace. 

Typical  plants  of  this  kind  are  shown  in  Figs.  25  and  26. 
A  dam  extends  across  the  river  and  impounds  a  large  body  of 
water  above  it.  It  is  built  with  a  spillway  section  for  the  entire 
length,  this  being  necessary  on  accoimt  of  the  large  flood  dis- 

70 


Digitized  by 


Google 


LOW-HEAD  DEVELOPMENTS 


71 


charges.  The  pondage  may  also  be  materially  increased  by 
placing  flashboards  on  top  of  the  dam,  and  by  this  means  an  addi- 
tional head  is  also  gained. 

Precautions  must  alwa3rs  be  taken  to  guard  against  floating 
1<^8,  debris,  ice,  etc.,  and  for  this  a  wing  dam,  having  submerged 
arches  through  which  the  water  enters  the  forebay,  has  been  built 
at  right  angles  to  the  main  dam,  between  which  and  a  rock-fiU 


Fig.  26. — SectioDal  Elevation  of  Power  House,  Cedars  Rapids  Mfg.  and 

Power  Company. 

above  there  are  floating  booms,  which  serves  to  deflect  such  ice, 
etc.,  which  is  carried  towards  the  forebay.  Care  should  be  taken 
that  the  arches  of  the  wing  dam  are  submerged  at  least  two  feet 
when  the  water  level  is  at  its  lowest. 

Any  ice  which  enters  the  forebay  despite  these  safeguards,  as 
weD  as  ice  which  may  be  formed  there,  can  be  diverted  by  provid- 
ing ice  chutes  from  the  forebay  toward  the  tailrace.    The  crests 


Digitized  by 


Google 


72 


CLASSIFICATION  OF  DEVELOPMENTS 


of  these  should  be  of  the  same  elevation  as  the  crest  of  the  main 
spillway. 

MEDIUM  AND  HIGH-HEAD  DEVELPMENTS 

To  this  class  belong  those  plants  which  consist  of  a  diversion 
dam  with  an  intake  at  the  head  waters  from  where  the  flow  is  led 


through  tunnels,  open  canals  or  flumes  to  a  forebay  pond.  This 
is  usually  located  on  the  hillside  above  the  power-house  and  pipe 
lines  carry  the  water  from  the  same  to  the  turbines.    In  other 


Digitized  by 


Google 


LOW-HEAD  DEVELOPMENTS  73 

instances  the  entire  water  conductor  from  the  diversion  dam  to 
the  wheels  may  be  of  enclosed  pressm^  type.  The  quantity  of 
water  is  usually  much  smaller  than  in  low-head  plants. 

High-head  developments  are  characteristic  of  the  California 
water  powers  where  the  high  mountain  storage  of  the  wiuter 
flood  waters  can  be  used  during  that  part  of  the  year  when  the 

run-off  is  a  Tninimnm. 

A  typical  high-head  installation  is  shown  in  Fig.  27.  It  con- 
sists of  a  diversion  dam  with  spillway  for  impounding  the  waters 
of  the  river,  thus  forming  a  reservoir  of  considerable  size.  The 
intake  is  located  at  right  angles  to  the  dam,  thus  lessening  the 
accumulation  of  ice,  logs,  trees  and  other  floating  debris  in  front 
of  the  intake  trash  racks. 

The  water  conductor  connecting  the  intake  and  the  turbines 
in  the  power-house  consists  of  five  sections;  a  reinforced  concrete- 
lined  tunnel  blasted  through  rock,  a  woodnstave  pipe,  a  steel 
pipe,  a  distributor  and  finally  the  steel  penstocks. 


Digitized  by 


Google 


CHAPTER  IV 

DAMS  AND  HEADWORKS 

1.  DAMS 

Classification.    Dams   may   be   classified   according  to  the 

material  used  in  their  construction,  as: 

.   Timber  crib  dams. 

Earth-fill  dams. 

Rock-fill  dams. 

Masonry  dams. 

The  choice  of  type  is  generally  dictated  by  natural  conditions. 
Sohd  rock  foimdations  usually  mean  masonry  dams,  whether  of 
overflow  type  or  not.  Absence  of  rock  foundations,  however, 
usually  means  the  choice  of  crib,  earth  or  rock-fill  dams,  and  which 
of  these  is  chosen  is  generally  determined  by  local  conditions, 
such  as  available  construction  material,  etc. 

Location.  Before  a  final  decision  can  be  reached  as  to  the 
exact  location  of  a  dam  there  are  niunerous  points  which  must  be 
carefully  investigated.  For  example,  with  low-head  developments, 
the  area  which  will  be  flooded  must  be  ascertained  as  this  will 
determine  the  available  head.  It  is,  therefore,  evident  that,  from 
this  point  of  view,  a  dam  would  be  preferable  at  a  point  where  the 
river  banks  are  steep  so  that  a  sufficient  pondage  and  head  could 
be  obtained  without  causing  a  flooding  of  too  much  adjacent  land. 

The  character  of  the  soil  is  also  of  the  utmost  importance,  and 
governs,  as  previously  stated,  the  type  of  dam  which  is  to  be 
selected.  It  should  be  impervious  and  able  to  withstand  the  load 
of  the  dam.  It  is  alwa3rs  advisable,  especially  where  a  sohd  rock 
foundation  is  not  to  be  had,  to  dig  or  drill  a  number  of  test  holes, 
from  which  the  character  of  the  underlying  strata  may  be  ascer- 
tained. It  may  then  be  found  that  one  site  will  require  a  very 
deep  foundation  but  a  smaller  dam  structure,  while  at  another 
site  the  reverse  may  be  true. 

Available  material  for  construction,  such  as  rock,  sand,  etc., 

74 


Digitized  by 


Google 


DAMS  76 

are  also  deciding  factors,  as  are  also  the  facilities  for  spillways  to 
take  care  of  the  overflow. 

It  is,  therefore,  evident  that  the  location  can  only  be  deter- 
mined after  a  careful  consideration  of  all  the  above  facts,  and  com- 
parative estimates  are  often  required  for  a  number  of  sites  before 
the  problem  can  be  intelligently  solved,  both  from  a  technical 
and  economical  standpoint. 

Timber  Crib  Dams.  These  dams  are  only  used  for  low  heads 
of  about  30  feet  and  less  and  in  locations  where  timber  is  plentiful 
and  cheap.    They  are  generally  used  for  diversion  purposes  and 


Pig.  28. — ^Timber  Crib  Dam,  Montana  Power  Company. 

mostly  entirely  submerged,  which  gives  them  a  long  life.  They 
are,  however,  often  used  for  temporary  structures  or  when  the 
cost  of  other  t3rpes  would  be  prohibitive  for  the  development  in 
question. 

They  consist  of  a  crib  or  framework  of  logs  or  sawed  timbers 
bolted  or  otherwise  fastened  together,  the  structure  being  filled 
with  rock,  gravel,  earth,  etc.,  and  the  sloping  sides  are  faced  with 
planks  to  prevent  leakage. 

Almost  any  kind  of  foundation  may  be  used  if  the  proper  pre- 
cautions are  taken.  With  solid  rock  the  framework  should  be 
securely  bolted  thereto  to  obviate  any  tendency  of  the  dam  to  slide. 


Digitized  by 


Google 


76 


DAMS  AND  HEADWORKS 


Soft  foundations  usually  require  a  dam  with  wider  base,  and 
it  may  be  necessary  to  first  fill  in  with  rock  or  gravel,  while  if  the 
soil  is  pervious  piling  may  also  be  required.  Undermining  should 
also  be  guarded  against  by  extejiding  the  facing  at  the  toe. 

Figs.  28  and  29  show  a  rock-filled  crib  dam  of  modem  design. 
The  up-stream  side  has  been  given  such  a  slope  that  the  stability 
of  the  dam  is  assured  even  under  the  greatest  floods,  the  weight 


itf^t  nnik.  ITM  i^i  h*  ^  «•*'  tr 

Fig.  29. — Cross-section  of  Timber  Crib  Dam  Shown  in  Pig.  28. 

of  the  water  acting  to  hold  it  down,  so  that  the  higher  the  flood 
the  greater  the  stabihty.  The  down-stream  side  is  also  sloping 
and  tapers  off  into  a  long  apron  so  designed  as  to  take  care  of  the 
overflow  without  shock  or  commotion.  A  sluiceway  is  provided 
at  one  end  of  the  dam,  and  at  the  other  there  is  a  concrete  cham- 
ber or  forebay  serving  as  intake  to  the  pipe  lines  supplying  the 
plant.  The  openings  to  this  forebay  are  controlled  by  gates  and 
are  provided  with  the  usual  screens  for  the  exclusion  of  trash. 

Earth-fill  Dams.  This  type  of  dam  generally  has  a  trapezoidal 
cross-section  and  consists,  as  the  name  impUes,  of  an  earth-fill 
faced  with  some  harder  material.  It  cannot  be  overturned  and  its 
stabihty  depends  on  the  imperviousness  of  the  material  used  in  its 
construction.  It  is  not  intended  to  be  used  as  a  weir,  and  in  case 
of  overflow  is  liable  to  be  disintegrated  and  washed  away.  For 
this  reason,  earth-fill  dams  must  be  provided  with  spillwa3n3  if 
there  is  danger  of  fiood-waters  passing  over  the  crest.  They  are 
not  intended  for  very  high  structures,  and  while  dams  of  this  type 
have  been  built  for  heights  above  100  feet,  about  50  and  75  feet 
is  more  common.  There  are  no  definite  rules  laid  down  for  cal- 
culating the  dimensions,  but  it  is  considered  good  practice  not  to 
let  the  slope  of  the  wetted  side  exceed  1  in  3,  while  the  outside 
slope  may  be  1  in  2.    The  height  should  be  at  least  10  feet  higher 


Digitized  by 


Google 


DAMS 


77 


than  the  high-water  level  and  the  width  of  the  crest  varies  any- 
where from  8  to  10  feet  for  low  dams,  to  20  or  more  for  the  highest 
one. 

One  of  the  most  important  things  in  its  construction  is  to  secure 
a  water-tight  foundation.  Hardpan  and  clay  are  good  founda- 
tions while  soft  soil  and  rocks  with  fissures  are  very  bad.  The 
site  must  be  cleared  of  tree  stumps,  roots,  etc.,  and  it  is  always 
necessary  to  remove  the  soil  for  a  depth  of  1  to  2  feet.    One  or 


— JWl 


Fig.  30.— Earth-Fm  Dam  with  Puddle  Core. 


OattiiL?  9urf  a«  to  b*  Drwed 
FUSia  vlth  SoU  ii*il  S 


<.  friw'i  nJil  tj  r  •  I  'J '.  J  -  ■ '  r 


Fig.  31. — ^Earth-Fill  Dam  with  Impervious  Puddle  Core. 

more  trenches  are  dug  parallel  to  the  axis  of  the  structure,  to  hold 
the  material,  and  if  the  soil  is  pervious  it  may  be  necessary  to 
provide  a  puddle  core,  as  shown  in  Fig.  30,  in  order  to  prevent  the 
water  from  seeping  under  the  dam,  or  piling  may  have  to  be 
driven  down  to  bedrock. 

The  material  which  goes  into  the  structiure  must  be  found  near 
the  dam  site,  and  its  character,  therefore,  determines  the  method 
of  construction  to  a  great  extent.    The  best  material  is  a  mixture 


Digitized  by 


Google 


78  DAMS  AND  HEADWORKS 

of  gravel,  sand  and  clay,  and  if  this  is  readily  obtained,  the  struc-- 
tiire  is  generally  built  homogeneous,  as  in  Fig.  31. 

There  are  many  different  methods  of  placing  the  material, 
such  as  providing  trestles  and  dump-cars,  cable  ways,  etc.  If  the 
material  is  taken  from  a  higher  elevation  than  the  dam,  and  water 
is  plentiful,  the  hydraulic  method  of  filling  may  be  used  and  is 
generally  found  very  economical. 

If  good  material  is  not  to  be  foimd  near  the  site,  puddle  or 
concrete  cores  must  be  built  to  insure  an  impervious  structiure,  as 
shown  in  Fig.  31. 

Such  a  puddle  core  is  pri^ferably  made  of  a  mixture  of  clay  and 
gravel,  this  being  considered  superior  to  clay  alone.  It  is  placed 
in  the  center,  with  the  finer  material  next  and  the  coarser  outside. 
It  should  be  protected  from  becoming  dry,  in  which  case  it  would 
crack  and  permit  the  water  to  se^p  through.  Enough  water  is, 
however,  generally  percolating  through  the  structure  to  keep  it 
moist.  The  fill  towards  the  outside  surface  should,  however,  be 
kept  as  dry  as  possible  to  keep  it  from  disintegrating,  and  it  is, 
therefore,  advisable  to  install  an  efficient  drainage  83^tem  on  this 
side. 

To  protect  the  wetted  side  from  the  effect  of  the  water  it  is 
usually  constructed  with  a  rip-rap,  and  sometimes  a  concrete 
facing  may  be  advisable  to  prevent  seepage.  The  other  side 
should  also  have  a  covering  of  rip-rap  or  gravel,  or  it  should,  at 
least,  be  sodded. 

Rock-fill  Dams.  A  typical  construction  of  this  type  of  dam 
is  shown  in  Fig.  32,  the  essential  difference  between  the  same  and 


E1.4SdO 


8p{lliini7  El.  48 W ^H.W.  El.  4m 


Fig.  32.— Rock-FiU  Dam. 

an  earth-filled  dam  being  the  rock-filled  part  which  forms  the 
down-stream  section,  while  the  other  side  is  filled  with  earth  and 
gravel. 

The  rock-fill  serves  as  a  support  for  the  earth-fill,  which  makes 
the  dam  impervious,  and  it  is,  therefore,  evident  that  this  type 


Digitized  by  VjOOQ IC 


DAMS 


79 


is  superior  to  the  plain  earth-filled  t3rpe,  in  that  less  damage  would 
be  caused  by  an  overflow. 

If  only  poor  material  can  be  obtained  for  the  earth-fiU,  it  is 
necessary  to  provide  a  puddle  or  concrete  core,  the  same  as  with 
the  previous  construction,  and  the  wetted  surface  should  also  be 
protected  by  a  rip-rap  or  concrete  facing. 

Masonry  Dams.  Masonry  dams  may,  according  to  their 
design,  be  divided  in  two  general  classes,  gravity  dams  and  arched 
dams,  and  these  further  into  solid  or  buttressed  structures. 

Gravity  Dams.  Gravity  dams  must  resist  any  tendency 
toward  sliding  or  overturning. 

Assimie  a  dam  structiure  of  a  trapezoidal  crossHsection  and 
with  the  water  surface  level  with  the  crest,  as  in  Fig.  33.  Then 
the  pressure  in  poimds 
acting  on  the  up-stream 
side  of  the  dam  per  foot 
length  is  equal  to 

P= rj X  sec  e. 


Where 

/f=Head  in  feet; 
(?= angle  of  dam  sur- 
face   with    the 
vertical; 
62.4  =  weight  of  1  cubic 
foot  of  water. 


Fig.  33. — Crofis-section  of  Gravity  Dam 
(Water  same  level  as  crest). 


This  pressure  acts  perpendicularly  to  the  surface  be  at  a  point 
two-thirds  the  height  of  the  dam,  figured  from  the  top.  The 
leverage  with  which  this  force  tends  to  overturn  the  structure 
about  point  d  is  equal  to  the  perpendicular  distance  between  this 
point  d  and  the  continuation  of  the  pressure  line  P,  i.e.,  dk.  The 
overturning  force  is,  therefore,  equal  to  Pxdk  foot-pounds. 

The  overturning  force  must  be  coimterbalanced  by  the  weight 
of  the  structure.  This  is  equal  to  W  and  it  acts  perpendicularly 
from  the  center  of  gravity.  Its  leverage  about  the  point  d  is  equal 
to  di  and  the  resisting  force  is,  therefore,  equal  to  WXdi  foot- 
pounds. 

The  center  of  gravity  of  a  trapezoid  may  graphically  be  found 


Digitized  by 


Google 


80 


DAMS  AND  HEADWORKS 


as  follows:  Draw  af  equal  to  cd  and  ec  equal  to  ab.  Divide  ab 
and  cd  in  two  equal  parts  and  connect  the  dividing  points.  Cour 
nect  e  and  fy  and  the  point  where  these  two  lines  intersect  is  the 
center  of  gravity.  It  may  also  be  calculated  from  the  following 
formula: 


.    H    H/cd-abX 


cd+abj' 

The  cross-section  A  of  the  dam  can  be  figured  from   the 
formula: 


and  by  multiplying  this  by  the  weight  of  masonry,  150  lbs.  per 
cu.  ft.,  the  weight  of  the  dam  per  foot  length  is  obtained. 

The  factor  of  safety 
1  witwLtTfi  against  overturning  S  of 

the  structure  is: 

-^  WXdi 


5  = 


PXdk' 


Fig.  34. — Croas-section  of  Gravity  Dam 
(Water  overflowing). 


34.    In  this  case  the  pressiure  P  is  equal  to 


It  is  seen  that  the  greater 
the  inclines  of  the  sur- 
faces the  more  stable 
will  the  structure  be. 

In  the  above  it  was 
assimied  that  the  water 
was  level  with  the  crest 
of  the  dam.  Suppose 
now  that  the  water  is 
flowing  over,  as  in  Fig. 


^=62.4/^^Vff-^ 


P=62.4r-lJ_*ViJ-A)xsec  6 


M^) 


sec  d  pounds. 


This  pressure  is,  however,  not  appUed  at  a  point  \H  from  the  top, 
as  in  the  previous  case,  but  at  a  point  x  from  the  top,  this  distance 
being  equal  to 


-i"-!^^- 


Digitized  by 


Google 


DAMS 


81 


The  resisting  mpment  due  to  the  weight  of  the  structure  is 
figured  as  in  the  previous  case,  except  that  the  weight  of  the  water 
should  also  be  considered.  The  factor  of  safety,  S,  is  found  from 
the  same  formula  as  before,  i.e.. 


S= 


WXdi 
PXdk' 


It  is  also  a  common  method  to  ascertain  if  the  design  is  safe 
by  completing  the  pressure  diagram,  as  in  Fig.  35.  The  two 
forces  P  and  W  are  scaled  off  from  the  intersection  point  X,  and 
if  their  resultant  Pi  falls 
inside  the  middle  third  of 
the  base,  the  dam  will 
safely  withstand  the  over- 
turning moment. 

It  is,  however,  not  sufloi- 
cient  to  determine  the  over- 
turning moment  for  the  full 
cross-section  about  the  toe. 
It  must  be  figured  for  sev- 
eral sections  such  as  a,  b,  c, 
d;  a,  b,  e,  /,  etc.,  Fig.  35, 
and  the  calculations  must 
show  that  every  part  of  the 
structure  is  sufficiently  thick 
to  withstand  the  pressiure. 

Besides  the  above  there  are  other  stresses  which  must  be  given 
due  consideration,  such  as  ice  thrust,  uplift  caused  by  seepage 
waters  and  internal  stresses  due  to  varjring  temperature  condi- 
tions. 

According  to  Mr.  A.  C.  Beardsley,  masonry  dam  design  should 
be  governed  by  the  following  rules: 

1.  Design  the  crest  and  apron  so  that  vacuums  cannot  form. 

2.  Underdrain  the  dam  to  eliminate  all  uplift. 

3.  Design  the  toe  of  the  dam  so  there  will  be  no  imcertainty 
as  to  the  exact  location  of  the  tipping  edge. 

4.  Allow  for  the  effect  of  floating  due  to  tail-water. 

5.  Allow  for  ice  expansion  and  use  the  maximum  crushing 
strength  of  ice  instead  of  average  values. 

6.  Take  care  of  expansion  and  contraction  stresses. 


Fia.  35. — Graphical  Determination  of 
Safety  of  Grarvity  Dams  (Middle-third 
method). 


Digitized  by 


Google 


82 


DAMS  AND  HEADWORKS 


7.  Allow  for  wave  action. 

8.  Where  necessary,  reinforce  the  dam  with  steel. 

On  account  of  their  great  weight,  gravity  dams  should  neces- 
sarily be  placed  on  bedrock  foundations,  and  the  materials  should 
be  carefully  tested  as  to  their  bearing  power.  The  fact  should 
also  be  kept  in  mind  that  the  pressure  is  not  uniform  over  the 
entire  base  but  varies  according  to  the  water  level  back  of  the  dam. 
For  example,  with  the  reservoir  full,  the  pressure  is,  of  course,  a 
maximum  at  the  toe  and  decreases  toward  the  other  side.     All 


Fig.  36. — ^Typical  Masonry  Dam  of  the  Gravity  Type. 

Company. 


Appalachian  Power 


tendency  of  seepage  should  be  prevented  by  sealing  all  fissures, 
and  drains  should  be  provided  for  carrying  any  waters  that  may 
reach  the  base  or  enter  the  structure. 

The  material  of  which  gravity  dams  are  built  consists  either 
of  concrete  or  rubble  masonry.  With  the  former  the  rocks  are 
crushed  to  a  uniform  size  making  an  even  mixture,  while  with  the 
rubble  masonry,  or  cyclopean  concrete  construction,  as  it  is  also 
termed,  large  stones,  weighing  up  to  ten  tons,  are  used.  These 
are  carefully  placed  in  position  and  the  spaces  between  them 


Digitized  by 


Google 


DAMS 


83 


filled  with  smaller  stones  and  cement  mortar,  forming  a  very 
strong  structure. 

In  most  low-  and  medium-head  developments  where  large  flood 
discharges  must  be  passed,  the  entire  dam  or  a  large  part  of  it 
must  be  built  in  the  form  of  a  spillway.  It  is  important  that  its 
size  be  sufficient  to  take  care  of  the  largest  known  floods,  and  in 


flW<iirm)aaJ  til  n«t\i 


HoUu-In  ealonlafing  pressures 

Concrete  assumed  to  weight 

8m  assumed  to  weigh  115  Vm. 
per  en.  ft. 


Dm^Qb  unJfft  toe  Qi  l>offl  t€  b& 


Fig.  37. — Cross-section  of  Masonry  Gravity  Dam  Shown  in  Fig.  36. 

order  to  be  on  the  safe  side  it  is  in  many  instances  designed  for 
10  to  15  per  cent  greater  discharge  capacity  than  any  previous 
record  would  show  to  have  taken  place.  The  downstream  face 
should  be  curved  so  that  the  water  will  follow  the  surface  and 
prevent  vacuum  from  forming,  and  also  so  that  it  is  discharged  in  a 
horizontal  direction,  protecting  the  bed  of  the  stream  against 


Digitized  by 


Google 


84 


DAMS  AND  HEADWORKS 


Digitized  by  CjOOQ IC 


DAMS  85 

undercutting  and  erosion  at  the  lower  end  of  the  toe  when  passing 
severe  floods,  and  permitting  a  quiet  discharge  without  subjecting 
the  masonry  structure  to  dangers  from  vibrations. 

The  illustrations  in  Fig.  36  and  37  show  a  typical  design  of  a 
masonry  dam  of  the  straight  gravity  tjrpe. 

Buttressed  Dams.  This  type  of  gravity  dam  has  been  devised 
with  a  view  of  utilizing  the  material  more  economically  than  is 
possible  in  a  gravity  structure,  a  typical  design  being  illustrated  in 
Fig.  38.  As  seen,  it  is  a  hollow  structure  consisting  of  a  concrete 
deck  supported  at  stated  intervals  by  buttresses  or  piers  per- 
pendicular to  the  axis  of  the  dam.  As  the  downward  pressure 
of  the  water  is  relied  on  to  a  great  extent  to  give  the  structme 
stability,  the  upstream  face  should  have  an  incline  of  not  more 
than  45^  with  the  horizontal.  The  thickness  of  the  deck  should 
be  proportioned  in  accordance  with  the  hydrostatic  pressure,  and 
it  should  vary  uniformly  from  the  base  to  the  top,  being  some- 
times reinforced  with  steel  to  increase  its  strength.  Careful 
precautions  should  be  taken  to  make  the  structure  water-tight, 
and  drains  should  be  provided  as  well  as  passageways  for  interior 
inspection. 

This  type  of  dam  requires  very  good  foimdations.  As  the 
entire  pressure  must  be  withstood  by  the  buttresses  alone,  it  is 
evident  that  the  base  width  of  these  at  right  angles  to  the  axis 
will  have  to  be  considerably  greater  than  for  a  gravity  type  struc- 
ture. 

Arched  Dams.  These  may  be  either  of  solid  or  buttressed 
design,  curved  in  a  horizontal  arch  with  the  abutments  braced  in 
the  rock  on  the  sides  of  the  gorge  or  canyon,  thus  giving  greatly 
increased  stability.  It  is  not  considered  good  practice,  however, 
to  rely  entirely  on  the  arch  action  and  dams  of  this  class  are, 
therefore,  as  a  rule  designed  as  a  combined  arch  and  gravity  type. 
In  fact,  the  dam  is  often  designed  purely  as  a  gravity  structure, 
and  the  added  strength  given  by  its  curved  form  is  simply  assiuned 
to  increase  its  safety  to  that  extent. 

It  has  been  the  general  practice  to  build  these  dams  in  one  con- 
tinuous arch,  and  Jorgensen  in  "Journal  of  Electricity"  states  that 
for  spans  less  than  600  feet  a  curved  dam  of  this  type  requires  less 
material  for  the  same  factor  of  safety  than  a  straight  gravity  dam. 
If  the  gap  to  be  closed  is  over  600  feet,  the  cross-sectional  area  of 
the  arch  becomes  nearly  as  great  as  the  cross-sectional  area  of  a 


Digitized  by 


Google 


86 


DAMS  AND  HEADWORKS 


Fig.  39.— Arched  Dam.    Orland  Project,  California. 


.■>...     -7^///"^     '     Confflonienite 


Fig.  40. — Cross-section  of  Arched  Dam  Shown  in  Fig.  39. 


Digitized  by 


Google 


DAMS 


87 


FiQ.  41. — Multiple-Arched  Dam.    Umatilla  Project,  Oregon. 


FiQ.  42. — CrofiSHsection  of  Multiple-Arched  Dam. 


Digitized  by 


Google 


88  DAMS  AND  HEADWOKKS 

gravity  dam  for  equal  stresses,  and  when  it  is  considered  that  the 
arch  is  alwa3rs  longer  than  the  chord,  it  is  evident  that  the  limit 
of  economy  for  a  single  arch  span  has  been  reached,  if  special 
conditions  are  not  present. 

Recently  the  multiple-arch  type  of  dam  has  come  into  exist- 
ence and  gives  promise  of  allowing  big  reductions  in  the  quan- 
tity of  material  required  for  structures  safely  spanning  gaps  of 
any  width. 

Figs.  39  and  40  illustrate  an  arched  dam  and  Figs.  41  and  42  a 
dam  of  the  multiple-arched  type. 

General  Rules  Governing  Design  of  Dams.  The  foDowing 
regulations  governing  the  design  and  construction  of  dams  were 
recently  issued  by  the  New  York  State  Conservation  Commission. 

"  Complete  plans,  elevations  and  sections  of  all  proposed 
dams  must  be  submitted  and  approved  of  by  this  commission 
before  any  work  on  the  dam  can  be  commenced,  and  the  site  must 
be  examined  and  approved  of  by  this  commission,  both  before 
and  after  it  has  been  prepared. 

**  Foundation  Bei:  Dams  must  be  built  upon  a  firm,  compact, 
impervious  and  natural  foundation  bed,  from  which  all  perish- 
able material  has  been  removed.  Earth  foundation  beds  must  be 
ploughed  or  trenched.  Masonry  must  be  carried  into  solid  rock 
at  the  base  and  sides,  wherever  practicable,  and  also  have  channels 
cut  into  the  rock  bed  sufficient  to  afford  a  firm  hold  for  the  dam. 
Rock  foundations  must  have  all  loose  material  removed;  the 
crevices  for  200  feet  above  and  for  100  feet  below  the  dam,  must 
be  thoroughly  filled  with  concrete  or  grout,  and  the  whole  surface 
under  the  dam  thoroughly  washed.  Masonry  dams  over  36  feet 
in  height  must  have  the  rock  bed  drilled  for  hidden  fissures  and 
tested  by  compressed  air;  these  holes  must  be  filled  with  grout 
under  a  pressure  equal  to  the  maximimi  ultimate  pressure. 

''  CoLcvlations:  Dams  must  be  stable  at  any  section  and  under 
all  conditions.  The  compression  upon  masonry  on  the  upstream 
face  shall  be  10,  14  and  18  tons  per  square  foot  and  for  the  down- 
stream face  8,  10  and  14  tons  per  square  foot,  depending  on  the 
mass;  the  first  for  walls  less  than  12  feet  thick  and  buttressed 
dams,  and  the  last  for  solid  masonry  dams  over  150  feet  in  height 
with  the  best  of  work  done  under  the  inspection  of  a  competent 
engineer  approved  by  this  commission. 

"  All  cement  must  be  Portland  and  up  to  the  standard  of  the 


Digitized  by 


Google 


DAMS  89 

New  York  City  Building  Law,  tested  as  prescribed  by  the  Amer- 
ican Society  of  Civil  Engineers,  and  must  more  than  fill  the  voids 
of  sand  and  stone  mixed  in  the  proportions  as  used.  The  sand 
and  stone  used  for  masonry  must  be  sound  and  permanent,  clean, 
hard,  and  not  easily  sheared  or  split. 

"  OvUeU:  All  dams  must  be  provided  with  approved  outlets 
of  sufficient  size,  and  so  located  as  to  completely  allow  the  im- 
pounded water  to  be  released  when  desired  or  necessary,  and 
precautions  must  be  made  to  prevent  leakage  along  the  outlets. 

"  Ice  Pressure:  From  Dec.  1  to  March  15  no  dam  shall  have 
the  water  higher  than  two-thirds  the  height  of  the  dam,  unless 
permission  is  granted  by  the  Conservation  Commission  to  keep 
the  water  above  at  a  higher  level.  Dams  Uable  to  be  full  during 
the  above  period  must  be  built  strong  enough  to  resist  any  possible 
ice  pressure  in  addition  to  the  water  pressure,  and  dams  not  so 
designed  must  have  an  outlet  at  two-thirds  the  height  of  the  dam. 

"  Aprons:  Spillways  of  all  dams  must  be  provided  with  aprons 
or  other  provision  on  the  downstream  side  to  prevent  the  under- 
mining of  the  dam  by  the  falling  waters. 

"  Wooden  Dams:  Wooden  dams  may  be  used  for  temporary 
purposes,  or  where  the  reach  of  the  water  impounded  above  the 
dam  is  not  over  300  feet  or  its  depth  more  than  10  feet.  The 
timber  of  the  dam  must  be  removed  at  the  end  of  five  years, 
unless  express  permission  is  granted  by  the  Conservation  Com- 
mission for  a  longer  period. 

"  The  crib  work  of  wooden  dams  shall  be  built  in  pockets  not 
more  than  8  feet  square,  well  fastened  together  with  at  least 
f-inch  spikes  or  bolts,  long  enough  to  pass  through  three  timbers, 
and  the  pockets  soUdly  packed  with  stone.  The  upstream  face  is 
to  be  built  at  an  angle  of  three  horizontal  to  one  vertical,  covered 
with  plank,  on  which  is  to  be  laid  a  good  layer  of  gravel.  If  the 
foundation  is  rock,  the  bottom  timbers  must  be  anchored  to  the 
rock. 

"  Earth  Dams:  The  upstream  half  of  earth  dams  shall  be 
composed  of  gravelly  earth  with  about  15  per  cent  of  clay,  with 
no  stones  over  4  inches  near  the  upstream  face,  or,  if  there  be  a 
core,  next  to  the  core  on  the  upstream  side.  The  earth  is  to  be 
moist,  not  wet,  well  rolled  in  12-inch  layers  slightly  sloping  down 
to  the  middle  of  the  dam.  The  downstream  half,  or  part  below 
the  core,  may  be  composed  of  coarser  materials  and  stones.    The 


Digitized  by 


Google 


90  DAMS  AND  HEADWORKS 

»'-■ 

top  should  be  slightly  convexed  and  of  a  minimum  width  of  8  feet 
plus  1  foot  in  width  for  every  5  feet  over  15  feet  in  height.  The 
slop)es  should  be  two  horizontal  to  one  vertical,  except  if  stone  is 
used  on  the  downstream  half  it  may  be  one  and  one-half  horizontal 
tc  one  vertical.  If  the  upstream  part  is  of  very  fine  material,  the 
slope  must  be  less.  A  berm,  or  horizontal  surface,  which  shall 
be  not  less  than  4  feet  wide,  shall  be  constructed  on  the  slopes  at 
every  20  feet  horizontally  from  the  top.  On  the  downstream 
face  these  berms  shall  be  provided  with  paved  gutters.  The  up- 
stream face  shall  have  an  18-inch  stone  pavement  laid  in  broken 
stone  or  gravel  from  the  top  to  the  upper  berm,  and  below  shall 
have  a  pavement  of  rip-rap.  The  downstream  face  is  to  be  sodded 
or  covered  with  12  inches  of  gravel  or  rip-rap. 

"  Every  earth  dam  must  be  provided  with  a  masonry  spillway 
of  sufficient  unobstructed  area  to  take  the  high  flow,  and  built 
with  the  same  requirements  as  for  masonry  dams.  The  height 
of  the  dam  shall  be  at  least  3  feet  above  high  flow,  plus  3  feet  for 
a  reach,  or  expanse  of  water  upstream,  of  one  mile,  plus  8  feet  for 
a  reach  of  two  miles,  and  proportional  for  an  intermediate  reach. 
"  Earth  dams  of  over  10  feet  in  height  shall  be  provided  with 
a  masonry  core  in  the  middle,  the  top  to  be  not  more  than  2  feet 
below  the  top  of  the  dam,  and  a  top  width  of  not  less  than  2  feet 
with  a  batter  of  1  horizontal  to  24  vertical  on  each  side.  Or,  the 
core  may  be  placed  on  the  upstream  side,  in  which  case  the  width 
of  the  core  at  any  point  must  be  equal  to  half  of  the  depth.  Or, 
the  core  may  be  omitted  and  the  dam  made  5  feet  wider  and  3  feet 
higher  than  above  specified;  in  this  case  the  hydraulic  process  of 
construction  may  be  employed. 

"  Masonry  Dams:  The  least  width  of  masonry  dams  shall  be 
one-tenth  of  the  height,  with  a  minimum  of  4  feet.  The  mini- 
mum width  at  any  depth  shall  be  two-thirds  the  depth  below  the 
highest  water  level. 

"  The  masonry  must  be  built  up  in  horizontal  sections  with 
center  grooves  in  the  top  and  sides  for  bonding,  formed  by  cm- 
bedding  beveled  timbers  in  the  concrete.  Concrete  masonry 
shall  have  vertical  cast-iron  bars  in  the  upstream  face,  placed  at 
least  2  feet  apart  and  of  sufficient  length  to  protect  the  masonry 
against  ice  and  floating  bodies. 

''Reinforced  Buttressed  Dams:  The  buttresses  shall  not  be 
over  20  feet  apart  for  dams  over  100  feet  high  on  rock  foundations, 


Digitized  by 


Google 


FLASHBOARDS  91 

and  nearer  for  others,  with  the  necessary  cross  stiflfening  girders. 
The  upstream  face  shall  be  at  an  angle  of  not  over  45°  with  the 
horizontal  and  the  downstream  face  not  over  60°.  No  part  of 
the  dam  shall  be  less  than  12  inches  thick. 

*'  If  the  dam  is  on  rock  foundations,  the  front  face  must  have  a 
heavy  cut-oflf  wall  built  into  the  rock.  If  on  gravel  and  clay  foun- 
dations both  faces  must  have  deep  cut-oflf  walls  and  a  heavy  rein- 
forced flooring  with  weep  holes  to  relieve  the  water  pressure  under 
the  flooring.  Drainage  must  be  provided  in  interior  pockets  for 
seepage  waters,  and,  if  practical,  the  interior  must  be  made  acces- 
sible to  allow  for  inspection. 

"  The  crest  of  the  spillway,  and  for  3  feet  below,  must  be  thick- 
ened and  heavily  reinforced,  and  the  entire  dam  and  bulkheads 
protected  from  ice  and  floating  bodies  the  same  as  masonry  dams. 
The  dam  must  be  well  anchored  to  the  bulkheads." 

2.  FLASHBOARDS 

The  maintaining  of  a  constant  water  level  above  the  dam  is 
naturally   very   desirable.     This   water   surface   fluctuates   con- 
siderably during  the  diflferent  seasons  of  the  year,  depending  on 
the  flow,  and  it  was  previously  shown  that  the  spillway  must  be  of 
ample  capacity  to  discharge  the  flood  waters  and  prevent  the 
water  above  the  dam  from  flooding  such  land  as  has  not  been 
included  in  the  flowage  area.     It  is  furthermore  desii*able  to  keep 
the  surface  at  approximately  the  same  level  during  the  low- 
water  periods  and  thus  maintain  a  constant  head.    This  is  accom- 
plished by  providing  flashboards,  which  are  placed  on  the  top  of 
the  dam,  and  arranged  to  be  raised  or  lowered  with  the  variation 
in  the  water  level.     It  has  also  been  found  that  for  installations 
with  steam  reserve  plants  the  operating  arrangement  that  will 
insure  the  most  efficient  use  of  the  river  flow  is  to  maintain  the 
level  in  the  storage  reservoir  at  nearly  the  crest  of  the  flashboards, 
carrying  by  the  auxiliary  plant  any  excess  load  until  such  time  as 
reports  from  the  watershed  above  indicate  a  freshet.    Then  the 
stream  plant  is  shut  down  and  the  water  drawn  down  in  the  reser- 
voir to  such  an  extent  as  to  allow  it  to  be  filled  by  the  anticipated 
freshet. 

There  are  numerous  designs  of  such  flashboards,  the  most 
common  being  as  follows: 

1.  Stationary  flashboards. 


Digitized  by 


Google 


92 


DAMS  AND  HEADWOKKS 


2.  Sliding  gates. 

3.  Tilting  gates. 

4.  Tainter  gates. 

5.  Rolling  gates. 

All  cf  these  with  the  exception  of  the  first  class  require  that  piers 
be  provided  on  the  crest  of  the  dam,  between  which  they  may  be 
supported.  The  number  of  these  piers  and  spillway  sections 
depends  then  on  the  maximum  length  to  which  the  gates  can  be 
successfully  built. 

Stationary  Flashboards.  This  arrangement  simply  consists 
in  placing  a  row  of  wooden  panels  on  top  of  the  dam  crest,  and 
supporting  them  by  iron  pins  which  are  set  vertically  in  holes 


np>8tream  Bide 


-GnwaPIiiB 


Fia.  43. — Stationary  Flash-board  Design. 

previously  provided  in  the  concrete  structure,  as  shown  in  Fig.  43. 
These  pins  are  so  dimensioned  that  when  the  water  reaches  a 
certain  elevation  they  will  give  way  and  readily  release  the  boards. 

R.  MuUer  ("  Engineering  Record,"  August  22, 1908),  gives  the 
following  formula  for  calculating  the  head  of  water  that  will 
cause  the  iron  pins  to  bend.    It  is  based  on  Wa3me  iron  pins: 

y_18.12(P    2. 

in  which 

X  =  height  of  water  in  feet  above  the  dam  crest  when  pins 

begin  to  bend; 
d= diameter  of  pins  in  inches; 
jS= spacing  of  pins  in  feet; 
A = height  of  flashboard  in  feet. 


Digitized  by 


Google 


FLASHBOARDS 


93 


The  ends  of  the  different  sections  overlap  each  other,  as  seen 
in  the  illustration,  and  a  fairly  water-tight  joint  is  thus  provided  by 
utiUzing  the  water  pressure  itself.  For  sealing  the  joint  between 
the  lower  edge  of  the  boards  and  the  masonry  it  has  been  found 
that  a  composition  of  cinders  and  straw,  well  mixed  before  appli- 
cation, is  very  satisfactory.  In  it  the  cinders  form  the  body,  while 
the  straw  is  the  elastic  tightening  medium. 

While  the  pins  are  ordinarily  removed  once  a  year,  the  flash- 
boards  are  Ukely  to  be  taken  up  a  number  of  times  each  season, 
and  speed  and  economy  in  their  handling  is,  therefore,  of  impor- 
tance. For  wide  streams  the  usual  method  of  handling  them  is  by 
means  of  a  scow  provided  with  a  steam-driven  derrick,  while  for 
narrower  streams  specially  designed  cableways  with  chain  hoist 
have  been  used  with  very  great  success. 

Sliding  Gates.  These  may  be  either  of  the  plain  friction  type 
or  they  may  be  provided  with  roller  guides  to  make  their  operation 
easier. 

The  gates  used  by  the  Mississippi  River  Power  Company  at 
Keokuk,  Iowa,  shown  in  Fig.  44,  indicate  probably  the  maximum 


^  r  rn  1  n 


^t£i» 


Fia.  44. — Spillway  Gates.  Mississippi  River  Power  CJompany, 
Keokuk,  Iowa. 

size  to  which  the  friction  type  can  be  built.  They  are  11  feet 
high  and  32  feet  long  over  all.  Each  gate  consists  of  a  frame- 
work of  18-inch  I-beams,  covered  with  |-inch  steel  plate  on  the 
upstream  side.  The  edges  are  milled  to  make  a  water-tight  joint 
with  the  iron  sill  plates  against  which  they  fit,  and  the  gates  are 


Digitized  by  VjOOQ IC 


94 


DAMS  AND  HEADWORKS 


operated  by  an  electrically  driven  crane  running  along  the  bridge, 
which  forms  the  top  of  the  dam. 

For  smaller  installations  a  much  simpler  structure  can,  of 
course,  be  used,  such  as  an  ordinary  hand-operated  sluice  gate. 
(See  section  on  "  Gates  and  Valves.") 

A  good  example  of  the  enormous  size  to  which  sliding  gates 
with  roller  guides  can  be  built  is  that  of  the  Gatim  spillway  of  the 
Panama  Canal,  as.  shown  in  Figs.  45  and  46.  Each  of  these 
gates  has  a  height  of  19  feet  and  an  over-all  length  of  47  feet.    The 


Fio.  45. — Gatun  Spillway,  Panama,  Showing  Spillway  Gates. 

operating  machinery  is  designed  to  raise  or  lower  the  gates  in 
approximately  ten  minutes.  It  consists  essentially  of  two 
counterweights,  one  at  each  end  of  the  gate,  which  practically 
balance  the  weight  of  the  gate,  so  that  the  machine  has  to  over- 
come only  the  resistance  to  movement  of  the  gate  due  to  the 
water  pressure.  These  counterweights  are  connected  to  the 
gate  by  a  screw  and  chain,  the  screw  being  moved  vertically  by 
means  of  a  worm  nut,  which  is  motor  driven  by  a  worm.  The 
two  screws  at  the  gate  ends  are  driven  simultaneously  through  a 
driving  shaft  which  is  provided  with  a  worm  at  each  end  for 


Digitized  by 


Google 


FLASHBOARDS 


95 


tf%\::^'  15(2  /  y  y^^ 


Digitized  by  VjOOQIC 


96 


DAMS  AND  HEADWORKS 


El.lSl4-0f>? 


ILISSM. 


operating  the  wonn  nuts.  The  screws  are  held  m  a  vertical  posi- 
tion and  the  hoisting  chains  pass  over  sheaves  at  the  tops  of  the 
gate  piers.  A  machinery  tunnel  extends  the  full  length  of  the 
spillway,  a  distance  of  approximately  800  feet,  and  is  built  within 
the  dam  and  contains  all  the  operating  machinery.  Limit  switches 
are  provided  to  prevent  overtravel  by  cutting  oflf  the  current  from 
the  motor  at  the  proper  instant. 

Tilting  Gates.  This  type  of  flood  gate  generally  consists  of  a 
flaishboard  which  is  hinged  at  its  lower  edge  to  the  crest  of  the 

spillway,  the  other  edge 
being  free  to  move  from 
a  more  or  less  vertical  to 
a  horizontal  position.  It 
maintains  its  upright  posi- 
tion imtil  the  water  level 
above  the  dam  reaches 
the  normal  level.  As  the 
water  continues  to  rise  the 
additional  pressure  on  the 
gate  will  cause  it  to  tilt 
over  further  until  it  finally 
rests  in  a  horizontal  posi- 
tion on  the  dam  crest. 
As  the  water  subsides  the 
gate  will  automatically 
rise  until  the  normal  water 
level  in  the  pond  is 
reached. 

Many  different  devices 
have  been  used  for  accom- 
plishing the  coimterbal- 
ancing  effect,  one  of  the 
latest  being  that  shown  in 
Fig.  47.  This  particular  installation  is  designed  to  operate  with 
a  maximum  fluctuation  in  water  level  of  three  inches. 

Each  flashboard  consists  of  a  steel-reinforced  timber  panel 
hinged  at  the  bottom  and  connected  at  the  top  to  a  17-ton  con- 
crete roller  counterweight  by  two  steel  cables,  which  are  wound 
in  grooves  around  each  end  of  the  roller.  These  rollers  travel  on 
inclined  tracks,  each  end  being  provided  with  a  geared  drum  which 


ELiaaLo. 


Fig.  47.— Tilting  Spillway  Gate  with 
Counter  Weight. 


Digitized  by 


Google 


FLASHBOARDS 


97 


£t  un.& 


engages  a  rack  to  prevent  slipping.  The  principle  of  operation  is 
simply  a  balancing  of  the  moments  of  force.  The  pressure  on  the 
flashboards  is  transmitted  to  the  dnmis  through  the  cables  which 
act  to  roll  the  coimterweight  up  the  track,  while  its  dead  weight 
tends  to  roll  it  down;  the  two  forces  balancing  each  other  when 
the  water  level  is  at  the  fixed  elevation.  Hand-operated  winches 
are  also  provided,  and  their  general  construction  is  clearly  shown  in 
the  illustration. 

The  above  dam  consists  of  10  spillway  openings,  6  of  which  are 
provided  with  these  automatic  spillway  gates.  The  other  4 
openings,  which  are  located  towards  the  intake  side,  are  pro- 
vided with  flashboards  of  the  ordinary  stationary  construction, 
and  are  so  designed  that  if  the  water  in  the  pond  rises  1  foot 
above  the  normal  level,  the  boards  will  give  away. 

Trainter  Gates.  This  type  of  gate  is  generally  built  of  steel 
throughout,  its  general  construction  being  clearly  shown  in  Fig.  48. 
In  order  to  make  it  water-  c  '  -  a 

tight  the  bottom   of  the  * — laV.    JL;    Igkitga 

gate  may  be  fitted  with  a 
sill  block  of  oak,  which 
takes  a  bearing  on  a  steel 
plate  set  in  the  top  of  the 
concrete  sill.  Along  the 
ends  may  also  be  fitted 
rubber  strips  for  making  a 
tight  joint  with  the  side 
walls. 

The  gates  are  usually 
raised    and    lowered    by ' 
chains    attached    to  the 
bottom  edge  of  the  gate 
and  wound  upon  drums  on  a  shaft  above, 
hand-  or  motor-operated. 

Rolling  Gates.  The  principle  of  these  gates  is  impUed  in  the 
name,  that  is,  the  weir  body  is  moved  away  from  its  closed  posi- 
tion by  rolling  on  an  inclined  track.  In  the  simplest  form  it  con- 
sists of  a  large  hollow  cylinder  of  a  diameter  corresponding  to  the 
height  to  which  it  is  desired  to  raise  the  water,  and  of  a  length 
equal  to  the  width  of  the  opening  to  be  closed.  This  cylinder  is 
built  up  of  boiler  plate,  substantially  braced  to  withstand  the 


Fig.  48.— TaiDter  Gate. 


They  may  be  either 


Digitized  by 


Google 


98  DAMS  AND  HEADWORKS 

• 
strains  to  which  it  is  subjected.    At  each  end  the  cylinder  is  pro- 
vided with  a  specially  designed  gear  engaging  a  rack  laid  in  an 
inclined  recess  in  the  abutment  or  pier.     By  means  of  a  sprocket 
chain  wrapped  around  one  end  of  the  cylinder  and  connecting  with 


1 

S 


I 


I 


(5 


the  operating  mechanism  the  dam  can  be  rolled  up  or  down  as 
desired,  see  Figs.  49  and  49A. 

For  larger  lifts  and  moderate  spans,  the  cylindrical  part  of  the 
weir  is  often  much  smaller  in  diameter  than  the  height  of  the  weir, 
the  upstream  side  of  the  gate  being  provided  with  a  metal  shield 
connected  by  strong  braces  to  the  cylindrical  body. 

This  type  of  gate  is  a  comparatively  new  invention  and,  while 


Digitized  by 


Google 


FISHWAYS 


99 


it  has  been  used  in  Europe  to  a  considerable  extent,  there  are  only 
a  few  installations  in  this  country.     It  possesses  many  advantages 


Pig.  49a. — ^Dam  of  Wasmngton  Water  Power  Company,  Showing  Arrange- 
ment of  Rolling  Gates. 

over  other  types  of  flood  gates  on  account  of  the  larger  size  in 
which  it  can  be  built.  For  example,  rolling  dams  have  been  built 
in  Europe  with  lengths  up  to  115  feet  and  depths  of  28  feet. 

8.  FISHWAYS 

In  many  States  the  law  demands  that  dams  be  provided  with 
means  whereby  fish  can  easily  ascend  and  descend  according  to 
their  natural  habits  in  search  of  spawning  grounds  and  of  food. 

Many  different  designs,  of  more  or  less  value,  are  in  use, 
the  illustrations  in  Fig.  50  showing  fishway  recommended  by  the 
New  York  State  Conservation  Commission.  This  type  is  termed 
the  Improved  Coil  Fishway,  and  consists  of  a  niunber  of  compart- 
ments arranged  in  steps  and  separated  by  cross-partitions.  These 
are  provided  with  orifices,  alternating  from  side  to  side,  through 
which  the  fish  may  pass  from  compartment  to  compartment, 


Digitized  by 


Google 


100 


DAMS  AND  HEADWORKS 


LOHGITUDINAL  SECTION 


ELEVATION  OF  INTAKE 


Fig.  50.— Fishway. 

or  they  may  leap  over  the  cross-partitions,  according  to  their 
habit. 

4.  INTAKES 

Intakes  of  many  kinds  are  employed,  and  their  design  and 
location  is  to  a  great  extent  governed  by  local  conditions. 

Trash  Racks.  An  essential  feature  common  to  all  types  and 
which  has  a  bearing  on  the  economic  use  of  water,  is  the  trash 
rack  and  its  design.  These  racks  should  be  so  constructed  as 
to  give  sufficient  area  for  passing  the  desired  quantity  of  water 
without  excessive  loss  in  head.  This  is  especially  important  in 
low-head  developments,  where  large  quantities  of  water  are 
utilized.     Considerable  loss  of  efficiency  may  result  from  restricted 


Digitized  by 


Google 


INTAKES  101 

water  passages  through  racks,  and  in  the  design  allowance  should 
be  made  for  the  accumulation  of  trash  as  a  factor  in  the  restriction 
of  water  passage. 

Low-head  Installations.  With  low-head  plants  the  intake 
generally  forms  a  part  of  the  dam  or  power-house,  as  shown  in 
Fig.  26  imder  section  "  Low-head  Developments."  The  up- 
stream bay  comprises  the  gate  room,  and  by  thus  installing  the 
gates  and  screens  indoors,  there  is  less  danger  from  ice  forming 
therein  during  cold  weather.  In  certain  stations  arrangements 
are  also  made  whereby  the  heated  air  from  the  generators  can  be 
led  to  the  gatehouse  for  preventing  the  formation  of  ice. 

The  water  from  the  forebay  enters  the  gatehouse  through 
arches  in  the  front  wall,  and  by  submerging  these  below  the  low- 
water  level  certain  floating  material  will  be  prevented  from 
entering. 

High-head  Installations.  For  high-head  plants  the  intakes 
are  often  built  as  independent  structures,  and  where  overflow 
diversion  dams  are  used,  they  should  preferably  be  located  at  a 
right  angle  to  the  dam.  This  arrangement  has  several  advantages, 
among  which  are  the  ease  with  which  logs,  trees  and  other  float- 
ing debris  can  be  cleared  away  by  simply  opening  one  or  two  of 
the  nearest  flashboards. 

The  intake  shown  in  Figs.  51  and  52  represents  a  typical 
installation  of  the  latest  design.  It  is  a  caisson-like,  self-contained 
structure,  divided  by  partitions  into  five  sections  in  order  to  resist 
the  stresses  on  the  outside  walls  due  to  the  hydrostatic  pressure 
when  the  intake  is  empty  and  the  water  in  the  pond  is  at  its 
maximum  elevation.  At  the  rear  of  each  division  wall  there  is  an 
opening  which  allows  the  water  to  pass  to  the  timnel  entrance 
located  at  the  center  and  bottom  of  the  rear  wall. 

There  are  two  sets  of  racks,  a  coarse  set  consisting  of  |-inch 
round  iron  rods  spaced  4  inches  apart,  being  placed  in  front  of  the 
head  gates  to  prevent  large  debris  from  interfering  with  their 
operation.  In  addition,  there  is  a  fine  set  mounted  in  an  inclined 
position  in  each  of  the  intake  chambers.  These  racks  are  made 
of  4X|-inch  flat  iron  bars,  spaced  IJ  inches  apart.  They  are 
pro\aded  with  a  rack  cleaner,  each  rack  section  being  cleaned  by 
three  rakes  placed  in  a  staggered  position  and  operated  at  a 
speed  of  3  feet  per  minute  by  means  of  link  chains  from  a  motor- 
driven  coimtershaft   located   on  top  of  the  structure.    At  the 


Digitized  by 


Google 


l&l 


DAMS  AND  HEADWORKS 


FiQ.  Sl.^Tunnel  Intake,  Showing  Its  Relation  to  the  Diversion  Dam. 


Fig.  62. — Cross-section  of  Tunnel  Intake  Shown  in  Fig.  51,  Illustrating  Racks, 
Rack-cleaners  and  Gates. 


Digitized  by 


Google 


INTAKES  103 

top  of  each  bay  an  adjustable  iron  comb  catches  the  debris  col- 
lected by  the  rakes  and  drops  it  on  the  floor. 

Influeoce  of  Ice.  In  cold  climates,  where  it  is  impracticable 
to  reduce  the  entering  velocity  of  water  to  a  sufficient  extent  to 
allow  the  surface  to  freeze  over,  and  where  considerable  quan- 
tities of  anchor  or  frazil  ice  are  likely  to  be  swept  against  the  racks, 
adherence  to  the  racks  may  be  reduced  either  by  maintaining  the 
portion  of.  the  racks  above  the  water  surface  at  a  temperature 
above  freezing  by  housing  or  otherwise,  or  by  constructing  the 
exposed  portions  of  the  racks  of  wood,  concrete  or  other  non- 
conductors of  heat,  the  portion  below  the  water  being  of  steel. 
Electric  heaters  have  been  used  in  some  cases  for  the  piurpose  of 
preventing  clogging  of  the  racks,  and  it  has  also  been  proposed  to 
so  arrange  the  bars  composing  the  rack  that  a  low-voltage  electric 
current  may  be  sent  through  them  in  series,  thus  heating  them 
sufficiently  to  prevent  the  adherence  of  ice. 

In  order  to  prevent  trouble  from  ice  in  the  wheel  casings,  it  is 
essential  that  effective  waternseals  be  provided  in  the  tailrace 
discharge  to  prevent  the  entrance  of  cold  air. 

For  further  information  on  precautions  to  be  taken  against  ice 
troubles,  the  reader  is  refered  to  the  N.E.L.A.  Prime  Mover 
Committee's  Report  for  1917. 


Digitized  by 


Google 


CHAPTER  V 

WATER  CONDUCTORS  AND  ACCESSORIES 

1.  WATER  CONDUCTORS 

Classification  of  Water  Conductors.  As  in  the  case  of  dams, 
there  is  a  great  variety  of  types  of  water  conductors,  the  par- 
ticular kind  to  be  used  being  entirely  governed  by  the  nature  of 
the  development  as  well  as  by  economy.  Where  the  power-house 
is  located  near  the  dam,  there  may  be  no  need  for  conduits  at  all,  as 
in  low-head  plants,  or  they  may  simply  consist  of  very  short  pipes. 
For  medium-  and  high-head  developments,  however,  a  more 
elaborate  system  of  conduits  must  as  a  rule  be  provided,  as  the 
water  must  in  many  such  instances  be  diverted  for  miles,  before 
it  finally  reaches  the  power-house. 

The  different  kinds  of  water  conductors  in  general  use  may  be 
divided  into  two  classes,  open  or  closed,  the  closed  construction 
being  either  of  the  low-  or  high-pressure  type. 

Classification  op  Water  CoNDUcroBa 
Open. 

Canals:  lined  or  unlined. 
Flumes:  wood,  concrete  or  steel. 
Closed. 

Low-pressure. 
Tunnels. 

Pipe:  wood,  concrete  or  steel. 
High-pressure. 
Pipe:  steel. 

Open  canals  and  flumes  are  often  used  for  carrying  the  water 
from  the  point  of  diversion  to  the  beginning  of  the  pressure  lines. 
This  method  was  extensively  used  in  earUer  developments,  and 
while  it  may  in  many  cases  be  the  cheapest,  a  higher  efficiency  can 
be  obtained  by  a  closed  system  of  tunnels  and  pipes,  in  that  the 
total  head  will  be  greater.  Where  the  contour  of  the  country  is 
very  irregular  the  cost  of  excavating  for  canals  and  of  building 
high  trestles  for  the  flumes  may  be  very  high,  and  in  such  instances 

104 


Digitized  by 


Google 


WATER  CONDUCTOBfl 


105 


TABLE  XXVII 
Table  of  n  for  Kutter'b  Formxtla 


Surface. 


Perfect. 


Good. 


Fair. 


Had. 


pipe 


Uncoated  c.-i.  pipe 

Coated  c.-i.  pipe 

Commercial  w.-i.  pipe,  black 

Commercial  w.-i.  pipe,  galv 

Smooth  brass  and  glass  pipe 

Smooth  lockbar  and  welded  "  OD 
Riveted  and  spiral  steel  pipe 

Vitrified  sewer  pipe 

Glazed  brickwork 

Brick  in  cement  mortar;  brick  sewers. 

Neat  cement  surfaces 

Cement  mortar  surfaces 

Concrete  pipe 

Wood-etave  pipe 

Plank  Flumes: 

Planed 

Unplaned 

With  battens 

Concrete-lined  channels 

Cement-rubble  surface 

Dry-rubble  surface 

Dressed-ashlar  surface 

Semicircular  metal  flumes,  smooth. .  .  . 
Semicircular  metal  flumes,  corrugated. 
Canals  and  Ditches: 

Earth,  straight  and  uniform 

Rock  cuts,  smooth  and  uniform .... 

Rock  cuts,  jag^d  and  irregular.  .  . . 

Winding  sluggish  canals 

Dredgea  earth  channels 

Canals  with  rough  stony  beds,  weeds 
on  earth  banks 

Earth  bottom,  rubble  sides . 
Natural  Stream  Channels: 

(1)  Clean,  straight  bank,  full  stage,  no 
rifts  or  deep  pools 

(2)  Same  as  (1),  but  some  weeds  and 
stones 

(3)  Winding,  some  pools  and  shoals, 
dean 

(4)  Same  as  (3),  lower  stages,  more  in- 
effective slope  and  sections 

(5)  Same  as  (3).  some  weeds  and  stones 

(6)  Same  as  (4),  stony  sections 

(7)  Sluggish     nver     reaches,     rather 
weedy  or  with  very  deep  pools 

(8)  Very  weedy  reaches 


0.012 

0.011 

0.012 

0.013 

0.009 

0.010 

0.013 

0.010  \ 

0.011/ 

0.011 

0.012 

0.010 

0.011 

0.012 

0.010 

0.010 
0.011 
0.012 
0.012 
0.017 
0.025 
0.013 
0.011 
0.0225 

0.017 

0.025 

0.035 

0.0225 

0.025 

0.025 
0.028 


0.025 

0.030 

0.035 

0.040 
0.033 
0.045 

0.050 
0.075 


0.013 

0.012* 

0.013 

0.014 

0.010 

0.011* 

0.016* 

0.013* 

0.012 
0.013 
0.011 
0.012 
0.013 
0.011 

0.012* 

0.013* 

0.015* 

0.014* 

0.020 

0.030 

0.014 

0.012 

0.025 


0.014 

0.013* 

0.014 

0.015 

0.011 

0.013* 

0.017* 

0.015 

0.013* 

0.015* 

0.012 

0.013* 

0.015 

0.012 

0.013 

0.014 

0.016 

0.016* 

0.025 

0.033 

0.015 

0.013 

0.0275 


0.020 

0.030 

0.040 

0.025* 

0.0275* 

0.0225* 

0.033* 

0.045 

0.0275 

0.030 

0.030 
0.030* 

0.035* 
0.033* 

0.0275 

0.030 

0.033 

0.035 

0.040 

0.045 

0.045 
0.035 
0.050 

0.050 
0.040 
0.055 

0.060 
0.100 

0.070 
0.126 

0.015 

0.015 
0.017 
0.013 


0.017 

0.015 
0.017 
0.013 
0.016 
0.016 
0.013 

0.014 
0.016 

0.018 
0.030 
0.036 
0.017 
0.016 
0.030 

0.026 
0.036 

0.030 
0.033 

0.040 
0.036 


0.033 

0.040 

0.050 

0.056 
0.045 
0.060 

0.080 
0.160 


*  Values  commonly  uaed  in  deaigninc. 


Digitized  by  CjOOQ IC 


106  WATER  CX)NDUCTORS  AND  ACCESSORIES 

the  closed  construction  generally  becomes  more  economical,  in 
that  tmmels  may  be  built  and  the  pipes  follow  more  or  less  the 
contour  of  the  country.  The  selection  of  the  particular  type  of 
conduit  construction  is,  therefore,  an  engineering  problem  of 
considerable  importance,  and  has  to  do  with  the  economic  oper- 
ating features  of  the  development. 

Canals.  The  velocity  of  water  in  a  canal  is  affected  by  the 
roughness  of  the  bed,  by  the  wetted  surface  of  the  form  of  the 
crossHsection,  and  finally  by  the  grade.  According  to  Chesy's 
formula  it  is  equal  to: 

where   i;= velocity  in  feet  per  second  ; 
c»  coefficient; 
r»hydraulic  radius  in  feet; 
«» grade  or  hydraulic  slope, 

the  values  of  c  may  be  obtained  from  the  following  two  formula, 
both  of  which  are  in.  common  use. 
Kutter's  formula: 

n  8 


where  n  is  the  coefficient  of  roughness,  the  values  of  which  are 
given  in  table  XXVIL* 

Bazin's  Formula: 

87 
c= . 

0.662+^ 

Vr 

TABLE  XXVIII 

Values  of  m 

Smooth  cement  and  planed  boards 0.06 

Planks  and  bricks 0. 16 

Rubble  masonry 0.46 

Earth  canab  in  excellent  condition 0.85 

Earth  canab  in  fair  condition 1 .30 

Earth  canab  in  bad  condition 1 .75 

*  R.  E.  Horton,  "  Engineering  News,"  February  24,  1916. 


Digitized  by 


Google 


WATER  CONDUCTORS  107 

on.   u  J      1-       J-  Area  of  Croes-fiection  .,         , .    , 

The  hydraubc  radius,  r=   „^    -  ,  ^    . — i — ,  the  wetted  per- 

Wetted  Perimeter  "^ 

imeter  of  the  croes-Bection  of  a  channel  being  that  part  which  is 

in  contact  with  the  water. 

For  an  open  canal,  the  grade  or  slope,  a,  is  the  ratio  of  the  fall  to 
the  length  in  which  the  fall  occurs.  For  a  closed  penstock  under 
pressure,  it  is  the  ratio  between  the  loss  in  head  due  to  friction 
to  the  length.     (See  also  pag^  117.) 

The  velocity  of  the  water  in  a  canal  should  be  kept  below  that 
which  would  cause  erosion  of  the  bed.  It  should,  however,  be 
large  enough  to  prevent  vegetable  growth  from  forming  or  silt 
from  being  deposited.  Assuming  the  bottom  velocity  to  be 
about  75  per  cent  of  the  mean  velocity,  the  figures  in  Table  XXIX 
represent  the  safe  values  which  are  widely  used  in  determining  the 
permissible  velocities  of  water  in  open  canals. 

TABLE  XXIX 
Safb  Mean  Velocitibs  * 

Very  fine  sandy  soil  or  loose  silt 0.50 

Pure  sand. 1.00 

light  sandy  soil,  15  per  cent  clay 1 .20 

light  sandy  loam,  40  per  cent  clay 1 .80-2.00 

Coarse  sand 1.50-2. 00 

Loose  gravelly  soil 2.60 

Ordinary  loam 2.60 

Ordinary  firm  soil  or  loam,  65  per  cent  clay. ...  3 .00 

Stiff  clay  loam 4.00 

Firm  gravelly  clay  soil 5.00-7.00 

Stiff  day 6.00 

Conglomerates,  soft  slate 6.60 

Stratified  rocks 8.00 

Small  botdders 8.00-15.00 

Hard  rock 13.33 

Concrete 16.00-20.00 

•  B.  A.  Etcheverry,  "  Journal  of  Electricity »  Power  and  Gas." 

The  most  advantageous  cross-section  to  use,  from  the  hydraulic 
point  of  view,  would  be  that  which  gives  the  smallest  wetted  per- 
imeter or  the  largest  value  of  the  hydraulic  radius.  This  would 
meari  a  semicircular  section,  but  it  is' seldom  used  on  account  of 
the  difficulties  in  building.  A  trapesoidal  section  is,  however, 
generally  used,  and  by  letting  the  bottom  and  sides  be  tangents 
to  an  inscribed  semicircle,  as  in  Fig.  53,  the  best  hydrauUc  results 
will  be  obtained;  the  slope,  i.e.,  the  angle  6,  being  60^ 


Digitized  by 


Google 


108 


WATER  CONDUCTORS  AND  ACCESSORIES 


The  ideal  croesHsection  from  the  hydraulic  point  of  view  is, 
however,  not  always  the  best  to  adopt.  There  are  other  factors 
which  must  be  considered,  such   as  the  cost   of   construetion, 

whether  lined  or  unlined, 
the  character  of  the  soil, 
seepage,  safety,  grade,  and 
velocity.  No  specific  rules 
can  be  laid  down  to  cover 
all  cases  and  each  installa- 
tion must  be  treated  indi- 
vidually. 

A  concrete-lined  canal  having  the  least  wetted  perimeter  will 
require  the  smallest  amoxmt  of  material,  while  the  steeper  sides 
mean  less  excavation.  Such  a  canal  can  furthermore  be  given  a 
steeper  grade,  if  sufficient  fall  is  available,  and  thus  a  higher 
velocity,  so  that  the  cross-section  can  be  small  for  a  given  quan- 


Fia.  53. — Cross-section  of  Canal. 


ITola:- 


Approx.  4}^  en.  ft.  of.o 
per  fool  leairth  of  ditdu 
Bnrf »ee  of  concrete  on  Inaide  of 
diteh  to  be  made  frnooth 


Fig.  54. — Open  Concrete-lined  Canal. 

tity  of  water.  This  fe  advantageous  especially  on  hillsides,  and,  if 
the  soil  is  hard  and  the  excavation  difficult,  a  concrete-lined  canal 
may  be  cheaper  than  an  unlined  one.  In  other  instances  the  sofl 
may  be  of  such  a  porous  nature  that  lining  is  essential  to  prevent 
excessive  seepage.     (See  Figs.  64  and  55.) 

Note.— For  "  Flow  of  Water  in  Channeb,"  see  Bulletin  No.  194  U.  S. 
Dept.  of  Agriculture. 


Digitized  by 


Google 


WATER  C0NDUC5S0HS 


109. 


f" 


Fig.  55. — Concrete-lined  Canal. 


From  the  standpoint  of  safety  a  shallow  canal  is  better  than  a 
deep  one.  The  pressure  on  the  banks  increases  with  the  depth  of  water 
and  may  cause  breaks,  especially  where  canals  are  built  on  side 
hills,  and  where  the  banks  may  have  been  weakened  due  to  erosion. 

The  slopes  should,  therefore,  in  the  first  place  be  such  that 
they  will  withstand  such  erosion  of  the  water,  the  values  given  in 
Table  XXX  being  representative  of  actual  practice. 

TABLE  XXX 
Side  Slopes 


Solid  rock  or  cement 

Hardpan  and  very  firm  soil 

Ordinary  firm  soil 

Ordinary  sandy  loam 

Loose  sandy  soil 


Vertical. 


Digitized  by 


Google 


no  WATER  CONDUC?rORS  AND  ACCESSORIES 

Evaporation  is  small  as  compared  with  seepage,  which  increases 
with  the  depth  of  the  water  and  with  the  wetted  perimeter,  but 
decreases  with  an  increase  in  velocity.  While  evaporation,  there- 
fore, can  be  neglected,  the  e£fect  of  seepage  must  usually  be  con- 
sidered in  determining  the  capacity  of  a  canal. 

The  velocity  decreases  with  an  increase  in  the  wetted  perim- 
eter, and  when  the  fall  is  great  it  may  be  advisable  to  use  a 
shallower  section  to  reduce  the  velocity,  or  vice  versa.  If  the 
actual  slope  of  the  country  is  so  great  that  the  corresponding 
velocities  would  cause  erosion,  it  is  necessary  to  limit  the  grade  to 
a  value  which  would  not  give  an  excessive  velocity,  and  to  con- 
centrate the  excess  fall  at  suitable  drops  along  the  canal. 

Flumes.  Where  the  contour  of  the  country  is  very  irregular 
or  the  soil  very  hard  and  difficult  to  excavate,  flumes  are  some- 
times used  for  diverting  the  water.  While  the  first  cost  of  such 
structures  may  be  very  low  where  timber  is  cheap,  their  upkeep  is, 
however,  usually  much  higher  than  for  a  canal,  and  every  pre- 
caution must,  therefore,  be  taken  in  their  design  and  construction. 

The  velocity  of  the  water,  which  can  be  foimd  from  the  for- 
mulae given  in  the  previous  section,  may  be  much  higher  than  for 
unlined  canals,  and  the  higher  the  velocity  the  smaller  cross-section 
is  required.  When  the  water,  therefore,  enters  a  flume  from  a 
canal,  it  becomes  necessary  to  provide  a  sufficient  drop  in  the  upper 
end  of  the  fliune  for  the  increased  velocity  head.  This  may  be 
found  from  the  formula: 

wbere 

/I = drop  necessary  to  increase  the  velocity  in  feet; 
v\ » velocity  of  flow  in  flume  in  feet  per  second; 
V2  =  velocity  of  flow  in  canal  in  feet  per  second; 
(7= acceleration  of  gravity =32. 16. 

Similarly  there  should  be  a  gain  in  head  when  the  water  again 
enters  a  canal  from  a  flume,  although  this  is  not  reaUzed  to  a  very 
great  extent  and  can  be  neglected. 

Flumes  may  be  classified  according  to  the  material  of  which 
they  are  built,  into: 

Rectangular  wooden  flumes. 
Semi-circular  wood-stave  flumes. 


Digitized  by 


Google 


WATER  CX)NDUCTORS 


111 


Reinforced  concrete  flumes. 
Steel. 

Also,-  according  to  their  general  design,  into  bench  flumes  and 
trestle  flumes. 

A  t3rpical  design  of  a  rectangular  wooden  flume  of  the  bench 
t3rpe  is  shown  in  Fig.  56,  the  width  being  from  1^  to  2  times  the 
depth  of  the  water.  The  illustration  clearly  shows  the  detail  of 
construction  and  this  type  is  used  on  hillsides  or  places  where  it 
may  be  located  directly  on  the  ground.  When  crossing  depressions 
it  is  supported  on  trestles.  Careful  consideration  must  be  given 
to  the  construction  of  the  foundations,  and  precautions  taken  so 


alk  Bowd  iW'x  U'z  le'o" 


C*p2"x8"xlf0" 


BBoanb Ik's  111 


J^xY'Batten/ 


I  Bottom  Boards |  li^'x  IS'x  liVi 


Sill  e"i 


*p-joloted)r 


Fig.  66. — ^Rectangular  Wooden  Flume. 

that  floods  will  not  undermine  the  same.  Drains  should,  there- 
fore, be  provided  if  there  is  any  such  danger.  Spillways  for  dis- 
charging any  overflow  should  also  be  installed  at  points  where 
the  water  can  be  readily  disposed  of.  This  refers  to  canals  as 
well  as  flumes. 

Fig.  57  shows  the  design  of  a  semicircular  woodnstave  flume. 
This  section  is,  as  before  stated,  very  advantageous  from  the 
hydrauUc  point  of  view.  It  is  easily  adjusted  to  curves,  and  it 
can  be  kept  water-tight  by  screwing  up  the  nuts  above  the  tie- 
beams  at  the  ends  of  each  threaded  band. 

Reinforced  concrete  flumes  have  been  used  in  some  installations 
of  late,  Fig.  58  showing  such  a  design.     While  the  first  cost  is 


Digitized  by 


Google 


112 


WATER  CONDUCTORS  AND  ACCESSORIES 


usually  much  higher  than  that  of  a  wooden  flume,  its  life  is  so 
much  longer  and  the  maintenance  cost  so  much  lower,  that  it  may 
prove  more  economical  in  the  long  run.  There  is  further  one 
advantage  of  such  flumes  and  that  is  the  omission  of  crosspieces 
over  the  top,  which  makes  cleaning  very  easy.  This  point  should 
be  carefully  considered  for  waters  which  are  prolific  in  moss  and 
vegetable  matters. 

Steel  flumes  are  generally  semicircular  in  crossHsection,  similar 
to  Fig.  67.    There  are  several  makes  of  such  flumes  but  their 


7  X  14  X  SS 


r    11  ■£  Wr     stringer* 


ThoM.' ., _  -  ,  , 


Fig.  57. — Semi-Circular  Wood-Stave  Flume. 


V8q.lMBtB«4 

Fig.  58. — Concrete  Flimie. 

construction  does  not  differ  materially.  They  consist  of  curved 
metal  sheets  with  a  bead  or  corrugated  groove  rolled  in  each  edge 
of  the  sheet.  The  sheets  are  put  together  by  means  of  an  inter- 
locking joint  formed  by  overlapping  the  edges,  which  fit  over 
each  other.  The  joint  is  made  tight  by  means  of  a  curved  rod 
which  fits  on  the  outside  of  the  corrugated  groove  and  a  curved 
beveled  bar  or  small  channel  on  the  inside.  The  steel  rods  carry 
the  weight  of  the  flume,  and  their  ends  are  threaded  for  nuts 
and  pass  through  a  carrier  or  tie-beam  which  is  supported  on 
stringers  about  16  feet  long. 


Digitized  by 


Google 


WATER  CONDUCTORS  113 

As  the  use  of  flumes  becomes  less  and  less  as  hydro-electric 
work  becomes  more  permanent  in  character,  it  is  suggested  that 
for  preliminary  estimating  purposes  the  cost  of  low-pressure  pipe 
lines  be  used  instead  of  using  the  presumably  lower  cost  of  flume 
construction. 

Tunnels.  Where  the  proposed  route  of  the  waterway  encoun- 
ters mountain  ridges  it  is  often  advantageous  if  not  absolutely 
necessary  to  go  through  these  by  means  of  tunnels  rather  than  to 


Are*  of  waterway  IBI  aq.  ft. 

W»t(«d  pmimetor  45  ft. 

Hydimalla  hmIIm  SLS 

FMetioB  ooaffielant  .014 

Ormdo  .00* 

C.  130 

Fia.  59. — Typical  Tunnel  Section. 

excavate  deep  cuts  or  go  around.  The  question  as  to  which  method 
should  be  chosen  is  one  of  first  cost  as  well  as  of  maintensmce. 
Tunnels  are,  of  course,  safer  and  their  upkeep  is  usually  low  as 
compared  with  open  canals,  especially  if  these  are  built  on  the 
hillsides  where  they  are  exposed  to  dangers  from  boulders  striking 
them,  undermining,  etc. 

Tunnels  may  be  either  of  the  pressure  or  non-pressure  type. 
When  of  considerable  length  they  are  usually  of  the  former  type 
so  that  the  drop  may  be  utilized  as  useful  head.  They  are  almost 
always  lined  with  concrete,  the  thickness  of  the  lining  varying  from 


Digitized  by 


Google 


114  WATER  CONDUCTORS  AND  ACCESSORIES 

4  to  12  inches  depending  on  the  grade  and  the  pressure  of  the  water. 
A  lining  serves  several  purposes.  It  holds  the  rocky  material  in 
place;  it  prevents  seepage  if  the  rock  is  porous;  and  finally  it  de- 
creases the  friction  which  is  of  greatest  importance  in  tunnel  work, 
as  it  permits  a  higher  velocity  with  a  correspondingly  reduced  sec- 
tion. The  velocity  may  be  obtained  from  Kutter's  formula,  and  the 
values  for  n  may  be  taken  as  0.014  for  lined  tunnels  and  0.028 
for  unlined.    The  safe  velocity  is  from  10  to  15  feet  per  second. 

While  the  circular  cross-section  would  be  most  advantageous 
from  the  hydraulic  point  of  view,  it  is  usually  given  a  horseshoe 
shape  (see  Fig.  59)  as  this  has  been  found  to  be  the  easiest  to 
excavate.  In  order  to  permit  quick  construction,  especially  of 
long  timnels,  one  or  more  adits  or  openings  are  usually  provided 
at  certain  intervals  so  that  the  work  can  proceed  from  several 
headings  at  the  same  time. 

Pipe  Lines.  Pressure  pipes  must  be  used  for  conveying  the 
water  from  the  upper  level  at  the  forebay  or  dam  to  the  wheels  at 
the  power-house.  These  may  be  constructed  of  steel,  wood,  and 
sometimes,  although  rarely,  of  concrete.  The  particular  kind 
to  use  depends  upon  the  head  and  the  corresponding  pressiue. 

Head,  The  total  or  gross  head,  as  ordinarily  understood, 
is  the  difference  between  the  elevation  of  the  water  in  the  fore- 
bay  and  the  tailrace.  It  must  be  distinguished  from  the  net  or 
effective  head  acting  on  the  turbine,  the  difference  between  the 
two  being  equal  to  the  head  lost  on  account  of  friction  in  the 
penstock,  etc. 

The  net  or  effective  head  at  any  point  on  the  pipe  line  is 
equal  to  the  sum  of  the  pressure  head  at  the  point  considered, 
plus  the  elevation  head  at  the  point  above  a  datum  plane  plus 
the  velocity  head  in  the  pipe.  Thus 

where 

A = effective  or  net  head  in  feet; 

p  =  pressure  head,  this  being  equal  to  the  pressure  in  pounds 

per  square  foot,  at  the  point  in  consideration,  divided 

by  62.4; 
2  =  the  elevation  of  the  point  above  any  arbitrary  datum 

plane,  in  feet; 
V  =  velocity  at  the  point  in  feet  per  second. 


Digitized  by 


Google 


WATER  CONDUCTORS  115 

The  efifective  head  at  one  point  in  a  pipe  will  differ  from 
that  at  another  point  upstream  or  downstream  from  it,  by  an 
amomit  corresponding  to  the  losses  and,  of  course,  to  any  work 
done  or  received  between  the  two  points  when  a  machine,  such 
as  a  turbine  or  pump,  is  placed  in  the  pipe  line.  Considering 
only  the  losses,  it  follows  that  the  effective  head  must  decrease 
in  the  direction  of  the  flow  by  an  amount  equal  to  the  head  lost. 
Therefore,  although  either  the  pressure,  elevation,  or  velocity 
may  increase  in  the  direction  of  the  flow,  the  sum  of  them  must 
continually  decrease  so  that  an  increase  in  one  of  these  items 
must  always  be  accompanied  by  a  corersponding  decrease  in 
on  3  or  both  of  the  others. 

In  regard  to  the  head  to  be  used  in  computing  the  efficiency 
of  an  installation  or  a  turbine,  the  turbine  testing  code  of  the 
turbine  builders  specifies  the  following: 

"For  the  purpose  of  computing  the  plant  efficiency  the 
total  or  gross  head  acting  on  the  plant  is  to  be  used,  and  is  to 
be  taken  as  the  difference  in  elevation  between  the  equivalent 
still-water  siuf ace  before  the  water  has  passed  through  the  racks, 
to  the  equivalent  still-water  surface  in  the  tailrace  after  dis- 
charge from  the  draft  tube.  When  the  water  in  the  forebay  in 
advance  of  the  racks  flows  with  sufficient  velocity  to  make  its 
velocity  head  an  appreciable  quantity,  the  actual  elevation 
of  the  water  surface  shall  be  increased  by  the  amount  of  this 
velocity  head.  The  same  process  shall  apply  to  the  point  of  meas- 
urement in  the  tailrace;  that  is,  the  velocity  head  at  the  point 
of  measurement  in  the  tailrace  shaU  be  added  to  the  actual  ele- 
vation of  the  surface,  the  sum  being  considered  the  equivalent 
still-water  elevation. 

"  In  computing  the  efficiency  of  the  turbine,  the  losses  through 
racks,  in  the  intake  to  the  penstocks,  and  in  the  penstocks  shall 
not  be  charged  against  the  turbine;  nor  shall  the  head  necessary 
to  set  up  the  velocity  required  to  discharge  the  water  from  the 
end  of  the  draft  tube  be  charged  against  the  tiurbine. 

"  The  net  or  effective  head  acting  on  turbines  equipped  with 
casings  is  to  be  taken  as  the  difference  between  the  elevation 
corresponding  to  the  pressure  in  the  penstock  near  the  entrance 
to  the  turbine  casing,  and  the  elevation  of  the  tail  water  at  the 
highest  point  attained  by  the  discharge  from  the  unit  under  test, 
the  above  difference  being  corrected  by  adding  the  velocity 


Digitized  by 


Google 


116  WATER  CONDUCTORS  AND  ACCESSORIES 

head  in  the  penstock  at  the  point  of  measurement  and  subtract- 
ing the  residual  velocity  head  at  the  end  of  the  draft  tube.  The 
velocity  head  in  the  penstock  shall  be  taken  as  the  square  of  the 
mean  velocity  at  the  point  of  measurement,  divided  by  2g;  the 
mean  velocity  being  equal  to  the  quantity  of  water  flowing  in 
cubic  feet  per  second,  divided  by  the  cross-Bectional  area  of  the 
penstock  at  the  point  of  measurement  in  square  feet.  The  residual 
velocity  head  at  the  end  of  the  draft  tube  shall  be  taken  as  the 
square  of  the  mesm  velocity  at  the  end  of  the  draft  tube,  divided 
by  2g,  the  mean  velocity  being  equal  to  the  quantity  flowing 
in  cubic  feet  per  second,  divided  by  the  final  cross-sectional 
discharge  area  of  the  closed  or  submerged  portion  of  the  draft 
tube  in  square  feet." 

The  loss  of  head  is  due  to  the  loss  in  the  entrance  of  the  pen- 
stock, to  the  friction  of  the  interior  surface,  to  curvature,  and  to 
various  other  obstructions  such  as  headgates,  racks,  and  valves. 
In  the  case  of  impulse  turbines,  there  is  a  further  loss  caused  by 
the  necessity  of  placing  the  wheel  clear  of  tailwater  so  that  after 
leaving  the  wheel  the  water  drops  freely  through  the  vertical 
height  between  the  wheel  and  the  tailwater  surface,  and  fails  to 
utilize  the  head  corresponding  to  this  free  fall.  It  is  customary 
in  computing  the  efiiciency  of  impulse  turbines  to  charge  against 
the  wheel  only  the  net  head  with  reference  to  the  elevation  of 
the  center  of  the  nozzle  taken  as  datum. 

Loss  of  Head  in  Entrance.  This  loss  of  head  is  probably 
due  to  internal  friction  of  the  particles  of  water  against  each 
other  when  they  converge  towards  the  contracted  entrance.  The 
loss  depends  on  the  shape  of  the  intake,  but  for  ordinary  purposes 
it  may  be  obtained  from  the  formula 

Loss  of  Head  in  Friction.    For  determining  the  loss  of  fric- 
tion in  pipe  lines  there  are  two  formulas  in  very  general  use: 
Chezy  formula: 

V = cVrs    (for  values  of  c  see  page  106). 

Williams  and  Hazen  formula: 
t;=1.32      cr"" 


Digitized  by 


Google 


WATER  CONDUCTORS  117 

where 

»= velocity  in  feet  per  second; 

r = hydraulic  radius = -  for  circular  pipes,  d  being  the  diameter 

in  feet; 

hf 
«= hydraulic  slope =y,  where  A/ represents  the  loss  in  head 

due  to  friction  and  I  the  length  of  pipe,  both  in  feet; 
c= friction  coefficient. 

In  using  the  latter  (Williams  ond  Hazen)  formula,  the  folloT^ing 
values  of  the  friction  coefficient  are  recommended: 

For  cast-iron  pipe c  =  120-110 

For  riveted  steel  pipe c  =  105-100 

For  wood-stave  pipe c  =  130-120 

To  facilitate  the  calculations  when  using  their  formula, 
Williflms  and  Hazen  have  published  a  book  entitled  "  HydrauUc 
Tables,"  which  contains  a  series  of  tables  giving  the  values  of 
friction  losses  for  pipes  of  different  materials  and  sizes,  and  also 
different  degrees  of  roughness  and  for  various  velocities.  This 
book  is  very  useful,  and  may  be  obtained  from  John  Wiley  & 
Sons,  Inc. 

Merriam  in  his  "  Treatise  on  HydrauUcs  "  states  the  following 
in  regard  to  the  friction  loss: 

1.  The  loss  of  head  in  friction  is  directly  proportional  to  the 
length  of  the  pipe. 

2.  It  is  inversely  proportional  to  the  diameter  of  the  pipe. 

3.  It  increases  nearly  as  the  square  of  the  velocity. 

4.  It  is  independent  of  the  pressure  of  the  water. 

5.  It  increases  with  the  roughness  of  the  interior  surface. 
Thus 

The  friction  factor,  /,  depends  upon  the  degree  of  roughness 
of  the  surface,  the  values  given  in  Table  XXXI  being  apphcable 
to  clean  cast-iron  or  wrought-iron  pipes. 


Digitized  by 


Google 


118 


WATER  CONDUCTORS  AND  ACCESSORIES 


TABLE  XXXI 
Fbiction  Factors  for  Clean  Iron  Pipes 


Diameter 

V«LOCITT 

m  Fbbt  p«b  Second. 

in 
Feet. 

1 

2 

3 

4 

6 

10 

15 

0.05 

0.047 

0.041 

0.037 

0.034 

0.031 

0.029 

0.028 

0.1 

0.038 

0.032 

0.030 

0.028 

0.026 

0.024 

0.023 

0.25 

0.032 

0.028 

0.026 

0.025 

0.024 

0.022 

0.021 

0.5 

0.028 

0.026 

0.025 

0.023 

0.022 

0.020 

0.019 

0.75 

0.026 

0.025 

0.024 

0.022 

0.021 

0.019 

0.018 

1. 

0.025 

0.024 

0.023 

0.022 

0.020 

0.018 

0.017 

1.25 

0.024 

0.023 

0.022 

0.021 

0.019 

0.017 

0.016 

1.5 

0.023 

0.022 

0.021 

0.020 

0.018 

0.016 

0.015 

1.75 

0.022 

0.021 

0.020 

0.018 

0.017 

0.016 

0.014 

2. 

0.021 

0.020 

0.019 

0.017 

0.016 

0.014 

0.013 

2.5 

0.020 

0.019 

0.018 

0.016 

0.015 

0.013 

0.012 

3. 

0.019 

0.018 

0.017 

0.015 

0.014 

0.013 

0.012 

3.5 

0.018 

0.017 

0.016 

0.014 

0.013 

0.012 

4. 

0.017 

0.016 

0.015 

0.013 

0.012 

0.011 

5. 

0.016 

0.015 

0.014 

0.013 

0.012 

6. 

0.015 

0.014 

0.013 

0.012 

0.011 

Table  XXXII  *  gives  the  loss  in  head  in  each  100  feet  of 
riveted  steel  pipe  for  diameters  from  2  to  12  feet  and  for  veloci- 
ties up  to  12  feet  per  second. 

Loss  of  Head  in  Bends.  This  may  be  obtained  from  the 
formula: 

where  /i  is  the  curve  factor.  The  values  for  the  same,  given  in 
the  following,  were  determined  by  Williams,  Hubbell  and  Fen- 
kell  by  experiments  made  on  a  30-inch  cast-iron  water  main, 
with  90  deg.  bends. 

Let  R  be  the  radius  of  the  circle  in  which  the  center  line  of 
the  pipe  is  laid  and  d  the  diameter,  then: 


For  1=  24 

16 

10 

6 

4 

2.4 

/i= 0.036 

0.037 

0.047 

0.060 

0.062 

0.072 

^S.  Morgan  Smith  Co.'s  Bulletin  No.  104. 


Digitized  by 


Google 


WATER  CONDUCTORS 


119 


Hydraulic  Gradient.  The  hydraulic  gradient  is,  strictly 
speaking,  a  line  representing  atmospheric  pressure  conditions, 
although  it  may  also  conveniently  be  used  as  a  graphical  repre- 
sentation of  the  internal  pressures  in  a  pipe  line  at  any  point. 
It  may  also  be  defined  as  the  line,  the  vertical  distance  between 
which  and  the  center  of  the  pipe  gives  the  pressure  heads  at 
the  respective  points.  For  example,  referring  to  Fig.  59A,  the 
hydraulic  gradient  or  grade  line  is  a  line  through  the  points  to 


Pig.  59a. — Hydraulic  Gradient. 

which  the  water  levels  would  rise  if  piezometer  tubes  were  inserted 
along  the  pipe,  as  shown.  The  line  will  be  approximately  straight 
when  the  head  is  lost  uniformly  along  the  pipe,  that  is,  if  the 
size  and  surface  of  the  entire  length  of  pipe  is  the  same. 

The  grade  line  should  be  drawn  from  a  point  A  near  the  upper 
water-level,  the  distance  AB  being  equal  to  the  velocity  head 
plus  the  entrance  head,  to  a  point  at  the  end  of  the  pipe.  For 
a  pipe  discharging  freely  in  the  air  this  would  be  the  center  of 


Digitized  by 


Google 


120  WATER  CX)NDUCTORS  AND  A(XESSORIES 

its  outlet,  but  for  a  pipe  with  submerged  discharge  it  would  be 
the  lower  water  level  instead  of  the  point  of  discharge. 

The  slope  or  drop  in  elevation  along  the  pipe  corresponds 
to  the  friction  loss,  so  that,  for  example,  the  vertical  distance 
between  D  and  E  would  be  equal  to  the  head  lost  on  account  of 
friction  between  these  two  points. 

If  the  pipe  is  laid  so  that  it  rises  above  the  hydraulic  gradi- 
ent ACj  as  at  F,  the  pr^ure  in  the  pipe  at  this  point  will 
be  less  than  that  of  the  atmosphere  by  a  head  corresponding  to 
FG;  thus  negative.  If  no  air  could  enter  the  pipe  it  would  act 
as  a  siphon  and  the  flow  would  continue  as  usual,  provided  the 
distance  FG-  did  not  exceed  about  25  feet,  the  theoretical  limit 
of  vacuum  being  34  feet. 

Air  is,  however,  always  present  in  the  water  and  will  collect 
at  the  smnmit  near  F  and  the  pressure  will  approach  atmospheric, 
in  which  case  the  gradient  would  shift  to  AF  and  the  discharge 
would  only  be  that  due  to  the  vertical  head  between  B  and  F 
instead  of  between  B  and  C.  The  remainder  of  the  pipe  from  F 
to  C  would  merely  act  as  a  channel  to  deliver  the  flow. 

From  the  above  it  is  evident  that  the  pipe  line  should  be  laid 
well  below  the  hydraulic  gradient,  and  much  trouble  may  be 
avoided,  if  from  the  outset  a  profile  of  the  proposed  route  is 
prepared  and  the  hydraulic  gradient  carefully  calculated  and 
drawn  in. 

Size  of  Pipe  Line:  In  determining  the  size  of  a  pipe  line  or 
penstock  the  first  thing  to  consider  is  the  number  of  pipes  and 
necessarily  also  the  amoimt  of  water  which  each  must  be  able  to 
carry.  As  to  the  number,  this  should  preferably  be  equal  to  the 
turbine  units,  as  this  secures  a  greater  flexibility  in  the  operation 
of  the  plant.  It  further  does  away  with  the  large  Y-distributing 
joints  at  the  bottom  of  the  penstocks,  as  well  as  with  large  size 
gate  valves  and  heavy  plate  thicknesses. 

In  determining  the  most  economical  pip>e-line  installation  for  a 
hydro-electric  plant,  several  factors  in  addition  to  the  primary 
consideration  of  the  grade  or  route  must  be  studied.  In  general, 
these  must  have  direct  relation  to  the  earning  capacity  with 
resp)ect  to  the  first  cost.  Usually  the  pip)e-line  investment  repre- 
sents one  of  the  principal  items  of  the  initial  cost  of  the  generating 
station.  Elspecially  is  this  apparent  in  connection  with  those 
installations  where  the  pij^e  line  is  long  and  subject  to  high  pres- 


Digitized  by 


Google 


WATER  CONDUCTORS  121 

sure.  Because  of  its  initial  high  relative  cost  and  consequent 
interest  charge,  a  careful  consideration  of  the  pipe  line  must  be 
made;  otherwise,  an  injudicious  monetary  expenditure  may  result. 

It  is  obvious  that  for  a  given  water  quantity,  the  size  of  the 
pipe  is  determined  by  the  velocity  at  which  the  water  is  allowed 
to  run.  This  is  the  difficult  point  to  settle,  and  varies  anywhere 
from  6  to  12  feet  per  second,  the  average  probably  being  around  9 
feet.  A  high  velocity  entails  a  considerable  friction  loss,  while  a 
low  velocity  necessitates  a  larger  pipe  and  thus  increases  the  cost 
of  construction.  For  a  low-head  development  a  rather  low  veloc- 
ity should  be  used,  because  the  loss  of  head  will  then  form  a  much 
larger  percentage  of  the  total  head  than  where  a  high  head  is 
available.  In  high-head  pipe  lines  of  some  length  it  is,  of  course, 
also  more  economical  to  use  smaller  diameter  and  larger  velocity 
at  the  bottom,  where  the  pressure  is  higher  and  thicker  pipe  is 
required.  1 

Consideration  must  also  be  given  to  the  load  factor  at  which 
the  turbine  is  running,  i.e.,  the  average  amount  of  water  which  the 
pipe  line  is  to  carry.  Some  plants  require  that  the  turbines  are 
run  continuously  at  full  gate  opening,  while  in  other  instances 
they  may  operate  normally  at  half  gate,  only  opening  up  occa- 
sionally to  full  gate  to  take  care  of  momentary  peak  loads.  In 
such  a  case  the  friction  loss  should  naturally  be  based  on  the  water 
conveyed  when  the  wheels  are  operating  at  half  gate  op>ening. 

Theoretically,  therefore,  the  economical  diameter  of  a  pipe 
line  for  a  water-power  development  should  be  such  that  any 
increase  in  the  diameter  of  the  pip)e  would  cost  more  than  the 
value  of  the  power  which  could  be  obtained  from  the  decrease  in 
loss  of  head  due  to  friction  from  such  increase  in  diameter.  Or, 
stated  in  other  words;  the  size  of  pipe  should  be  such  that  the 
value  of  the  power  annually  lost  in  friction  plus  the  annual  interest, 
profit  and  depreciation  charges  on  the  pip)e  line  should  be  a  min- 
imum.   For  a  steel  pipe  this  leads  to  the  following  formula:^ 


^4 


\S20XYXX^X(^Xe 


Where  '    if^XtXmXiXc^ 

d= economic  diameter  in  feet  for  thickness  t; 

y= weight  of  water  in  poimds  per  cubic  foot =62.4; 

»  By  courtesy  of  J.  G.  White  &  Co. 


Digitized  by 


Google 


122  WATER  (CONDUCTORS  AND  ACCESSORIES 

«= thickness  of  pipe  in  feet; 
m^:  weight  of  material  in  pipe  line  in  pounds  per  cubic  foot 
=490. 

g= average  flow  of  water  through  pipe  during  twenty-four 
hours,  expressed  in  cubic  feet  per  second. 

c=sale  value  of  1  foot-pound  per  second  for  one  year,  meas- 
ured in  water  l^efore  delivery  to  turbine. 

t  =  annual  interest,  profit  and  depreciation  charge  on  1  pound 
of  material  in  pipe  line  in  place,  expressed  as  a  ratio. 
This  value  should  be  multiplied  by  whatever  factor  is 
necessary  to  make  allowance  for  excess  of  actual 
weight  of  pip)e  line  over  theoretical  weight  due  to  lap, 
rivets,  etc. 
c= friction  coefficient.     (See  page  106.) 

The  factor  X  for  a  50  per  cent  load  factor  will  generally  vary 
from  1.3  to  1.5.     It  may  be  figured  from  the  formula: 


Average  of  the  cubes  of  load  curve  ordinates 
Cube  of  the  average  of  load  curve  ordinates  ' 


This  means  that  the  load  cwrve  may  be  divided  into  as  many 
sections  as  desired  for  accuracy,  and  the  mean  ordinate  of  each 
section  used  in  the  formula. 

Having  determined  the  economic  diameter  for  a  given  thick- 
ness, that  for  any  other  thickness,  all  other  conditions  remaining 
the  same,  varies  inversely  as  the  sixth  root  of  the  thickness.^ 

Speed  regulation  must  also  be  considered  in  determining  the 
size  of  a  pipe  line,  and  this  point  is  probably  of  more  importance 
than  the  economical  consideration.  Load  changes  on  the  tur- 
bine cause  the  governor  to  open  or  close  the  turbine  gates  rapidly, 
thus  causing  pressure  changes  in  the  penstock.  These  pressure 
changes  are  due  to  the  acceleration  or  deceleration  of  the  water 
column  in  the  pipe  line,  and  the  magnitude  of  the  same  depends 
upon  both  the  length  of  the  penstock  and  the  change  of  velocity 
in  same. 

The  pressure  changes  always  act  in  opposition  to  the  action 
of  the  governor;   thus,  when  a  load  suddenly  goes  oflF  the  gener- 

»See  also  "Economical  Penstock  Size"  by  M.  Warren  A.S.C.E., 
Dec.  2,  1914. 


Digitized  by 


Google 


WATER  CONDUCTORS 


123 


TABLE  XXXII 
Loss  IN  Head  in  Each  100  Feet  Length  op  Pipe  at  Different  Velocities 


PaicTioM  llEAti  iitr  Fekt  fo«  Prffjft  100  Frist  Long  rnoM 

** 

2  Toe 

hKisv  DiAMKT«n  Inci.tj»ive  with  Ciimc   FfcET 

a 

J)1«CH 

A  ROB      ^K:B 

MjVUTE      I1hl*fc*l      ViHiOClTJEfl     FROM 

5 

1^ 

^i. 

1 

tu  12  Fekt  Inclijhivi£  i*wu  S 

eCOKD, 

-1 

it 

i-^%. 

3S£ 

2'  DIam. 

3*  Dlam. 

r  Dlarn. 

5"  Dlam, 

6'  Diam. 

3 

II 

11 

|| 

II 

ii 

1=: 

1^ 

> 

m 

^ 

u^ 

u 

El. 

O 

£ 

O 

ki 

o 

ii. 

U 

1  Q 

.015.^2 

00776 

.024 

188 

.015 

424 

754 

IJ78 

L696 

1   2 

02336 

oniet 

.033 

226 

022 

509 

016 

905 

013 

K414 

oil 

2,036 

1   4 

1  fl' 

03043 

01521 

043 

264 

.029 

504 

.021 

10. "iO 

017 

1,649 

Ot4 

2.375 

.03975 

.01087 

055 

303 

036 

670 

.027 

1206 

022 

1,885 

018 

2,714 

1   S 

06031 

02515 

068 

330 

045 

703 

.034 

1357 

027 

2,120 

022 

3,054 

2  0 

06211 

03105 

.082 

377 

054 

848 

.041 

1508 

032 

2,356 

027 

3,393 

2  2 

,07515 

03  787 

097 

415 

064 

033 

048 

16.^*9 

039 

2,. 592 

032 

3,732 

2  4 

.O»044 

04473 

lU 

453 

075 

1018 

057 

t8lO 

045 

2,H27 

037 

4,071 

2  6 

104M 

05248 

131 

400 

.087 

1 1  03 

065 

1960 

(t52 

3,063 

043 

4.411 

2  n 

.12173 

0608fi 

150 

528 

.09'.J 

1  18K 

075 

2111 

.  060 

3,299 

049 

4.750 

3.0 

13075 

0^087 

.169 

565 

112 

1272 

.084 

2262 

067 

3,534 

056 

5,080 

3  2 

,15» 

0795 

190 

603 

I2ri 

13.'->7 

O05 

2413 

076 

3.770 

063 

5,420 

3  4 

I7i>5 

8975 

.212 

641 

.141 

1442 

106 

2563 

085 

4,006 

070 

5,768 

a  e 

20124 

.ICM>62 

,235 

679 

.i.^>e 

1  r.27 

117 

2714 

094'    4.241 

078 

6,107 

as 

22422 

lun 

.260 

716 

173 

1612 

.130 

2H65 

104:    4.477 

OH  6 

0.446 

4  0 

24^44 

.12422 

,285 

754 

IKO 

1607 

T42 

3016 

114 

4.712 

004 

6,78© 

4  2 

.27^91 

.13695 

.311 

791 

207 

17X1 

155 

3167 

124 

4MAH 

103 

7J2S 

4  4 

30062 

.15031 

339 

829 

226 

1B66 

169 

3317 

t:^.' 

5, 1  84 

112 

7.464 

4  6 

328SB 

16444 

368 

867 

245 

1051 

184 

3468 

147 

5.410 

122 

7  804 

4  B 

35770 

I7fi88 

307 

0O5 

264 

2036 

198 

3619 

159 

5.655 

132 

N.I  43 

5  0 

38«19 

19400 

.428 

042 

285 

2121 

214 

3770 

171 

5.801 

142 

8,482 

S  Z 

4I9&7 

20993 

46 

080 

.300 

2205 

230 

3921 

1H4 

6  J  26 

1  53 

8.821 

5  4 

.45270 

22630 

403 

1018 

328 

2290 

246 

4071 

107 

6,362 

164 

0.1 6t 

5  « 

.411605 

24347 

,527 

1050 

351 

2375 

263 

4222 

210 

6.507 

^fi 

O.-^iOO 

S  R 

..V2235 

26117 

654 

1003 

.374 

2460 

2S1 

4373 

224 

0.833 

187 

9.830 

B  0 

.559 

279 

59ft 

1131 

308 

2545 

209 

4524 

239 

7.060 

101» 

19,179 

6  2 

,50SS9 

20844 

.035 

nfio 

423 

2630 

317 

4675 

254 

7.304 

211 

i(r.5tft 

6  4 

63602 

.31801 

,673 

1206 

448 

2714 

336 

4P25 

269 

7.540 

224 

10,8,57 

e  6 

67639 

.33819 

.712 

1244 

474 

^70^( 

356 

4076 

285 

7.775 

237 

1L107 

fl  8 

7I80I 

.35000 

753 

1282 

.501 

21^84 

376 

5127 

301 

8,0  n 

250 

( 1 .536 

7  0 

760B6 

38043 

794 

1319 

529 

2960 

.397 

5278 

317 

8,247 

264 

n.N7S 

7  2 

S04ti6 

.40248 

836 

1357 

h557 

3054 

.418 

5420 

334 

8.482 

278 

J  2.2 14 

7  4 

.85031 

.42515 

.  mt) 

1395 

586 

3138 

440 

5570 

352 

8,7  IH 

293 

rj,5,'i4 

7  0 

.89689 

44844 

024 

U33 

616 

3223 

462 

5830 

:^60 

8,954 

307 

Ili.MJS 

7  ft 

94472 

.47236 

070 

1470 

646 

3.H08 

,485 

5881 

38J^ 

0,1  SO 

32;i 

13,232 

&  0 

09378 

.49380 

1.01 

1  508 

67" 

3393 

.'iOrt 

6032 

406 

9.425 

33  h 

l;i,572 

S  2 

1  G44DO 

62204 

i.oe 

1546 

709 

3478 

532 

6182 

425 

9.660 

3^4 

13.011 

»  4 

1  mm^ 

54782 

1. 11 

1583 

741 

3563 

556 

6333 

445'    0.fi96: 

370 

14.2.-10 

R  a 

1    14S44 

57422 

1    16 

1621 

774 

>647 

58  J 

6484    , 

465    10.132 

387 

14.580 

IB 

1   2024« 

60124 

1    2] 

1650 

808 

3732 

606 

6635 

485 

10,367 

404 

1  i  ,929 

ft  0 

1   25776 

62888 

1    26 

1606 

843 

3817 

632 

6786 

506 

10.603 

421 

1  5.268 

t  a 

J   31428 

65714 

1.31 

1734 

,878 

3002 

65  K 

6036 

527 

10,830 

439 

1  5,607 

9  4 

1   372,^ 

68602 

1.37 

1772 

.013 

3087 

685 

7087 

548 

11.0741 

456    15.947 

&  fl 

I  43lO/i 

71552 

1.42 

1809 

050 

4072 

713 

7238 

570 

11,310 

475    I6.2H6 

«  S 

149190 

74565 

t   48 

1847 

om: 

4156 

741 

7389 

502 

11.545 

493|  Hi025 

10  0 

1   55279 

77630 

1    53 

1885 

I    02 

424  J 

76n 

7540 

615 

11,7S1 

512    16,964 

10  2 

I   61552 

. 80776 

1    59 

1923 

I    06 

4326 

798 

7600 

638 

12,017 

532    17.304 

10  4 

1    6795 

S3975 

1    65 

1060 

I    10 

4411 

.827 

7841 

662 

12.252 

551 

17.043 

10  6 

1   74472 

.87236 

1,71 

1008 

1    14 

4496 

.857 

7002 

686 

12,488 

571 

17,082 

m  8 

1   8U18 

90559 

1.77 

2036 

1    18 

4580 

8143 

710 

12,733 

592 

JH.322 

11  0 

1  87Sft8 

,93044 

1    83 

2073 

1    22 

4R65 

919 

fii2a4 

735 

12,959 

612 

18,661 

H  2 

1   947R2 

073fll 

1.90 

2111 

I    26 

4750 

950 

H444 

760 

13.195 

633 

1  9,000 

n  4 

2  OlMOl 

I    009 

1    06 

2149 

1    31 

483.=^' 

082 

8505 

786 

13,430 

655    («t,:^J9 

iL  e 

2  08944 

1.04473 

2  02 

2187 

1    35 

4920 

1   01 

h746 

811    13,6661 

67r    (1L*;79 

u  & 

2   I62n 

I    08105. 

2,09 

2224 

1    30 

5005 

1    04 

88r^7   1 

838    13.9021    HUK    2il,018 

ta  Q 

2  23602 

IIKOI- 

2   16 

2262 

1    44 

5089 

1  ow 

0048    1 

Kr;5 

I4,3:i7' 

72<s. 

^o,;ri7 

Digitized  by 


Google 


124 


WATER  (CONDUCTORS  AND  ACCESSORIES 


TABLE  XXXIL— Con^inueei 


FmcTirm  Head  in  J-'kkt  won  Pipe«  100  Fbet  Lotto  rmov 

** 

7  TO  12  Feet  Diauktkr  Incluhivb  with  Cubic  Fk«t 

m 

D[0CK\HI1E      PRtt       MiMtTTK      UfWl»E»      VKLOClTlEfl       rHOM 

s 

B& 

1  TO  12  Feet  Imclubivic  peh  Beconp. 

Pi 

3|£ 

|4S 

7'  DIara, 

8'  Diftiii. 

0'  Diam. 

10'  Dlftm. 

12'  Dt»ni. 

1^ 

it 

II 

-1 

H 

|l 

II 

II 

11 
1^ 

|l 

> 

X 

m 

bk 

u 

u* 

0 

U4 

U 

Urn 

0 

ta 

U 

1.0 

.01553 

00776 

2,309 

3,016 

3.817 

4.712 

6,786 

1.2 

.03236 

oni8 

2.771 

3,619 

4.580 

6.655 

S.I  43 

1.4 

.03043 

01521 

.012 

3.233 

4,222 

5,344 

6,. 597 

9.500 

16 

.0397.^ 

oiasr 

015 

3,695 

013 

4,825 

012 

6,107 

oii 

7.540 

10,858 

1.8 

.05031 

02515 

019 

4,156 

017 

5.429 

015 

6.871 

013 

8,482 

.011 

12.215 

2  0 

.062U 

03105 

023 

4,618 

020 

6,032 

018 

T,634 

016 

9,425 

013 

13.572 

2   2 

.07-51.5 

,03757 

027 

5,080 

024 

6.635 

021 

8.397 

019 

10.367 

.016 

14.929 

2  4 

.08944 

.04472 

032 

5.542 

028 

7.238 

025 

&,161 

022 

11,310 

018 

16.2S6 

2ft 

. 10496 

05248 

037 

6,004 

032 

7,841 

029 

9.924 

026 

1  2.252 

.021 

17,644 

2  8 

12173 

.06086 

042 

ft. 465 

039 

8,445 

033 

10.688 

,030 

13.195 

034 

19.001 

3  0 

13975 

06987 

04  H 

6.927 

042 

9.048 

.037 

11,451 

033 

14.137 

.028 

20.358 

3   2 

159 

J>7H"i    ■    O.M 

7,389 

047 

9,651 

043 

12.214 

038 

15,080 

031 

21,715 

3   4 

t79-> 

O.H'.i7.-. 

4HpO 

7,851 

053 

10,254 

047 

12,978 

043 

16.022 

035 

23,072 

3ft 

20124 

,  HHH]2 

007 

8,313 

0.58 

10.857 

052 

13,741 

047 

16.965 

039 

24,430 

3  8 

22423 

112M 

074 

8.775 

065 

11.460 

057 

14„5fl5 

053 

17.907 

.043 

25.787 

4  0 

,24ft44 

12423 

081 

9p236 

071 

1  2,064 

063 

15.368 

057 

18.850 

.047 

27,144 

4.2 

27391 

.13605 

088 

9.698 

077 

12,667 

069 

16.031 

062 

19.793 

051 

28,501 

4  4 

.30062 

L^31 

097 

ID,  160 

OM 

13,370 

075 

16.795 

067 

20,735 

056 

29*S5S 

4  d 

32S»S 

.16444 

,105 

10,622 

092 

1  3,873 

081 

17,558 

073 

21.677 

061 

21.216 

4.g 

.35776 

178SS 

113 

11.084 

099, 

14,476 

088 

18.323 

079 

22.620 

066 

32.573 

5  0 

3smu 

.  11*409 

133 

11,54ft 

197 

1  5.080 

095 

19.08,^1 

085 

23.502 

071 

33.930 

5   2 

.41987 

,30nP3 

131 

1 2.007 

115 

15.683 

.102 

19,^49 

092 

24„^4 

076 

35.287 

5.4 

.  45279, 

2263i)| 

140 

12,4fl9 

.123 

1  6,286 

,109 

20,612 

098 

25,447 

032 

36.664 

5  6 

.48095 

.24347 

.150 

12.931 

131 

1  6,889 

117 

21,375 

.105 

26.389 

.087 

38,003 

6  8 

. 52235 

.26117 

160 

13,393 

140 

17.492 

.124 

22,139 

112 

27,332 

.093 

39,^59 

6  0 

559 

.2795 

170 

13,855 

149 

I8,6t»5 

.132 

22.902 

.119 

28,274 

.099 

40.716 

d.2 

.  Bumv 

2W844 

181 

14,316 

.  158 

18,699 

.141 

23,666 

127 

29.317 

.105 

42.073 

6.4 

636*12 

.31801 

192 

14,778 

168 

19,302 

.149 

24,439 

.134 

30,1,'>& 

U2 

43.430 

6  e 

. 6763W 

.33810 

203 

15.340 

178 

19.905 

.158 

25,192 

142 

31.102 

AIS 

44.788 

6.8, 

.71«01 

3590 

215 

15,702 

188 

20,.50S 

.167 

25 .9  ,"16 

ITjO 

32.044 

.125 

46.145 

7.0 

.7608ft 

3843 

226 

16.164 

198 

21.111 

.175 

26,719 

158 

32,987 

.132 

47,502 

7.2 

.804Ua 

40248 

238 

16,626 

209 

21,714 

\S:t 

i;7.483 

,167 

33.929 

139 

48.859 

7.4 

.85031 

42515 

251 

17.087 

220 

22.318 

1  '15 

2N,246 

176 

34.872 

.146 

50«3I6 

7.6 

.8§6S© 

44844 

264 

17„^49 

231 

22.921 

.  2U.-» 

29,009 

.184 

3.5,814 

154 

51.fi74 

7,8 

.94472 

47236 

277 

18.011 

242 

23,524 

2  I  5 

29,773 

194 

36.757 

lil 

53,931 

8  0 

.99.178 

49689 

290 

18.473 

254 

24.137 

,  235 

30,536 

203 

37,699 

169 

54,288 

8.2 

1    04100 

.52204 

303 

18,U35 

266 

24,730 

.2.36 

31,300 

.212 

38.642 

.177 

55,645 

8.4 

1.09561 

547h;j 

317 

19,396 

278 

35,334 

.247 

:i2.nfi3 

222 

39.584 

ISfi 

57.002 

8  6 

1    14844 

. 57422 

332 

19,858 

290 

25JI37 

•jr.^ 

Hl!,H2fi 

2,33 

40.527 

,193 

58,300 

g.o 

1    20248 
1    2577H 

60 11^4 
.62HH8 

34  rt 
361 

20.320 
20,782 

303 
310 

36.540 
27,143 

:{H..-,90 
:i-l,353 

.242 
253 

41.469 
43.412 

.202 

.210 

&9,717 
61,074 

a  2 

1,31428 

65714 

376 

21,2^4 

.329 

27,74fi 

2*J2 

35,117 

263 

43.354 

.319 

62,431 

9  4 

1    37204 

.68002 

.391 

2K7tl6 

342 

2^,349 

AiU 

35.880 

274 

44.207 

.22fi 
.237 

163.788 

9   6 

1    43105 

. 7  1 552 

407 

22,167 

'A  5fl 

2s,9,'i'i 

3  1  () 

36,643 

2S5 

45.230 

65,146 

©   8 

I    4B130 

745f>5 

423 

22,62a 

370 

2il,556 

339 

37.407 

296 

4ft.  182 

.247 

66,. 503 

10  0 

1  .  5527il 

77tt39 

4^19 

23,091 

384 

30,159 

341 

38,170 

307 

47,124 

.3.56 

67.860 

10  2 

1    61552 

807  7 1> 

455 

23.,553 

399 

30,762 

354 

38.943 

319 

48,066 

.266 

69.217 

10,4 

1    67115 

S3  97  5 

472 

24.015 

413 

31.365 

307 

39.697 

331 

49.009 

.275 

70.574 

10.6 

1.74172 

S7336 

490   24.4  7ft 

428 

31,9H9 

381    4O.-lfi0 

343 

49,951 

285 

71.932 

10  8 

1    81118 

905511 

507    24, 93s 

444 

32,575 

394 

41.224 

355 

50,894 

.296 

73,289 

11.0 

1.87888 

l^mAi 

525   25.400 

459 

33,175 

4  OK 

4  1 .987 

387 

51.836 

306 

74.646 

11    2 

J    94782 

517391 

543   2.5,862 

475 

33.778 

43^: 

42,751 

,380 

52,779 

316 

76,003 

11    4 

2  0I8U1 

1    000 

561    26,324 

491 

34,381 

4M\ 

43,514 

393 

53,721 

327 

77,360 

lie 

2,08944 

1    01472 

r,7\t  ztii^n 

507 

34.9S1 

\r,ii 

44,277 

.406 

54,ft64 

33a 

78,718 

11. s 

2    16211 

1    OKI 05     ftiP^'  27  247' 

5:>4 

:{.^^,^h.^ 

hfiFi 

45.04  1 

419 

5£i<IOft 

,349 

§0.075 

12.0 

2   33602 

I     MMDI      tAl    27. 7139 j 

540 

3rtj-ji 

4. SO 

I 

15,804 

432 

5&,4S9 

.300 

§1,432 

Digitized  by  CjOOQ IC 


WATER  CONDUCTORS  125 

ator  and  the  turbine  gates  close,  there  is  an  increase  in  pressure 
in  the  penstock  which  tends  to  develop  more  horse-power,  and 
vice  versa,  when  a  load  comes  on  the  generator  and  the  turbine 
gates  op)en,  there  is  a  drop  in  pressure  in  the  penstock,  tending 
to  decrease  the  output  of  the  turbine.  As  the  length  of  the  p)en- 
stock  for  any  particular  installation  is  fixed,  it  is  necessary  to  Umit 
the  changes  in  velocity  in  the  penstock,  in  order  to  give  reasonably 
good  speed  regulations. 

Excessive  rises  in  pressure  may  be  eliminated  by  the  use  of 
pressure  regulators,  by-pass  relief  valves  or  surge  tanks.  After 
the  size  of  the  penstock  has  been  tentatively  settled  as  most  suit- 
able for  economical  considerations,  it  must  then  be  investigated 
for  speed  regulation,  and  this  may  indicate  that  a  larger  pipe  may 
have  to  be  used  than  consistent  with  the  highest  economy.  (See 
also  Surge  Tanks  and  Pressure  Regulators.) 

Steel  Pipe.  These  may  be  made  of  rolled  steel  plates,  riveted 
together.  Fig.  60,  or  lap-welded,  the  latter  only  being  used  for 
very  high  heads  where  the  pressure  is  excessive  and  where  the  use 
of  the  riveted  construction  would  greatly  increase  the  thickness  of 
the  plate.  In  figuring  the  thickness  of  the  plate,  this  should  be 
based  not  only  on  the  pressure  due  to  the  net  head  but  also  on 
the  additional  pressure  caused  by  the  water  hanmier. 

The  formula  for  the  strength  of  riveted  steel  pipe  is 


Pdf 


2Te' 
where 

^  =  thickness  of  plate  in  inches; 
P= pressure  in  pounds  p)er  square  inch; 
d= diameter  of  pipe  in  inches; 
/= factor  of  safety,  based  on  the  ultimate  tensile  strength.    4 

is  a  factor  generally  used  in  this  country. 
T=tensUe  strength  =  50,000  for  mild  steel; 

=60,000  for  wrought  iron; 
e=efl5ciency  of  riveted  joint  =  approximately  0.60  for  single 
rivets  and  0.70  for  double  rivets. 

Table  XXXIII  ^  gives  the  safe  working  heads  and  weights  of 
riveted  steel  pipes. 

» Pelton  Water  Wheel  Company. 


Digitized  by 


Google 


126 


WATER  CONDUCTORS  AND  ACCESSORIES 


It  would  seem  advisable  in  proportioning  the  thickness  of 
penstocks  to  provide  in  addition  to  the  thickness  computed  by  the 
above  formula,  an  allowance  for  corrosion,  that  is,  the  addition 


Fig.  60.— Ten  Five-foot  Riveted  Steel  Penstocks. 

of  a  constant  term  to  the  thickness,  say  |  inch  or  whatever  is  con- 
sidered advisable  under  the  conditions  of  installation. 

Another  point  which  must  be  given  very  careful  consideration 
in  connection  with  the  calculation  of  pipe-line  sizes  and  thicknesses 
is  their  safety  from  collapsing,  due  to  sudden  drop  in  pressure. 
The  following  formula  gives  the  maximum  difference  between 
the  external  and  the  internal  pressures  which  a  circular  steel  pipe 
can  withstand: 

p  =  50,200,000Hy. 

Where 

p  =  pressure  difference  in  pounds  per  square  inch; 
i  =  thickness  of  plate  in  inches; 
d= diameter  of  pipe  in  inches. 


Digitized  by 


Google 


WATER  CONDUCTORS 


127 


TABLE  XXXIII 
Safb  Wobkinq  Heads  and  Weights  of  Riveted  Steel  Pipes 

Heavy-face  figures  -weight  per  foot. 

Light-lace  figures  -safe  head  in  feet. 

Safety  factor -4. 

Tensile  strength  -55,000  pounds  i>er  square  inch. 

EfBciency  of  riveted  joint  -70  per  cent. 


Platb  Thickness 

Diam- 

Diam- 

eter. 
In. 

No. 
18 

No. 
10 

No. 
8 

A" 

i" 

A" 

1" 

A" 

*" 

A" 

1" 

eter. 
In. 

10 

15.8 

483 

18.8 

592 

88.3 

730 

10 

IS 

17.9 

401 

11.1 

493 

16.5 

607 

31.0 

600 

44.6 

921 

U 

14 

10.6 

343 

15.8 

423 

81.6 

511 

86.7 

592 

49.4 

790 

14 

16 

83.8 

300 

18.9 

368 

85.8 

456 

41.4 

518 

66.0 

692 

16 

18 

16.0 
268 

31.1 

320 

40.0 

405 

46.1 

460 

68.1 

615 

78.3 

768 

18 

10 

18.6 

85.3 

44.0 

50.6 

68.5 

86.4 

104.3 

10 

240 

206 

365 

414 

553 

691 

830 

tt 

81.4 

88.9 

48.1 

65.6 

74.9 

94.9 

114.9 

135.8 

154.9 

11 

218 

269 

331 

377 

503 

630 

755 

880 

1006 

M 

84.1 

41.1 

58.3 

60.8 

81.1 

101.9 

114.3 

146.4 

168.6 

14 

201 

247 

303 

346 

461 

576 

602 

807 

922 

S6 

86.8 

45.3 

56.4 

64.9 

87.5 

110.7 

184.0 

160.0 

185.9 

M 

184 

228 

280 

320 

426 

534 

640 

748 

856 

» 

89.6 

48.7 

60.4 

69.8 

93.9 

119.0 

143.7 

170.8 

197.8 

114.0 

150.0 

18 

172 

211 

260 

296 

395 

494 

593 

692 

790 

873 

988 

ao 

41.8 

58.1 

64.6 

74.6 

99.1 

116.8 

153.4 

181.4 

111.3 

839.4 

166.8 

30 

161 

197 

243 

276 

368 

461 

553 

645 

738 

830 

922 

n 

45.0 

55.5 

68.7 

79.1 

106.4 

134.7 

168.9 

191.1 

119.7 

851.3 

183.0 

31 

150 

184 

228 

250 

346 

432 

519 

605 

692 

777 

865 

34 

47.8 

58.8 

71.9 

83.9 

111.9 

141.7 

171.5 

104.1 

186.0 

167.8 

199.1 

84 

141 

173 

214 

243 

325 

406 

488 

569 

651 

731 

813 

86 

50.6 

68.0 

76.9 

88.8 

U8.1 

149.6 

181.0 

813.0 

147.1 

181.0 

314.8 

86 

133 

164 

202 

230 

307 

384 

461 

538 

615 

692 

768 

88 

66.8 

81.0 

91.3 

115.1 

167.0 

189.0 

114.0 

159.0 

193.8 

818  6 

88 

155 

191 

217 

289 

362 

435 

507 

578 

652 

725 

10 

68.7 

86.1 

98.1 

131.8 

167.0 

199.1 

135.9 

171.9 

310.1 

347.9 

40 

148 

182 

206 

276 

346 

414 

484 

553 

622 

691 

48 

71.0 

89.3 

101.8 

138.0 

173.1 

108.1 

146.0 

183.7 

384.0 

364.1 

41 

140 

173 

197 

263 

329 

396 

461 

517 

593 

658 

44 

98.4 

107.4 

144.1 

180.9 

117.3 

157.8 

197.8 

339.1 

880.0 

44 

164 

188 

250 

314 

377 

439 

502 

565 

628 

46 

97.6 

111.0 

150.8 

188.3 

816.5 

168.8 

810.8 

353.6 

896.4 

48 

158 

181 

239 

300 

360 

420 

480 

541 

601 

48 

101.6 

117.1 

158.1 

196.4 

135.1 

179.1 

813.5 

366.5 

410.0 

49 

151 

173 

230 

288 

346 

403 

461 

519 

576 

60 

111.7 

168.6 

104.1 

146.1 

190.1 

334.5 

881.6 

418.5 

50 

165 

220 

276 

331 

386 

442 

497 

552 

88 

116.6 

169.7 

111.6 

155.6 

308.8 

350.8 

398.1 

445.1 

51 

159 

212 

265 

318 

371 

424 

477 

531 

64 

131.8 

176.1 

110.8 

166.1 

313.5 

361.0 

411.1 

468.0 

54 

153 

205 

256 

307 

358 

410 

461 

512 

66 

136.1 

188.4 

816  6 

176  0 

316.1 

377.8 

416.1 

477.6 

56 

149 

198 

249 

298 

348 

398 

447 

498 

No. 
11 

No. 
10 

No. 
8 

A" 

i" 

A" 

r 

A" 

i" 

A" 

1' 

Digitized  by 


Google 


128  WATER  CONDUCTORS  AND  ACCESSORIES 

A  study  must  be  made  of  the  entire  penstock  from  the  head- 
gates  to  the  turbine  casing,  and  the  exact  drop  in  pressure  cal- 
culated at  each  section  imder  the  most  severe  conditions,  which 
possibly  would  occur  when  a  turbine  unit  is  running  light,  and  a 
short  circuit  occurs  on  the  generator,  in  which  case  the  turbine 
gates  open  wide  very  quickly,  and  there  is  a  tendency  to  accelerate 
the  water  in  the  various  sections  of  the  pipe  line. 

There  may  be  some  section  in  a  long  penstock  where  the 
water  colmnn  below  this  section  has  sufficient  head  to  accelerate 
the  lower  column  quicker  than  the  water  colunm  above  may  be 
accelerated.  This  may  cause  a  break  in  the  water  coltmin  at  the 
section  in  question,  and  a  considerable  vacuum,  which  is  very 
likely  to  collapse  the  penstock.  To  prevent  this  air  vents  (see 
page  157)  may  be  provided  at  the  points  along  the  pipe  line  where 
dangers  are  expected,  as  whenever  the  pipe  greatly  increases  its 
slope  or  rate  of  fall.  The  amoimt  of  air  which  must  be  admitted 
to  keep  the  pressure  from  going  below  a  certain  given  value  must 
be  such  as  will,  at  the  given  pressure,  replace  the  water  which 
has  run  away  from  the  section. 

On  account  of  the  imcertainty  of  the  calculation  of  the  col- 
lapsing strength  of  a  riveted  steel  pipe,  and  in  order  to  provide  a 
margin  of  safety,  it  would  seem  to  be  the  best  practice  to  pro- 
vide against  any  excess  of  external  over  the  internal  pressure 
at  any  point  in  the  pipe  line,  rather  than  attempt  to  compute 
the  collapsing  pressure.  .  The  critical  points  subject  to  a  deficient 
internal  pressure  can  best  be  located  by  drawing  a  hydraulic 
gradient  under  conditions  of  accelerated  or  retarded  flow  in  the 
pipe  line.  , 

For  a  more  complete  treatise  on  this  important  subject,  the 
reader  is  referred  to  an  article  by  Enger  and  Seely,  in  "Engineering 
Record^' for  May  23,  1914. 

Expansion  joints  are  not  usually  employed  in  this  country,  and 
if  the  pipe  is  carefully  laid  and  buried  or  kept  with  water  flowing 
at  all  times,  are  not  required  except  in  special  cases.  Whether 
the  pipes  are  buried  or  not,  they  should  be  carried  on  concrete 
piers.  Heavy  anchorage  blocks  should  be  inserted  at  all  vertical 
and  horizontal  bends,  and  with  considerable  temperature  varia- 
tions, expansion  joints  should  in  such  instances  be  provided  to 
take  care  of  the  expansion  and  contraction  of  the  pipe.  While  the 
stress  may  be  well  within  the  elastic  limit  of  the  pipe  material. 


Digitized  by 


Google 


WATER  CONDUCTORS 


129 


and  would  have  little  influence  on  the  pipe  itself,  the  thrust 
caused  by  the  expansion  may  throw  a  very  high  stress  on  the 
anchorage  blocks.     By  pro- 
viding  expansion  joints  a 
material   saving  can  often 
be  effected  in  the  cost  of 
anchorage  blocks  and  piers, 
esp)ecially  where  their  con- 
struction involves  difiicul- 
ties  owing  to  the  steepness 
of  the  grade  and  lack  of 
handling  faciUties.     A    de-        Fig.  61. — Pipe  Line  Expansion  Joint, 
tail  design  of  an  expansion 

joint  is  shown  in  Fig.  61  and  in  Figs.  62  and  63  are  shown  a  typical 
penstock  installation  and  details  of  supporting  and  anchoring  pier& 


Fio.  62. — Large  Hydro-Electric  Power  Station  at  Rjukan,  Norway,  Showing 
Ten  Five-foot  Penstocks  and  Method  of  Anchoring  Same. 

In  order  to  prevent  freezing  it  is  often  essential  to  know  the 
amount  of  water  necessary  to  pass  through  the  penstock,  as  for 


Digitized  by 


Google 


130 


WATER  CONDUCTORS  AND  ACX!ESSORIES 


Digitized  by 


Google 


WATER  CONDUCTORS  131 

example  during  the  shut-down  of  a  unit.    This  may  be  obtained 
from  the  following  formula  by  Boucher:  . 

T 

rp   -^  to 

Where 

Q= Water  discharge  in  cubic  meters  p)er  hour; 
Ta  =  Lowest  air  temperature   in  degrees  Centigrade;  without 

negative  sign; 
T«,= Water  temperature  in  degrees  Centigrade*  (may  be  taken 

as  rC); 
S= Exposed  surface  of  penstock  in  square  meters. 

Wooden-stave  Pipe}  This  kind  of  pipe  is  extensively  used 
in  the  West  where  redwood  or  fir  is  cheap  and  plentiful.  It  is 
admirably  adapted  for  heads  up  to  about  200  feet,  and  for  high- 
head  developments  it  is  often  used  for  the  upper  sections.  For 
heads  above  200  feet,  steel  pipe  is  preferable,  as  the  spacing  of 
the  bands  for  wooden-fitave  pip)e  becomes  so  close  that  the  cost 
of  the  pip)e  may  equal  or  exceed  that  of  steel. 

Woodennstave  pipe. has  a  greater  carrying  capacity  than  steel 
pipe  on  account  of  the  smooth  surface  of  the  planed  wood,  and  its 
carrying  capacity  will  not  decrease  with  age,  as  deposits  will  not 
adhere  to  the  inside  of  the  pipe. 

A  woodennstave  pipe  should  always  be  in  use  so  that  the  staves 
are  thoroughly  saturated.  Under  these  conditions  they  will  not 
decay  and  leakages  are  prevented.  Provisions  are,  however, 
made  so  that  the  staves  may  readily  be  drawn  firmly  together  by 
tightening  the  bands. 

Continuous  woodnstave  pipe  is  constructed  in  place  and 
should  preferably  be  located  above  ground  and  free  from  all 
contact  with  it,  cracHes  being  provided  at  certain  intervals  for 
the  support  (Figs.  ^  and  65). 

In  erecting  the  pipe  the  staves  are  assembled  and  put  together 
to  form  a  circle  of  the  diameter  of  the  pipe  and  the  bands  put 
around  the  outside  and  tightened  to  hold  the  staves  together. 
The  end  joints  in  the  staves  should  be  broken  by  a  lap  of  not  less 

1  An  excellent  treatise  on  wood-stave  pipe  is  found  in  Bulletins  Nos.  155 
and  376  U.  S.  Dept.  of  Agriculture. 


Digitized  by 


Google 


132 


WATER  (CONDUCTORS  AND  ACCESSORIES 


than  1  foot,  and  they  can  be  made  tight  by  inserting  a  metal 
tongue  or  plate  in  the  saw  kerf  cut  in  the  ends  of  the  staves. 


Fig.  64. — Wooden-Stave  Pipe  Showing  Method  of  Installation  in  Difficult 

Territory. 


Fig.  65. — Montana  Power  Company.     Dam  and  Wooden  Penstocks  for 
Madison  No.  2  Plant. 

After  the  pipe  is  completed  and  before  the  water  is  turned  on,  the 
bands  should  be  tightened  imiformly  so  as  to  give  tension  on  all 


Digitized  by 


Google 


WATER  CONDUCTORS  133 

the  bands.  When  the  pipe  is  filled  with  water  the  staves  swell 
sufficiently  to  bed  the  bands  slightly  into  the  wood  and  make 
the  longitudinal  joints  water-tight. 

The  size  of  the  bands  and  the  spacing  are  naturally  related, 
and  when  properly  designed  they  should  be  strained  to  their  safe 
resisting  value,  and  the  bearing  pressure  on  the  stave  must  not  be 
greater  than  the  safe  bearing  value  of  the  wood.  It  has  been 
found  from  actual  experience  that  the  width  of  contact  between 
the  band  and  pipe  is  equal  to  about  the  radius  of  the  band  before 
the  fibers  of  the  wood  are  crushed  beyond  safety.  The  safe 
crushing  stress  for  wood  is  generally  taken  as  650  pounds  per  square 
inch,  and  putting  the  safe  stress  in  the  band  equal  to  the  safe- 
bearing  pressure,  we  get 

irr^«  =  (/J+0650r, 
or 

^(fl+0650 

Where 

r= radius  of  band  in  inches; 
£==  internal  radius  of  pipe  in  inches; 
(= thickness  of  stave  in  inches; 

8 = safe  tensile  strength  of  band.  Taking  the  ultimate  strength 
of  steel  as  60,000  poimds,  and  assuming  a  factor  of 
safety  of  4,  the  safe  strength  is  15,000  pounds  per 
square  inch. 

The  number  and  thus  the  spacing  of  the  bands  depends  on  the 
stresses  due  to  the  water  pressiure  and  to  the  swelling  of  the  wood. 
The  sum  of  these  two  stresses  should  be  equal  to  the  safe  strength 
of  the  band,  as  determined  by  the  previous  formula. 
Thus 

7n^8==j)dR+tdE, 
suid 

"pR+tE' 
where 

d==  spacing  of  bands  in  inches; 
p= water  pressure  in  pounds  per  square  inch; 
£= swelling  force  of  wood  per  square  inch.    This  is  usually 
assumed  to  be  approximately  equal  to  100. 


Digitized  by 


Google 


134 


WATER  CONDUCTORS  AND  ACCESSORIES 


TABLE  XXXIV 
Flow  of  Water  through  Wooden-stavb  Pipe 


2  Feet  Diameter. 

3  Feet  Diameter. 

4  Feet  Diametes. 

Dis- 

charsre, 

Cu.ft. 

per 

Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

1.6 

0.48 

0.003 

4 

0.57 

0.003 

6 

0.48 

0.002 

3.0 

0.95 

0.012 

8 

1.13 

0.010 

12 

0.95 

0.006 

4.5 

1.43 

0.025 

12 

1.70 

0.021 

18 

1.43 

0.011 

6.0 

1.91 

0.042 

16 

2.26 

0.035 

24 

1.91 

0.018 

7.5 

2.39 

0.064 

20 

2.83 

0.054 

30 

2.39 

0.028 

9.0 

2.86 

0.090 

24 

3.40 

0.077 

36 

2.86 

0.040 

10.5 

3.34 

0.122 

28 

3.96 

0.105 

42 

3.34 

0.054 

12.0 

3.82 

0.159 

32 

4.53 

0.137 

48 

3.82 

0.070 

13.5 

4.30 

0.201 

36 

5.09 

0.173 

54 

4.30 

0.088 

15.0 

4.77 

0.248 

40 

5.66 

0.213 

60 

4.77 

0.108 

16.6 

5.25 

0.300 

44 

6.22 

0.258 

66 

5.25 

0.131 

18.0 

5.73 

0.356 

48 

6.79 

0.306 

72 

5.73 

0.156 

19.6 

6.21 

0.416 

52 

7.36 

0.358 

78" 

6.21 

0.183 

21.0 

6.68 

0.482 

56 

7.92 

0.415 

84 

6.68 

0.212 

22.6 

7.16 

0.553 

60 

8.49 

0.476 

90 

7.16 

0.243 

24.0 

7.64 

0.629 

64 

9.05 

0.542 

96 

7.64 

0.276 

26.5 

8.12 

0.709 

68 

9.62 

0.613 

102 

8.12 

0.311 

27.0 

8.59 

0.793 

72 

10;  19 

0.687 

108 

8.59 

0.349 

28.5 

9.07 

0.881 

76 

10.75 

0.764 

114 

9.07 

0.389 

30.0 

9.55 

0.974 

80 

11.32 

0.846 

120 

9.55 

0.431 

31.5 

10.03 

1.073 

84 

11.88 

0.933 

126 

10.03 

0.475 

33.0 

10.50 

1.178 

88 

12.45 

1.024 

132 

10.50 

0.521 

34.5 

10.98 

1.287 

92 

13.02 

1.118 

138 

10.98 

0.569 

36.0 

11.46 

1.400 

96 

13.58 

1.216 

144 

11.46 

0.619 

37.5 

11.94 

1.519 

100 

14.15 

1.319 

150 

11.94 

0.671 

39.0. 

12.41 

1.643 

104 

14.71 

1.427 

156 

12.41 

0.725 

40.5 

12.89 

1.772 

108 

15.28 

1.539 

162 

12.89 

0.781 

42.0 

13.37 

1.907 

112 

15.85 

1.655 

168 

13.37 

0.840 

43.5 

13.85 

2.046 

116 

16.41 

1.775 

174 

13.85 

0.901 

45.0 

14.32 

2.189 

1 

120 

16.98 

1.900 

180 

14.32 

0.965 

Digitized  by 


Google 


WATER  CONDUCTORS 
TABLE  XXXIW— Continued 


135 


5  Fbkt  Diamxtbb. 

6  Fbbt  Diambtbb. 

7  Fbbt  Diambtbb. 

Dis- 
charge. 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

10 

0.51 

0.001 

15 

0.63 

0.001 

20 

0.52 

0.002 

20 

1.02 

0.004 

30 

1.06 

0.004 

40 

1.04 

0.004 

30 

1.53 

0.009 

45 

1.59 

0.008 

60 

1.56 

0.007 

40 

2.04 

0.016 

60 

2.12 

0.014 

80 

2.08 

0.012 

50 

2.55 

0.025 

75 

2.65 

0.022 

100 

2.60 

0.018 

60 

3.06 

0.036 

90 

3.18 

0.032 

120 

3.12 

0.026 

70 

3.57 

0.048 

105 

3.71 

0.043 

140 

3.64 

0.035 

80 

4.07 

0.062 

120 

4.24 

0.056 

160 

4.16 

0.045 

90 

4.58 

0.078 

135 

4.77 

0.070 

iro 

4.68 

0.057 

100 

5.09 

0.096 

150 

5.31 

0.086 

200 

5.20 

.0.070 

110 

5.60 

0.116 

165 

5.84 

0.104 

220 

5.72 

0.085 

120 

6.11 

0.138 

170 

6.37 

0.124 

240 

6.24 

0.102 

130 

6.62 

0.162 

195 

6.90 

0.145 

260 

6.76 

0.120 

140 

7.13 

0.188 

210 

7.43 

0.168 

280 

7.28 

0.139 

150 

7.64 

0.216 

225 

7.96 

0.193 

300 

7.80 

0.169 

160 

8.15 

0.246 

240 

8.49 

0.219 

320 

8.32 

0.180 

170 

8.66 

0.277 

255 

9.02 

0.247 

340 

8.83 

0.203 

180 

9.17 

0.310 

270 

9.55 

0.276 

360 

9.35 

0.227 

190 

9.68 

0.345 

285 

10.08 

0.307 

380 

9.87 

0.263 

200 

10.19 

0.882 

300 

10.61 

0.340 

400 

10.39 

0.280 

210 

10.70 

0.421 

315 

11.14 

0.375 

420 

10.91 

0.308 

220 

11.20 

0.462 

330 

11.67 

0.412 

440 

11.43 

0.337 

230 

11.71 

0.505 

345 

12.20 

0.451 

460 

11.96 

0.368 

240 

12.22 

0.550 

360 

12.73 

0.491 

480 

12.47 

0.401 

250 

12.73 

0.597 

375 

13.26 

0.532 

600 

12.99 

0.436 

260 

13.24 

0.646 

390 

13.79 

0.575 

520 

13.51 

0.472 

270 

13.75 

0.696 

405 

14.32 

0.620 

540 

14.03 

0.509 

280 

14.26 

0.748 

420 

14.85 

0.666 

560 

14.55 

0.548 

290 

14.77 

0.802 

435 

15.38 

0.714 

580 

15.07 

0.588 

300 

15.28 

0.858 

450 

15.92 

0.7C4 

600 

15.59 

0.629 

Digitized  by 


Google 


136  WATER  CONDUCTORS  AND  ACCESSORIES 

TABLE  XXXIV— Ccm/iniied 


8  FbbT  DiAlCSTBB. 

10  Fbbt  Diambteb. 

Dis- 
charge. 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 

Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Loss  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

Dis- 
charge, 
Cu.ft. 
per 
Sec. 

Veloc- 
ity. 
Feet 
per 
Sec. 

Lose  of 
Head  in 
Feet  per 
100  Feet 
of  Pipe. 

30 

0.60 

0.001 

30 

0.47 

0.001 

40 

0.51 

0.001 

60 

1.19 

0.004 

60 

0.94 

0.002 

80 

1.02 

0.002 

90 

1.79 

0.008 

90 

1.41 

0.004 

120 

1.63 

0.004 

120 

2.39 

0.014 

120 

1.89 

0.008 

160 

2.04 

0.008 

150 

2.98 

0.021 

150 

2.36 

0.012 

200 

2.55 

0.013 

180 

3.58 

0.030 

180 

2.83 

0.017 

240 

3.06 

0.018 

210 

4.18 

0.041 

210 

3.30 

0.023 

280 

3.57 

0.024 

240 

4.77 

0.053 

240 

3.77 

0.030 

320 

4.07 

0.032 

270 

5.37 

0.067 

270 

4.24 

0.038 

360 

4.58 

0.040 

306 

5.97 

0.083 

300 

4.72 

0.046 

400 

5.09 

0.049 

330 

6.56 

0.100 

330 

5.19 

0.056 

440 

5.60 

0.059 

360 

7.16 

0.119 

360 

5.66 

0.067 

480 

6.11 

0.070 

390 

7.76 

0.139 

390 

6.13 

0.078 

520 

6.62 

0.082 

420 

8.36 

0.161 

420 

6.90 

0.090 

560 

7.13 

0.095 

450 

8.95 

0.185 

450 

7.07 

0.104 

600 

7.64 

0.109 

480 

9.55 

0.211 

480 

7.55 

0.118 

640 

8.15 

0.124 

510 

10.15 

0.238 

510 

8.02 

0.133 

680 

8.66 

0.140 

540 

10.74 

0.267 

540 

8.49 

0.149 

720 

9.17 

0.157 

570 

11.34 

0.297 

570 

8.96 

0.165 

760 

9.68 

0.175 

600 

11.94 

0.329 

600 

9.43 

0.183 

800 

10.19 

0.194 

630 

12.53 

0.362 

630 

9.90 

0.202 

840 

10.70. 

0.214 

660 

13.13 

0.397 

660 

10.38 

0.222 

880 

11.20 

0.235 

690 

13.73 

0.434 

690 

10.85 

0.243 

920 

11.71 

0.257 

720 

14.32 

0.437 

726 

11.32 

0.264 

960 

12.22 

0.280 

750 

14.92 

0.514 

750 

11.79 

0.286 

1000 

12.73 

0.303 

780 

15.52 

0.556 

780 

12.26 

0.309 

1040 

13.24 

0.328 

810 

16.11 

0.599 

810 

12.73 

0.333 

1080 

13.75 

0.364 

840 

16.71 

0.644 

840 

13.20 

0.358 

1120 

14.20 

0.381 

870 

17.31 

0.690 

870 

13.68 

0.385 

1160 

14.77 

0.408 

900 

17.90 

0.738 

900 

14.15 

0.413 

1200 

15.28 

0.437 

Digitized  by 


Google 


WATER  CONDUCTORS 


137 


For  large  size  pipes  and  high  pressures  the  stress  due  to  the 
swelling  action  is  relatively  small  and  may  be  neglected,  in  which 
case  the  equation  can  be  written, 


d= 


The  friction  losses  may  be  obtained  from  Hazen  and  Williams' 
formula  on  page  116,  and  the  Table  XXXIV  ^  gives  the  discharge, 
velocity  and  loss  of  head  per  100  feet  for  pipes  of  different  di- 
ameters. 

Concrete  Pipe.  Reinforced  concrete  pipes  (Figs.  66  and  67) 
for  power  work  are  used  to  a  limited  extent  for  low-pressure  con- 


PiG.  66.— Concrete  Pipe,  Showing  Steel  Forms  for  Pouring. 

duits,  but  there  is  every  indication  that  they  may  in  the  future 
be  extensively  used  in  place  of  open  flumes  and  canals.  This  will 
not  only  tend  to  increase  the  total  head  of  the  plant,  but  it  will 
prevent  leaves,  branches,  etc.,  from  falling  into  the  conduit,  which 
is  often  the  case  when  they  are  of  the  open  type. 

*  As  given  by  Washington  Pipe  and  Foundry  Company. 


Digitized  by  VjOOQ IC 


138 


WATER  CONDUCTORS  AND  ACCESSORIES 


Concrete  pipes  are  in  use  for  heads  up  to  150  feet.  They  are 
very  smooth,  being  in  this  respect  nearly  on  a  par  with  wooden- 
stave  pipe,  and  thus  offer  Httle  resistance  to  the  flow  of  water. 


Fig.  67. — Cross-section  of  Concrete  Pipe. 

They  are  especially  adapted  for  use  where  raw  material  such  as 
sand,  stone  or  gravel  and  cement  are  available  locally,  in  which 
case  the  pipes  are  generally  manufactured  on  the  job  where  they 
are  used. 

2.  WATERHAMMER  AND  SURGE  TANKS  ^ 

Waterhammer.  When  the  gates  at  the  lower  end  of  a  pen- 
stock are  closed  and  the  water  column  suddenly  checked,  the 
pressure  immediately  rises  and  may  reach  very  high  and  destruc-* 
tive  values  if  not  provided  for  or  prevented.  This  rise  of  pressure 
is  known  imder  the  name  of  "  waterhammer." 

When  the  gate  begins  to  close  the  pressure  rises  first  at  this 
point,  and  a  pressure  wave  or  vibriEttion  begins  to  travel  towards 
the  upper  end  of  the  pipe.  If  the  pipe  is  absolutely  rigid,  the 
velocity  at  which  it  would  travel  would  have  been  about  4650  feet 
per  second,  or  the  same  as  that  of  sound.    On  account  of  the 

*  See  also  sections  on  "  Water  Conductors  "  and  **  Governors." 


Digitized  by 


Google 


WATERHAMMER  AND  SURGE  TANKS  139 

flexibility  of  the  penstock  walls,  however,  the  velocity  is  reduced 
and  may  be  computed  from  the  following  formula: 


'idk 


Where 

a  =  velocity  of  pressure  wave  or  vibration  in  feet  per  second; 

g= acceleration  of  gravity =32.16; 

l/= specific  weight  of  water  =  62.4  pounds  per  cu.  ft.; 

A: = elasticity  of  water  in  compression  =42,000,000  poxmds  per 

sq.  ft. 
d= inside  diameter  of  penstock  in  inches; 
<= thickness  of  plate  in  inches; 
J? = elasticity  of  penstock  material  in  tension: 
For  steel  plate =4,032,000,000  lbs.  per  sq.  ft.= 

(28,000,000  lbs.  per  sq.  in.); 
For  cast  iron  =2,160,000,000  lbs.  per  sq.  ft.= 
(15,000,000  lbs.  per  sq.  in.) 

The  value  of  a  varies  from  2500  to  4000  feet  per  second  as  the 
size  of  pipe  decreases,  and  the  time  required  for  the  pressure  wave 
to  reach  the  top  of  the  penstock  and  return  is  evidently  equal  to 

m        2L 

a 

Where 

Ti  =time  required  for  round  trip  of  pressure  wave  in  seconds; 

L  =  Length  of  penstock  in  feet. 

If  now  the  gate  is  closed  instantaneously,  or,  in  a  time  T,  which 
or 
is  equal  to  or  less  than  — ,  i.e.^  before  the  reflected  pressure  wave 

has  had  time  to  return  to  the  gate  and  reduce  the  pressure  there, 
we  obtain  a  maximum  excess  pressure  head  which  is  equal  to 

while  the  total  pressure  will  be  equal  to  the  above  plus  that  caused 
by  the  static  head,  or 


Digitized  by 


Google 


140  WATER  CONDUCTORS  AND  ACCESSORIES 

where 

Hmax=i^eB,d  corresponding  to  maximum  pressure; 

t;= velocity  of  water  in  penstock  in  feet  per  second,  cor- 
responding to  the  normal  water  flow; 
Ao= static  head  in  feet. 

It  is  impossible  for  the  pressure  to  rise  above  this  value,  Hum. 

2L 
The  time  — ,  therefore,  represents  the  critical  time  in  which  the 
a 

turbine  gates  may  be  closed,  and  it  is  evident  that  the  time  of 

2L 
closure  should  always  be  greater  than  — ,  in  which  case  the  water- 

hanmier  can  never  reach  a  maximum  value. 

or 
When  the  time  is  greater  than  — ,  the  excess  pressure  head 

may  be  calculated  from  the  following  formula  by  Warren:^ 

,  Lv 

and  the  total  pressure  head  becomes: 


The  above  pressure  will  also  be  obtained  if  the  gate  is  only 
closed  partially,  as  long  as  the  closing  is  at  such  a  rate  that  T  is 
the  time  which  it  would  require  to  completely  close  it. 

Example:  Assume  a  steel  pipe  line  having  a  length  of  1000  feet, 
a  diameter  of  4  feet  and  a  plate  thickness  of  J  inch.  The  water 
velocity  is  6  feet  per  second  and  the  net  head  100  feet.  What  is 
the  minimum  time  in  which  the  turbine  gates  may  be  closed,  in 
order  that  the  excess  pressiure  due  to  waterhammer  shall  not  exceed 
50  per  cent  of  the  normal  pressure  due  to  the  net  head? 

The  first  thing  is  to  ascertain  the  velocity  of  the  pressure  wave 
which  is  computed  as  follows: 

°=  //         1     "^'^     48'       y^"^- 

'  ^^•*42,00O,O0O''"iX4,O32,00O,000J 

'  For  derivation  of  formula  see  Transactions  Am.  Soc.  Civil  Engre., 
Vol.  79, 1915,  page  238. 


Digitized  by 


Google 


WATERHAMMER  AND  SURGE  TANKS  141 

The  permissible  excess  pressure  is  equal  to 

^X100=50feet; 

and  thus 

1000X6 


50= 


T=4.1  seconds. 

An  extensively  used  formula  for  calculating  waterhammer  is 
also  the  following  one,  derived  by  L.  Allievi: 


»..5^«+sf^. 


where 

This  formula  is  applicable  for  a  slow  closing  of  the  valve  when 

2Z/ 
T  is  considerably  greater  than  — ,  but  may  be  incorrect  for  a 

a 

or 
quick  closing  as  when  the  value  of  T  is  close  to  — . 

Surge  Tanks.  In  plants  with  long  pipe  lines  under  medium 
and  high  heads  it  is  often  foimd  that  not  only  the  pressure  rise, 
but  also  the  pressure  drop  will  be  excessive,  and  in  such  cases 
it  may  be  necessary  to  provide  both  a  relief  valve  and  a  surge 
tank  to  equalize  the  pressure  variation.  Synchronous  relief 
valves  (see  page  258)  are,  of  course,  only  of  use  against  a  pressure 
rise  when  the  load  is  going  off  and  not  when  the  load  is  coming  on, 
because  they  cannot  supply  to  the  moving  water  colunm  the 
kinetic  energy  which  it  has  lost  and  which  it  must  regain  before  it 
can  flow  at  the  higher  velocity  required  by  an  increase  of  load. 
To  accomplish  this,  surge  tanks,  or  standpipes  as  they  are  also 
conmionly  termed,  must  be  used. 

There  are  two  kinds  of  surge  tanks,  the  simple  and  the  differ- 
ential. The  former  consists  of  an  open  standpipe  or  storage  tank 
placed  at  the  downstream  end  of  the  pipe  line  (Fig.  68).  When 
the  gates  are  closed  the  inertia  of  the  water  column  in  the  penstock 
causes  a  rise  of  the  water  in  the  standpipe,  and  the  velocity  is 
thus  gradually  reduced.    On  the  other  hand,  when  the  load  comes 


Digitized  by 


Google 


142 


WATER  CONDUCTORS  AND  ACCESSORIES 


on  suddenly,  the  standpipe  furnishes  the  water  quickly  without 
waiting  for  the  velocity  in  the  long  pipe  line  to  pick  up,  and  thus 
greatly  aids  the  regulation. 

To  be  most  efficient,  the  surge  tank  should  be  located  as  near 
the  power-house  as  possible,  and  if  operating  under  atmospheric 
pressure,  its  height  should  evidently  be  above  that  of  the  high- 
water  level  in  the  forebay  or  storage  pond.    It  is  obvious,  however. 


Fig.  68. — Pressure  Variations  with  Stand  Pipe. 

that  such  an  open  design  would  not  be  feasible  for  high-head  devel- 
opments, and  in  such  cases  a  closed  standpipe  is  usually  provided, 
the  increased  air  pressure  being  obtained  by  the  static  head.  In 
many  plants  both  open  and  closed  surge  tanks  are  provided,  the 
open  type  being  installed  at  the  upper  end  of  the  pipe  line,  where  it 
passes  over  the  brow  of  the  hill  above  the  power-house,  while 
closed  air  chambers^  are  installed  just  outeide  the  power-house. 
For  pipe  lines  several  miles  in  length  it  is  also  advisable  to  provide 
equalizing  reservoirs  at  intervals  along  the  pipe  line,  so  that  changes 
in  the  velocity  of  the  water  coliunn  will  be  as  gradual  as  possible. 
The  differential  surge  tank  consists  of  a  standpipe  of  about  the 
same  diameter  as  the  conduit,  freely  connected  to  it,  and  a  storage 
tank  of  larger  dimensions,  surrounding  the  standpipe  and  con- 
nected to  the  conduit  by  a  properly  restricted  passage.  In  a 
simple  tank,  the  level  of  the  stored  water,  following  a  demand  for 
more  power,  represents  the  accelerating  level  which  is  urging  more 
water  from  the  forebay,  and  measures  the  head  acting  on  the 
water  wheel.    In  the  differential  type,  owing  to  the  resistance 

^  For  the  use  of  air  tanks  for  pipe  line  regulating  purposes,  nee  Proceed- 
ings American  Society  of  Civil  Engineers  for  August,  1917. 


Digitized  by 


Google 


WATERHAMMER  AND  SURGE  TANKS  143 

interposed  between  tank  and  ponduit,  the  level  of  the  stored 
water  is  quite  independent  of  the  acceleration,  and  does  not  affect 
the  waterwheel  governor  directly.  The  water  in  the  standpipe 
takes  care  of  these  things,  and  acts  Uke  a  simple  tank  of  small 
dimensions  which  is  supplemented  by  the  steadying  action  of  the 
stored  water,  fed  into  the  system  in  an  independent,  nonnsyn- 
chronous  manner,  meeting  all  demands  for  water  without  causing 
the  unstable  pendulum-like  behavior  which  is  so  characteristic 
of  the  simple  surge  tank. 

Mr.  R.  D.  Johnson^  has  derived  the  following  equation  for 
determining  the  maximum  surge  in  simple  surge  tanks: 


^-4 


^^^,,. 


Ag 
where 

5ni»x  =  maximum  surge  up  or  down,  in  feet,  measured  in  starting, 
from   reservoir  or  head-water  level,  and,  jn  stopping, 
from  a  distance  below  this  equal  to  the  friction  head,  h/; 
P  =  cross-sectional  area  of  pipe  Une,  in  square  feet; 
L= length  of  pipe  Une  in  feet;   • 
r= velocity  of  water  in  pipe,  in  feet  per  second; 
A  =  cross-section  area  of  surge  tank,  in  square  feet; 
flf  =  acceleration  of  gravity; 

A/ =  total  feet  of  head  lost  due  to  friction  in  pipe  between  res- 
ervoir and  siurge  tank. 

Fig.  68  illustrates  the  pressure  variations  with  simple  surge 
tanks,  the  upper  curve  to  the  right  illustrating  the  rise  in  pressure 
with  the  closing  of  the  gates  and  the  lower  curve,  the  drop  in 
pressure  due  to  the  opening  thereof. 

Figs.  69  and  27  show  the  design  and  arrangement  of  a  large 
differential  surge  tank.  This  particular  tank  consists  of  a  cylin- 
drical shell,  50  feet  in  diameter  and  80  feet  high,  with  a  hemi- 
spherical bottom  which  adds  25  feet  to  the  height,  and  its  capacity 
is,  therefore,  1,400,000  gallons.  The  tank  is  supported  on  ten 
colunms  with  heavy  concrete  footings.  It  and  the  riser  are  housed 
in  with  a  frame  wooden  structure  providing  a  surrounding  air 
space  which  can  be  heated  when  necessary  from  a  small  house* 
below.     The  top  of  the  roof  of  this  structure  is  205  feet  above  the 

^  American  Society  of  Civil  Engineers,  Vol.  79,  1915,  p.  265. 


Digitized  by 


Google 


144 


WATER  CONDUCTORS  AND  ACCESSORIES 


ground,  and  the  top  of  the  tank  is  high  enough  above  the  crest 
of  the  dam  so  that  if  the  flow  of  the  water  in  the  pipe  hne  were 


FiQ.  69. — Cross-section  of  Salmoo  River  Power  Company's  Development. 

suddenly  interrupted  its  energy  would  be  absorbed  by  the  rise 
in  level  in  the  tank  without  overflow. 

A  complete  treatise  of  the  surge  tank  problem  is  to  be  found 
in  two  excellent  papers  by  Messrs.  R.  D.  Johnson  and  M.  M. 
Warren  in  the  Transactions  of  the  American  Society  of  Civil 
Engineers,  Vols.  78  and  79,  1915. 

3.   GATES  AND  VALVES  ^ 

Reqtiirements.  For  the  control  of  water  flow  in  hydro- 
electric developments  gates  and  valves  are  generally  used.  They 
may  be  either  of  the  sluice  gate  or  gate  valve  type  and  the  selection 
of  the  type,  as  well  as  the  number  required,  is  governed  by  the 
nature  of  the  development.  So,  for  example,  in  low-head  plants, 
only  one  set  of  sluice  gates  are,  as  a  rule,  needed,  these  being 
installed  in  front  of  the  turbine  intakes,  either  in  a  gatehouse,  as 
in  Fig.  89,  or  outside  the  power-house  building  at  the  dam  struc- 
ture, as  in  Fig.  70. 

For  high-head  developments,  however,  two  and  sometimes 
three  sets  of  controlling  devices  are  required,  depending  on  the 
pipe-line  arrangement,  and  in  order  that  the  water  may  be  projH 

1  See  also  section  on  "  Flashboards." 


Digitized  by 


Google 


GATES  AND  VALVES 


145 


erly  shut  oflf  in  case  of  emergency  should,  for  example,  one  of  the 
valves  become  damaged  or  stick.  In  such  plants  sluice  gates 
are  installed  as  headgates  at  the  forebay  or  reservoir  intake,  while 
gate  valves  are  provided  in  the  pipe  line  at  the  point  where  this 
branches  off  to  the  different  turbine  units,  and  sometimes  also 
at  a  point  close  to  the  wheel  casing  in  addition. 

The  gates  should  be  of  suflScient  size  to  pass  the  required  max- 


Fia.  70. — Sectional  Elevation  of  Power  House,  Turners  Falls  Power  and 
Electric  Company,  Showing  Headgate  Arrangement. 

imum  flow  of  water,  and  also  of  suflScient  strength  to  withstand 
the  shocks  and  excessive  pressures  resulting  from  a  quick  closing 
in  case  of  emergency.  This  is  a  point  which  must  be  considered 
in  determining  the  minimum  time  in  which  the  gates  may  be 
closed.  As  mentioned  under  the  chapter  on  Waterhammer  and 
Surge  Tanks,  the  longer  time  allowed  for  closing  the  gates  the 
less  will  the  excessive  pressure  caused  by  waterhammer  be. 


Digitized  by 


Google 


146 


WATER  CONDUCTORS  AND  ACCESSORIES 


Sluice  Gates.  These  may  be  either  of  structural  steel  or  cast 
iron,  the  former  generally  being  used  for  large  intake  openings. 
With  low-head  developments  these  openings  are  now  generally 
divided  in  a  number  of  vertical  sections  in  order  to  insure  a  more 
even  distribution  of  the  water  to  the  speed  ring  of  the  turbine, 


Fig.  71. — Rising-Stem  Sluice  Gate 
with  Floor  Stand.  (Ludlow  Valve 
Mfg.  Co.) 


Fig.  72.— Rising  Stem  Sluice 
Gate.  Side  View  of  Gate 
Shown  in  Fig.  71. 


and  this,  of  course,  also  very  considerably  reduces  the  size  of  the 
gates,  one  set  being  provided  for  each  section.  Sometimes  the 
sections  are  also  divided  horizontally,  as  shown  in  Fig.  26,  in 
order  to  still  further  reduce  the  size  of  the  gates.  At  the  junction 
of  the  upper  and  lower  sections  there  is  a  reinforced  concrete 
beam  which  serves  as  a  support  and  as  a  seal.     The  two  gate 


Digitized  by 


Google 


GATES  AND  VALVES 


147 


sections  are  provided  with  separate  guide  slots  so  that  they  may 
be  manipulated  independently.  This  type  of  gate  is  generally 
lifted  by  means  of  chains. 

The  gates  shown  in  Fig.  70  are  the  Broome  type  and  are  con- 
structed of  heavy  steel  plates  run  on  a  continuous  chain  of  rollers 
between  the  tracks  on  the  gates  and  guides.     A  gantry  crane, 


Fig.  73. — Sectional  Elevation  of  Power  House  of  the  Hydro-Electric  Com- 
pany of  West  Virginia,  Showing  Application  of  Tainter  Gate. 

electrically  driven  and  running  on  the  head  wall,  operates  the 
gates.     This  crane  also  carries  a  mechanical  rack-raking  device. 

Gates  which  are  raised  or  lowered  by  means  of  stems  may  be 
either  of  the  rising  or  non-rising  stem  type,  the  former  being 
preferable  at  intakes  where  there  is  no  danger  of  the  operating 
stands  being  submerged,  and  where  the  rising  stem  may  serve 


Digitized  by  CjOOQ IC 


148 


WATER  CX)NDUCTORS  AND  ACCESSORIES 


as  an  indicator  of  the  gate  position  (see  Figs.  71  and  72).  For 
gates  which  are  installed  in  diversion  dams  for  sluicing  off  excess 
flood  water  in  forebay  ponds  or  reservoirs,  the  non-rising  type  is 

preferable,  as  it  permits  be- 
ing submerged  without  being 
damaged  by  floating  ice. 

Tainter  Gates.  This  type 
of  gate  is  occasionally  used  for 
controlling  the  water  passages 
to  the  wheel  chambers  in 
low-head  developments,  the 
methods  of  application  being 
shown  in  Fig.  73.  They  are, 
however,  more  used  in  connec- 
tion with  diversion  dams. 

Gate  Valves.  There  are 
niunerous  different  designs  of 
gate  valves,  the  details  of  one 
of  the  most  improved  designs 
being  illustrated  in  Fig.  74.  It 
is  intended  for  high  pressures 
and  consists  of  tlie  stem,  a 
double  disc  and  two  bevel- 
faced  wedges,  the  wedges  being 
entirely  independent  of  the 
discs  and  working  between 
them. 

By  the  action  of  the  stem, 
which  works  through  a  nut  in 
the  upper  wedge,  the  discs  de- 
scend parallel  with  their  seats 
until    the  lower  wedge  strikes 
the  stop  in  the  bottom  of  the 
case.      The   discs  and   upper 
wedge,  however,  continue  their 
downward  movement  until  the 
face    or   bevel    of    the    upper 
wedge  comes  in  contact  with  the  face  or  bevel  of  the  lower 
wedge.    The  discs  then  being  down  opposite  the  valve  opening, 
the  face  of  the  upper  wedge  moves  across  the  face  of  the  lower 


Fig.  74. — Ludlow  Bronze  Mounted 
Doubl3  Gate  Valve  with  Bolted 
Stuffing  Box. 

A— Caae.  B— Cover  of  Bonnet.  C — 
Stem  or  Spindle.  D— Packing  Plate  or 
Stuffing  Box.  J]— Stuffing  Box  Gland  or 
Follower.  F— Stem  Nut.  GG— Gates. 
H — Gate  Rings.  I — Case  Rings.  J — Top 
Wedge.  K— Bottom  Wedge.  L— Throat 
Flange  Bolts.  M— Stuffing  Box  or  Fol- 
lower Bolts. 


Digitized  by 


Google 


GATES  AND  VALVES 


149 


wedge,  bringing  pressure  to  bear  on  the  backs  of  both  discs,  from 
central  bearings,  thus  forcing  them  apart  and  squarely  against 
their  seats. 

In  opening  the  valve,  the  first  turn  of  the  stem  releases  the 
upper  wedge  from  contact  with  the  lower  wedge,  thereby  instantly 
releasing  both  discs  from  their  seats  before  they  commence  to  rise. 

All  gate  valves  and  sluice  gates  should  be  fully  bronze-mounted 
to  prevent  corrosion.  That  is,  the  disc  and  seat  rings  should  be 
made  of  bronze,  as  well  as  the  threaded  portion  of  the  stem,  the 
operating  nut  and  the  wedging  apphances. 

Where  the  water  pressure  is  very  great,  by-pass  valves  may 
be  provided  for  equalizing  the  pressure  on  both  sides  of  the  valve 
before  it  is  opened. 

Operatioii  and  ControL  Sluice  gates  and  gate  valves  may  be 
operated  either  by  hand,  water  or  electrically,  the  two  latter 


Fig.   75. — Gatehouse,  Showing  Gate-Lowering  Mechanism.      Mississippi 
River  Power  Company. 

methods  being  used  extensively,  resulting  in  a  saving  of  labor, 
while  on  the  other  hand  acting  as  a  protection  in  the  case  of 


Digitized  by 


Google 


150 


WATER] CX)NDUCTORS  AND  ACCESSORIES] 


trouble.  This  is  evident  by  considering  that  some  large  valves 
would  require  hours  to  close  by  hand.  When  sluice  gates  are 
installed  in  gatehouses  a  traveling  crane  is  often  provided  for 
lifting  them.    Their  dosing  is  then  done   through  their  own 

weight,  and  brakes  are  installed 
for  regulating  the  same  (Fig.  75). 
Whatever  method  of  operation 
is  chosen  it  should  be  simple  and 
positive  in  its  action. 

There  are  numerous  hand- 
operated  lifting  devices  such  as 
rack-and-pinion  with  an  operating 
lever,  windlass,  floorstands  with 
threaded  gate  stems  and  operating 
wheels,  etc.  Gear  trains  should 
always  be  provided  where  there  is 
considerable  pressure  on  the  gate, 
or,  otherwise,  it  may  be  impos- 
sible for  the  operator  to  start  the 
gates  especially  when  Ihey  have 
been  closed  for  some  time. 
Arrangements  are,  however,  gen- 
erally made  for  shifting  the  hand- 
wheel  directly  to  the  stem  after 
the  gate  has  been  opened  slightly, 
in  order  that  the  opening  may  be 
accomplished  more  rapidly.  Roll- 
ers and  ball  bearings  are  also 
sometimes  provided,  either  with 
the  discs  or  the  Ufting  devices  so 
as  to  reduce  the  friction. 

Gate  valves  may  also  be 
operated  by  hydraulic  cyUnders. 
The  regulating  valve  consists  of  a  flat  valve  which  is  operated 
by  a  piston,  this  in  turn  being  moved  by  releasing  the  pressure 
on  either  side  by  means  of  small  poppet  valves  which  may  be 
operated  by  hand  or  electrically  f)'om  any  convenient  point.  A 
valve  of  the  latter  type  is  shown  in  Fig.  76,  and  the  electrical  con- 
nections in  Fig.  77. 

A  double-throw  control  switch  and  an  alarm  bell  are  mounted 


Fig.  76.  —  Ludlow  Hydraulic 
Operated  Cylinder  Valve  with 
Electric  Control. 


Digitized  by 


Google 


GATES  AND  VALVES 


151 


on  a  panel.  The  upper  contact  on  the  switch  operates  the  valve, 
and  the  lower  is  for  the  bell,  which  will  ring  only  when  the  valve  is 
closed  or  open.  This  depends  on  how  the  connections  are  made, 
and  the  operator  can  at  all  times  ascertain  what  position  the  valve 
is  in.  If  the  valve  is  connected  so  that  the  bell  will  ring  when  the 
valve  is  closed,  and  the  operator  closes  the  bell  circuit,  and  the 
bell  will  not  ring,  he  will  readily  understand  the  valve  is  open; 
if  he  closes  the  valve  circuit,  or  upper  pole  pf  switch  for  a  few 
seconds,  and  again  closes  the  lower  or  bell  circuit  and  the  bell 


ValTtf  CIrenIt 


Fig.  77. — ConnectioDs  for  Electro-Hydraulic  Valve  Shown  in  Fig.  76. 

rings,  he  will  understand  the  valve  is  closed.  The  valve  can  be 
operated  by  hand  simply  by  lifting  the  small  armature  on  the  con- 
trolling device. 

Cylinder  valves  are,  as  a  rule,  more  economical  for  smaller 
Hizes,  while  for  larger  the  electric  motor  operated  valve  (Fig.  78) 
is  to  be  recommended.  Such  valves  are  very  reliable  and  can  be 
closed  in  a  comparatively  short  time.  Besides  this,  remote 
control  from  the  main  control  board  in  the  power-house  can  readily 
be  provided. 

The  service  of  valve  motors  is  exceedingly  intermittent  and 
may  vary  from  comparatively  short  intervals,  such  as  once  every 


Digitized  by 


Google 


152 


WATER  CONDUCTORS  AND  ACCESSORIES 


hour,  to  weeks  or  even  months.  When  the  apparatus  at  the  end 
of  long  periods  of  idleness  is  called  upon  to  operate  it  must  per- 
form its  function  without 
fail,  and  must,  therefore 
be  designed  accordingly, 
totally  inclosed  motors  of 
a  moisture-proof  design 
being  preferable.  Metal- 
line bearings  are  generally 
used,  as  the  motors  may 
be  mounted  in  any  posi- 
tion from  vertical  to  hori- 
zontal. Due  to  the  inter- 
mittent natiu^  of  the  ser- 
vice, efficiency  or  power 
factor  need  not  be  con- 
sidered, the  main  consid- 
eration being  reliable 
operation. 

The  proper  size  of  a 
motor  for  driving  a  valve 
will  vary  with  the  duty 
and  conditions  under 
which  the  valve  operates. 
A  small  valve  may  only 
require  a  1 -horse-power 
motor,  while  very  large 
valves  require  up  to  25- 
H.P.  motors.  The  size  of 
the  valve  is,  however,  not 
the  only  factor  determin- 
ing the  required  motor 
capacity,  which  also  de- 
pends to  a  very  large  ex- 
tent on  the  pressure  on 
the  valve  and  the  time  of 


Fig.  78. — Heavy  Pressure  Motor-operated 
Gate  Valve  Showing  Motor  Equipment 
and  Limit  Switch. 


openmg. 
The  torque  requirements  vary  greatly  during  the  operating 
cycle.     It  is  maximimi  shortly  after  the  time  of  imseating  the 
valve;  that  is,  after  the  wedges  have  been  released  and  the  actual 


Digitized  by 


Google 


GATES  AND  VALVES 


153 


motion  begins.  It  then  drops  somewhat  until  the  valve  has 
opened  about  one-fourth,  after  which  it  takes  comparatively 
little  power  to  complete  the  opening,  as  the  pressure  on  the  valve 
is  then  comparatively  small.  When  closing  the  valve,  friction 
only  needs  to  be  overcome  in  starting  and  there  is  no  pressure  on 


BsmoM  OoBtrol  FMial 


Op«i 


limit  Swlteh 

Fig.  79. — A.  C.  One-station  Remote  Control  Equipment. 

the  valve  until  it  has  begim  to  close.  The  torque  is  therefore 
not  very  high  imtil  the  valve  is  about  three-fourths  closed,  after 
which  the  pressure  causes  the  torque  to  increase  rapidly.  At  the 
end  of  the  closing  cycle  the  torque  does  not,  however,  reach  the 
value  it  did  during  the  period  of  starting. 

Valve  motors  are,  therefore,  generally  rated  for  maximum 


Digitized  by 


Google 


154 


WATER  CONDUCTORS  AND  ACCESSORIES 


starting  torque  and  either  direct  or  alternating  current  motors 
may  be  used.  The  former  are  mostly  compound  wound  with  a 
sufficient  shunt  field  to  limit  the  speed  at  light  load.  With  the 
latter  the  squirrel-cage  induction  motor  seems  to  be  most  widely 
used  for  small  and  mediimi-size  valvef?,  principally  on  account  of 
its  simplicity.  It  should  be  designed  with  a  high-resistance 
rotor  to  increase  the  starting  torque,  and  it  is  generally  found 


B«mot«  Control  PmB«l 


Clow-^ 


AraTl   I 

I — :qjivww« 


Limit  Switch 

Fig.  80. — D.  C.  One-station  Remote  Control  Equipment. 

necessary  to  select  a  motor  somewhat  larger  than  what  would  be 
the  case  with  a  compound-wound  direct-current  motor  to  perform 
the  same  duty. 

With  certain  valves  it  becomes  necessary  to  overcome  the 
sticking  due  to  wedging  action  when  opening,  and  the  drive  is 
therefore  provided  with  a  "  lost  motion  "  so  as  to  give  a  hammer- 
blow.    For  alternating  current  motors  this  is  furthermore  of  value 


Digitized  by 


Google 


GATES  AND  VALVES 


155 


in  that  it  permits  the  motor  to  speed  up  some  and  gain  in  torque 
before  the  load  comes  on,  the  maximum  torque,  as  a  rule,  occurring 
slightly  above  zero  speed. 

Valve  motors  are  generally  thrown  directly  on  the  line,  and 
the  control  is  accomplished  by  means  of  contactors  for  remote 
control  and  large  equipments.  For  hand  control  of  smaller  equip- 
ments, ordinary  knife  switches  are  sufficient.  Fuses  give  better 
protection  than  automatic  circuit  breakers,  in  that  they  will 
protect  against  a  stalled  motor  but  will  not  blow  during  start  or 
running. 

Limit  switches  which  will  open  the  circuit  when  the  gate  has 
reached  its  limit  of  travel  should  always  be  provided.  Such 
switches  are  geared  to  the  valve  stem  and  arranged  to  open  the 
contactors  at  a  predetermined  point  of  travel  of  the  gate  in  either 
direction.  Provision  is  also  made  so  that  the  open  or  closed  valve 
positions  are  indicated  on  the  control  board  by  means  of  two  lamps. 
When  only  one  lamp  bums  it  indicates  open  or  closed  valve  posi- 
tion, Es  the  case  may  be, 
while  both  lamps  burn  in 
any  mid  position. 

Connection  diagrams  for 
D.C.  and  A.C.  remote-con- 
trol equipments  are  shown  in 
Fig3.  79  and  80.  These  are 
for  single-station  control,  and 
for  multiple-station  ^control 
push  buttons  are  substi- 
tuted for  the  single-pole 
double-throw  pilot  switch. 

Pivot  Valve.  A  type  of 
valve  used  in  a  number  of 
large  plants  for  the  purpose 
of  shutting  off  the  turbine 
from  the  penstock  is  the 
pivot  or  "  butterfly  "  type  of 
valve,  illustrated  in  Fig.  81. 

This  type  of  valve  is  simple  in  construction  and  takes  up  very 
little  space.  The  operation  is  by  means  of  a  hydraulic  cylinder, 
having  a  trunk  piston  connected  to  the  crank  or  lever  shown.  It 
is  reliable  in  service,  but  is  not  as  tight  against  leakage  as  either 


Fig.  81.— Pivot  Valve.   (Built  by  I. 
Morris  Company.) 


Digitized  by 


Google 


156 


WATER  CONDUCTORS  AND  ACCESSORIES 


the  Johnson  valve  or  gate  valve.  In  cases  where  leakage  through 
the  valve  can  be  carried  off  through  an  ample  drain,  the  valve 
can  be  used  very  satisfactorily. 

The  Johnson  Hydraulic  Valve.    This  valve  consists  essen- 
tially of  a  circular  body  forming  an  enlargement  of  the  pipe  line 


Open 


Closed 


Fig.  82. — Johnson  Hydraulic  Valve.      (Wellman-Seaver-Morgan  Company.) 

or  penstock,  and  having  an  internal  cylindrical  chamber  contain- 
ing a  sliding  plunger  (Fig.  82).  The  closed  end  of  the  internal 
chamber  and  the  nose  of  the  plunger  are  of  conical  form.  They 
are  designed  to  guide  the  water  smoothly  as  it  enters  and  leaves 


Digitized  by 


Google 


GATES  AND  VALVES  157 

the  valve.  The  waterways  throughout  the  valve  offer  no  obstruc- 
tion to  the  flow  and  consequently  there  is  no  appreciable  loss  of 
head. 

No  external  source  of  power  is  required  for  operation.  When 
the  plimger  is  withdrawn  into  the  internal  or  operating  chamber, 
the  valve  is  open  and  presents  an  unobstructed  passage  for  the 
water.  When  the  plunger  protrudes  from  the  operating  chamber, 
it  seats  against  a  ground  ring  in  the  neck  of  the  valve  body,  form- 
ing a  water-tight  joint.  The  standard  control  mechanism  pro- 
vides for  only  the  open  and  closed  positions  of  the  plunger,  but  it 
may  be  specially  arranged  to  hold  the  plunger  at  intermediate 
positions  if  desired. 

The  valve  plimger  is  of  the  differential  type,  forming  an  annular 
chamber  A  within  the  operating  cylinder,  in  addition  to  the  cen- 
tral chamber  B.  By  means  of  a  suitable  external  control  valve 
and  piping,  either  pipe-line  pressure  or  atmospheric  pressure  may 
be  alternately  applied  to  the  chambers  A  and  B.  Admitting  pipe- 
Une  pressure  to  A  and  exhausting  it  from  B  opens  the  valve; 
reversing  the  operation  closes  it. 

The  external  control  valve  may  be  operated  by  hand  or 
electricity  and  may,  therefore,  be  located  remotely  from  the 
valve  as  from  the  switchboard,  if  desired. 

Another  application  of  this  valve  is  for  automatic  pressure 
relief.  The  valve  plunger  is  held  closed  by  air  pressure  so  arranged 
that  it  is  automatically  released,  permitting  the  valve  plunger  to 
open  when  the  pipe-line  pressure  exceeds  normal  by  some  pre- 
determined margin.  The  advantage  of  using  air,  rather  than 
water,  lies  in  the  rapidity  with  which  air  may  be  discharged,  and 
the  consequent  rapid  opening  of  the  relief  valve. 

Air  Valves.  In  addition  to  sluice  gates  and  gate  valves  pre- 
viously described,  air  valves  are  often  required  in  connection  with 
the  pipe  lines  of  hydro-electric  developments.  These  may  be  of 
two  kinds:  the  automatic  lever  and  float  valve  and  the  automatic 
poppet  valve. 

The  former  is  for  use  on  pipe  lines  which  follow  the  contour 
of  hilly  country  and  where  air  may  accumulate  at  nigh  summits 
and  obstruct  the  flow  of  water.  The  valve  is  connected  to  the 
outside  of  the  pipe  at  its  highest  point  or  points,  and  when  air 
takes  the  place  of  water  about  the  float  in  the  valve  chamber,  the 
float  which  is  attached  to  a  lever  drops,  thus  opening  a  small  valve, 


Digitized  by 


Google 


158  WATER  CONDUCTORS  AND  ACCESSORIES 

allowing  the  air  to  escape.    As  the  water  returns,  it  lifts  the  float 
thereby  closing  the  valve. 

The  poppet  valve,  on  the  other  hand,  is  intended  for  use  on 
pipe  lines  to  permit  air  to  enter  when  water  is  being  drawn  off 
and  thus  eliminates  any  danger  of  collapse  from  vacuum  forming 
in  the  pipe  lines  as,  for  example,  when  the  head  gates  are  closed. 
Similarly,  they  may  be  provided  to  allow  air  to  escape  when  the 
pipes  are  being  filled.  The  valve  remains  open  until  the  water 
reaches  and  lifts  the  copper  float  and  closes  the  same,  after  which 
it  remains  closed  while  the  pressing  is  on.. 


Digitized  by 


Google 


CHAPTER  VI 
STORAGE  RESERVOIRS  1 

Many  watersheds  have  some  natural  storage  features  tending 
to  equalize  the  stream-flow  as  compared  with  the  rainfall,  while 
with  others  surplus  water  in  times  of  high  flow  can  only  be  held 
back  for  use  in  times  of  low  flow  by  the  construction  of  artificial 
reservoirs. 

Storage  and  Pondage.  The  impounding  and  acciunulation 
of  surplus  water  which  may  be  utilized  when  needed  is  termed 
either  "  storage "  or  "  pondage."  The  former  generally  refers 
to  reservoirs  located  on  a  watershed  at  some  distance  from  the 
power-house,  and  where  large  quantities  of  water  may  be  im- 
pounded for  use  during  the  dry  season.  "  Pondage,"  on  the  other 
hand,  refers  to  the  storage  for  taking  care  of  the  daily  fluctuation 
in  the  load  curve,  otherwise  canals,  flumes  and  pipe  lines  will 
have  to  carry  the  peak  flow  of  water  instead  of  the  average.  It  is 
often  the  case  that  the  average  demand  for  power  during  twelve 
or  fourteen  hours  of  the  day  is  twice  as  great  as  the  demand  for 
the  remaining  ten  or  twelve  hours.  The  small  volume  of  power 
required  during  a  portion  of  the  day  permits  an  accumulation  of 
water  at 'the  power  dam  itself  which  can  be  used  as  a  reserve  force 
to  meet  the  higher  demand  during  the  other  portions  of  the  day. 
Thus,  a  stream  that  during  the  twenty-four  hours  might  develop 
a  continuous  horse-power  would,  if  relieved  of  half  of  the  demand 
for  half  of  the  day,  be  able,  with  small  pondage,  to  supply  con- 
siderably more  than  the  average  during  the  remaining  portion  of 
the  day. 

The  importance  of  pondage  should,  however,  not  be  exag- 
gerated, as  it  can  only  be  utilized  at  the  expense  of  operating  head, 
but  to  counteract  this  it  is  possible  to  provide  temporary  flash- 
boards  by  which  the  normal  level  may  be  raised  several  feet. 

The  storage  is,  however,  of  the  greatest  importance,  as  it  wiU 
usually  greatly  increase  the  earning  capacity  of  any  development. 

^  See  also  section  on  **  Water  Storage/' 
159 


Digitized  by 


Google 


160 


STORAGE  RESERVOIRS 


Limitations  to  Storage.  There  is,  however,  a  limit  to  storage 
and  in  no  ease  can  sufficient  impounding  be  maintained  to  give 
to  any  stream  the  power  representing  anything  like  its  maximum 
flow.  The  excess  run-oflf  from  any  watershed  varies  greatly  from 
year  to  year,  and  it  is  generally  considered  to  be  the  best  prac- 
tice to  base  the  reservoir  capacity  on  the  run-ofif  for  the  minimum 
year,  as  impounding  the  water  in  years  of  heavy  run-off  for  holding 
over  in  storage  to  dry  seasons  is  generally  considered  uneconomical, 
among  other  things  on  account  of  the  loss  due  to  evaporation. 
In  general,  there  are  two  factors  determining  the  practicable 
amount  of  storage.  The  first  consideration  ia  usually  the  topog- 
raphy of  the  locality.  In  some  locaUties  a  sufficiently  high  dam 
may  be  built  at  a  very  reasonable  cost,  and  it  may  provide  storage 
for  an  immense  volume  of  water  and  thus  greatly  enhance  the 
minimum  power  of  the  stream.  In  other  cases  the  conditions 
may  be  entirely  the  reverse.    A  further  practical  consideration 

is  the  value  of  the 
land.  Even  with 
favorable  topographic 
conditions  the  cost  of 
acquiring  lands  to  be 
flooded  may  be  so 
great  as  to  make  any 
great  amount  of  stor- 
age impracticable. 

Location  of  Reser- 
voir. The  relative 
location  of  the  pro- 
posed reservoir  site  in 
the  drainage  area 
must,  of  course,  also 
be  considered,  and 
likewise  its  location 
with  respect  to  the 
point  of  distribution 
so  that  proper  outlets  and  conduits  can  be  provided  at  a  reason- 
able cost. 

Before  accurate  surveys  are  justified,  it  may  become  desirable 
to  approximately  determine  the  quantity  of  water  that  a  proposed 
reservoir  may  hold.    This  is  usually  done  by  means  of  contour 


Bw. 

M                                   f\ 

8«!.M 

1 

5i 

f/x^               Sec. 

-^      H        V 

Boalc  oC  Milei 

Sec.  91 

Fig.  83. — Ck)ntour  Map  of  Reeervoir  Site. 


Digitized  by 


Google 


STORAGE  RESERVOIRS 


161 


maps,  the  topography  being  taken  by  means  of  transits  and  stadia, 
and  the  contours  plotted  as  in  Fig.  83.  The  area  is  found  by 
planimeters  and  the  volume  by  multiplying  the  vertical  distance 
between  the  contour  levels  with  the  mean  area  of  the  sections.  A 
certain  dead  space  must  be  allowed  at  the  bottom  of  the  reservoir 
as  it  is  not  advisable  to  draw  off  the  water  from  the  bottom  level 
on  account  of  the  silt  and  mud  which  accumulates  there.  The 
foUowing  table  gives  the  capacity  of  the  reservoir  site  outlined 
in  the  above  figure: 

TABLE  XXXV 
RxBEBVora  Capactft 


Height  of  Water 
above  Bottom  in  Feet. 

Area  in  Acres. 

Capacity  of  Section 
in  Acre-feet. 

Total  Capacity 
in  Acre-feet. 

10 

10 

0 

0 

20 

36 

230 

230 

30 

74 

550 

780 

35 

110 

460 

1240 

50 

188 

2235 

3475 

60 

274 

2310 

5785 

The  imit  measure  of  stored  water  is  generally  the  "  acre-foot," 
representing  43,560  cubic  feet,  and  the  curves  in  Figs.  84  and  85, 
show  the  kilowatt-hours  for  different  acre-feet  storage  on  various 
heads,  and  vice  versa,  the  over-all  hydro-electric  efficiency  being 
assumed  to  be  65  per  cent. 

It  has  also  been  proposed  to  adopt  the  ''square-mile  foot" 
as  a  unit  for  expressing  large  quantities  of  stored  water.  This" 
is  equivalent  to  27,878,400  cubic  feet  or,  640  acre-feet. 

The  building  of  storage  reservoirs  involves  many  engineering 
problems,  the  most  important  being  the  dam  construction,  which 
was  treated  in  Chapter  IV.  Spillways  must  be  provided  for  dis- 
charging excess  flood  waters,  and  with  earthen  dams  or  masonry 
dams  of  considerable  height,  outlets,  in  the  form  of  tunnels  or 
otherwise,  are  generally  provided  some  distance  from  the  dam  to 
prevent  any  possibility  of  damage  to  the  same.  Provision  must 
also  be  made  for  outlets  at  the  bottom  of  the  reservoir,  so  that 
excess  accumulation  of  silt  and  mud  may  be  sluiced  away. 


Digitized  by 


Google 


162 


STORAGE  RESERVOIRS 


/y 

/ 

y 

OQOQMn 

7" 

r 

r 

r 

/  1 

/// 

7/ 

J 

f 

/ 

— 

— 

— 

— 

— 

4 

4 

■/ 

1 

/ 

/ 

J 

//'- 

7/ 

/ 

/ 

f 

/ 

J 

f 

r 

b// 

!// 

/ 

/ 

I 

y 

J 

/ 

/ 

!  mil  /I  / 

/ 

/ 

|flOi»000 

^a 

f 

/ 

/ 

J 

?  / 

/ 

J 

f 

h 

J 

J 

/ 

J 

/ 

^Ml  J 

f/ 

/ 

/ 

1 

r* 

4 

/ 

/ 

/ 

^  if^f^i 

[*/ 

/ 

/ 

- 

J 

Ij 

y 

iB 

I 

/ 

lull  ni  i^ 

r 

/ 

Ij 

/ 

n 

f 

/ 

1//// 

y 

/ 

y 

h 

/ 

/ 

} 

^ 

J 

A 

f//  ' 

r 

*"'! 

^ 

/ 

i i 

f  i 

i 

/ 

/ 

7 

V 

y 

/j 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

4^ 

Mlh 

> 

r 

/ 

^ 

^ 

sowwt 

/ 

ili 

0 

/ 

J 

/ 

^> 

^ 

Z 

7^ 

r^ 

P 

^ 

^ 

/J 

/V 

V 

f 

/ 

/ 

y^ 

A 

,^ 

^ 

,> 

^    J 

l^y 

Jr 

4 

/ 

y 

^ 

.'J 

\^ 

/ 

^ 

^ 

:'^ 

// 

/ 

/ 

^ 

\^ 

r^ 

J/ 

^ 

^ 

- 

— 

- 

w//. 

/' 

/ 

^ 

s**^ 

ji  ^ 

f^ 

^ 

V. 

/^ 

•r- 

.^^> 

V 

Y^ 

^ 

P 

r^ 

,y 

1 

r^ 

^ 

^ 

w^ 

* 

^ 

-^ 

"^ 

_ 

L_ 

0       ^       BO      Y^      iDO      «>o     fm  D       4^      tm     i^rxr     tiioo     »mo     t4W    an 

Head  la  Feet  Head  Hi  Feci 

Aasamed  Hydroelectric  Effldency  V^ 

FiQ.  84. — Curves  Showing  Kilowatt  Hours  for  Diflferent  Acre  Feet  Storage 

on  Various  Heads. 


to  000 

h 

\\ 

[^ 

n 

-^ 

— ' 

tUDO 

\ 

iM 

V 

V<s 

\ 

\ 

\y 

\ 

\ 

^ 

1000 

I 

\ 

\ 

w 

\ 

\ 

Ki 

1 

1 

\ 

\  ^ 

\ 

7fW0 

L 

\  \'^ 

y 

\ 

^ 

\ 

\\ 

\ 

lA 

LN 

\ 

1 

s. 

<aoii 

1 

«  5000 

1 

\ 

\ 

\  ^ 

^ 

V 

V 

\ 

\ 

\ 

1 

\ 

\ 

I  \ 

^ 

^^ 

^ 

s 

V 

V 

\ 

I 

\ 

\£^ 

r 

V 

\ 

\ 

s 

\ 

iOOO 

V 

\^ 

\ 

\ 

\ 

^ 

\ 

\ 

"S 

\ 

\ 

It 

^ 

\ 

\ 

^ 

\ 

N 

V 

\ 

\, 

s. 

\ 

V- 

N 

3000 

vi 

\ 

\ 

\ 

^x 

N 

\ 

\ 

\J 

K 

^ 

-> 

1000 

\ 

s 

s^ 

N. 

s 

"^ 

'^I 

lOOO 

V, 

' 

r^ 

^ 

_ 

***1 

— 

0 

"ii 

- 

IT 

\w 

Ih5 

'  u\ 

V 

:  lm\ 

l«0 

\\ 

lio 

\\\\\^ 

V 

- 

llWv 

A 

l£0 

\v\> 

L 

lAuV^^ 

\j 

^a^ 

m 

- 

A\  v\\^ 

1 

^j 

K 

_ 

- 

HD 

rn^^ 

f^.^ 

S 

h  u%\%> 

sJkS 

0 

s, 

60 

\M\ 

\ 

^iii 

^ 

t^ 

V 

J 

\*\ 

V 

S^^^ 

% 

V 

:^ 

"^ 

40 

\ 

\ 

\ 

^^S^ 

k 

^ 

^ 

■^ 

^. 

s 

o-^ 

^ 

■^ 

-, 

EO 

- 

s 

s 

■— . 

:= 

-^^ 

k^ 

^ 

1 

1 

c 

l€ 

>6 

)tO 

0 

la 

JO 

li 

90 

eoi 

m 

m 

» 

©« 

0       10       w      i»      isa      no      eto 

Head  in  Feet  Head  in  Feet 

Assumed  Hydroelectric  Efflciency  OSK 

Fig.  85. — Curves  Showing  Acre  Feet  Storage  Required  for  Different  Kilowatt 
Hours  on  Various  Heads, 


Digitized  by  CjOOQ IC 


STORAGE  RESERVOIRS 


163 


Intakes.  The  intake  should  be  located  sufficiently  far  back 
from  the  dam  in  order  that  the  water  may  be  drawn  out  when 
at  its  lowest  level.     It  is  also  preferable  to  provide  several  intake 


Iwld^r  t&  Bottom  of  Tq^wcr 


[□(«!« 


Fig.  86. — Concrete  Intake  Tower. 


openings  at  different  elevations,  especially  where  the  depth  of 
water  is  considerable.  The  upper  openings  should  then  be  used 
when  the  water  level  is  highest  and  the  others  in  order,  as  the  water 
is  drawn  out  and  the  level  lowered.     In  this  manner  the  pressure 


Digitized  by 


Google 


164  STORAGE  RESERVOIBS 

and  erosive  effect  is  reduced,  while,  on  the  other  hand,  there  is 
less  danger  of  a  shut-down  in  case  there  were  only  one  gate  open- 
ing at  the  bottom,  which  would  be  Uable  to  be  clogged  up  by  silt 
and  mud. 

Such  intakes  are  often  built  in  the  form  of  towers,  a  typical 
design  being  shown  in  Fig.  86.  There  are  four  square  intake 
openings  placed  from  18  to  21  feet  apart  vertically  and  at  angles 
60^  to  each  other  in  the  plan.  The  openings  are  provided  with 
screens  and  sUding  steel  gates  which  are  controlled  from  the  oper- 
ating floor.  There  is  also  a  secondary  intake  placed  entirely  inside 
the  tower,  consisting  of  a  standpipe  42  inches  in  diameter,  built  up 
in  four  separate  sections.  Each  section  has  a  conical  seat  at  the 
upper  and  lower  ends,  and  is  seated  on  the  one  next  below,  the 
bottom  section  seating  on  a  heavy  cast-iron  elbow  which  connects 
with  the  intake  pipe.  The  water  entering  the  intake  openings 
in  the  tower  wall  must,  therefore,  pass  through  the  top  of  the 
vertical  standpipe  and  in  this  manner  any  silt  or  mud  is  pre- 
vented from  being  carried  along.  As  the  water  level  goes  down, 
sections  of  the  standpipe  are  removed.  This  is  readily  accom- 
plished by  means  of  a  lifting  gear,  the  pipe  sections  being  closely 
guided. 

Seepage  and  Evaporation.  Consideration  must  also  be  given 
to  seepage  and  extreme  care  should  always  be  taken  to  insure 
imperviousness  of  the  reservoir  bottom.  It  may  thus  be  neces- 
sary to  strip  the  top  soil  until  impervious  strata  are  reached,  while 
fissures  may  have  to  be  closed. 

Ev^aporation  must  necessarily  be  taken  into  account  when 
determining  the  reservoir  capacity.  This  loss  can,  however, 
not  be  regulated,  although  a  deeper  and  narrower  reservoir  will 
have  a  less  evaporation  loss  than  a  wider  and  shallower. 


Digitized  by 


Google 


CHAPTER  VII 

POWER-HOUSE  DESIGN 

1.  BUILDING 

General  Design.  The  design  of  power-houses  differs  greatly, 
depending  on  the  conditions  which  are  to  be  met.  It  is  affected, 
to  a  very  great  extent  by  natural  conditions  such  as  the  location 
with  respect  to  the  stream,  the  condition  of  the  soil,  etc.  Low 
and  high-head  developments  require  different  types  of  turbines, 
and  these  may  furthermore  be  of  a  horizontal  or  vertical  con- 


iiiiiiiiiiiiiilll  I 


*     i     t     t     t 


Fig.  87.— Power  House,  Mississippi  River  Power  Company,  Keokuk,  Iowa. 

struction,  necessitating  entirely  different  layouts.  The  number 
and  capacity  of  the  generating  units  is  obviously  a  determining 
factor,  and  the  location  of  the  development  is  generally  such  that 
a  high-tension  transmission  is  necessary  so  that  provision  must  be 
made  for  housing  the  transforming  and  high-tension  switching 
apparatus. 

In  designing  the  building,  the  arrangement  of  the  apparatus 
should  natiu-ally  be  given  first  consideration,  but  this  does  not 
mean  that  the  architectural  features  sould  be  neglected.    It  is  not 

165 


Digitized  by 


Google 


166  POWER-HOUSE  DESIGN 

necessary  that  the  building  should  be  too  ornamental.  Sim- 
plicity in  design  and  harmony  with  the  surroundings  is  very 
desirable  so  as  not  to  injure  the  scenic  conditions,  but,  on  the 
other  hand,  attract  the  attention  of  visitors.  Figs.  87  and  88  are 
good  examples  of  a  pleasing  architecture. 

A  hydro-electric  px)wer-house  building  is  generally  divided 
into  two  longitudinal  bays,  a  front  or  main  bay,  containing  the 
turbines  and  generators,  and  a  rear  bay  containing  the  trans- 


Fig.  88. — Cohoe8  Hydro-Electric  Power  Development,  Cohoes,  N.  Y. 

formers,  switching  apparatus,  etc.  (see  Fig.  89).  The  two  bays 
are  separated  either  by  a  wall  or  by  a  row  of  supporting  colunms, 
and  the  rear  bay  is  divided  into  two  or  more  floors,  and  these  in 
turn  into  various  rooms  or  compartments.  When  the  space  is 
very  limited,  as  on  steep  hill  slopes,  where  the  cost  of  excavation 
becomes  extra  high,  it  is  sometimes  desirable  to  locate  the  switch- 
and  transformer-house  some  distance  back  from  the  generating 
station  and  connect  the  two  by  a  tunnel  through  which  the  cables 
can  be  run. 

Basements.  In  modem  low-head  developments,  where  ver- 
tical turbines  are  used,  the  substructure  not  only  serves  as  foun- 
dation for  the  superstructure  of  the  building,  but  is  really  the 
hydraulic  structure,  in  that  the  intakes,  turbine  casings  and  draft 
tubes  are  molded  directly  in  the  concrete.  In  such  plants  one 
or  more  basements  or  tunnels  are  necessary  for  providing  access  to 
the  turbines,  and  for  housing  the  various  oil-pressure  pumps  for 
the  governors  and  step  bearings. 

Where  the  floors  must  carry  heavy  loads,  or  when  they  are  to 
support  the  generator  frames,  step  bearings,  etc.,  they  must  be 


Digitized  by 


Google 


BUILDING 


167 


Fig.  89. — Sectional  Elevation  of  Power  House,  Mississippi  River  Power  Com- 
pany, Keokuk,  Iowa. 

heavily  reinforced  with  I-beams  and    supported  with  concrete 
piers. 

With  horizontal  turbines  no  basement  is  needed,  although 
tunnels  are  then,  nevertheless,  usually  installed  below  the  main 
floor  for  cables,  oil  and  water  piping,  etc.  Ventilating  ducts  for 
carrying  fresh  air  from  the  outside  to  the  different  generators  are 
also  essential,  especially  in  low-head  plants  with  slow-speed  imits. 
This  subject  is  treated  more  fully  under  "  Ventilation." 


Digitized  by  CjOOQ IC 


168 


POWER-HOUSE  DESIGN 


Foundation.  The  most  important  part  of  the  building  is  the 
foundation,  and  careful  soundings  must  be  made  to  ascertain  the 
underl3ring  strata.  If  bedrock  is  found  within  moderate  depth, 
the  foundation  should  be  carried  down  to  the  same.  For  all  soils 
there  is  a  certain  safe  bearing  load,  and  if  this  is  exceeded  the 
structure  suppx)rted  thereby  is  apt  to  settle.  The  safe  loads 
usually  allowed  in  this  country  are  given  in  the  following  Table 
XXXVI. 

TABLE  XXXVI 

Safe  Bearinq  Power  of  Soils  ^ 


Bbabino  Powbr  in  Tons  pxr 
Squarb  Foot. 

Minimum. 

Maximum. 

Rock,  hardest  kind »: 

200 

25 

15 

4 

4 
2 
1 
8 
4 
2 
0.5 

"Rnolc.  AniiflJ  tn  Aflhlfir  maflOTirv 

30 

Brick,  eaual  to  ashlar  masonrv 

20 

Brick  of  Door  aualitv 

7 

Clav  in  thick  beds  alwavs  drv 

6 

Clav  in  thick  beds  moderately  dry 

4 

Clay,  soft 

2 

Gravel  and  coarse  sand 

10 

Sand,  fine  and  comoact 

6 

Sand,  clean  and  drv.  .  ,    ,  , 

4 

Alluvial  soils  and  uncertain  sand 

1 

1  From  "  Treatise  on  Masonry  ConBtniction,"  by  Baker. 

For  the  machinery  foundations  it  is  considered  good  practice 
to  use  somewhat  lower  values.  About  one-half  is  a  good  working 
basis  for  such  work,  thus  allowing  a  maximum  load  of  about  1000 
pounds  per  square  foot  for  ordinary  alluvial  soils.  Clean,  sharp 
sand  is  considered  to  be  a  good  bearing  soil,  and  may  only  be 
necessary  to  cover  it  with  a  concrete  mat,  which  requires  a  mini- 
mum of  concrete.  For  soft  or  alluvial  soils  piling  is  almost  always 
required.  The  piles  may  be  of  wood,  although  in  the  last  few 
years  much  use  has  been  made  of  concrete  piles,  both  plain  and 
reinforced.    Such  piles  are  less  apt  to  decay  and  their  bearing 

Note.  A  complete  bibliography  on  the  subject  of  "  Bearing  Value  of 
Soils  "  is  contained  in  the  Proceedings  of  the  American  Society  of  Civil  Engi- 
neers for  August,  1917. 


Digitized  by 


Google 


BUILDING  169 

power  is  higher  dtie  to  their  greater  friction.  They  may  also 
be  made  of  larger  diameters  than  can  be  obtained  with  wood  piles, 
and  a  less  number  is  therefore  required  to  support  a  given  load. 

When  designing  foundations  the  first  step  is  to  ascertain  the 
total  weight  that  will  be  sustained  by  the  soil  and  then  to  pro- 
vide a  sufficient  number  of  square  feet  of  area  of  the  base  to  bring 
the  pressure  per  square  foot  within  the  safe  value.  The  weight 
should  include  the  machines,  fittings,  the  weight  of  the  founda- 
tion itself,  and,  in  the  case  of  the  turbines,  the  weight  due  to  the 
water  thrust  unless  this  is  balanced.  Separate  foundation  should 
be  provided  for  the  different  imits  so  as  to  isolate  any  failure  as 
far  as  possible. 

Concrete  is  alwa3rs  used  for  the  foundations.  They  should 
be  solid  for  the  machinery,  while  the  building  may  be  supported 
on  columns  or  arches  so  as  to  economize  on  the  concrete.  Where 
there  is  danger  of  high  water  in  the  tailrace,  the  outside  founda- 
tion walls  should  necessarily  be  made  water-tight  so  as  to  prevent 
water  from  entering  and  flooding  the  basement.  For  such  cases  a 
sump  is,  therefore,  generally  provided  into  which  the  seepage 
may  collect  and  from  where  it  can  readily  be  pumped  out.  A  mix- 
ture of  one  part  cement,  three  parts  sand  and  six  parts  gravel 
or  broken  stone  forms  a  concrete  that  is  extensively  used,  and 
which  has  given  perfect  satisfaction  for  machine  foundations. 

On  small  machines  the  foundation  bolts  and  plates  may  be 
placed  in  position  before  the  concrete  is  put  in.  They  should  be 
hung  in  place  by  a  wooden  template  and  the  bolts  surrounded  by 
stove  pipe,  conveyor  pipe,  or  scrap-iron  pipe,  several  inches  larger 
than  the  bolts  themselves.  This  allows  for  mistakes  in  location 
and  variation  in  the  machine  parts,  the  holes  being  filled  when  the 
base  is  grouted.  With  large  machines,  however,  it  is  better  to 
have  pockets  in  the  concrete  large  enough  for  the  foundation  plates 
to  be  dropped  in.  These  holes  can  be  fiUed  in,  grouting  the 
base,  and  serve  the  further  purpose  of  making  a  good  bond  between 
the  foundation  proper  and  the  grout  in  and  under  the  base. 

Grout  is  preferable  mixed  half  sharp  clean  sand  and  half  cement. 
It  should  be  thin  enough  to  flow  readily  and  should  be  well  pud- 
dled into  place.  Before  pouring,  all  dust  and  trash  should  be 
cleaned  off  and  the  foundation  thoroughly  wet  down.  It  is  better 
to  use  a  fairly  slow-setting  cement  on  large  castings.  In  some 
cases  cement  for  grouting  has  been  set  aside  and  aged  a  year 


Digitized  by 


Google 


170  POWER-HOUSE   DESIGN 

before  using.  Fresh  or  quickHsetting  cement  may  heat  enou^ 
while  setting  to  cause  expansion  and  distortion  of  large  castings.  A 
record  of  the  grout  and  room  temperature  should  be  taken  as  a  check. 

Floors.  No  combustible  material  of  any  kind  should,  if  pos- 
sible, be  used  in  the  construction  of  a  px)wer-house.  As  the  sub- 
structure of  the  building  is  generally  built  of  concrete  it  is  but 
natural  that  the  floors  should  also  be  of  concrete.  A  dark  color  is 
preferable  so  as  to  render  drops  of  oil  inconspicuous.  Tile  or  mosaic 
floors  are  possibly  the  best  floor  finish  for  a  generating  room.  It  is 
smooth,  easy  to  keep  clean  and  has  a  very  handsome  apperance 
if  made  to  conform  with  the  general  interior  finish  of  the  station. 

Walls.  The  walls  may  be  either  of  reinforced  concrete  con- 
struction or  of  brick  with  a  steel  skeleton  framework.  Where 
future  extensions  are  contemplated  a  false  wall  is  provided  on  one 
end  of  the  building.  The  interior  should  be  kept  as  light  as  pos- 
sible, and  it  is  therefore  advisable  to  apply  a  smooth  siurface  of 
cement  plaster  and  whitewash  or  paint  the  same.  For  more 
important  stations  the  walls  may  be  faced  with  pressed  brick  and 
up  from  the  floor  to  about  10  feet  with  enameled  brick.  Where 
the  extra  expense  is  warranted,  the  walls  may  be  entirely  lined 
with  enameled  brick  and  a  wainscoting  of  contrasting  color, 
preferably  olive-green. 

Roof.  The  roof  of  the  building  should  alwajrs  be  supported 
on  the  steel  trusses,  carried  on  the  side  of  the  walls  or  on  the 
steel  columns.  The  slope  should  not  be  excessive,  2  inches  per 
foot  being  suflicient  with  gravel  covering.  This  construction 
requires  less  material,  and  is  advantageous  when  the  transmission 
wires  are  to  enter  the  station  through  roof  entrance  bushings, 
or  where  the  lightning  arrester  horns  are  to  be  installed  on  the  roof. 

The  roof  covering  may  simply  consist  of  boards  covered  with 
roofing  paper,  tar  and  gravel.  Reinforced  concrete  is  some- 
times used  in  place  of  boards  so  as  to  make  an  absolutely  fireproof 
construction.  Roofs  covered  with  red  tile  are  often  used  and 
present  a  very  pleasing  appearance.  Corrugated  iron  roofs  are, 
however,  objectionable  due  to  the  liability  of  moisture  condensing 
on  the  inner  surface  and  dripping  into  the  station.  They  may 
also  cause  the  station  to  be  extremely  hot  in  the  summer  unless  an 
insulating  lining  is  provided  below  the  roof  trusses  to  keep  out  the 
heat.  This,  however,  is  objectionable  and  corrugated  iron  roofs 
are  therefore  seldom  used  for  power-houses.    For  tile  or  metal 


Digitized  by 


Google 


BUILDING  171 

roofs  it  is  necessary  to  provide  steeper  inclines  than  with  gravel 
roofs  so  that  the  water  may  run  off  rapidly.  The  height  of  the 
trusses  should  be  about  one-third  of  the  span.  Monitors  are 
sometimes  provided  so  as  to  give  additional  ventilating  facilities. 

Roof  trusses  with  a  raised  chord,  as  in  Fig.  101,  are  in  many 
instances  of  great  advantage  in  that  they  provide  an  increased 
headroom  without  unnecessarily  raising  the  walls  of  the  building. 
This  is  of  special  importance  in  the  high-tension  part  of  the  sta- 
tion, where  ample  headroom  must  be  provided  for  the  busbars. 

Windows.  A  good  lighting  is  imperative,  and  large  windows 
are  therefore  essential.  They  should  be  symmetrically  located 
with  regard  to  the  generating  units  and  their  design  should  be 
such  as  to  harmonize  with  the  building,  arched  windows  being 
very  generally  used.  Skylights  of  glass  tile  placed  in  the  roof 
will  also  add  considerably  to  the  lighting.  The  window  sashes 
should  preferably  be  metallic  and  the  glass  reinforced  with  wire 
netting  so  as  to  prevent  shattering  when  broken.  Ribbed  or  non- 
transparent  glass  is  also  desirable,  because  it  keeps  out  the  intense 
rays  of  the  sun.  In  order  to  provide  for  ventilation  provision 
should  be  made  so  that  the  windows  can  be  readily  opened,  and 
in  large  stations  they  are  operated  by  electric  motors  controlled 
from  the  main  switchboard.  Precautions  should  also  be  taken 
so  that  rain,  snow  or  dust  will  not  blow  in  on  the  machinery  or 
apparatus.  This  is  especially  important  on  the  switchboard  side 
where  the  wiring  is  exposed  and  it  is,  therefore,  better  practice 
not  to  provide  any  means  for  opening  the  windows  on  that  side. 
For  tropical  climates  all  windows  which  are  liable  to  be  opened 
should  be  equipped  with  mosquito  screens. 

Doors.  The  location  of  the  doors  is  naturally  governed  by 
local  conditions.  One  of  the  openings  should  be  of  a  sufficient 
size  to  admit  a  railroad  car  and  tracks  should  therefore  also  be 
provided.  Very  often  these  doors  are  of  the  rolling  type,  this 
design  being  most  economical  as  regards  space. 

Traveling  Crane.  Provision  should  always  be  made  for  sup- 
porting the  track  for  a  traveling  crane,  which  should  span  the 
generator  room  and  run  the  full  length  of  the  station.  The  track 
is  generally  supported  on  pilasters  in  the  outside  wall  and  on  the 
steel  colmnns  separating  the  generator  and  switch  rooms.  There 
should  be  ample  headroom  allowed  so  that  the  various  machine 
parts  can  be  readily  removed  when  repairs  arc  to  be  made.    This 


Digitized  by 


Google 


172  POWER-HOUSE  DESIGN 

is  especially  important  with  vertical  units  where  the  water-wheel 
rotor  is  mounted  on  the  same  shaft  as  the  generator  field,  and  in 
which  case  it  should  be  possible  to  lift  out  the  whole  revolving 
element  by  simply  removing  the  top  bracket  and  bearing  of  the 
generator. 

The  type  of  crane  depends  largely  on  the  size  of  the  units, 
weight  of  heaviest  pieces  and  the  number  of  units  in  the  station. 
In  small  stations  a  hand-operated  crane  may  be  ample,  while  very 
large  stations  will  require  two  electrically  operated  cranes.  A  few 
stations  have  been  equipped  with  a  gantry  type  of  crane  just 
long  enough  to  straddle  the  generators  and  high  enough  for 
the  highest  lift.  This  type  deserves  more  careful  consideration 
than  it  has  had  heretofore.  The  span  is  shorter  and  consequently 
lighter  than  an  overhead  crane.  The  building  framework  can  be 
designed  simply  for  the  roof  load,  with  a  material  reduction  in  the 
steel  required. 

The  crane  should  be  of  sufficient  capacity  to  lift  the  total 
revolving  element  of  vertical  wheels  and  generators  imless  some 
special  arrangement  of  jacks  under  the  generator  rim,  or  on  the 
shaft,  is  provided.  This  support  is  necessary  to  reUeve  the  thrust 
bearing  for  inspection  or  repairs.  Jacks  or  supporting  blocks 
under  the  generator  field  rim  are  also  of  great  assistance  d^jring 
the  erection  olF  vertical  units. 

The  question  of  armature  repairs  should  be  considered  when 
designing  the  crane  equipment.  A  few  coils  can  be  replaced  in  a  ver- 
tical machine  by  removing  two  or  more  field  poles.  Extensive  re- 
pairs are  best  handled  by  lifting  the  entire  armature  above  the  field 
rim  and  supporting  it  on  substantial  blocking.  A  temporary  floor 
is  laid  on  the  top  of  the  field  spider  for  a  working  platform.  This 
arrangement  does  not  disturb  the  line  up  of  the  revolving  parts 
and  usually  makes  a  very  material  saving  in  time  and  expense. 

Some  special  arrangenlent  is  usually  necessary  to  provide 
power  for  the  electric  cranes.  The  exact  details  depend  largely 
on  local  conditions  and  a  careful  analysis  should  be  made.  In 
some  cases  a  motor  generator  set  may  be  purchased  in  advance 
and  later  used  as  part  of  the  permanent  exciter  equipment.  In 
others,  an  engine  or  turbine-driven  generator  set  may  be  the  best 
solution.  In  any  case,  sufficient  capacity  for  the  heaviest  lifts 
must  be  provided.  An  under-powered  equipment  where  the 
heavy  lifts  have  to  be  jumped  a  few  inches  at  a  time  is  decidedly 


Digitized  by 


Google 


BUILDING  173 

unsatisfactory  as  well  as  dangerous.  If  ample  driving  power  is 
not  available  a  flywheel  will  assist  materially. 

Slings,  lifting  devices,  hooks,  etc.,  should  be  designed  with 
ample  safety  factors  and  to  allow  safe,  accurate  and  rapid  assembly. 

Wire  slings  should  be  oiled  to  prevent  rusting  and  protected 
from  kinking  or  cutting  on  sharp  comers  by  pads  or  their  equiva- 
lent. Angle  pieces  made  from  boiler  plate  are  good,  cheap  and 
durable.  Any  slings  that  show  wear  or  weakening  should,  of 
course,  be  replaced. 

Ventilation.^  Particular  attention  must  be  given  to  the  ven- 
tilating problem  in  the  design  of  the  building;  especially  for  large 
installations  where  the  heat  to  be  carried  away  from  the  generators 
is  very  great.  The  oversight  of  this  important  feature  in  stations, 
otherwise  well  designed,  has  led  to  considerable  trouble  from  over- 
heating the  machines;  for  if  no  provision  is  made  for  admitting 
fresh  air,  the  air  in  the  machine  pit  and  in  the  space  around  the 
machine  is  used  over  and  over  again.  Fresh  cool  air  can  be  taken 
to  the  generator  pit  through  ventilating  ducts  especially  built  for 
this  ptu-pose  below  the  floor,  and  from  the  pit  the  air  is  drawn  up 
through  the  machine  by  the  fanning  action  of  the  rotor  or  forced 
circulation  may  be  provided  by  motor-operated  fans,  the  heated 
air  escaping  through  openings  in  the  roof.  The  size  of  the  inlets 
and  outlets  depends  uix)n  the  losses  to  be  dissipated,  the  allowable 
difference  in  temperature  between  the  inside  and  outside  air  and 
the  height  of  the  building. 

Mr.  R.  C.  Muir  in  the  "  General  Electric  Review  "  gives  the 
following  recommendations:  "  The  maximum  difference  between 
inside  or  room  temperatmre  and  outdoor  temperature  should  not 
exceed  20**  F.  (11.1**  C),  dining  hot  weather,  since  the  air  entering 
the  machine  is  taken  from  the  room  and  the  air  leaving  the  machine 
is  considerably  warmer  than  the  room  temperature.  The  ven- 
tilation scheme  should  be  laid  out  for  most  severe  or  hot  weather 
conditions.  It  is  very  impx)rtant  to  make  the  difference  in  height 
between  inlet  and  outlet  openings  as  great  as  the  station  will  per- 
mit, as  is  shown  from  Table  XXXVII. 

The  amount  of  air  required  for  the  generating  room  can  be 
easily  calculated,  as  follows: 

One  Kw.-hour  will  raise  the  temperatmre  of  10,000  cubic  feet 
of  air  from  80**  F.  to  100**  F.,  a  rise  of  20**  F.  (11.1**  C). 

'  See  also  "Generator  Ventilation." 


Digitized  by 


Google 


i74 


JPOWERrHOUSE  DESIGN 


The  total  loeses  in  generating  room  equals  (total  Kv.A.  gen- 
erator capacity)  —  (total  Kv.A.  generator  capacity  X  generator 
eflBciency). 

The  total  amount  of  air  required  for  the  generator  room  in 
cubic  feet  per  minute  equals 

10,000  X  total  loss  per  hoiu'  in  kw-hr. 
60  • 

The  above  method  will  give  approximately  twice  as  much 
air  as  that  required  with  the  forced  or  positive  ventilation  schemes, 
for  the  reason  that  when  the  ventilating  scheme  is  such  that  a 
definite  amount  of  outside  air  will  pass  through  the  machine,  a 
temperature  diflference  of  30**  F.  to  40^  F.  (16.7**  C.  to  22.7°  C.) 
between  ingoing  and  outgoing  air  is  not  excessive. 

TABLE  XXXVII 

QuANnrr  of  Air  in  Cubic  Feet  Discharged  per  MimrrB  through  a 
Ventilating  Duct  of  1  Square  Foot  in  Cross-sectional  Abea. 
Difference  in  Temperature  op  Air  in  Duct  and  OursroB— 20*  F. 


Height  of  Vent.  Duct  in  Feet. 

Cubic  Feet  per  Minute, 

10 

153 

20 

217 

30 

265 

40 

306 

to 

342 

60 

375 

niumination.  This  is  mostly  done  by  tungsten  lamps,  the 
proper  location  and  spacing,  of  com^e,  being  governed  by  the 
general  layout  and  arrangement  of  the  apparatus.  In  the  gene- 
rator room  500-watt  lamps  are  used  very  generally  and  are 
mounted  on  the  roof  trusses  and  provided  with  intensity  reflectors, 
giving  a  very  uniform  and  satisfactory  illumination.  In  addition 
the  lamps  are  also  mounted  on  brackets  along  the  walls.  For 
other  parts  of  the  station  the  lamps  vary  in  size  from  25  to  500 
watts. 

The  current  for  the  lighting  may  be  taken  from  the  exciter 
system,  if  not  fluctuating  too  widely,  or  by  means  of  step-down 
transformers  from  the  main  biis.    As  a  protective  measure  it  is  a 


Digitized  by 


Google 


ARRANGEMENT  OF  APPARATUS  175 

good  method  to  arrange  about  one-third  of  the  lights,  well  dis- 
tributed in  the  station,  on  a  separate  circuit,  which,  in  case  of 
trouble,  may  be  switched  over  to  the  exciter  battery  or  other 
reserve  sotu-ce.  In  some  stations  this  is  accomplished  auto- 
matically. 

For  illuminating  outdoor  equipments  flood-lighting  has,  of 
late,  been  used  with  very  great  success. 

Heating.  The  heating  of  the  power-house  building  is  ordina- 
rily, to  a  very  great  extent,  done  by  the  heat  radiated  from  the 
machines,  and  provision  is  often  made  whereby  during  cold 
weather  the  ventilating  air  may  be  used  over  and  over  again  until 
it  reaches  a  certain  temperature.  In  many  stations  separate 
provision  must  be  made  for  heating.  In  some  this  is  done  by 
means  of  electrical  heaters,  while  in  others  complete  steam-heating 
systems  are  installed.  In  connection  with  these  a  steam-cleaning 
plant  for  waste,  which  necessarily  is  used  in  considerable  quanti- 
ties in  large  stations,  can  readily  be  provided. 

Miscellaneous.  Provision  should,  of  course,  also  be  made  for 
necessary  repair  shops,  store  rooms,  offices,  toilets,  etc.,  and  pro^ 
tective  measures  for  accidents  and  fire  must  not  be  neglected.  A 
vacuum  compressed-air  system  may  be  required  for  cleaning  or 
other  purposes  and  a  complete  water-supply  system  to  various 
parts  of  the  building  is,  of  course,  also  necessary.  Elevators  and 
ample  stairway  provision  is  essential  so  as  to  permit  a  ready  access 
to  important  paints,  as,  for  example,  between  the  generator  room 
and  the  switchboard  gallery. 

2.  ARRANGEMENT  OF  APPARATUS 

General  Considerations.  The  arrangement  of  the  apparatus 
should  be  very  carefully  considered  from  the  standpoint  of  sim- 
plicity and  reliability  of  operation.  The  purpose  of  the  station 
being  to  give  reUable  service  consideration  must  also  be  given  to 
the  causes  of  disturbances  and  means  for  minimizing  their  effects. 
In  anticipating  these  abnormal  or  so-called  emergency  conditions, 
the  failure  of  every  piece  of  apparatus  must  be  considered  as  a 
possibility,  and  a  definite  plan  worked  out  for  limiting  the  mag- 
nitude and  area  of  such  disturbances. 

Turbines.  With  horizontal  sets  the  turbines  may  be  located, 
together  with  the  generators,  in  the  generator  room  or  in  separate 
wheel  chambers  built  in  the  dam  or  partition  towards  the  fore- 


Digitized  by 


Google 


176  POWER-HOUSE  DESIGN 

bay.  The  latter  practice  is  only  used  for  very  low-head  develop- 
ments, where  one  of  the  power-house  walls  forms  part  of  the 
dam  structure.  With  vertical  units  the  turbines  are  always 
located  in  a  basement,  the  thrust  bearing  being  suppx)rted  on  an 
intermediate  floor  below  the  main  floor,  imless  suspension  bear- 
ings are  used,  these  being  mounted  on  top  of  the  upper  generator 
bearing  bracket. 

Govemors.  The  governors  should  be  located  on  the  generator 
room  floor  close  to  the  units  which  they  are  to  control,  and  con- 
nected to  the  operating  cylinders  on  the  turbines  directly 
below.  The  governor  oil  pumps  with  their  pressure  and  storage 
tanks  should  also  be  installed  in  the  basement,  and  similarly 
the  oiling  system  for  the  turbo-generator  units. 

Generators.  The  turbo-generator  units  are  located  on  the 
main  floor  and  are  almost  always  arranged  in  a  line  along  the  long 
axis  of  the  station  (Fig.  90).  They  should  be  spaced  far  enough 
apart  so  that  ample  space  for  passage  is  provided  between  them. 
Horizontal  sets  may  be  installed  either  at  right  angles  (Fig.  91) 
or  parallel  (Fig.  92)  to  the  long  axis,  the  latter  method  being 
necessary  for  high  heads  where  impulse  wheels  are  used.  The 
arrangement  of  the  rest  of  the  equipment,  such  as  the  trans- 
formers, may  also  be  a  determining  factor  in  regard  to  which 
direction  the  sets  should  be  installed.  If  one  transformer  bank, 
consisting  of  single-phase  units,  is  to  be  installed  for  each  gen- 
erator, the  space  occupied  by  them  may  be  of  such  a  length  that 
it  would  be  more  economical  to  install  the  turbo-generator  sets 
parallel  to  the  long  axis,  thus  reducing  the  width  of  the  building. 
Exciters.  The  exciters  are,  as  a  rule,  installed  on  the  same 
floor  as  the  main  generators  and  in  the  center  of  the  station. 
The  advantage  of  such  an  arrangement  is  that  the  exciters  will  be 
located  close  to  the  operating  switchboard,  and  the  amount  of 
copper  required  for  the  exciter  leads  is  thus  a  minimum.  The 
system  may  readily  be  sectionaUzed,  one  exciter  serving  the 
generators  located  in  one-half  of  the  station,  and  the  other  the 
generators  on  the  opposite  side.  This  does  not,  of  course,  refer 
to  direct-connected  exciters  or  to  individual  motor-driven  exciters, 
which  are  located  near  their  respective  generators. 

Transfonners.  Due  to  their  weight,  the  step-up  transformers 
should  preferably  be  located  on  the  main  floor.  They  are  gen- 
erally installed  in  isolated  compartments  in  the  rear  bay,  sep- 


Digitized  by 


Google 


ARRANGEMENT  OF  APPARATUS 


177 


Fia.  90. — Interior  of  Generating  Station,  Cedar  Rapids  Mfg.  and  Power 
Company.  Present  Equipment,  Ten  10,000  Kv.A.  Generators.  Ultimate 
Eighteen  Similar  Units. 

arated  from  the  generating  room  by  fireproof  steel  curtains. 
These  compartments  should  be  suf5ciently  large  to  allow  a  good 
ventilation.  A  car  track  is  provided  on  the  generator  room  floor 
in  front  of  the  transformer  compartments  whose  floors  are  raised 
so  that  the  transformers  can  be  run  out  on  the  car  and  moved  to 
some  convenient  place  in  the  station  where  repairs  can  be  readily 
made.     For  large  units  it  may  be  necessary  to  provide  a  hole  in 


Digitized  by 


Google 


178 


POWER-HOUSE  DESIGN 


Fig.  91. — Interior  View  of  Generating  Station,  Connecticut  Power  Company, 
Falls  River  Development. 


the  floor  above  the  repair  room  so  as  to  enable  the  transformer 
core  to  be  lifted  out  of  the  tank,  or  a  pit  may  be  provided  into 
which  the  transformer  may  be  lowered  so  that  suflieient  head- 
room is  obtained  for  lifting  out  the  core.  Sometimes  the  repair 
room  is  so  situated  that  the  main  crane  cannot  be  utilized  for 
dismantling  the  units.  In  such  a  case  a  chainfall  supix)rted  from 
a  heavy  I-beam  in  the  floor  above  may  be  provided.  This,  how- 
ever, as  a  rule,  only  refers  to  smaller  plants. 

The  oil  tanks  should  be  located  in  the  basement,  and  par- 
ticular care  should  be  taken  to  avoid  any  fire  risk.  For  this 
reason  it  is  advisable  to  install  the  tanks  in  separate  enclosed 
compartments  and  in  certain  cases  these  have  been  filled  with 
sand.  Their  location  should  also  be  such  that  in  case  of  fire  the 
oil  can  readily  be  drained  in  the  tailrace. 

Current  Limiting  Reactors.  As  these  are  inserted  either 
between  the  low-tension  bus  sections  or  in  the  low-tension  trans- 
former leads,  their  location  is  in  the  low-tension  switchroom,  close 


Digitized  by 


Google 


ARRANGEMENT  OF  APPARATUS 


179 


Fig.  92. — Interior  View,  Big  Creek  Development,  Pacific  Light  and  Power 
Company.     17,500  Kv.A.  Generators. 

to  the  apparatus  which  they  are  to  protect.  It  is  advisable  to 
enclose  them  in  compartments,  like  the  transformers,  and  pro- 
vision should  be  made  so  that  they  can  be  securely  anchored. 
They  should  be  installed  at  a  distance  of  approximately  half  their 
diameter  from  any  iron  or  steel  structure  so  as  to  prevent  any 
heating  of  this  and  consequently  increased  losses. 

Switchboards.  The  different  pieces  of  apparatus  comprising 
the  switching  equipment  are  distributed  on  the  various  floors 
in  the  switch  section  of  the  station,  each  story  being  partitioned 
to  suit  the  various  purposes.  The  operating  room  with  the  con- 
trol switchboard  is  generally  located  on  the  second  floor  and  in 
such  a  position  that  the  operator  may  have  an  unobstructed  view 
of  the  station  and  be  able  to  readily  communicate  with  the  tur- 
bine operators.  A  balcony,  somewhat  overhanging  the  generator 
room  in  front  of  the  switchboard,  is  often  provided,  or  the  operat- 
ing room  is  built  with  a  curved  front  wall  extending  out  over 
the  generator  room. 


Digitized  by 


Google 


180 


POWER-HOUSE  DESIGN 


1  rmBtoilMlon  LioM 
No.  00  Copper  Wire  ^ 
Horn  Osptv 


Fia.  93. — Power-house  Arrangement.    Alabama  Traction,  Light  and  Power 
Company.    Lock  No.  12  Development. 


Digitized  by  VjOOQ IC 


ARRANGEMENT  OF  APr.^^^ATUS 


181 


Digitized  by  CjOOQ IC 


182 


lO^ER-HOUSE  DESIGN 


Fig.  94a. — Floor  Plan  of  Big  Creek  Power-house.    Pacific  Light  and  Power 
Company.    (For  crossHsection  see  Fig.  94.) 


Fig.  95. — Typical  Hydro-Electric  Power-house  Arrangement.    Cross-Section. 
(For  floor  plan  see  Fig.  95a.) 


Digitized  by 


Google 


ARRANGEMENT  OF  APPARATUS 


183 


Digitized  by 


Google 


184 


POWER-HOUSE  DESIGN 


Fig.  96. — Sectional  View  of  Hydro-Electric  Power-house  Arrangement  with 
Limited  Space.     (For  floor  plans  see  Fig.  96a.) 


Digitized  by  LjOOQ IC 


ARRANGEMENT  OF  APPARATUS 


185 


4 


^ 


iD- 


1: 


o 
o 


o 

z 

3 


Digitized  by  VjOOQIC 


186 


POWER-HOUSE  DESIGN 


Digitized  by  VjOOQ IC 


ARRANGEMENT  OF  APPARATUS 


187 


Digitized  by  CjOOQ IC 


188 


POWER-HOUSE  DESIGN 


Fig.  98. — Sectional  View  of  Power-house  Arrangement.    Mexican  Northern 
Power  Company.     (For  floor  plan  see  Fig.  98a.) 


Fig.  99. — Sectional  View  of  Power-house  Arrangement.  ^  Montana    Power 
Company,  Rainbow  Falls  Development.     (For  floor  plan  see  Fig.  99a.) 


Digitized  by  CjOOQ IC 


ARRANGEMENT  OF  APPARATUS 


189 


IS 


o 


o 

a 

B 
o 
O 

I 


1 


IS 


o 

I 


J 


Digitized  by  CjOOQ IC 


190 


POWER-HOUSE  DESIGN 


Digitized  by  CjOOQ IC 


ARRANGEMENT  OF  APPARATUS 


191 


Fig.  101. — Power-house  Arrangement,  Georgia  Railway  and  Power  Com- 
pany.    Tallulah  Falls,  Georgia. 


Digitized  by  CjOOQ IC 


192  POWER-HOUSE  DESIGN 

The  switchboard  containing  the  switches,  etc.,  for  the  exciters 
and  other  station  auxiliaries,  should  be  located  on  the  main  floor 
at  some  convenient  point,  usually  below  the  control-board  gallery. 

Oil  Circuit  Breakers.  The  low-tension  oil  circuit  breakers  are 
generally  of  the  enclosed  type  and,  together  with  the  low-tension 
busbars,  are  located  in  compartments  on  the  main  floor  back  of 
the  transformer  compartments.  The  switches  themselves  should 
preferably  be  set  in  parallel  rows  and  opposite  the  generator  and 
transformer  bank  which  they  control,  so  as  to  call  for  as  short  a 
connection  as  possible  and  in  order  that  these  connections  may  be 
of  equal  length.  The  high-tension  oil  switches  and  busbars,  and 
also  as  a  rule  the  lightning  arrester  tanks,  are  installed  on  the 
floor  above. 

Lightning  Arresters.  The  aluminum  arrester  is  now  generally 
used  in  all  high-voltage  stations.  Both  the  arrester  tanks  and  the 
associated  horn  gaps  may  be  located  within  the  building,  or  the 
horn  gaps  may  be  placed  outside  and  the  tanks  inside,  or  both  may 
be  placed  outside,  provided  there  is  no  danger  of  the  electrolyte 
freezing.  Standard  equipments  of  27,000  volts  and  below  are 
usually  designed  as  complete  units  to  be  installed  inside  the 
station,  whereas  for  those  above  27,000  volts  the  horn  gaps  should 
preferably  be  installed  outside  the  station,  although  the  tanks 
may  be  inside.  There  is,  however,  a  growing  tendency  to  in- 
stall the  entire  lightning  arrester  equipment  outdoors  for  these 
higher  voltages. 

The  arrester  should  naturally  be  placed  close  to  the  line  en- 
trances, and  the  location  should  also  be  such  that  the  path  for  the 
discharge  from  the  line  conductors  to  the  arresters  and  ground 
will  be  as  straight  as  possible.  When  installed  out  of  doors,  it 
may  be  placed  on  the  roof  of  the  building  or  on  a  separate  struc- 
t\u«  at  the  side  of  the  building. 

A  number  of  modern  station  layouts  illustrating  some  of  the 
nimierous  manners  in  which  the  apparatus  may  be  arranged  are 
shown  in  Figs.  93  and  101. 

Outdoor  Apparatus.  With  the  introduction  and  successful 
operation  of  the  outdoor  sub-station,  this  method  of  installing  at 
least  part  of  the  generating  station  apparatus  outdoors  should  be 
given  careful  consideration.  A  large  installation  of  this  kind  is 
that  of  the  Utah  Light  and  Power  Company,  where  only  the 
generating  and  exciter  units  and  the  low-tension  switching  equip- 


Digitized  by 


Google 


TRANSPORTATION  AND  ERECTION 


193 


ment  is  located  indoors,  while  the   transformers,  high-voltage 
switches  and  lightning  arresters  are  located  outdoors. 

A  still  more  revolutionary  power-house  design  has  been  sug- 
gested by  Mr.  R.  J.  McClelland.^    As  seen  from  Fig.  102,  the 


Fig.  102.— Perspective  of  50,000  Kv.A.  Outdoor-Type  Generating  Station. 

plan  proposes  to  put  the  generators  and  transformers,  as  well  as 
the  high-tension  equipment  outdoors,  while  the  exciters  and  the 
more  delicate  control  equipment  are  put  under  cover,  where  pro- 
vision is  also  made  for  the  repair  shop. 

8.  TRANSPORTATION  AND  ERECTION 

Transportation.  The  transportation  of  such  large  machines 
as  are  generally  involved  in  hydro-electric  power  stations  requires 
a  careful  consideration  of  the  limitations  imposed  by  the  rail- 
roads or  carriers.  It  is  furthermore  evident  that  these  points 
must  be  considered  at  the  time  when  the  machines  are  selected 
or  designed. 

The  shipping  limitations  are  the  clearances  (height  and  width) 
and  the  weight.    The  former  are  governed  by  tunnels,  bridges, 

.}  Electr.  World,  Sept.  25, 1916. 


Digitized  by 


Google 


194  POWER-HOUSE  DESIGN 

platforms,  etc.,  and  the  latter  by  the  carrying  capacity  of  the 
bridges  as  well  as  the  cars.  Both  vary  for  different  roads  and 
even  divisions  or  sections  of  the  same  road,  and  in  many  instances 
considerable  advantages  may  be  gained  by  detoming.  For 
example,  it  may  be  found  that  the  extra  expense  of  dividing  cer- 
tain parts  of  a  machine  in  sections  may  be  so  high  that  a  consider- 
able saving  may  be  made  by  detouring  the  shipment  over  a 
route  whose  limitations  are  such  that  the  parts  can  be  built  and 
shipped  as  one  piece,  even  if  the  extra  distance  were  quite  great. 

Special  cars  may  occasionally  be  obtained  which  will  facilitate 
the  shipments  of  large  capacity.  These  may  be  provided  with 
pits  in  which  part  of  the  machines  may  be  recessed,  thus  decreasing 
the  over-all  height,  or  they  may  be  of  extra  large  carrying  capacity. 

Unloading.  The  question  of  imloading  and  transporting 
material  and  machine  parts  from  the  nearest  point  on  the  railroad 
should  have  careful  consideration  early  in  the  design  work.  The 
dimensions  of  the  largest  and  the  weights  of  the  heaviest  pieces 
should  be  obtained  from  all  companies  interested.  Also,  one 
should  know  how  these  pieces  will  be  boxed  and  shipped. 

It  is  always  preferable  to  deliver  the  machinery  in  the  cars 
under  the  station  crane.  Unfortunately  this  is  often  impossible 
or  impracticable  on  account  of  the  expense  involved.  Local  con- 
ditions, however,  usually  determine  the  best  arrangement  for  each 
installation.  In  each  case  careful  consideration  should  be  given 
to  the  job  as  a  whole,  and  to  all  the  material  which  must  come  in. 
Steel  cranes,  water  wheels,  generators,  transformers,  switch- 
board equipment,  cable,  piping,  etc.,  must  all  be  handled. 

A  carefully  designed  erection  equipment,  with  the  job  as  a 
whole  in  mind,  will  effect  material  savings,  as  various  contractors 
will  either  pay  for  the  use  of  this  equipment  or  make  correspond- 
ing reductions  in  the  total  price. 

In  difficult  country,  or  far  from  the  railroads,  it  may  be  nec- 
essary to  arrange  with  the  various  manufactm^rs  for  shipment 
partially,  or  totally,  knocked  down.  The  increased  price  should, 
in  such  a  case,  be  balanced  against  the  transportation  costs. 

Car  ferries,  inclined  railways  with  car  (Fig.  103),  skidwa3rs  or 
heavy  trucking  equipment,  whatever  is  decided  on,  had  best 
remain  under  the  direct  supervision  of  the  resident  engineer  or 
general  superintendent,  who  can  determine  the  best  schedule  for 
handling  all  of  the  material. 


Digitized  by 


Google 


TRANSPORTATION  AND  ERECTION 


195 


Fig.  103. — Inclined  Railway  with  Special  Car. 

Apparatus  Storage.  In  many  cases  material  must  be  delivered 
at  certain  times  before  it  is  needed,  i.e.,  during  the  summer  navi- 
gation, before  the  rainy  season,  while  the  ground  is  frozen,  etc. 
The  question  of  storage,  therefore,  needs  careful  consideration, 
as  there  usually  is  insuflScient  room  in  the  power-house,  espe- 
cially before  the  building  is  completed. 

The  castings  and  rough  machine  parts  may  be  stored  in  the 
open.  A  derrick  for  unloading  and  reloading  will  answer  on  small 
jobs.  On  large  installations  it  may  prove  advantageous  to  install 
one  of  the  main  cranes  or  a  forebay  crane  on  a  temporary  track 
supported  on  timber  framework  over  a  skidway.  Finished  parts 
must  be  protected  from  the  weather,  fittings  and  small  parts  from 
sneak  thieves. 


Digitized  by 


Google 


196  POWER-HOUSE  DESIGN 

Electrical  apparatus  must  be  stored  in  a  dry  place  and  kept 
above  the  freezing-point.  The  best  arrangement  is  an  electric 
heater  which  is  large  enough  to  keep  the  storage  building  above 
freezing  and  arranged  so  that  the  temperature  will  alwajrs  be 
higher  than  that  outside.  Great  care  should  be  taken  to  prevent 
fires.  In  the  majority  of  cases  a  responsible  watchman  on  duty 
at  all  times  is  the  best  insurance  against  fire  and  thieves. 

Schedule  of  Erection.  A  careful  schedule  of  the  erection 
work  should  be  made  to  insure  rapid,  efficient  work  and  prevent 
congestion.  At  least  part  of  the  building  steel  and  the  crane 
should  be  erected  before  any  of  the  heavy  machine  parts  are 
delivered.  The  delivery  of  water  wheel  parts  should  be  arranged 
for  in  the  order  required  and  with  sufficient  time  allowance  to 
permit  of  the  assembly  work  keeping  step  with  the  wheel  pit 
construction. 

On  large  installations  space  and  equipment  must  be  provided 
for  the  necessary  assembly  of  the  machine  parts  before  they  are 
pl0,ced  in  their  final  position.  Any  convenient  open  space  under 
the  crane,  and  centrally  located  in  regard  to  the  final  location  will 
do  for  the  wheel  parts. 

The  generators  must  be  protected  from  the  weather  and  from 
the  dirt,  smoke  and  cement  dust  usually  present  during  the  build- 
ing construction.  In  the  case  of  large  generators  it  is  often 
necessary  to  assemble  the  punchings  and  wind  the  armatures  on 
the  groimd.  This  is  best  handled  in  a  temporary  house,  under  a 
crane.  The  roof  can  be  made  in  sections,  with  eyebolts,  to  per- 
mit easy  removal  with  the  crane  and  the  handling  of  the  armature 
sections.  This  temporary  building  will  protect  the  machines  from 
dirt,  moisture  and  mechanical  injury.  The  winders  will  also  do 
more  and  better  work  when  protected  from  the  noise,  confusion 
and  dirt  of  the  power-house  imder  construction. 

In  most  cases  the  coils  must  be  warmed  before  using.  Where 
this  is  necessary,  convenient  heating  ovens  should  be  made  part 
of  this  temporary  house.  These  ovens  should  contain  wooden 
racks  for  holding  the  coils,  and  steam  coils  or  electric  heaters 
under  the  racks  for  supplying  the  heat.  In  general  the  ovens 
should  range  from  150°  F.  to  200°  F.  and  should  be  large  enough 
to  permit  of  coils  being  heated  several  hours.  A  little  care  in 
arranging  the  ovens  for  the  ready  placing  of  cold  and  removing  of 
hot  coils  will  affect  materially  the  speed  and  costs  of  the  winding 


Digitized  by 


Google 


TRANSPORTATION  AND  ERECTION  197 

work.  Some  very  large  coils  must  be  heated  internally  with 
current.  Direct  current  is  best  for  this  piu-pose.  Usually  one  of 
the  exciter  sets  will  be  of  the  proper  capacity;  failing  this  it  may 
be  necessary  to  secure  an  electrolytic  generator  of  the  proper 
capacity  and  drive  this  by  motor  or  engine.  Current  is  also 
needed  for  heating  the  coils  in  the  split  where  the  armatures  are 
shipped  in  sections. 

Crane  Service.  This  is  usually  the  cause  of  considerable 
friction  between  the  various  erectors  and  oftentimes  one  man  will 
tie  up  the  crane  unnecessarily  simply  to  prevent  some  other  gang 
from  using  it,  although  this  action  is  delaying  the  job  as  a  whole. 

The  general  superintendent,  or  resident  engineer,  should 
allot  the  crane  without  fear  of  favor,  considering  the  progress  of 
the  work  as  a  whole,  or  else  allot  it  to  the  various  gangs  for  stated 
periods.  In  some  cases  the  scheme  of  allowing  the  wheel  erector 
the  crane  mornings  and  the  generator  erector  afternoons  has 
worked  well.  Both  men  can  then  plan  their  work  ahead  and  avoid 
delays. 

Protective  Features.  All  electrical  apparatus  and  finely 
finished  parts  of  all  machines  must  be  protected  from  injiuy  by 
water,  dirt  and  falling  material  during  the  erection  and  imtil  the 
power-house  is  roofed  and  glazed.  In  most  cases  a  liberal  supply 
of  tarpaulins  will  answer,  although  some  cases  warrant  a  tem- 
porary shelter  of  lumber  and  roofing  paper. 

Some  fire-fighting  equipment  should  be  installed  before  start- 
ing the  erection.  Trash,  excelsior,  packing  cases  and  skidding 
should  be  cleaned  out  promptly,  as  the  fire  danger  is  great  under 
the  best  conditions.  Competent  watchmen  should  be  in  charge 
whenever  the  erectors  are  not  working,  to  guard  against  fire, 
thieves  and  malicious  mischief.  This  last  is  by  no  means  a  negli- 
gible item,  as  every  large  installation  sooner  or  later  shows  dam- 
age, or  attempted  damage,  of  this  character. 

It  is  unsafe  and  almost  foolhardy  to  start  any  machine  while 
the  general  construction  is  going  on  without  a  thorough  inspection 
just  before  turning  over.  There  are  numberless  cases  where  these 
inspections  have  brought  to  light  bolts,  tools,  rocks  and  miscella- 
neous metal  that  had  no  excuse  for  being  anywhere  near  the 
machine.  These  pieces  are  always  in  the  air  gap  or  at  some 
adjacent  point  where  the  motion  of  the  magnetic  field  will  draw 
them  into  the  air  gap. 


Digitized  by 


Google 


198  POWER-HOUSE  DESIGN 

Cooperation.  A  conference  of  all  interested  parties  should  be 
arranged  before  starting  the  erection,  and  the  various  steps  of 
the  erection  discussed  and  settled.  This  is  especially  important 
in  the  wheel  and  generator  erection  as  the  successful  operation 
depends  almost  entirely  on  the  careful  line-up  of  these  units. 
Arrangements  should  be  made  at  this  meeting  for  checking  up  the 
line-up  of  the  various  parts,  as  this  is  nearly  always  a  loophole 
for  future  discussion. 

In  case  of  trouble  there  is  always  the  tendency  to  place  the 
blame  on  the  other  fellow's  work.  This  can  be  absolutely  avoided 
by  having  ai]  work  checked  by  the  wheel  erector,  the  generator 
erector  and  the  resident  engineer  or  his  authorized  representative 
and  all  three  signing  a  statement,  in  triplicate,  each  party  keeping 
his  copy.    This  should  read  something  as  follows:    "  We  agree 

that  unit   No. is  on  the  longitudinal  center  line  within 

mils.    The  cross  center  line  within  mils.    On   the   proper 

elevation  within mils,  and  is  level  within mils." 

Where  a  two-piece  shaft  is  used  insert  a  clause,  "  The  water 

wheel  coupling  is  true  within mils,  the  rim  is  true  within 

mils.     The  generator  coupling  face  is  true  within mils,  the 

rim  within mils."  Where  it  is  impossible  to  test  the  coup- 
lings on  the  ground  this  test  can  be  made  at  the  factories  and  a 
statement  furnished.  These  statements  should  be  called  for  in 
placing  the  orders  for  the  apparatus. 

4.   STARTING  UP 

General  Precautions.  Before  starting  the  machines  for  the 
first  time  they  should  be  carefully  inspx3cted  and  guarded  to  pre- 
vent damage  from  tools  or  other  foreign  material  being  carelessly 
or  naaliciously  left  where  they  will  cause  damage.  The  machines 
should  be  blown  out  with  compressed  air  to  remove  dust  and  dirt. 
The  bearings  should  be  flushed  with  kerosene  or  oil,  and  when 
self-lubricated,  filled  with  clean  oil  of-  the  grade  reconmiended  by 
the  machine  manufacturers. 

If  the  station  is  equipped  with  a  central  oiling  system,  all  the 
piping  should  be  flushed  with  oil  and  the  oil  carefully  filtered 
before  it  is  fed  to  the  bearings.  A  temporary  by-pass  from  the 
feed  pipe  to  the  returns  at  the  generator  will  be  of  great  assistance 
in  cleaning  and  testing  the  oiling  system.  All  piping  should  be 
examined  and  tested  for  leaks. 


Digitized  by 


Google 


STARTING  UP  199 

The  electrical  connections  should  be  carefully  inspected  by 
men  of  known  responsibility.  Loose  bolted  contacts,  oil  switches 
with  no  oil,  or  insufficient  oil  in  the  pots,  dinner  pails  stored  on 
top  of  the  oil  pots,  or  in  the  bus  compartment  are  common  sources 
of  troubles. 

After  the  machines  are  ready  for  operation  the  switchboard 
instruments  must  be  looked  over  and  any  necessary  changes  made 
in  the  wiring.  The  synchronizing  devices  must  be  checked  very 
carefully.  The  majority  of  them  are  single  phase  and  it  often 
happ>ens  that  mistakes  in  connections  cause  incorrect  indication 
on  the  meter.  DifiFerent  phases  on  the  two  machines  may  be  con- 
nected to  the  synchronism  indicator  or  the  phase  rotation  of  the 
two  machines  may  be  different.  The  phase  rotation  must  be 
checked  either  by  potential  transformers  and  lamps  connected 
across  a  machine  switch  on  all  three  phases  at  once,  or  a  small 
induction  motor  may  be  run  in  turn  on  all  the  generators.  When 
a  motor  is  available  to  check  the  phase  rotation,  the  synchronism 
indicator  can  be  checked  single  phase  with  a  potential  transformer 
and  lamps. 

Drying  Out  Exciters  and  generators  will  need  more  or  less 
drying  out,  depending  on  the  amount  of  moisture  they  have 
absorbed.  It  is  assumed  that  they  have  been  protected  from  rain 
and  leaking  water  from  concrete  forms.  The  only  other  way 
moisture  can  get  into  the  machines  is  by  sweating  or  condensa- 
tion, due  to  the  machines  being  colder  than  the  surrounding  air. 
This  condition  can  be  largely,  if  not  altogether,  avoided,  by  keeping 
the  p)Ower-house  at  an  even  temperature.  Where  heating  the 
whole  building  is  impossible,  and  the  humidity  is  high,  the  machines 
may  be  enclosed  in  a  temporary  shelter  with  steam  or  electric 
radiators.  In  winter  weather  the  machines  should  be  kept  above 
the  freezing-point.  In  most  cases  it  is,  however,  impossible  to 
prevent  some  condensation  arid  some  drying  is  usually  necessary. 

The  exciters  should,  of  course,  have  the  first  attention.  If 
possible  they  should  be  started  up  and  run  for  several  dajrs  with- 
out'field.  The  windage  will  then  assist  materially  in  drying  and  a 
plumber's  torch  or  stove  can  be  placed  under  the  commutator. 
Care  should  be  used,  however,  to  prevent  overheating.  The 
temperature  should  not  get  higher  than  60°  C. 

Where  it  is  impossible  to  operate  the  exciter  for  any  length 
of  time  before  starting,  the  preliminary  drjdng  can  be  accom- 


Digitized  by 


Google 


200  POWER-HOUSE  DESIGN 

pliflhed  by  hot  air.  This  air  can  be  forced  through  the  exciter  by 
a  blower,  or  boxing  and  barriers  arranged  to  cause  the  hot  air  to 
circulate  through  the  machine  by  natural  draft.  This  hot  air 
may  be  obtained  from  a  steam  radiator,  a  hot-air  furnace,  electric 
heater  or  a  bank  of  incandescent  lamps.  In  any  case  the  air 
should  not  be  higher  than  80^  C.  When  a  blower  is  used  a  cheese- 
cloth screen  over  wire  mesh  is  advisable  on  the  blower  intake  to 
reduce  the  amount  of  dust  blown  into  the  windings.  The  exciter 
may  then  be  brought  slowly  up  to  voltage  as  soon  as  the  insulation 
to  ground  tests  satisfactorily. 

The  large  A.C.  generators  should  be  brought  slowly  up  to 
speed,  and  a  short-circuit  heat  run  put  on  as  soon  as  the  bearings 
are  in  satisfactory  condition.  This  short-circuit  current  should 
be  the  full  load  current  only  on  maximum  rated  machines.  Ma- 
chines with  an  overload  guarantee  may  be  run  at  the  overload 
current.  The  short-circuit  run  should  be  continued  until  the 
proper  insulation  resistance  is  reached.  It  is  advisable,  however, 
to  run  twenty-four  hours  on  short  circuit,  even  when  the  insula- 
tion tests  satisfactorily. 

After  the  short-circuit  run,  the  machine  should  be  brought  up 
to  normal  voltage  and  run  for  several  hoiurs  before  going  on  the 
load.  This  is  to  heat  the  iron  thoroughly.  On  very  large  gen- 
erators it  is  advisable  to  continue  the  drying  for  twenty-four  hours 
as  follows:  Two  hours  10  per  cent  over  voltage,  then  two  hours 
full  load  current  on  short  circuit,  the  drying  continuing  by  alter- 
nating open  and  short  circuit  every  two  hours. 

For  transformer  drjdng  see  section  on  "  Transformers." 

Insulation  Resistance.  Insulation  resistance  may  be  obtained 
by  the  following  method  when  a  megger  or  bridge  is  not  available: 

Connect  one  side  of  a  direct-current  source  of  power  to  the 
windings  to  be  tested;  connect  the  other  side  of  the  direct-current 
circuit  to  a  portable  voltmeter  and  then  read  the  voltage  when  the 
free  side  of  the  meter  is  connected  to  the  other  side  of  the  circuit 
where  it  is  attached  to  the  windings.  Call  this  reading  V.  Then 
connect  to  the  frame  of  the  machine,  being  careful  to  get  a  good 
contact;  call  this  reading  Fi.    Then 

where  i2=the  cold  resistance  of  the  insulation,  and  Ri^ihe  resist- 


Digitized  by 


Google 


STARTING  UP  201 

anoe  of  the  voltmeter  itself,  this  value  usually  being  given  inside 
the  cover  of  the  instrument. 

Before  using  power  from  a  commercial  circuit  for  testing  insu- 
lation, testd  should  be  made  to  determine  if  the  supply  circuit  is 
grounded.  One  side  of  the  circuit  must  be  free  from  grounds  and 
the  ungrounded  side  should  be  used  in  series  with  a  voltmeter  in 
taking  resistance  readings. 

It  is  impossible  to  give  any  hard-and-fast  rules  regarding  the 
minirmiTn  value  of  the  insulation  resistance  that  will  cover  all 
classes  and  sizes  of  machines,  and  the  results  must  be  used  with 
judgment  and  common  sense.  The  insulation  resistance  of  a 
machine  indicates,  as  a  fact,  little  more  than  the  condition  of  the 
insulation  as  regards  the  moisture;  and  the  rate  of  change  of  the 
resistance  as  the  machine  is  being  dried  is,  perhaps,  the  best  indi- 
cation as  to  when  the  drying  has  been  carried  far  enough. 

The  following  approximate  rules  have  been  developed  to  give 
what  may  be  termed  a  fair  value  of  what  the  insulation  resistance 
should  be.  It  must  be  understood,  however,  that  they  are  to 
be  used  merely  as  a  guide: 

For  A.C.  generators 

p    3,000,000  X  Rated  volts 
^"  Rated  Kw. 

For  exciters  and  D.C.  generators 

D    300,OOOXRated  volts  ,--^,,^ 
Rated  current 

The  above  formulae  give  the  insulation  resistance  in  ohms, 
but  as  a  rule  it  is  given  in  megohms,  which  is  equal  to  the  ohms 
as  obtained  above  divided  by  one  million. 


Digitized  by 


Google 


CHAPTER  VIII 

HYDRAULIC  EQUIPMENT 

1.  TURBINES 

Modern  turbines  inay  be  divided  into  two  classes:  Pressure, 
reaction  or  Francis  turbines.  Pressureless,  impulse  or  Pelton 
turbines. 

Reaction  Turbines.  This  type  is  a  combined  potential  and 
kinetic  energy  wheel,  or  more  properly  speaking  a  turbine,  since  it 
admits  water  all  around  the  ^riphery  of  the  runner  and  all  parts 
of  the  same  perform  useful  work.  The  water  enters  the  runner 
at  a  speed  which  is  lower  than- the  spouting  velocity,  and  a  pres- 
sure head  is  left  to  be  used  for  the  acceleration  of  the  flow  of  water 
through  the  nmner. 

The  water  may  pass  either  radially  inward  or  outward  or  it 
may  enter  the  runner  radially  toward  the  shaft  but  leave  in  an 
axial  direction,  i.e.,  in  a  direction  parallel  with  the  shaft.  In 
this  case  the  turbine  is  of  the  mixed-flow  type,  this  being  most 
extensively  used  in  this  country. 

The  runner  rotates  partly  from  velocity  action  and  partly 
from  reaction  due  to  pressmre  and  consequent  acceleration  in 
buckets.  As  the  draft  tube  is  closed,  the  nmner  is  full  of  water 
and  practically  the  total  difference  in  head  between  head-water 
and  tail-water  is  useful. 

The  speed  of  a  reaction  turbine  can  be  varied  not  only  by  vari- 
ation of  the  runner  diameter  but  also,  and  very  effectively,  by 
varying  the  bucket  angle  and  the  angle  between  the  entrance 
speed  and  the  peripheral  speed.  Combining  both  these  means  it 
is  possible  to  vary  the  speed  of  a  pressure  tiurbine  for  a  given  head 
and  capacity  in  the  ratio  6:1. 

Three  different  designs  for  reaction  turbine  runners  are  shown 
in  Figs.  104,  105,  and  106.  The  first,  Fig.  104,  represents  a  low- 
speed  runner  which  would  be  used  for  relatively  high  heads  and 
relatively  small  quantities  of  water.    The  bucjcet  angle  j8  is  less 

202 


Digitized  by 


Google 


TURBINES 


203 


Fig.  104. — Low-speed  Runner. 


j=i 


Fig.  105. — Medium-speed  Runner. 


Flo.  106.— High-speed  Runner. 


Digitized  by 


Google 


204  HYDRAULIC  EQmPMENT 

than  90^  and  the  angle  a  of  the  water  leaving  the  guides  is  also 
small. 

Fig.  105  represents  a  medium-speed  runner,  the  angle  P  being 
approximately  90^  and  the  angle  a  larger  than  in  the  previous 
type. 

Fig.  106  represents  a  high-speed  runner  for  low  heads  and  rel- 
atively large  quantities  of  water.  It  is  seen  that  the  angle  fi  is 
larger  than  90^  and  angle  a  also  considerably  larger  than  before, 
thus  giving  a  very  high  peripheral  velocity. 

Comparing  the  above  types  it  is  also  seen  the  width  of  the 
buckets  varies  from  a  comparatively  narrow  to  a  wide  size,  while, 
on  the  other  hand,  the  shape  changes  from  a  forward  to  a  back- 
ward curved  bucket  as  the  speed  increases. 

Impulse  Turbines.  The  impulse  or  tangential  turbine  is 
generally  known  as  the  Pelton  turbine.  It  is  a  kinetic  energy 
wheel,  the  water  being  discharged  from  one  or  more  nozzles  against 
a  number  of  buckets  attached  to  the  periphery  of  the  runner, 
and  the  momentum  of  the  mass  of  water  in  its  impulse  upon  the 
runner  buckets  is,  therefore,  the  main  principle  utilized  in  the 
energy  transformation.  When  the  water  leaves  the  buckets  it  is 
moving  at  so  slow  an  absolute  velocity  that  practically  its  entire 
energy  has  been  imparted  to  the  runner. 

Since  usually  the  number  of  nozzles  is  small  as  compared  with 
the  number  of  buckets,  the  latter  are  in  active  use  only  during 
piart  of  a  revolution,  and  hence  this  type  of  prime  mover  is  some- 
times called  a  water  wheel  instead  of  a  turbine.  This  distinction 
seems,  however,  rather  arbitrary,  ancl  can  probably  be  traced  to 
an  attempt  to  show  that  the  impulse  turbine  is  derived  from  an 
undershot  water  wheel.  It  seems,  therefore,  better  to  consider 
both  types  as  turbines,  since  they  both  really  involve  the  same 
principles  and  action.  As  a  matter  of  fact,  the  term  water  wheel 
is  loosely  applied  to  all  sorts  of  turbines. 

The  description  of  the  tangential  type  of  impulse  turbine  is 
given  on  page  242. 

The  speed  of  an  impulse  turbine  of  a  given  diameter  is  variable 
only  within  very  small  limits.  The  speed  is  practically  deter- 
mined by  the  head,  and  can  be  varied  only  by  variation  of  the 
runner  diameter. 

Selection  of  Turbines.  In  deciding  upon  the  ntunber,  capacity 
and  speed  of  the  units  in  a  water-power  station,  the  combination 


Digitized  by 


Google 


TURBINES  205 

of  the  tiirbine  and  the  generator  must  necessarily  be  considered 
together.  Besides  hydraulic  conditions  such  as  the  head  and  its 
variations^  storage  facilities,  etc.,  and  the  limitations  of  the 
turbine  design,  a  proper  selection  is  governed  by  the  load  factor, 
the  natiu^  of  the  load,  the  reserve  capacity,  the  reliability  and 
flexibiUty  of  the  service,  the  cost  and  operating  expenses,  etc. 
The  units  should  be  operated  as  near  full  load  as  possible  and  new 
units  should  preferably  be  started  as  the  load  increases  instead  of 
utilizing  overload  capacities.  Where  sudden  overloads  of  con- 
siderable magnitude  come  on  the  system  for  short  periods  it  is, 
of  course,  necessary  to  have  turbine  capacity  sufficient  to  care  for 
them.  Single  units  are  never  desirable  except  for  multiple-plant 
83rstems,  in  which  case  the  necessary  reserve  can  be  obtained  from 
other  stations.  For  single-plant  systems  the  number  of  units 
should  preferably  not  be  less  than  three  or  fom*,  but  above  this 
the  number  should  be  governed  by  the  upper  limit  in  design,  con- 
sidered both  from  a  technical  and  economical  standpoint.  With 
a  small  number  of  large  units  the  first  cost,  the  maintenance 
charge  and  the  necessary  floor  space  is  reduced,  and  the  efficiency 
is  also  usually  better  than  for  a  larger  number  of  smaller  units. 
The  ultimate  development  may  also  influence  the  size,  and  it  may 
be  found  advisable  to  provide  larger  units  for  the  initial  develop- 
ment than  would  otherwise  have  been  chosen. 

In  water  developments  by  far  the  larger  majority  of  installa- 
tions are  subject  to  wide  variations  in  the  head.  In  many  of 
the  low-head  installations  the  back  water  may  bring  about  a 
change  in  head  which  is  l)eyond  the  capacity  of  one  wheel  or 
runner  to  accommodate,  and  in  some  cases  additional  runners 
must  be  mounted  on  the  same  shaft  and  cut  into  service  at  times 
of  low  head.  In  many  of  the  large  developments  this  change  in 
head  is  the  limiting  feature  in  design  of  the  water  wheel  as  related 
to  the  generator  capacity,  for  in  all  electrical  work  it  is  essential 
that  the  speed  of  the  generator  be  kept  constant. 

It  is  very  generally  known  that  the  peripheral  speed  of  a 
water  wheel  bears  a  certain  ratio  to  the  spouting  velocity  of  the 
water  under  any  given  head,  this  ratio  as  a  percentage  varying 
between  40  and  50  per  cent  for  impulse  turbines  and  between  60 
and  80  per  cent  for  reaction  turbines.  Hence  the  percentage 
variation  from,  say,  a  mean  of  60  per  cent  is  only  33|  above  and 
33|  below  for  any  given  head.    But  the  diameter,  and  conse- 


Digitized  by 


Google 


206  HYDRAULIC  EQUIPMENT 

quently  the  R.P.M.  corresponding  to  the  peripheral  speed  may 
.vary  widely  according  to  type,  make,  and  number  of  runners  or 
jets. 

Specific  Speed.  Turbine  runners  of  different  makes  are  best 
compared  on  the  basis  of  their  specific  speeds,  this  being  the 
nmnber  of  revolutions  per  minute  at  the  point  of  maximum 
efficiency  that  a  homologous  or  geometrically  similar  wheel  would 
give  if  it  were  to  deUver  1  horse-power  under  unit  head,  usually 
1  foot.  With  the  same  specific  speeds  the  different  designs  vary 
comparatively  little,  it  being  the  aim  of  manufacturers  to  produce 
a  line  of  turbines  covering  all  specific  speeds  with  the  highest  effi- 
ciencies possible  at  each  specific  speed,  and  turbines  for  use  under 
low  heads  should  have  as  high  a  specific  speed  as  possible  with- 
out sacrificing  efficiency  or  other  desirable  characteristics.  After 
a  certain  design  has  been  adopted  for  a  certain  specific  speed,  a 
full  series  of  such  tiurbines  can  be  laid  out,  all  of  identical  design 
with  the  original,  each  being  an  enlargement  or  reduction  of 
another. 

If 

Q  =  quantity  of  water; 

A=head; 

D  =  diameter  of  runner; 

then  for  any  ^.ven  tmrbine: 

Q  varies  as  h^^^; 

H.P  varies  as  QXh  or  A'^*; 

R.P.M.  varies  as  h^^\ 

Hence,  the  horse-power  delivered  under  1  foot-head  will  be 
fi^i  =  -TS7r*^d  the  speed  will  be  R.P.M.i  =  — ^tt^ — ' 

If  now  the  head  is  kept  constant  and  it  is  assumed  that  all 
dimensions  of  the  runner  are  reduced  proportionally,  then  the 
dimensions  will  all  remain  in  fixed  ratio  to  the  diameter,  Z>,  and 
all  areas  of  passages  through  the  runner  will  vary  in  proportion  to 
D^;  the  velocities  remaining  constant  on  account  of  the  constant 
head. 

Therefore,  for  tm-bines  of  homologous  or  geometrically  similar 


Digitized  by 


Google 


TURBINES  207 

design,  but  built  in  various  sizes  and  operated  under  the  same 
head: 

Q  varies  as  D^; 

H.P.  varies  as  D^; 

R.P.M.  varies  as  yr. 

Hence,  the  speeds -of  a  set  of  similar  runners,  operating  under 
the  same  head,  will  vary  inversely  as  the  square  roots  of  their 
horse-powers,  and  if  one  runner  gives  a  speed  of  R.P.M.  with  a 
power  H.P.,  it  follows  that  the  speed  of  a  1  H.P.  turbine  will  be 
R.P.M.xVH.P.  Thus,  if  the  head  be  1  foot,  the  speed  of  the 
1  H.P.  runner  or  its  specific  speed,  iV„  will  be 


^,  _R.P.M.^    /H.R. 

^'-        h'/2        ^\    h^/t  ' 

or  

Vhp 

.V.=R.P.M.X-^. 

If  it  is  desired  to  obtain  the  specific  speed  according  to  the  metric 
system  with  English  units  (Ft.  and  H.P.)  used  in  the  formula,  mul- 
tiply the  values  obtained  from  the  above  formula  by  4.45.     In 

transferring  we  have  1  foot  equal  to  ^-^  meter  and  1  English  H.P. 

equal  to  0.986  metric  H.P.    Thus 


A6/4xV0.986  /iV4 

The  value  A*/*  may  readily  be  figured  out  as  follows: 

The  diagram  in  Fig.  107  supplies  a  convenient  graphic  method 
of  deducing  the  specific  speed  of  a  runner  from  any  given  set  of 
conditions  without  the  use  of  the  formula. 

In  figuring  the  specific  speed  of  a  turbine  with  more  than  one 
runner  or  nozzle,  the  H.P.  used  should,  of  course,  be  the  output 
from  each  runner  or  nozzle.  Furthermore,  as  the  above  formula 
applies  to  single-runner  turbines,  it  follows  that  in  the  case  of  a 
turbine  of  the  same  capacity  having  n  runners  of  ||ie  same  spe- 
cific speed,  it  is  seen  that  the  R.P.M.  would  be  Vn  times  the 


Digitized  by 


Google 


208 


HYDRAULIC  EQUIPMENT 


R.P.M.  of  the  single-runner  turbine.    It  is  also  readily  seen  that 
for  a  given  value  of  R.P.M.  and  A,  the  H.P.  output  is  propor- 

Bpeciflc  Speed 

'-,  \\ 


^  SPECIFIC   BPEtD  CHWIT 

Itirii  lIiH  Fi-lnt  LBflhntrt  "       "■ 

lUcd-JlfLH     H.P.H.*1T4b 
Em  ^^>^  xf^*!^ 


Fig.  107.— Specific  Speed  Chart. 
(By  Courtesy  of  Wellman-Seaver-Morgan  Company). 

tional  to  the  square  of  the  specific  speed,  and  also  that  for  a  given 
head  and  H.P.  ti|e  R.P.M.  of  a  turbine  is  proportional  to  the  spe- 
cific speed. 


Digitized  by  CjOOQ IC 


TURBINES 


209 


By  comparison  of  the  specific  speeds  it  is  possible  to  judge  the 
characteristics  of  water-wheel  runners  without  considering  their 
actual  speed;  power  or  head.  Other  things  being  equal,  a  high 
specific  speed  means  a  high  actual  speed,  and  a  low  specific  speed, 
a  low  actual  speed  in  revolutions  per  minute.  With  low-hiead 
developments  the  speed  must  be  selected  as  high  as  is  good  en- 
gineering practice  in  order  to  keep  down  the  weight  and  conse- 
quently the  cost  of  the  generators.  With  very  high  heads  it  is 
mostly  a  question  of  keeping  the  speed  reasonably  low  so  as  to 
avoid  the  use  of  costly  high-speed  generators.  The  limit  of  high 
speed  for  low-head  developments  is  fixed  by  the  progress  of  the  art 
of  designing  high-speed  runners,  and  the  limit  of  low  speeds  under 
high  heads  is  fixed  by  the  risks  involved  in  designing  runners  for 
operation  with  very  low  coeflBicient  of  specific  speed. 

High  speed  imder  low  heads  also  means  large  discharge  capacity 
per  unit  diameter,  resultuig  in  a  large  power  capacity,  while  low 
speeds  under  high  heads  mean  a  small  discharge  capacity  per  unit 
diameter  resulting  in  a  small  width  of  runner. 

Reaction  turbines  have  been  built  for  specific  speeds  as  low  as  7, 
but  12  is  probably  as  low  as  should  be  used  in  normal  practice. 
Starting  at  this  speed,  the  efficiency  will  increase  as  the  specific 
speed  approaches  more  normal  values,  the  efficiency  reaching 
the  highest  values  between  specific  speeds  of  25  and  75.  Above  90 
it  drops  again  at  a  rapidly  increasing  rate,  and  above  100  specific 
speed  results  are  much  more  problematic,  at  least  for  the  present, 
but  efficiencies  over  90  per  cent  have  been  obtained  with  specific 
speeds  all  the  way  from  25  to  90. 

Pfau  iQ  his  paper  before  the  International  Elngineering  Con- 
gress in  San  Francisco  classifies  tentatively  Francis  reaction  tur- 
bines as  follows,  the  types  being  claimed  to  operate  successfully 
under  the  heads  given. 

TABLE  XXXVni 


Type. 

Specific  Speed. 

Head.  Feet. 

A 

Very  low,         20-25 

750 

B 

Low,                 25-30 

400 

C 

Medium  low,    30-40 

175 

D 

Medium  high,  40-60 

90 

B 

High,                60-80 

40 

P 

Very  high,        80-100 

20 

Digitized  by 


Google 


210  HYDRAULIC  EQUIPMENT 

For  impulse  turbines  of  the  Pelton  type,  specific  speeds  down 
to  very  low  values  may  be  obtained  with  good  results.  The 
highest  efficiencies  may  perhaps  be  obtained  with  specific  speeds 
varying  from  1  to  4,  and  will  then  be  increasingly  reduced  as  the 
speed  is  increased  up  to  about  6|  or  7,  which  might  be  taken  as  the 
extreme  limit,  the  figures  appl3dng  to  single-nozzle  wheels.  The 
maximum  obtainable  efficiencies  with  an  impulse  wheel  may  be 
taken  as  between  85  and  89  per  cent,  these  being  figures  to  the 
center  line  of  the  nozzle.  If  two  or  more  nozzles  are  used  on  the 
same  wheel,  a  reduction  of  several  per  cent  will  result,  and  of  course 
the  power  will  be  increased  in  proportion  to  the  number  of  nozzles. 

Where,  therefore,  the  specific  speed  characteristics  exceed  the 
above  values,  multi-runner  turbines,  more  than  one  nozzle  or 
smaller  units  must  be  used. 

There  is  no  hard  and  fast  rule  for  the  choice  of  a  reaction  tur- 
bine or  an  impulse  turbine,  the  field  of  their  respective  usefulness 
overlapping  to  a  considerable  degree.  The  thing  which  limits  the 
specific  speed  of  reaction  turbines  under  high  heads  is  the  risk  of 
corrosion  of  the  buckets,  while  the  reason  for  the  non-use  of 
impulse  turbines  imder  very  low  heads  is  the  lack  of  economy 
due  to  large  dimensions  and  low  speed  required.  The  reaction 
type  of  turbine  is  generally  used  for  heads  from  10  to  600  feet 
and  the  impulse  type  between  300  and  3000  feet.  The  proper 
system  seems,  therefore,  to  be  very  well  fixed  for  low  or  high 
heads,  while  for  the  intermediate  range  between  about  300  and  600 
feet,  the  proper  system  must  be  determined  by  referring  to  the 
specific  speed  characteristics. 

Assume,  for  example,  an  installation  having  a  head  of  1900 
feet  and  where  the  generators  would  require  turbines  of  10,000 
H.P.  capacity  running  at  375  R.P.M.  What  type  of  wheel 
should  be  installed?  

_      v^io;^_ 

and,  consequently,  an  impulse  turbine  should  be  selected. 

On  the  other  hand,  with  a  10,000  H.P.  wheel  to  operate  at 
57.7  R.P.M.  under  a  32-foot  head,  we  get 

Ar.=57.7X^=76, 
and  a  reaction  turbine  must  be  used. 


Digitized  by 


Google 


TURBINES  211 

In  selecting  a  type  of  wheel,  the  number  of  runners  or  nozzles, 
their  capacity  and  speed  must  be  chosen  with  a  view  of  obtaining 
not  only  the  highest  eflSiciency,  but  also  the  most  economical  com- 
bination of  the  prime  mover  and  generator.  For  example,  a 
wheel  is  to  be  operated  under  a  head  of  400  feet  and  develop  1500 
H.P.  How  many  nozzles  should  it  have  and  at  what  speed  may  it 
operate?  Assume  that  a  specific  speed  of  4  will  give  a  good  efS- 
ciency  for  the  wheel,  then  the  actual  speed  of  the  imit  will  be 

R.P.M.=i><S=185. 
Vl500 

This  speed,  however,  may  be  entirely  too  low  for  the  generator, 
and,  by  providing  two  nozzles,  each  supplying  750  H.P.  the 
speed  would  be  increased  to 

R.P.M.  =  186XV2  =  260. 
and  with  four  nozzles 

R.P.M.  =  186XV4  =  370. 

Let  us  also  see  what  the  result  would  be  if  we  tried  to  apply  a 
reaction  turbine  running  at  720  R.P.M.  The  specific  speed  would 
then  be 

JV.=720X^=16. 

and,  consequently,  this  type  would  undoubtedly  be  the  most 
advantageous  to  use  for  our  case. 

The  efficiencies,  especially  at  partial  load,  are  related  to  the 
specific  speed,  the  curves  of  high  specific  speed  runners  being  more 
pointed  than  with  the  low  specific  speed  type,  thus  allowing  a 
narrower  margin  for  operation  under  the  best  conditions.  This 
is  clearly  shown  in  the  curves  in  Fig.  108. 

The  maximum  full-load  capacity  of  a  turbine  is  that  point 
beyond  which  the  output  decreases  with  an  increase  in  gat«  open- 
ing. The  margin  between  the  point  of  maximum  efficiency  and  of 
maximimi  capacity  depends  upon  the  specific  speed  of  the  runner, 
and  is  smaller  the  higher  the  specific  speed.  This  is  illustrated 
in  Fig.  108,  which  shows  that  as  the  specific  speed  is  increased  the 
point  at  which  maximum  efficiency  occurs  approaches  nearer  to 
the  power  delivered  at  full  gate  opening.  The  specific  speed  may 
thus  be  increased  to  such  an  extent  that  the  point  of  maximum 


Digitized  by 


Google 


212 


HYDRAULIC  EQUIPMENT 


efficiency  and  maximum  output  coincide.  With  low  heads  and 
high  specific  speeds  it  is,  therefore,  desirable  to  operate  wheels 
near  their  point  of  maximum  output,  and  to  obtain  the  best  re- 
sults the  generator  should  be  designed  with  consideration  to  this 
point. 

Referring  again  to  the  curves  in  Fig.  108,  it  will  be  noted  that 
the  full-load  capacity  occiuB  at  about  6  per  cent  above  normal  or 
rated  full  load  in  all  three  cases.  This  is  in  accordance  with  the 
general  practice,  the  margin  being  allowed  for  governing.  It  is 
also  noted  that  for  ciures  B  and  C  the  efficiency  is  falling  off  very 
rapidly  at  6  per  cent  overload,  and  that  should  the  gate  be  opened 
still  further  the  output  would  reduce  instead  of  increase.    If, 


I" 

A  80 


-= 

^ 

r 

^ 

-- 

=:=: 

r:: 

== 

"^ 

^ 

?^ 

■ 

/ 

/ 

A-  Specific  Speed  -25  Head-UO  Ft. 
B-                  "    -80    •«     -  «  " 
C-       "            "     -M     ♦'      -  90  " 

/ 

/ 

/ 

V 

0 

// 

/ 

J 

7 

L 

5     10    16     20    tf 


80    35    40    45    50    65    60    66     WTO    80    86 
Hone  Power  in  Per  cent  of  Normal  Load 


90    96  100  106  U( 


Fig.  108. — Performance  Curves  of  Several  Turbines  for  Various  Heads  and 
Specific  Speeds  as  Shown. 

with  low  specific  speed  wheels,  as  represented  by  Curve  A,  the 
gates  were  still  further  opened,  the  power  would  continue  to 
increase  to  some  extent. 

The  point  of  maximum  efficiency  for  wheels  represented  by 
curve  A  occmB  at  about  90  per  cent  of  normal  full  load,  in  the  case 
of  B  at  93.5  per  cent,  while,  in  the  case  of  C,  the  maximum  effi- 
ciency occurs  just  at  the  point  of  normal  or  rated  full  load.  Thus, 
as  stated  before,  the  power  at  which  the  maximum  efficiency  occurs 
approaches  nearer  to  full  load  as  the  specific  speed  increases. 

For  wheels  of  low  or  moderate  specific  speed,  as  represented  by 
curve  A,  the  efficiency  remains  very  high  over  a  very  large  range 
in  power,  while  for  wheels  of  high  specific  speed,  curve  C,  the 
efficiency  falls  off  rapidly  as  the  power  is  reduced  below  the  normal 


Digitized  by 


Google 


TURBINES  213 

full  load.  For  this  reason  it  is  desirable  to  run  low-head  wheels 
under  practically  full  load  conditions.  With  high-head  wheels 
this  is  not  so  important,  as  the  efficiency  is  still  high  at  partial 
loads.  With  wheels  as  represented  by  curve  C,  it  is  also  neces- 
sary to  allow  some  margin  above  the  normal  full  load  for  govern- 
ing, as  it  IB  desirable  to  operate  the  turbine  at  its  point  of  max- 
imum efficiency.  With  high-head  wheels,  curve  A,  such  a  margin 
need  not  be  allowed. 

The  curves  plotted  in  Fig.  108  represent  operating  conditions 
under  constant  head.  This,  however,  is  not  always  realized, 
especiaUy  in  low-head  plants  where  floods  and  dry  seasons  some- 
times cause  quite  a  variation  in  the  head,  and  this  has,  as  pre- 
viously mentioned,  quite  a  bearing  on  the  selection  of  the  water 
wheel,  and  should,  therefore,  be  given  careful  consideration. 

If  the  speed  of  the  unit  could  be  allowed  to  vary  at  all  times  the 
square  root  of  the  ratio  of  the  heads,  the  shape  of  the  performance 
curve  for  any  head  other  than  normal  would  be  the  same  as  that 
secured  at  normal  head,  but  the  output  would  vary  as  the  J 
power  of  the  ratio  of  the  heads.  In  the  case  of  wheels  driving 
alternating-current  generators  a  speed  variation  is  not  permissible 
and  the  speed  must  be  kept  constant  irrespective  of  any  varia- 
tion in  head  which  may  occur,  and  this  will  still  further  lower  the 
output  due  to  the  reduced  efficiency  when  operating  at  the  best 
head  and  speed. 

In  Fig.  109  is  plotted  a  set  of  curves  illustrating  the  effect  of 
a  varying  head.  A  10,000-H.P.  tiu*bine  is  assumed  to  operate 
normally  under  a  32-foot  head,  the  speed  to  be  constant  for  a 
range  of  heads  from  26  to  38  feet.  As  the  head  goes  up  to  38  feet 
the  shape  of  the  curve  approaches  more  closely  curve  B  in  Fig.  108, 
while,  when  the  head  falls  to  26  feet,  the  speed  being  constant,  it 
approaches  more  closely  to  curve  C  In  other  words,  when  oper- 
ating under  a  38-foot  head,  the  speed  is  lower  than  the  best  speed 
for  the  runner  under  that  head,  while,  when  operating  under  the 
2&-foot  head,  the  speed  of  the  wheel  is  higher  than  the  best  speed. 
Under  38-foot  head  the  point  of  maximum  efficiency  is,  further- 
more, considerably  below  the  normal  full  load  at  that  head,  while, 
under  26-foot  head,  the  power  at  which  maximum  efficiency  occurs 
is  the  actual  full  load,  illustrating  the  points  discussed  above  in 
reference  to  the  relation  of  the  power  at  which  maximum  efficiency 
occurs  and  the  normal  full-load  power  for  various  specific  speeds. 


Digitized  by 


Google 


214 


HYDRAULIC  EQUIPMENT 


Let  us  assume  that  a  selection  of  a  wheel  is  to  be  made  for  an 
installation,  and  that  performance  curves  are  desired,  showing  the 
expected  efficiency  for  various  loads  and  speeds.  Curves  A,  B,  and 
C,  in  Fig.  108  may  each  represent  a  possible  curve,  dependent 
upon  the  revolutions  selected  for  the  turbine  in  question,  the  rev- 
olutions being  directly  proportional  to  the  specific  speeds,  and 
they  will  illustrate  the  manner  in  which  the  efficiencies  at  partial 
gate  openings  will  fall  off  in  any  one  case,  depending  upon  the 
actual  revolutions  per  minute  selected  for  the  design  of  the  wheel. 


Qrt 

~ 

g 

^ 

g 

f 

- 

flO 

% 

n 

% 

S 

1 

"n 

-«v 

s 

1" 

Sj  An 

tf 

^ 

d 

% 

i 

i 

1 

1 

1 

% 

^ 

P 

«0 

dS 

a 

Iff 

IS 

h 

s 

y 

^ 

1 

•3 

•a 

1 

1 

^ 

•j 

^ 

W 

^ 

n 

1 

1 

1 

n 

iS 

?: 

^30 

— 

J 

M 

f 

^20 

^ 

10 

jf 

• 

/] 

5    10Ua0253035  40   45   60fi56065T0   7580869095  100  106  UO 115120  IS  ISO  IS 
Hone  Power  In  Per  cent  of  Normal  Full  Load  at  Somial  Head 

Fig.  109. — 10,000  Horse-power  Turbine-curves  Showing  Efficiency  and  Power 
for  Constant  Speed  and  a  Normal  Head  of  32  Feet  for  Various  Heads 
as  Shown. 

They  will  also  give  an  idea  as  to  the  margin  between  the  normal 
full  load  and  the  power  at  which  the  points  of  maximum  efficiency 
will  occur.  In  the  selection  of  a  speed  for  any  installation,  there- 
fore, aside  from  the  cost  of  the  generators,  the  question  of  the 
wheel  efficiencies  at  partial  gate  openings  has  a  considerable 
bearing.  Where  a  unit  is  likely  to  operate  under  a  very  wide 
range  in  power,  it  would  be  advisable  to  select  a  wheel  repre- 
sented by  curve  A,  giving  a  high  efficiency  for  a  considerable 
range  in  power. 

Actual  Speed.  For  speeds  used  in  a  number  of  hydro-electric 
developments  and  corresponding  heads  and  capacities,  see  table 
on  page  316. 

Characteristic  Curves.  For  studying  the  action  of  a  turbine 
under  different  conditions  of  operation,  the  characteristic  curves, 
as  shown  in  Figs.  110,  111,  and  112,  are  extensively  used,  and 
from  these  curves  it  can  at  a  glance  be  seen  at  what  speed  the 


Digitized  by 


Google 


Fig.  110. 


^.Efficiency  Curves 

of  an 
I^.  Morris  Ranner 


1.0      i.i       1.2       1.3       1.4 


.2        .S        .4 

Fig.  111. 


.6        .1         .8         .9 

^•H.Pi.  Ouryes 

of  au 

I.P.  Morris  Ranner 


1.0       1.1      1.2       1.3      1.4 


Fig.  112. 


.6         .7         .8         .9        1.0       1.1       1.2       lot       1.4 

Eqaal— Efficiency 

ContoarCnrree 

of  an 

LP.  Morris  Ronnex 


215 


Digitized  by 


Google 


216  HYDRAULIC  EQUIPMENT 

turbine  should  be  run  for  the  best  efficiency  at  any  gate  open- 
ing. 

To  show  how  the  curves  are  constructed  a  typical  example  is 
given,  based  on  actual  tests  given  in  Table  XXXIX.  The  values  of 
the  abscissse  in  the  curves  represent  0,  the  coefficient  of  peripheral 
velocity,  although  sometimes  values  of  HPi,  are  used  for  this 
piupose.  These  two  quantities  are,  however,  directly  propor- 
tional, so  that  the  change  merely  affects  the  abscissa  scale  of  the 
curves. 

From  the  report,  the  values  of  head,  revolutions  per  minute, 
horse-power  and  efficiency  are  taken  from  the  corresponding  col- 
umns for  the  various  runs  at  each  gate.  From  these  are  com- 
puted the  corresponding  values  of  4>,  which  is  equal  to 

DXR.P.M. 


0  = 


Q0X^2gh' 


in  which  D  is  the  nominal  diameter  of  the  runner  in  feet,  in  the 
case  taken  25  inches  or  2.0833  feet,  and  h  is  the  head 
in  feet.  The  corresponding  values  of  HPi  or  the  horse-power 
reduced  to  1  foot  head  are  also  computed  by  dividing  the 
given  horse-power  by  the  three-halves  power  of  the  head, 
thus 

The  result  represents  the  power  which  the  given  runner  would 
develop  if  operated  at  the  corresponding  speed,  that  is,  at  the  same 
value  of  4>  under  a  head  of  1  foot  instead  of  the  head  used  in  the 
test.  The  0-efficiency  and  i^-HPi  curves  are  now  plotted  for 
each  gate  opening.  Figs.  110  and  111. 

In  order  to  construct  the  curves  of  equal  efficiency,  Fig.  112, 
the  <t>-HPi  diagram  is  selected  and  points  having  the  same 
efficiency  on  the  ciures  for  different  gate  openings  are  joined  in  a 
curve.  In  order  to  keep  the  diagrams  clear  the  <t>-HPi  curves 
have  been  repeated  in  dotted  lines  on  a  separate  diagram. 

To  illustrate  the  construction  of  one  of  the  equal  efficiency 
contours  for  instance,  take  the  line  for  86  per  cent  efficiency. 
Referring  to  the  0-efficiency  diagram,  a  horizontal  line  repre- 
senting the  efficiency  selected  will  intersect  the  curve  for  2.5  gate 
at  0=0.64,  and  will  again  intersect  the  same  gate  at  0=0.725. 


Digitized  by 


Google 


TURBINES 


217 


TABLE  XXXIX 

TE8TING   FlUMB   OF  THE   HOLTOKE   WaTEB   PoWEB   Co.,   HoLTOEE,    MaSS. 

Report  of  tests  of  a  25-inch  Right-hand  I.P.  Morris  Company  turbine  wheel. 


h 

1 

«a 

II 

5^ 

1^ 

©  1^ 

11 

1^ 

11 
1^ 

if 

Z 

O 

& 

W 

Q 

cr 

(U 

s 

55 

4.00 

0.917 

17.74 

3 

182.33 

97.79 

146.49 

74.62 

54 

4.00 

0.924 

17.72 

3 

200.33 

98.49 

161.27 

76.60 

53 

4.00 

0.939 

17.71 

226.75 

99.99 

167.62 

78.61 

56 

4.00 

0.948 

17.70 

241.25 

100.92 

160.30 

79.31 

52 

4.00 

0.961 

17.66 

266.25 

102.21 

162.63 

79.68 

51 

4.00 

0.984 

17.52 

275.00 

104.20 

166.12 

80.42 

49 

4.00 

0.999 

17.53 

295.00 

106.86 

169.29 

80.63 

50 

4.00 

1.005 

17.51 

302.00 

106.44 

169.66 

80.45 

48 

4.00 

1.015 

17.48 

312.67 

107.40 

169.99 

80.02 

47 

4.00 

1.018 

17.51 

321.76 

107.76 

166.20 

77.38 

46 

4.00 

1.006 

17.54 

340.76 

106.68 

144.09 

68.05 

45 

4.00 

0.985 

17.60 

362.60 

104.66 

109.49 

62.58 

43 

3.50 

0.837 

17.84 

183.76 

89.62 

140.97 

78.01 

42 

3.50 

0.850 

17.83 

206.33 

90.87 

149.66 

81.58 

44 

3.50 

0.860 

17.76 

221.60 

91.76 

163.87 

83.44 

41 

3.50 

0.862 

17.78 

221.76 

91.99 

164.04 

83.23 

38 

3.50 

0.877 

17.79 

241.50 

93.67 

160.47 

86.20 

37 

3.50 

0.894 

17.72 

261.60 

96.27 

166.86 

86.83 

40 

3.50 

0.905 

17.67 

272.80 

96.29 

168.09 

87.31 

39 

3.50 

0.911 

17.67 

279.76 

96.87 

168.99 

87.25 

36 

3.50 

0.909 

17.69 

287.76 

96.76 

166.14 

86.26 

35 

3.50 

0.906 

17.69 

302.26 

96.41 

166.19 

80.42 

34 

3.50 

0.897 

17.71 

323.00 

95.60 

136.68 

71.37 

33 

3.50 

0.879 

17.75 

361.60 

93.68 

106.16 

66.42 

82 

3.10 

0.784 

17.88 

191.20 

83.80 

138.60 

81.65 

76 

3.10 

0.814 

17.75 

236.76 

86.76 

164.46 

88.65 

76 

3.10 

0.821 

17.75 

244.80 

87.63 

166.76 

89.26 

77 

3.10 

0.821 

17.75 

246.80 

87.63 

167.28 

89.67 

74 

3.10 

0.824 

17.73 

249.76 

87.86 

168.41 

89.82 

81 

3.10 

0.824 

17.73 

260.00 

87.97 

168.67 

89.69 

78 

3.10 

0.827 

17.73 

261.00 

88.08 

158.44 

89.67 

79 

3.10 

0.828 

17.72 

5 

262.40 

88.19 

168.66 

89.67 

80 

3.10 

0.827 

17.75 

4 

264.26 

88.19 

168.96 

89.74 

Digitized  by 


Google 


218 


HYDRAULIC  EQUIPMENT 


TABLE  XXXIX— Continued 

Tksting  Flume  of  the  Holtoke  Water  Power  Co.,  Holtoke,  Mass. 

Report  of  tests  of  a  25-mch  Right-hand  I. P.  Morris  Company  turbine  wheel. 


H 
II 

roportional  Part  of 
the  Full  Discharge 
of  the  Wheel  in  Per 
Cent. 

1^ 

5-! 

g5i 

si 

uantity  of  Water 
Discharged  by  the 
Wheel.     Cu.ft.  per 
Sec. 

u 

i 

1^ 

z 

^ 

P^ 

» 

Q 

tf 

Ct 

£ 

w 

73 

3  10 

0.S24 

17.74 

4 

266.75 

87,86     ! 

156,30 

88.62 

72 

:i.io 

0,823 

17.75 

4 

265,75 

87,75 

152.60 

86  63 

71 

3.10 

0.823 

17.76 

3 

274.00  ' 

87.75 

148,96 

84  47 

70 

3.10 

0.S20 

17.77 

4 

187.53 

87.53 

140.87 

80.04 

60 

3,10 

0,820 

17,77 

4 

309.50 

87.53 

130.87 

74,36 

6S 

3  10 

0,817 

17.80  ' 

4 

346.25 

87.20 

104.58 

59  54 

25 

3,00 

0.760 

17.98 

4 

181.00 

80.31 

130.11 

79,00 

30 

3.00 

0.776 

17.87 

4 

221.25 

83,02 

147  01 

87,68 

23 

3.00 

0.785 

17.83 

4 

233,25 

83.90 

150.76 

89.07 

22 

3.00 

0.790 

17,83 

3 

239.00 

84.44 

153.03 

89,83 

19 

3.00 

0,791 

17,85 

4 

242.00 

84,55 

153.48 

89.88 

21 

3,00 

0,792 

17.83 

4 

242.50 

84.66 

153,81 

90.05 

24 

3.00 

0.792 

17,81 

4 

243,00 

84,56 

152  66 

89.59 

IS 

3.00 

0.790 

17.85 

4 

249-50 

84.44 

150.71 

88.37 

17 

3,00 

0,788 

17.86 

4 

257.25 

84.23 

147.63 

86.73 

16 

3,00 

0,784 

17.86 

10 

265.10 

83.90 

144,12 

85.00 

15 

3.00 

0.786 

17.84 

3 

293.67 

84,01 

133,06 

78.45 

14 

3.tK) 

0.787 

17.84 

4 

323.25 

84.12 

117.16 

68.99 

13 

3.rK) 

0,780 

17. 87 

3 

361.00 

83.46 

87.23 

51  69 

9 

2-.50 

0.047 

18,12 

4 

169.75 

69.69 

114  85 

SO  37 

8 

2-50 

0.662 

18.07 

4 

197.00 

71.24 

124.95 

85.78 

7 

2.50 

0.668 

18.06 

3 

210,67 

71,86 

127.26 

86.66 

12 

2.60 

0,669 

18.03 

5 

213,20 

71.86 

128.79 

87.85 

10 

2.50 

0,670 

18.04 

6 

217.50 

72.06 

128.76 

87.53 

6 

2.50 

0.666 

18.07 

5 

221.80 

71.65 

127,28 

86,83 

5 

2.50 

0.664 

18.07 

230,00 

71.44 

126.04 

86.80 

4 

2.50 

0.662 

18.07 

240,00 

71,24 

123  23 

84.60 

3 

2.50 

0,665 

18.05 

251,75 

71.55 

121.66 

83.25 

2 

2.50 

0.665 

18.07 

274.50 

71.55 

116.07 

79.34 

1 

2.50 

0.065 

18,05 

320,00 

71.55 

96.65 

66.14 

11 

2.50 

0.657 

18.10 

356.50 

70.72 

64.60 

44,60 

66 

2.00 

0.S3S 

18.26 

171,50 

58.16 

98.42 

81.90 

67 

2.00 

0,642 

18.24 

186.25 

58,55 

102,38 

84.72 

Digitized  by 


Google 


TURBINES 


219 


TABLE  XXXIX'-CwUinued 
Testing  Flume  of  the  Holyoke  Water  Power  Co.,  Holtoee,  Mass. 
Report  of  tests  of  a  25  Right-hand  I.  P.  Morris  Company  turbine  wheel. 


^1 

It 

1 

|5 

Xrtional  Part  of 
Full  Discharge 
of  the  Wheel  in  Per 
Cent. 

s 

i 

III 

i 

s 

l^ 

uantity  of  Water 
Discharged  by  the 
Wheel.     Cu.ft.  per 
Sec 

II 

4 

2 

O 

^ 

» 

p 

OS 

cr 

Ah 

H 

65 

2.00 

0.542 

18.25 

188.75 

58.64 

102.62 

84.74 

64 

2.00 

0.537 

18.28 

196.75 

58.16 

101.02 

83.98 

63 

2.00 

0.535 

18.29 

205.40 

57.87 

99.26 

82.88 

62 

2.00 

0.535 

18.29 

216.50 

57.87 

98.09 

81.90 

61 

2.00 

0.534 

18.40 

226.00 

57.78 

95.56 

79.87 

60 

2.00 

0.532 

18.31 

238.40 

57.59 

93.61 

78.45 

59 

2.00 

0.535 

18.31 

255.75 

57.97 

92.69 

77.18 

58 

2.00 

0.538 

18.32 

284.75 

58.26 

86.00 

71.21 

57 

2.00 

0.533 

18.33 

5 

334.40 

57.78 

60.60 

50.57 

32 

1.25 

0.332 

18.53 

3 

142.00 

36.18 

55.76 

73.60 

31 

1.25 

0.332 

18.65 

3 

158.33 

36.26 

57.39 

74.99 

30 

1.25 

0.326 

18.66 

3 

185.00 

35.69 

55.88 

74.15 

29 

1.25 

0.323 

18.65 

3 

215.00 

35.29 

51.95 

69.76 

28 

1.25 

0.318 

18.65 

3 

247.33 

34.80 

44.82 

61.03 

27 

1.25 

0.315 

18.65 

3 

276.33 

34.39 

33.38 

46.00 

26 

1.25 

0.310 

18.65 

3 

302.33 

33.83 

18.26 

25.58 

89 

4.00 

0.915 

17.65 

4 

423.00 

97.33 

88 

3.50 

0.842 

17.80 

4 

426.50 

89.86 

87 

3.10 

0.751 

17.95 

3 

418.67 

80.52 

86 

3.00 

0.720 

18.02 

4 

415.75 

77.31 

85 

2.50 

0.604 

18.21 

4 

401.25 

65.23 

84 

2.00 

0.482 

18.39 

4 

381.50 

32.28 

83 

1.25 

0.305 

18.70 

4 

323.50 

33.35 

Passing  to  the  4>-HPi  diagram,  points  are  located  on  the  2.5  gate 
curve  at  the  two  values  of  0  just  found.  These  are  two  points  on 
the  desired  curve.  Similar  points  of  intersection  of  the  86  per 
cent  efficiency  line  with  the  0-efficiency  curves  for  other  gates 
are  similarly  located  and  the  resulting  points  joined  up. 

The  diagram  obtained  may  be  viewed  as  a  contour  map  of  a 
moimd  or  hill,  the  heights  of  which  in  a  direction  perpendicular 
to  the  paper  may  be  imagined  to  represent  efficiency.    If  the  hill 


Digitized  by 


Google 


220 


HYDRAULIC  EQUIPMENT 


is  imagined  to  be  cut  by  a  plane  perpendicular  to  the  paper  and 
intersecting  the  paper  in  an  ordinate  at  any  given  value  of  ^, 
the  resulting  intersection  would  be  a  performance  curve  such  as 
one  of  those  plotted  in  Fig.  109. 

In  Fig.  109,  of  course,  the  horse-power  has  been  stepped  up  to 
represent  a  large  runner  operating  under  a  given  head.  The 
performance  curves  can  be  conveniently  plotted  from  the  4^ 
efficiency  and  4^HPi  curves  by  finding  the  corresponding  values 
of  HPi  and  efficiency  for  certain  required  values  of  4>  which  are 
determined  by  the  head  and  speed  in  a  given  installation. 

Speed  Regulation.^  The  most  generally  used  method  for 
governing  the  speed  of  reaction  turbines  is  by  means  of  wicket 
gates  or  guide  vanes  which  change  the  amount  of  water  suppUed 
by  simply  altering  the  water  passage  (see  Fig.  113).    The  vanes 


Fia.  113. — Typical  Arrangement  of  Vertical  Reaction  Turbine,   Showing 
Relation  of  Wicket  Guide  Vanes  to  Casing  and  Runner. 

rotate  about  pivots  and  are  fastened  to  a  shifting  ring  by  link 
motion,  the  ring  being  operated  by  pressure  cylinders,  actuated 
by  the  governor.  If  the  velocity  of  the  water  is  checked  too 
suddenly  dangerous  pressures  may  be  set  up  in  the  pipe  lines  and 
the  speed  regulation  may  be  affected.  In  order  to  avoid  this, 
^  See  also  sections  on  ''  Governors  "  and  "  Waterhammer." 


Digitized  by 


Google 


TURBINES  221 

relief  valves  are  often  provided,  either  of  the  pressure  or  the 
synchronous  by-pass  type.  The  former  is  analagous  to  the 
safety  valve  on  a  boiler  and  does  not  open  until  a  certain  pressure 
has  been  obtained.  The  latter,  however,  is  operated  by  the  gov- 
ernor at  the  same  time  as  the  turbine  gates  but  in  opposite  direc- 
tion, thus  affording  a  by-pass  so  that  there  is  no  reduction  in  the 
flow.  To  prevent  waste  of  water  these  by-passes  may  be  slowly 
closed  by  some  auxiliary  device.  It  is  obvious,  however,  that 
such  water-saving  reUef  valves  are  inoperative  when  the  load  is 
thrown  on,  and,  therefore,  cannot  then  assist  the  speed  governor 
or  prevent  siu-ges  in  the  pipe  lines  caused  by  the  same.  For  pre- 
venting these,  siu-ge  tanks  or  suflSicient  flywheel  effect  of  the  tur- 
bine imit  must  be  relied  upon. 

For  the  speed  regulation  of  impulse  wheels  there  are  three 
methods  in  general  use,  viz.  : 

1.  Hand-regulated  needle  nozzle  with  jet  deflector. 

2.  Needle  regulating  and  deflecting  nozzle. 

3.  Auxiliary  relief  needle  nozzle. 

Either  of  the  above  involve  the  use  of  the  characteristic  needle 
and  nozzle  tip,  a  sectional  view  of  which  is  shown  in  Fig.  114,  the 


Fia.  114. — Sectional  View  of  Pelton-Doble  Needle  Nozzle. 

full  lines  illustrating  the  position  of  the  needle  when  the  nozzle 
is  closed  and  the  dotted  lines  the  needle  position  with  jet  dis- 
charging. 

The  first  sjrstem  consists  of  a  nozzle  body  in  which  is  inserted 
a  concentric  tapered  needle  as  just  described.  By  means  of  this 
needle,  which  is  manually  controlled  for  this  tjrpe  of  nozzle,  the 
jet  area  is  adjusted  intermittently  to  correspond  to  either  the 


Digitized  by 


Google 


222  HYDRAULIC  EQUIPMENT 

stream-flow  or  the  maximum  anticipated  load  likely  to  be  carried 
within  a  certain  time  limit.  The  automatic  speed  regulation  is 
obtained  by  means  of  a  governor  which  actuates  a  deflector, 
which  is  placed  in  front  of  the  nozzle  tip  and  regulates  the  speed 
by  intercepting  or  deflecting  the  stream.  It  is,  therefore,  obvious 
that  this  system  of  regulation  thus  does  not  permit  of  any  economy 
in  the  water  consumption,  unless  the  station  attendant  frequently 
changes  the  needle  adjustment  by  following  closely  the  load 
curve.  It  is,  therefore,  mainly  intended  for  plants  that  are  located 
on  streams  where  water  storage  is  not  feasible,  or  where  other 
power  plants  are  located  on  the  same  stream,  making  it  necessary 
to  allow  the  full  flow  of  the  stream  to  pass  the  plant;  or  on  those 
streams  where  irrigators'  or  riparian  rights  have  a  prime  con- 
trol, thus  preventing  the  storage  of  water.  The  above  holds  also 
for  the  second  class  of  control,  i.e.,  deflecting  nozzles,  as  described 
in  the  next  paragraph  and,  of  course,  also,  to  a  certain  extent,  to 
water-wasting  by-pass  reUef  valves  for  reaction  turbines  as  pre- 
viously described. 

The  needle-deflecting  regulating  nozzle,  as  shown  in  Fig.  115, 
consists  of  an  ordinary  needle  nozzle  which  is  provided  with  a 
ball-and-6ocket  joint,  permitting  it  to  be  raised  or  lowered  so  as 
either  to  direct  the  full  jet  into  the  buckets  of  the  wheel  or  to  par- 
tially or  entirely  direct  the  jet  outside  of  the  path  of  the  buckets 
of  the  wheel. 

In  both  the  stationary  needle  nozzle  with  the  jet  deflector  and 
the  needle-regulating  deflecting  nozzle,  the  needle  is  usually  oper- 
ated by  hand  control,  the  needle  being  set  to  utilize  to  full  advan- 
tage the  available  supply  of  water.  In  plants  where  either  of  these 
types  of  nozzles  is  installed  and  where  there  are  forebay  reservoirs, 
economy  in  the  use  of  water  is  secured  by  setting  the  needle  at 
different  times  during  the  day  to  cany  the  maximum  load  on  the 
plant,  the  needle  being  set  to  follow  the  general  load  curve  of  the 
plant,  while  the  momentary  load  changes  and  speed  control  are 
taken  care  of  by  the  governor  either  operating  the  jet  deflector  or 
deflecting  the  nozzle. 

In  such  plants,  where  large  units  are  installed,  the  control  of 
the  needle  setting  may  be  by  means  of  an  electric  motor  with 
remote  control  from  the  switchboard,  so  that  the  power  plant 
operator  can,  from  the  switchboard,  set  the  position  of  the  needle 
so  as  to  carry  any  predetermined  load  that  is  desired,  the  needle 


Digitized  by 


Google 


TURBINES 


223 


setting  being  changed  from  time  to  time  as  the  general  condition 
of  the  load  changes.  In  such  plants  the  overall  consumption  of 
'water  approximates,  in  a  series  of  steps,  the  load  curve  on  the 
mover. 

The  deflecting  nozzle  may  be  equipped  with  an  automatic 


Fig.    116. — Combination    Needle   and    Deflecting   Nozzle.     (Pelton   Water 

Wheel  Company.) 

regulating  device  so  that  the  governor  in  rejecting  the  load  on  a 
plant  first  operates  the  deflecting  means  and  then  brings  about  a 
gradual  resetting  of  the  needle  and  nozzle  opening. 


Digitized  by 


Google 


224 


HYDRAULIC  EQUIPMENT 


The  ideal  type  of  nozzle,  and  the  one  that  insures  the  most 
sensitive  speed  regulation  and  highest  economy  of  water  consump- 
tion is,  however,  the  "  axuiliary  relief  needle  nozzle,"  Kg.  116. 


I 

I 


I 


This  consists  of  a  main  needle  nozzle  and  a  synchronous  by-pass 
in  the  form  of  an  axuiliary  needle  nozzle  which  discharges  into  the 
tailrace.    Both  nozzles  are  operated  by  the  power  mechanism  of 


Digitized  by 


Google 


TURBINES 


225 


the  speed  governor  simultaneously,  but  in  opposite  directions. 
The  auxiliary  nozzle  opens  when  the  power  nozzle  closes  and  vice 
versa,  the  volumetric  relationship  between  the  two  being  adjusta- 
ble, according  to  the  conditions  at  the  plant.  This,  in  itself,  would 
prevent  any  pressure  rise  in  the  pipe  conduit,  but  would  not 
afford  any  economy  in  water  consumption.  In  order  to  save 
water,  it  is  necessary  to  keep  the  auxiliary  relief  nozzle  closed 
during  a  partial  and  slow  motion  of  the  main  needle  and  also  to 
have  it  close  at  a  safe  rate  of  speed  after  it  has  been  opened.    This 


Fia.   117. — 10,000  Horse-power  Auxiliary  Relief  Type  Nozzle.     (Built  by 
Pelton  Water  Wheel  Company.) 

result  is  accomplished  by  a  cataract  cylinder  or  dashpot  which  is 
inserted  in  the  operating  gear  of  the  auxiliary  rehef  needle. 

Fig.  117  illustrates  a  10,000-H.P.  auxiliary  relief  tjrpe  nozzle 
with  direct-motion  governor.  Alongside  the  wooden  scaffolding 
is  the  governor  oil-pumping  system.  To  provide  pressure  oil  to 
operate  the  governor  piston,  this  pumping  set  is  operated  by  a 
water  wheel  arranged  with  a  control  so  that  the  wheel  and  pump 
operate  only  when  the  level  of  oil  falls  below  a  predetermined 
point. 


Digitized  by 


Google 


226  HYDRAULIC  EQUIPMENT 

Over-speed.  Due  to  the  action  of  the  governor  the  nonnal 
speed  of  the  turbine  is  usually  maintained  constant  under  oper- 
ating conditions.  If  the  load  changes,  however,  take  place  with- 
out a  corresponding  regulation  of  the  admitted  quantity  of  water, 
the  speed  will  necessarily  vary,  increasing  as  the  load  decreases 
and  vice  versa.  If  the  load  should  suddenly  drop  ofiF  with  the 
gates  wide  open  and  remain  so  for  some  reason  or  other,  the  speed 
will  rise  considerably,  sometimes  resulting  in  disaster  to  the 
direct-connected  generators,  and  these  should,  therefore,  always 
be  designed  safely  to  withstand  such  runaway  speeds  of  the 
water  wheels.  These  depend  to  a  great  extent  on  the  hydraulic 
development  and  the  type  of  wheel  used.  For  high-head  plants, 
where  impulse  wheels  are  used,  the  over-speed  should  preferably 
be  estimated  at  100  per  cent  of  the  normal  speed.  For  low  heads 
with  reaction  turbines,  when  the  same  are  working  at  the  most 
eflicient  speed,  and  the  head  is  constant,  the  over-speed  may  be 
from  60  to  80  per  cent  above  the  normal  speed.  Under  low-head 
conditions  with  a  wide  variation  in  the  head  and  with  wheels 
designed  for  an  intermediate  speed  to  work  imder  these  different 
conditions,  a  runaway  speed  of  up  to  200  per  cent  may  then  be 
reahzed  under  the  maximum  head.  The  above  values  are  only 
general,  and  it  is  most  desirable  that  in  all  cases  a  detail  analysis 
is  made,  based  on  test  data  for  the  particular  type  of  wheel  which 
is  to  be  used,  considering  the  extreme  range  of  heads  and  the  other 
conditions  under  which  the  wheel  is  anticipated  to  operate. 

To  prevent  dangerous  over-speeds  several  types  of  overnspeed 
devices  are  being  used.  One  of  these  consists  of  a  fly-ball  mechan- 
ism, independent  of  the  turbine  governor,  driven  from  the  shaft 
of  the  unit  which,  in  the  event  of  excessive  speeds,  by  means  of 
control  valves  admits  water  behind  the  piston  in  an  auxiliary 
cylinder  on  the  governor.  This  causes  it  to  move  in  such  a  man- 
ner as  to  overcome  the  oil  pressure  in  the  control  element  of  the 
governor  and  shut  down  the  unit. 

Mechanical  Designs.  Reaction  type:  There  is  a  very  great 
variety  of  turbine  designs  and  while  overlapping  to  a  certain  degree, 
each  has  its  particular  field  of  apphcation.  For  example,  the  units 
may  be  horizontal  or  vertical,  the  latter  being  now  almost  entirely 
used  for  low  and  medium-head  installations.  As  a  fact,  in  about 
90  per  cent  of  the  large  installations  built  during  the  past  two  or 
three  years  the  turbines  have  been  of  the  vertical-shaft  type.    The 


Digitized  by 


Google 


TURBINES 


227 


units  may  also  have  one  or  more  runners,  as  previously  explained, 
and  when  a  pair  of  runners  are  used  the  question  arises  whether 
an  outward  discharge,  requiring  two  draft  tubes,  or  a  center  dis- 
charge, requiring  only  one  draft  tube,  is  to  be  used. 

Horizontal  Turbines:    The  multi-runner  horizontal  turbine  of 
the  open  flume  type  is  open  to  the  objection  that  the  gate  mechan- 


PiG.  118. — Horizontal  Double-runner  Open  Flume  Turbine  with  Cast-iron 
Draft  Chest  and  Steel  Draft  Tube.     (S.  Morgan-Smith  Company.) 

ism  is  submerged  and  cannot  be  efliciently  lubricated,  while  the 
entire  machine  is  less  accessible  for  inspection  and  repairs.  The 
only  advantage  which  can  be  claimed  for  this  unit  is  its  higher 
speed. 

The  most  approved  type  of  horizontal  imit  at  present  is  either 


Digitized  by  VjOOQ IC 


228  HYDRAULIC  EQUIPMENT 

the  single  or  double  discharge,  both  admitting  of  an  exposed  gate 
mechanism.  The  double  discharge  has  some  advantages  over  the 
single  in  that  it  is  hydraulically  balanced  against  end  thrust.  On 
the  other  hand,  if  it  has  a  central  discharge,  i.e.,  both  runners  dis- 
charging into  a  conmion-draft  tube  (Fig.  118),  the  draft-tube  con- 
ditions are  not  so  favorable,  unless  the  runners  are  spaced  well 
apart. 

Horizontal  turbines  for  very  low  heads  are  necessarily  set  in 
open  flumes  or  wheel  pits.    For  high  heads,  the  volute  or  spiral 


Fig.  119. — Single-runner  Horizontal  Turbine  with  Cast-iron  Spiral  Caae  and 
Single  Discharge  Tube.     (Built  by  I.  P.  Morris  Company.) 


casing  is  the  preferable  type,  the  question  of  central,  double  or 
single  discharge  depending  on  the  conditions  to  be  met  (see  Figa. 
119,  120,  and  121).  For  intermediate  heads,  the  cylindrical  plate 
steel  casing,  Fig.  122,  has  been  conmionly  used  in  the  past.  It  is 
not  as  efficient,  hydraulically,  as  the  spiral  casing,  but  is  some- 
times cheaper  in  first  cost.  In  order  to  avoid  prohibitive  losses,  it 
is  necessary  to  make  the  plate  steel  cylindrical  casing  much  larger 
than  a  volute  casing,  and  the  additional  space  and  material  would 
tend  to  neutraUze  or  reverse  the  reduction  in  cost.    If  the  pen- 


Digitized  by 


Google 


TURBINES 


229 


stock  connection  is  at  the  top  or  the  side,  the  gate  mechanism  may 
be  exposed,  which  is  not  the  case  if  the  penstock  is  connected  at 
the  end.  In  the  latter  case,  however,  the  hydraulic  conditions 
are  better.  In  general,  the  plate  steel  cyUndrical  type  of  casing  is 
more  or  less  out  of  date.  Fig.  122  shows  a  unit  of  the  end  intake 
t)rpe  which  has  given  high  efficiency  in  tests  made  on  the  com- 
pleted installation. 


Pig.  120. — ^Double-runner  Horizontal  Turbines  with  Cast-iron  Spiral  Case  and 
Double  Discharge  Tube.  Hydraulic  Power  Company,  Niagara  Falls. 
(Built  by  I.  P.  Morris  Company.) 

Vertical  Turbines,  Multi-runner  vertical  turbines  are  open  to 
the  same  objections  as  horizontal  units  in  that  the  gate  mechanism 
is  submerged  and  the  machine  more  complicated.  The  best  prac- 
tice of  to-day,  therefore,  adheres  to  the  single  vertical  turbine. 
The  casing  is  of  volute  or  spiral  form  and  for  low  heads  is  usually 
molded  in  the  concrete  foimdations  of  the  power-house  (Fig.  123). 
For  higher  heads  it  is  made  of  cast-iron,  cast-steel  or  riveted-steel 
plate,  as  conditions  may  require.  Sometimes  the  metal  casing  is 
imbedded  in  concrete  under  the  floor  which  supports  the  generator 


Digitized  by 


Google 


230 


HYDRAULIC  EQUIPMENT 


Fig.  121. — Double-Runner  22,500  Horse-power  Horizontal  Turbine  with  Cast- 
iron  Spiral  Cases  and  One  Common  Discharge  Tube.  Long  Lake  Station 
of  the  Washington  Power  Company.  (Built  by  the  I.  P.  Morris  Com- 
pany.) 


Fig.  122. — Double-runner  Horizontal  Turbine.     Cylindrical  Case  with  End 
Intake  and  Central  Discharge.     (1.  P.  Morris  Company.) 


Digitized  by 


Google 


TURBINES 


231 


(Figs.  124  and  125).  The  thrust  bearing  is  occasionally  located 
between  the  generator  and  the  turbine,  and  supported  by  the 
latter,  but  it  is  usually  and  preferably  placed  on  top  of  the 
generator,  and  supported  by  a  spider  mounted  on  the  generator 


Fig.  123. — Single-runner  Vertical  Turbine  with  Volute  Casing  and  Draft 
Tube  Molded  in  the  Concrete  Substructure. 


frame.  The  gate  mechanism  is  of  the  exposed  type,  no  parts 
being  in  the  water  except  the  gates  themselves,  and  all  bearings 
and  pin  connections  are  accessible  for  lubrication. 

Runners:  The  nmners  are  mostly  made  in  one  piece  (Fig.  126), 
except  for  very  large  sizes  where  it  becomes  preferable  to  make 
them  in  sections  on  account  of  shipping  Uhiitations  and  so  as  to 
assure  sound  castings.  While  bronze  was  used  previously  to  a 
very  great  extent  so  as  to  prevent  corrosion,  experience  has  proved 
that  this  effect  is  primarily  due  to  defective  designs.  For  this 
reason  cast  iron  is  now  used  to  a  much  greater  extent  for  runners 


Digitized  by  CjOOQ IC 


232 


HYDRAULIC  EQUIPMENT 


Fia.  124.— Vertical  Turbine  with  Imbedded  Circular  Plate  Steel  Spiral  Caaing. 
(Allis-Chalmers  Company.) 


Digitized  by  CjOOQ IC 


TURBINES 


233 


Fig.  126# — Single-runner  Vertical  Turbine  with  Cast-iron  Spiral  Casing  and 
Steel-lined  Concrete  Draft  Tube.     (I.  P.  Morris  Company.) 

than  formerly,  especially  for  low  and  medium  heads,  while  cast 
steel  is  not  considered  a  very  desirable  metal  from  considerations 
of  corrosion,  on  account  of  the  unavoidable  roughness  of  the  sur- 
face. 

Damage  to  turbine  nmners  may  be  caused  by  both  corrosion 
and  erosion,  the  two  being  of  an  entirely  different  nature.  Mr. 
H.  B.  Taylor  thus  explains  their  difference  as  follows:  "  Erosion  is 
entirely  a  mechanical  action,  while  corrosion  or  pitting,  is  the 
result  of  chemical  action.  The  abrasive  action  of  foreign  sub- 
stances in  the  water  has  the  effect  of  first  polishing  the  vane  sur- 
faces, and  eventually  cutting  away  the  metal  until  the  vanes 
are  worn  entirely  through.  The  eroded  parts  are,  therefore, 
smooth  and  can  be  readily  distinguished  from  the  pitted  marks 
which  result  from  corrosion. 

''It  has  been  demonstrated  that  corrosion  is  primarily  a  ques- 
tion of  design  and  it  has  been  clearly  shown  in  practice  where 
sharp  curves  are  resorted  to,  where  contraction  is  not  sufficient, 


Digitized  by 


Google 


234  HYDRAULIC  EQUIPMENT 

or  where  there  are  pockets  formed  in  the  surface  of  the  vanes,  pit- 
ting or  corrosion  inevitably  develops.  It  has  also  been  demon- 
strated that  where  air  in  large  quantities  is  entrained  in  the  water 
carried  to  the  turbine  corrosion  seems  to  take  place  very  rapidly 
if  the  design  is  not  correct. 

"A   corroded  vane  surface  has  an  appearance  resembling  a 
sponge,  the  surface  being  extremely  irregular  and  the  pitted  spots 


Fig.  126. — Runner  for  Reaction  Turbine.     (Built  by  I.  P.  Morris  Company.) 

often  opening  holes  entirely  through  the  vane.  Chemical  analysis 
of  the  corroded  surfaces  has  brought  out  the  fact  that  the  metal 
has  been  oxidized.  In  runners  made  of  bronze  or  an  alloy,  mod- 
ifications in  the  composition  have  been  detected  in  the  corroded 
portions. 

"The  theory  of  corrosion  as  now  generally  accepted  is  that  the 
water  in  passing  over  any  pocket  or  depressed  surface,  or  in  failing 
to  adhere  to  the  surface  of  the  vane,  leaves  spaces  which  are  filled 


Digitized  by 


Google 


TURBINES 


235 


with  eddies  possessing  high  velocities  and  very  low  static  pressure, 
in  which  oxygen  is  liberated  from  the  water.  This  oxygen  is 
beheved  to  be  in  the  nascent  state  and  rapidly  attacks  the  sur- 
face of  the  metal,  forming  an  oxide  doating,  the  greater  part  of 
which  is  rapidly  washed  away  by  the  water.  When  once  the 
depth  of  this  pocket  is  increased  by  corrosion,  it  is  natural  that, 
due  to  the  greater  area  exposed,  the  pitting  action  should  con- 
tinue at  an  accelerated  rate  imtil  the  vane  is  entirely  eaten 
through." 

Gate  Mechanism:  For  controUing  the  flow  of  reaction  tur- 
bines there  are  two  principal  types  of  gates  in  use,  the  cylinder 
gate  and  the  wicket  or  swivel  gate.    The  latter,  Figs.  127  and  128, 


Fig.  127. — T)rpical  Arrangement  of  Gate  Mechanism  for  a  Small  Vertical 

Reaction  Turbine. 

offers  decided  advantages  of  the  two.  Wear  and  tear  is  greatly 
reduced  for  small  fractional  loads  due  to  better  flow  conditions, 
resulting  in  higher  eflSciencies  than  can  be  obtained  with  cyhnder 


As  previously  stated,  the  exposed  or  so-called  "  outside " 


Digitized  by 


Google 


236 


HYDRAULIC  EQUIPMENT 


type  of  gate  mechanism,  is  much  superior  to  the  older  t3rpe8  in 
which  the  moving  parts  are  more  or  less  submerged.  The  exposed 
t3rpe,  as  apphed  to  spiral  casing  turbines  for  high  heads,  is  the 
ideal  arrangement  in  that  all  bearings  may  be  lubricated  and  the 
gate-stem  packings  may  be  arranged  to  exclude  water  and  grit. 
The  exposed  mechanism  has  a  further  advantage  in  that  it  per- 
mits a  more  direct  connection  between  the  operating  ring,  to  which 
the  gate-stem  levers  are  connected,  and  the  regulating  cylinders 


Fig.  128. — ^Vertical  Reaction  Turbine,  Showing  the  Gate-operating  Mechanism 
and  Speed  Ring.     (Built  by  I.  P.  Morris  Company.) 

or  "  servo-motors  "  of  the  governor  system.  When  the  operating 
ring  is  outside  the  wheel  casing,  it  may  frequently  be  directly 
attached  to  the  connecting  rod  of  the  regulating  cylinder.  Large 
imits  should  have  two  regulating  cylinders  connected  to  the  ring 
at  diametrically  opposite  points,  so  as  to  insure  a  balanced  con- 
dition. 

The  wicket  gates,  or  movable  guide  vanes,  are  mostly  made  of 
cast  steel.  They  are  subjected  to  rough  usage  on  account  of  ice, 
stones  and  rubbish  in  the  water,  and  cast  iron  is  too  brittle  for 
such  service.    In  very  large  units,  the  gate  stems  or  fulcnims 


Digitized  by 


Google 


TUBBINES  237 

should  be  detachable  from  the  gates.  The  stem  may  then  be 
withdrawn  from  the  gate  and  the  latter  removed  without  dis- 
turbing the  crown  plate  of  the  turbine.  This  is  a  great  convenience 
but,  unfortunately,  is  feasible  only  in  connection  with  large  units, 
and  on  smaller  work  the  stems  must  either  be  cast  or  forged  inte- 
gral with  the  gates.  The  gate  stems  must,  furthermore,  be  of 
ample  strength  to  resist  the  strain  in  case  an  obstruction  is  caught 
between  two  gates  and  the  full  power  of  the  governor  is  con- 
centrated upon  them.  The  links  which  connect  the  gate  stem 
levers  to  the  operating  ring  should  be  the  weakest  element  of  the 
gate  mechanism,  and  should  be  designed  to  break  before  the 
stress  reaches  the  elastic  limit  of  the  material. of  any  of  the 
other  parts. 

Speed  Rings.  These  were  introduced  in  connection  with  the 
large  single-runner  vertical  turbine  with  volute  casings  molded 
directly  in  the  concrete.  They  consist  of  a  series  of  curved  vanes 
outside  of  the  turbine  guide  vanes,  forming  together  with  an 
upper  and  lower  crown  (Figs.  113  and  128),  a  rigid  frame  to  sup- 
port the  weight  of  the  portions  of  the  turbine  and  of  the  concrete 
substructure  of  the  power-house  located  above  the  casing,  as  well 
as  the  generator  and  thrust  bearing.  The  vanes  are  shaped  to 
suit  the  free  passage  of  water  entering  the  movable  guide  vanes, 
and  this  arrangement  is  preferable  in  every  way  to  round  stay 
bolts,  the  large,  projected  area  and  circular  form  of  which  causes 
considerable  hydrauUc  losses.  Besides  this,  there  is  a  mechanical 
advantage  in  the  use  of  a  rigid  cast-iron  connection  between  the 
upper  and  lower  speed-ring  crowns. 

Casings.  The  most  efficient  form  of  turbine  casing  in  use  at 
present  is  that  of  volute  or  spiral  shape,  Fig.  119.  This  t3rpe  has 
been  in  common  use  under  high  heads  for  some  years,  and  is  now 
being  adopted  with  increasing  frequency  for  low  heads,  partic- 
ularly where  the  turbines  are  of  large  capacity.  The  materials 
most  conunonly  used  for  medium  and  high  heads  are  cast  iron 
and  cast  steel,  the  choice  between  them  being  influenced  chiefly 
by  consideration  of  the  stresses  imposed.  Large  casings  for  high 
heads  are  usually  made  of  cast  steel.  Cast  iron,  although  more 
suitable  for  medium  heads,  may  properly  be  used  for  high  heads 
if  the  casings  are  small  and  the  material  is  worked  at  low  stress  to 
provide  an  ample  factor  of  safety  against  pressure  surges  which  are 
of  more  common  occurrence  in  high-head  than  in  low-head  plants. 


Digitized  by 


Google 


238 


HYDRAULIC  EQUIPMENT 


As  compared  to  plate  steel,  cast-iron  casings  have  certain  ad- 
vantages, such  as  the  lack  of  rigidity  of  the  plate  steel,  its  danger  of 
local  weaknesses  at  the  riveted  joints,  possibility  of  corrosion  and 
leakage  developing  undetected,  especially  corrosion  on  the  outside 
surface.  Cast  casings  have,  furthermore,  the  advantage  that 
they  may  be  tested  in  the  shops  to  a  hydrostatic  pressure  well  in 
excess  of  that  which  they  can  ever  be  subjected  to  after  installa- 


Fig.  129.— Wooden  Forms  for  Concrete  Turbine  Casings. 

tion.    On  account  of  their  strength  and  rigidity,  they  can  also 
serve  as  an  excellent  bed  plate  for  the  entire  unit. 

For  low  heads,  and  especially  with  large  turbines,  the  casings 
are  usually  molded  in  the  concrete  foundations  of  the  power-house 
by  means  of  wooden  forms  (Fig.  129).  If  the  casings  are  large 
enough  and  the  head  high  enough  to  produce  serious  stresses  in 
the  concrete,  they  may  be  made  of  metal  and  imbedded  in  the 
concrete.  The  principal  controlling  factor  in  this  case  is  the 
relative  cost  of  such  casings  as  compared  with  the  cost  of  adequate 
reinforcing  steel  for  the  concrete,  which  would  be  required  if  the 
metal  lining  were  omitted. 


Digitized  by 


Google 


TURBINES 


239 


Where  the  intake  openings  are  large  it  has  become  general 
practice  to  divide  the  openings  by  means  of  vertical  piers  in  a 
number  of  channels  (Fig.  130).    This  insures  a  more  uniform  dis- 


FiG.  130. — Sectional  Plan  of  Cedar  Rapids  Wheel  Chambers. 

tribution  of  the  water  around  the  runner,  while,  on  the  other  hand, 
it  strengthens  the  casing  by  subdividing  the  span.  It  also  greatly 
facilitates  the  appUcation  of  the  gates,  which  otherwise  would  be 
of  a  size  hardly  possible  to  manipulate. 

Draft  Tubes.  A  correct  draft-tube  design  is  absolutely  es- 
sential in  order  to  obtain  the  maximum  efficiency  of  a  turbine 
as  a  whole.  It  is  an  integral  part  of  the  design  of  the  turbine  and 
should  be  furnished  by  the  turbine  builder.  The  fundamental 
principles  underlying  their  design  and  construction  are  that  the 
water  shall  leave  the  draft  tube  with  as  small  velocity  as  possible 
so  that  the  maximimi  amount  of  kinetic  energy  is  abstracted  from 
the  water.  The  velocity  of  the  water  in  the  tailrace  must,  fur- 
thermore, be  suflicient  to  prevent  it  from  backing  up  and  it  is, 
therefore,  necessary  that  the  water  emerging  from  the  draft  tube 
must  have  a  velocity  at  least  equal  to  that  in  the  tailrace.  In 
order  to  accomplish  this  the  draft  tube  should  be  constructed  on  a 


Digitized  by 


Google 


240 


HYDRAULIC  EQUIPMENT 


long  radius  so  as  to  change  the  direction  of  discharge  from  a  ver- 
tical to  a  horizontal  plan.  The  section  of  the  draft  tube  must  also 
be  gradually  increased  from  the  discharge  ring  of  the  turbine 
to  the  tailrace  so  as  to  gradually  reduce  the  velocity  of  the  water 
from  the  turbine  to  the  tailrace,  and  it  is  conmion  practice  to 
gradually  increase  the  section  from  the  circular  form  at  the  tur- 
bine to  an  oblong  section  at  the  end,  the  long  axis  being  horizontal. 
Good  draft-tube  design  is  fundamentally  dependent  upon  the 
proper  elevation  of  the  turbine  above  tail-water.    The  runner 


Fig.  131.— Placing  of  Wooden  Forms  for  Draft  Tubes  of  Three  10,000  Horse- 
power Turbines. 

should  be  so  located  that  the  total  draft  head  at  the  top  of  the 
tube  (i.e.,  the  static  elevation  of  the  runner  above  tail-water  added 
to  the  velocity  head  at  the  throat  of  the  runner)  is  well  within  the 
theoretical  limits  of  a  vacuum;  namely,  approximately  34  feet, 
depending  on  the  barometer  reading.  If  not,  the  water  coliunn  in 
the  draft  tube  will  break,  retimiing  with  a  surge  and  causing  water- 
hammer.  If,  on  the  other  hand,  the  vacuum  in  the  draft  tube  is 
near  the  breaking  point,  the  continuity  of  the  flow  may  be  inter- 
rupted at  the  discharge  end  of  the  water  passages  through  the 
runner,  resulting  in  corrosion  and  pitting  of  the  vanes. 


Digitized  by 


Google 


TURBINES 


241 


The  residual  velocity  at  the  point  where  the  discharge  is 
released  to  the  atmosphere  is  an  irreclaimable  loss  and  should  be 
made  as  small  as  possible.  The  fact  that  this  loss  is  not  charge- 
able to  the  turbine  should  always  be  taken  into  account  in  making 
eflBciency  tests. 

It  used  to  be  conmion  practice  to  make  all  draft  tubes  of  steel 
plate,  but  of  late  years  they  are  usually  like  the  wheel  casings 
molded  in  the  concrete  foundation  of  the  power-house,  except  in 
the  case  of  small  turbines  (Figs.  131  and  132).  It  is  not  feasible 
to  build  large  draft  tubes  of  plate,  nor  is  it  possible  to  obtain  the 


Fia.  132.— Lower  End  of  a  Molded  Concrete  Draft  Tube. 

smooth  ciu^es  and  eflScient  design  characteristic  of  concrete 
tubes. 

Bearings,  Most  bearings  of  horizontal  turbines  are  of  the 
ordinary  babbitted  generator  type,  except  where  submerged,  in 
which  case  lignum  vitse  bearings  are  used.  Where  water  thrust  is 
to  be  taken  care  of  thrust  bearings  must  also  be  provided. 

For  vertical  miits  the  thrust  bearing  is  almost  always  located 
above  the  generator  on  a  cast-iron  supporting  truss,  which  at  the 
same  time  forms  the  generator  head  cover.  The  upper  guide 
bearing,  which  is  located  immediately  below  the  thrust  bearing,  is 
usually  of  the  oil-lubricated  babbitted  type,  while  the  lower  one  is  a 


Digitized  by 


Google 


242  HYDRAULIC  EQUIPMENT 

water-lubricated  lignum  vitaB  bearing,  permitting  it  to  be  located 
very  close  to  the  runner. 

The  lignum  vitae  is  dovetailed  into  the  bearing  boxes  in  the 
form  of  strips  running  parallel  to  the  axis  of  the  shaft  and  with  the 
end  grain  of  the  wood  placed  normally  to  the  surface  of  the  shaft. 
Twenty  or  more  of  these  strips,  evenly  spaced  in  a  liberal  length 
and  separated  by  spaces  for  circulation  of  cooling  water,  are  so 
proportioned  as  to  present  sufficient  area  to  the  shaft  to  insure 
very  satisfactory  performance. 

In  the  case  of  tiu-bines  operated  in  clear  water,  the  supply  for 
the  bearing  may  be  taken  through  a  pipe  directly  from  the  wheel- 
casing.  A  duplex  strainer  should  be  connected  in  the  line  to 
remove  any  foreign  substances  which  might  otherwise  reach  the 
bearing  and  damage  it.  In  installations  in  which  the  water  ca- 
ries large  quantities  of  foreign  matter  in  suspension,  a  suitable 
central  filtering  system  should  be  provided. 

For  a  description  of  the  various  types  of  thrust  bearings,  see 
page  334. 

Impulse  Type,  like  the  reaction  type,  impulse  tiu-bines  are 
built  in  many  different  designs,  the  controlling  factors  differing  so 
materially  in  each  installation  that  they  not  only  affect  the  general 
type  or  arrangement  of  the  design,  but  also  of  details. 

Horizontal  and  Vertical  Wheels,  Impulse  tm-bines  are  almost 
exclusively  of  the  horizontal  type.  This  not  only  represents  the 
most  economical  design,  but  it  has  many  advantages  of  simplicity 
of  construction  and  arrangement  of  parts  available  for  inspection, 
lubrication,  and  cleaning.  Vertical  wheels  have,  however,  been 
built  and  operate  satisfactorUy,  and  they  may  be  used  for  com- 
paratively low-head  plants,  where  the  water  contains  large  quan- 
tities of  sand  or  grit.  With  this  type  up  to  six  jets  can  be  installed 
in  a  single-wheel  runner. 

Runners.  There  are  two  general  types  of  wheel-runners,  the 
double-lug  bucket  type  and  the  chain  or  triple-lug  bucket  type. 
In  the  former  the  wheel  center  consists  of  a  single  rim  and  the 
buckets  have  two  lugs  which  are  machined  to  a  press  fit  over  the 
rim  of  the  wheel  center  and  held  in  position  by  two  bolts.  In 
the  latter  type,  a  double  or  U-shai:)ed  wheel  rim  is  required  aiid 
the  buckets  have  three  lugs,  a  forward  center  lug  and  two  rear  . 
lugs.  The  forward  center  lug  is  a  close  fit  between  the  two  rims 
forming  the  duplex  wheel  center,  and  the  two  rear  lugs  straddle 


Digitized  by 


Google 


TURBINES 


243 


the  rims,  the  arrangement  of  the  lugs  being  so  designed  that  the 
rear  lugs  of  one  bucket  come  directly  in  line  with  the  forward  lug 
of  the  next  following  bucket.  A  single  bolt,  therefore,  passes 
through  the  rear  lugs  of  one  bucket,  the  rims  and  the  central  or 
forward  lug  of  the  next  following  bucket,  thus  connecting  up  jaXl 
of  the  buckets  into  a  con- 
tinuous chain.  Fig.  133 
shows  such  a  type  of  wheel. 

In  the  chain-type  wheel 
the  base  line  of  the  buckets 
or  the  distance  between  the 
supporting  bolts  is  very 
much  greater  than  it  is  with 
double-lug  buckets.  This 
type  of  construction  is, 
therefore,  particularly  suit- 
able for  all  installations 
where  the  ratio  between  the 
diameter  of  the  jet  and  the 
pitch  diameter  of  the  wheel 
is  small,  that  is,  where  a 
large  diameter  of  jet  is 
applied  to  a  comparatively 
small  diameter  of  wheel. 
This  is  always  the  case 
where  a  very  large  power  output  is  required,  with  a  turning  speed 
comparatively  high,  as  proportional  to  the  head  of  water,  thus 
calling  for  large  buckets  on  a  comparatively  small  wheel.  It  is  also 
especially  suitable  for  extreme  cases  of  large  horse-power  and  high 
heads,  making  the  wheel  runner  of  the  most  stable  construction. 

The  buckets  are  ellipsoidal,  which  causes  the  water  jet  to 
impinge  without  shock  or  disturbanbe,  and  it  is  discharged  along 
natural  lines  over  the  entire  bucket  surface.  The  central  portion 
of  the  front  entering  wedge  or  lip  of  the  bucket  is  cut  away  in  the 
form  of  a  semicircular  notch,  and  this  opening  allows  the  solid 
circular  water  jet  to  discharge  upon  the  central  dividing  wedge 
of  the  bucket  without  being  split  in  a  horizontal  plane,  with  the 
result  that  all  eddy  currents  are  avoided  and  the  full  force  of  the 
jet  is  expended  for  useful  work,  resulting  in  the  maximum  bucket 
eflSciency. 


Fig.  133.— Tangential  Water  Wheel 
Equipped  with  Ellipsoidal  Inter- 
locking Chain  Type  Buckets.  (Built 
by  Pelton  Water  Wheel  Company.) 


Digitized  by 


Google 


244  HYDRAULIC  EQUIPMENT 

Arrangement  of  Runners.  The  two  principal  runner  arrange- 
ments are  the  single-overhung  and  the  double-overhung.  In 
addition  there  is  also  the  self-contained  type.  The  first-named  is 
mounted  on  an  overhung  extension  to  the  generator  shaft  (Figs. 
116  and  134),  no  extra  outboard  bearings  being  provided,  and  the 
second  type  comprises  simply  two  single-overhung  wheels,  one 
being  mounted  on  a  shaft  extension  at  each  end  of  the  generator. 
This  is  the  ideal  construction  for  large  units  and  is  extensively 
used.    With  the  double-overhung  type  it  is  possible  to  make  a 


Fig.  134. — Single  Overhung  Impulse  Turbine,  Governor  Regulated  by  Jet 
Deflector.     (Built  by  Pelton  Water  Wheel  Company.) 

prime  mover  of  double  the  power  output,  maintaining  the  same 
speed  of  rotation  with  the  same  conditions  of  water  pressure. 
For  very  large  units,  two  wheels  on  each  side  of  the  generator  may 
be  used,  making  foiu-  wheels  per  unit.  This  usually  requires  four 
bearings,  the  generator  rotor  being  mounted  between  the  two  main 
bearings,  with  an  outboard  bearing  at  each  end,  two  wheels  being 
located  between  one  main  bearing  and  one  outboard  bearing.  The 
self-contained  type  has  its  own  shaft,  bearing,  base  and  housing, 
one  or  more  runners  being  mounted  on  the  same  shaft  and  in  the 
same  housing.    It  is  mainly  used  for  small  capacity  units. 


Digitized  by 


Google 


TURBINES 


245 


Referring  again  to  Fig.  116,  a  water  connection  for  throwing  a 
fine  spray  of  water  through  the  hollow  shaft  will  be  noted  on  the 
outer  end  of  the  right  bearing.  Within  the  housing  of  a  tangential 
wheel  there  is  a  very  definite  vacuum  due  to  the  action  of  the 
revolving  wheel  as  a  centrifugal  blower  and  the  action  of  the  jet 
of  water  acting  as  an  injector.  Therefore  discharging  a  fine  spray 
of  water  into  the  open  end  of  the  shaft,  this  is  drawn  through  and 
is  most  effective  in  cooling. 

The  illustration  (Fig.  116),  also  shows  what  is  termed  a  "  tail- 
race  ventilator."    This  is  a  labyrinth  passage  from  the  bottom  of 


Fig.  136. — Impulse  Turbine  with  Two  Nozzles  per  Wheel.    Arranged  with 

Auxiliary  Relief. 

the  generator  pit  to  the  water-wheel  pit,  tne  vacuum  existing  in 
the  water-wheel  pit  bringing  about  a  very  definite  circulation  of 
air  which  it  draws  out  of  the  bottom  of  the  generator  pit. 

The  three  principal  types  of  nozzles  used  with  impulse  tur- 
bines were  described  under  "  Speed  Regulation,''  page  221. 
While  one  jet  per  wheel  is  used  in  most  cases,  there  may  be  installa- 
tions where  the  head  of  water  available  is  low,  as  compared  with 
the  quantity  of  water  and  where  it  is  desired  to  maintain  a  com- 
paratively high  speed  of  rotation.  Under  such  conditions  two 
jets  of  water  may  be  applied  to  each  wheel  from  the  same  nozzle 


Digitized  by 


Google 


246  HYDRAULIC  [EQXnPMENT 

body,  the  jets  being  approximately  90°  apart.  Such  an  arrange- 
ment is  shown  in  Fig.  135,  this  sketch  representing  a  unit  with 
four  wheels  and  eight  jets,  developing  10,500  H.P.  under  an 
effective  head  of  380  feet  at  200  R.P.M. 

Housings,  The  general  tjrpe  and  construction  of  wheel 
housings  or  casings  for  impulse  turbines  is  illustrated  in  Fig.  134, 
the  best  practice  being  to  provide  a  separate  housing  for  each 
wheel  to  prevent  interference  from  discharged  water.  The  lower 
part  is  usually  made  of  iron  castings  and  the  upper  housing  or 
cover  of  steel  plate  riveted  into  a  cast-iron  frame.  This  type  of 
housing  for  large  imits  is  claimed  to  be  preferable  to  a  housing 
made  entirely  of  cast  iron,  as  it  is  Ughter  to  handle  and  elimi- 
nates any  danger  of  breakage  where  the  shaft  of  the  nmner  passes 
through  the  sides  of  the  housing,  and  water  leakage  is  prevented 
by  means  of  a  centrifugal  disc  and  water  guard,  which  device  in- 
sures a  frictionless  packing.  For  small  units,  on  the  other  hand, 
the  self-contained  cast-iron  housing  is,  as  previously  stated,  to  be 
preferred. 

2.   GOVERNORS' 

Before  the  advent  of  the  automatic  voltage  regulator,  close 
speed  regulation  by  water-wheel  governors  was  of  much  greater 
importance  than  now.  for  anv  deoarture  from  normal  had  an 
immediate  effect  on  the  voltage.  With  the  automatic  regulator 
in  operation,  reasonable  changes  in  speed  have  no  appreciable 
effect  on  the  voltage,  but  they  do,  of  course,  affect  the  frequency. 
A  sUght  variation  in  frequency,  is  to  be  expected,  for  like  all  gov- 
ernors for  prime  movers,  the  water-wheel  governor  requires  a 
certain  change  in  speed  to  ensure  good  governing. 

Factors  Affecting  Speed  Regulation.  While  primarily  the 
regulation  for  speed  originates  with  the  governor,  it  also  involves 
the  consideration  of  the  pipe-line  conditions  and  those  devices 
required  for  Umiting  the  pressure  rise  therein,  and  besides  the 
effective  flywheel  effect  of  the  rotating  elements  of  the  generator 
and  water  wheel. 

Variations  in  the  velocity  of  the  water  in  the  pipe  line  will 
always  occur,  and  every  retardation  in  velocity  of  the  moving 
water  column  will  bring  about  an  increase  in  the  pressure,  in- 
versely in  proportion  to  the  time  occupied  for  a  given  change.    It 

1  See  also  ''  Pipe  Lines/'  "  Water  Hammer  and  Surge  Tanks." 


Digitized  by 


Google 


GOVERNOBS  247 

is  thus  evident  that  the  quicker  the  governor  movement,  the  greater 
the  pressure  rise  will  be,  while,  if  the  governor  movement  is  made 
slower,  the  speed  increase  will  be  greater,  and  a  proper  balance 
between  the  two  is,  therefore,  the  correct  time  for  adjusting  the 
governor  closing  stroke.  Few  conditions  will,  however,  warrant 
a  stroke  quicker  than  1}  seconds. 

In  addition,  the  flywheel  effect  must  be  considered  the  greater 
the  inertia  of  the  rotating  masses  and  the  higher  their  rotation, 
the  smaller  the  speed  variation  will  be.  A  sufficient  rotating  mass 
to  supply  stored  energy  (WIP)  must,  therefore,  also  be  intro- 
duced to  keep  the  speed  within  permissible  limits. 

To  secure  a  constant  speed  with  a  water  wheel  operated  under 
a  variable  load,  the  energy  produced  by  the  water  wheel  must  be 
varied  proportionally  to  the  load,  and  the  method  of  achieving  this 
in  practice,  except  for  tangential  impulse  wheels  with  deflecting 
nozzles,  consists  essentially  of  varying  the  size  of  the  gate  or  valve 
openings  through  which  the  water  to  the  wheels  is  admitted  (see 
"  Speed  Regulation,"  page  220). 

The  regulation  of  hydro-electric  imits,  as  stated,  requires  a 
certain  departure  from  normal  speed  before  the  governor  can  act. 
Since  the  immediate  effect  of  the  gate  motion  is  opposite  to  that 
intended,  the  speed  will  depart  still  further  from  the  normal,  which, 
in  turn,  tends  to  cause  the  governor  to  move  the  gate  too  far,  with 
the  result  that  the  speed  will  not  only  return  to  normal  as  soon  as 
the  inertia  of  the  water  and  the  rotating  parts  is  overcome,  but 
may  rush  far  beyond  normal  in  the  opposite  direction. 

A  given  gate  opening  will  produce  a  certain  velocity  of  the 
water  in  the  penstock  and  the  energy  of  the  moving  water  will  be 
equal  to  the  weight  of  the  water  in  the  penstock  multiplied  by  the 
square  of  the  velocity  and  dividing  this  product  by  64.4.  For 
example,  with  a  penstock  300  feet  long  and  6  feet  in  diameter,  the 
weight  of  the  water  would  be  530,000  pounds,  and  assuming  a 
velocity  of  5  feet  per  second,  corresponding  to  the  head  and  full 
gate  opening,  the  total  kinetic  energy  of  the  water  would  be 

^^^=206,752  foot-pounds. 

If  the  gates  are  now  instantly  closed  to  about  one-quarter 
gate  opening  so  that  the  velocity  would  be  reduced  to  1.6  feet  per 
second,  the  corresponding  kinetic  energy  would  only  be  18,517 


Digitized  by 


Google 


248  HYDRAULIC  EQUIPMENT 

foot-pounds.  The  loss  of  energy  is,  therefore  equal  to  205,752— 
18,517  foot-pounds,  and  this  amount  will  be  transferred  to  the 
water  issuing. from  the  gate  apertures,  which,  therefore,  wiU  have 
its  velocity  increased  until  the  187,235  foot-pounds  of  energy  has 
been  absorbed.  The  kinetic  energy  of  the  water  column  will, 
therefore,  be  transferred  to  the  water  wheel  at  the  very  moment 
when  it  is  desired  to  reduce  the  energy  produced  by  the  wheel. 
In  the  same  manner,  if  the  load  be  thrown  on  and  the  gate  again 
instantly  opened  full,  the  same  amount  of  energy  which  the  water 
column  gave  out  on  being  retarded  in  the  previous  case  will  be 
absorbed  by  the  water  column  in  accelerating  its  velocity  to  5 
feet  per  second.  The  energy  delivered  to  the  wheel  will,  therefore, 
be  reduced,  causing  its  speed  to  drop  off,  just  when  the  opposite 
is  required,  and  this  action  cannot  be  overcomie  by  rapid  move- 
ment of  the  gate,  but,  on  the  contrary,  is  intensified  by  more 
rapid  gate  movement.  It  is,  therefore,  obvious  that  after  the 
governor  has  been  set  in  motion  by  a  change  of  speed,  some  means, 
other  than  the  return  of  the  speed,  must  be  provided  to  stop  it 
when  it  has  moved  the  gates  the  amount  required  by  the  change 
of  load  which  was  the  cause  of  the  change  in  speed  that  originally 
set  the  governor  in  motion.  The  means  provided  for  this  pur- 
pose is  a  dashpot,  known  as  the  '^  compensating  "  mechanism, 
and  is  an  essential  feature  of  all  quick-acting  water-wheel  govern- 
ors. Compensation  may  thus  be  considered  the  act  of  stopping 
and  waiting  for  the  result  of  the  gate  movement. 

It  is  a  comparatively  easy  matter  to  calculate  the  speed- 
regulation  in  cases  where  the  inertia  of  the  moving  water  column 
is  a  negUgible  quantity,  such  as  with  open  flumes  and  short  draft- 
tubes.    For  such  conditions,  the  following  formula  apphes: 

d = 81,000,000  yff^yjpj 

where         d  =  percentage  temporary  change    in  speed  for  load 
thrown  off; 
H.P.=  maximum  horse-power  capacity  of  the  turbine; 
r=time  in  seconds  occupied  by  the  governor  in  moving 
the  turbine  gates  through  their  range; 
W IP = weight  of  rotating  parts  multiplied  by  square  of 
radius  of  gyration  of  generator; 
iV= normal  speed  of  rotating  parts  in  R.P.M. 


Digitized  by 


Google 


GOVERNORS  249 

For  installations  with  long  penstocks  the  regulation  becomes 
much  more  serious  and  is  difficult  to  calculate  acciu-ately  due  to 
the  many  variable  factors  involved,  such  as  the  length  of  the  pipe 
line,  the  effective  head  and  velocity  of  flow,  time  of  governor 
action,  flywheel  effect  and  effect  of  standpipes,  etc. 

The  final  speed  after  a  load  change  will  be  that  due  to  the 
initial  kinetic  energy  of  the  rotating  parts  and  the  excess  or 
deficiency  above  or  below  the  load  requirements  during  the  time 
of  gate  adjustment.  This  excess  or  deficient  energy  is  due  to  the 
excess  or  deficiency  in  the  quantity  of  water  during  the  change  in 
addition  to  that  of  the  energy  required  to  accelerate  or  retard  the 
water  colimrn  in  the  penstock. 

The  effect  of  a  standpipe  must  also  be  considered  in  absorbing 
the  excess  power.  When  such  a  structure  of  sufficient  size  is 
installed  close  to  the  wheels,  the  conditions  will  approach  those  of 
an  open  flume,  while,  if  located  some  distance  from  the  plant, 
they  become  similar  to  those  of  a  closed  penstock  of  a  Jength  equal 
to  the  distance  from  the  draft-tube  to  the  standpipe. 

Action  of  Governor.  The  obvious  tendency  of  a  governor,  as 
explained  above,  is  to  permit  the  speed  to  oscillate  above  and  below 
normal.  A  successful  governor  must,  therefore,  anticipate  the 
effect  of  any  gate  movement,  and  in  order  to  overcome  the  effect 
of  the  pressure  change  in  the  penstock  the  governor  must  move 
the  gate  sUghtly  beyond  the  final  position,  in  order  to  restore  the 
speed  to  normal;  the  final  motion  of  the  gate  being  a  slow  move- 
ment back  to  the  final  position.  This  last  slow  movement  is  con- 
troUed  by  the  compensation  device  as  explained.  The  percentage 
variation  in  speed  which  will  occur  before  the  governor  begins  to 
move  the  gate,  or  the  limits  within  which  the  governor  is  inopera- 
tive, shoidd  be  a  minimum.  This  is  generally  realized  with 
hydraulic  governors  where  it  varies  from  practically  nothing  to 
0.75  per  cent. 

Tlie  speed  with  which  the  governor  moves  the  gate  is  a  most 
essential  element  of  a  good  governor.  No  general  rule  can  be 
given  of  the  rate  at  which  the  governors  should  open  or  close  the 
gates.  It  can  be  more  rapid  the  shorter  the  penstock  and  the 
lower  the  velocity  of  the  water.  The  effect  of  both  rapid  opening 
and  closing  of  the  gates  should  be  investigated  in  every  projected 
plant,  in  order  to  guard  against  drawing  down  the  pressure  at 
critical  points  in  the  penstock  below  that  of  the  atmosphere,  and 


Digitized  by 


Google 


250  HYDRAULIC  EQUIPMENT 

thereby  causing  danger  of  collapse,  or  permitting  increases  of 
pressure  beyond  the  strength  of  the  penstock. 

The  duration  of  the  momentary  variation  between  the  first 
departure  of  the  speed  from  normal  and  its  complete  restoration  to 
steady  speed  should  also  be  a  minimum.  It  is  governed  by  the 
energy  contained  in  the  water  column  as  well  as  the  flywheel 
effect.  By  increasing  the  flywheel  capacity  the  speed  variation 
can  be  reduced,  and  thus  a  plant  with  a  moderate  length  of  pen- 
stock and  a  small  flywheel  capacity  will  give  a  large  momentary 
variation  over  a  short  interval  of  time,  while  the  same  plant  with 
a  larger  flywheel  capacity  will  give  a  comparatively  small  varia- 
tion over  a  longer  time  interval.  The  latter  is,  however,  more 
favorable  from  an  operating  standpoint. 

The  speed  regulation  of  a  number  of  10,000-H.P.  recently 
installed  reaction  turbines,  operating  under  a  96-foot  head,  at 
185  R.P.M.,  is  given  in  the  foUowing.  It  is  based  on  a  total  fly- 
wheel effect  {WR^)  of  1,700,000,  a  pipe  Une  diameter  of  11  feet, 
and  a  pipe  length  of  190  feet. 

Load  change,  per  cent 10         25  50  100 

Speed  change,  per  cent 0.9        2.1        3.5        16 

These  governors  were  furthermore  guaranteed  to  restore  the 
speed  of^the  units  to  within  0.5  per  cent  of  normal  from  any  change 
in  load,  and  will  begin  to  act  before  the  speed  has  changed  more 
than  0.5  per  cent  from  normal. 

An  imnecessarily  close  regulation  should  not  be  required  when 
considering  such  extreme  conditions  as  full  load  thrown  suddenly 
on  or  off  a  unit,  conditions  which  sddom  occur  in  a  plant,  and 
when  they  do  occur  it  is  usually  due  to  a  short  circuit  or  a  dropping 
of  load  under  circumstances  in  which  regulation  ceases  to  be  a 
consideration. 

Pipe-line  Pressure  Caused  by  Governor  Action.  In  order  to 
arrive  at  the  maxtmimi  pressing  developed  in  the  pipe  lines  by  the 
governor  action,  the  following  formula  may  be  used  with  sufficient 

^^^y^  ^_0.027XLXt; 

where  P=maximirai  pressure  change  in  pounds  p)er  square  inch; 
L  =  length  of  water  column  in  feet; 
t;  =  velocity  in  feet  per  second; 
r=time  in  which  water  column  is  stopped  in  seconds. 


Digitized  by 


Google 


GOVERNORS  251 

Relief  valves  at  the  turbine  case  are  sometimes  employed  to 
obviate  the  difficulties  of  long  feed  pipes,  but  it  is  evident  that 
they  can  be  of  use  only  upon  the  load  going  off.  They  can  be  of 
no  use  upon  the  load  going  on,  for  they  cannot  supply  to  the 
moving  water  column  kinetic  energy  which  it  has  lost  and  which 
it  must  regain  before  it  can  flow  at  the  higher  velocity  required 
by  an  increase  of  load.  Standpipes  are  better;  in  fact,  they  are 
often  imperative.  If  of  improper  design,  or  of  insufficient  capacity, 
they  frequently  add  to  the  difficulty  of  obtaining  regulation.  If 
of  proper  design,  they  simply  result  in  shortening  the  closed  water 
column;  that  is,  they  bring  the  tin-bine  nearer  to  being  set  under 
open-water  conditions  which  are  the  most  favorable  conditions. 
Unfortunately,  the  conformation  of  the  country  is  often  such  that 
a  standpipe  is  unfeasible,  and  reliance  must  be  placed  on  reUef 
valves  to  prevent  dangerous  water  pressures  being  developed  and 
upon  flywheels  to  Uberate  or  absorb  kinetic  energy  as  the  closed 
water  colunm  absorbs  or  liberates  it. 

Energy  Output  of  Governor.  In  order  to  be  of  ample  capacity 
to  control  the  gates  promptly  and  still  have  a  margin  for  speed 
regulation  of  the  wheels,  it  is  necessary  that  the  governor  should  be 
capable  of  developing  an  effort  in  excess  of  the  maximimi  effort 
required  to  merely  operate  the  gates  themselves.  Practical  expe- 
rience seems  to  indicate  that  this  margin  should  be  about  100  per 
cent  of  the  maximum  effort  required  to  move  the  gates. 

Governors  are  nominally  rated  in  foot-pounds  at  a  given  pres- 
sure, the  rated  effort  being  equal  to  the  nominal  rating  divided  by 
the  length  of  the  stroke,  expressed  in  feet.  It  has,  however,  been 
suggested  to  rate  a  governor  by  its  maximimi  torque  produced  or 
also  by  its  energy  produced  per  second.  This  latter  term  would  be 
an  indication  of  both  the  power  suppUed  for  and  the  rate  of  the 
gate  motion  to  be  produced  by  the  governor. 

Arrangement  and  Operation.  The  movement  of  turbine  gates 
requires  a  relatively  large  amount  of  energy  and  indirect-acting 
governors  are  therefore  almost  exclusively  used,  employing  either 
mechanical  energy  as  with  the  so-called  mechanical  governor  or  a 
compressed  fluid  as  with  the  hydraulic  governor. 

Mechanical  governors  obtain  their  energy  mechanically  by 
belt  drive  from  the  prime  mover  and  transmit  it  by  friction  coup- 
lings, etc.,  to  the  gate  shaft.  They  are  not  very  sensitive  but 
exposed  to  considerable  wear,  for  which  reason  they  are  only 


Digitized  by 


Google 


252  HYDRAULIC  EQUIPMENT 

used  for  very  small  units.    In  fact,  they  are  being  rapidly  dis- 
carded. 

A  hydraulic  pressure  governor  system  can  be  divided  in  two 
distinct  parts — ^the  pumping  outfit  and  the  governor  imit  proper. 

The  pumping  outfit  in  its  simplest  form  consists  of  a  power 
pimip,  a  pressure  tank,  a  receiving  tank  and  suitable  connecting 
pipes,  valves,  gauges,  etc.  The  fluid  which  is  used  to  operate 
the  power  cylinder  of  the  governor  is  obtained  from  the  pressing 
tank,  which  normally  should  be  about  half  filled  and  of  sufficient 
capacity  to  provide  for  a  series  of  governor  strokes,  even  though 
the  pump  be  temporarily  inoperative.  The  receiving  tank  receives 
the  fluid  after  it  has  performed  its  work  in  the  governor,  the  func- 
tion of  the  pump  being  to  draw  the  fluid  from  the  receiving  tank 
and  force  it  into  the  pressure  tank  together  with  a  sufficient  amount 
of  air  to  obtain  a  pressure  of  from  100  to  200  pounds  per  square 
inch.  This  compressed  air  is  the  immediate  source  of  energy  for 
operating  the  governor,  and  although  the  pump  accumulates  or 
renews  this  energy  at  a  comparatively  slow  rate,  it  is  available 
for  use  in  the  governor  as  rapidly  as  the  requirements  of  regulation 
demand.  It  is  this  principle  which  makes  possible  the  rapid 
movement  of  the  gates,  which  is  essential  to  close  speed  regu- 
lation. 

Two  general  sjrstems  of  pressure  supply  are  in  successful  use, 
one  utilizing  oil  and  the  other  water.  Water  is  advantageous  in 
the  case  of  large  plants.  High-epeed,  multi-stage  centrifugal 
pumps  of  relatively  small  size  may  be  used,  while  for  oil,  plunger 
or  gear  pumps  are  required.  The  cost  of  oil  necessary  for  the 
pressure  sjrstem  of  a  large  plant  is  also  an  important  item,  but,  on 
the  other  hand,  the  wear  on  valves  and  valve  sleeves,  etc.,  is  un- 
questionably less  with  oil  than  with  water.  Each  of  these  two 
systems  has  its  advantages  and  disadvantages  which,  should  be 
carefully  considered  in  each  installation. 

Water  treated  with  a  soluble  oil  may  also  be  used  as  a  gov- 
ernor fluid.  A  small  percentage  of  soluble  oil  will  supply  the 
required  lubricating  qualities,  and  will  prevent  rusting  or  cor- 
rosion. The  use  of  this  fluid,  handled  by  centrifugal  pimips,  is 
probably  the  best  practice  in  the  case  of  large  stations. 

Many  large  plants  are  now  equipped  with  central  pressure 
systems.  The  pumps  are  sometimes  motor-driven  with  auto- 
matic pressure  control.    Sometimes  the  motors  are  allowed  to 


Digitized  by 


Google 


GOVERNORS  253 

run  continuously  and  the  pumps  are  equipped  with  unloading 
valves.  Each  unit  has  its  own  accumulator  or  pressure  tank 
situated  close  to  the  goveror  to  eliminate  the  effect  of  inertia  in 
the  supply  pipe,  and  unless  the  discharge  piping  is  of  hberal  size, 
each  unit  should  have  a  local  simip  tank  from  which  the  oil  or 
water  returns  by  gravity  to  the  central  reservoir  supplying  the 
pumps.  This  latter  method  complicates  the  piping,  and  it  is 
better  to  use  a  large  return  pipe  and  only  one  sump  tank.  Both 
oil  and  water  systems  are  now  generally  of  the  open  type;  that  is, 
they  are  arranged  to  discharge  under  atmospheric  pressure.  The 
closed  or  vacuum  system,  at  one  time  conmaonly  used,  has  been 
discarded,  even  with  individual  pumping  systems,  because  of  the 
tendency  to  break  down  the  oil. 

The  principal  elements  of  the  governing  unit  proper  are: 
One  or  two  power  cylinders  or  servo-motors,  suitable  mechanism 
for  transmitting  the  movement  of  the  power  piston  to  the  gate 
shaft,  a  main  or  relay  valve,  a  pilot  or  regulating  valve,  a  safety 
stop,  a  centrifugal  speed  governor  and  a  compensating  device. 
In  addition  they  are  also  usually  arranged  so  as  to  permit  of  hand 
control  as  well.as  remote  control  and  when  required  a  load  limiting 
device  may  be  provided. 

The  admission  of  the  fluid  from  the  pressure  tank  to  the  power 
cylinder  and  back  to  the  receiver  tank  is  regulated  by  the  main  or 
relay  valve.  This  must,  in  most  cases,  be  of  such  a  large  size  as 
to  make  it  impossible  to  control  directly  from  the  centrifugal  speed 
elements,  and  for  this  reason  an  intermediate  pilot  or  regulating 
valve  is  provided.  This  is  connected  to  the  centrifugal  mechanism 
and  r^ulates  the  admission  of  the  pressure  fluid  to  the  main 
relay  valve,  and  this,  in  turn,  to  either  end  of  the  piston  in  the 
power  cylinder,  which  transmits  its  motive  power  to  the  gates  or 
nozzle  mechanism  of  the  turbine  when  a  speed  variation  occurs. 
The  movement  of  the  relay  valve  is  always  such  as  to  return  the 
pilot  valve  to  its  neutral  position  after  a  load  variation  has  oc- 
curred, resulting  in  a  movement  of  the  governor  piston. 

The  rapid  growth  in  size  of  units  has  brought  about  a  corre- 
sponding change  not  only  in  the  size  of  governors,  but  also  in  the 
arrangement.  Standard  governors  were  formerly  self-contained; 
that  is,  the  control  and  power  elements  were  combined  in  the  gov- 
ernor itself.  It  was  necessary  only  to  connect  the  centrifugal 
element  to  the  turbine  shaft  and  the  power  element  to  the  tur- 


Digitized  by 


Google 


254  HYDRAULIC  EQUIPMENT 

bine  gate  mechanism,  and  the  installation  was  complete  excepting 
the  pumping  system.  While  this  arrangement  is  still  in  use  with 
small  units,  it  is  no  longer  used  for  large  imits.  In  the  latter  case 
the  centrifugal  control  mechanism  and  regulating  valves  are  now 
combined  and  localized  in  an  ''  actuator  "  placed  in  any  con- 
venient position  near  the  imit,  and  the  power  element  or  servo* 
motor  is  incorporated  in  the  design  of  the  turbine.  By  separating 
these  elements,  each  of  them  may  be  located  in  the  most  advan- 
tageous position  with  respect  to  the  individual  function  it  has  to 
perform.  For  example,  in  the  case  of  vertical  units,  the  actuators 
may  be  placed  on  the  generator  floor  and  the  servo-motors  in  the 
wheel  pit,  directly  connected  to  the  gate  mechanism. 

Methods  of  ControL  Governors  up  to  about  60,000  foot- 
pounds capacity  are  often  equipped  with  mechanical  hand  con- 
trol independent  of  the  servo-motor.  This  is,  however,  scarcely 
feasible  with  larger  governors  on  accoimt  of  the  time  that  would  be 
required  to  develop  so  much  power  by  hand.  They  are,  there- 
fore, equipped  with  hand  control  of  the  operating  pressure  only. 
This  control  is  independent  of  the  centrifugal  speed  element, 
and  is  of  great  value  for  adjusting  the  load  on  thjB  unit  and  for 
synchronizing  purposes.  In  addition  to  local  hand  control  all 
governors  are  now  as  a  rule  also  equipped  with  manual  remote 
control.  The  mechanism  is  equipped  with  a  small  reversible 
motor  electrically  connected  with  a  double-throw  control  switch  on 
the  switchboard,  and  enables  the  operator  to  control  the  load  and 
speed  from  the  switchboard. 

Numerous  plants  can  be  found  where  the  units  must  first  be 
paralleled  by  hand  and  the  governor  "  cut  in  "  after  the  generators 
are  tied  into  the  power  sjrstem.  The  reason  may  be  found  in  the 
lack  of  harmony  of  flywheel  effect,  the  velocity  of  the  water  and 
the  length  of  the  pipe  line.  If  one  of  the  three  could  be  properly 
altered,  the  trouble  could  possibly  be  eliminated. 

Sometimes  it  is  required  to  carry  a  fixed  load  irrespective  of 
the  load  or  speed  variation  of  the  system  and  such  fixed  loads  may 
be  less  than  that  developed  at  full  gate  opening.  This  reqiiire- 
ment  necessitates  the  use  of  a  load-limiting  device,  which  prevents 
the  distributor  of  the  regulating  valve  from  attaining  a  position 
beyond  the  amount  desired.  A  load-limiting  device  also  allows  of 
an  adjustment  according  to  the  head  or  quantity  of  water  available 
at  various  times,  and  it  should  preferably  be  remote  control. 


Digitized  by 


Google 


GOVERNORS 


255 


Typical  Designs.  A  description  of  the  many  governors  on 
the  market  and  their  operation  can  be  readily  obtained  from  the 
numerous  trade  catalogs  and  will  not  be  gone  into  here.    Fig.  136 


Fio.  136. — Pelton  Oil  Pressure  Governor  of  Moderate  Capacity  with  Self- 
contained  Rotary  Pumping  Set  Located  within  Governor  Base.  (Built 
by  Pelton  Water  Wheel  Company.) 

shows  a  Pelton  oil  pressure  governor  with  self-contained  pumping 
set.  Fig.  137  illustrates  a  Lombard  governor  of  very  modem 
design,  which  is  built  in  sizes  from  3000  to  30,000  foot-pounds. 


Digitized  by 


Google 


256 


HYDRAULIC  EQUIPMENT 


It  is  very  homologous  in  design  and  is  intended  for  direct  con- 
nection to  either  horizontal  or  vertical  wicket-gate  turbines.  The 
illustration  shows  the  governor  arranged  for  connection  to  a 
horizontal  gate  shaft,  and  when  it  is  to  be  connected  to  a  vertical 
shaft  it  is  provided  with  a  Lnshaped  sub-base  which  constitutes  a 
steady  and  supporting  bearing  for  the  upper  end  of  the  gate  shaft. 


Fig.  137.— Lombard  Type  T  Governor. 

This  governor  is  also  provided  with  a  hydraulic  hand  control 
which  may  be  instantly  thrown  in  and  out  of  action.  A  hand 
pump,  mounted  on  the  governor  frame,  is  also  provided  for  moving 
the  turbine  gates  when,  for  any  reason,  the  pressure  is  let  down  in 
the  pressing  tank. 

Fig.  138  is  the  type  of  actuator  governor  furnished  by  the 
Lombard  Governor  Company  for  the  Mississippi  River  Power 


Digitized  by 


Google 


GOVERNORS 


257 


Company,  the  servo-motors  or  power  cylinders  developing  250,000 
foot-pounds  being  located  on  the  floor  below.  The  distinguishing 
feature  of  this  design  is  that  all  adjustable  parts  are  enclosed 
within  the  cast-iron  frame,  thus  preventing  their  being  tampered 
with  by  unauthorized  persons.  The  back  consists  of  plate-glass 
doors  through  which  all  working  parts  may  be  readily  inspected. 
The  face  of  the  actuator  carries  the  various  dials  which  indicate 


Fig.  138. — Lombard  Governor  Actuator.     Mississippi  River 
Power  Company. 

speed  of  turbines,  gate  opening  of  turbine  gates,  pressure  on  the 
accumulator  system,  and  back  pressure,  if  any.  There  are  also 
provided  hand  wheels  controlling  the  main  throttle  of  the  gov- 
ernor system,  throttle  for  the  relay  valve  system  and  a  hydraulic 
hand  control.  The  central  hand  wheel  controls  an  adjustable 
gate-setting  device  by  means  of  which  the  maximum  amount 
of  turbine  gate  opening  may  be  regulated  at  the  will  of  the 
operator. 


Digitized  by 


Google 


258 


HYDRAULIC  EQUIPMENT 


Fig.  139. — Large  Capacity  Hydraulic  Governor. 

Company.) 


(Built  by  I.  P.  Morris 


Fig.  139  illustrates  a  double  floating-lever  type  hydraulic 
governor  adapted  to  large  turbines  as  built  by  the  I.  P.  Morris 
Company. 

3.  PRESSURE  REGULATORS  OR  RELIEF  VALVES 

Pressure  regulation  is  a  problem  which  must  be  considered  in 
connection  with  the  speed  regulation  of  a  plant.  As  previously 
stated,  if  the  pipe  lines  are  long,  either  suflScient  flywheel  effect 


Digitized  by 


Google 


PRESSURE  REGULATORS  OR  RELIEF  VALVES  259 

must  be  provided  to  permit  a  slower  governor  action  or  a  surge 
tank  must  be  used.  If  such  a  surge  tank  cannot  be  located  close 
to  the  power-house,  relief  valves  must  £dso  be  provided,  and  some 
conditions  may  require  all  the  devices. 

Relief  valves  are  of  two  principal  types,  the  synchronous  by- 
pass governor  operated,  and  the  direct-pressure  operated.  Their 
design  is  essentially  the  same,  except  for  the  control  mechanism. 
The  first  becomes  immediately  operative  with  the  closing  gate 
motion  and  this  action  continues  imtil  the  gates  stop  moving. 
The  water  rejected  by  the  turbine  as  the  gates  close  is  discharged 
through  the  regulator.  Thus  the  penstock  velocity  instead  of 
being  suddenly  checked,  resulting  in  waterhanuner,  remains 
practically  unchanged.  The  device  is  made  water-saving  by  the 
use  of  a  dashpot,  which  permits  of  a  relative  motion  of  the  con- 
nection between  the  turbine  gate  mechanism  and  the  by-pass 
valve.  The  adjustment  of  the  dashpot  is  made  such  that  the 
by-pass,  after  having  been  opened  due  to  a  sudden  closing  of  the 
turbine  gates,  is  closed  within  a  period  which  is  suflSciently  long 
to  prevent  a  dangerous  pressure  rise  in  the  pipe  line.  If  the  load 
goes  ofif  gradually  and  the  gates  are  closed  at  a  rate  slower  than 
that  produced  by  the  dashpot,  the  pressure  regulator  remains 
inactive.  If  the  gates  are  again  opened  before  the  dashpot  has 
closed  the  pressing  regulator,  then  it  should  close  synchronously 
with  the  gate  motion,  otherwise  an  e?cess  quantity  of  water 
is  discharged,  causing  a  drop  of  pressure  in  the  pipe  line. 

The  second  class  of  relief  valves,  or  the  direct-pressure  operated 
type,  do  not  act  imtil  the  pressure  in  the  pipe  hne  or  turbine 
casing  has  risen  above  normal  and  are,  therefore,  a  more  direct 
means  of  protecting  pipe  Unes  against  dangerous  pressure  rises, 
such  as  caused  by  the  clogging  up  of  the  gates,  etc.  Governor- 
operated  pressiu-e  regulators  are,  however,  made  which  permit  of 
an  automatic  action  independent  of  the  gate  motion  to  take  care 
of  such  emergencies. 

In  order  to  obtain  ideal  results,  the  maximum  capacity  of 
the  regulator  should  be  equal  to  the  full-load  discharge  of  the 
turbine  less  the  discharge  required  to  run  at  synchronous  speed 
without  load.  Ordinarily  some  sacrifice  is  made  to  reduce  the 
size  of  the  regulators.  They  are  seldom  installed  in  excess  of 
75  per  cent  of  the  maximiun  turbine  discharge,  and  in  many  cases 
not  more  than  40  per  cent  or  50  per  cent  is  provided.     In  such 


Digitized  by 


Google 


260 


HYDRAULIC  EQUIPMENT 


FiQ.    140. — Governor-operated    Relief    Valve.      (Wellman-SeaveivMoiigan 

Company.) 


Digitized  by 


Google 


prTMAU 


PRESSURE  REGULATORS  OR  RELIEF  VALVES 


261 


cases,  of  course,  some  pressure  rise  occurs  in  the  penstock.  The 
size  of  regulator  depends  largely  upon  the  water  velocity  in  the 
penstock  and  upon  the  length  of  penstock  between  the  turbine  and 
the  forebay  or  between 
the  turbine  and  the  surge 
tank,  if  one  is  used.  It  is 
usually  attached  directly 
to  the  turbine  casing  and 
discharges  into  the  tail- 
race,  and  the  discharge 
should  not  be  connected 
to  the  draft  tube. 

Fig.  140  illustrates  a 
governor-operated  pres- 
sing regulator  used  with 
reaction  turbines.  It  is 
mechanically  connected  to 
the  gate  mechanism  of  the 
turbine  but  the  power  re- 
quired to  operate  it  is  sup- 
plied by  the  pressing  in 
the  penstock.  No  load  is 
imposed  upon  the  gov- 
ernor nor  any  pressure 
drawn  from  the  governor 
system.  The  connection 
to  the  turbine  gate 
mechanism  simply  oper- 
ates the  pilot  valve  of  the 
regulator  which  controls 
its  action. 

Fig.  141  shows  another 
type  of  governor-operated 
relief  valve,  in  which  the 
valve  is  interconnected 
with  the  turbine  gates 
through    a  self-contained 

oil-pressiure  system,  the  operation  of  the  relief  valve  being  pro- 
duced directly  by  the  motion  of  the  turbine  gate.  Above  the 
elbow  forming  the  body  of  the  reUef  valve  casing  will  be  noticed 


Fig.  141.    Governor-operated  Relief  Valve. 
(I.  P.  Morris  Company.) 


Digitized  by 


Google 


262  HYDRAULIC  EQUIPMENT 

a  large  cylinder  containing  a  balancing  piston,  the  purpose  of 
which  is  to  equalize  the  load  on  the  valve,  allowing,  however, 
a  small  residual  force  tending  to  close  the  relief  valve.  Above 
the  balancing  cylinder  is  a  smaller  cylinder  containing  a  piston 
for  operating  the  valve.  The  two  pipe  connections  shown  at  the 
ends  of  this  small  cylinder  are  joined  by  pipes  to  the  two  ends 
of  a  jack  cylinder  moimted  on  the  tailrod  on  one  of  the  operating 
cylinders  of  the  turbine.  The  jack  cylinder  and  the  operating 
cylinder  of  the  reUef  valve  displace  equal  volimies  when  their 
respective  pistons  move  through  the  full  stroke.  The  relief  valve 
is  thus  forced  to  move  by  an  incompressible  fluid  coliunn,  and  the 
operation  is  similar  to  that  which  would  be  obtained  by  a  direct 
mechanical  connection  between  the  turbine  gates  and  the  relief 
valve. 

The  slow-closing  feature  of  the  valve  operation  is  obtained  by 
by-pass  connections  joining  the  two  ends  of  the  operating  cylinder. 
A  needle  valve  permits  the  rate  of  closing  to  be  adjusted.  The 
method  of  operating  this  reUef  valve  has  several  advantages. 
One  of  these  is  the  positive  action  obtained,  the  effect  of  which  is 
to  prevent  the  turbine  gates  moving  at  a  rapid  rate,  if  for  any 
reason  the  reUef  valve  should  fail  to  move  owing  to  any  accidental 
cause,  such  as  lodging  of  obstructions  in  the  reUef  valve.  Thus,  if 
the  reUef  valve  is  imable  to  open,  the  turbine  gates  will  be  auto- 
matically prevented  from  closing,  except  at  a  slow  rate  which  will 
not  endanger  the  penstock. 

For  relief  valves  used  with  impulse  wheels  see  section  on  "  Tur- 
bines," page  221. 

4.  WATER-FLOW  METERS 

One  of  the  most  convenient  means  of  measuring  the  amount  of 
water  taken  by  a  hydrauUc  station  and  for  ascertaining  the 
efiiciency  of  the  turbines  is  the  Venturi  meter. 

Venturi  Meter.  It  consists  of  a  meter  tube,  which  is  inserted 
in  the  pipe  line  similar  to  a  section  of  pipe,  and  of  a  register  which 
is  piped  to  the  tube  and  which  can  be  located  at  any  convenient 
place  in  the  station,  as  shown  in  Fig.  142. 

The  interior  contour  of  the  meter  tube  is  shown  in  Fig.  143, 
and  the  accuracy  of  the  meter  greatly  depends  upon  its  proper 
design.  As  the  water  flows  from  A  toward  the  throat  B,  its 
velocity  rapidly  increases  and  the  pressure  at  B  becomes  materi- 


Digitized  by 


Google 


WATER-FLOW   METERS 


263 


ally  less  than  the  pressure  at  A,  This  difference  in  pressure 
between  A  and  B  can  be  accurately  measured,  and  bears  an  exact 
ratio  at  all  times  to  the  rate  of  xflow  through  the  throat  B.  After 
passing  the  throat,  the  velocity  begins  to  decrease  with  an  accom- 
panying rise  in  pressure,  and  when  C  is  reached  the  pressure  tem- 
porarily lost  at  B  has  been  almost  entirely  regained.  Therefore,  a 
properly  proportioned  tube  not  only  provides  a  basis  for  accurate 


FiQ.  142.— Method  of  Installing  Venturi  Meter. 


Fig.  143. — Principle  of  Venturi  Meter  Tube. 

measurement  of  the  flow,  but  it  will  deliver  practically  the  same 
amount  of  water  as  a  straight  pipe  of  equal  length  and  diameter. 

Conmiercial  tubes  are  made  in  two  or  more  sections,  as  seen 
from  Fig.  142,  and  near  the  inlet  and  at  the  throat  are  annular 
chambers  communicating  with  the  interior  of  the  pipe  by  numer- 
ous ventholes.  The  throat  portion  is  lined  with  bronze  accurately 
bored  to  a  definite  diameter  and  contour. 


Digitized  by  VjOOQ IC 


264 


HYDRAUUC  EQUIPMENT 


Connections  to  the  registering  instrument  are  made  by  two 
small  pipes,  one  at  the  inlet  pressm-e  chamber  and  the  other  at  the 
throat  pressure  chamber.  No  water  flows  through  these  pipes  as 
they  simply  transmit  the  two  pressures,  the  difference  in  which 
controls  the  readings  of  the  instrument. 

Registers.  There  are  different  kinds  of  registers,  the  most 
complete  being  illustrated  by  Fig.  144.    At  the  back  there  are 

two  vertical  wells  connected  at  the  bot- 
tom. One  well  is  subjected  to  the 
inlet  and  the  other  to  the  throat 
pressing  of  the  Venturi  Meter  Tube, 
these  pressures  being  transmitted  by 
the  two  small  pipes  as  previously 
mentioned.  In  one  well  is  a  heavy- 
metal  float  resting  upon  the  mercury, 
a  part  of  which  flows  from  this  well 
to  the  other  well  in  direct  proportion 
to  the  changes  in  flow  through  the 
Venturi  Meter  Tube.  This  is  accom- 
plished by  having  the  receiving  well 
of  a  variable  cross-section.  Conse- 
quently, the  large  float  descends  in 
direct  proportion  to  the  change  in  rate 
of  flow  and  its  motion  is  transferred 
to  the  main  shaft  of  the  instrument 
by  means  of  a  rigid  float  rod  and  suit- 
able gearing.  The  movement  of  the 
shaft  is  in  turn  transferred  by  means 
of  rack-and-spur  gearing  to  the  long 
main  lever  of  the  instrument  which 
carries  the  chart  pen  and  the  integrat- 
ing coimter. 

The  recorder  dial  contains  a  large 
circular    chart    giving    an    unbroken 
autographic  record  of  the  rate  of  flow  through  the  meter  tube. 

The  coimter  dial  shows  the  total  amoimt  of  water  (gallons, 
cubic  feet,  etc.)  which  has  passed  through  the  tube. 

The  indicator  shows  the  exact  rate  of  flow  in  gallons  per  day 
or  other  imits  at  the  moment  of  observation. 

Where  the  expense  of  installing  a  complete  registering  outfit 


Fig.  144. — Venturi  Meter 
Register. 


Digitized  by 


Google 


WATER-FLOW  METERS 


265 


\  !■! 


!l 


is  prohibitive,  or  for  testing  the  accuracy  of  register  instru- 
ments, the  manometer  may  be  advantageously  used,  and  it  may 
be  connected  with  the  same  pipes 
that  serve  to  connect  the  tube     ^  „     ^ 

with  the  registering  apparatus. 

Manometer.  The  Venturi 
Meter  Manometer  as  illustrated 
in  Fig.  145,  consists  essentially  of 
a  U-tube  using  the  same  principle 
as  a  barometer.  The  large  mer- 
cury well  is  connected  to  the  up- 
stream of  the  Venturi  Meter  Tube 
and  the  throat  of  the  Venturi 
Meter  Tube  is  connected  to  the 
small  vertical  glass  tube,  thus  the 
downward  motion  of  the  mercury 
surface  in  the  mercury  well  is 
very  slight  in  comparison  with 
the  upward  motion  of  the  mer- 
cury surface  in  the  small  glass 
tube.  The  slight  motion  of  the 
large  surface  is  properly  corrected 
in  the  fixed  scales  of  the  instru- 
ment. The  rate  of  flow  corre- 
sponding to  the  difference  in  height 
of  the  mercury  surfaces  is  read  on 
the  graduated  scale.  This  instru- 
ment is  absolutely  accm^te,  con- 
taining no  moving  parts  whatever 
except  the  mercury  iself . 

5.  WATER-STAGE  REGISTERS 

Automatic  water-stage  regis- 
ters are  divided  into  two  classes — 
those  making  a  printed  record, 
and  those  making  a  graphic  record. 
In  the  first  type  a  printed  record  of  the  gauge  height  and  time 
is  made,  while  in  the  second  type  the  record  is  traced  by  a  pen  or 
pencil  on  the  surface  of  a  paper  sheet,  both  moving  in  harmony 
with  time  and  height. 


Fia.  145. — Barometric  Venturi 
Manometer. 


Digitized  by 


Google 


i 


Fig.  146. — Automatic  Stage  Register  Making  Printed  Record.    (Manufactured 

by  W.  &  L.  E.  Gurlcy,  Troy,  N.  Y.) 

266 


Digitized  by 


Google 


WATERrSTAGE  REGISTERS 


267 


The  first  tjrpe  of  register  is  designed  to  give  printed  records  of 
the  rise  and  fall  of  water  continuously  for  a  long  period  of  time, 
and  is  especially  adapted  for  stations  where  it  is  impracticable  or 
impossible,  by  reason  of  inaccessibility,  for  the  observer  to  visit 
the  station  for  long  intervals  of  time  and  where  the  record  to  be 


Fig.  147. — Tape  Reel  for  Use  with  Water  Stage  Printing  Register. 

of  service  should  be  continuous.     The  records  are  given  at  inter- 
vals of  fifteen  or  thirty  minutes. 

Fig.  146  shows  an  automatic  water-stage  register  making  a 
printed  record,  and  Fig.  147  shows  a  tape  reel  for  handling  and 
examining  the  records.  .  A  graphic  recording  register  is  shown  in 
Fig.  148. 


Digitized  by 


Google 


268 


HYDRAUUC  EQUIPMENT 


In  installing  an  automatic  register  (Fig  149)  it  is  necessary 
to  provide  a  well  for  the  float,  connected  with  the  river  by  an 
intake  pipe,  a  house  to  shelter  the  register,  and  staff  or  hook 


i 


Fig.  148. — Automatic  Graphic  Recording  Water  Stage  Register.     (Manufao 
factured  by  W.  &  L.  E,  Ourley,  Troy,  N.  Y.) 


Digitized  by 


Google 


WATER-STAGE  REGISTERS 


269 


gauges  with  bench  marks  for  checking  the  record  and  maintaining 
the  datum.  The  well  and  the  house  should  be  located  far  enough 
back  from  the  river  to  be  out  of  danger  from  floating  ice  or 
drift  and  to  provide  sufficient  protection  for  the  well  and  pipes 
to  prevent  freezing.  A  permanent  ladder  should  extend  to  the 
bottom  of  the  well,  so  that  the  float  and  intake  can  be  readily 


Fig.  149. — Method  of  Installing  Automatic  Water  Stage  Register. 

inspected.  If  the  register  is  to  be  maintained  for  a  long  period 
the  well  should  be  lined  with  concrete,  otherwise  a  heavy  plank 
lining  may  be  used.  The  intake  pipe  should  be  placed  well 
below  the  lowest  stage  of  the  river  and  provided  with  a  screen 
for  keeping  out  silt  and  foreign  material.  It  should  also  be  pro- 
vided with  check  gate  as  it  enters  the  well,  so  that  the  flow  can 
be  reduced  if  necessary  to  eliminate  wave  action. 


Digitized  by 


Google 


CHAPTER  IX 

ELECTRICAL  EQUIPMENT 

1.  GENERAL  CONSIDERATIONS 

Before  entering  into  a  detailed  study  of  the  apparatus  com- 
prising the  electrical  equipment,  there  are  two  broad  problems 
which  require  a  more  careful  consideration  and  must  first  be 
decided  on,  inasmuch  as  they  have  an  important  bearing  on  the 
entire  equipment.  These  problems  deal  with  the  voltage  and  the 
frequency. 

Voltage.  There  are  three  voltages  between  which  a  distinc- 
tion must  be  made  in  a  hydro-electric  power  system;  viz.,  the 
generator  voltage,  the  transmission  voltage  and  the  distribution 
voltage. 

Generator  VoUage.  When  additions  to  an  existing  plant  or 
system  are  made,  the  voltage  of  the  new  generators  is  generally 
determined  by  that  of  the  old  machines,  or  by  some  other  con- 
dition of  the  installation.  In  new  installations,  however,  the  gen- 
erator voltage  can  be  determined  only  after  considering  a  number 
of  factors.  For  example,  a  compromise  must,  as  a  rule,  be  found 
between  the  increased  cost  of  a  high-voltage  machine  and  its 
control  equipment  as  compared  with  the  reduced  cost  of  the  bus- 
bars and  connections  caused  by  the  smaller  amoimt  of  copper 
required.  Whether  generators  are  to  be  wound  for  a  high  voltage 
for  direct  transmission,  or  for  low  voltage  and  step-up  transformers, 
is  to  a  certain  extent  also  decided  by  the  relative  cost  of  the  two 
methods.  If  economically  feasible  the  latter  method  with  step-up 
transformers  is,  however,  the  most  reliable  and  to  be  recommended. 
In  other  instances  the  nature  of  a  local  load  may  be  such  that, 
by  installing  high-voltage  generators,  power  for  this  load  may  be 
directly  transmitted  at  the  generator  voltage;  while  at  the  same 
time  step-up  transformers  may  be  provided  for  raising  the  pres- 
sure of  the  current  which  is  to  be  transmitted  for  greater  dis- 
tances. The  standard  generator  voltages  are  given  imder  "  Syn- 
chronous Generators." 

270 


Digitized  by 


Google 


GENERAL  CONSIDERATIONS  271 

Transmission  Voltage.  The  transmission  voltage  should  be 
chosen  to  insure  the  most  economical  ensemble.  Many  factors 
affect  the  problem  variously,  and  their  nature  makes  a  mathe- 
matical expression  difficult  and,  as  a  rule,  unsatisfactory.  The 
distance  of  the  transmission  is  naturally  the  factor  which  governs 
the  choice  of  the  voltage  to  the  greatest  extent.  The  crossHsection 
area  and,  consequently,  the  weight  to  the  transmission  conduc- 
tors, varies  inversely  as  the  square  of  the  voltage  for  a  given  load. 
The  cost  of  the  conductors  is,  therefore,  reduced  75  per  cent  every 
time  the  voltage  is  doubled,  and  it  would,  consequently  seem 
proper  to  use  the  highest  voltage  possible  in  any  given  case. 
Though  with  increasing  voltage,  the  cost  of  the  conductors  de- 
creases, the  cost  of  other  apparatus  and  appliances  increases. 
This  involves  transformers,  switching  equipment,  lightning 
arresters  and  line  structure  and  insulators,  while,  of  course,  the 
necessary  safety  requirements  become  stricter  with  higher 
voltages. 

With  very  high  voltages  and  long  lines  the  capacity  current  of 
the  lines  becomes  considerable,  especially  in  sixty-cycle  systems, 
and  may  reach  values  higher  than  the  full-load  ciurent.  Its 
greatest  objection  is  that  it  loads  the  generators  with  current  which 
represents  no  power,  and  where  small  units  are  used  it  may  often 
render  it  impossible  to  throw  one  machine  on  the  line  alone. 
Much  more  serious,  however,  is  the  impairment  of  the  voltage 
regulation  incident  to  very  long  lines,  i.e.,  the  voltage  variation 
between  no  load  and  full  load,  especially  for  inductive  loads.  By 
providing  S3mchronous  condensers,  it  is,  however,  possible  to 
compensate  for  the  wattless  currents  and  improve  the  regulation. 
Another  factor  which  has  a  limited  bearing  on  high  potentials 
for  transmission  purposes  is  corona,  as  experience  has  shown  that 
if  the  voltage  on  a  given  line  is  raised  beyond  a  certain  point  the 
air  at  the  surface  of  the  conductors  breaks  down  as  an  insulating 
medium  and  becomes  luminous.  The  most  serious  objection  to 
corona  comes  from  the  losses,  which  increase  at  a  high  rate  as  the 
voltage  is  raised  above  this  luminous  or  so-called  visual  critical 
point.  This  critical  voltage  increases  with  the  size  of  the  con- 
ductors and  their  spacing,  and  by  properly  choosing  these  values 
the  losses  may  be  materially  reduced  or  entirely  eliminated.  For 
high  altitudes  corona  starts  at  lower  voltages  and  this  should  be 
given  careful  consideration  (see  section  on  "  Station  Wiring.'* 


Digitized  by 


Google 


272 


ELECTRICAL  EQUIPMENT 


The  factors  determining  the  proper  transmission  voltage  are, 
as  a  rule,  of  an  economical  nature,  and,  while  no  fixed  formula 
for  determining  the  voltage  can  be  given,  in  general  it  may  be 
said  that  the  most  economical  voltage  is  the  one  for  which  the 
annual  cost  of  the  energy  loss  added  to  the  annual  cost  for  depre- 
ciation and  interest  on  the  first  cost,  becomes  a  minimum.  In 
determining  the  value  of  the  energy  loss,  a  mean  value  for  a 
number  of  years  should  evidently  be  taken,  and  the  value  should 


160,000 


123,000 


.100,000 


I  76,000 


60,000 


86,000 


/ 

^ 

"^ 

/ 

/ 

i 

/ 

/ 

/ 

0                 6 

B               K 
Transmiss 

»              1, 
ion  Distanc 

50          a 

;e  in  Milea 

w          a 

n 

FiQ.  150. — Approximate  Voltages  for  Power  Transmission  of 
Various  Lengths. 

be  based  on  the  cost  for  which  the  power  can  be  produced.  The 
interest  and  depreciation  as  well  as  operating  charges  should  only 
be  applied  to  such  items  that  will  vary  with  changes  in  the  volt- 
age, such  as  the  line  conductors  and  tower  line,  the  generating 
and  substation  buildings,  transformers,  switching  equipment  and 
Hghtning  arresters. 

An  approximate  average  scale  of  voltages  for  transmission 
lines  up  to  250  miles  in  length  is  given  in  Fig.  150. 


Digitized  by 


Google 


GENERAL  CONSIDERATIONS  273 

Disiribviion.  Voltage  The  selection  of  the  proper  distributing 
voltage  is  also  an  important  matter.  Where  large  territories 
have  to  be  served  from  high  voltage  transmission  circuits,  the 
general  practice  seems  to  indicate  that  the  most  economical  volt- 
ages for  such  systems  are  in  the  neighborhood  of  from  22,000  to 
33,000  volts.  A  second  or  even  third  transformation  is,  therefore, 
necessary  before  the  power  can  be  used  for  motors  or  lighting. 

The  distribution  of  alternating  current  for  general  commercial 
purposes  is  accomplished  almost  universally  by  2300  volt  mains, 
suppl3dng  step-down  tran^ormers  located  near  groups  of  con- 
sumers, whose  premises  are  served  by  secondary  mains  at  115  to 
230  volts.  Single-phase  circuits  are  quite  generally  used  for 
lighting  service,  while  power  service  is,  as  a  rule,  given  from  two- 
phase  or  three-phase  mains.  The  former  S3rstem  is  used  chiefly 
where  this  method  of  distribution  was  established  in  the  early 
period  of  the  development,  and  where  it  is  too  extensive  to  warrant 
a  change  to  the  three-phase  system,  which  is  standard  for  all  new 
installations  where  a  polyphase  supply  is  wanted  for  power  service. 

For  small-  and  medium-sized  cities  a  three-wire,  "  delta  "- 
connected,  2300-volt  system  is  very  generally  used  for  power  dis- 
tribution, while  for  larger  cities  there  is  a  steady  trend  toward  the 
four-wire,  "  Y  '^-connected  system  operating  at  2300-4000  volts. 
There  are  numerous  advantages  with  this  system  where  feeders 
are  extended  more  than  two  miles  from  the  point  of  supply,  and 
where  adjacent  towns  within  a  radius  of  five  miles  may  be  served 
without  step-up  transformers  or  substations.  It  is  possible  to 
regulate  the  phases  separately,  and  there  is  not  so  much  of  a 
necessity  for  maintaining  a  carefully  balanced  load.  Even  for 
secondary  distribution  the  four-wire,  three-phase  system,  oper- 
ating at  approximately  115-200  volts,  is  being  generally  used. 
With  this  system  lighting  and  motor  service  may  be  given  for  all 
ordinary  retail  purposes  from  the  same  circuit,  the  principal  dis- 
advantages being  that  there  are  three  phases  to  be  kept  balanced. 

Frequency.  The  subject  of  frequency  for  commercial  power 
and  lighting  systems,  far  from  being  settled,  is  discussed  again  with 
every  new  installation.  Frequency  affects  the  operating  charac- 
teristics of  circuits  and  apparatus,  and  also  their  cost. 

The  frequencies  most  commonly  employed  in  this  country 
are  25  and  60.  In  general  it  may  be  said  that,  where  lighting 
load  is  predominating,  60  cycles  should  preferably  be  selected; 


Digitized  by 


Google 


274 


ELECTRICAL  EQUIPMENT 


while,  if  the  load  mainly  consists  of  power,  25  cycles  is  often  prefer- 
able, especially  if  the  load  consists  of  a  large  number  of  syn- 
chronous converters.  With  a  large  induction  motor  load  it  may, 
on  the  other  hand,  be  more  advantageous  to  use  60  cycles  on 
account  of  the  greater  number  of  speeds,  which  are  possible  with 
this  frequency. 

In  the  following  discussion  the  influence  of  frequency  will  be 
treated  in  connection  with  frequency  changers,  generators,  trans- 
formers, transmission  lines,  induction  motors,  synchronous  con- 
verters, railroad  work  and  illumination. 

Frequency  Changes.  Frequency  changers  are  primarily  used 
for  efifecting  a  change  in  frequency.  They  are  either  utilized  for 
obtaining  a  frequency  high  enough  for  lighting  purposes  from  a 
low-frequency  system,  or,  as  a  means  of  interchanging  power 
between  systems  operating  at  different  frequencies. 

The  change  from  25  to  60  cycles  or  vice  versa  requires  a  set 
running  at  300  R.P.M.,  which  is  a  serious  limitation  because  this 
speed  is  much  too  low  for  the  economical  design  of  frequency 
changers  of  small  or  moderate  size.  If  an  exact  ratio  is  not  abso- 
lutely necessary,  as  when  power  is  taken  from  an  existing  system 
for  lighting  and  industrial  purposes,  and  the  frequency  changer 
is  not  intended  for  tying  two  generating  systems  together,  the 
available  range  of  speed  is  greatly  increased  as  shown  in  the  fol- 
lowing table. 

TABLE  XL 
Frequency-Changer  Combinations 


Fbbquknct. 

Polka. 

Speed. 

Generator 

Frequency. 

Motor. 

Generator. 

Motor. 

Generator. 

25 

62.5 

4 

10 

750 

4.17  per  cent  high 

25 

62.5 

8 

20 

375 

4.17  percent  high 

25 

60 

10 

24 

300 

Exact 

25 

58.3 

6 

14 

500 

2.78  per  cent  low 

25 

56.3 

8 

18 

375 

6. 18  per  cent  low 

60 

26.7 

18 

8 

400 

6.8    per  cent  high 

60 

25.7 

14 

6 

514 

2.8    per  cent  high 

60 

25 

24 

10 

300 

Exact 

60 

24 

20 

8 

360 

4        per  cent  low 

60 

24 

10 

4 

720 

4       per  cent  low 

Digitized  by  VjOOQ IC 


GENERAL  CONSIDERATIONS  275 

While  synchronous  motors  are  abnost  invariably  used  with 
frequency  changers,  induction  motors  may  be  used  if  proper 
arrangements  are  provided  for  adjusting  the  slip  so  as  to  insure 
a  satisfactory  parallel  operation.  This  adjustment,  of  course, 
means  the  introduction  of  a  permanent  resistance  and  a  corre- 
sponding loss,  and  is,  therefore,  undesirable  unless  other  advan- 
tages of  greater  importance  can  be  obtained.  Where  only  one 
set  is  required  speed  adjustment  is  not  necessary,  and  the  motor 
may  be  designed  with  a  sUp  which  will  just  be  sufficient  to  bring 
the  generator  frequency  to  the  right  value. 

Generalors.  'The  frequency  of  sjmchronous  generators  in 
alternations  per  minute  is  equal  to  the  number  of  poles  times  the 
revolutions  per  minute,  and  the  periodicity  or  cycles  per  second 
is  shown  by  the  following  equation: 

^    ,        Number  of  poles  X  rev.  per  min. 
Cycles ^ ^ . 

Due  to  the  fact  that  there  is  a  natural  relation  between  the 
windings  of  electrical  apparatus  which  varies  inversely  as  the 
square  of  the  frequency,  the  higher  the  frequency  the  greater  is, 
in  general,  the  peripheral  velocity  at  the  same  revolutions  per 
minute.  Increase  in  peripheral  velocity  means  a  larger  diameter 
with  a  smaller  length  and  a  better  natiu'al  ventilation.  The 
higher  periodicity  in  definite  pole  machines  is  also  preferable 
in  that  the  load  of  the  rim  of  the  spider  is  better  distributed  and 
smaller  in  amoimt  at  the  point  of  attachment  of  poles. 

The  induced  e.m.f.  is  directly  proportional  to  the  frequency 
and,  due  to  the  lower  core  loss  with  lower  frequencies,  the  effi- 
ciency is  naturally  better  at  25  than  at  60  cycles.  The  cost  is 
also  increased  by  the  frequency  there  being  a  natural  tendency 
for  25-cycle  apparatus  to  be  heavier  than  60-cycle.  As  a  general 
rule,  the  labor  item  is  higher  on  the  higher  frequency  machines, 
and  the  material  item  higher  with  the  lower  frequencies. 

Parallel  operation  is  more  satisfactory  at  low  frequencies,  so 
far  as  the  variation  in  angular  velocity  is  concerned.  Due  to 
other  factors,  the  conditions  for  parallel  operation  depend  more 
upon  the  relations  between  natural  and  impressed  frequencies, 
rather  than  upon  the  absolute  value  of  either. 

Traruformera,  The  frequency  has  a  very  important  bearing 
both  on  the  design  and  operation  of  transformers.    With  trans- 


Digitized  by 


Google 


276  ELECTRICAL  EQUIPMENT 

fonners  and  other  electric  apparatus  using  two  windings  and  an 
iron  core,  the  ratio  of  turns,  other  factors  remaining  the  same,  will 
be  approximately  inversely  as  the  square  root  of  the  frequency. 
The  lower  the  frequency  the  larger  the  flux,  and  the  larger  the 
number  of  turns  for  the  same  voltage.  Therefore,  transformers 
increase  in  cost  and  weight  as  the  frequency  decreases. 

The  regulation  of  25-cycle  transformers  is  not  quite  as  good  as 
for  60-cycle  on  account  of  the  increased  drop,  due  to  the  great 
number  of  turns  and  their  increased  mean  length,  and  the  efli- 
ciency  is  also  somewhat  less. 

Operating  25-cycle  transformers  on  a  60-cycle  circuit  decreases 
the  flux  density  and  the  core  loss.  Operating  a  60-cycle  trans- 
former on  a  25-cycle  circuit  increases  the  density  and  core  loss,  and, 
in  general,  gives  a  prohibitive  exciting  current.  Frequency  also 
enters  into  the  mechanical  forces  to  which  a  transformer  may  be 
subjected,  as  the  reactance  increases  with  the  frequency,  and, 
while  the  mechanical  force  varies  directly  as  the  square  of  the 
current,  a  25-cycle  transformer  operating  on  a  60-cycle  circuit 
would  be  subject  to  about  one-half  the  mechanical  strains  on  short- 
circuit.  The  limit  of  reactance  in  a  transformer  is  from  8  to  10 
per  cent  at  60  cycles  and  somewhat  higher  at  25  cycles. 

Transmission  Lines.  Transmission  lines  are  designed  from 
considerations  of  regulation  and  efficiency  and,  as  explained  more 
fully  imder  "  Voltage,"  the  regulation  is  better  as  the  frequency  is 
lower,  and  so  for  commercial  work  25  cycles  is  preferable  to  60 
cycles,  considering  the  line  alone.  The  capacity  current  plays, 
as  stated,  also  an  important  part  with  small  units  and  high  volt- 
ages, rendering  it  often  impossible  to  throw  one  machine  on  the 
line  alone.  Both  the  reactance  and  the  capacity  current  of  the 
line  are  proportionate  to  the  frequency  as  shown  by  the  following 
equations: 

Reactance = 2irfL ; 

Capacity  current  =  2ir/CJ?; 

The  resistance  of  wires  and  cables  canying  alternating  cur- 
rents is  also  affected  by  the  frequency,  in  that  the  current  is  not 
distributed  uniformly  over  the  cross-section  of  the  conductors,  the 
current  density  being  higher  near  the  periphery.  This  is  known  as 
"  skin  effect  **  and  results  in  an  increased  resistance.    The  eflfect 


Digitized  by 


Google 


GENERAL  CONSIDERATIONS  277 

is,  however,  negligible  for  low  frequencies  and  small  conductors, 
but  increases  rapidly  for  higher  frequencies  and  large  conductors. 
With  magnetic  material  it  is  much  higher  than  with  non-magnetic, 
and  its  effect  should  be  considered  where  iron  conductors  are  used 
and  for  heavy  copper  work. 

Induction  Motors.  The  speeds  of  25-cycle  induction  motors 
for  general  application  are  practically  limited  to  750,  500  and  375 
R.P.M.,  while  the  corresponding  speeds  for  60-cycle  motors  would 
be  1200,  900,  720,  600,  514,  450,  and  400  revolutions.  Twenty- 
five-cycle  motors  could,  of  course,  be  woimd  for  two  poles,  giving  a 
speed  of  1500  revolutions,  but  this  is  rarely  done  except  in  the 
very  small  sizes.  The  objection  is  that  since  the  flux  per  pole  is 
twice  as  large  as  in  the  four-pole  type,  the  section  of  iron  back  of 
the  slots  must  be  twice  as  great,  for  the  same  rotor  diameter. 
Moreover,  the  end  connections  become  very  long  and  the  machine 
difficult  to  wind  and  consequently  the  cost  is  very  materially 
increased. 

The  efficiency  depends  upon  a  number  of  features.  The  lower 
frequency  will,  of  course,  tend  to  make  the  iron  loss  less,  but  on  the 
other  hand,  the  copper  loss  will  be  considerably  greater  on  accoimt 
of  the  longer  end  connections,  and,  as  a  rule,  the  efficiency  is  foimd 
to  be  somewhat  lower  for  low-  than  for  high-frequency  motors. 

The  power  factor  of  an  induction  motor  is  expressed  by  the 

ratio  tt — 1  •       x-    It  is  affected  by  the  reactance  and  the  mag- 
Kv.A.  mput 

netizing  current.  At  constant  line  voltage  the  latter  remains 
practically  constant,  while  the  former  varies  with  the  current. 
The  shape  of  the  power  factor  curve,  that  is,  the  power  factor  at 
fractional  loads  and  overloads,  therefore,  depends  upon  the  rela- 
tive values  of  the  magnetizing  ciurent  and  the  reactance. 

R 
Power  factor = cos  0 = •=. 

A  motor  with  a  relatively  large  magnetizing  current  and  a  low 
reactance  will,  in  general,  have  a  low-power  factor  at  fractional 
loads  and  a  rapidly  increasing  power  factor  at  higher  loads,  while 
a  motor  with  a  relatively  low  magnetizing  cmrent  and  a  high  reac- 
tance will  have  a  high-power  factor  at  fractional  loads  and  only  a 
slightly  greater  power  factor  at  overloads. 

The  25-cycle  motor  has  an  inherently  lower  reactance  and 


Digitized  by 


Google 


278  ELECTRICAL  EQUIPMENT 

requires  less  magnetizing  current,  for  which  reason  its  power 
factor  is  considerably  higher  than  for  high-frequency  motors. 

The  starting  torque  and  the  maximum  torque  depend  inversely 
on  a  function  of  the  reactance,  and  are,  therefore,  higher  for  low 
frequencies. 

The  starting  torque  of  an  induction  motor  is  equal  to: 

the  starting  current  is  equal  to 

E 
Z' 

the  running  torque  is  equal  to 

,  E^sri 


the  maximum  torque  is  equal  to 

k ^ . 

[{2{r+VW+X^)]' 

where  k  =  constant ; 

J? = applied  voltage; 

«  =  slip; 

r=stator  resistance  per  phase; 
fi  ==  rotor  resistance  per  phase; 
X= total  reactance; 
Z  =  total  impedence ; 

Comparing  the  weights  based  on  motors  of  the  same  capacity 
and  speed,  it  is  found  that,  on  the  average,  25-cycle  motors  will 
weigh  about  15  per  cent  more  than  60-cycle  motors.  For  the 
smaller  sizes  there  is  very  httle  difference  in  the  cost,  but  as  the 
sizes  increase  there  is  a  marked  difference  in  favor  of  the  60-cycle 
motors. 

Synchronous  Converters.  Ji  synchronous  converter  being  in 
effect  a  combination  in  one  machine  of  a  synchronous  motor  and  a 
direct-current  generator,  the  important  factors  in  which  the  fre- 
quency is  concerned  have  to  do  almost  entirely  with  the  contin- 
uous-current side.  The  continuous-current  generator,  as  a  rule, 
nms  at  frequencies  much  below  25  cycles,  and  at  the  frequencies 


Digitized  by 


Google 


GENERAL  CX)NSIDERATIONS  279 

of  synchronous  converters,  especially  for  60  cycles  and  above, 
the  problems  of  commutation  and  commutator  construction  be- 
come of  importance.  The  pole  pitch  on  the  commutator,  arma- 
ture or  field,  is  the  space  passed  through  in  one  alternation,  and 
it  is  thus  seen  that  there  is  a  natural  tendency  for  higher  per- 
ipheral speeds  at  the  higher  frequencies,  and  it  is  the  limitation 
of  peripheral  speed  which  fixes  the  limits  of  design. 

With  direct-current  machines  this  occurs  with  turbine-driven 
generators  and  in  the  commutators,  which  are  necessarily  mechan- 
ical in  construction,  consisting,  as  they  must,  of  a  certain  amoimt 
of  insulation.  Direct-current  generators  are,  therefore,  more 
limited  in  speeds  than  alternating-ciurent,  and  the  same  holds 
true  when  they  are  combined  as  in  rotary  converters. 

Improvements  in  design  have  made  the  60-cycle  synchronous 
converter  entirely  satisfactory  for  the  conditions  under  which 
such  machines  operate.  In  eflBiciency  26-cycle  converters  are 
slightly  higher  than  the  60-cycle. 

Railroad  Work.  Twenty-five  cycles  has  been  recognized  as 
the  standard  frequency  for  railway  systems  in  this  country. 
Until  not  long  ago  all  systems  were  of  the  altemating-current- 
direct-current  type,  alternating  current  being  generated  and 
transmitted  to  the  various  substations,  where  it  was  changed  to 
direct  current  by  means  of  sjmchronous  converters.  The  choice 
of  this  frequency  was,  therefore,  chiefly  caused  uy  the  less  satis- 
factory operation  of  the  earlier  types  of  60-cycle  converters. 

Even  with  the  successful  operation  of  the  present  60-cycle 
converters,  there  is  no  reason  for  changing  the  standard  25-cycle 
frequency.  While  60  cycles  would  be  preferable  as  far  as  the 
generators  and  transformers  are  concerned,  this  is  ofiFset,  however, 
by  the  advantages  of  the  25-cycle  transmission  system  and  the 
lower  cost  of  synchronous  converters  for  larger  capacities.  Where^ 
the  supply  is  60  cycles,  synchronous  motor-generator  sets  are  very 
often  used  for  the  conversion. 

With  the  introduction  of  the  alternating-current  railway  motor, 
60  cycles  is  obviously  entirely  eliminated,  due  to  the  excessive 
impedence  drop  and  "  skin  effect "  caused  by  the  alternating; 
ciurent  flowing  in  the  rails.  The  25-cycle  system,  on  the  other 
hand,  is  fully  satisfactory  for  this  service,  and,  although  the  16-' 
cycle  sjrstem  has  been  advocated,  its  advantages  over  the  25-cycle 
system  have  not  been  proved  to  be  of  suflicient  weight  to  neces- 


Digitized  by 


Google 


280  ELECTRICAL  EQUIPMENT 

sitate  a  change  in  the  present  standard  frequency.  In  Europe, 
however,  a  few  single-phase  systems  are  using  this  frequency. 

Illumination,  Where  alternating  cmrent  is  used  for  lighting, 
the  60-cycle  frequency  is  generally  used.  No  arc  lamp  has  as 
yet  been  developed  that  will  operate  with  entire  satisfaction  on 
frequencies  of  less  than  40  cycles,  and  incandescent  lamps  cannot 
be  used  to  advantage  on  frequencies  of  less  than  30  cycles. 
Low-voltage  incandescent  lamps  show  no  flicker;  but  the  effect 
of  fatiguing  the  eye  is  noticeable  at  25  cycles,  especially  in  high- 
voltage  lamps. 

In  systems  where  lighting  predominates  a  60-cycle  frequency 
should,  therefore,  be  selected,  while,  if  most  of  the  energy  is  to  be 
used  for  power  purposes  the  condition  may  be  such  that  25  cycles 
would  prove  to  be  preferable,  in  which  case  frequency-changers 
can  be  provided  for  changing  the  current  required  for  lighting 
purposes  to  60  cycles. 

2.  SYNCHRONOUS  GENERATORS 

Alternating-current  generators  may  be  classified  into  two 
general  classes  according  to  their  general  characteristics:  S3rn- 
chronous  generators  and  Induction  generators.  The  former  type 
is  used  almost  entirely  while  the  latter  is  used  only  occasionaUy 
for  special  cases  as  explained  under  the  section  of  Induction 
Generators. 

The  generator  forms  one  of  the  most  important  parts  of  the 
equipment  in  a  hydro-electric  development  and  a  thorough  knowl- 
edge of  its  characteristics  and  design  is  of  the  utmost  importance. 
The  subject  will,  therefore,  be  treated  somewhat  more  in  detail 
than  would  at  first  seem  desirable. 

General  Description.  Most  alternating-current  generators 
are  of  the  revolving  field  type.  The  armature,  which  is  then 
stationary,  consists  of  a  laminated  iron  core  supported  by  a  cast- 
iron  frame,  the  inside  periphery  of  the  core  being  slotted  to  carry 
the  armature  winding.  Inside  the  stator  revolves  the  rotor  or 
revolving  field  system,  and  as  S3mchronous  generators  are  not 
self-exciting,  the  field  excitation  is  obtained  from  some  external 
direct-current  soiu'ce. 

Induced  E.M.F.  The  e.m.f's.  and  currents  are  alternating, 
i.e.,  have  one-half  wave  or  alternation,  first  positive  and  then 
negative,  for  each  pole  passed  by  a  given  armature  conductor. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  281 

A  cycle  is  a  complete  wave  of  two  alternations  and  the  frequency 
is  equal  to  the  product  of  the  number  of  pairs  of  poles  and  the 
speed  of  the  machine  in  revolutions  per  second;  it  is,  therefore, 
strictly  proportional  to  the  speed  of  the  machine. 

The  wave  shape  of  the  e.m.f.\  depends  on  the  distribution  of 
the  magnetic  flux  at  the  armature  surface,  and  the  total  e.m.f. 
is  the  sum  of  the  e.m.f .  waves  in  the  different  armature  conductors, 
added  in  the  proper  phase  relation.  The  instantaneous  values  of 
the  e.m.f.  and  current  are  constantly  changing  from  maximum 
positive  to  maximum  negative  and  the  specified  or  effective  value 
is  equal  to  the  square  root  of  the  average  value  of  the  square  of 
the  instantaneous  values.  For  a  true  sine  wave  shape  it  is  equal  to 
the  maximum  value  divided  by  V2. 

The  phase  relation  differs  symmetrically  for  pol3rphase  sys- 
tems. In  the  two-phase  system  the  terminal  voltages  of  the  two 
circuits  differ  in  phase  by  90  electrical  degrees,  Fig.  151,  and  in  the 


Fig.  151. — ^Two-phaae  Alternating  Current. 

three-phase  system,  the  terminal  voltages  of  the  three  circuits 
differ  in  phase  by  120  electrical  degrees,  Fig.  152.  The  terminal 
voltage  of  two-phase  generators  is  equal  to  the  e.m.f.  of  the  arma- 
ture circuits  and  the  line  current  equal  to  the  current  in  these  cir- 
cuits. For  three-phase  machines,  however,  the  armature  winding 
can  be  connected  either  Y  or  A,  which  will  be  discussed  more  fully 
later.  If  the  winding  is  Y-connected,  then  the  terminal  voltage 
is  equal  to  Vs  times  the  e.m.f.  per  armature  circuit  and  the  line 
current  equal  to  the  armature  current.  If  the  winding  is  A-con- 
nected,  then  the  terminal  voltage  is  equal  to  the  e.m.f.  per  circuit 
and  the  line  current  equal  to  Vs  times  the  current  in  the  armature 
circuit.    In  general,  when  speaking  of  current  and  voltage  in 


Digitized  by 


Google 


282  ELECTRICAL  EQUIPMENT 

in  a  three-phase  system,  under  current  the  Y-current  or  current 
per  line  and  under  voltage  the  A-voItage  or  voltage  between  lines 


Fig.  152. — ^Three-phase  Alternating  Current. 

wires  is  understood.    This  subject  is  covered  more  fully  in  the 
section  on  "  Armature  Connections." 

The  e.m.f.  induced  in  the  armature  circuit  is  determined  by 
the  following  formula: 

in  which    A?/ = wave  form  factor; 
fc,=slot  factor; 
ii0= winding  pitch  factor; 
/=  frequency  in  cycles  per  second; 
n=s  number  of  armature  conductors  connected  in  series 

per    phase    (twice    the   number   of   turns   per 

phase); 
0=flux  per  pole  in  maxwells. 

The  form  factor  of  an  e.m.f.  wave  is  defined  as  the  ratio 

5- ^    and  for  a  sine  wave  this  value  is  equal  to  1.11. 

average  voltage 

The  armature  winding  is  generally  distributed,  that  is,  the 
armature  conductors  are  placed  in  more  than  one  slot  per  pole  per 
phase.  The  principal  advantages  of  such  a  distribution  is  the 
closer  approximation  toward  a  sinusoidal  wave  form,  while,  on  the 
other  hand,  the  total  radiating  surface  of  the  coils  is  increased. 

With  a  distributed  winding  the  e.m.f.  will,  however,  be  some- 
what reduced  because  the  voltage  induced  in  the  conductors  in 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


283 


the  different  slots  are  somewhat  out  of  phase  with  one  another, 
and  for  this  reason  the  slot  factor  h,  for  which  the  values  are 
given  in  Table  XLI,  must  be  introduced  in  the  formula.  With 
two-layer  windings  the  value  of  k,  should  correspond  to  the  num- 
ber of  slots  per  layer  per  pole  per  phase  and  not  to  the  total  num- 
ber of  slots  per  pole  per  phase. 

TABLE  XLI 
Values  of  Slot  Factor  k. 


Slots  per  Pole 
per  Phase. 

Two-phase. 

Three-phase. 

1 
2 
3 
4 
5 
6 

1.000 
0.924 
0.911 
0.907 
0.904 
0.903 

1.000 
0.966 
0.960 
0.958 
0.957 
0.956 

The  windings  may  be  arranged  for  full  or  fractional  pitch.  In 
the  former  case  the  coil  spans  a  distance  exactly  equal  to  pole 
pitch  while  in  the  latter  case  it 
spans  a  lesser  distance.  Frac- 
tional pitch  windings  are  very 
generally  used,  the  advan- 
tages being  a  better  wave  and 
shorter  end  connections  of  the 
windings,  resulting  in  a  sav- 
ing of  annature  copper  be- 
sides making  the  machine 
shorter.  This  is  especially 
the  case  for  machines  with  a 
small  number  of  poles.  It  is 
evident  that  the  e.m.f's.  in- 
duced on  both  sides  of  the 
same  coil  are  not  exactly  in 

phase  with  each  other  in  a  fractional  pitch  winding,  so  that  a 
larger  flux  will  be  required  than  with  a  full-pitch  winding.  This 
is  allowed  for  in  the  voltage  formula  by  introducing  a  winding 


IM 

^ 

=^ 

■^ 

^Aon 

^/^ 

^ 

•g 

/ 

5oJio 

/ 

f 

/ 

/ 

60  70  80  90 

Per  cent  Wlodlnc  Pitch 


100 


FlQ.  153.- 


-Valuea  of  Winding-pitch 
Factor. 


Digitized  by  VjOOQ IC 


284  ELECTRICAL  EQUIPMENT 

pitch  factor,  K,  its  values  for  different  per  cent  pitch  being  given 
in  Fig.  153.    They  are  simply  based  on  the  formula: 


ia0»Bin 


ln(Xx(K.-), 


where  x  is  the  per  cent  pitch. 

For  single-phase  generators  the  annature  is  generally  wound 
similar  to  a  three-phase  machine,  one-phase  being  left  normally 
idle.  With  this  arrangement  the  slot  factors  ks  are  the  same  as 
given  for  three-phase  windings.  If  the  winding  is  furthermore 
distributed  as  with  purely  single-phase  generators,  when  it  covers 
considerably  more  than  two-thirds  of  the  annature  surface,  the 
values  of  these  slot  factors  should  be  reduced. 

The  flux  0,  obtained  from  the  previous  formula  is  that  which  is 
necessary  in  the  armature  for  inducing  the  required  e.m.f.,  i.e., 
the  useful  flux.  Due  to  the  leakage  between  the  poles  it  is,  how- 
ever, necessary  to  provide  a  greater  flux  in  the  field  poles  and  the 
yoke  to  compensate  for  this  leakage,  and  this  must  be  considered 
when  calculating  the  ampere  turns  of  the  field  winding.  This 
increased  flux  is  obtained  by  multiplying  the  useful  flux  by  a  leak- 
age coefficient.  The  average  values  for  this  factor  at  no  load, 
depending  on  the  diameter  per  pole,  may  be  obtained  from  Table 
XLII. 

TABLE  XLII 

Pole  Leakaqb  CosFncnBMTB 

Diameter  per  pole,  inches:    2  3  4  5  6  7  8 

Leakage  coefficients:  1.4      1.35      1.3      1.26      1.22      1.18      1.16 

Effect  of  Power  Factor  on  Operation.  Assuming  all  conditions 
except  the  load  constant,  the  terminal  voltage  of  an  alternating- 
current  generator  will  fall  as  the  load  increases.  This  is  due  to 
the  resistance  of  the  armature  conductors  and  the  synchronous 
reactance,  the  latter  combining  the  effects  of  the  annature  reac- 
tion and  the  armature  reactance  or  self-induction.  For  a  con- 
stant terminal  voltage  with  increased  load,  the  armature  resistance 
and  self-induction  require  an  increase  in  voltage  while  the  demag- 
netizing effect  requires  only  an  increase  in  the  magnetic  flux  to 
make  up  for  the  reduction  in  flux  caused  by  the  armatm^  current. 
The  latter  does  not  require  any  increase  in  the  generated  voltage 
since  the  action  is  conflned  to  the  magnetic  flux. 

The  drop  in  voltage,  due  to  the  armature  resistance,  requires 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


285 


Fio.  154. — ^Armature  Reaction. 
Current  in  Phase. 


no  explanation  beyond  the  statement  that  the  voltage  drop  is  iii 
phase  with  the  current  flowing. 

The  armature  reaction,  which  represents  the  resultant  e.m.f. 
of  the  armatiu^  currents,  de- 
pends on  the  current  and  the 
number  of  effective  turns  in 
series  per  pole  per  phase.  It 
noiay  have  a  magnetizing  or 
demagnetizing  effect,  or  it  may 
shift  the  field  flux  from  one 
side  of  the  pole  to  the  other, 
or  its  effect  may  be  a  com- 
bination of  the  two.  The 
energy  component  of  the  cur- 
rent will  only  cause  a  shifting 
or  distorting  effect,  while  the 
wattless  component  will  cause 
a  demagnetizing  or  magnetiz- 
ing effect,  depending  on 
whether  the  current  is  lag- 
ging or  leading.  These  effects 
are  illustrated  in  Figs.  154  to 
156. 

Rg.  154  represents  two 
conductors  of  an  armature 
coil.  These  are  midway  under 
a  north  and  south  pole,  re- 
spectively, and  the  e.m.f.  in- 
duced in  the  coil  is  obviously 
a  maximum  for  this  position. 
The  current  in  the  coil  will 
also  have  the  maximum  value 
as  it  is  in  phase  with  the  e.m.f. 
and  the  flux  produced  by  the 

same  will  have  a  crofis-magnetizing  effect  not  directly  opposing 
the  field  ampere-tiuns,  but  simply  causing  a  distortion  of  the  field 
flux.  The  current  in  the  armature,  however,  always  lags  behind 
the  induced  e.m.f.  by  reason  of  the  inductance,  and  even  with 
unity  power  factor  in  the  external  circuit,  the  armature  reaction  is 
demagnetizing  to  a  certain  extent. 


Fig.  155. — ^Armature  Reaction. 
Current  Lagging. 


Fio.  156.    Armature  Reaction. 
Current  Leading. 

•  Annatare  OoDdactoc 
0    In-pbase  GompoDeni 

of  Armatore  Carrent 
0   WatttoM  CompoDent 

of  Armature  OaxrcBit 


Digitized  by 


Google 


286  ELECTRICAL  EQUIPMENT 

In  the  position  shown  in  Fig.  155  the  e.m.f.  generated  in  the 
coil  has  a  value  somewhere  between  zero  and  maximum,  zero  cor- 
responding to  a  coil  position  midway  between  the  field  poles. 
The  armature  current,  which,  in  this  case,  is  lagging  somewhere 
between  0^  and  90^,  can  be  considered  as  made  up  of  two  com- 
ponents, an  in-phase  component  having  a  cross-magnetizing 
effect,  and  a  90^  lagging  component  having  a  demagnetizing 
effect.  At  zero  power  factor  the  wattless  armature  current, 
lagging  90^,  has  a  maximum  value,  and  consequently  the  greatest 
demagnetizing  effect. 

In  Fig.  156  the  current  is  leading  and  its  effect  is  just  opposite 
to  that  when  the  current  was  lagging.  It  is  thus  seen  that,  in  a 
generator,  the  field  is  weakened  by  a  lagging  current  and  strength- 
ened by  a  leading  current. 

The  armature  reaction  in  polyphase  generators  is  materially 
different  from  that  in  single-phase  machines.  In  the  former  its 
total  effect  combines  that  of  the  several  phases  and  hajs  a  constant 
value  provided  the  load  is  balanced.  If  unbalanced  it  will  be  of  a 
more  or  less  pulsating  nature  of  double  frequency,  as  is  always  the 
case  in  single-phase  generators. 

The  armature  self-induction  is  caused  by  the  leakage  flux 
which  is  set  up  by  the  armature  current  and  which  does  not  inter- 
link with  the  field  flux.  Since  the  armature  current  is  alternating, 
the  local  or  leakage  flux,  which  does  not  become  linked  with  the 
main  field,  will  be  continually  altering  in  magnitude  and  direction, 
so  that  there  is  set  up  a  self-induced  e.m.f.  proportional  to  the 
leakage  flux  of  each  phase  and  lagging  90^  behind  the  current. 
The  armature  leakage  is  usually  local,  and  thus  a  distributed 
winding  with  many  slots  will  have  a  smaller  leakage  inductance, 
since  the  flux  generated  by  each  imit  of  current  will  be  linked  with 
a  smaller  number  of  ampere  turns. 

The  exact  value  of  the  self-induction  of  an  armature  winding  is 
somewhat  difficult  to  determine,  its  magnitude  depending  upon 
the  reluctance  of  the  paths  taken  by  the  leakage  flux.  There  are, 
however,  several  methods  in  use  which  give  results  which  agree 
very  closely  with  those  afterwards  obtained  experimentally. 

If  L  is  the  self-induction,  expressed  in  henrys,  and  /  the  fre- 
quency, the  inductive  reactance  is  equal  to  2vfL,  It  is  of  the  same 
nature  as  resistance  and  is  expressed  in  reactance  ohms.  The 
counter  e.m.f.  caused  by  it  is  lagging  90^  behind  the  current,  and 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


287 


the  e.m.f.  which  it  consumes  and  which  has  to  be  impressed,  must 
thus  be  leading  90^  ahead  of  the  current.    This  is  illustrated  in 


Fig.  157.— E.m.f.  and  M.M.F.  Diagram.    Non-inductive  Load.] 


Fig.  158. — E.M.F.  and  M.M.F.  Diagram.    Lagging  Inductive  Load.] 
0 


Fig.  159.— E.M.F.  and  M.M.F.  Diagram.    Leading  Inductive  Load. 

the  diagram,  Fig.  157,  where  the  vector  XI  denotes  the  e.m.f. 
consumed  by  the  reactance  X. 


Digitized  by 


Google 


288  ELECTRICAL  EQUIPMENT 

The  vector  Et  represents  the  terminal  e.m.f.  and  /  the  current 
which  in  this  case  is  in  phase  with  the  terminal  e.m.f.,  the  load 
being  non-inductive.  The  e.m.f.  consumed  by  the  resistance  is 
equ^  to  RI,  in  phase  with  /,  and  Eg  is  the  e.m.f.  which  must  be 
induced  to  obtain  a  terminal  e.m.f.  Et  and  overcome  the  effects 
of  the  resistance  and  reactance,  thus  causing  a  current  to  flow. 

The  flux  required  to  produce  Eg  is  90^  ahead  of  this  e.m.f .,  the 
magneto-motive  force  or  ampere-turns  to  produce  the  same  being 
represented  by  Fg.  Due  to  the  demagnetizing  effect  of  the  arma- 
ture current,  t.6.,  the  armature  reaction,  the  vector  Fg  is  the 
resultant  of  the  m.m.f .  of  the  armature  current  Fa,  and  the  total 
impressed  m.m.f.  or  field  excitation  Fe.  The  m.m.f.  Fa  is  in 
phase  with  the  current,  and  after  having  determined  the  value  of 
Fg  and  Fa,  the  necessary  field  excitation  F«  is  obtained  by  com- 
pleting the  parallelogram. 

The  effect  of  a  lagging  inductive  load  is  shown  in  Fig.  158  and 
of  a  leading  inductive  load  in  Fig.  159.  For  the  same  terminal 
voltage  Et,  it  is  seen  that,  as  compared  with  a  non-inductive  load, 
a  much  higher  field  excitation  is  required  with  a  lagging  inductive 
load,  and  a  lower  field  excitation  with  a  leading  inductive  load. 
The  field  excitation  required  to  produce  the  terminal  voltage  Eg 
at  open-circuit  would  be  obviously  less  than  the  field  excitation 
with  non-inductive  load. 

Field  Excitation.  The  excitation  or  filed  ampere-turns 
required  to  produce  the  magnetic  flux  which  is  necessary  in  order 
to  induce  a  desired  e.m.f.  depends  on  the  character  of  the  mag- 
netic circuit,  i.e.  J  on  its  dimensions  and  on  the  material  of  which  it 
is  made  up.  The  values  are  readily  obtained  by  reference  to 
standard  saturation  curves,  similar  to  the  ones  shown  in  Fig.  160, 
these  curves,  of  com-se,  depending  upon  the  quaUties  of  the  iron 
or  steel  which  is  used.  The  total  magneto-motive  force  per  mag- 
netic circuit  is  equal  to  the  siun  of  the  m.m.f's.  necessary  for 
establishing  the  required  flux  in  the  separate  parts  of  the  circuit 
which  are  in  series;  viz.,  the  pole  pieces,  the  field  spider,  the  air 
gaps,  the  teeth  and  the  armature  core. 

The  relation  of  the  e.m.f.  produced  by  an  alternator  at  no-load, 
i.e.,  open  circuit,  to  the  field  current  when  the  alternator  is  driven 
at  constant  speed  is  represented  by  the  no-load  saturation  curve. 
Such  a  curve  is  shown  by  curve  A,  Fig.  161,  and  it  is  seen  that  this 
curve  is  almost  a  straight  Une  for  small  exciting  currents.    At  low 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


289 


excitation,  the  reluctance  of  the  air  gap  is  very  high  and  that  of 
the  iron  very  low,  and,  therefore,  the  former  may  be  considered 
as  constituting  the  entire  reluctance  of  the  magnetic  circuit. 
Since  the  reluctance  of  air  is  constant  regardless  of  the  flux  density, 
at  small  excitations  the  flux  will  be  proportional  to  the  mag- 
neto-motive force,  and,  therefore,  the  open  circuit  voltage  is  pro- 
portional to  the  field  current,  hence  the  curve  is  straight.  As 
the  field  becomes  stronger,  however,  the  proportion  of  the  air-gap 
reluctance  to  the  entire  reluctance  decreases  because  the  per- 
meability of  iron  decreases  with  increased  flux  density,  and, 
therefore,  the  e.m.f .  increases  less  rapidly  with  increased  excitation. 


ISO 
120 

— 

= 

ATiAealJ 

481 

pU 

Wg 

— 

■ 

-^ 

rwR 

.8tB 

Bl 

_  - 

— ■ 

— 

— 

t^O 

/ 

/^' 

^ 

— 

»-■ 

^90 

/ 

/^ 

^ 

/ 

§  eo 

/ 

/ 

. 

— 

' 

^80 
?  10 

f 

. 

^ 

SJIS 

a— 

1   •*' 

4A 

-^ 

^ 

y^ 

. 

For 
Am 

Alt 
RToiaia-^ 

988 

1A 

/ 

•^ 

0 

/ 

• 

0     10    »    10    40    60    60    70    80    00   100  110  120  180  140  160  160  170  180  190  200  210  Z» 
Ampere  Tama  per  Inch 

Fig.  160. — Saturation  Curves. 


It  was  pointed  out  in  the  previous  section  that  when  a  current 
is  flowing  in  the  armature  circuit,  i.e.,  under  load,  the  field  ampere- 
turns  required  to  maintain  normal  terminal  voltage,  exceed  the 
no-load  ampere-turns  required  for  normal  voltage,  due  to  the 
resistance  and  the  synchronous  reactance  of  the  armature  circuit. 
A  number  of  more  or  less  accurate  methods  have  been  proposed 
for  calculation  of  the  above  components,  and  thus  determining  the 
required  field  excitation  at  full  load.  IQiowing  the  resistance 
and  the  leakage  reactance  or  self-induction  of  the  armature,  the 
voltage  drop  caused  by  these  is  added  vectorically  to  the  terminal 
voltage,  this  giving  the  voltage  which  must  be  induced  (see  Figs. 


Digitized  by 


Google 


290 


ELECTRICAL  EQUIPMENT 


SIM 

•MM 

^^ 

"^ 

■■^ 

^^ 

:== 

A 

^ 

^ 

■jJAA 

y 

/ 

^ 

-^ 

S  91An 

/ 

/ 

/ 

< 

r 

A 

IHIM 

^7 

1 

< 

f 

900 

iUMl 

y 

/ 

Am 

y 

i 

•w 

Z 

J 

8      4      5      6      7 
Amperes  Field 


8     9     10 


Fig.  161. — Alternator  Characteristics. 


157  to  159).    Knowing  from  the  no-load  saturation  curve  the 
required  net  excitation  at  this  voltage,  and  correcting  it  for  the 

effect  of  the  armature  re- 
action, the  necessary  total 
field  ampere-turns  are  ob- 
tained. The  result  of  such 
calculations  for  different  loads 
and  power  factors  are  repre- 
sented by  the  load-character- 
istic curves.  Such  a  full-load 
characteristic  of  an  alternator 
is  shown  by  curve  J?,  Rg.  161. 
In  order  to  get  the  best 
combination  for  automatic 
voltage  r^ulation  an  alter- 
nator should  preferably  have 
a  range  in  excitation  from 
no-load  to  maximum  load,  with  approximately  80  per  cent 
power  factor,  of  the  ratio  of  not  more  than  one  to  two.  With 
125  volts  excitation,  the  voltage  should,  therefore,  not  be  allowed 
to  exceed  125  volts  at  maximum  load,  80  per  cent  power  factor, 
and  the  corresponding  no-load  excitation  should  be  about  70 
volts.  Should  the  excitation  voltage  be  250,  the  same  ratio  should 
hold  true. 

The  excitation  required  varies  considerably  for  different 
machines,  depending  upon  the  size,  the  number  of  poles,  the  speed 
and  the  regulation.  For  alternators  of  different  capacities,  but 
otherwise  similar,  the  relative  excitation  naturally  decreases  as 
the  size  of  the  alternator  increases.  High-speed  machines  gen- 
erally require  less  excitation  than  slow  speed,  due  to  the  smaUer 
number  of  poles.  With  a  large  number  of  poles,  however,  the  air 
gap  is  usually  smaller,  and  this  will  somewhat  offset  the  higher 
excitation  for  slow-speed  machines.  In  general,  it  may  be  said 
that  small  machines  with  many  poles  require  a  proportionally 
large  excitation  and  large  machines  with  few  poles  a  small  exci- 
tation. The  curves  given  in  Fig.  162  give  the  approximate  exci- 
tation required  by  water-wheel  driven  synchronous  generators. 
The  values  given  are  per  Kv.A.  per  R.P.M.  of  the  generator  capa- 
city, and  is  based  on  a  maximum  continuous  rating  at  80  per  cent 
power  factor. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


291 


r 

r 

186 

19ft 

JJSD 

^ 

11« 

lAA 

L^ 

x" 

y 

y^ 

1U» 

.X* 

^ 

WO 

88 
80 
72 
64 

^ 

^ 

1 

> 

^ 

16 

U 
18 
10 

X 

^ 

-^ 

•** 

^' 

y 

^* 

% 

ffi- 

ui 

^ 

^ 

^ 

-^ 

y^ 

i^ 

> 

X^ 

^ 

g 

^ 

[y 

/ 

/^ 

^' 

6 
4 

40 

^ 

^ 

/ 

t4 

3S 
S4 
16 

/^ 

^ 

/ 

% 

J 

/^ 

L 

' 

/ 

1 .' 

\  .' 

J 

1  •' 

1    1 

0  1 

2  1 

ri|«  1}8  210  2 

2~2|t  216  2j8  S|0 

8 

_j 

K 

[lA 

lA 

rfi 

£i 

bl. 

_ 

_ 

_ 

_ 

J 

0    4     8    12  16S024»82a640i4486256e06468  72  76  8084889296  100 

Fig.  162. — ^Approximate  Excitation  of  Water  Wheel-driven  Polyphase  Syn- 
chronous Generators.  Per  Kv.A.  per  R.P.M.  Based  on  Maximum 
Rating  at  80  per  cent  Power  Factor. 

Regulation.  The  regulation  of  an  alternating-current  gener- 
ator is  the  rise  in  voltage  when  a  specified  load  at  specified  power 
factor  is  reduced  to  zero;  the  speed  and  field  excitation  remaining 
constant.  It  is  expressed  in  per  cent  of  normal  rated-load  voltage, 
and  unless  otherwise  specified  understood  to  refer  to  a  non-induc- 
tive load. 

A  close  inherent  regulation  was  formerly  considered  one  of 
the  essential  requirements  of  a  good  generator,  but  fortunately 
this  idea  is  now  entirely  changed.  A  low  percentage  regulation 
may  be  obtained  in  two  ways;  first ,  by  designing  the  generator 
with  a  field  magnetically  strong  as  compared  with  the  armature, 
so  that  the  self-induction  and  demagnetizing  effect  of  the  armature 
is  comparatively  small,  resulting  in  a  small  increase  in  field  current 
required  to  maintain  normal  voltage  with  increase  in  load;  and, 
second,  by  saturating  the  magnetic  circuit,  particularly  in  the 
field,  where  high  densities  do  not  increase  the  losses  or  temperature 
rise.-  Both  of  these  methods  are,  however,  detrimental  to  the 
present-day  operating  practice.  A  decrease  in  the  synchronous 
reactance,  would  proportionally  increase  the  short-circuit  currents 


Digitized  by 


Google 


292  ELECTRICAL  EQUIPMENT 

of  the  machine  and  dangerously  increase  the  severe  mechanical 
strains  produced  by  the  same  on  the  apparatus,  as  these  increase 
with  the  square  of  the  current.  A  highly  satiu*ated  machinei 
on  the  other  hand,  is  detrimental  to  the  use  of  automatic  voltage 
regulators.  With  these  regulators  a  close  inherent  r^ulation 
machine  is  not  necessary  as  a  good  regulation  of  the  system  can, 
nevertheless,  be  maintained.  The  regulator  automatically  in- 
creases the  field  excitation  as  the  load  increases  and  thus  main- 
tains a  constant  terminal  voltage.  If  desired,  it  can  also  be 
adjusted  so  as  to  increase  the  voltage  with  the  load  and  comr 
pensate  for  the  line  drop. 

Modem  water-wheel-driven  alternators  have  a  regulation  at 
unity  power  factor  of  aroimd  20  to  25  per  cent.  This  is  con- 
sidered entirely  satisfactory  as  the  voltage  regulation  is  best  taken 
care  of  by  automatic  voltage  regulators. 

Short-circuit  Current.  In  speaking  about  the  short-circuit 
current  of  an  alternator,  distinction  must  be  made  between  the 
instantaneous  short-circuit  current  and  the  sustained  or  permanent 
short-circuit  current. 

The  sustained  short-circuit  current  of  an  alternator  is  limited 
by  the  armatiu^  resistance  and  reactance,  as  well  as  its  reaction 
on  the  field.    It  is  equal  to 

z: 

where  E  is  the  generated  e.m.f .  corresponding  to  the  field  excita- 
tion, and  Z,  the  "synchronous  impedance,"  representing  the 
combined  effect  of  the  above  three  factors.  This  formula,  there- 
fore, gives  the  value  of  the  sustained  short-circuit  current,  while  its 
instantaneous  value  wUl  be  very  much  higher.  This  is  due  to  the 
fact  that  in  the  first  instant,  when  the  generator  is  short-circuited, 
the  current  is  limited  only  by  the  resistance  and  self-induction  of 
the  armatiu^  circuit,  while  a  time  lag  of  sometimes  a  few  seconds 
takes  place  before  the  armature  reaction  becomes  effective.  The 
armature  resistance  and  reactance  are  thus  the  only  two  quan- 
tities that  limit  the  instantaneous  short-circuiting  current.  This 
limiting  effect  is,  however,  not  constant,  but  decreased  slightly 
with  high  short-circuiting  currents  due  to  their  saturation  of  the 
magnetic  field. 

Fig.  163  represents  an  oscillogram  of  a  typical  three-phase 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  293 

short  circuit,  the  generator  being  short  circuited  at  the  terminals 
of  the  armature  winding.  Comparing  the  currents  for  phases 
A  and  C,  it  is  noticed  that  the  latter  gives  an  approximately  sym- 
metrical relation  of  the  current  crests  with  respect  to  the  zero- 
axis,  while  in  the  former  case  the  wave  is  displaced  so  that  the 
maximnm  peak  of  the  initial  current  is  nearly  double  that  of  phase 
A,  the  actual  ratio  for  the  average  machine  being  about  1.8.  In 
calculating  the  instantaneous  short-circuit  current  which  may  occur 
under  the  worst  conditions,  an  unsymmetrical  current  wave  should, 


■W/WVWWwwwvwv//'sri 
/vWWWWWWWM^^vv^ 


Cnmniin 


^CmreDtia 
^  FImmO 


Fia.  163. — Oscillogram  of  Three-phase  Alternator  Short  Circuit. 

therefore,  be  considered  as  well  as  the  fact  that  the  short  circuit 
may  occur  when  the  generator  is  excited  for  full  load,  which  would 
mean  a  still  further  increase  of  say  10  per  cent  in  the  flux  and  in 
the  short-circuit  current.    Thus,  for  a  generator,  with  20  per  cent 

reactance,  the  maximum  peak  would  be  -^r^r-  or  five  times  the  normal 

ZU 

mean  effective  current  times  V2  times  2. 

The  sustained  short-circuit  current  is,  as  previously  stated, 
limited  by  the  synchronous  impedance,  or  less  exactly,  the  syn- 
chronous reactance,  of  the  generator,  and,  neglecting  saturation, 
it  is  directly  proportional  to  the  field  exciting  current.     Although 


Digitized  by 


Google 


294 


ELECTRICAL  EQUIPMENT 


synchronous  reactance  is  a  fictitious  quantity,  expressing  as  it 
does  in  a  single  quantity  both  the  armature  reaction  and  the 
armature  self-induction  or  reactance,  it,  nevertheless,  represents 
the  equivalent  of  a  true  reactance  and  may  be  expressed  in  ohms 
and  taken  just  as  any  other  reactance  in  determining  the  sus- 
tained short-circuit  current.  It  can  also  be  combined  in  the 
ordinary  way  with  any  external  reactance. 

The  per  cent  synchronous  reactance  is  determined  from  the 


Amperes  Field 
Fig.  163a. — Saturation  and  Synchronous  Impedance  Curves. 

saturation  and  synchronous  impedance  curves,  Fig.   163a,  as 
follows: 

In  order  to  produce  the  normal  current  In  a  field  current  F« 
is  required,  which  would  cause  an  open-circuit  voltage  e.  A  field 
current  Fb  would  produce  on  open  circuit  a  normal  voltage  E  if 
there  were  no  saturation.  Hence,  e  is  consumed  in  the  syn- 
chronous reactance  with  normal  current  flowing,  and  the  per 
cent  synchronous  reactance  is 

x.«|xioo=^xioo. 

This,  combined  with  the  per  cent  reactance  and  resistance  of  the 
external  circuit,  will  give  the  sustained  short-circuit  current  Ie, 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  295 

corresponding  to  the  field  current  Fg,  and  it  is  then  only  necessary 
to  increase  Is  in  the  ratio  of  the  actual  field  current  on  the  alter- 
nator at  the  time  of  short  circuit  to  Fb  That  is,  the  sustained 
short-circuit  current  at  load  excitation  Fi  is 

PB 

If  a  voltage  regulator  is  used,  the  generator  field  current  corre- 
sponding to  the  maximum  voltage  across  the  collector  rings  must 
be  taken  as  Fi. 

For  water-wheel-driven  alternators  the  sustained  short-circuit 
current  based  on  full-load  excitation  is  generally  from  two  to  three 
times  the  normal  full-load  current. 

When  a  short  circuit  takes  place  the  current  becomes  lagging 
and  its  effect  will  be  to  demagnetize  the  field  poles.  Assume,  for 
example,  a  generator  with  short-circuit  current  ratio  of  ten  times 
the  normal  full-load  current.  Then  tan  0  =  10  and  0=84.5^ 
Thus  cos  0  or  the  power  factor  under  short  circuit  is  equal  to  0.09. 
However,  it  requires  an  appreciable  time  to  reduce  the  magnetic 
flux  to  its  low  short-circuit  value,  since  it  is  surrounded  by  the 
field  coils,  which  act  as  a  short-circuited  secondary  opposing  a 
rapid  change  in  the  field  flux,  that  is,  in  the  moment  when  the 
short  circuit  starts  it  begins  to  demagnetize  the  field,  and  the 
magnetic  field  flux,  therefore,  begins  to  decrease.  In  decreasing, 
however,  it  generates  an  e.m.f.  in  the  field  coils,  which  opposes 
the  change  of  field  flux,  that  is,  increases  the  field  current  so  as  to 
momentarily  maintain  the  full  field  flux  against  the  armature 
reaction.  The  field  flux,  however,  gradually  decreases,  and 
also  the  field  current  which  increased  considerably  the  first 
moment.  This  is  clearly  illustrated  in  the  oscillograms  shown  in 
Pig.  163. 

Armature  Connections.  Synchronous  generators  may,  as 
previously  mentioned,  be  connected  either  single-phase,  two-phase 
or  three-phase.  Single-phase  machines  are  rarely  used,  and  when 
two-phase  machines  are  required  it  is,  as  a  rule,  in  connection 
with  some  existing  system.  Three-phase  machines,  on  the  other 
hand,  are  used  almost  exclusively,  due  to  the  many  advantages 
of  this  system  over  the  other  two. 

A  three-phase  cmrent  may  be  obtained  from  an  ordinary 
closed  coil  winding  by  making  connections  to  point  on  the  winding 


Digitized  by 


Google 


296 


ELECTRICAL  EQUIPMENT 


spaced  120^  apart  as  in  Fig.  164.  Such  a  method  is,  however, 
rarely  used  because  the  e.m.f.'8  of  the  sections,  which  are  com- 
bined with  each  other  to  form  one-phase  of  the  three-phase  cir- 
cuit, are  out  of  phase  with  each  other,  and  the  resultant  e.m.f. 


Fig.  164. 


Fio.  165. 


Fig.  166. 


Fig.  168. 


Fig.  167. 


and,  consequently,  the  capacity  of  the  machine  is  reduced,  simply 
because  the  most  effective  use  of  the  windings  is  not  obtained. 
The  highest  output  is,  however,  obtained  with  the  delta  and  star 
connections  where  groups  of  similar  phase  relations  are  connected 
in  series  or  paraUel  as  in  Figs.  165  to  168.    Of  these,  however,  the 


Digitized  by 


Google 


SYNCHRONOUS  GENERATOBS  297 

star  connection  is  preferred,  the  main  advantages  of  this  con- 
nection being: 

1.  It  is  possible  to  bring  out  a  lead  from  the  neutral  point, 
which  is  useful  for  various  purposes. 

2.  The  cost  is  less  than  with  delta  connection,  requiring 
approximately  only  58  per  cent  of  the  turns. 

3.  It  is  not  possible  for  circulating  ciurents  of  triple  frequency 
to  flow  in  the  windings. 

If  E  represents  the  effective  e.m.f.  of  each  group  and  /  the  lim- 
iting current  which  can  be  carried  by  the  same,  the  corresponding 
three-phase  capacities  of  the  various  arrangements  will  be 

Rg.  164:  3XV3£X/=6.196£/; 

Rg.  166:  3X2EXl=6EI; 

Fig.  166:  3X2IXE=6EI; 

Fig.  167:  2V3EXIXV3^6EI; 

Rg.  168:  V3^X2/XV3=6£/. 

For  two-phase  connections  the  capacities  are  the  same  for  the 
different  combinations  shown  in  Rgs.  169  to  172.  If  Ei  repre- 
sents the  e.m.f.  of  each  group  and  /  the  permissible  current  it 
equals  4EiL 

The  armature  winding  of  single-phase  generators  can  be 
arranged  either  for  purely  single-phase  duty  or  on  the  basis  of  the 
same  winding  being  used  both  for  polyphase  and  single-phase 
flervice,  the  latter  method  being  the  one  mostly  used.  When 
intended  for  three-  and  single-phase  service  any  one  of  the  con- 
nections shown  in  Rgs.  173  to  175  can  be  used,  although  the  star 
connection  in  Rg.  175  is  by  far  the  most  common.. 

The  single-phase  e.m.f's.  will  be  the  same  as  three-phase  with 
the  exception  of  the  arrangement  shown  in  Rg.  174,  where  the 
single-phase  connection  is  obtained  from  diametrically  opposite 
points  on  the  closed-coU  winding. 

The  comparative  capacities  of  the  machines  when  used  for 
smgle-phase  and  three-phase  service  should  obviously  be  based 
on  the  losses  and  heating  in  the  individual  armature  coils  or  group 
of  coils  and  not  on  the  total  armature  losses.  The  reason  for  this 
is  that  the  armature  loss  is  not  equally  divided  among  the  differ- 
ent groups  of  coils  and  the  heating  therein  will  consequently  be 


Digitized  by 


Google 


298 


ELECTRICAL  EQUIPMENT 

^ E- 


FiQ.  169. 


Fig.  170. 


Fig.  171. 


Fig.  172. 


Fig.  173. 


Fig.  175. 

Digitized  by  CjOOQ IC 


SYNCHRONOUS  GENERATORS  299 

higher  in  groups  carrying  the  highest  current.  When  a  poly- 
phase machine,  therefore,  is  loaded  single-phase  its  capacity  is  lim- 
ited by  the  current  which  any  individual  coil  can  carry,  and  this 
current  is  obviously  the  same  whether  polyphase  or  single-phase. 
With  the  connection,  as  shown  in  Fig.  173,  the  three-phase 
rating  is  the  same  as  in  Fig.  164;  viz.,  3xV§-BX/,  or  equal  to 
5.19&EL  The  corresponding  single-phase  rating  is  V3EX1.5I 
or  equal  to  2.598  E;  the  two  groups  of  the  winding  carrying  the 
limiting  current  /  while  the  other  four  groups  carry  a  ciurent  equal 

to  -,  the  total  current  thus  being  1.5/.    The  single-phase  capacity 

with  this  connection  is,  therefore,  equal  to  50  per  cent  of  the  cor- 
responding three-phase  rating. 

The  diametrical  connection  shown  in  Fig.  174  gives  a  much 
higher  rating  than  the  previous  one.  The  relative  capacities, 
however,  depend  on  whether  the  six  terminals  are  utilized  in 
connection  with  transformers  for  obtaining  three-phase  power. 
As  each  half  of  the  winding  can  carry  the  limiting  ciurent  /,  the 
total  current  is  equal  to  2/,  and  the  single  phase  rating  4EI,  the 
single-phase  e.m.f .  being  2E.  The  corresponding  six-phase  rating 
is  equal  to  &EI,  and  the  single-phase  rating  is,  therefore,  66.7 
per  cent  of  this  rating.  For  straight  three-phase  connection, 
however,  the  three-phase  rating  becomes  5.196-B/  and  in  this  case 
the  single-phase  capacity  is  77  per  cent  of  the  three-phase. 

With  star  connection  shown  in  Fig.  175,  two  of  the  phases 
carry  all  of  the  current  while  the  third  phase  is  idle  and  could  be 
omitted,  although  it  is  generally  added,  being  a  reserve  in  case 
of  accident  to  either  of  the  other  phases.  With  the  star  arrange- 
ment, as  shown,  two-thirds  of  the  winding  is  almost  in  phase  with 
the  single-phase  terminal  e.m.f.,  being  86.6  per  cent  effective,  and 
this  arrangement  is,  therefore,  about  15  per  cent  more  effective 
than  the  delta  connection  shown  in  Fig.  173. 

The  three-phase  rating  is  SEXixVs  or  equal  to  5.196S/, 
while  the  single-phase  rating  is  equal  to  3EI;  thus  57.7  per  cent 
of  the  three-phase  rating.  This  is  by  far  the  most  common  method 
of  connecting  armature  windings  for  single-phase  service. 

The  general  practice  in  building  single-phase  generators  is  to 
use  a  Y-wound  stator  and  give  it  a  rating  from  65  per  cent  to  70 
per  cent  of  the  three-phase  rating.  This  is  possible,  since  one- 
third  of  the  armatiu'e  slots  will  either  be  vacant  or  filled  with 


Digitized  by  VjOOQ IC 


300 


ELECTRICAL  EQUIPMENT 


Fig.  176. 


coils  in  which  no  current  is  flowing,  and  so  serve  to  cany  away  the 
heat  from  the  two-thirds  of  the  stator  in  which 
there  is  current. 

With  two-phase  alternators,  single-phase 
current  may  be  taken  off  from  two  of  the  termi- 
nals, and  assmning  the  same  limiting  current  / 
per  coil  and  a  coil  e.m.f.  Eu  we  get  the  single- 
phase  capacity  for  Fig.  169,  V2Ei  X2/,  or  equal 
to  2,828^i/  which  is  70.7  per  cent  of  the  cor- 
responding two-phase  rating  4JEiI. 

For  the  arrangements  shown  in  diagrams 
Figs.  170  to  172,  the  single-phase  rating  will  be 
equal  to  2JE^i/,  while  for  Figs.  176  and  177  it 
will  equal  2.828EiL 

A  comparison  of  the  two-phase  and  three-phase  capacities 
both  with  respect  to  each  other  and 
to  the  single-phase  ratings  obtained 
is  readily  made.  As  Ei  is  equal  to 
V2XEy  the  two-phase  ratings  4-Bi/, 
when  put  in  terms  of  three-phase, 
will  be  4Xy/2XEXl  or  equal  to 
5.656^7.  When  comparing  this  with 
the  ratings  obtained  from  the  various 
arrangements  given  on  page  297,  it  is 
seen  that  the  three-phase  closed-coil 
arrangement  gives  a  less  output  than 
for  two-phase,  while  the  other  three- 
phase  arrangements  give  an  increased 
rating. 

The  best  single-phase  rating  ob- 
tained from  a  three-phase  winding 
occurred  with  the  closed-coil  arrange- 
ment, Fig.  174  and  was  equal  to  4EL 
For  a  two-phase  winding,  on  the  other 
hand,  the  best  single-phase  rating  was 
shown  to  be  equal  to  2.828-Bi/.  As  Ei  is  equal  to  V2E,  this 
equals  2.828xV2XBX/,  or  4EI;  thus  the  same  as  with  closed- 
coil  winding,  shown  in  Fig.  174. 

The  above  capacities  have,  as  previously  stated,  only  refer- 
ence to  machines  which  can  be  adapted  to  both  polyphase  and 


Fig.  177. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  301 

single-phase  service.  For  machines  designed  for  purely  single- 
phase  duty,  the  ratings  can,  however,  be  somewhat  higher.  This 
is  due  to  the  fact  that  the  armature  winding  can  be  more  efficiently 
spaced  and  proportioned,  in  which  case  the  limit  in  output  as  a 
rule  is  determined  by  the  temperature  rise  in  the  field. 

Wave  Shape.  The  e.m.f.  in  a  conductor  is  proportional  to  the 
rate  of  cutting  the  lines  of  force,  and  has,  therefore,  a  wave  form 
of  the  same  shape  as  the  curve  of  flux  distribution.  Due  to  the 
non-uniform  flux  distribution  in  definite  pole  machines,  caused 
by  the  slots,  the  shapes  of  the  pole-pieces,  the  armature  reaction, 
etc.,  the  wave  will  never  have  a  perfect  sine  shape.  It  may,  how- 
ever, be  considered  as  the  resultant  of  a  number  of  sine  waves 
consisting  of  a  fimdamental  and  harmonics.  The  third  and  fifth 
harmonics  are  generally  predominating  in  three-phase  machines, 
while  even  harmonics  are  seldom 
found  in  the  e.m.f.  wave  of  an 

alternator.    This  is  due  to  the         j/^  \\  /^         PMiuTetnd 
fact  that  the  resultant  of  a  fimda- 
mental and  an  even  harmonic 
gives  an  unsymmetrical  ciurve,  as 

shown  in  Fig.  178,  where  the  re-   .,     ^^    -.  *  •    i  ri-  *   *  j 

-,      ^  .1  -         Pig.  178. — Unsymmetncal  Distorted 

sultant  curve  is  made  up  of  a  E.M.F.  Wave, 

fundamental  and  a  second  har- 
monic.   If,  therefore,  the  e.m.f.  wave  is  sjrmmetrical,  it  may  be 
assumed  that  no  even  harmonics  are  present. 

With  fractional  pitch-windings  certain  harmonics  are  elim- 
inated, depending  on  the  pitch.    For  example,  if  the  pitch  of  the 

coQ  can  be  shortened  by  -  of  the  pole  pitch,  then  the  nth  harmonic 
n 

and  its  multiples  wiU  be  eliminated. 

The  analyzation  of  a  wave  involves  a  considerable  amoimt  of 
work,  but,  in  general,  it  is  possible  to  tell  at  a  glance  which  har- 
monics are  predominating.  With  a  positive  third  harmonic,  that 
is,  if  counting  from  the  zero  point  of  the  complex  wave  the  har- 
monic wave  rises,  the  complex  wave  will  be  flat-topped.  If,  how- 
ever, the  harmonic  is  negative,  that  is,  if  after  crossing  the  base 
line,  it  rises  in  opposition  to  the  complex  wave,  its  effect  will  be  to 
produce  a  distorted  wave  of  the  peaked  type.  A  fifth  harmonic, 
however,  if  positive  will  give  rise  to  a  peaked  saw-toothed  wave, 
and  if  negative  to  a  flat-topped  wave. 


Digitized  by 


Google 


302 


ELECTRICAL  EQUIPMENT 


Complex  alternating  current  waves  as  mentioned  above  can 
be  represented  by  their  equivalent  sine  wave,  having  the  same 
efifect  as  the  complex  wave.  They  have  the  same  effective  value, 
that  is,  the  same  square  root  of  mean  square  of  the  instantaneous 
values  of  the  complex  wave.  Thus,  considering  all  complex 
alternating  currents  as  represented  by  equivalent  sine  waves,  all 
investigations  become  applicable  to  any  alternating  current  cir- 
cuit irrespective  of  the  wave  shape.  Terms  such  as  reactance, 
impedance,  etc.,  are  based  on  the  assumption  of  a  sine  wave  or 
equivalent  sine  wave. 

The  objections  to  higher  harmonics  are,  among  other  things, 
their  effect  in  increasing  the  maximum  value  of  the  e.m.f.  and 


JUnnoaie 


N«gAt]««Kk 


Fig.  179.— Symmetrical  Distorted  E.M.F.  Waves. 

the  correspondingly  increased  insulation  strain,  as  shown  by  the 
peaked  waves  in  Fig.  179.  In  certain  cases  the  triple  frequency 
voltage  estabhshed  by  the  generator  is  of  sufficient  value  to  cause 
heavy  triple  frequency  currents  to  circulate.  A  considerable 
distortion  of  wave  shape  might  also  affect  the  performance  of 
induction  or  synchronous  motors.  Here,  if  the  distortion  of  the 
voltage  wave  acting  at  the  motor  terminals  is  considerable,  the 
rotating  field  produced  will  be  more  or  less  of  a  pulsating  charac- 
ter. Induction  motors  might  operate  uneconomically  with  a  pos- 
sibility of  dead  points  in  the  starting  torque,  or  with  a  considerable 
coimter  torque  during  running.  Synchronous  motors  or  con- 
verters may  hunt,  or  even  fall  out  of  step.    Or  if  the  wave  shape 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  303 

of  the  induced  counter  electro-motive  force  greatly  differs  from  the 
pressure  wave  acting  at  the  terminals  of  a  synchronous  motor  or 
converter,  excessive  heating  might  result,  thus  lowering  the 
efficiency  of  the  system.  These  results  are,  of  course,  to  be 
expected  only  if  the  distortion  is  considerable,  and  for  this  reason 
it  has  become  a  general  practice  to  limit  the  maximum  permissible 
deviation  of  the  complex  wave  from  a  true  sine  wave  to  10  per 
cent.  This  deviation  is  to  be  determined  by  superimposing  upon 
the  actual  wave,  as  measured  by  an  oscillograph,  the  equivalent 
sine  wave  of  equal  length,  in  such  a  manner  as  to  give  the  least 
difference,  and  then  dividing  the  maximum  difference  between 
corresponding  ordinates  by  the  maximum  value  of  the  equivalent 
sine  wave. 

For  three-phase  machines  the  three  circuits  are,  as  previously 
stated,  connected  either  in  star  or  delta.  The  line  voltages  of 
the  three  phases  are  120^  apart  and  their  simi  must,  at  any  instant, 
be  zero.  Since  the  third  harmonics  are  in  phase  with  each  other, 
they  would  not  add  up  to  zero  and,  therefore,  cannot  exist,  and  for 
the  same  reason  a  third  harmonic  of  the  line  ciurent  cannot  be 
present.  In  a  balanced  system,  third  harmonics  can  exist  only 
in  the  voltage  from  line  to  neutral  or  Y-voltage,  and  in  the  current 
from  line  to  line  or  delta  ciurent,  as  will  be  explained  in  the  fol- 
lowing. 

Fig.  180  represents  a  delta-connected  three-phase  generator 
with  a  predominating  third  harmonic  e.m.f.  in  each  phase.  As 
the  three  triple  harmonics  are  in  phase,  the  machine  is  really 
running  imder  short  circuit,  as  far  as  the  triple  harmonic  is  con- 
cerned. This  triple-frequency  current  is  internal  in  the  windings, 
and  the  e.m.f.'s  which  causes  it  to  flow  are  short-circuited  in  the 
closed  delta,  and  will,  therefore,  not  appear  in  the  terminal  e.m.f.'s. 
The  circulating  current  may  be  of  great  magnitude,  entailing  large 
PR  losses  in  the  windings  with  corresponding  loss  of  efficiency. 

If  the  generator  is  Y-connected,  as  in  Fig.  181,  the  terminal 
e.m.f.  between  A  and  B  is  the  resultant  of  the  two  e.m.f.  vectors 
OA  and  OB,  thus  OA—OB,  the  negative  sign  of  the  latter  on 
accoimt  of  its  direction.  The  triple  harmonics  are  the  same  as 
in  the  previous  case,  but  by  adding  the  e.m.f.  waves  in  a  and  b, 
corresponding  to  OA  and  OB,  we  get  the  resultant  c.  OB,  that  is 
6,  must,  of  course,  be  reversed  and  the  triple  harmonics  will  cancel 
and  no  triple  harmonic  can,  therefore,  exist  in  the  terminal  e.m.f.. 


Digitized  by 


Google 


304 


ELECTRICAL  EQUIPMENT 


but  the  fundamental  e.m.f .  wave  is,  of  coursei  larger  than  in  each 
of  the  phases. 

If  the  neutral  is  grounded  the  potential  difference  from  line  to 
groimd  may  not  be  the  line  voltage  divided  by  VS;  but,  super- 
imposed on  this  voltage,  there  may  be  the  triple-frequency  e.m.f. 

A  B 


Fig.  180. 


Fig.  181. 


and  the  maximum  value  of  the  wave  may  be  greatly  increased, 
thus  increasing  the  insulation  strain. 

In  a  balanced  three-phase  system,  third  harmonics  can, 
therefore,  only  exist  in  the  voltage  from  line  to  neutral  or  Y-voltage; 
in  the  current  from  line  to  line,  or  generator  delta  current;  and 
in  the  line  current  only  if  the  generator  neutrals  are  groimded  or  a 
return  circuit  provided. 

Grounding  of  Generator  Neutral.  With  two  generators  oper- 
ating in  parallel,  a  difference  of  potential  will  exist  between  their 
neutrals  equal  to  the  vector  difference  between  their  phase  e.m.f  s. 
With  the  neutrals  interconnected  a  local  current  would  flow,  lim- 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


305 


ited  by  the  generator  impedance  at  triple  frequency;  while,  if 
the  triple-frequency  e.m.f.'s  in  the  two  generators  were  equal  and 
exactly  in  phase,  there  could  be  no  neutral  potential  or  current. 
Owing,  however,  to  the  difference  in  the  angular  velocity  of  the 
machines,  different  wave  forms,  different  excitation,  etc.,  this 
condition  never  exists;  and  a  triple-frequency  current,  therefore, 
always  flows  between  the  neutrals,  if  interconnected.    This  cur- 

JNo.1  |No.8  |No.8  |No.4 

V    ^K    i^x    >\. 

[ \  ^  i { 

iReflistanoe 


Fio.  182. — System  of  Grounding  Generator  Neutrals. 


|No.1  |No.8  ::No.8  ;|no.4 

.,.4^  -»4v  -^"x  >^ 


Fio.  183. — System  of  Grounding  Generator  Neutrals, 
can  be  grounded  at  one  time.) 


(Only  one  neutral 


rent  may  be  of  considerable  magnitude  with  low  reactance  ma^ 
chines,  and,  if  excessive,  precautions  must  be  taken  for  preventing 
it.  This  can  be  done  by  grounding  only  one  generator  at  a  time, 
if  the  generators  are  to  be  groimded,  leaving  the  neutrals  of  the 
other  machines  isolated.  Arrangements  must  then  also  neces- 
sarily be  made  so  that  any  one  of  the  generator  neutrals  can  be 
grounded,  as  shown  in  Figs.  182  and  183. 


Digitized  by 


Google 


306  ELECTRICAL  EQUIPMENT 

Whether  the  generator  neutral  should  be  grounded  or  not 
depends  on  the  operating  conditions.  If  an  uninterrupted 
service  is  the  most  essential  consideration,  the  system  should  not 
be  grounded,  while  if  it  is  more  desirable  to  limit  the  voltage 
strains,  imposed  by  grounds,  it  may  be  advisable  to  ground  the 
neutral,  thus  limiting  the  stress  to  the  Y-voltage.  Groimding  may 
also  be  advisable  where  selective  action  is  desired  on  a  number 
of  outgoing  feeders,  especially  imderground,  so  that  individual- 
feeders  may  be  disconnected  even  in  the  case  of  grounds. 

The  use  of  a  resistance  in  the  grounded  neutral  of  a  system 
offers  the  advantages  of  limiting  the  current  which  flows  through  a 
ground  on  one  phase,  and  thereby  eliminates  the  danger  of  mechan- 
ical destruction  due  to  the  excessive  ciurents  at  the  dead  short- 
circuit,  which  would  occur  with  a  ground  on  one  phase  of  a  S3rstein 
with  the  neutral  grounded  without  resistance.  Such  a  groimding 
resistance,  however,  abandons  the  advantage  of  the  dead  grounded 
system  that  the  voltage  between  Unes  and  ground  can  never  exceed 
the  Y  voltage.  It  is,  therefore,  not  permissible  where  the  appa- 
ratus cannot  safely  stand  the  delta  voltage  of  the  line.  This  is 
the  case  with  low-voltage  generators  feeding  a  Une  or  cable  through 
step-up  auto-transformers.  With  dead-grounded  generator  neu- 
tral the  voltage  between  generator  and  ground  is  fixed,  but  with 
resistance  in  the  neutral  ground,  a  dead  ground  on  the  high- 
potential  phase  puts  nearly  delta  voltage  of  the  high-potential 
circuit  on  the  low-potential  generator,  and  thereby  seriously 
endangers  it,  if  the  step-up  ratio  of  the  auto-transformer  is  1 :2. 
If  it  is  higher  there  is  every  likelihood  that  the  generator  will  be 
destroyed.  The  same  is  the  case  when  connecting  together  trans- 
mission systems  of  diflferent  voltages  through  auto-transformers. 
In  any  case,  if  auto-transformers  are  not  considered  safe,  trans- 
formers must  be  used. 

A  grounding  resistance  should  have  a  value  high  enough  to 
limit  the  neutral  current,  but  still  low  enough  to  insure  that,  if  a 
ground  occurs  in  one  phase,  it  will  permit  a  sufficiently  large  cur- 
rent to  flow  in  the  neutral  to  open  the  protective  circuit-breakers. 
Non-inductive  resistances  are  preferable  to  reactances,  since  they 
eliminate  the  danger  of  high-frequency  oscillations  beween  lines 
and  ground  through  the  generator  reactance  in  the  path  of  the  third 
harmonic,  by  damping  the  oscillation  in  resistance.  The  ground- 
ing of  the   neutral   through   a   reactance    may,    therefore,   be 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  307 

very  dangerous,  owing  to  the  poesibilities  of  a  resonance  voltage 
rise. 

If  three  auto-transformers,  Y-connected  with  the  neutral 
grounded  as  the  only  ground,  are  used  to  step  up  the  generator 
voltage,  abnormal  potentials  to  ground  may  result,  due  to  the 
presence  of  high  harmonics.  The  distortion  does  not  appear  in 
the  voltage  between  the  lines  because  the  distortion  between  one 
line  and  the  neutral  is  canceled  by  that  of  the  other  two  lines. 
The  voltage  distortion  may  be  eliminated  by  providing  a  path 
for  the  triple-frequency  exciting  current  which  is  required  for  the 
magnetization  of  the  transformer.  This  is  done  by  connecting 
the  transformer  neutral  to  the  generator  neutral. 

Rating.  Synchronous  generators  should  be  rated  by  the 
electrical  output,  and  this  should  be  expressed  in  kilo-volt- 
amperes  (Kv.A.)  and  not  in  kilowatts  (Kw.)  \mless  the  power 
factor  of  the  load  is  also  given.  Preferably  both  should  be  given, 
80  as  to  avoid  any  misunderstanding  whether  Kv.A.  or  Kw.  is 
meant,  for  example  2000  Kv.A.  (1600  Kw.-.8  P.F.). 

Most  water-wheel-driven  generators  are  now  given  a  maximiun 
continuous  rating,  without  any  overload  provision,  except  that 
they  must  be  able  to  carry  momentary  loads  of  150  per  cent  of 
the  amperes  corresponding  to  the  continuous  rating,  keeping  the 
rheostat  set  for  load  excitation. 

The  rated  full-load  current  is  that  ciurent  which,  with  rated 
voltage,  gives  the  rated  kilowatts  or  rated  kilo-volt-am'peres. 
In  machines  in  which  the  rated  voltage  differs  from  the  no-load 
voltage,  the  rated  current  should  refer  to  the  former.  The  rated 
output  may  be  determined  as  follows: 

If  B=full  load  terminal  voltage  and  7= rated  current,  then 

EI 
for  a  single-phase  generator  Kv.A.  =t7w^q- 

For  a  two-phase  generator  the  total  output  is  equal  to  the  out- 
put of  the  two  single-phase  circuits,  and  if  7,  in  this  case,  is  the 
rated  current  per  circuit,  the  output  for  a  two-phase  generator  is 

TT    A      2B7 

^^•^=iooo- 

For  a  three-phase  generator  there  are  three  circuits  to  be  con- 
sidered, whether  the  machine  is  star  or  delta  connected.  If  £  is 
the  terminal  voltage  and  7  the  Une  current,  then  for  a  three-phase 

,     rr    A  3EI         V3EI 

generator  ^^'^'-^3X1000^1660' 


Digitized  by 


Google 


308  ELECTRICAL  EQUIPMENT 

The  rating  of  a  generator  is  usually  determined  by  its  permis- 
sible temperature  rise  caused  by  the  current.  This  rise  neces- 
sarily increases  with  increasing  load  and  also  with  decreasing 
power-factor.  Thus,  for  a  given  Kv.A.  output,  the  total  heat 
losses  are  larger  for  low  than  for  high  power  factors,  the  differ- 
ence being  due  to  the  heat  generated  by  the  increased  field  cur- 
rent which  is  required  to  overcome  the  armature  reaction  and 
maintain  the  given  current  and  terminal  voltage. 

Alternating-current  generators  are  generally  designed  to  oper- 
ate a  normal  load  and  80  per  cent  power  factor  without  ex- 
ceeding a  specified  temperature  rise;  and  should  such  a  machine 
have  to  be  operated  with  a  load  having  a  lower  power  factor,  its 
rating  will  be  reduced  when  based  on  the  same  temperatine  guar- 
antee. The  true  operating  power  factor  should,  therefore,  be 
carefully  considered  in  selecting  the  capacity  of  the  generating 
units.  The  power  factor  depends  not  only  on  the  t3rpe  of  appa- 
ratus comprising  the  load,  but  also  on  the  load  factor  at  which 
they  are  operated. 

To  obtain  the  total  Kv.A.  capacity  of  a  system,  the  sum  of 
the  wattless  components  of  the  different  loads  should  be  calcu- 
lated, the  efficiency,  power  factor  and  load  factor  being  duly 
considered.    The  total  capacity  is  then  equal  in  Ev.A.  to 

V(Total  Kw.  energy)2+ (Total  Kv.A.  wattless)^, 
and  the  combined  power  factor  of  the  load 

Total  Kw.  energy 
Total  Kv.A.     • 

It  is  obvious  that  a  generator  must  not  be  permitted  to  be 
operated  under  such  conditions  that  it  will  attain  such  excessive 
temperatm^  which  will  cause  the  insulation  employed  in  their 
construction  to  deteriorate,  and  the  A.I.E.E.  Standardization 
Rules  contains  the  following  table,  giving  the  highest  tempera- 
tures and  temperatiu^  rises  to  which  various  classes  of  in«iilfif.ing 
materials  may  be  subjected.  While  it  was  recognized  that  the 
manufacturers  could  successfully  employ  class  B  insulation  at 
150^  C.  and  even  higher,  it  was  not  felt  that  sufficient  data  was 
available  to  recommend  this  and  the  institute  adopted  125^  C. 
as  a  conservative  limit  for  this  class  of  insulation,  any  increase 
above  this  figure  being  considered  a  special  guarantee.   The  ambient 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


309 


temperature  of  reference,  that  is,  the  cooling  air  surrounding 
the  machine  is  given  as  40^  C,  and  by  deducting  this  ambient 
temperature  from  the  maximum  permissible  temperature  given 
above,  the  permissible  temperature  rise  is  obtained.  As  it  is 
usually  impossible  to  determine  the  maximum  temperature 
(hottest  spot)  attained  in  insulated  windings,  a  correction  factor 
must  be  applied  to  the  observable  temperature,  so  as  to  approx- 
imate the  difference  between  the  actual  maximum  temperature 
and  the  observable  temperatiu^  by  the  method  used.  This 
correction,  or  margin  of  security,  is  provided  to  cover  the  errors 
due  to  fallibility  in  the  location  of  the  measuring  devices,  as  well 
as  inherent  inaccuracies  in  measurement  and  methods. 

TABLE  XLIII 

Permibsibub  Temferatcrbs  and  Temperature  Rises  for  iNauLATiNG 

Matekialb 


1 

2 

CUn. 

Description  of  Material. 

Maximum 
Temperature 
to  which  the 
material  may 
be  subjected. 

Maximum 

Temperature 

Rise. 

A. 

Cotton,  silk,  paper  and  similar  mate- 
rials, when  so  treated  or  impreg- 
nated as  to  increase  the  thermal 
limit,   or  when  permanently  im- 
mersed in  oil;  also  enameled  wire*. 

105^  C. 

66**  C. 

B. 

Mica,  asbestos  and  other  materials 
capable  of  resisting  high  tempera- 
tw^,  in  which  any  Class  A  material 
or  binder  is  used  for  structural  pur- 
poses only,  and  may  be  destroyed 
without  impairing  the  insulating  or 
mechanical  qualities  of  the  insulation. 

126«C. 

86*»C. 

C. 

Fireproof  and  refractory  materials, 
such   as   pure   mica,   porcelain, 
quartz,  etc 

No  limits  ftHAnifiAd 

*  For  cotton,  silk,  paper  and  similar  materials,  when  neither  impregnated  nor 
immersed  in  oil,  the  highest  temperatures  and  temperature  rises  shall  be  10*>  C.  below 
the  UmlU  fixed  for  Class  A. 


Digitized  by 


Google 


310 


ELECTRICAL  EQUIPMENT 


There  are  three  different  methods  provided  for  determining 
the  temperature  of  different  parts  of  a  machine.  These  will  be 
briefly  described  in  the  following  and  the  respective  permissible 
temperature  rises  given,  based  on  class  A  insulation. 

I.  Thermometer  Method.  This  consists  in  appl3ring  a  thermom- 
eter to  the  hottest  accessible  part  of  the  completed  machine. 
With  this  method  a  correction  of  15°  C.  must  be  made, 
that  is,  the  permissible  observable  temperature  rise  as 
read  by  the  thermometer  cannot  exceed  60°  C. 

An  exception  to  this  rule  is  the  case,  when  thermometers  are 
appUed  directly  to  the  surfaces  of  bare  windings,  as  the  field  coils. 
Then  only  a  5°  C.  correction  has  to  be  made,  so  that  the  permis- 
sible observable  temperature  rise  is  limited  to  60°  C. 

II.  Resistance  Method,  This  consists  in  the  measurement  of  the 
temperature  of  windings  by  their  increase  in  resistance,  cor- 
rected to  the  instant  of  shut-down  when  necessary.  In  the 
application  of  this  method,  careful  thermometer  measure- 
ments should  also  be  made,  whenever  practicable,  in  order 
to  increase  the  probabiUty  of  revealing  the  highest  observ- 
ble  temperature.  Whichever  measurement  yields  the 
higher  temperature  that  temperature  shall  be  taken  as 
the  *'  highest  observable "  temperature  and  a  hottest- 
spot  correction  of  10°  C.  added  thereto.  The  permissible 
temperature  rise  with  this  method  is,  therefore,  65°  C. 

TABLE  XLIV 
Temperature  Coefficiemtb  of  Copper  Resistance 


Temperature  of  the  Winding, 

in  <>  C.  at  which  the  Initial 

Resistance  is  Measured. 

Increase  in  resistance  of 

Copper  per  «>  C.  per  Ohm  of 

Initial  Resistance. 

0 

0.00427 

5 

0.00418 

10 

0.00409 

15 

0.00401 

20 

0.00393 

25 

0.00385 

30 

0.00378 

35 

0.00371 

40 

0.00364 

Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


311 


The  temperature  coefficient  of  copper  may  be  deducted  from 
1 


the  formula 


-.    Thus,  at  an  initial  temperature 


(234.5+0* 

i=40°  C,  the  temperature  coefficient  of  increase  in  re- 
sistance per  degree  centigrade  rise,  is  =0.00364. 

Table  XLIV  deduced  from  the  formula,  is  given  for  con- 
venience of  reference. 

III.  Efkbedded  Temperature — Detector  Method.  This  method  con- 
sists in  the  use  of  thermo-couples  or  resistance  tempera- 
ture detectors,  located  as  nearly  as  possible  at  the  esti- 
mated hottest  spot.     When  this  method  is  used,  it  shall. 


Double-layer  Winding.  Single-layer  Winding. 

Fig.  184. — Methods  of  Locating  Temperature  Detectors. 

when  required,  be  checked  by  Method  II;  the  hottest  spot 
shall  then  be  taken  to  be  the  highest  value  by  either 
method,  the  required  correction  factors  being  applied  in 
each  case.  Temperature  detectors  should  be  placed  in  at 
least  two  sets  of  location,  as  shown  in  Fig.  184. 
The  corrections  to  be  added  to  the  **  observable  "  tempera- 
ture when  Method  III  is  used,  are  as  follows:  In  the  case 
of  two-layer  windings,  with  detectors  between  coil  sides, 
and  between  coil  side  and  core,  add  5**  C.  to  the  highest 


Digitized  by 


Google 


312  ELECTRICAL  EQUIPMENT 

reading.  In  single-layer  windings,  with  detectors  be- 
tween coil  side  and  core  and  between  coil  side  and  wedge, 
add  to  the  highest  reading  lO""  C.  plus  l""  C.  per  1000  volts 
above  5000  volts  of  terminal  pressure. 

Thus,  for  a  three-phase  machine  with  an  11,000-volt  single- 
layer  winding,  the  correction  to  be  added  to  the  maximum  "  ob- 
servable "  temperature  in  estimating  the  "  hottestnspot "  tem- 
perature, is  16°  C,  and  the  permissible  temperature  rise  is, 
therefore,  49**  C.  For  double-layer  windings  the  permissible 
rise  is  60°  C.  and  for  single-layer  windings  for  5000  volts  or  less 
55°  C. 

Increased  altitude  has  the  effect  of  increasing  the  temperature 
rise  of  some  types  of  machinery.  In  the  absence  of  information 
in  regard  to  the  height  above  sea  level  at  which  the  machine  is 
intended  to  work  in  ordinary  service,  this  height  is  assumed  not  to 
exceed  1000  meters  (3300  feet).  For  machinery  operating  at  an 
altitude  of  1000  meters  or  less,  a  test  at  any  altitude  less  than 
1000  meters  is  satisfactory,  and  no  correction  shaU  be  applied  to 
the  observed  temperature.  Machines  intended  for  operation 
at  higher  altitudes  shall  be  regarded  as  special,  and  when  a  ma- 
chine is  intended  for  service  at  altitudes  above  1000  mecers  (3300 
feet)  the  permissible  temperature  rise  at  sea  level  shall  be  reduced 
by  1  per  cent  for  each  100  meters  (330  feet)  by  which  the  altitude 
exceeds  1000  meters. 

Efficiency.  The  efficiency  of  a  generator  is  the  ratio  of  the 
kilowatt  output  to  the  kilowatt  input  at  the  rated  Kv.A.  and 
power  factor.  The  difference  between  these  two  quantities  is 
equal  to  the  losses.  The  method  commonly  and  most  readily 
used  for  obtaining  the  efficiency  is  to  determine  these  losses  and 
then  compute  the  efficiency  by  dividing  the  power  output  by  the 
sum  of  the  power  output  plus  the  losses. 

The  guaranteed  efficiency  should  always  refer  to  the  energy 
load  and  it  is  most  important  that  the  power  factor  of  the  load  is 
also  given.  In  certain  cases  the  guaranteed  efficiency  is  based  on  a 
Kv.A.  output,  but  the  inconsistency  of  such  a  method  is  apparent, 
as  the  following  example  will  illustrate: 

Assume  a  generator  rated  100  Kv.A.  (100  Kw.  1.0  P.F.)  or 
100  Kv.A.  (80  Kw.  .8  P.F.),  and  that  the  losses  at  unity  and  80 
per  cent  power  factors  are  10  and  11  Kw.  respectively,  the  effi- 
ciency is  then: 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  313 

Based  on  100  Kw.  1.0  P.F. 

100 
Eflf.^^QQ      Q«91  per  cent. 

Based  on  80  Ew.  .8  P.F. 

80 
Eff.  =  g^q:^=88  per  cent. 

Based  on  100  Kv.A.  .8  P.F. 

^'"  100+11  "^  per  cent. 

From  the  last  two  values  it  is  seen  that  for  80  per  cent  power 
factor  if  based  on  the  Kv.A.,  a  2  per  cent  greater  efficiency  guar- 
antee can  be  made,  although  this  value  has  no  meaning,  as  it  is 
based  on  apparent  power. 

It  is,  of  course,  equally  important  that  all  the  losses  are  included 
and  that  they  are  figured  on  the  same  basis,  in  order  that  a  fair 
comparison  may  be  made  of  the  efficiencies  guaranteed  by  differ- 
ent manufacturers. 

The  A.I.E.E.  Standardization  rules  require  that  for  syn- 
chronous generators  the  following  losses  are  included  in  deter- 
mining the  efficiency:  (1)  core  losses,  (2)  PR  loss  in  all  windings 
based  upon  rated  Kv.A.  and  power  factor,  (3)  stray  load  losses, 
(4)  friction  of  bearings  and  windage,  (5)  rheostat  losses  corre- 
sponding to  rated  Kv.A.  and  power  factor. 

Bearing  Friction  and  Windage  may  be  determined  as  follows: 
Drive  the  machine  from  an  independent  motor,  the  output  of 
which  shaD  be  suitably  determined.  The  machine  under  test 
shall  not  be  excited.  This  output  represents  the  bearing  friction 
and  windage  of  the  machine  imder  test. 

Core  Loss.  Follow  the  above  test  with  an  additional  reading 
taken  with  the  machine  separately  excited  so  as  to  produce  at 
the  terminals  a  voltage  corresponding  to  the  calculated  internal 
voltage  for  the  load  under  .consideration.  The  difference  between 
the  output  obtained  by  this  test  and  that  obtained  by  the 
previous  one  shall  be  taken  as  the  core  loss,  neglecting  the  brush 
friction.  The  internal  voltage  shall  be  determined  by  correcting 
the  terminal  voltage  for  the  resistance  drop  only. 

PBLoss  may  be  calculated  directly  from  the  resistance  meas- 


Digitized  by 


Google 


314  ELECTRICAL  EQUIPMENT 

urement,  the  current  being  based  on  the  rated  Kv.A.  and  power 
factor.  The  resistance  of  the  windings  should  be  taken  at  75®  C, 
or  the  values  corrected  for  this  temperature.  It  is  important  that 
this  is  followed. 

Stray  Load  Losses.  These  include  iron  losses,  and  eddy- 
current  losses  in  the  copper,  due  to  fluxes  varying  with  load  and 
also  to  saturation. 

Stray  load  losses  are  determined  by  operating  the  machine  on 
short  circuit  and  at  rated-load  current.  This,  after  deducting 
the  windage  and  friction  and  PR  loss,  gives  the  stray  load  loss 
for  polyphase  generators.  For  single-phase  generators  they  are 
much  larger. 

Field-Rheostat  Losses  shall  be  included  in  the  generator  losses 
where  there  is  a  field  rheostat  in  series  with  the  field  magnets  of 
the  generator,  even  wheYi  the  machine  is  separately  excited. 

In  making  eflSciency  tests  after  installation  in  the  power 
station,  it  may  occasionally  be  possible  to  drive  the  unit  by  its 
exciter  when  the  same  is  direct  connected;  but  for  large  units 
and  when  the  direct-connected  exciters  are  not  provided  the 
retardation  or  deceleration  testing  method  is  resorted  to.  This 
test  is  based  on  the  principle  that  every  moving  body  possesses 
a  certain  definite  amount  of  energy,  due  to  its  motion.  It  is 
described  in  detail  in  an  article  by  Mr.  R.  Treat  in  the  General 
Electric  Review  for  June,  1916. 

A  convenient  and  most  satisfactory  method  of  detennining 
the  efficiency  of  a  generator  after  installation  may  be  employed 
where  there  are  two  or  more  units  in  the  power  house  available 
for  the  use  of  the  test,  or  where  the  unit  under  test  may  be  varied 
in  conjunction  with  some  other  imit  of  sufficient  size  located 
elsewhere  in  the  system  but  which  may  be  segregated  for  the 
purpose.  The  method  for  determining  the  core  losses  and  the 
friction  windage  losses  consists  in  operating  the  generator  as 
a  synchronous  motor  and  measurmg  the  input  by  watt- 
meters. 

When  the  retardation  method  of  testing  is  used,  it  is  to  be 
recommended,  if  possible,  to  check  such  tests  by  means  of  the 
input  method. 

A  new  method  of  artificially  loading  generators  for  tests  in 
hydro-electric  power  stations  is  described  in  an  article  in  the 
General  Electric  Review  for  April,  1917. 


Digitized  by 


Googk 


SYNCHRONOUS  GENERATORS 


315 


# 

96 
97 

It-T-i 

iod-1  iO-P« 

P.— 

— 



_F« 

ITm 

L<L0.8P,, 

p. 

&9b 

^ 

^ 

.^ 

■""■ 

=£S 

1^ 

y 

y^ 

^ 

7f>^ 

Cd  94 

h 

/ 

-y 

91 

90 

0      1 


3      4      5      6      7      8      9     10     11    12     18    U    15    16     17     18 
Generator  Capacity  la  Thousand  Kv.A. 


Fig.  185. — Approximate  Efficiencies  of  Polyphase  Water-wheel-driven 

Alternators. 

The  curves  in  Fig.  185  represent  approximate  efficiencies  of 
polyphase  water-driven  alternators,  the  ratings  being  maximum 
continuous. 

Speed.  The  speed  of  water-wheel-driven  generators  is  deter- 
mined by  the  frequency  of  the  system  and  by  the  hydraulic  con- 
dition, that  is  the  speed  of  the  wheel,  which,  in  tiun,  is  governed 
by  the  size  of  the  unit  and  the  head. 

With  a  fixed  frequency  the  number  of  poles  must  be  increased 
in  inverse  proportion  to  a  reduction  in  the  speed.  To  accommo- 
date this  increased  number  of  poles  the  diameter  must  necessarily 
be  larger  and  with  this  follows  also  an  increased  amount  of  material 
and  labor.  The  cost  of  slow-speed  machines  must,  therefore, 
necessarfly  be  much  higher  than  for  machines  of  higher  speeds. 

Table  XLV  shows  what  has  actually  been  the  practice  in  re- 
gard to  the  speed  of  hydro-electric  units.  This  table  covers  a 
number  of  years'  manufacture  of  wheels  and  generators,  and  can 
hardly  be  said  to  represent  latest  practice,  and  future  speeds  may 
be  considerably  higher  than  these,  particularly  in  the  smaller  units 
on  the  higher  heads.  It  is  seen  that  the  speeds  range  from  as  low 
as  55  R.P.M.  to  as  high  as  600  R.P.M.,  these  figures,  of  course 
referring  to  direct-connected  units. 


Digitized  by 


Google 


316 


ELECTRICAL  EQUIPMENT 


TABLE 
Actual  Speeds  op  Water- 


H«iUl 
In 

Kv-A 

.  Capacity  oi 

CSNERATO&. 

200 

300 

400 

500 

600 

700 

800 

000 

1000 

1250 

1500 

1750 

2000 

2500 

3000 

3500 

4000 

4500 

5000 

5500 

9000 

15 

100 
180 

20 

ICKl 
171 

m 

■I2b 
225 

120 

25 

120 
112 
133 

80 

35 

180 
157 

97 

225 

164 
180 

40 

225 

45 

164 
116 

164 

50 

55 

00 

65 

, 

257 

225 
225 

70 

.... 

150 

75 

iio 

200 

180 

85 

100 

375 
514 

300 

225 
300 

225 

125 

.... 

150 

340 

175 

400 

200 

J57 

400 

375 

375 

... 

225 

250 

164 

800 

450 

200 

514 

350 

400 

ibn 

:i00 

300 

450 

500 

400 

600 

375 

450 
600 

m 

360 

514 

700 

300 

800 

300 

900 

250 

1000 

400 

1100 

1200 

600 

450 

1800 

1400 

1500 

600 

420 

1600 

1700 

1800 

1900 

2000 

600 

Voltage.  Standard  generator  voltages  for  all  frequencies 
are  240,  480,  600,  2300,  4000,  6600,  with  the  correspondiDg 
motor  voltages  220,  440,  550,  2200,  6000.  There  is  no  motor 
voltage  corresponding  to  4000  volts,  since  this  is  only  used 
on  three-phase,  four-wire  lighting  distributing  systems.    In  addi- 


Digitized  by  CjOOQ IC 


SYNCHRONOUS  GENERATORS 


317 


XLV 

WHEEL-DRIVEN  GENERATORS. 


KTA.  CAPAcrrr  or  Ginbrator. 

Head 

ID 

Feet. 

6500 

7000 

7500 

8000 

8500 

9000 

ftSOO 

10000 

IIGOO 

12000 

125O0 

13000 

UOD( 

mm 

16(KK) 

1 

iriMX) 

15 

20 

25 

57.7 

55.4 

30 

35 

40 

45 

M 

94 

116 

50 

56 

60 

100 

65 

70 

75 

144 

85 

100 

250 

125 

250 

180 
250 

225 

150 

187 

250 

175 

187 
500 
420 

200 

428 



225 

360 

250 

300 

375 

350 

400 

400 

500 

400 
514 
600 

450 

400 

500 

600 

700 

800 

200 
315 

900 

300 

1000 

1100 

300 

1200 

300 
360 

1300 

1400 

400 

1500 

1600 

300 

1700 

1800 

375 

1900 

2000 

1 

tion,  11,000  volt  is  also  standard  for  60  cycles,  and  13,200  volt 
for  25  cycles. 

When  a  generator  is  wound  for  240  volts  it  does  not  necessarily 
follow  that  it  may  be  reconnected  for  480  volts;  and,  vice  versa,  a 
480-volt  machine  cannot  always  be  reconnected  to  240  volts  by 


Digitized  by  CjOOQ IC 


318  ELECTRICAL  EQUIPMENT 

changing  the  number  of  circuits.  The  above  is  particularly  true 
of  generators  with  large  diameters  and  a  great  number  of  poles. 
Small  machines  with  few  poles  can,  as  a  rule,  be  reconnected  or 
rewoimd  for  any  voltage  up  to  and  including  2300.  It  is  a  com- 
mon but  erroneous  idea  that  machines  wound  for  2300  volts, 
delta  connected,  can  be  simply  reconnected  to  4000  volts  Y. 
While  this  is  all  right  so  far  as  mere  voltage  is  concerned,  the  slot 
in  the  armature  may  not  be  large  enough  to  accommodate  the 
extra  insulation  required  for  the  higher  voltage.  In  large  machines 
the  above  change  may  sometimes  be  made  without  much  diffi- 
culty, but  small  machines  require  as  a  rule,  new  coils  and  fre- 
quently new  punchings. 

Parallel  Operation.  In  order  that  an  alternating-current  gen- 
erator shall  be  able  to  carry  a  load,  a  current  must  flow  corre- 
sponding to  this  load.  The  e.m.f.  required  to  generate  this  cur- 
rent is  the  resultant  of  the  terminal  and  the  induced  e.m.f.'s  of 
the  generator,  the  displacement  between  these  e.m.f.'s  being  due 
to  the  impulse  of  the  prime  mover.  In  the  same  manner  when 
two  or  more  generators  are  operating  in  parallel  the  division  in 
load  between  the  different  units  is  entirely  dependent  on  the 
turning  efforts  of  the  prime  movers,  and  a  change  in  the  field 
excitation,  as  with  direct-current  generators,  will  have  no  eflFect 
whatsoever. 

For  a  satisfactory  parallel  operation  it  is  important  that  the 
e.m.f.'s  of  the  generators  are  the  same  and  that  they  are  operated 
in  perfect  synchronism,  as  if  this  is  not  the  case  cross  currents 
will  flow  between  the  units.  These  cross  currents  may  be  either 
wattless  or  they  may  represent  a  transfer  of  energy,  depending  on 
whether  they  are  caused  by  a  difference  in  the  e.m.f.  or  a  speed 
variation  of  the  machines. 

When  two  alternators  are  operating  in  parallel  at  the  same 
speed,  their  e.m.f.'s  are  naturally  in  opposition  as  shown  in  Fig.  186. 

Let  OA  be  the  e.m.f.  of  generator  No.  1  and  OB  the  e.m.f.  of 
generator  No.  2,  the  difference  in  their  values  being  caused  by  a 
stronger  excitation  of  the  latter  machine.  The  resultant  e.m.f. 
OC  will  be  in  phase  with  OB,  and,  being  impressed  on  the  syn- 
chronous impedance  of  the  two  generator  armatures  in  series,  it 
will  produce  a  cross  current,  lagging  nearly  90^  behind  the  e.m.f. 
of  generator  No.  2  and  leading  nearly  90"*  in  advance  of  the  e.m.f. 
of  generator  No.  1.    This  is  practically  true,  as  the  impedance 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


319 


can  be  considered  to  consist  almost  entirely  of  the  reactance  of  the 
circuit.  The  cross  current  will,  therefore,  have  a  magnetizing 
effect  on  generator  No.  1  and  a  demagnetizing  effect  on  generator 
No.  2,  and  consequently  keep  the  voltages  the  same.  The  cross 
current  is  wattless,  consimiing  no  power  except  that  corresponding 
to  the  PR  loss  in  the  circuit.  It  is  thus  evident  from  the  above 
that  a  change  in  the  field  excitation  can  hs^-ve  no  effect  on  the  load 
of  the  machine. 

If  the  excitation  of  the  two  machines  is  the  same,  but  the  gov- 
ernor adjustments  differ,  a  cross  current  will  also  be  produced  as 
shown  in  Fig.  187.  OA  represents  the  induced  e.m.f.  of  generator 
No.  1,  leading  6  degrees  in  advance  of  the  bus-bar  voltage,  while 
OB  represents  the  induced  e.m.f.  of  generator  No.  2,  lagging  $ 
degrees  behind  the  bus-bar  voltage.    The  resultant  OC  will  cause 


A-^ 


Qm.*» 


-►B 


0        C 
FiQ.  186. 


Fig.  187. 

a  cross  current  to  flow  and  as  the  resistance  of  the  circuit  is  small 
compared  to  the  reactance,  it  will  lag  nearly  90°  behind  OC,  and 
practically  be  in  phase  with  the  e.m.f.  of  generator  No.  1,  and  in 
opposition  to  the  e,m.f.  of  generator  No.  2.  It  will  thus  consume 
p)0'^er  of  the  leading  machine  No.  1,  that  is,  retard  it,  and  supply 
power  to  the  lagging  machine  No.  2,  that  is,  accelerate  it,  and  thus 
pull  the  two  machines  together.  It  is  evident  from  the  diagram 
that  it  is  the  reactive  component  ID  of  the  cross  current  that  pro- 
duces the  synchronizing  power,  and  that  the  power  component 
OD  has  no  effect  in  this  respect.  A  certain  amount  of  reactance  is 
therefore  necessary  for  a  satisfactory  synchronous  operation,  and 
the  larger  the  reactance  is,  compared  to  the  resistance,  the  larger 
is  the  synchronizing  component  of  the  cross  current.  Increasing 
the  reactance  would,  therefore,  increase  the  synchronizing  force, 


Digitized  by 


Google 


320  ELECTRICAL  EQUIPMENT 

but  there  is  a  limit  hereto  also,  as  with  a  very  high  reactance  the 
total  cross  current  would  be  reduced,  and  thus  also  the  syn- 
chronizing current. 

The  synchronizing  force  is  a  function  of  the  short-circuit  cur- 
rent ratio  of  the  generator  and  may  be  defined  as  the  torque  per 
d^ree  displacement. 

The  torque  in  foot-pounds  corresponding  to  a  given  Kw. 
energy  load  is: 

Kw.X33,000     ^Kw.X7040 
R.P.M.xarX.746       R.P.M.    ' 

The  synchronizing  torque  is  then  equal  to 

^_Kw.X7040 

where  6  is  the  angle  of  displacement. 

Assume  a  generator  rated  475-72-1250  Kw.  1.0  P.F.-IOO- 
2300  V.  having  a  synchronous  impedance  limiting  the  short-cir- 
cuit cmrent  to  three  times  normal.  The  current  flowing  can,  with 
sufficient  accuracy,  be  assumed  to  be  proportional  to  the  sine  of 
the  displacement  between  the  terminal  or  bus-bar  e.m.f.  and  the 
induced  generator  e.m.f.  At  short  circuit,  this  displacement 
would  be  approximately  90**,  thus  the  short-circuit  current  would 
correspond  to  sin  90"*  =  1.  As  this  current  has  been  assumed  to 
be  three  times  full-load  current,  the  latter  would  correspond  to  a 
displacement  of  0^^  the  sine  of  which  would  be  equal  to  J. 

Sine  6=i  and  ^  =  19.5  degrees. 

The  synchronizing  torque  of  this  generator  with  a  certain  dis- 
placement, for  example,  10  degrees,  would  be: 

rn      1250X7040      .-^-.     ^  , 

^'^  100X19.5  ^^^^^  foot-pounds. 

The  cross  current  of  the  above  generator  with  a  certain  dis- 
placement, for  example,  10°,  would  be: 
Sin.  10*^=0.17. 
Full-load  cmrent =315  amp. 

0  17 
Cross  current =^r^X  315  =  160  amperes. 

Strictly  speaking,  this  is  not  a  cross  current  but  the  transfer 
of  current  to  the  generator  in  question  from  the  others,  which  are 
reUeved  of  a  corresponding  amoimt. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  321 

Where  troubles  from  excess  cross  ciirrents  are  found,  it  can 
usually  be  found  due  to  a  too  close  regulating  machine,  having  a 
too  high  short-circuit  ratio  in  combination  with  insufficient  fly- 
wheel capacity. 

In  considering  the  function  of  flywheel  effect,  a  sharp  distinc- 
tion should  be  made  between  momentary  speed  changes  or  speed 
fluctuations  and  slow  changes  or  adjustments  due  to  the  speed- 
load  characteristic  of  the  water  wheel  and  governor,  or  what  is 
properly  called  speed  regulation.  All  prime  movers  that  operate 
together  to  supply,  power  to  a  common  load  must  operate  at  a 
lower  speed  when  loaded  than  when  unloaded,  in  order  that  the 
several  prime  movers  will  properly  divide  the  load.  It  is  also  well 
to  differentiate  between  the  fimction  of  flywheel  effect  in  water- 
wheel-driven  generators  and  in  reciprocating  engine-driven  gen- 
erators. In  the  former  the  single  purpose  is  to  restrain  speed 
changes  during  the  necessarily  long  period  of  adjustment  of  input 
to  output.  In  the  latter  the  most  important  function  is  to  pre- 
vent the  excessive  changes  in  angular  velocity  during  a  single 
revolution  that  would,  otherwise,  be  caused  by  the  varying  torque 
delivered  by  the  engine  cylinders.  WhUe  with  engine-driven  imits 
flywheel  effect  is  important  from  the  standpoint  of  steady  parallel 
operation,  this  is  not  the  case  with  water-wheel  installations. 
With  the  latter  the  flywheel  effect  influences  the  speed  only  with 
sudden  changes  in  load,  and  during  the  short  time  interval  during 
which  the  hydraulic  conditions  are  changing  to  meet  the  new-load 
conditions. 

The  division  of  the  load  was  entirely  dependent  on  the  angular 
displacement  between  the  bus-bar  and  induced  generator  e.m.f.'s 
caused  by  the  turning  movements  of  the  prime  movers.  It  is, 
therefore,  evident  that  the  speed  regulation  of  the  prime  movers 
must  be  the  same,  i.e.,  they  must  drop  in  speed  from  no  load  to 
full  load  by  the  same  percentage  and  in  the  same  manner.  If  this 
is  not  the  case,  the  alternator  connected  to  the  prime  mover  of 
closer  speed  regulation  will  take  more  than  its  share  of  the  load 
under  heavy  loads  and  less  under  light  loads,  and  a  too  close  speed 
regulation  is,  therefore,  not  desirable  for  parallel  operation  of 
alternators.  To  illustrate  this  further:  Assume  prime  movers  of 
different  speed  regulation  as  shown  in  Fig.  188.  When  operating 
in  parallel  it  has  previously  been  proven  that,  if  an  irregular  speed 
exists,  a  transfer  of  energy  will  take  place  between  the  alternators, 


Digitized  by 


Google 


322 


ELECTRICAL  EQUIPMENT 


tending  to  retard  the  machine  of  the  higher  speed  and  accelerate 
the  machine  of  the  slower  speed,  thus  tending  to  hold  the  machines 
in  S3mchronism  at  a  speed  corresponding  to  the  load.  The 
division  of  the  load  between  the  imits  depends  then  only  on  the 
action  of  the  governors,  and  it  is  seen  from  the  cmires  that  at  a 
load  Cf  the  machines  will  divide  the  load  equally.  For  other  loads, 
the  ratio  will  be  different,  for  example,  at  a  certain  lighter  load 

the  ratio  may  be  ~,  while  for  a  certain  heavier  load  it  may  be  -^\ 

The  division  of  load  between  two  alternators  depends,  there- 
fore, as  stated,  primarily  upon  the  speed-load  characteristics  of 
the  prime  movers,  the  governors  of  which  must  be  adjusted  for 


Fig.  188. 

a  definite  drop  in  speed  from  no  load  to  full  load.  With  flat  speed 
characteristics  the  division  of  the  load  will  be  of  an  unstable 
nature.  By  adjusting  the  field  4he  form  of  the  energy  deUvered 
by  the  generator  can  be  changed  but  not  the  amount.  What 
really  occurs  with  a  change  in  field  adjustment  of  any  piece  of 
synchronous  apparatus  operating  in  parallel  with  another,  is  a 
change  of  the  power  factor  of  that  machine. 

The  above  refers  also  to  different  stations  operating  in  parallel 
on  the  same  system,  and  the  division  of  load  and  wattless  current 
between  the  stations  must,  therefore,  be  handled  differently.  On 
a  network  supplying  power  over  a  large  territory,  the  power  factor 
will  often  be  low  and  there  will  be  considerable  wattless  current 
to  be  taken  care  of. 


Digitized  by 


Google 


SVNCHRONOUS  generators  323 

A  successful  parallel  operation  of  several  stations  on  a  system 
is,  as  a  rule,  not  difficult,  inasmuch  as  the  line  characteristics,  i.e., 
resistance  and  reactance,  are  generally  such  that  they  little  inter- 
fere with  the  synchronizing  force  of  the  generators.  This  force  is, 
as  stated,  greatest  when  the  machines  are  over-excited,  and  the 
only  case  where  a  machine  would  drop  out  of  step  would  be  on 
extensive  S3rstems  where  large  lagging  cmrents  are  required  for 
voltage  regulation.  These  currents  naturally  greatly  reduce  the 
S3mchronizing  force  in  that  they  weaken  the  field,  but  there  is 
generally  no  danger  of  a  shut-down  imless  a  very  heavy  load  should 
suddenly  come  on. 

Many  different  methods  are  used  for  dividing  and  regulating 
the  load  on  a  large  system.  In  some  cases  one  or  more  generators 
in  a  large  station  or  one  or  more  stations  in  a  large  system  will  do 
the  governing,  taking  care  of  the  load,  the  other  generators  or 
stations  being  then  operated  with  constant  gate  opening  and  con- 
stant load.  Plants  having  large  pondage  are  usually  selected  to 
take  care  of  the  load  fluctuations  while  those  with  little  or  no 
storage  should  preferably  be  operated  so  as  to  take  the  full  flow  of 
the  stream.  In  many  systems  such  stations  are  equipped  with 
induction  generators  which  require  very  Uttle  attention,  possibly 
only  once  a  day.  They  may  be  started  up  in  the  morning  or  kept 
running  all  the  time,  and  as  they  are  dependent  on  the  other 
synchronous  apparatus  on  the  system  for  their  excitation,  their 
speed  and  frequency  is  determined  by  them.  As  there  are  no 
governing  devices,  means  must  be  provided  for  disconnecting  the 
units  from  the  system  as  well  as  shutting  the  gates,  should  the  load 
be  dropped  for  some  reason  or  other,  thus  preventing  overspeed. 

When  steam-turbine  stations  are  used  as  auxiliaries  these  carry, 
as  a  rule,  little  load  ordinarily,  but  on  the  contrary,  often  a  full 
load  of  wattless  current,  and  besides  they  are  always  ready  in 
case  of  emergency  to  pick  up  the  load. 

In  this  connection  it  may  be  well  to  point  out  a  fallacy  that 
often  exists  with  large  customers,  in  that  they  specify  that  their 
lines  shall  be  independent  of  the  rest  of  the  system  and  that  their 
load  be  suppUed  by  separate  generators.  Such  requirements  are, 
of  coiu'se,  based  on  an  assumption  that  a  better  service  can  be 
obtained  in  this  way,  as  his  lines  or  generators  are  not  affected  by 
the  fluctuation  on  the  rest  of  the  system.  This  is,  however,  in 
most  instances  not  the  case,  as  changes  in  his  load  will  affect  the 


Digitized  by  VjOOQ IC 


324  ELECTRICAL  EQUIPMENT 

speed  on  his  generators  and  the  regulation  of  his  lines  much  more 
than  if  the  fluctuations  were  divided  among  a  greater  number  of 
generators  and  lines.  So,  for  example,  in  a  large  system,  what 
would  be  50  per  cent  load  thrown  on  or  oflf  one  generator  if  it  were 
feeding  a  separate  customer  would,  perhaps  be  only  5  or  10  per  cent 
load  on  the  entire  system  and  neither  speed  nor  voltage  would  be 
materially  affected.  In  general,  it  may,  therefore,  be  said  that 
in  many  cases  it  is  preferable  to  operate  everything  in  parallel 
and  to  have  the  governors  on  as  many  machines  as  feasible.  This 
naturally  reduces  the  work  of  the  governors,  as  a  change  in  load 
then  only  requires  each  governor  to  work  through  a  small  range, 
allowing  a  more  sensitive  adjustment  and  less  speed  deviation 
than  would  be  the  case  if  the  system  were  divided  up  into  sections 
with  different  generators  supplying  individual  loads. 

Mechanical  Design.  Revolving  Field  Type.  Alternating  cur- 
rent generators  are  almost  always  of  the  revolving  field  type,  this 
construction  being  preferable  as  compared  with  the  revolving 
armature  t3rpe.  Besides  relieving  the  high  potential  armature 
winding  from  strains  imposed  by  a  centrifugal  force,  it  gives  an 
increased  space  for  the  winding,  which  is  of  greatest  importance. 
Only  two  collector  rings  are  required  for  handling  the  field  cur- 
rent, the  energy  and  voltage  of  which  is  relatively  small  com- 
pared to  that  which  would  have  to  be  handled  in  the  case  of  a 
revolving  armature  generator  of  the  same  capacity. 

Method  of  Drive.  With  regard  to  the  method  of  drive  water- 
wheel-driven  generators  are  almost  always  of  the  direct  connected 
type,  only  the  very  smallest  sizes  being  belt  or  rope  driven. 

Horizontal  or  Vertical.  Water-wheel-driven  generators  may 
be  either  of  the  horizontal  or  vertical  type,  the  latter  being  now 
very  extensively  used  in  low-head  developments  where  it  becomes 
desirable  to  place  the  generators  above  the  highest  flood  level. 
This  arrangement  requires  less  excavation,  and  obviates  the  neces- 
sity for  special  construction  to  protect  from  flood  water,  which 
would  be  necessary  with  horizontal  units.  In  order  to  obtain 
commercial  speeds  for  direct  connection  to  horizontal  generators 
it  has  been  necessary  for  extreme  low-head  developments  to  put 
a  nimaber  of  runners  on  the  same  shaft.  Recent  improvements  in 
the  design  of  single  nmner  turbines  for  low  heads  resulting  in 
increased  speeds,  as  well  as  the  comparatively  low  cost  of  vertical 
generators  operating  at  from  one-third  to  one-half  the  speed  of 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  325 

horizontal  generators,  have  made  the  construction  of  vertical 
units  for  extreme  low  heads  lAuch  simpler  than  horizontal  units. 
The  draft-tube  excavation  required  is,  of  course,  much  less  and 
involves  less  expense.  For  high-head  developments  with  impulse 
wheels,  horizontal  imits  are  of  com-se  preferable. 

Stator  Frame,  The  main  function  of  the  stationary  armature 
frame  is  to  support  the  punchings  and  it  should,  therefore,  be  of  a 
rigid  construction  so  as  to  prevent  any  sag  of  the  punchings  due 
their  weight  and  an  unbalanced  magnetic  pull.  It  is  usually  of  a 
box  type  construction,  and  for  smaller  sizes  they  are,  as  a  rule, 
made  in  one  piece,  while  for  larger  units  they  are  split  so  as  to 
facilitate  an  easy  handling  and  shipping.  A  number  of  openings 
are  provided  for  ventilation,  a  subject  which  is  treated  more  in 
detail  in  the  latter  part  of  this  section. 

The  core  consists  of  sheet-iron  laminations  carefully  annealed 
and  treated  so  as  to  minimize  both  hysteresis  and  eddy-current 
losses.  The  punchings  are  stacked  together  so  that  the  lamina- 
tions overlap  each  other.  They  are  held  rigidly  in  place  by  heavy 
steel  clamping  fingers,  air  circulation  being  provided  for  by  air 
ducts  formed  by  spacing  blocks  inserted  at  frequent  intervals 
between  the  laminations.  The  outer  circumference  is  dove- 
tailed for  fastening  to  the  frame,  while  the  slots  for  the  windings 
are  punched  at  the  inner  circumference,  the  slots  generally  being 
of  the  open  type  so  as  to  permit  the  use  of  form-wound  coils,  which 
can  easily  be  removed  and  replaced  in  case  of  damage.  With  the 
open  slot  construction  means  must  be  provided  to  guard  against 
the  generation  of  eddy  currents,  due  to  the  unequal  flux  distribu- 
tion. This  is  done  by  subdividing  the  individual  conductors 
either  by  using  several  wires  in  parallel  or,  in  the  case  of  con- 
ductors of  large  cross-section,  by  using  pressed  cable;  the  eddy 
durents  are  thus  reduced  to  a  negligible  quantity. 

Armature  Winding.  The  armatiu-e  winding  is  generally  of  the 
lap  or  barrel-wound  type.  Fig.  189,  and  the  chain  winding  has 
b^n  practically  abandoned  as  it  requires  coils  of  different  shapes, 
especially  with  the  widely  distributed  windings  which  are  used  in 
modem  machines. 

The  coils  should  be  taped  and  treated  with  an  impregnating 
compound,  the  number  of  layers  and  dippings  being  determined 
by  the  operating  voltage.  The  materials  used  should  be  very 
carefully  selected  to  avoid  deterioration  or  diminution  of  the  dielec- 


Digitized  by 


Google 


326  ELECTRICAL  EQUIPMENT 

trie  strength,  this  being  especiaJIy  important  for  high  potentials. 

After  being  tested  the  coils  are  inserted  in  the  armature  slots  in  an 

armor  of  horn  fiber  or  mica, 

.  ,  .__      ,      ^^  J  retaining  wedges  of  wood 

^^^^^to  1^  ^c  dovetailed  into  the  sides 

f  id^HB^  p^^        of  the  slots  near  the  top.   Ac- 

i  /S^^^S^  1^9        cording  to  the  A.I.E.E.  rules 

the  insulation  should  be  such 
that  the  winding  will  with- 
stand a  test  voltage  for  one 
minute  continuously  of  twice 
the  normal  voltage  plus  1000 
volts.  The  frequency  of  the 
testing  circuit  shall  not  be  less 
than  the  rated  frequency  of 
the  generator. 

J  lie  '  »  *  t    AWV^^^ff  Where  heavy  windings  pro- 

P  /  »  »'  /  *fflviK3^B  ^^^^  beyond  the  laminations, 

"jt^-'I^   ^uWS^V  _         g^  additional  support  is  pro- 
FiG.  189.— Lap  or  Barrel-type  vided  by  means  of    an    insu- 

Armature  Winding.  lated  metal  ring  or  brackets 

to  which  the  outer  ends  of 
the  coils  are  fastened,  thereby  protecting  them  from  mechanical 
displacement  or  distortion  due  to  magnetic  disturbances  caused 
by  violent  fluctuations  or  short  circuits.  This  bracing  of  the 
armature  winding  is  particularly  necessary  with  single-phase 
generators  where  the  severe  mechanical  strains  are  imposed  on 
the  armature  windings  by  the  pulsating  flux. 

Flexible  terminal  leads  provided  with  suitable  connection  joints 
should  be  brought  through  the  frame  near  the  bottom.  With 
three-phase  machines  it  is  in  many  cases  necessary  to  bring  out  the 
neutral  lead,  as  the  machine  may  have  to  be  operated  on  the  four-, 
wire  principle  or  it  may  be  desired  to  ground  the  neutral. 
!  Field  Spider,  The  rotating  field  generally  consists  of  pole 
pieces  mounted  on  a  cast-iron  or  steel  ring  connected  to  the  hub 
by  means  of  arms  of  ample  crossngection.  For  smaller  and 
medium-size  machines  the  field  centers  may,  however,  consist  of 
built-up  punchings  to  which  the  pole  pieces  are  dovetailed.  Where 
shipping  conditions  permit,  the  field  spider  and  the  rim  may  be 
cast  in  one  piece,  otherwise  it  must  be  split  into  sections.    When 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  327 

split,  this  can  be  done  either  lengthwise  or  crosswise  to  the  shaft. 
The  former  method  is  used  when  the  diameter  of  the  rotor  is  very 
large  and  the  latter  method  when  the  length  is  large  (see  Fig.  190). 
Tte  sections  should  be  securely  held  together  by  heavy  bolts  and 
link  keys,  and  when  the  field  is  split  crosswise  to  the  shaft  one  set 
of  arms  should  preferably  be  provided  for  each  section  so  as  to 
insure  a  rigid  construction. 

Field  Poles.  The  pole  pieces  are  built  up  of  laminated  sheet 
steel  punchings,  spreading  at  the  pole  face  so  as  to  secure  not  only 
a  wide  polar  arc  for  the  proper  distribution  for  the  magnetic  flux, 
but  also  for  holding  the  field  coils  in  place.  These  pimchings  are 
either  riveted  or  bolted  together  and  reinforced  by  two  stiff  end 
plates.  For  machines  of  moderate  speed  the  poles  are  simply 
bolted  to  the  rim,  while  for  machines  of  higher  speeds  they  are 
solidly  mounted  on  the  spider  by  means  of  dovetail  slots  in  the 
rim  (see  Fig.  190).  These  dovetailed  grooves  should  be  made 
somewhat  larger  than  the  corresponding  part  of  the  punchings 
and  a  tight  fit  is  obtained  by  means  of  steel  wedges,  which  are 
guarded  from  falling  out  by  two  bolted  end  rings.  For  high-speed 
water-wheel-driven  generators  which  must  be  designed  for  a  nm- 
away  speed  of  twice  normal,  it  often  becomes  necessary  to  pro- 
vide additional  precautions  against  the  increased  centrifugal 
stresses  at  such  occasions.  SoUd  steel  rings  as  shown  in  Fig.  191, 
are  then  often  provided  at  each  end  of  the  rotor,  these  rings  being 
securely  bolted  both  to  the  rim  and  each  pole  piece.  On  some 
very  high-speed  machines,  a  design  as  shown  in  Fig.  192,  is  often 
used.  The  field  centers  are  here  constructed  of  rolled  steel  plates 
and  the  pole  pieces  are  securely  dovetailed  thereto,  thus  making 
a  very  substantial  construction. 

The  revolving  parts  of  water-wheel-driven  generators  should 
be  designed  so  as  to  keep  the  stresses  due  to  centrifugal  force, 
well  within  the  elastic  limit  of  all  the  material  at  the  run-away 
speed  of  the  water  wheel.  This  speed  varies  with  different 
types  of  wheels  and  different  conditions  of  installation;  but  the 
general  practice  is  to  design  the  rotors  with  a  100  per  cent  over- 
speed  in  view. 

Flywheel  Effect.  This  problem  should  be  considered  when  the 
design  of  the  rotor  is  decided  on,  as  well  as  when  a  comparison 
between  different  proposed  generators  is  made.  This  is  really  a 
hydraulic  problem,  and  where  additional  flywheel  effect  is  required 


Digitized  by 


Google 


328 


ELECTRICAL  EQUIPMENT 


I 
f 


I 

I 
I 

O 


o 


8 

d 


Digitized  by 


Google 


SYNCHR.ONOUS  GENERATORS 


329 


Fig.  191.— Revolving  Field  of  3000-Kv.A.,  600-R.P.M.,  Horizontal  Alternator. 


I 


Fio.  192.— Rotor  of  10,000-Kv.A.  High-speed  Water-wheel-driven  Generator. 
Field  Center  Made  up  of  Rolled  Steel  Plates,  into  which  the  Pole  Pieces 
are  Dovetailed. 


Digitized  by 


Google 


330  ELECTRICAL  EQUIPMENT 

it  should,  properly  speaking,  be  put  into  the  runner  of  the  wheel 
or  an  external  flywheel  provided.  It  is,  however,  invariably 
found  that  such  an  arrangement  is  objectionable  and  nearly  alwa3rs 
more  costly  than  to  design  the  generator  rotor  for  the  desired  fly- 
wheel effect,  which  means  additional  material  in  the  rotor  rim  or 
by  increasing  the  diameter  of  the  same.  A  good  value  of  the  WR^ 
for  water-wheel-driven  generators  has  been  given  as  10,000,000 
per  Kw.  maximum  rating,  divided  by  the  square  of  the  speed  ex- 
pressed in  revolutions  per  minute;  thus 

TI7D2       xr      /         N     10,000,000 
WIP  perKw.  (max.)  =  ^^^  p  j^  .g. 

Field  Winding.  Two  methods  are  used  for  winding*  the  field 
coils;  viz.,  the  wire  wound  and  the  strip  wound.  For  small 
machines,  where  even  for  moderate  exciting  voltages  it  is  neces- 
sary to  have  many  turns  of  small  section,  the  cotton-covered  wire- 
wound  coil  is  usually  selected.  The  necessary  insulation  may  be 
placed  on  the  assembled  pole  piece  and  the  winding  wound  directly 
thereon.  Heavy  metal  and  fiber  collars  are  provided  at  the  ends 
and  serve  to  clamp  the  conductors  together  and  prevent  move- 
ment due  to  mechanical  stresses. 

The  wire-wound  field  coil,  however,  has  its  limitations  both 
mechanically  and  electrically.  As  the  centrifugal  force  of  the 
field  coil  increases,  the  vertical  component  of  the  force  will  reach 
a  critical  value  where  the  crushing  stress  on  the  cotton  insulation 
around  the  individual  wires  becomes  excessive,  while  at  the  same 
time  the  horizontal  component  tends  to  tear  the  wires  from  the 
pole.  From  the  electrical  standpoint  the  hmitation  is  that  of 
heating.  It  is  evident  that  the  heat  generated  in  the  inner  layers 
of  the  winding  can  reach  the  outside  surface  of  the  winding  only 
by  passing  through  the  insulation  of  each  succeeding  layer.  This, 
of  course,  results  in  a  very  considerable  difference  in  temperature 
between  the  inner  and  outer  layers  and  in  order  to  operate  the 
former  at  safe  temperatures  it  is  necessary  to  adapt  compara- 
tively low-current  densities  in  the  copper;  this,  in  turn,  resulting 
in  a  heavy  winding  and  consequently  high  centrifugal  forces. 

In  order  to  obviate  these  difficulties,  inherent  to  the  wire- 
woimd  field,  it  is  customary  to  construct  the  winding  of  copper 
strip  wound  on  edge,  as  shown  in  Fig.  190.  The  method  of  insu- 
lating this  type  of  winding  is  similar  to  that  described  for  wire- 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS    .  331 

wound  coils  with  the  exception  that  the  insulation  between  turns 
consists  of  vamisli,  paper,  asbestos,  etc.  It  is  evident  that  this 
type  of  coU  will  not  only  stand  much  greater  vertical  forces,  but 
also,  on  account  of  the  high  moment  of  inertia  of  this  flat  strip,  it 
is  better  able  to  resist  the  horizontal  component  of  the  centrifugal 
force.  In  extreme  cases  it  is  necessary,  however,  even  with  strip 
winding  to  place  brackets  between  the  field  coils  to  overcome  this 
tendency  toward  lateral  distortions  of  the  coil,  as  shown  in  Fig. 
192.  Means  should  also  be  provided  to  thoroughly  fasten  the 
connections  between  the  coils,  and  prevent  them  from  working 
loose,  due  to  the  strains  imposed  by  the  centrifugal  force. 

The  bare  outside  edge  of  the  copper  strip  is  exposed  to  the 
direct  fanning  action  of  the  rotor,  and  since  the  temperature  drop 
in  the  copper  itself  is  negligible,  that  is,  for  the  widths  of  strip 
ordinarily  used,  the  heating  of  the  coils  is  due  almost  entirely  to 
surface  drop.  As  a  result,  a  much  higher  current  density  can  be 
used  than  would  be  permissible  with  the  wire-wound  field. 

The  exciter  current  is  conveyed  to  the  revolving  field  through 
two  collectorrings  mounted  on  the  shaft  of  the  machine. 

According  to  the  A.I.E.E.  rules  field  windings  for  A.C.  gen- 
erators must  withstand  a  one-minute  test  voltage  of  a  value  ten 
times  that  of  the  exciter  voltage;  but  in  no  case  less  than  1500 
volts  nor  more  than  3500  volts. 

Shaft,  Shafts  are,  as  a  rule,  furnished  with  water-wheel- 
driven  generators  and  provided  for  couplings  to  be  connected 
to  the  water-wheel  shaft.  Occasionally  one  single  piece  shaft  is 
used  for  mounting  both  the  water-wheel  runner  and  the  generator 
field. 

Provision  is  often  made  for  moving  the  frame  along  the  shaft 
for  convenience  in  repairing  the  windings.  With  the  construction 
shown  in  Fig.  193,  this,  of  course,  means  an  extra  long  and  con- 
sequently larger  and  more  expensive  shaft,  and  in  many  cases  the 
advantages  are  hardly  worth  the  extra  cost.  With  the  base  con- 
struction shown  in  Fig.  194  a  movement  of  the  armature  frame  is 
obtained  without  the  additional  expense  of  a  heavier  shaft  and 
sometimes  also  larger  bearings. 

Bearings.  The  bearings  of  horizontal  units  are  ordinarily  of 
the  self-aligning  pedestal  type  arranged  for  oil  ring  lubrication. 
In  large  bearings,  particularly  for  highnspeed  service,  it  often 
becomes  necessary  to  provide  artificial  water  cooling  for  carrying 


Digitized  by 


Google 


332 


ELECTRICAL  EQUIPMENT 


Fig.  193. — Horizontal  Generator  Showing  Arrangement  for  Moving  Frame 
in  Case  of  Repair,  and  Method  of  Mounting  Direct-connected  Ebcciter. 


Fig.  194. — ^Large  Horizontal  Generator,  Showing  Method  of  Moving  Fnune 

in  Case  of  Repair. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


333 


oflf  the  heat  generated.  Thin,  coil-shaped  copper  pipe  is  embedded 
in  the  lower  bearing  half  just  below  the  surface  of  the  babbitt  and 
oooling  water  is  forced  through  the  coil.  If  the  water  wheel  is 
of  the  overhung  type,  the  size  of  the  bearing  nearest  the  wheel  must 
be  of  suflScient  size  to  take  care  of  the  extra  weight.  Whether  the 
water  thrust  is  balanced  or  not  must  also  be  considered. 

With  vertical  units  the  present  practice  is  to  support  the 


Fig.  1»5.— Typical 


of  Modem  Vertical  Generator  with  Direct-con- 
nected Exciter. 


revolving  element  of  the  entire  imit  from  a  thrust  bearing  moimted 
on  top  of  the  generator  frame.  Two  guide  bearings  are  usually 
provided  with  the  generators,  one  in  the  upper  bracket  directly 
below  the  thrust  bearing,  and  the  other  one  supported  in  a  bracket 
below  the  revolving  field  (Fig.  195).  Generally,  one  guide  bearing 
is  provided  in  connection  with  the  water  wheel.  This  is  usually 
a  babbitted  bearing,  although  sometimes  the  lining  is  of  Ugnum 
vits.    In  case  of  very  low-speed  machines,  where  it  is  possible  to 


Digitized  by 


Google 


334 


ELECTRICAL  EQUIPMENT 


use  an  exceptionally  short  shaft,  it  is  sometimes  possible  to  omit 
the  bearing  immediately  below  the  revolving  field,  but,  in  gen- 
eral, it  seems  preferable  to  have  a  bearing  at  this  point.  The 
thrust  bearing  must  sustain  not  only  the  weight  of  the  revolving 
element  but  also  the  unbalanced  water-thrust,  and  the  top 
bracket  must,  therefore,  be  of  adequate  strength  and  is  usually 
heavily  reinforced,  as  shown  in  Fig.  205. 

There  are  two  general  classes  of  thrust  bearings;  those  which 
depend  upon  a  film  of  oil  between  two  plates,  and  those  which  have 
hardened  rollers  between  two  hardened  surfaces.  The  first  class 
may  be  subdivided  into  (a)  those  which  are  supplied  with  oil 
under  pressure  and  (6)  those  which  revolve  in  a  bath  of  oil  imder 
atmospheric  pressure.  There  are  also  combinations  of  the  two 
classes.  In  either  case  the  bottom  plate  is  stationary  and  some- 
times mounted  on  a  spherical  self-aUgning  washer,  while  the  top 
one  rotates  with  the  shaft. 

Oil-pressure  Bearings.  In  this  type  of  bearing,  oil  imder  high 
pressure  is  pumped  into  an  annular  chamber  between  a  revolving 
and  a  stationary  disc  (see  Fig.  196),  and  the  pressure  required  to 


Fig.  196. — Assembly  of  an  Oil-pressure  Bearing. 

separate  the  plates  is,  of  course,  dependent  on  the  superincumbent 
weight  and  the  area  of  the  bearing  plates.  This  type  of  bearing 
is  not  extensively  used  with  water-wheel-driven  units,  as  when 
taken  in  connection  with  the  necessary  pumps  and  auxiliaries  it  is 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


335 


usually  more  expensive  than  the  other  types.  A  drop  in  the 
pressure  of  the  oil  supply  or  a  momentary  failure  of  the  same 
would  cause  serious  damage  to  the  bearing. 


^ 


Fig.  197. — Kingsbury  Thrust  Bearing. 

Contact-plate  Bearings.  To  this  class,  which  is  the  mopt  gen- 
erally used,  belong  the  Kingsbury  and  the  Spring-thrust  bearings. 
The  former  consists  of  a  stationary  and  a  revolving  plate  sub- 
mersed in  a  bath  of  oil  under  atmospheric  pressure.     The  lower 


Digitized  by 


Google 


336 


ELECTRICAL  EQUIPMENT 


stationary  plate  is  divided  into  a  number  of  babbitted  s^ments 
spaced  sufficiently  apart  to  permit  a  free  circulation  of  oil  (see 
Fig.  197).  Each  segment  or  shoe  has  a  single  pivot  support  located 
toward  one  end  of  the  shoe,  slightly  beyond  the  center  of  gravity 
in  the  direction  of  rotation.    This  arrangement  causes  the  space 


Fig.  198. — Spring-supported  Thrust  Bearing,  Showing  Rubbing  Surface  of 
Rotating  Ring;  Stationary  Ring  with  Sawcut  is  Raised  to  Show  Arrange- 
ment of  Springs. 

between  the  shoe  and  the  thrust  block  on  the  shaft  to  open  slightly 
at  the  other  end  of  the  shoe,  where  the  oil  is  drawn  in  by  the 
rotation  of  the  thrust  block.  The  film  of  oil  on  the  face  of  the 
shoe  thus  assumes  the  form  of  a  very  fine  wedge  constantly  urged 
forward  by  the  rotation  of  the  thrust  block.    This  bearing  may  be 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


337 


operated  with  surface  pressures  of  400  to  500  pounds  per  square 
inch.  A  considerable  excess  of  area  must  be  provided,  however, 
to  take  care  of  the  starting  and  stopping  conditions  which  are 
much  more  severe  than  the  running  conditions. 

The  spring-thrust  bearing  (Figs.  198  and  199),  automatically 
adjusts  itself  to  imequal  loading  due  to  inaccuracies  in  workman- 
ship or  in  alignment.  This  is  of  the  utmost  importance  as,  while  a 
bearing  may  be  properly  adjusted  when  installed,  a  distinctive 
feature  of  the  spring-supported  bearing  is  that  it  will  automatically 
adjust  ^tself  while  in  operation  if  there  is  a  loss  of  aUgnment  due  to  a 
settling  of  foundation  or  to  other  causes. 

As  seen  from  the  illustrations,  the  thrust  collar  is  keyed  to  the 


yEvtAlidntf  ma^ 


Fig.  199. — Spring  Thrust  Bearing. 


shaft  and  transmits  the  weight  of  the  revolving  parts  to  the 
rotating  ring  of  the  bearing.  This  ring  has  a  smooth  rubbing 
surface  and  is  so  designed  that  a  rapid  circulation  of  oil  is  main- 
tained. The  upper  surface  of  the  stationary  ring  is  the  stationary 
rubbing  surface  of  the  bearing.  The  ring  rests  on  springs  held  in 
position  by  means  of  center  pins,  while  dowel  pins  are  provided  to 
keep  the  stationary  ring  from  revolving. 

The  rubbing  surfaces  are  in  a  bath  of  oil,  the  quantity  of  oil 
circulated  from  an  outside  source  depending  on  the  losses  and  the 
cooling  conditions.  Water  cooling  coils  may  be  installed  in  the 
bearing  housing  which  will  reduce  the  amount  of  oil,  and  for 
smaller  bearings,  at  low  speed,  no  circulation  of  oil  from  a  source 
outside  the  bearing  housing  is  required. 


Digitized  by 


Google 


338 


ELECTRICAL  EQUIPMENT 


A  combined  guide  bearing  and  spring-thmst  bearing  has  also 
been  de\reloped  for  carrying  moderate  weights.  These  bearings 
are  very  economical  in  the  space  required  and  usually  do  not 
require  oil  circulation  from  a  source  outside  the  bearing  housing. 

Roller  Bearings.  This  bearing  consists  also  of  a  revolving  and 
a  stationary  plate,  which,  in  this  case,  however,  are  separated  by 


r  r 

Fig.  200. — Assembly  of  a  Roller  Suspension  Bearing. 

hardened  steel  rollers,  held  in  a  brass  retainer  and  arranged  radially 
to  the  shaft  (see  Fig.  200).  The  oil  enters  at  the  inner  periphery 
of  the  brass  cage  and  discharges  between  the  rolls  into  the  sur- 
rounding chamber. 

Combination  Bearings.  Roller  bearings  for  large  units  are 
constructed  in  some  cases  to  incorporate  the  oil  pressing  feature 
also.  The  latter  is  combined  with  the  rollers  in  such  a  manner 
that  the  weight  may  be  lifted  off  the  rollers  for  ordinary  oj)eration 
and  carried  by  them  only  in  the  event  that  the  pressure  should 
accidentally  fail.  Or  it  may  be  carried  ordinarily  on  the  rollers, 
the  pressure  being  held  in  reserve  in  case  of  trouble  w^ith  the 
rollers. 

A  platform  with  ladders  leading  to  it  should  be  installed  on  the 
top  of  the  machines  so  as  to  facilitate  the  inspection  of  the  bear- 
ings. Bridges  are  often  provided  from  such  platforms  to  a  gallery 
running  along  one  wall  of  the  power-house. 

Lubrication.  The  advent  of  the  adoption  of  thrust  bearings 
has  presented  a  new  engineering  problem,  that  is,  the  proper 
design  of  an  oil-circulating  and  oil-filtering  system  so  that  these 


Digitized  by 


Google 


SYNCHRONOUS  GENERATOBS  339 

bearings  will  at  all  times  be  supplied  with  continuous  streams  of 
cool,  clean  oil. 

The  lubrication  of  pressure  bearings  requires  a  positive  dis- 
placement type  of  pump  such  as  a  triplex  pump,  preferably  directly 
geared  or  chain-driven  from  the  turbine  shaft.  As  a  continuous 
supply  is  absolutely  essential  two  pumps  are  sometimes  installed 
for  each  unit,  one  being  motor  driven.  A  central  oil  supply  may 
also  be  used  in  starting  up  or  in  case  of  emergency.  The  inter- 
connection of  the  thrust  bearing  and  the  governor  oil-supply  sys- 
tem by  the  use  of  one  set  of  pumps  is  not  to  be  reconunended. 

For  the  lubrication  of  thrust  bearings,  requiring  no  pressure, 
a  central  oiling  and  filtering  system  of  the  gravity  type,  as  shown 
in  Fig.  201,  is  generally  employed.  Clean  oil  is  stored  in  over- 
head reservoirs,  then  is  distributed  to  the  thrust  and  guide  bear- 
ings on  each  unit  by  means  of  a  suitable  system  of  piping.  After 
passing  through  the  thrust  and  guide  bearings  the  used  oil  flows 
by  gravity  to  filters  located  in  the  basement,  where  it  passes 
through  the  filtering  medium  and  over  cooling  coils,  and  the  puri- 
fied oil  is  then  returned  by  automatically  controlled  pumps  to  the 
overhead  reservoirs  ready  for  re-use. 

The  oil  piping  should  be  laid  out  carefully  to  permit  of  readily 
draining  and  cleaning  the  pipes,  and  air  pockets  should  be  avoided. 
Return  drain  should  be  amply  large  and  properly  pitched  to  rapidly 
and  thoroughly  remove  used  oil.  It  is  better  to  err  on  the  safe 
side  and  have  the  returns  a  size  or  two  too  large  rather  than  too 
small  with  consequent  flooding  of  machines  and  wastage  of  oil. 
AU  feed  pipes  should  be  of  brass  or  reamed  steel  pipes.  All  joints 
should  be  carefully  reamed  and  the  piping  blown  out  with  steam 
or  compressed  air  as  they  are  installed.  Arrangement  for  a  tem- 
porary connection  from  the  feed  pipes  to  the  return  drains  at  the 
machines  is  advisable.  This  allows  of  thoroughly  flushing  out  all 
dirt  by  kerosene  or  oil  before  any  oil  is  fed  to  the  bearings.  The 
piping  should  be  equipped  with  valves  and  unions  to  permit 
readily  disconnecting  a  machine  for  repair  work. 

All  bearings  should  be  equipped  with  sight  feeds  or  some  sim- 
ilar arrangement  to  show  when  the  oil  is  feeding  profusely.  This 
should  preferably  be  in  the  return  as  this  indicates  that  oil  is 
actuaUy  going  through  the  bearings.  Also  the  oil  temperature  for 
each  bearing  can  be  measured  when  necessary.  There  are  many 
indicators  on  the  market  for  this  purpose.     One  of  the  best  schemes 


Digitized  by 


Google 


340 


ELECTRICAL  EQUIPMENT 


is  a  fitting  with  a  spring  cover  on  one  side  that  permits  the  operator 
to  actually  put  his  fingers  in  the  return  oil.    Inspection  is  quickly 


made,  the  oil  stream  is  clearly  seen,  or  may  be  tested  by  the 
fingers  when  the  hght  is  poor,  and  there  is  no  chance  of  a  dirty 
sight  glass  giving  fake  indications. 


Digitized  by 


Google 


SYNCHRONOUS  GENERATOBS 


341 


Digitized  by 


Google 


342 


ELECTRICAL  EQUIPMENT 


The  details  of  a  filter  commonly  used  are  shown  in  Fig.  202, 
and  the  detail  of  the  filter  miits  in  Fig.  203.  It  is  the  devel- 
opment of  this  type  of  filter  miit  which  has  made  possible  the 
continuous  filtration  of  the  enormous  quantities  of  oil  required  in 
modem  hydro-electric  plants.  As  will  be  noted  from  Fig.  202, 
this  design  permits  of  installing  a  very  large  amount  of  filtering 


Fig.  203. — ^Peterson  Filtering  Unit,  Showing  Method  of  Placing 
Bag  Over  Frame. 

surface  in  a  comparatively  small  space,  and  inasmuch  as  the  cloth 
on  individual  units  is  free  from  folds  or  plaits  every  square  inch 
of  it  is  eflfective  in  filtering  the  oil.  Each  unit  consists  of  gal- 
vanized wire  screens  held  in  a  metal  frame.  The  cloth  is  in  the 
form  of  a  bag  which  is  brought  up  over  the  top  of  the  filter  unit  and 
retained  in  place  by  a  cover  which  is  held  down  by  two  thumb 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


343 


nuts.  The  oil  passes  from  the  outside  to  the  inside  of  the  filter 
units,  then  out  through  nozzles  which  project  through  the  wall  of 
the  filtering  compartment  to  the  clean  oil  compartment.  The 
nozzles  on  each  unit  fit  into  spring-actuated  valves  so  that  any 
individual  imit  can  be  withdrawn  and  cleaned  without  interfering 
with  the  continuous  operation  of  the  filter.  When  the  filter  unit 
is  withdrawn  this  valve  closes  and  prevents  imfiltered  oil  from 
flowing  into  the  clean  oil  compartment. 

In   order  to  afford  the  operators  complete  control  over  the 
operation  of  the  oiling  system  and  also  to  provide  necessary  plant 


Fio.  204. — Oil  Piping  Arrangement,  Showing  Indicators. 

records  it  is  customary  to  arrange  the  oil  piping  at  the  individual 
units  as  shown  in  Fig.  204.  Such  equipments  include  sight- 
flow  indicators,  oil  meters  and  recording  as  well  as  indicating 
thermometers. 

Instead  of  installing  a  central  system,  as  described  above,  it  is 
occasionally  found  desirable  to  provide  each  generator  with  its 
own  individual  oiling  and  filtering  system.  A  plant  equipped  in 
this  manner  is  shown  in  Fig.  205.    The  filter  is  of  the  same  general 


Digitized  by 


Google 


344 


ELECTRICAL  EQUIPMENT 


design  as  described  above  and  sets  alongside  of  each  generator. 
Dirty  oil  from  the  bearings  flows  by  gravity  into  the  filter,  while 
the  clean  oil  is  pumped  direct  to  the  bearings  by  means  of  a  rotary 
pump,  belt  or  chain  driven  from  the  governor  shaft.  The  dis- 
charge of  this  pump  is  provided  with  a  reUef  valve  with  by-pass 
leading  back  into  the  clean  oil  compartment  of  the  filter.  The 
piping  at  these  units  is  arranged  practically  the  same  as  shown 
in  Fig.  204,  that  is,  the  inlet  and  outlet  lines  are  provided  with 
thermometers,  sight  flow  indicators,  etc. 

Fig.  205  shows  the  installation  at  the  plant  of  Colimibia  Mills, 
Inc.,  Minetto,  N.  Y.    This  plant  contains  six  2000  H.P,  imits 


Fig.  205. — Individual  Oiling  and  Filtering  Systems  for  each  Generating  Unit. 

Peterson  System. 


and  in  order  to  insure  continuous  operation  an  auxiliary  filter, 
with  necessary  pumps,  oil  storage  tanks,  etc.,  is  located  at  the  end 
of  the  generator  room,  with  clean  oil  and  dirty  oil  manifolds  con- 
necting with  the  individual  filters  on  each  machine.  If,  for  any 
reason,  it  is  desired  to  cut  out  one  of  the  individual  oiling  systems 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS  346 

this  auxiliary  system  at  the  end  of  the  room  can  immediately  be 
thrown  on  to  any  one  of  the  generators. 

While  there  are  many  modifications  of  the  systems  described 
above,  they  will  serve  to  indicate  the  general  types  of  oiling  and 
filtering  systems  now  in  vogue.  The  design  of  these  oiling  sys- 
tems is  a  highly  specialized  branch  of  engineering  because  in  laying 
them  out  and  determining  the  pipe  sizes  it  is  necessary  to  take  into 
consideration  the  kind  of  oil  to  be  used,  and  especially  its  viscosity, 
the  flow  of  oil  being  dependent  upon  the  viscosity  of  the  oil,  which, 
in  turn,  varies  with  the  temperature  of  the  plant.  These  factors 
all  have  to  be  considered  in  la3ring  out  the  piping,  calculating  the 
quantity  of  filtering  surface  and  designing  the  pumps. 

Ventilation.  With  large  generating  units  the  question  of 
ventilation  becomes  of  great  importance;  and  modern  machines 
are,  therefore,  being  designed  to  control  the  path  and  utilize  the 
cooling  effect  of  the  moving  air  to  the  greatest  extent.  Such  a 
machine  is  shown  in  Fig.  194.  The  frame  is  provided  with  ven- 
tilating holes  only  above  the  base  line,  no  outlets  being  provided 
toward  the  pit.  The  end-shields  are  so  designed  that  they  enclose 
the  end  of  the  rotor;  and  all  of  the  air  for  ventilating  the  machines 
is  forced  by  means  of  fans  on  the  rotor  into  the  end  shields  where 
it  is  put  imder  pressure,  thus  ventilating  the  end  windings.  The 
air  which  passes  through  the  core  and  windings  below  the  base  is 
forced  out  of  the  large  openings  in  the  feet  of  the  armature  frame. 
This  will  prevent  the  collection  of  heated  air  in  the  pit,  which  may 
again  be  returned  to  the  field,  and  so  used  over  and  over,  and 
become  more  and  more  heated.  In  certain  instances  no  fans 
need  be  provided,  the  field  poles  themselves  providing  the  required 
fan  action.  Another  very  noticeable  feature  of  this  construction 
is  the  quiet  running  of  the  machines. 

For  machines  requiring  a  large  amount  of  cooling  air  it  is 
becoming  general  practice  to  provide  ducts  whereby  fresh  air  may 
be  taken  directly  from  the  outside  to  the  generator  pit.  With 
moderate  and  high-speed  machines,  which  have  a  suflBcient  fan- 
ning action  in  themselves,  it  is  only  necessary  to  provide  hoods 
for  enclosing  the  ends  of  the  machine  over  the  pit,  as  shown  in 
Fig.  206.  The  air  is  then  drawn  directly  from  the  outside  and 
enters  both  ends  of  the  generator  and  is  forced  through  the  stator 
and  out  in  the  station.  The  bottom  of  the  frame  has  no  holes  so 
as  to  prevent  the  heated  air  from  re-entering  the  pit. 


Digitized  by 


Google 


346 


ELECTRICAL  EQUIPMENT 


This  method  of  ventUation  is  also  readily  adopted  with  vertical 
units.  The  fresh  air  is  drawn  from  the  pit  and  forced  through 
holes  in  the  spider  between  the  pole  pieces  through  the  ducts  in 
the  stator  arid  then  out  in  the  station  through  the  opening  in  the 
top.  In  case  it  is  objectionable  to  let  the  heated  air  out  in  the 
station,  as  in  the  summer  time,  it  may  instead  be  piped  to  the  out- 
side. The  top  of  the  generator  may  be  covered  with  a  sheets 
steel  hood  to  which  a  duct  leading  to  the  outside  may  be  attached. 


Fig.  206. — Horizontal  Water-wheel-driven  Generator,  Showing  Hoods  Pro- 
vided for  Ventilation. 

During  the  winter  months  it  is,  of  course,  advisable  to  discharge 
the  heated  air  into  the  station  in  order  to  heat  the  same. 

With  the  advent  of  very  slow-speed  machines  and  low  periph- 
eral velocities  where  fans  attached  to  the  rotor  cannot  be  effec- 
tively used,  it  may  become  necessary  to  resort  to  forced  ventila- 
tion by  providing  motor-driven  fans,  as  shown  in  Fig.  207.  The 
ventilating  system  should  preferably  be  sectionaUzed,  each  section 
being  provided  with*  at  least  two  fans — one  for  spare.  Where 
three  fans  are  provided,  the  combined  capacity  of  two  must  be 


Digitized  by 


Google 


SYNCHRONOUS  GENERATORS 


347 


able  to  provide  the  required  amount  of  air  for  the  section  in 
question,  the  third  being  kept  in  reserve.  Each  fan  inlet  should 
be  equipped  with  a  damper  for  controlling  the  air  admission  and 
an  automatic  shut-off  damper  on  the  discharge  so  that  when  one 
fan  is  shut  down  no  leakage  will  occur  through  the  fan  from  the 
air  chamber.  The  amount  of  air  to  each  generator  is  regulated 
by  dampers  in  the  ducts  leading  from  the  air  chamber  to  the  wheel 
pits,  and  these  dampers  may  be  regulated  from  the  generator  floor. 
The  entrances  to  the  wheel  pits  should  be  provided  with  air-tight 
doorSy  and  the  pressure  in  the  wheel  pits  should  be  kept  approxi- 


Air  Chamber  for  Units  1  to  8  Air  llirht  Door, 


Vw  Set  H 


Fan  Set  «f  2  Fan  Set '^S 

PLAN  VIEW 


ew.,-!a«^....rv..^.^^  £le4aTiON 

SECTION  A-A 

Fig.  207. — Plan  and  Sectional  Elevation  of  Ventilating  System  for  Large 
Horizontal  Slownspeed  Generators. 

mately  one  inch  of  water,  or  just  enough  to  insure  a  positive  air 
passage  through  the  generator. 

The  air  ducts  should  be  as  straight  as  possible,  and  the  per- 
missible air  velocity  may  vary  from  as  low  as  300  or  400  feet  per 
minute  for  small  or  very  slow-speed  machines  to  as  much  as  1000 
or  1500  feet  for  higher  speed  machines. 

An  approximate  figure  for  the  amount  of  air  required  is  from 
125  to  150  cubic  feet  per  minute  per  kilowatt  loss.  Should, 
however,  the  air  in  passing  through  the  machine  rise  in  tempera- 
ture more  than  15^  to  20^  C,  it  indicates  that  the  air  does  not 


Digitized  by 


Google 


348  ELECTRICAL  EQUIPMENT 

effectively  conduct  away  the  heat  and  the  supply  should  then  be 
increased. 

For  a  further  consideration,  of  the  ventilation  of  the  generator 
room,  see  page  173. 

Brakes.  In  large,  modem  water-power  installations  the  units 
are  very  often  provided  with  brakes,  in  order  to  stop  them  quickly. 
Foreign  material  may  obstruct  the  gates,  preventing  their  closure; 
so  that,  imless  a  brake  is  provided,  it  may  not  be  possible  to  stop 
the  wheel  without  closing  the  emergency  gates.  The  brakes  are 
generally  appUed  to  the  generator  rotor,  the  wooden  face  bearing 
directly  on  the  field  rim,  and  the  required  pressure  being  obtained 
by  means  of  the  oil  pressure  which  is  used  for  operating  the  gov- 
ernors, or  air  pressure  from  the  compressed  air  system. 

A  band  brake  is  also  sometimes  used,  consisting  of  a  flanged 
pulley  mounted  on  the  main  shaft  and  rotating  in  a  steel  brake 
band  into  which  are  bolted  blocks  of  maple.  The  band  may  be 
tightened  around  the  pulley  by  a  worm  gear  operated  by  a  hand 
wheel  on  the  main  generator  floor  above. 

8.  mDUCTION  GENERATORS 

Output  and  Excitation.  The  induction  generator  is  simply 
an  induction  motor  driven  above  its  synchronous  speed.  It 
requires  a  wattless  exciting  current  for  its  operations  and  can, 
therefore,  not  be  operated  as  a  self-contained  unit,  but  alwajrs  in 
connection  with  synchronous  machines,  generators  or  motors. 
These  machines  will  then  furnish  the  necessary  excitation,  and 
also  entirely  govern  the  voltage  and  frequency  of  the  induction 
generator. 

The  output  depends  on  its  speed  above  synchronism,  and,  with 
the  speed  of  the  induction  generator  constant,  it  can  only  be 
increased  by  decreasing  the  speed  and  thus  the  frequency  of  the 
synchronous  machinery.  There  can  be  no  permanent  short-cir- 
cuit current  flowing  inasmuch  as  the  exciting  current  disappears 
when  a  short  circuit  takes  place,  and  the  momentary  current  rush 
is  also  very  small. 

Comparative  Capacity  of  Induction  and  Sjmchronous  Gen- 
erators. Inasmuch  as  the  induction  generator  cannot  furnish 
any  wattless  exciting  current  for  the  inductive  load  on  the  system 
or  for  its  own  excitation,  it  follows  that  this  must  be  furnished 
entirely  by  the  synchronous  machines,   thus  increasing  their 


Digitized  by 


Google 


INDUCTION  GENERATORS  349 

capacity.  For  example,  assume  a  system  with  a  load  of  8000  Ew. 
0.80  P.F.,  and  that  it  is  desired  to  install  an  induction  generator 
having  a  capacity  of  5000  Kw.  0.80  power  factor.  What  would 
the  required  capacity  of  the  synchronous  generators  then  be? 

The  wattless  components  of  the  load  and  the  induction  gen- 
erator which  the  synchronous  generators  must  supply  will  be  6000 
Kv.A.  and  2660  Kv.A.,  respectively,  and,  as  in  addition  they  must 
furnish  the  remaining  energy  of  3000  Kw.  their  capacity  would 

have  to  be  ' 

Kv.A. «  V30002+86602  =  9150, 

thus  almost  twice  that  of  the  induction  generator.  A  somewhat 
larger  generator  could,  therefore,  carry  the  entire  load  without 
any  induction  generator. 

For  a  higher  power  factor,  however,  the  condition  would  be 
different.  If  the  power  factor,  for  example,  were  0.95  instead  of 
0.80  the  total  wattless  Kv.A.  to  be  suppUed  would  only  be  2600+ 
2650 »  5250  and  the  capacity  of  the  synchronous  generators 

Kv.  A.  =  V30002+52502  =  6045. 

For  low-power  factors  it  is,  therefore,  not  very  advantageous  to 
use  induction  generators. 

Operation.  When  putting  an  induction  generator  into  opera- 
tion it  is  only  necessary  to  bring  it  up  to  speed  and  close  the  switch. 
Synchronizing  is  not  needed  inasmuch  as  the  machine  cannot 
generate  any  e.m.f .  until  excited  from  the  Une,  and  when  so  excited 
it  will,  of  course  be  in  phase. 

The  first  current  rush  is  only  exciting  current  because  the  load 
cannot  be  {Hcked  up  until  the  field  is  established.  If  the  current 
rush  should  be  undesirably  large  it  can  readily  be  reduced  by 
inserting  reactances  when  the  machine  is  thrown  on  the  circuit. 
These  coils  can  then  be  cut  out  as  soon  as  a  steady  condition  is 
reached. 

When  driven  by  governor-controlled  water  wheels,  the  speed 
of  the  induction  generator  will  drop  slightly  with  the  load,  and 
in  order  to  divide  the  load  properly  it  will  be  necessary  for  the 
speed  of  the  synchronous  generators  to  drop  still  more.  The 
best  method  of  operating  induction  generators  is,  therefore,  to 
drive  them  with  wheels  without  governor  control.    In  this  manner 


Digitized  by 


Google 


350  ELECTRICAL  EQUIPMENT 

their  output  will  be  kept  constant  and  the  load  fluctuations  will 
be  taken  care  of  by  the  synchronous  generators. 

Places  of  Utilization.  The  foremost  use  of  induction  generators 
is,  therefore,  to  be  expected  in  stations  where  no  storage  is  pro- 
vided and  where  the  entire  output  must  be  utilized  or  wasted. 
Such  stations  will  need  very  little  attendance,  due  to  their  sim- 
plicity; probably  only  once  or  twice  a  day.  Means  must,  how- 
ever, be  provided  for  disconnecting  the  unit  from  the  system  and 
shutting  the  gates  should  the  power  for  some  reason  or  other  go 
ofif  the  line.  This  would,  of  course,  mean  that  the  generator  would 
be  unloaded  and  the  unit  reach  an  overspeed  which  must  be 
automatically  guarded  against. 

General  Construction.  The  construction  of  an  induction  gen- 
erator is  identical  to  that  of  an  induction  motor  with  a  low- 
resistance  squirrel-cage  secondary  winding.  The  machine  requires 
a  very  small  air  gap  and  carefid  consideration  must  be  given  to 
the  ventilation. 

4.  EXCITERS 

One  of  the  problems  in  connection  with  large  generating  sta- 
tions which  has  been  given  comparatively  Uttle  attention  until 
lately,  is  that  of  excitation.  It  is,  however,  of  the  greatest  im- 
portance, as  upon  it  depends,  to  a  large  extent,  the  successful 
operation  of  the  plant.  The  capacity  of  the  exciter  imits,  the 
proper  division  of  the  required  exciter  capacity  into  several 
units,  the  method  of  drive,  whether  by  separate  prime-movers, 
by  individual  motors,  or  whether  direct  connected  to  the  main 
generating  units,  the  arrangements  and  connections  of  the  dif- 
ferent imits,  the  proper  system  of  automatic  voltage  regula- 
tion, etc.,  are  all  factors  which  demand  a  careful  consideration 
when  designing  a  power  plant. 

Separate  Excitation.  With  very  rare  exceptions  all  synchron- 
ous machines  are  separately  excited,  the  excitation  being  obtained 
from  some  direct-current  supply  source.  Generally,  separate 
direct-ciurent  generators  are  provided  for  this  purpose,  and  when 
so  utilized  are  termed  "  exciters." 

A  separately  excited  generator  has  no  inherent  tendency 
toward  regulation,  this  being  either  eflfected  by  a  rheostat  in  the 
field  circuit  or  by  means  of  diflferent  systems  of  automatic  voltage 
regulation,  as  treated  more  fully  in  the  next  section. 


Digitized  by 


Google 


EXCITERS  351 

Capacity  and  Rating.  The  exciters  should  have  a  capacity 
suflScient  to  excite  ail  of  the  synchronous  apparatus  in  the  station 
when  these  machines  are  operating  at  their  maximum  load  and 
at  the  true  operating  power-factor.  It  is  not  enough  to  provide 
for  the  excitation  when  operating  at  unity  power-factor,  because 
the  excitation  which  is  required  at  lower  power-factors  is  con- 
siderably higher  than  at  unity  power-factor.  It  is  considered 
good  practice  to  make  the  combined  capacity  of  all  the  exciters 
equal  to  the  excitation  required  for  all  the  generators,  when  these 
are  operating  at  their  maximum  load  and  stated  power-factor 
(usually  80  per  cent),  plus  a  20  per  cent  addition  for  possible 
variations  in  the  required  excitation. 

Auxiliary  station  apparatus  should  not  be  operated  from  the 
exciter  system,  since  troubles  are  always  Ukely  to  occur  in  these 
circuits  that  may  damage  the  exciters  at  times  when  such  damage 
would  cause  considerable  inconvenience  in  the  operation  of  the 
station.  In  many  stations,  station  auxiUaries  are  now  entirely 
operated  by  alternating  current,  and  the  direct  current  for  the 
control  circuits  can  be  easily  taken  care  of  by  the  use  of  a  small 
motor-generator  set  combined  with  a  storage  battery.  No  com- 
plications are  then  introduced  by  voltage  fluctuations  caused 
by  automatic  voltage  regulators.  Reserve  capacity  in  case  of 
breakdowns  should,  of  course,  be  provided,  the  amount  depend- 
ing on  the  number  of  units. 

Exciters  are  now  given  a  maximum  continuous  Kw.  rating 
based  on  a  temperature  rise  not  exceeding  50**  C,  as  measured 
by  thermometer,  above  an  ambient  room  temperature  of 
40°  C. 

Voltage.  The  pressure  most  commonly  used  for  excitation 
is  125  volts.  For  A.C.  machines  of  very  large  capacity  requiring 
a  large  excitation,  it  will,  however,  usually  be  found  more  econom- 
ical to  use  a  250-volt  excitation.  This  higher  voltage  will  permit 
the  use  of  smaller  exciter  and  field  switches,  while  leads  of  reduced 
size  from  the  exciters  to  the  bus-bars  and  from  the  bus-bars  to 
the  generator  field  may  be  used,  and  the  cross-section  of  the  bus- 
bars cut  in  two;  all  this  being  of  importance  in  reducing  the  cost, 
especially  in  large  installations.  A  considerable  saving  can  also 
generally  be  accomplished  in  the  exciter  itself.  Machines  for 
125  volts  require  a  commutator  twice  as  large  as  those  for  250 
volts;   and  with  water-wheel-driven  units,  where  they  must  be 


Digitized  by 


Google 


352  ELECTRICAL  EQUIPMENT 

designed  to  safely  withstand  double  speed,  the  construction  often- 
times involves  considerable  difficulties  and  expense. 

Characteristics.  When  exciters  are  to  be  operated  in  connec- 
tion with  automatic  voltage  regulators,  as  is  almost  always  the 
case,  it  is  most  important  that  they  are  designed  with  this  point 
in  view.  The  densities,  especially  in  the  fields,  should  be  fairly 
low,  as  with  high  density  the  time  element  required  to  vary  the 
voltage  from  one  point  to  another  would  be  so  long  as  to  materially 
afifect  the  regulation.  The  operating  range  should,  therefore,  be 
below  the  bend  of  the  saturation  curve. 

The  exciter  should  preferably  have  a  time  element  so  that  it 
will  be  responsive  to  changes  in  the  field  excitation  to  the  extent 
that,  by  inserting  an  external  resistance  equal  to  about  three  times 
the  resistance  of  the  field,  the  voltage  will  fall  from  125  to  25  volts 
in  from  six  to  eight  seconds.  An  ideal  exciter  designed  along 
these  lines  should  also  give  at  full  field  165  volts  and  the  increase 
in  the  field  current  from  125  volts  to  150  volts  should  not  be  over 
60  per  cent. 

For  alternators  operating  at  maximum  inductive  load  125  volts 
is  generally  required  for  the  excitation,  and  in  order  to  get  a  satis- 
factory regulation  when  an  automatic  regulator  is  used,  the 
exciter  must  be  designed  so  as  to  be  able  to  give  165  volts  momen- 
tarily. It  is  also  necessary  that  the  increase  in  the  exciter  field 
cmrent  should  be  small,  so  that  the  exciter  will  respond  quickly 
to  the  short-circuiting  of  the  rheostat,  and  thus  insure  the  desired 
alternator  excitation.  Should  the  excitation  voltage  be  any  other 
value  than  125,  viz.,  250  volts,  the  above  values  would  be  pro- 
portionally changed. 

Shunt  vs.  Compound  Wound.  While  an  exciter  can  be  either 
compound  wound  or  shunt  wound,  the  former  is  considered  prefer- 
able for  parallel  operation  with  automatic  voltage  regulation. 

Non-regulating  exciters  should  be  more  or  less  highly  saturated 
in  order  to  insure  a  stable  parallel  operation.  If  such  exciters 
were  to  be  used  with  automatic  regulation,  they  would  be  rather 
slow  to  correspond  to  the  changes  in  field  excitation.  If  a  shunt- 
wound  exciter  is  designed  for  a  low  saturation  so  as  to  make  it  a 
good  regulating  exciter,  the  tendency  might  be  an  unstable  oper- 
ation when  running  in  parallel  without  a  regulator.  Shunt-wound 
exciters  are,  however,  generally  provided  with  commutating  poles 
to  overcome  the  above  difficulties. 


Digitized  by 


Google 


EXCITERS  353 

The  series  field  excitation  of  regulating  exciters  should  not 
exceed  30  per  cent  of  the  total  excitation,  and  the  resistance 
of  the  rheostat  should  be  about  three  times  that  of  the  resistance 
of  the  exciter  shunt  field  when  hot. 

For  regulating  exciters,  which  are  not  to  be  operated  in  parallel, 
the  shunt-wound  type  is  entirely  satisfactory,  provided  it  has  been 
designed  with  this  point  in  view,  that  is,  for  low  saturation. 

Speed.  The  speed  of  an  exciter  depends  on  the  method  of  its 
drive  and  on  its  capacity.  Extremely  slow  or  high  speeds  mean 
excessive  cost  with  the  addition  of  mechanical  difficulties  for 
high  speed.  This  is  especially  important  in  hydro-electric  in- 
stallations, where  the  exciters  are  turbine  driven,  in  which  case 
they  must  be  designed  to  withstand  the  increased  stresses  due  to  a 
double-speed.  This  fact  should  not  be  neglected  when  making  a 
decision  on  the  speed  of  a  water-wheel-driven  exciter. 

Method  of  Drive.  While  the  exciters  can  be  either  belt-driven 
or  direct-connected  to  the  machines  driving  them,  the  latter  prac- 
tice is  almost  exclusively  used  except  in  the  very  smallest  plants. 
The  direct  connection  may  be  either  to  the  main  generators,  to 
separate  water  wheels  or  to  motors,  usually  of  the  induction  type. 
Sometimes,  although  rarely,  an  exciter  may  be  foimd  that  is 
connected  both  to  a  motor  and  a  turbine,  the  latter  running  idle 
when  the  motor  is  carrying  the  load,  and  vice  versa. 

Mechanical  Design.  The  mechanical  design  of  exciters  does 
not  dififer  from  other  direct-current  generators.  They  may  be 
either  of  the  horizontal  or  vertical  type,  the  latter  construction 
being  used  for  units  direct  connected  to  vertical  main  generators 
or  directly  to  vertical  water  wheels.  When  intended  for  direct 
connection  to  horizontal  water  wheels  they  are  almost  invariably 
of  the  pedestal  bearing  type,  the  shaft  being  provided  with  the 
necessary  coupling.  Care  should  be  taken  in  designing  the  bear- 
ings to  see  that  the  water  thrust,  if  any,  is  provided  for.  The 
same  construction  is  also  generally  used  for  large  motor-driven 
sets.  Fig.  208,  the  two  units  being  mounted  on  a  conmion  base. 
Occasionally  only  two  bearings  are  used  and  a  conmion  shaft. 
For  horizontal  imits  direct-connected  to  the  main  units,  shaft 
and  bearings  are  generally  omitted,  the  exciter  armature  being 
mounted  on  an  extension  to  the  generator  shaft  and  the  frame 
supported  on  an  extension  to  the  generator  subbase,  as  shown  in 
Pig.  209. 


Digitized  by 


Google 


354 


ELECTRICAL  EQUIPMENT 


Vertical  direct-driven  exciters,  Fig.  210,  are  ordinarily  pro- 
vided with  one  or  two  guide  bearings  and  a  short  shaft  with 


Fig.  208. — Induction  Motor-drivan  Exciters. 


Fig.  209.— 3000-K.W.  Frequency  Changer  Set,  Showing  Mounting  of  Direct- 
connected  Exciter. 

coupling.    The  rotating  element  is  supported  by  means  of  a 
thrust  bearing  located  on  the  upper  bearing  bracket.     It  should 


Digitized  by 


Google 


EXCITERS  355 

be  of  suflScient  size  to  take  care  not  only  of  the  weight  of  the 
exciter  armature,  but  also  of  the  revolving  element  and  water 
thrust  of  the  turbine. 

In  the  case  of  a  vertical  generator  the  direct-connected  exciter 


Fig.  210. — ^Vertical  Water-wheel-driven  Exciter,  Showing  Thrust 
Bearing  at  Top. 

is  usually  carried  by  the  thrust-bearing  bracket,  as  shown  in 
Fig.  211. 

Arrangement  and  Connections.^  The  question  always  arises 
whether  direct-connected  exciters  should  be  provided  for  each 
unit  or  whether  a  central  supply  source,  consisting  of  as  few  units 
as  possible  is  preferable.  Either  arrangement  has  its  advantages 
and  disadvantages.  The  great  advantage  in  a  water-wheel-driven 
exciter  arrangement  lies  in  the  fact  that  it  is  independent  of  the 
A.C.  system  with  its  load  and  speed  fluctuations.  Two  units  are 
then  usually  provided,  either  of  which  has  a  capacity  to  take  care 
of  the  entire  excitation  of  the  plant,  thus  providing  a  100  per  cent 

*  See  also  Voltage  Regulation. 


Digitized  by 


Google 


356 


ELECTRICAL  EQUIPMENT 


Fig.  211. — Vertical  Generator  with  Direct-connected  Exciter. 

reserve  capacity.  Occasionally  three  units  are  installed,  two  of 
which,  combined,  can  take  care  of  the  entire  excitation,  the  third 
unit  being  the  reserve.  This  method  may  be  the  most  desirable, 
on  account  of  the  possibiUty  of  debris  or  ice  clogging  up  the  small 
exciter  turbines  and  shutting  them  down.  Under  such  condi- 
tions it  would  naturally  be  more  advantageous  to  keep  a  motor- 
driven  set  in  reserve  for  such  an  emergency.  From  an  economical 
point  of  view,  however,  it  is  evident  that  two  motor-driven  units 
with  a  spare  water-driven  unit  would  cost  less.  An  objection  to 
motor-driven  sets  which  is  occasionally  raised  is  that  they  are 
liable  to  drop  out  of  step  when  a  short  circuit  occurs  on  the  system. 
This  is,  however,  not  the  case  with  well-designed  sets  under 
momentary  short  circuits,  and  where  it  has  occurred,  it  has  been 
prevented  by  equipping  the  sets  with  flywheels.  This,  of  course, 
increases  the  expense  of  the  sets  and  is,  as  a  rule,  not  justified. 


Digitized  by 


Google 


EXCITERS 


357 


Exciters  direct  connected  to  each  of  the  main  units  are,  as  a 
rule,  used  in  plants  having  a  small  number  of  units.  They  are,  of 
course,  afifected  by  the  speed  fluctuations  of  the  main  imits,  and 
at  runaway  speeds  they  may  cause  over-voltages  amounting  to 
two  or  three  times  the  normal  voltage,  thus  greatly  endangering 
the  apparatus  on  the  system.  Such  over-voltages  must,  there- 
fore, be  guarded  against  either  by  providing  means  for  artificially 
loading  the  system,  should  the  outside  load  drop,  or,  preferably, 
by  providing  high-voltage  cut-out  relays,  which  will  automatically 
insert  resistance  in  the  exciter  field  circuits  and  thus  prevent  an 
excess  voltage  rise. 

Where  two  or  three  units  are  used,  each  exciter  should  have  a 
capacity  suflSicient  to  excite  two  generators,  while  with  four  or 


iiQ.  212. — SyiStem  of  Exciter  Connections. 

more  units  it  will  undoubtedly  be  more  advantageous  to  make  the 
capacity  of  each  exciter  correspond  to  the  excitation  require- 
ments of  one  generator  and  provide  a  motor-driven  exciter  for 
spare.  This  may  then  have  the  same  capacity  as  one  of  the  direct- 
connected  exciters  or,  for  larger  stations,  it  may  have  twice  the 
capacity  or  two  sets  may  be  installed. 

The  economical  question  should,  of  course,  also  be  consid- 
ered in  deciding  between  the  two  systems.     Direct-connected 


Digitized  by 


Google 


358 


ELECTRICAL  EQUIPMENT 


exciters  will,  as  a  rule,  be  of  a  rather  slow  speed  and  thus  more 
expensive  per  Kw.  than  water-wheel-driven  units,  the  difference, 
however,  diminishing  as  the  head  and  number  of  units  increase. 
On  the  other  hand,  water-wheel-driven  exciters  involve  the  cost 
of  the  hydraulic  equipments,  and  besides  the  additional  expense 
of  the  building  caused  by  the  space  occupied  by  these  units.  The 
eflSiciency  is,  however,  mostly  in  favor  of  direct-connected  units. 

The  general  practice  is  to  provide  one  or  two  sets  of  common 
bus-bars  to  which  all  the  exciters  are  connected  in  parallel  and 


ifftli    IF1I¥BTT 

'  T  r    r     r  T  r »--  r 

I  *  X  X  ill        Rheoautl 


SUtion 
Liffhtins 


f 


Fig.  213. — System  of  Exciter  Connections. 

from  which  the  fields  of  the  different  generators  are  excited,  a 
rheostat  being  inserted  in  each  field  circuit. 

In  the  arrangement  shown  in  Fig.  212  there  are  three  exciters, 
two  of  which  are  water-wheel-driven,  the  reserve  being  motor- 
driven.  Only  one  set  of  exciter  bus-bars  is  shown,  although 
frequently  an  auxiliary  set  i§  also  installed.  The  equaUzer  con- 
nection and  the  exciter  shunt  fields  are  left  out  so  as  to  simplify 
the  diagram.  Means  are  provided  for  sectionaUzing  the  bus,  as 
shown.  Power  for  the  induction  motor  is  taken  from  the  main 
bus,  and  any  number  of  motors  can  be  started  by  one  compen- 
sator if  a  common  nmning  and  starting  bus  is  provided. 

Fig.  213  represents  a  comparatively  large  system  with  not  less 
than  six  direct-connected  exciters  operating  in  parallel.  There 
are  two  sets  of  bus-bars,  one  for  excitation  and  the  other  for 
emergency  or  auxiliary  service,  and  switches  are  provided  so  that 
the  exciters  can  be  connected  to  either  set  as  desired.    One  exciter 


Digitized  by 


Google 


EXCITERS 


359 


can,  if  necessary,  be  connected  to  the  auxiliary  bus  while  the  others 
are  operating  on  the  field-bus.  As  previously  stated,  however,  it 
is  not  considered  good  practice  to  use  the  excitation  system  for 
the  auxiUary  service.  To  provide  spare  capacity  for  the  system 
shown,  a  motor-driven  exciter  can  be  installed,  feeding  the  iaux- 
iliary  bus  and  the  field  switches  made  double-throw  instead. 

In  very  large  plants  the  general  tendency  is  to  so  arrange  the 
system  that  each  generating  unit  shall  form  a  complete  plant  in 
itself,  capable  of  independent  operation,  although  normally  the 
units  are  operated  in  parallel.    Each  generator  is,   therefore, 


^  lL  llllltll' 


Ci^&l^&i^ 


6  mm 


-        _  witliJJinat 


6- 

Fig.  214. — System  of  Exciter  Connections. 


provided  with  its  individual  exciter,  which  may  be.  either  direct 
connected,  as  previously  described,  or  also  motor  driven  with 
power  from  a  separate  s6urce.  Which  system  is  the  most  eco- 
nomical and  advantageous  depends  entirely  on  the  conditions. 

There  are  two  large  systems  in  operation  which  use  the  latter 
scheme,  differing  only  somewhat  with  respect  to  the  power  supply, 
which,  however,  is  entirely  independent  in  either  case. 

One  of  these  arrangements  ^  is  illustrated  by  the  representative 
diagram  in  Fig.  214.  The  exciters,  which  have  a  capacity  cor- 
responding to  that  required  by  their  respective  generators,  are 
not  operated  in  parallel,  but  have  their  terminals  connected 
directly  to  the  generator  fields  through  the  collector  rings.  The 
regulation  is  accomplished  by  adjusting  the  exciter  fields  (see 
Voltage  Regulation),  thus  eliminating  large  field  rheostats  in  the 

*  Mississippi  River  Power  Co. 


Digitized  by 


Google 


360 


ELECTRICAL  EQUIPMENT 


main  field  circuits^  as  well  as  their  losses.  The  exciter  sets  (Fig. 
215),  receive  their  driving  power  normally  from  an  entirely  inde- 
pendent som-ce,  consisting  of  two  auxiliary  water-wheel-driven 
low-voltage  alternators  with  their  own  individual  direct-connected 
exciters  (Fig.  216).  These  alternators  feed  into  a  set  of  bus-bars, 
to  which  the  exciter  motors  are  normally  connected.  Provision 
is  also  made,  however,  so  that  the  exciter  sets  can  be  fed  from 
the  main  bus.    One  step-down  transformer  is  provided  for  each 


Fig.  216. — ^Exciter  Set  for  Generating  Unit  in  MissiBsippi  River  Power  Com- 
pany's Plant  at  Keokuk. 


bus  section  and  supply  power  to  an  auxiliary  exciter  bus  which  is 
sectionalized  in  the  same  number  of  groups  as  the  main  bus. 
Connection  can  also  be  established  (not  shown  in  diagram),  in 
case  of  emergency,  with  a  storage  battery  which  ordinarily  is 
used  for  the  operation  of  the  oil  switches. 

Besides  supplying  power  for  the  exciter  sets,  the  auxiliary 
alternators  also  supply  power  for  the  station  service  and  lighting, 
although  provision  is  made  so  that  it  can  also  be  taken  from  the 
main  bus. 


Digitized  by 


Google 


EXCITERS 


361 


In  the  other  installation  ^  mentioned  the  supply  system  for 
the  exciter  sets  consists  of  two  low-voltage  generators  arranged 
for  combination  drive,  one  end  being  connected  to  a  water  wheel 
and  the  other  to  an  induction  motor  which  obtains  its  driving 
power  through  step-down  transformers  from  the  main  busses. 
The  water  wheel  is  used  for  normal  operation.  In  either  system, 
each  auxiliary  alternator  is  capable  of  carrying  the  entire  exciter 
load  of  the  station. 


FiQ.  216. — ^Auxiliary  Generators  with  Direct-connected  Exciters. 
Supply  for  Motor-driven  Exciters  shown  in  Fig.  215. 


Power 


It  is  advisable  to  keep  a  spare  exciter  set  on  hand  to  replace 
any  one  that  may  break  down,  inasmuch  as  each  exciter  has  only 
sufficient  capacity  to  excite  one  generator  only.  The  process  of 
changing  requires  but  a  very  short  time. 

Exciter  Batteries.  The  use  of  storage  batteries  as  a  reserve 
source  for  field  excitation  has  of  late  been  increasing  considerably. 
The  method  of  connecting  and  operating  such  batteries  varies 
somewhat    with   the   arrangement   adopted   for   furnishing   the 

*  Ontario  Power  Company. 


Digitized  by 


Google 


362 


ELECTRICAL  EQUIPMENT 


normal  supply  of  exciting  current,  and  the  method  employed  for 
controlling  the  field  excitation.  Where  the  exciting  ciurent  for 
all  of  the  machines  is  taken  from  a  common  exciter  bus,  the  bat- 
tery would  ordinarily  be  floated  directly  across  this  bus,  provided 
its  voltage  is  substantially  constant.    If  the  exciter  bus  voltage 


Yoltmster  OouwctiaH 
Ft.  Na  1  ChMg*  "A" 
Pt.NaS  "  "r 
Pt.NaSTotalBattnr 
Pt.NowlBu 


Fig.  217. — Diagram  of  Connections  for  Exciter  Battery  for  Emergenc^^  Service, 
Normally  off  Line. 

is  varied  from  time  to  time  by  manual  control,  the  battery  can 
still  be  kept  constantly  connected  to  the  bus,  but  the  number  of 
cells  in  circuit  must  be  adjusted  by  means  of  an  end  cell  switch 
whenever  the  exciter  bus  voltage  is  changed. 

If  the  exciter  bus  voltage  is  constantly  varied  automatically, 


Digitized  by 


Google 


VOLTAGE  REGULATION  363 

as  by  T.A.-regiilator  control,  the  battery  cannot  be  connected 
directly  across  the  bus.  In  such  a  case  two  difiFerent  methods  of 
handling  the  battery  have  been  used.  The  first  consists  in  pro- 
viding a  constant  potential  exciter  bus  to  which  the  battery  is 
normally  connected,  and  introducing  between  this  bus  and  the 
conmion  excitation  circuit  a  booster  whose  voltage  is  automatically 
controlled  by  the  T.A.  regulator.  This  gfystem  is  described  in 
detail  in  the  section  on  "  Voltage  Regulation/'  page  369. 

The  second  method  consists  in  connecting  the  two  outer  ter- 
minals of  the  battery  to  the  corresponding  sides  of  the  exciter  bus 
and  opening  the  battery  circuit  in  the  middle,  with  an  automatic 
switch  at  this  point  for  connecting  the  battery  in  one  series  in  case 
of  failure  of  the  normal  source  of  exciting  current.  The  two  halves 
of  the  battery  are  provided  with  a  trickling  charge  to  keep  the  cells 
in  a  healthy  and  fully  charged  condition,  by  connecting  through 
high  resistance  to  the  opposite  side  of  the  bus,  as  shown  in  diar 
gram.  Fig.  217. 

Where  there  is  no  common  excitation  circuit,  but  each  alter- 
nator is  provided  with  its  own  independent  exciter,  a  different 
arrangement  is  adopted.  In  such  a  case  an  emergency  exciter 
bus  is  used  to  which  the  battery  is  normally  connected  and  to 
which  a  spare  exciter  may  be  connected  when  required.  Should 
the  som'ce  of  excitation  for  any  one  of  the  alternators  fail,  its  field 
circuit  is  automatically  connected  to  the  emergency  exciter  bus 
and  the  spare  exciter  may  then  be  started  up,  if  it  is  not  already 
in  service,  to  relieve  the  battery  as  soon  as  this  can  conveniently 
be  done. 

5.  VOLTAGE  REGULATION 

Hand  Regulation.  The  simplest  system  of  regulation  is  by 
means  of  hand-operated  rheostats  connected  in  the  field  circuits  of 
each  generator.  The  pressmre  of  the  exciter  bus  is  then  generally 
kept  constant  at  the  rated  exciter  voltage  and  all  the  regulation 
is  done  by  manipulating  the  generator  rheostats.  In  order  to 
regulate  the  exciter  voltage  it  is,  of  course,  also  necessary  to  pro- 
vide rheostats  in  the  exciter  fields. 

T.A.  Regulator.  Of  the  various  schemes  proposed  for  auto- 
matic voltage  regulation,  the  T.A.  regulator  is  now  most  widely 
used.  With  this  system  the  desired  voltage  is  maintained  by 
rapidly  opening  and  closing  a  shunt  circuit  across  the  exciter  field 


Digitized  by 


Google 


364 


ELECTRICAL  EQUIPMENT 


rheostat.  The  rheostat  is  first  turned  in  until  the  exciter  voltage 
is  greatly  reduced  and  the  regulator  circuit  is  then  closed.  This 
short-circuits  the  rheostat  through  contacts  in  the  regulator  and 
the  voltage  of  the  exciter  and  generator  immediately  rises.  At  a 
predetermined  point  the  regulator  contacts  are  automaticaUy 
opened  and  the  field  current  of  the  exciter  must  again  pass  through 
the  rheostat.  The  resulting  reduction  in  voltage  is  arrested  at 
once  by  the  closing  of  the  regulator  contacts  which  continue  to 
vibrate  in  this  manner  and  keep  the  generator  voltage  within  the 
desired  limits. 

Method  of  Operation.  An  elementary  diagram  of  the  type 
T.A.  regulator  connections  with  an  alternating-current  generator 
and  exciter  is  shown  in  Fig.  218.    The  regulator  has  a  directr 


D.OL  Oaatrol 


Fig.  218.— Elementary  Connections  of  Type  T.A.  Automatic  Voltage 

Regulator. 

current  control  magnet,  an  alternating-current  control  magnet, 
and  a  relay.  The  direct-current  control  magnet  is  connected  to 
the  exciter  bus-bars.  This  magnet  has  a  fixed  stop-core  in  the 
bottom  and  a  movable  core  in  the  top  which  is  attached  to  a  piv- 
oted lever  having  at  the  opposite  end  a  flexible  contact  puDed 
downward  by  four  spiral  springs.  For  clearness,  however,  only 
one  spring  is  shown  in  the  diagram.  Opposite  the  direct-current 
control  magnet  is  the  alternating-current  which  has  a  potential 
winding  connected  by  means  of  a  potential  transformer  to  the  alter- 
nating-current generator  or  bus-bars.  There  is  an  adjustable  com- 


Digitized  by 


Google 


VOLTAGE  REGULATION 


365 


peDsating  winding  on  the  alternating-current  magnet  connected 
through  a  current  transformer  to  the  principal  lighting  feeder.  The 
object  of  this  winding  is  to  raise  the 
voltage  of  the  alternating-current 
bus-bare  as  the  load  increases.  The 
alternating  current  control  magnet 
has  a  movable  core  and  a  lever  and 
contacts  similar  to  those  of  the  direct- 
current  control  magnet,  and  the  two 
combined  produce  what  is  known 
as  the  "  floating  main  contacts." 

The  number  of  relays  vary  ac- 
cording to  the  number  and  size  of 
the  exciters,  and  while  the  funda- 
mental principle  of  operation  of  all 
the  forms  of  T.A.  regulators  is  the 
same,  certain  modifications  are 
necessary.  The  relay  consists  of  a 
U-shaped  magnet  core  having  a 
differential  winding  and  a  pivoted 
armature  controlling  the  contacts 
which  open  and  close  the  shunt  cir- 
cuit across  the  exciter  field  rheostat. 
One  of  the  differential  windings  of 
the  relay  is  permanently  connected 
across  the  exciter  bus-bars  and  tends 
to  keep  the  contacts  open;  the  other 
winding  is  connected  to  the  exciter 
bus-bars  through  the  floating  main 
contacts  and  when  the  latter  are 
closed  neutralizes  the  effect  of  the 
first  winding  and  allows  the  relay 
contacts  to  short-circuit  the  exciter 
field  rheostat.  Condensers  are  con- 
nected across  the  relay  contacts  to 
prevent  severe  arcing  and  possible 
injury. 

The  regulator  may  be  mounted  on  the  switchboard  or  on  pedes- 
tals, as  in  Fig.  219,  this  particular  form  having  twenty  relays, 
divided  into  two  groups. 


Fig.  219.— Type  T.A.  Automatic 
Voltage  Regulator  Mounted 
on  Pedestal. 


Digitized  by 


Google 


S66  fiLECtRlCAL  EQUIPMENT 

Cyde  of  Operation.  The  circuit  shunting  the  exciter  field 
rheostat  through  the  relay  contacts  is  opened  by  means  of  a  single- 
pole  switch  at  the  bottom  of  the  regulator  panel  and  the  rheostat 
turned  in  until  the  alternating-current  voltage  is  reduced  65  per 
cent  below  normal.  This  weakens  both  of  the  control  magnets 
and  the  floating  main  contacts  are  closed.  This  closes  the  relay 
circuit  and  demagnetizes  the  relay  magnet,  releasing  the  relay 
armature,  and  the  spring  closes  the  relay  contacts.  The  single- 
pole  switch  is  then  closed  and  as  the  exciter  field  rheostat  is  short- 
circuited  the  exciter  voltage  will  at  once  rise  and  bring  up  the 
voltage  of  the  alternator.  This  will  strengthen  the  alternating- 
current  and  directrcurrent  control  magnets  and  at  the  voltage  for 
which  the  counterweight  has  been  previously  adjusted  the  main 
contacts  will  open.  The  relay  magnet  will  then  attract  its 
armature  and  by  opening  the  shimt  circuit  at  the  relay  contacts 
will  throw  the  full  resistance  into  the  exciter  field  circuit  tending 
to  lower  the  exciter  and  alternator  voltage.  The  main  contacts 
will  then  be  again  closed,  the  exciter  field  rheostat  short-circuited 
through  the  relay  contacts  and  the  cycle  repeated.  This  operation 
is  continued  at  a  high  rate  of  vibration,  due  to  the  sensitiveness  of 
the  control  magnets,  and  maintainB  not  a  constant,  but  a  steady 
exciter  voltage* 

Regvlatar  Arrangements.  The  most  generally  used  r^ulator 
arrangement  consists  of  one  common  regulator  for  several  exciters 
operating  in  parallel.  Such  a  regulator  should  have  sufficient 
capacity  to  take  care  of  all  the  exciters,  whether  it  is  necessary  to 
operate  them  all  at  one  time  or  not.  EquaUzing  rheostats  must 
also  be  provided  with  such  an  arrangement  in  order  that  each 
exciter  shall  carry  its  share  of  the  load.  The  full  field  voltage  of 
one  exciter  may,  for  example,  be  considerably  higher  than  another 
and  it  may  build  up  quicker  when  its  rheostat  is  short-circuited 
by  the  automatic  regulator.  Assuming  that  the  field  rheostats  of 
the  two  exciters  are  set  so  that,  with  the  regulator  contacts  open, 
the  voltages  are  equal,  the  more  sluggish  exciter  will  tend  to  main- 
tain its  voltage  at  a  lower  point  than  the  more  active  one.  The 
contacts,  of  course,  open  and  close  at  the  same  speed  on  both. 
The  more  active  exciter  would,  therefore,  tend  to  take  more  than 
its  share  of  the  load.  To  cause  proper  division,  the  resistance  in 
the  field  circuit  of  the  more  active  naachine  should  be  increased. 
When  an  exciter  requires  more  than  one  relay,  the  resistance  of  its 


Digitized  by 


Google 


VOLTAGE  REGULATION  367 

field  rheostats  is  divided  between  the  relays  and  a  change  in  posi- 
tion of  the  movable  arm  would  unbalance  the  load  on  the  different 
contacts.  An  external  resistance  is,  therefore,  provided  called 
the  equalizing  rheostat  and  inserted  in  the  field  circuit  of  the  more 
active  exciter  (usually  the  higher  speed),  as  shown  in  the  dia- 
gram. EquaUzing  rheostats  are  required  for  all  but  one  of 
several  exciters  in  parallel.  Compoimd  wound  exciters  in  par- 
allel are  also  provided  with  equaUzer  connections  in  the  same  way 
as  other  D.C.  generators. 

It  is  also  possible  to  operate  a  conunon  regulator  in  connection 
with  two  or  more  exciters  when  these  are  not  operated  in  parallel 
on  the  exciter  bus,  although  such  an  arrangement  is  not  recom- 
mended as  the  best  operating  conditions.  With  certain  modifi- 
cations, the  connections  are  just  the  same  as  if  the  exciters  were 
in  parallel.  If  the  exciters  to  be  thus  operated  have  similar  char- 
acteristics, very  satisfactory  regulation  will  probably  be  obtained 
over  the  whole  saturation  range,  but  if  the  exciters  have  different 
characteristics,  it  may  happen  that  if  satisfactory  parallel  opera- 
tion of  the  alternators  is  obtained  at  one  point  of  the  saturation 
curve  of  the  exciters,  successful  operation  will  probably  not  be 
obtained  at  a  different  point.  Under  various  load  conditions  it 
will,  therefore,  be  necessary  for  the  operator  to  either  adjust  the 
generator  field  rheostats  or  the  equahzing  rheostats,  which  should 
be  provided  as  with  the  previous  arrangement. 

A  third  arrangement  is  that  of  individual  regulator  operation. 
In  large  central  stations,  where  there  are  installed  a  large  number 
of  A.C.  generators  and  exciters,  and  it  is  desired  to  operate  the 
generators  in  parallel  but  not  on  the  exciters,  each  exciter  being 
arranged  to  excite  its  own  individual  generator  fsee  Fig.  214,  page 
350),  it  is  possible  to  operate  a  voltage  regulator  on  each  combina- 
tion of  generator  and  exciter.  The  generator,  exciter  and  regu- 
lator then  form  an  operating  unit  in  itself,  and  can  be  operated 
without  affecting  the  operation  of  the  other  units.  This  is  accom- 
plished by  simply  placing  a  current  transformer  in  the  opposite 
phase  from  that  to  which  the  potential  transformer  for  each  regu- 
lator is  connected  (Fig.  220). 

At  unity  power-factor  the  phase  angle  between  the  current  and 
the  potential  transformer  acting  on  the  regulator  magnet  core  is 
90®  and  the  current  winding  of  the  regulator  has  no  effect  upon 
the  voltage  of  the  regulator.     However,  should  the  voltage  of 


Digitized  by 


Google 


368 


ELECTRICAL  EQUIPMENT 


one  alternator  tend  to  increase  above  that  of  any  of  the  others, 
a  circulating  current  would  flow  between  this  alternator  and 
the  ones  having  the  lower  voltage.  This  exchange  current,  of 
course,  Would  be  out  of  phase  with  the  voltage  and,  therefore, 
would  swing  the  current  in  the  current  coil  of  the  A.C.  magnet 
in  phase  with  that  of  potential  current  of  this  magnet.  This 
would  cause  the  regulator  on  this  unit  to  reduce  the  generator 
voltage,  which  would,  of  course,  eliminate  the  possibility  of  any 
cross  currents  between  the  different  alternators  operating  upon 
the  bus-bars.    Of  coiu*se,  if  the  voltage  on  one  machine  tended  to 


EseilerFMd 


FiQ.  220. — Individual  T.A.  Regulator  Operation  with  Exciters  not  in  ParaUel. 
Main  Generators  Operating  in  Parallel. 

drop,  the  regulator  would  operate  in  the  opposite  direction,  causing 
the  voltage  on  this  generator  to  rise,  which  would  aldo  eliminate 
the  above-mentioned  cross  currents. 

Very  complete  instruction  for  the  connection  and  operation  of 
the  diflferent  forms  of  T.A.  regulators  can  be  obtained  from  the 
manufacturer. 

Line  Drop  Compensation.  Compensation  for  line  drop  may 
also  be  obtained  with  these  regulators.  For  ordinary  installations 
the  compensating  winding  on  the  alternating  current  control 
magnet  is  connected  to  a  current  transformer  in  the  main  feeder. 


Digitized  by 


Google 


VOLTAGE  REGULATION 


A  dial  switch  is  provided  by  which  the  strength  of  the  alternating- 
current  control  magnet  may  be  varied  and  the  regulator  made  to 
compensate  for  any  desired  line  drop  up  to  15  per  cent,  according 
to  the  line  requirements. 

This  arrangement  is  very  satisfactory  for  general  use  but  where 
the  power-factor  of  the  load  has  a  wide  range  of  variation,  as  in 


IblnContMU 


Fig.  221. — ConnectioDB  for  Line  Drop  Ck)mpeDsator. 

transmission  lines,  better  results  can  be  obtained  with  a  special 
line  drop  compensator,  adapted  to  the  regulator.  This  compen- 
sator (see  diagram  Fig.  221),  has  two  dial  switches  which  are  con- 
nected to  a  number  of  taps  of  a  resistance  and  a  reactance  coil,  so 
that  the  value  of  these  can  be  adjusted  to  compensate  accurately 
for  line  losses  with  loads  of  varjring  power-factor. 

KR  System  of  Regulation.  This  system  is  particularly 
adapted  to  plants  where  it  is  necessary  to  maintain  a  constant 
exciter  voltage  as  in  cases  where  it  is  desirable  to  operate  motors 
and  other  auxiliary  station  apparatus  from  the  exciter  bus.  This 
system  also  permits  of  the  use  of  a  storage  battery  in  multiple 
with  the  main  exciters. 

By  teferring  to  Fig.  222  it  will  be  noted  that  there  is  a  third 
bus  employed  and  a  D.C.  booster  connected  between  this  bus  and 
one  of  the  exciter  busses.  The  main  generator  fields  then  are 
connected  across  the  outside  bus,  the  voltage  of  which  is  deter- 
mined by  the  voltage  of  the  booster.  This  booster  is  usually 
excited  from  a  separate  exciter  whose  field  is  connected  from  the 
neutral  of  the  above-mentioned  battery  and  the  neutral  of  a  set 


Digitized  by 


Google 


370 


ELECTRICAL  EQUIPMENT 


of  resistances  marked  72-1,  ft-2,  and  fi-3,  respectively.  These 
resistances  in  series  are  connected  in  parallel  with  the  storage 
battery  and  the  main  exciter.  The  booster  exciter  field  connec- 
tion is  made  between  resistance  R-1  and  R-2,  while  resistance  R-3 
is  short-circuited  by  means  of  the  regulator  relay  contacts.  These 
resistances  are  so  proportioned  that  R-l  is  considerably  greater 
than  R-2  J  and  that  R-2  plus  R-S  is  greater  than  R-l.  It  will 
be  readily  seen  that  when  R-3  is  short-circuited  by  the  regulator 
contacts  the  direction  of  excitation  upon  the  booster  exciter  field 


L4J  ^t\ 


KirHtrr  - 


EsilU«T 


Swkteh  ''A" 


FflOHT  VIEW  KoB  TntluotlYO 

Fig.  222. — Type  T.A.  Voltage  Regulator  in  Connection  with  KR  System  of 

Regulation. 

will  be  in  one  direction;  and  when  this  resistance  jR-3  is  inserted 
in  circuit  by  means  of  the  relay  contacts  being  open,  the  direc- 
tion of  excitation  through  the  exciter  field  will  be  in  the  opposite 
direction. 

The  design  of  the  above  resistance  is  also  such  that  there  will  be 
full  excitation  upon  the  booster  exciter  in  each  instance,  making  it 
possible  to  obtain  the  full  boosting  and  bucking  condition  upon  the 
main  D.C.  booster.  Assuming  that  the  voltage  of  the  main 
exciters  is  250  and  that  the  D.C.  booster  is  capable  of  giving  50 
volts  in  each  direction  it  will  at  once  be  noted  that  the  voltage 


Digitized  by 


Google 


VOLTAGE  REGULATION 


371 


obtainable  across  the  main  generator  fields  will  be  from  200  (the 
difference  between  250  and  60)  to  300  volts  (the  sum  of  250  and  50 
volts). 

High-voltage,  High-current  Relays.  A  cut-out  relay  has  been 
devised  to  be  used  in  connection  with  T.A.  regulators  for  guarding 
against  short  circuits  and  voltage  rises  in  transmission  systems.  If 
a  voltage  regulator  is  Used  and  a  short  circuit  should  occur  some- 
where on  the  system, — ^for  example,  in  the  transmission  lines, — the 
action  of  the  regulator  would  naturally  be  to  deUver  the  maximum 
excitation  to  the  fields  of  the  exciters  and  generators,  so  as  to  keep 
up  the  voltage  of  the  system.    This,  in  turn,  necessitates  that  the 


Fig.  223. — Connections  of  High-voltage,  High-current  Cut-out  Relay  with 
Type  T.A.  Voltage  Regulator  and  Two  Exciters  in  Parallel. 

governors  of  the  prime  movers  be  wide  open,  and  if  the  short  cir- 
cuit should  be  suddenly  relieved,  the  voltage  often  rises  to  very 
high  values,  owing  to  the  time  element  involved  in  closing  the 
governors  and  in  demagnetizing  the  fields.  The  connections  for  a 
high-voltage,  high-current  relay  operating  in  connection  with  two 
exciters  and  one  T.A.  regulator  are  shown  in  Fig.  223.  The  relay 
is  provided  with  a  current  coil  and  a  potential  coil,  and  will 
automatically  insert  resistance  in  the  exciter  field  and  thus  reduce 
the  exciter  voltage  in  case  of  excessive  loads  or  voltages  on  the 
main  system. 

Synchronous  Condenser  Regulation.    The  question  of  regula- 
tion of  large  high-voltage  systems  involves  a  number  of  diflScul- 


Digitized  by 


Google 


372  ELECTRICAL  EQUIPMENT 

ties  not  encountered  in  low-voltage  work.  In  the  latter  case  the 
energy  loss  is  generally  the  limiting  factor  and  the  regulation  can 
often  be  improved  by  installing  larger  conductors,  which  at  the 
same  time  will  reduce  the  line  loss.  With  high-voltage  systems 
the  gain  of  doing  so  is  very  slight  and  other  means  must  be  resorted 
to  for  keeping  the  regulation  within  commercial  limits.  The 
effect  of  the  inductance  and  capacity  of  the  line  causes  the  voltage 
to  vary  within  very  wide  limits  from  full  to  no  load.  At  no  load 
the  large  capacity  current  causes  a  rise  of  voltage  from  the  gener- 
ating station  to  the  receiving  end,  while  at  full  load  the  lagging 
inductive  current  taken  by  the  load,  in  general,  more  than  offsets 
the  effect  of  the  capacity  current  and  causes  a  drop  of  voltage 
from  the  generating  station  to  the  receiving  end.  It  is  evident 
then  that  by  installing  a  synchronous  condenser  at  the  receiving 
end  and  by  taking  advantages  of  the  characteristics  of  this  ma- 
chine, the  receiving  voltage  can  be  kept  constant  at  a  determined 
value  or  approximately  so,  by  adjusting  the  synchronous  con- 
denser field  causing  the  condenser  to  draw  a  lagging  current  from 
the  line  at  no  load  and  a  leading  current  at  full;  thus,  by  varying 
the  power-factor. 

The  automatic  regulation  of  the  condenser  field  current  is 
readily  accomplished  by  means  of  a  T.A.  regulator.  In  this 
instance  the  regulator  does  not,  therefore,  hold  a  constant  power- 
factor,  but,  by  varjdng  the  same,  holds  a  constant  A.C.  voltage 
provided  there  is  the  proper  capacity  in  the  synchronous  con- 
denser upon  which  it  is  operating.  The  regulator  endeavors  to 
hold  just  as  much  leading  current  upon  the  condenser  as  there  is 
lagging  current  upon  the  main  transmission  line;  or  else  it  will 
endeavor  to  maintain  the  proper  lagging  current  to  counteract 
for  any  leading  current  that  exists  upon  the  transmission  system. 
The  connections  and  adjustment  for  the  regulator  are  the  same 
as  when  being  used  upon  an  A.C.  generator  with  the  exception 
that  greater  care  should  be  exercised  in  the  adjustment. 

In  a  system  of  this  kind,  if  the  synchronous  condenser  has  not 
ample  capacity,  there  is  danger  of  burning  out  the  fields,  due  to 
the  fact  that  the  regulator  is  trying  to  maintain  constant  A.C. 
voltage  upon  the  system.  It  is  verj'  important,  therefore,  that 
the  highest  safe  voltage  at  which  to  operate  the  condenser  fields 
be  determined,  and  the  regulator  adjusted  for  this  limiting  value, 
which  may  be  about  135  volts  for  a  125-volt  excitation. 


Digitized  by 


Google 


TRANSFORMERS  373 

The  regulator  then  cannot  hold  a  higher  voltage  than  135, 
and  should  the  voltage  reach  this  value  and  tend  to  go  higher,  the 
regulator  would  maintain  a  constant  exciter  voltage  of  this  value 
of  136;  but  the  A.C.  voltage  would  necessarily  drop  due  to  the 
fact  that  it  would  be  requiring  a  higher  exciter  voltage  than 
this  value  in  order  to  maintain  the  A.C.  voltage  for  which  the  reg- 
ulator might  be  adjusted.  The  above  value  of  135  is  selected  only 
as  a  matter  of  convenience  and  the  regulator  may  be  set  for 
whatever  value  it  is  safe  to  operate  the  condenser  fields.  If 
they  could  be  operated  to  as  high  as  145  volts  the  regulator  should 
be  adjusted  at  145  instead  of  135. 

For  a  fmlher  study  of  the  subject  of  "  Sjrnchronous  Condenser 
Regulation,"  the  reader  is  referred  to  an  article  by  F.  W.  Peek,  Jr., 
in  the  General  Electric  Review  for  Jime,  1913. 

6.  TRANSFORMERS' 

Fundamental  Principles.  A  constant  potential  transformer 
consists  essentially  of  an  iron  core  upon  which  are  wound  two 
wiadings,  a  primary  and  a  secondary.  When  one  \rinding  is 
connected  to  an  alternating-current  supply  of  power,  an  alternating 
magnetic  flux  is  excited  in  the  iron  core  and  an  alternating  voltage 
is  induced  in  the  secondary  winding,  as  its  turns  are  surrounded  by 
the  same  flux  as  the  primary.  If  the  now  secondary  winding 
is  closed  through  a  resistance  or  other  load  a  current  will  flow 
therein. 

In  an  "  ideal "  transformer,  power  would  be  transmitted 
from  primary  to  secondary  without  any  loss.  In  actual  practice, 
however,  this  is  not  quite  possible  on  account  of  the  losses  which 
take  place  in  the  iron  core  and  the  windings.  Similarly,  in  an 
ideal  transformer,  the  ratio  of  primary  to  secondary  voltage 
would  be  equal  to  the  ratio  of  the  number  of  turns  in  the  respective 
windings.  In  a  real  transformer  there  is,  however,  also  a  voltage 
drop  caused  by  the  resistance  and  leakage  reactance  of  the  wind- 
ings. This  reactance  is  due  to  the  leakage  flux  which  links  with 
the  turns  or  part  of  the  turns  of  one  winding  only. 

The  action  of  a  transformer  can  best  be  understood  by  means 
of  a  vector  diagram  (Fig.  224).    Consider  first  the  open-circuit 

'  Part  of  this  section  is  taken  from  an  article  on  "  Transformer  Connec- 
tions ''  in  the  General  Electric  Review  by  one  of  the  authors  and  Mr.  L.  F. 
Blume. 


Digitized  by 


Google 


374 


ELECTRICAL  EQUIPMENT 


condition,  i.e.,  no  current  flowing  in  the  secondary  winding.  The 
primary  e.m.f.  OB,  causes  an  exciting  current  OMi  to  flow,  this 
current  consisting  of  two  components  MMi,  and  OM.  The  com- 
ponent MM  I  is  in  phase  with  the  e.m.f.  and  supplies  the  iron  core 
loss  due  to  hysteresis  and  eddy  currents,  while  OM,  which  is  in 
quadrature  with  the  e.m.f.,  represents  the  magnetizing  current 
and  is  thus  in  phase  with  the  flux.  The 
secondary  e.m.f.  OB2  is  exactly  opposite 
^c  the  primary  in   phase  and  its  value  is 

equal  to  that  of  the  primary  times  the  in- 
verse ratio  of  the  turns  of  the  two  wind- 
ings. 

Suppose  now  that  the  transformer  is 
loaded,  in  which  case  a  secondary  current 
OA2  will  flow,  proportional  to  the  load. 
If  the  load  was  non-inductive  this  current 
would  be  in  phase  with  the  secondary 
terminal  e.m.f.  OD2,  thus  lagging  behind 
the  induced  e.m.f.  (^82^  due  to  the  leak- 
age reactance.  In  this  particular  case  how- 
ever, the  load  is  inductive  and  the  current 
OA2  lags f behind  the  terminal  e.m.f.  0D2^ 
degrees,  the  corresponding  power  factor  of 
the  load  being  cos  0.  The  secondary  ter- 
minal e.m.f.  OD2  is  less  than  the  induced 
e.m.f.  OB2  on  account  of  the  resistance 
drop  B2C2  and  the  reactance  drop  C2D2. 
These  values  are  the  product  of  the  secondary  current  times  the 
resistance  and  the  reactance,  respectively,  of  the  secondary  wind- 
ing, the  former  being  in  phase  with  the  current  and  the  latter  in 
quadrature. 

When  the  secondary  current  flows  it  disturbs  the  equilibrium 
by  tending  to  demagnetize  the  core,  and  the  primary  current 
increase  until,  in  addition  to  'the  exciting  current  OAf  1,  a  current 
flows,  the  magnetizing  effect  of  which  just  balances  the  magnet- 
izing effect  of  the  secondary  current.  This  additional  current 
is  represented  by  MiAi  and  it  is  just  equal  and  opposite  to  the 
secondary  current  OA2  times  the  inverse  ratio  of  the  number  of 
turns  in  the  windings. 

The  total  primary  current  0-4 1  is,  therefore,  seen  to  be  corn* 


Fig.  224.  —  Theoretical 
Transformer  Diagram. 


Digitized  by 


Google 


TRANSFORMERS  375 

posed  of  the  exciting  current  OMi,  which  is  practically  constant 
for  all  loads  and  the  load  current  MiAi.  The  impressed  primary 
e.m.f.  ODi  is  a  Uttle  greater  than  the  primary  counter  e.m.f. 
OBi  on  account  of  the  resistance  drop  BiCi  and  the  reactance 
drop  CiDi,  the  values  being  the  product  of  the  primary  current 
OAi  times  the  resistance  and  leakage  reactance^  respectively,  of 
the  primary  winding.  The  former  is  in  phase  with  the  current, 
the  latter  in  quadrature. 

Induced  e.mi.  The  relation  between  the  coimter  e.m.f.  of 
a  transformer  and  the  various  factors,  such  as  flux  density,  number 
of  turns,  frequency,  etc.,  are  determined  by  the  following  formula: 

£=4.44X/XnX«X10-8; 

in  which  J? = mean  effective  e.m.f.; 

/= frequency  in  cylces  per  second; 

n=  total  number  of  turns  of  the  primary  winding; 

^= total  magnetic  flux  in  maxwells. 

This  equation  is  based  on  the  assumption  that  the  e.m.f.  is  a 
true  sine  wave. 

Ratio.  The  A.  I.  E.  E.  Standardization  Rules  state  that 
"The  voltage  ratio  of  a  transformer  is  the  ratio  of  the  r.m.s. 
primary  terminal  voltage  to  the  r.m.s.  secondary  terminal  voltage 
under  specified  conditions  of  load."  It  also  defines  "  the  ratio  of 
a  transformer,  unless  otherwise  specified,  as  the  ratio  of  the 
number  in  turns  in  the  high-voltage  winding  to  that  in  the  low- 
voltage  winding,  i.e.,  the  turn-ratio." 

The  two  ratios  are  equal  when  one  of  the  windings  is  open 
and  the  transformer  does  not  carry  any  load.  When  loaded,  the 
resistance  and  inductance  of  the  windings  cause  a  drop  in  the 
voltage,  thus  modifying  the  ratio  of  transformation  sUghtly. 

The  ratio  of  a  transformer  refers,  of  course,  to  the  turns 
which  are  connected  in  series,  high-voltage  as  well  as  low-voltage. 
In  many  instances  it  is  desirable  for  the  sake  of  interchangeability 
and  standardization  to  split  up  the  windings  in  groups  of  sections 
which  may  be  connected  either  in  series,  parallel,  or  series-parallel. 
This  is  almost  always  the  case  with  distributing  transformers, 


Digitized  by 


Google 


376 


ELECTRICAL  EQUIPMENT 


where  the   low-voltage  winding  may  be  connected  for  115-230 
volts.    This  makes  possible  the  following  connections: 


HIOH-VOLTAOB. 

LOW-VOI.TAQB. 

Rfttio. 

Connection. 

Voltage. 

2300 
2300 

ParaUel 
Series 

115 
230 

20:1 
10:1 

For  transformers  of  very  high  voltages  it  is  often  requested 
that  the  high-voltage  winding  be  designed  for  series-parallel  con- 
nection. So,  for  example,  by  designing  a  transformer  with  a  high- 
voltage  of  say  110,000-66,000  volts,  it  is  possible  to  operate  the 
system  at  the  lower  voltage  mitil  the  load  has  increased  to  a  point 
necessitating  a  change-over  to  the  higher  transmission  voltage. 

Magnetizing  Current.  The  effect  of  the  magnetizing  current 
in  transformers  sometimes  leads  to  the  question  of  considering  its 
proper  limitations.  It  was  previously  shown  that  this  current  is 
wattless  with  the  exception  of  a  small  PR  loss  and  has  Uttle  influ- 
ence on  the  values  of  the  total  current  in  the  transformer  when  it 
is  operating  at  full  load,  but  as  the  load  decreases  the  effect  be- 
comes more  prominent  until  at  no  load  it  is  most  noticeable, 
and  the  power-factor  naturally  very  low.  This  is  an  important 
point  where  a  large  number  of  small  transformers  are  operating 
on  a  system,  and  for  such  cases  it  has  become  quite  conmion  to 
limit  the  magnetizing  current  to  a  value  not  exceeding  about  10 
per  cent  of  the  full-load  current,  a  value  which  cannot  be  consid- 
ered detrimental  to  the  system. 

There  is  also  another  limitation  which  is  given  consideration  in 
connection  with  large  transformer  units.  Such  transformers  are 
nowadays  built  of  high-grade  steel,  which  has  a  core  loss  per 
pound  much  less  than  formerly,  and  this  has  in  many  instances 
made  it  advisable  to  increase  the  core  densities.  If,  however, 
these  are  increased  much  above  the  bend  of  the  saturation  curve 
an  unstable  operation  is  Uable  to  follow,  and  for  such  conditions, 
the  limitation  of  the  magnetizing  current  is  governed  by  the  per- 
missible core  density,  usually  around  90.000  lines  per  square  inch. 
With  over-voltages  causing  a  saturation  of  the  core  the  mag- 
netizing current  increases  very  rapidly,  but  with  the  above  limi- 


Digitized  by 


Google 


TRANSFORMERS  377 

tation,  based  on  normal  voltage,  an  over-voltage  of  around  10  per 
cent,  which  is  to  be  expected,  should  not  cause  an  excessive  mag- 
netising current. 

With  regard  to  efficiency  and  regulation  the  effect  of  the  mag- 
netizing current  is  insignificant. 

Reactance.  The  percentage  of  the  total  flux  that  links  with 
the  primary  but  does  not  link  with  the  secondary  winding,  plus 
that  which  links  with  the  secondary  but  not  the  primary,  is  the 
per  cent  reactance  of  a  transformer.  Thus,  if  96  per  cent  of  the 
primary  flux  cuts  both  primary  and  secondary,  the  transformer  is 
said  to  have  a  5  per  cent  inherent  reactance. 

The  factors  affected  by  the  reactance  of  a  transformer  are  its 
regulation,  parallel  operation,  mechanical  stresses  and  eddy- 
current  losses.  A  low-reactance  transformer  has  naturally  a 
better  regulation  than  one  of  high  reactance,  especially  for  highly 
inductive  loads,  and  in  order  to  obtain  a  good  voltage  regulation 
it  waa  formerly  the  custom  to  design  transformers  with  a  reac- 
tance as  low  as  1}  to  2  per  cent.  Such  a  low  reactance  is,  however, 
often  detrimental  to  the  safe  operation  of  a  transformer  from  the 
mechanical  point  of  view.  If  a  short  circuit  sho\ild  occur  at  the 
secondary  terminals  of  a  transformer,  and  the  power  supply  at 
the  primary  is  sufficient  to  maintain  the  primary  terminal  voltage^ 
as  may  be  the  case  in  very  large  generating  systems,  the  primary 
and  secondary  currents  of  the  transformer  are  limited  by  the 
impedance  only,  and  with  the  exception  of  very  low  reactance 
transformers  it  is  essentially  the  reactance  which  determines  the 
total  impedance  and  thus  the  short-circuit  current. 

As  the  primary  and  secondary  currents  are  opposite  in  phase, 
they  repel  each  other,  the  force  being  approximately  propor- 
tional to  the  square  of  the  current.  It  therefore  follows  that  the 
repulsion,  which  is  small  at  full  load,  may  reach  enormous  values 
under  short-circuit  conditions  if  the  transformer  reactance  is  low. 
For  example,  in  a  transformer  having  a  2  per  cent  reactance  the 
short-circuit  current  will  be  50  times  normal  and  the  mechanical 
stresses  will  increase  as  the  square  of  this  or  2500  times,  amount- 
ing to  many  hundred  tons.  This  clearly  illustrates  the  necessity 
of  a  very  rigid  construction  and  also  the  advisability  of  reducing 
the  short  circuit  to  a  safe  value.  This  may  be  done  by  increasing 
the  transformer  reactance,  and  modem  practice  tends  toward  the 
use  of  considerably  higher  internal  reactances  than  was  formerly 


Digitized  by 


Google 


378  ELECTRICAL  EQUIPMENT 

used.  In  general  it  may  be  said  that  it  is  usually  difficult  to  go 
above  an  8  to  10  per  cent  reactance  in  a  60-cycle  moderate  size 
transformer  (1000  to  2000  Kv. A.),  without  undue  eddy-current 
losses,  and  that  the  allowable  maximum  would  be  considerably 
less  than  this  in  low  voltage  designs.  For  26-cycle  transformers, 
a  higher  reactance  may  be  obtained,  since  the  eddy-current  losses 
are,  of  course,  less  at  a  given  density. 

Regulation.  The  regulation  of  a  constant-potential  trans- 
former is  defined  by  the  A.I.E.E.  rules  as  the  difference  between 
the  no-load  and  rated-load  values  of  the  secondary  terminal 
voltage  at  specified  power-factor  (with  constant  primary  impressed 
terminal  voltage),  expressed  in  per  cent  of  the  rated-load  secondary 
voltage,  the  primary  voltage  being  adjusted  to  such  a  value  that 
the  transformer  delivers  rated  output  at  rated  secondary  voltage. 
All  parts  of  the  transformer  affecting  the  regulation  should  be 
maintained  at  constant  temperature  between  the  two  loads,  and 
where  the  influence  of  temperature  is  of  consequence,  a  reference 
temperature  of  75^  C.  shall  be  considered  as  standard.  If  a 
change  of  temperature  occurs  during  the  test,  the  result  shall  be 
corrected  to  the  above  reference  temperature. 

For  non-inductive  load  the  regulation  varies  approximately 
from  less  than  1  per  cent  for  large  sizes  to  around  3  per  cent  for 
smaller  units.  For  inductive  load  it  is  naturally  higher.  It  can 
be  determined  by  loading  the  transformer  and  measuring  the 
change  in  voltage  with  change  in  load  at  specified  power- 
factor. 

The  A.I.E.E.  recommends  the  foDowing  method  for  comput- 
ing the  regulation  for  any  specified  load  and  power-factor  from 
the  measured  impedance  watts  and  impedance  volts. 

Let     P= impedance  watts,  as  measured  from  short-circuit  test 
and  corrected  to  75°  C; 

Eg = impedance  volts  ; 

7X= reactance  drop  in  volts; 

Z= rated  primary  current; 

£= rated  primary  voltage; 

gr=per  cent  drop  in  phase  with  current; 

(y,=per  cent  drop  in  quadrature  with  current. 


Digitized  by 


Google 


TRANSFORMERS  379 

IX- 


■=^/-.•-Q^■ 


Then: 

1.  For  unity  power-factor,  we  have  approximately: 

Per  cent  regulation  =  gr+^. 

2.  For  inductive  loads,  where  the  power-factor  (cos  if>)  equals 
m  and  the  reactive  factor  (sin  ^)  equals  n.    Per  cent  regulation 

^mqr+nqx-\ 200 ' 

Core  and  Shell  Tjrpe«  Transformers  are  of  two  fundamental 
designs,  namely:  The  shell  and  the  core  type,  and  occasionally  a 
combination  of  these  two  is  also  used  (Fig.  225).  In  the  shell  type 
the  iron  circuit  surrounds  the  transformer  coils,  while  in  the  core 
type  the  copper  windings  surround  the  iron  core.  While  the  sheD- 
type  transformers  have  been  most  extensively  used  in  the  past, 
core-type  transformers  are  now  built  for  the  largest  sizes,  and  are 
rapidly  superseding  the  former  type.  With  the  core  type  design, 
the  arrangements  of  cores  and  the  circular  coils  present  a  con- 
struction which  offers  a  maximum  resistance  to  the  mechanical 
distorting  forces.  This  mechanical  strength,  combined  with  the 
inherent  reactance  of  this  type  of  transformer,  produces  a  unit 
which  is  exceptionally  able  to  withstand  severe  service.  On  the 
other  hand,  the  circular  coils  can  readily  be  insulated  for  the  very 
highest  voltages  in  use. 

Method  of  Cooling.  Transformers  may  be  divided  into  four 
classes,  depending  upon  the  method  of  cooling,  viz.,  natural  draft, 
air  blast,  oil  immersed  self-cooled  and  oil  immersed  water-cooled. 
Natural-draft  transformers  have  the  core  and  coils  exposed  directly 
to  the  air,  and  depend  entirely  upon  the  natural  circulation  of 
the  air  for  their  cooling.  They  are  built  only  for  very  low  voltages 
and  stnall  sizes.  Air-blast  transformers  depend  upon  a  forced  cir- 
culation of  air  over  the  surface  of  the  core  and  coils  to  carry  away 


Digitized  by 


Google 


380 


ELECTRICAL  EQUIPMENT 


the  heat.  They  may  be  built  for  large  capacities,  but  the  voltage 
rarely  exceeds  30,000  because  of  the  difficulty. of  insulating  them 
properly. 

Oil  immersed  self-cooled  or  water-cooled  transfonners  are  gen- 
erally used  with  hydro-electric  power  developments,  the  latter  in 


Single-Phase 

Three-Phase 

Vertical 

J 

1 

1 

^ 

^ 

1 

1   1 

1 

= 

View 

n 

1 



= 

n 

1 

1   1 

1 

= 

OopeTJT» 


Shell  Trpe 


Combimitioo 

Core  i  Shell 

Type 


Fig.  225. — Different  Types  of  Transformer  Core  Construction. 

the  generating  station,  while  either  type  may  be  used  with  the  sub- 
stations, depending  upon  the  availability  of  cooling  water.  Both 
types  are  built  for  the  largest  capacities  and  the  highest  voltages. 
Self-cooled  oil  immersed  transformers  have  the  core  and  cofls 
immersed  in  a  tank  of  oil,  the  tank  usually  being  corrugated  so  as 
to  increase  the  surface  available  for  dissipating  the  heat  generated 


Digitized  by 


Google 


TRANSFORMERS  381 

in  the  core  a^d  coils.  Sometimes  external  tubes  or  radiators, 
through  which  the  oil  circulates,  are  used  for  the  same  purp)08e. 
Water-cooled  oil  immersed  transformers  depend  upon  the  circu- 
lation of  water  through  a  coil  placed  in  the  top  of  the  tank  to  carry 
away  the  heat  from  the  oil,  about  J  gallon  being  required  per 
minute  per  Kw.  loss,  the  temperature  of  the  incoming  water  being 
15°  C. 

For  conditions  where  long  and  definite  periods  of  light  and 
heavy  load  occur,  as  in  small  winter  and  large  sunmier  service, 
a  combination  self-cooling  and  water-cooUng  design  has  been 
provided.  Such  transformers  are  placed  in  the  regular  sheet- 
steel  tanks  of  the  self-cooled  design,  excepting  that  they  have 
smaller  surfaces  and  are  in  addition  provided  with  water-cooling 
coils  to  take  care  of  the  super-load.  They  can  readily  be  designed 
to  carry  50  per  cent  of  the  maximum  load  without  water  circula- 
tion and  not  exceed  the  rated  temperature  rise.  The  increase  in 
the  cost  over  the  water-cooled  design  is  slight  and  will  often  be 
found  a  good  investment  when  cooling  water  has  any  appreciable 
value. 

Special  precautions  must  naturally  be  taken  to  protect  trans- 
formers of  the  outdoor  type  both  from  the  extreme  heat  and  from 
the  cold  in  the  winter.  The  former  can  readily  be  obtained 
by  providing  sunshades,  and  in  certain  instances  very  good  results 
have  been  obtained  by  simply  painting  the  tanks  white.  It  is 
more  difficult,  however,  to  provide  for  the  cold  winter  temperatures, 
especially  with  water-cooled  transformers.  With  the  trans- 
formers in  service  there  seems  to  be  no  danger  of  freezing,  and  if 
such  should  be  the  case  some  sort  of  heating  grids  could  readily 
be  provided  in  the  bottom  of  the  tanks.  The  main  difficulty  lies 
in  the  formation  of  moisture  which  takes  place  when  the  temper- 
ature of  the  transformer  is  allowed  to  faU  below  that  of  the  sur- 
rounding air;  this  appUes  also  to  indoor  transformers.  Pre- 
cautions must,  therefore,  be  taken  that  this  does  not  happen,  and 
may  be  accomplished  by  either  reducing  the  water  rate  at  times  of 
cold  weather,  or  by  using  the  cooling  water  over  and  over  again. 
An  oil  with  special  low  freezing-point  may  be  used  in  transformers 
in  rare  locations  experiencing  extreme  low  temperatures. 

Single  and  Polyphase  Transformers.  Transformers  are  made 
either  as  single  or  polyphase  units,  the  latter  being  generally  of  the 
three-phase  type.    The  single-phase  design  is  by  far  the  most 


Digitized  by 


Google 


382  ELECTRICAL  EQUIPMENT 

flexible,  as  by  different  connections  an}'-  combination  can  be 
obtained.  Economical  considerations  are,  however,  often  the 
determining  factor  in  deciding  on  what  type  to  use. 

Three-phase  designs  may  be  connected  either  in  delta  or  Y, 
and  the  units  may  be  either  of  the  sheD  or  the  core  type  con- 
struction. In  delta-connected  shell-type  transformers,  should  one 
phase  be  damaged,  it  is  possible  to  operate  the  remaining  two 
phases  in  open-delta  at  58  per  cent  of  the  combined  capacity,  by 
simply  disconnecting  the  damaged  unit  of  the  three  single-phase 
transformers,  or  in  the  case  of  three-phase  shell-type  units  by  dis- 
connecting and  short-circuiting  the  damaged  phase,  both  high- 
and  low-voltage.  This  will  reduce  the  flux  passing  through  the  part 
of  the  core  surrounded  by  these  windings  and  limit  the  current  in 
the  damaged  winding  to  a  fraction  of  the  normal  full-load  current. 

Y-connected  shell-type  transformers  of  both  the  single-  and 
three-phase  t3rpes  cannot  be  operated  with  one  phase  damaged, 
except  where  the  neutral  is  grounded,  in  which  case  they  may  be 
operated  at  58  per  cent  of  their  total  capacity  by  short-circuiting 
both  the  high-  and  low-voltage  windings  of  the  damaged  phase. 
Such  a  scheme  is,  nevertheless,  not  very  satisfactory  for  motor 
operations  on  account  of  the  unbalancing  of  the  phases  and  the 
reduced  voltage.  Lights  can,  however,  be  operated  successfully 
by  connecting  them  between  the  live  single-phase  wires  and  the 
neutral. 

In  the  case  of  three-phase  core-type  transformers,  even  though 
the  windings  are  delta-connected,  it  is  impossible  to  operate  when 
one  phase  becomes  short-circuited.  This  is  due  to  the  fact  that 
the  three  phases  are  magnetically  interlinked  in  such  a  manner  that 
any  one  phase  is  a  return  path  for  the  fluxes  in  the  other  two 
phases.  This  means  that  when  one  phase  is  short-circuited  the 
short  circuit  is  transmitted  magnetically  to  the  other  two  phases 
in  such  a  manner  that  when  the  two  phases  are  excited  large 
short-circuit  currents  flow,  the  short-circuit  phase  acting  as  sec- 
ondary and  the  remaining  phases  as  primary.  In  the  three-phase 
shell-type  transformer  this  does  not  occur,  because  the  fluxes  in 
the  three  phases  are  independent  of  each  other,  and,  therefore, 
the  flux  in  one  phase  can  be  reduced  to  zero  without  affecting  the 
other.  However,  if  the  damaged  winding  can  be  open-circuited 
or  removed  from  the  core,  the  transformer  will  operate  satisfactory 
connected  open-delta. 


Digitized  by 


Google 


TRANSFORMERS  383 

Rating.  A  transformer  should  be  rated  by  its  tdlovolt-ampere 
(Kv.A.)  output.  It  is  simply  eqiial  to  the  product  of  the  voltage 
and  current,  and  is,  therefore,  the  same  whether  the  different  coils 
are  connected  in  series  or  parallel.  If  the  load  is  of  imity  power- 
factor,  the  kilowatt  output  is  the  same  as  the  kilovolt-ampere 
output,  but  if  the  power-factor  is  less,  the  kilowatt  output  will  be 
correspondingly  less.  For  example,  a  100  Kv.A.  transformer  will 
have  a  full-load  rating  of  100  Kw.  at  100  per  cent  power-factor, 
90  Kw.  at  90  per  cent  power-factor,  etc. 

The  A.  I.  £.  £.  Standardization  Rules  identify  self-  and  water- 
cooled  oil  immersed  transformers  as  to  Kv.A.  rating  by  their 
maximimi  continuous  capacity  at  55^  rise.  With  an  ambient 
room  temperature  of  40**  C.  air  for  the  former  and  26**  C.  incoming 
water  for  the  latter,  the  observable  temperatures  would  be  95° 
and  80°  C.  respectively.  The  rules  further  specify  that  the  tem- 
perature of  the  windings  of  transformers  is  always  to  be  ascer- 
tained by  Method  II,  i.e.,  the  resistance  method.  (See  page  310, 
"  Rating  of  Generators  ")•  This  method  allows  for -a  correction 
factor  of  10°  C,  so  that  for  self-cooled  transformers  the  hottest- 
spot  temperature  is  limited  to  106°  C.  and  for  water-cooled  to 
90°  C.  The  oil  shall  in  no  case  have  a  temperature,  observable 
by  thermometer,  in  excess  of  90°  C. 

For  air-blast  transformers  the  rules  specify  that  a  correction 
sludl  be  applied  to  the  observed  temperature  rise  of  the  windings, 
and  it  is  to  be  noted  that  air-blast  transformers  constitute  the 
only  instance  wherein  it  is  required  that  a  correction  shall  be  ap- 
plied to  take  into  account  the  precise  ambient  temperature  at 
time  of  the  test.  This  is  due  to  the  difference  in  resistance,  when 
the  temperature  of  the  ingoing  cooling  air  differs  from  that  of  the 
standard  reference.  This  correction  shall  be  the  ratio  of  the 
inferred  absolute  ambient  temperature  of  reference  to  the  inferred 
absolute  temperature  of  the  ingoing  cooling  air,  i.e.,  the  ratio 

274  5 

— -— ^ — r ;  where  t  is  the  ingoing  cooling-air  temperature. 

Thus,  a  cooling-air  temperature  of  30°  C.  would  correspond 

to  an  inferred  absolute  temperature  of  264.6°  on  the  scale  of  copper 

resistivity,  and  the  correction  to  40°  C.  (274.6°  inferred  absolute 

274  6 
temperature)  would  be  =1.04,  making  the  correction  factor 

^04.0 

1.04;  so  that  an  observed  temperature  rise  of  say  60°  C.  at  the 


Digitized  by 


Google 


384  ELECTRICAL  EQUIPMENT 

testing  ambient  temperature  of  30**  C.  would  be  corrected  to 
60X1.04=62°  C,  this  being  the  temperature  rise  which  would 
have  occurred  had  the  test  been  made  with  the  standard  ingoing 
cooling-air  temperature  of  40**  C. 

Efficiency.  The  efficiency  of  a  transformer  is  the  ratio  of  the 
kilowatts  output  measured  at  the  secondary  terminals  to  the  kilo- 
watts input  measured  at  the  primary  terminals.  The  difference 
between  these  two  values  equals  the  losses,  which  consist  of  the  no- 
load  losses,  the  PR  losses  and  the  stray-load  losses.  The  no-load 
losses  consist  of  the  hysteresis  and  eddy-current  or  core  loss  in 
the  laminations,  the  PR  loss  due  to  the  exciting  ciurent  and  the 
dielectric  hysteresis  loss  in  the  insulation.  The  PR  losses  should 
include  the  copper  loss  in  all  the  windings,  primary  as  well  as 
secondary,  and  the  stray-load  losses  consist  of  the  eddy-current 
loss  in  the  windings  and  core,  due  to  fluxes  varying  with  the  load. 
They  should  also  include  the  stray  loss  in  other  parts  of  the  trans- 
former. In  determining  these  losses  care  shall  be  taken  that  they 
are  corrected  to  a  reference  temperature  of  76**  C. 

The  efficiency  is  generally  given  at  miity  power-factor,  but 
can  readily  be  figured  out  for  any  power-factor  as  the  losses  are 
independent  of  the  same  a&  long  as  the  Kv.A.  is  not  changed. 
For  example,  assimie  a  1000  Kv.A.  transformer  having  a  total  loss 
of  14  Kw.  or  1.4  per  cent  based  on  1000  Kw.  at  unity  power-factor. 
Based  on  800  Kw.  80  per  cent  power-factor  the  loss  would  be 
1.76  per  cent.  In  the  former  case  the  efficiency  at  full-load  would 
be  98.62  and  the  latter  98.28,  which  illustrates  the  importance  of 
basing  the  efficiency  identically. 

The  efficiency  depends  upon  the  voltage  and  the  sLee  of  the 
unit  and  varies  from  about  97  to  as  high  as  99  per  cent  for  trans- 
formers generally  used  in  hydro-electric  work.  For  26  cycles  the 
losses  are  somewhat  higher  and  the  efficiency  somewhat  lower  on 
account  of  the  larger  amount  of  material  required  for  this  fre- 
quency as  compared  to  60  cycles. 

Sometimes  the  all-day  efficiency  of  a  transformer  is  required 
for  comparison,  and  this  may  readily  be  figured  from  the  following 
simple  formula: 

A]l-D»7  BffloiflDoy-  „    .  „  _     -.. 

KV.A.  HOiiM  per  Day  Output _____ 

KvJL.  Hn.  per  Dbv  Output + 24  X  No-load  Lom+No.  of  9n.x/>i2+Stny-load  Liaa 

Voltage.    In  regard  to  the  use  of  the  terms  high-voltage, 


Digitized  by 


Google 


TRANSFORMERS  385 

low-voltage,  primary  and  secondary,  the  A.I.E.E.  standardiza- 
tion rules  read  as  follows: 

"  The  terms  high-voUage  and  hw-voUage  are  used  to  distinguish 
the  winding  having  the  greater  from  that  having  the  lesser  number 
of  turns.  The  terms  'primary  and  secondary  serve  to  distinguish 
the  windings  in  regard  to  energy  flow,  the  primary  being  that  which 
receives  the  energy  from  the  supply  circuit,  and  the  secondary 
that  which  receives  the  energy  by  induction  from  the  primary." 

The  terms  primary  and  secondary  are,  however,  often  confused, 
and  in  order  to  avoid  any  misunderstanding  it  is  preferable  to 
use  the  terms  high-voltage  and  low-voltage  instead  of  primary 
and  secondary. 

In  every  S3rmmetrical  three-phase  circuit  there  are  two  voltages 
which  should  be  clearly  distinguished: 

(1)  The  voltage  between  lines,  called  the  "  delta-voltage  " 
and  (2)  the  voltage  from  line  to  neutral,  called  the  "  Y-voltage." 
Under  balanced  conditions 

Y-voltage= delta-voltage  divided  by  V3,  and 

Delta-voltage =Y-voltage  times  Vs. 

Transformers  designed  to  be  suitable  for  use  in  either  delta  or 

Y-connection  have,  as  a  rule,  on  the  name  plates  the  line  voltages 

which  apply  for  both  connections.     The  line  voltage  resulting 

from  Y-connection  is  followed  by  the  letter  "  Y  " — ^for  example, 

10  000 
if  the  transformer  voltage  is  given       Jfkn  v>  *^  signifies  that 

both  voltages  are  line  voltages  but  the  latter  is  the  voltage  result- 
ing when  the  transformer  is  connected  in  Y.  The  sjonbol  **  Y  " 
is  used  as  an  abbreviation  to  indicate  that  sufficient  insulation  has 
been  provided  so  that  the.  transformer  may  be  connected  in  Y  for 
the  line  voltage  with  which  the  letter  is  used,  but  this  S3rmbol 
should  not  be  confused  with  "  Y-rvoltage."  The  expressions 
"  delta-voltage  "  and  '*  Y-voltage  "  are  often  loosely  used  for 
"voltage  when  connected  in  delta "  and  "  voltage  when  cgn- 
nected  in  Y  "  and  misunderstandings  are  often  caused  thereby. 
If,  however,  the  facts  are  kept  clearly  m  mind  that  a  "  Y  "  in  the 
voltage  rating  of  a  transformer  stands  for  "  Y-connection  "  and 
that  "  Y-voltage  "  is  only  a  part  of  the  line  voltage,  there  should 
be  no  cause  for  misunderstanding. 

The  transformer  voltage  depends,  of  course,  on  the  nature  of 
the  system.    The  primary  voltage  of  the  step-up  transformers  is, 


Digitized  by 


Google 


386 


ELECTRICAL  EQUIPMENT 


for  example,  governed  by  the  generator  voltage  and  may  be  any- 
thing up  to  13,200  volts.  The  secondary  of  the  step-up  trans- 
formers and  the  primary  of  the  step-down  transformers  is  deter- 
mined by  the  most  economical  transmission  voltage,  which  may 
be  as  high  as  150,000  volts.  The  secondary  of  the  step-down 
transformers  is  finally  governed  by  the  potential  of  the  distrib- 
uting system.  Where  this  is  extensive  its  voltage  may  be  com- 
paratively high,  may  be  33,000  or  even  higher,  while,  for  smaller 
systems  it  may  only  be  2300  volts  and  even  lower.  The  voltages 
generally  used  for  power  transformers  are  as  follows: 


Low-Voltage. 

High-Voltage. 

2,300 

6.600 

11,000 

13,200 

16,500 
22,000 
33,000 
44,000 

66,000 

88,000 

110,000 

150,000 

The  test  voltage  which  shall  be  applied  to  determine  the 
dielectric  strength  of  the  insulation  is  specified  by  the  A.I.E.E. 
rules  as  twice  the  normal  voltage  of  the  circuit  to  which  the 
transformer  is  connected  plus  1000  volts.  The  test  shall  be  made 
at  the  temperature  assumed  under  normal  operation,  and  the 
frequency  of  the  test  circuit  shall  not  be  less  than  the  rated  fre- 
quency of  the  apparatus  tested.  The  duration  of  the  application 
of  the  voltage  shall  be  one  minute,  and  it  shall  be  successively 
applied  between  each  electric  circuit  and  all  other  electric  cir- 
cuits and  metal  parts  grounded.  Inter-connected  pol3rphase 
windings  are  considered  as  one  circuit  and  all  windings  except 
that  under  test  shall  be  connected  to  ground.  Transformers 
which  may  be  used  in  Y-connection  on  three-phase  circuits  shall 
have  the  test  based  on  the  delta  or  line  voltage. 

The  following  exceptions  to  the  above  rule  are  given: 

(1)  Alternating  current  apparatus  connected  to  permanently 
grounded  single-phase  systems,  for  use  on  permanently  grounded 
circuits  of  more  than  300  volts,  shall  be  tested  with  2.73  times  the 
voltage  of  the  circuit  to  ground  plus  1000  volts.  This  does  not, 
however,  refer  to  three-phase  apparatus  with  grounded  neutral. 

(2)  Distributing  transformers  for  primary  pressures  from 
660  to  6000  volts,  the  secondaries  of  which  are  directly  connected 


Digitized  by 


Google 


TRANSFORMERS  387 

to  consumer's  circuits,  shall  be  tested  with  10,000  volts  from 
primary  to  core  and  secondary  combined.  The  secondary  winding 
shall  be  tested  with  twice  normal  voltage  plus  1000  volts. 

Under  certain-  conditions  it  is  permissible  to  test  transformers 
by  inducing  the  required  voltage  in  their  windings,  in  place  of 
using  a  separate  testing  transformer.  By  '^  required  voltage  "  is 
meant  a  voltage  such  that  the  line  end  of  the  windings  shall  receive 
a  test  to  ground  equal  to  that  required  by  the  above  general  rules. 

Transformers  with  *'  graded  "  insulation  shall  be  so  marked, 
and  shall  be  tested  by  inducing  the  required  test  voltage  in  the 
transformer  and  connecting  the  successive  line  leads  to  ground. 
The  term  "  graded  "  is  used  to  indicate  the  emplo3rment  of  less 
insulation  towards  the  end  of  the  windings  where  the  insulation 
stresses  are  low,  i.e.,  towards  the  groimd,  and  more  insulation  at 
the  high-potential  ends.  Such  transformers  usually  have  the  wind- 
ing groimded  within  the  tank  and  all  transformers  so  connected 
shall  be  tested  by  induced  voltage. 

Until  the  adoption  of  the  sphere  gap  as  a  method  of  voltage 
measurement,  transformers  were  generally  tested  by  the  use  of 
the  needle  gap.  This  resulted  in  more  or  less  inconsistent  tests, 
due  mainly  to  the  effect  of  the  variation  in  humidity  and  also  to 
some  extent  due  to  temperature,  barometric  pressure  and  corona. 
Accordingly,  when  needle  gaps  were  used  for  voltage  measure- 
ments, the  actually  appUed  voltage  depended  upon  the  particular 
season  of  the  year  and  the  atmospheric  conditions  at  that  time, 
and  this  naturally  resulted  in  that  in  many  instances  the  trans- 
former tested  did  not  receive  the  required  voltage.  With  the 
adoption  of  the  sphere  gap  the  variation  in  the  applied  voltage  is 
eliminated  and  by  insisting  upon  this  method  of  measurement  it  is 
safe  to  assume  that  the  full  potential  is  actually  applied.  This  is, 
of  course,  of  great  importance  with  very  high  voltage  transformers. 

Taps.  It  is  customary  to  provide  the  high-voltage  transformer 
windings  with  taps  for  four  2J  per  cent  steps  below  the  normal 
operating  voltage  so  as  to  compensate  for  voltage  drop  in  the  line. 
Fig.  226  illustrates  this  point,  the  diagram  representing  a  single- 
phase  system  for  the  sake  of  simplicity. 

For  the  step-up  transformers  in  the  generating  station  it  is 
obvious  that  taps  are  not  required,  but  they  are  sometimes  pro- 
vided for  the  sake  of  imiformity  with  the  step-down  transformers. 
Thus,  with  a  10  per  cent  voltage  drop  in  the  line,  the  conductors 


Digitized  by 


Google 


388  ELECTRICAL  EQUIPMENT 

can  be  connected  to  the  10  per  cent  tap,  thereby  compensating 
for  the  line  drop.  As  this  tap  is  used  when  the  load  is  greatest 
it  follows  that,  theoretically,  the  taps  should  be  of  full  capacity; 
i.e.,  the  current  carrying  capacity  of  the  high-voltage  winding 
should  correspond  to  the  lower  voltage  value.  Often,  however, 
reduced  capacity  taps  are  specified,  and  reUance  is  placed  on  the 
ability  of  the  transformer  to  carry  the  increased  current  safely. 


o 


WTap 


t^ittiVf 


Fig.  226. 

On  account  of  the  low-voltage  winding  the  capacity  of  the  trans- 
former is,  however,  based  on  the  full  rated  voltage. 

Sometimes  large  power  transformers  have  their  high-voltage 
windings  so  arranged  that  the  two  halves  can  be  connected  either 
in  parallel  or  series.  The  former  connection  corresponds  to 
only  half  the  voltage  of  the  latter  and  is  for  use  during  the  first 
period  of  operation  of  a  system  when  the  load  is  light  and  when  the 
lower  operating  voltage  is  sufficient.  When  the  load  has  increased 
so  as  to  necessitate  a  higher  voltage,  the  two  windings  are  con- 
nected in  series,  thereby  doubling  the  transmission  voltage. 

Transformers  are  sometimes  arranged  for  supplying  simul- 
taneously two  loads,  one  at  full  voltage  and  the  other  at  half 
voltage.  The  question  then  often  arises  as  to  how  much  each 
side  may  be  loaded  without  causing  overheating  of  the  trans- 
formers. This  can  readily  be  ascertained  from  the  curves  in  Fig. 
227.  For  example,  with  a  full  voltage  load  of  75  per  cent  it  is  pos- 
sible to  load  the  half  voltage  circuit  for  40  per  cent  current,  which 
is  equal  to  20  per  cent  the  capacity. 

Where  taps  are  not  essential  for  the  satisfactory  operation  of  a 
system,  they  should  be  avoided  as  much  as  posable,  especially  in 
very  high-voltage  transformers,  and  standard  practice  does  not 
contemplate  any  taps  for  voltages  below  6600,  nor  above  66,000. 
It  is  evident  that  taps  are  difficult  to  insulate  and  bring  out  to  the 
connection  board  and  that  they,  therefore,  introduce  additional 


Digitized  by 


Google 


TRANSFORMERS 


389 


weaknesses  in  the  design  of  a  transformer  and  thus  decrease  the 
reliability  of  operation. 

Induction  motors,  synchronous  motors  and  synchronous  con- 


"=^<. 

M             ^^ 

^^ 

I"                                             \ 

f  g                                                                                                                       ^ 

|s*                                     '^. 

ll»                       ^^ 

11                                    ^ 

IL                        ^^ 

11                           s 

llD                                                                                                      X 

^r- 

^ 

^ 

"             t 

ftl— J 1          1          '          1          1 1— 1 i          1          ' '          '          ' 1 1 1 ' l-J 

Per  Ceot  Current 
aO        40        60        GO        70 

Per  Cent  Camcity 

15         20         29         JO        86 

Half-Yoltaffo  Winding 

FlQ.  227. 


100 


verters  started  from  the  A.C.  side  require  frequently  transformers 
with  taps  for  reducing  the  potential  at  starting  in  order  to  prevent 
a  heavy  rush  of  current. 

Fig.  228  shows  the  arrangement  of  taps  for  starting  three-phase 
converters,  leads  1,  2,  and  3  being  the  operating  terminals,  and 
leads  1,  4,  and  5  those  for  starting  at  half  voltage.  Lead  6  is 
merely  for  the  purpose  of  making  the  three  transformers  dupli- 
cates. With  some  converters  it  has  been  found  advantageous 
to  insert  resistance  in  the  starting  connections  so  as  to  still  further 
lower  the  applied  voltage. 

Large  converters  are  usually  connected  six-phase  diametrical, 
and  when  started  from  the  alternating  current  side,  it  has  been 
customary  to  provide  taps  on  the  transformers  for  one-third  and 
two-thirds  voltage,  as  shown  in  Fig.  229.  Leads  1  to  6,  inclusive, 
are  the  operating  terminals;  leads  1, 3,  5,  7,  8,  and  9  are  for  the 


Digitized  by 


Google 


390 


ELECTRICAL  EQUIPMENT 


first  step,  and  leads  1,  3,  5,  10,  11,  and  12  for  the  second  st^p. 
Leads  2,  4,  and  6  are  for  the  final  or  full-voltage  step.    Leads 


WVWVV     WWWV     NAA/WVV 


71 


f«'  Ts'iir  li  I^Tilr^ 


■JzrTy      i|o 


Armature 


Collector 
RU«8 


Azmatore 


a 


Collector 

Rll«B 


FlQ.  228. 


1X8466 

Fig.  229. 


1, 3,  and  5  are  connected  directly  to  the  converter  and  the  starting 
is  done  by  two  triple-pole,  double-throw  switches,  as  shown. 

Recent  improvements  in  the  design  of  synchronous  converters 
have,  however,  made  possible  the  elimination  of  the  second  start- 
ing tap,  and  it  is  now  general  practice  to  use  only  one  partial 
starting  voltage,  requiring  one  three-pole,  double-throw  switch 
for  six-phase  converters. 

Number  and  Size  of  Units.  The  number  and  size  of  the  trans- 
former units  and  whether  they  should  be  single-  or  three-phase 
depends  entirely  on  the  nature  of  the  development  and  on  the  con- 
ditions to  be  met.  With  moderate  voltage  developments  it  has 
in  the  past  been  the  general  practice  to  install  one  transformer 
bank  for  each  generator  and  of  equal  capacity  to  the  same,  even  if 
the  size  was  not  the  most  economical.  With  a  large  number  of 
units  it  was  then  natiu'ally  more  advantageous  to  install  three- 
phase  transformers,  while  in  plants  consisting  of  one  or  two  gen- 
erating units,  where  the  cost  of  a  spare  three-phase  unit  was  not 
warranted,  it  was  foxmd  preferable  to  install  single-phase  units. 

With  present  modern  high-voltage  systems  where  it  is  unde- 
sirable to  paiallel  the  outgoing  transmission  lines  on  the  high- 
tension  side  of  the  transformers  or  to  carry  out  any  high  tension 
switching,  which  may  cause  siu'ges,  it  has  become  a  general 
practice  to  install  the  transformers  in  groups,  each  having  a 
capacity  corresponding  to  one  line;  the  transformer  group  and 
the  line  thus  being  considered  as  a  unit.    Transmission  lines 


Digitized  by 


Google 


TRANSFORMERS 


391 


may  have  a  capacity  up  to  30,000  or  40,000  kilowatts,  depending 
on  the  voltage,  and  inasmuch  as  it  is  now  possible  to  build  single- 
phase  transformers  for  one-third  this  capacity,  the  arrangement 
is  entirely  feasible;  otherwise  it  would  be  possible  to  install 
two  banks  in  parallel  for  each  Une. 

Connections.  Among  the  great  variety  of  transformer  ma- 
nipulations in  power  and  general  distribution  work,  either  for 
straight  voltage  transformation  or  for  phase  transformation,  the 
following  are  the  most  generally  used: 

Voltage  transformation: 
Single-phase; 
Two-phase; 

Three-phase,  delta-delta; 
Three-phase,  delta- Y,  and  vice  versa; 
Three-phase,  Y-Y; 
Three-phase,  open-delta; 
Three-phase,  T. 

Phase  transformation: 

Two-  or  three-phase  to  single-phase; 
Two-phase  to  six-phase; 
Three-phase  to  six-phase. 

Voltage  Transformation.  Single-phase.  The  windings  may  be 
divided  into  sections  and  variously  connected  to  meet  different 
requirements.  So,  for  example,  are  most  standard  distributing 
transformers  made  with  two  low-voltage  coils. 

Fig.   230  represents  the  straight  connection  of  two  trans- 


< — 10«H-J  * — ^1000 — A 

vvvw      wwv 

AAAAA        AAAAA 


— loo — 


<--100 ► 


^1000 

WVWWNAAAAA/ 
AAAAAA  /WVAAA 

<--100--»^    \* — 100 > 


Fig.  230. 


100 H 

FiQ.  231. 


formers  to  1000-volt  mains,  the  transformer  consisting  simply  of 
single  high-  and  low-voltage  windings. 

Figs.  231  to  233  represent  different  connections  of  transformers 


Digitized  by 


Google 


392 


ELECTRICAL  EQUIPMENT 


which  are  provided  with  two  coils.  So,  for  example,  in  Fig.  231 
the  two  low-voltage  coils  are  connected  in  parallel  to  supply  100 
volts. 

In  many  instances  it  is  deemed  advisable  to  operate  a  three- 
wire  circuit  from  the  low-voltage  side  of  transformers,  and  thereby 
reduce  the  cost  of  copper  for  the  feeders.  Such  a  connection  is 
represented  in  Fig.  232,  where  the  low-voltage  coils  are  con- 
nected in  series  and  their  junction  connected  to  the  neutral  wire. 
This  method  of  connection  is  used  very  extensively  and  is  known 
as  the  Edison  Three-wire  System.  When  used  for  combined 
power  and  lighting  load,  the  motors  are  usually  connected  to  the 
two  outside  wires,  and  the  lights  between  the  outside  and  neutral. 

The  neutral  wire  generally  carries  less  current  than  the  out- 
side wires,  except  in  the  case  where  the  entire  load  is  on  one  side. 
The  neutral  wire  should,  for  this  reason,  be  of  sufficient  cross- 


— > 

Wwwwwwv 


WWWWWWV 


FiQ.  232. 


Fig.  233. 


section  to  safely  carry  a  current  which  will  blow  out  the  main  fuses 
in  case  of  short  circuit  on  one  side  of  the  system. 

Fig.  233  shows  the  three-wire  distribution  where  a  grounded 
neutral  wire  is  employed,  this  system  also  being  widely  used  for 
general  distribution,  lighting,  small  motors,  etc. 

The  four  terminals  of  the  low-voltage  coils  are,as  a  rule, brought 
outside  the  case  in  such  proximity  that  they  can  readily  be  con- 
nected in  any  desired  manner  by  joining  adjacent  terminals. 
Connection  blocks  are  seldom  used  for  the  low-voltage  winding  of 
distributing  transformers,  because  of  the  large  current-carrying 
capacity  required. 

The  voltage  stress  on  the  windings  naturally  depends  on  the 
voltage  of  the  mains  to  which  they  are  connected,  and  also  on 
abnormal  operating  conditions  such  as  accidental  grounds,  light- 
ning surges,  etc.     For  the  arrangement  shown  in  Fig.  231  it  is 


Digitized  by 


Google 


TRANSFORMERS  393 

obvious  that  under  normal  conditions  the  maximum  voltage  stress 
between  the  high-voltage  leads  is  1000  volts,  and  to  ground  500 
volts.  If  a  ground  should  occur  at  one  of  the  high-voltage  con- 
nections to  the  mains,  the  stress  will  be  1000  volts. 

In  the  case  of  the  low-voltage  winding,  if  the  two  coils  are  con- 
nected in  series  and  non-grounded,  the  stress  to  ground  imder 
normal  conditions  is  100  volts,  which  is  also  the  maximum  stress  if 
the  junction  point  or  neutral  is  grounded.  If  not,  and  with  one 
lead  grounded,  the  stress  becomes  200  volts.  The  stress  between 
the  two  windings  is  equal  to  the  high-voltage  plus  or  minus  the 
low-voltage,  depending  on  the  arrangement  and  connections  of  the 
coils. 

In  order  to  avoid  the  danger  of  excessive  voltages  being  im- 
pressed on  the  low-voltage  circuits,  caused  by  crosses  between 
the  high-voltage  and  low-voltage  lines  or  windings,  grounding  of 
the  low-voltage  circuit  is  now  generally  advocated  for  all  voltages 
up  to  250  volts.  No  point  of  the  circuit  can  then,  except  under 
unusual  conditions,  rise  above  its  normal  potential,  and  such 
grounding,  therefore,  prevents  accidents  to  persons  and  damage 
by  fire  to  property.  If  the  low-voltage  side,  on  the  other  hand, 
is  not  groimded,  and  the  transformer  breaks  down,  the  high- 
voltage  may  be  impressed  on  the  low-voltage  circuit,  and  a  per- 
son touching  any  bare  part  of  the  low-voltage  circuit  is  Uable 
to  receive  the  full  shock  of  the  high  voltage,  if  he  were  grounded 
by  contact  with,  for  example,  a  gas  fixture,  etc.  Furthermore, 
if  the  low-voltage  side  is  not  grounded  and  there  is  a  ground  on 
the  high-voltage  circuit,  the  high-voltage  impressed  on  the  fittings 
of  the  low-voltage  circuit  might  cause  a  fire. 

For  a  two-wire  110-volt  circuit  it  is  common  practice  to  con- 
nect the  ground  to  one  side,  while  with  a  three-wire  Edison  cir- 
cuit the  neutral  wire  is  grounded,  limiting  the  potential  from  either 
outside  wire  to  groimd  to  110  volts.  On  a  220-volt  single-phase 
power  circuit  the  middle  or  neutral  point  of  the  transformer  wind- 
ing should  be  groimded. 

To  prevent  any  increase  of  the  potential  stress  between  ground 
and  either  low-voltage  wire,  the  ground  should  be  well  made  so 
that  it  cannot  readily  be  broken.  It  should  not  be  fused  and 
should  consist  of  a  conductor  which,  without  overheating,  can 
carry  a  current  sufiicient  to  blow  the  main  fuses. 

Two-phase.    This  system  practically  consists  of  two  separate 


Digitized  by 


Google 


394 


ELECTRICAL  EQUIPMENT 


single-phase  circuits,  the  two  e.m.f/s  and  currents  being  90  elec- 
trical degrees  or  one-fourth  of  a  cycle  out  of  phase  with  each  other 
(Rg.  234). 

Two  single-phase  transformers  are  mostly  used  for  two-phase 
systems,  and  the  most  common  connection  is  that  shown  in 
fig.  235.  The  high-voltage  windings  of  the  two  transformers  are 
connected  respectively  to  the  two  phases  of  the  supply  mains. 


— " 

f  r 

? 

♦-lOOQ— ♦ 

a           h 
WVWV 

♦  -1000— • 

e           d 
WWW 

/WWNA 
c,         di 

/WWNA 

Fig. 

h 
234. 

Fig.  235. 

It  is  sometimes  also  desirable  to  operate  a  three-wire  two-phase 
distribution,  as  shown  in  Fig.  236.  In  this  case  the  voltage  across 
the  outside  wires  is  \/2  or  1.41  times  the  voltage  of  each  individual 
transformer.    This  is  clearly  xmderstood  by  a  reference  to  the 


FiQ.  236. 


vector  diagram  in  Fig.  237,  and  is  due  to  the  90°  phase  difference 
between  the  two  e.m.f  .'s,  so  that  instead  of  adding  them  numer- 
ically they  must  be  added  vectorially.  The  current  in  the  neutral 
wire  is  also  1.41  times  the  current  in  either  of  the  outside  wires, 
provided  the  load  is  balanced. 

Transformers   in   two-phase  work  are  sometimes  intercon- 
nected, as  shown  in  Fig.  238,  where  a  common  return  is  used  od 


Digitized  by 


Google 


TRANSFORMERS 


395 


both  high-  and  lo\^-voltage  sides.  Very  few  systems  are,  however, 
operated  on  this  plan. 

By  connecting  together  the  middle  points  of  the  low-voltage 
windings,  as  shown  in  Fig.  239,  two  100-volt  main  circuits  ac 
and  be  are  obtained.  Also  four  70-volt  (50X  V2)  side  circuits  ab, 
bcy  cd,  and  da. 

This  method  of  connection  is  used  when  the  neutral  is  to  be 
brought  out  in  connection  with  Edison  three-wire  service  of  rotary 
converters.    If  the  converter  is  started  from  transformer  with 


Fig.  238. 


VWNAA/ 


«! 


-iOO- 


d, 


-100- 


FiG.  239. 

one-third  and  two-third  voltage  taps,  provision  must  be  made  for 
opening  the  neutral  connection  when  starting,  so  as  to  avoid  short- 
circuit. 

Another  two-phase  arrangement  is  shown  in  Fig.  240,  and  is 
commonly  called  the  five-wire  system.  It  is  accomplished  simply 
by  connecting  the  low-voltage  windings  at  the  middle  and  bring- 
ing out  an  extra  wire  from  these  points. 

With  the  connections  shown  in  Fig.  235  the  maximum  insula- 
tion stress  in  case  of  a  permanent  groimd  is  1000  volts  on  either 
phase  of  the  high-voltage  side,  but  a  simultaneous  grounding  of 


Digitized  by 


Google 


396 


ELECTRICAL  EQUIPMENT 


lines  1  and  4,  1  and  3,  2  and  3,  or  2  and  4  or  their  connection, 
causes  insulation  stresses  y/2  times  this  value  or  1414  volts.  On 
the  low-voltage  winding  the  corresponding  stress  would  be  141 
volts. 

With  the  two  low-voltage  windings  connected  for  a  three-wire 
distribution,  as  in  Fig.  236,  the  maximum  stress  when  one  of  the 
outside  wires  becomes  grounded  is  141  volts,  while,  if  the  junc- 
tion or  neutral  point  is  grounded  it  is  limited  to  100  volts. 

Some  systems  are  supplied  with  two-phase  generators  in  which 
the  neutral  points  of  each  winding  are  connected  together.  In 
this  case  simultaneous  grounding  or  connection  of  any  two  lines 


Fig.  240. 

from  the  generator  cause  a  short-circuit  on  one-half  the  generator 
winding. 

For  grounding  two-phase  systems  several  methods  are  em- 
ployed. With  a  four-wire  distribution  the  mid-point  of  each 
transformer  winding  should  be  independently  grounded  unless 
the  motor  windings  served  are  interconnected  so  as  to  prevent  it. 
In  that  event  the  neutral  of  one  transformer  only  should  be 
grounded.  With  the  three-wire  system  the  neutral  point  should 
be  grounded  and  the  same  appUes  to  the  systems  shown  in  Figs. 
239  and  240. 

Three-^hdse.  The  following  are  the  most  common  methods 
in  which  transformers  may  be  connected  for  a  three-phase  system: 

Delta-Delta. 

Delta- Y,  or  vice  versa. 

Y-Y. 

Open-delta. 

T-connection. 


Digitized  by 


Google 


TRANSFORMERS 


397 


DeltordeUa.  With  the  delta-delta  system  the  leads  of  three 
single-phase  transformers  are  connected  to  the  mains  as  shown 
in  Fig.  241.  The  e.m.f.'s  and  currents  differ  in  phase  120 
electrical  degrees,  and  the  line  voltage  is  equal  to  the  individual 


a  b 

WWW 


♦-4000— 4  - 

0  d\  t 

VWWV  V 

/wwy\ 


1000- 

WWW 


AAAAM     AW\M    j2AWV\ 


Fig.  241. 


transformer  voltages.  This  voltage  is  commonly  denoted  the 
"  delta-voltage  "  to  distinguish  it  from  the  "  star  or  Y-voltage  ^ 
in  the  star-connected  combination.  Similarly  the  line  current 
must  be  distinguished  from  the  cinrent  flowing  in  the  closed  delta 
winding. 

The   voltage  and  current  relations  are  easily  explained  by 
referring  to  the  vector  diagram  in  Fig. 
242. 

If  we  denote:  S= delta-voltage,  or  volt- 
age between  phases; 

€= Y-voltage,  or  voltage 
between  phases  and 
neutral; 

/=Y-ciUTent  or  line  cur- 
rent; 


then: 
and 


Fig.  242. 
i=delta-ciuTent  or  current  in  delta  winding; 


E 

E=eV3ore^—=y 
V3 

J=2V3ort=— =. 


Digitized  by 


Google 


398  ELECTRICAL  EQUIPMENT 

When  speaking  of  the  voltage  and  current  or  line  voltage  and 
line  current  of  a  three-phase  S3rstem,  without  further  qualifica- 
tions, the  delta-voltage  and  the  Y-current  are  understood. 

Delta-connected  transformers  must  be  wound  for  the  full-line 
voltage  but  for  only  58  per  cent  line  current.  The  windings  must, 
therefore,  have  a  greater  number  of  turns  than  for  star  connec- 
tion, while  they  can  be  of  a  smaller  size. 

The  maximum  insulation  stress  in  case  a  permanent  ground 
occurs  does  not  exceed  the  normal  voltage  stress,  pro\dded  the 
ground  is  at  the  transformer  terminals.  When,  however,  the 
ground  occurs  on  the  transmission  line  at  some  distance  from  the 
transformer  terminal  the  reactance  drop  due  to  the  charging  current 
adds  to  this  stress.  On  this  account  with  long  distance  high-volt^ 
age  transmission  lines  operating  on  the  deltardelta  system,  a  dead 
ground  of  one  wire  may  cause  the  potential  of  the  other  two  wires 
to  rise  above  ground  considerably  above  normal  potential,  thereby 
increasing  the  insulation  stress.  This  increased  stress  may  exist 
both  at  the  generating  and  receiving  ends  of  the  transmission  line. 

With  a  delta-connected  220-volt  distributing  system  the  ground 
connection  should  be  made  to  the  mid-point  of  the  winding  of  one 
transformer.  This  gives  110  volts  to  ground  from  the  phase  wires 
next  to  the  groimd  connection  and  about  200  volts  from  the  other 
phase  to  ground. 

Where  2200-220  volt  transformers  are  connected  delta-delta 
for  three-phase  power  service,  one  of  the  imits  is  occasionally  made 
larger  than  the  other  two,  and  a  tap  from  the  middle  point  of  the 

low-voltage  winding  brought  out  so  that  a  o^Tpvolt  single-phase 

three-wire  service  may  be  obtained  for  lighting  purposes. 

If  one  transformer  or  one  phase  of  the  three-phase  transformers 
is  disabled,  the  other  two  may  then  be  used  in  open-delta. 

The  capacity  of  a  group  of  delta-connected  transformers  is 
equal  to  VsxEXl  Kv.A.,  where  E  represents  the  transformer  or 
line  voltage  and  /  the  line  current.  The  current  in  the  trans- 
former windings  is  equal  to  — =. 

V3 

DeUorY,  Delta-Y  connection  or  vice  versa,  as  shown  in  Fig- 
243,  is  used  to  a  great  extent,  and  it  is  especially  convenient  and 
economical  in  distributing  systems,  in  that  a  fourth  wire  may  be 
led  from  the  neutral  point  of  the  low-voltage  windings. 


Digitized  by 


Google 


TRANSFORMERS 


399 


The  current  and  voltage  relations  in  the  delta  side  are  the  same 
as  in  the  deltar-delta  connection.  On  the  Y-connected  side,  how- 
ever, one  end  of  each  winding  is  connected  to  a  common  neutral 
point  and  the  other  three  ends  to  the  lines.  With  this  connection 
the  number  of  turns  in  a  transformer  winding  is  58  per  cent  of 
that  required  for  delta-connected  transformers,  but  the  cross- 
section  of  the  conductors  must  be  correspondingly  greater  for  the 
same  output.  For  high  voltages  the  currents  are,  however,  gen- 
erally so  small  that,  in  may  cases,  the  size  of  wire  in  the  high- 
voltage  winding  must  be  governed  by  mechanical  considerations, 


1                             1 

t-1000-» 

a        b 

vwvw 

AVWVN 

c         d 

VWVW 

NW\N\ 
c,        d, 

*— 57.7— » 

<-1000-> 

e         f 

wwvv 

AAMAA 

El«4 

Fig.  243. 

and  the  size  of  wire  may  have  to  be  the  same  for  either  system. 
The  delta  connection  is,  therefore,  sometimes  somewhat  more 
expensive. 

If  the  neutral  point  of  the  Y-connected  system  is  ungrounded, 
the  transformer  insulation  must  be  capable  of  standing  the  stress 
of  the  full  line  voltage,  since  a  ground  on  any  line  will  throw  full 
voltage  on  parts  of  the  transformers.  With  grounded  Y  the 
stress  is,  of  course,  limited  to  the  Y-voltage.  This  is,  however, 
only  true  for  step-up  transformers  at  the  generating  end  of  trans- 
mission line,  and  also  only  when  the  neutral  is  sohdly  grounded. 
When  the  neutral  is  grounded  through  a  resistance  the  insulation 
in  transformer  may  be  subjected  to  full  voltage  stress,  and  under 
any  conditions  the  step-down  transformers  may  be  subjected 
to  full  voltage  stress. 

For  distributing  service  the  transformers  have,  as  previously 
stated,  often  their  low-voltage  windings  Y-connected  and  the 
neutral  brought  out,  forming  a  four-wire  system,  as  shown  in  Fig. 
244.    The  single-phase  service  is  then  obtained  by  tapping  between 


Digitized  by 


Google 


400 


ELECTRICAL  EQUIPMENT 


any  line  and  the  neutral,  while  for  three-phase  work  the  line  wires 
are  tapped  directly,  the  voltage  between  these  beingVs  times  the 
single-phase.  This  system  results  in  a  copper  saving  of  56  per 
cent,  assuming  that  the  four  wires  are  of  the  same  cross-section. 

If  the  main  three-phase  hne  potential  is  fixed,  this  method 
offers  no  saving;  on  the  contrary,  it  requires  33  per  cent  more 
copper.  In  any  case,  however,  the  use  of  the  four-wire  system 
gives  increased  flexibility,  and  the  neutral  wire  carries  all  un- 
balanced currents. 

This  system  is  mostly  used  for  a  combination  of  motor  and 


a  b 

www 

/WWV\ 
til  6i 

■*— 57.7— 3 


*— 1000 — > 

c  d 

VWWV 

AAAAAA 
c,         d, 

< — 57.7 — > 


<— 1000 — > 

e  f 

WWW 

ei  /, 

<— 57.7 — » 


Fig.  244. 

lighting  loads.  The  lighting  service  is  operated  from  a  2300-volt 
phase  voltage  and  the  power  service  from  the  4000-volt  line  voltage. 

Transformers  are  sometimes  designed  so  as  to  be  suitable 
for  either  delta-delta  or  delta-Y  connection,  in  order  to  permit  the 
user  to  increase  the  capacity  of  a  transmission  line  by  raising 
the  line  voltage,  which  can  be  accomplished  by  changing  the  con- 
nection from  delta  to  Y  on  the  high  voltage  side.  Such  trans- 
formers are  necessarily  more  expensive  than  they  would  be  if 
designed  for  straight  delta-delta,  and  used  at  the  lower  voltage 
only,  because  they  must  be  insulated  to  withstand  the  higher  line 
voltage. 

The  rating  of  a  group  of  delta- Y-connected  transformers  is 
the  same  as  for  the  straight  delta-delta  connection. 

Where  power  is  transmitted  with  delta-Y  step-up  and  Y-delta 
step-down  transformers,  service  may  be  maintained  with  one 
step-down  transformer  cut-out,  the  connections  being  made  as 
shown  in  Fig.  245. 


Digitized  by 


Google 


TRANSFORMERS 


401 


A'  B'  C  represents  the  Y-connected  high-voltage  winding  of 
the  step-up  transformers  and  a  b  c  the  high-voltage  winding  of 
the  step-down  transformers,  of  which  the  phase  od  is  out  of  ser- 
vice. A  three-phase  open-delta  connection  a'  6'  c'  is  thus  obtained 
on  the  low-voltage  side. 

The  capacity  is  reduced  to  57  per  cent  of  the  original  value,  and 


Fig.  245. 

care  must  be  taken  not  to  connect  a'  c'  in  the  position  a'  c",  since 
this  will  not  give  a  three-phase  relation.  The  neutral  connection 
on  the  high-voltage  side  should  preferably  be  made  through  a 
wire,  but  can  be  made  by  solidly  grounding  the  neutral  of  both 
transformers.  The  system  will,  however,  be  electrostatically 
and  electro-magnetically  unbalanced,  and  the  usual  disturbances 
characteristic  of  such  a  condition  will  be  observed,  the  severity 
depending  oh  the  circuit  characteristics. 

S3mchronous  converters  are  frequently  installed  in  connection 
with  Edison  systems,  where  three-wire  direct-current  is  required. 


Fia.246. 

The  three-wire  feature  is  readily  obtained  by  connecting  the  neu- 
tral wire  directly  to  the  neutral  point  of  the  low-voltage  winding 
of  the  step-down  transformers.  Care  should,  in  such  a  case,  be 
taken  in  using  only  such  connections,  that  the  transformer  will 
act  as  an  auto-transformer,  that  is,  that  the  direct  current  in 
each  transformer  divides  into  two  branches  of  equal  m.m.f. 


Digitized  by 


Google 


402  ELECTRICAL  EQUIPMENT 

If  this  is  not  done,  the  direct  current  will  produce  a  uniKiirectional 
magnetism  in  the  transformer,  which,  superimposed  on  the  mag- 
netic cycle,  would  tend  to  raise  the  magnetic  induction  beyond 
satiu^ation,  and  thus  cause  excessive  exciting  current  and  heating 
except  where  the  unbalanced  ciurent  is  comparatively  small. 
Such  a  connection  is  shown  in  Fig.  246  which  represents  a  delta-Y- 
connected  step-down  transformer  with  the  neutral  brought  out. 
It  is  evident  that  in  this  case  each  transformer  low-voltage  winding 
receives  one-third  of  the  neutral  current,  and  if  this  current  is  not 
small,  as  compared  with  the  exciting  current  of  the  transformer, 
it  will  cause  an  increase  in  the  magnetic  density. 

A  sjnstem  with  a  distributed  Y  or  "  zig-zag  "  connected  low- 
voltage  winding,  as  shown  in  Fig.  247,  has,  however,  been  devised, 
and  will  eliminate  the  flux  distortion  due  to  the  imbalanced 
direct  current  in  the  neutral.  Two  separate  interconnected  wind- 
ings are  used  for  each  leg  of  the  Y.  The  unbalanced  neutral  cur- 
rent flowing  in  this  system  may  be  compared  in  action  to  the 
effect  of  a  magnetizing  current  in  a  transformer.  The  effect  of 
the  main  transformer  currents  in  the  high-  and  low-voltage  wind- 
ings is  balanced  with  regard  to  the  flux  in  the  transformer  core, 
which  depends  upon  the  magnetizing  current.  When  a  direct- 
current  is  passed  through  the  transformer,  unless  the  fluxes 
produced  by  the  same  neutralize  one  another,  its  effect  on  the 
transformer  iron  varies  as  the  magnetizing  current.  For  example, 
assume  a  transformer  having  a  normal  ^mpere  capacity  of  100  and, 
approximately,  six  amperes  magnetizing  current,  and  assume 
that  three  such  transformers  are  used  with  Y-connected  low- 
voltage  windings  for  operating  a  synchronous  converter  connected 
to  a^three-wire  Edison  system.  Allowing  25  per  cent  unbalancing, 
the  current  will  divide  equally  among  the  three  legs  giving  8.33 
amperes  per  leg,  which  is  more  than  the  normal  magnetizing  cur- 
rent. The  loss  due  to  this  current  is,  however,  inappreciable, 
but  the  increased  core  losses  may  be  considerable.  If  a  dis- 
tributed winding  is  used  the  direct  current  flows  in  the  opposite 
direction  around  the  two  halves  of  each  core,  thus  entirely  neu- 
traUzing  the  flux  distortion. 

Whether  the  straight  Y  or  the  interconnected  Y  connection  is 
to  be  used  is  merely  a  question  of  balancing  the  increased  core  loss 
of  the  straight  Y  connection  against  the  increased  copper  loss  and 
the  greater  cost  of  the  interconnected  Y  system.    The  straight  Y 


Digitized  by 


Google 


TRANSFORMERS 


403 


connection  is  much  simpler,  and  it  would  be  quite  permissible  to 
use  it  for  transformers  of  small  capacities  where  the  direct  current 
circulating  in  the  neutral  is  less  than  30  per  cent  (10  per  cent  per 
transformer)  of  the  rated  transformer  ciurent. 

When  three-phase  core-type  transformers  are  used,  it  is  not 
necessary  to  resort  to  the  zig-zag  connection,  as  in  such  trans- 
formers the  direct  current  flows  along  the  core  from  end  to  end  in 
the  same  direction  on  all  three  legs,  and  since  the  direct  mag- 


T^u. I 


vwwwwwww  wv\/v\a/\nv^aam}   vwwwwwww 
AWW\     /yVWA    <yWWN|     (WW^    |AAA/Wi|     |WWN^ 


Keatral 


2 
FiQ.  247. 


netism  must  find  its  return  path  through  the  air  and  the  case 
outside  of  the  core,  its  effects  are  practically  negligible. 

On  account  of  the  30°  displacement  between  the  voltage  from 
line  to  neutral  and  that  across  each  half  of  the  transformer  legs 
of  the  zig-zag  connected  windings,  the  low-voltage  side  operates 
only  at  86.6  per  cent  of  the  normal  capacity,  which  it  would  have 
if  operated  straight  Y. 

Y-Y.  This  connection  is  not  ordinarily  to  be  recommended 
for  a  bank  of  three  single-phase  transformers  or  a  three-phase 
Bhell-t3rpe  unit.  This  is  due  to  the  fact  that  the  triple  frequency 
component  of  the  exciting  current  necessary  for  normal  magnet- 
ization cannot  flow,  which  results  in  a  third  harmonic  and  its 
odd  multiples  appearing  in  the  e.m.f.  from  line  to  neutral,  and 


Digitized  by 


Google 


404  ELECTRICAL  EQUIPMENT 

thus  causes  an  excessive  stress  on  the  windings.  No  triple  fre- 
quency *  harmonic  appears,  however,  in  the  line  voltage,  which 
remains  normal,  because  the  third  harmonics  across  the  three 
transformers  are  in  phase  with  each  other. 

The  triple  frequency  component  does  not  exceed  75  per  cent 
of  the  fundamental  and  with  densities  commonly  used  has  an 
average  value  of  50  per  cent  of  the  fundamental.  An  exception 
to  this,  however,  is  the  case  when  the  transformers  are  operated 
with  grounded  neutral  and  connected  to  a  transmission  line  pos- 
sessing electrostatic  capacity.  In  such  a  case  the  induced  triple 
harmonics  may  be  intensified  to  values  as  high  as  two  or  three 
times  normal. 

To  obviate  the  above  increase  in  voltage,  it  is  necessary  to 
make  neutral  connections  in  such  a  maimer  that  the  triple  har- 
monic exciting  currents  necessary  for  sine  wave  excitation  can 
flow,  thereby  eliminating  the  triple  harmonic  voltage.  This  is 
accompUshed  first,  when  the  transformer  neutral  is  grounded, 
and  a  Y-delta  bank  of  transformers  with  grounded  neutral  of 
sufficient  Kv.A.  capacity  is  coimected  to  the  line,  second,  when 
the  primary  neutral  is  connected  to  the  neutral  of  the  generator, 
this  case  only  being  possible  for  step-up  transformers.  It  should 
be  noted  that  by  grounding  the  high  voltage  neutrals  of  both  step- 
up  and  step-down  transformers  the  danger  from  triple  voltage 
intensification  is  not  eliminated. 

It  should  be  kept  in  mind,  however,  that  when  such  ground 
connections  are  relied  upon  for  eliminating  triple  third  harmonic 
voltages,  such  voltages  are  restored  by  disconnecting  any  ground 
connection,  and  also  that  the  third  harmonic  groimd  currents  are 
Uable  to  subject  parallel  telephone  or  telegraph  systems  to  serious 
interference. 

The  above  does  not  refer  to  three-phase  core-type  transformers, 
which,  owing  to  their  construction,  are  not  subject  to  these  addi- 
tional strains. 

No  stable  neutral  can  be  maintained  on  a  bank  of  transformers 
with  both  high-  and  low-voltage  windings  Y-coimected  when  un- 
grounded, since  it  may  shift  to  any  position. 

Open-delta.  When  single-phase  or  three-phase  shell-type  trans- 
formers are  used,  it  is  possible  to  maintain  operation  if  one  phase 
is  damaged.  Such  a  combination  is  shown  in  Fig.  248,  and  is 
termed  the  open-delta  or  V  connection.    In  three-phase  core  type 


Digitized  by 


Google 


TRANSFORMERS 


405 


designs  it  is  possible  to  operate  open-delta  when  the  damaged 
winding  is  open-circuited.  With  V-connected  three-phase  shell- 
type  transformers  the  damaged  phase  should  be  short-circuited 
to  prevent  stray  fluxes  from  the  other  phase  from  inducing  volt- 
ages in  the  damaged  windings. 

With  the  V  connection  the  current  in  each  transformer  is  30® 
out  of  phase  with  the  transformer  voltage,  so  that  each  trans- 
former under  non-inductive  load  operates  at  only  86.6  per  cent 
power-factor.  Based  on  a  three-phase  load,  the  cutting  out  of 
one  transformer  would   therefore   reduce   the   current-carrying 


lOGO ^ 

1000- 


1000 


\a/wwwWwvwv 
/wwwvyvwwwv 

i 100^^ ^ 

< 100 


d,. 


C^bi 


Fig.  248. 


capacity  not  to  two-thirds  of  100  per  cent,  which  equals  66.6  per 

cent,  but  to  two-thirds  of  86.6  per  cent  which  equals  58  per  cent. 

Assiuning  that  each  transformer  shall  have  a  capacity  of 

OPT 

-^=1.5^7,  it  must  be  capable  of  carrying   1.73E/  kilovolt- 

amperes,  because  the  transformer  voltage  is  equal  to  the  line 

voltage  E,  and  the  transformer  cmrent  equal  to  the  line  current 

1.737.    Therefore,  the  single-phase  rating  of  each  transformer 

1  73 
must  be  -:p=-=  1.155  or  15i  per  cent  greater  than  one-half  the 

group  rating. 

Sometimes  it  is  desired  to  parallel  a  number  of  transformers 
in  such  a  way  that  certain  of  the  transformers  will  form  a  delta 
group  while  the  others  may  be  connected  in  open-delta  or  V. 
Such  a  combination  may  be  caused  by  the  desire  to  increase  the 
capacity  by  adding  spare  transformers  of  insuflScient  number  to 
form  a  group  of  complete  deltas  or  through  the  failure  of  one  or 


Digitized  by 


Google 


406 


ELEOTRICAIi  EQUIPMENT 


more  units  originally  installed.  It  is  not,  however,  generally 
realized  that  such  an  arrangement  will,  in  general,  prove  either 
uneconomical  as  to  capacity,  if  all  the  units  are  kept  to  rated 
currents,  or  disastrous  to  the  imits  on  the  legs  having  the  smaller 
numbers,  if  it  be  attempted  to  work  all  units  at  overloads  guar- 
anteed for  single-phase  operation.  Not  only  is  this  from  the  addi- 
tional 16i  per  cent  capacity  required  on  imits  for  open-delta 
service,  but  a  fmlher  increase  in  current  takes  place  in  the  V-con- 
nected  transformers  due  to  change  in  phase  relation,  and  for  this 
reason  when  delta  and  V  groups  are  operated  in  parallel  the  result- 
ant capacity  is  not  the  sum  of  the  individual  delta  and  V  ratings. 
More  than  one  V  group  cannot  be  used  advantageously  with  a 


TABLE 

XLVI 

Number  of 
TraDsformen 

Connectioo 

Three-phase  CapacltF 
of  Oroap  Id  per  cent 
orSiDgle-phawRattng 

A 

100 

A 

aej} 

A^ 

A 

88JB 
100 

A 

A 

80 

A 

A 

aao 

A 

A 

82 

A 

A 

A 

100 

A 

A 

A 

»1 

A 

A 

A 

T» 

A 

A 

A 

88 

delta  group  of  transformers  nor  with  two  or  more  paralleled  delta 
groups.  Three  delta-connected  transformers  when  added  to 
another  delta  group  will  give  more  capacity  than  if  four  trans- 
formers, connected  in  two  V  groups,  were  added  to  the  same  delta 
group.  This  is  because  the  four  transformers,  which  would  form 
two  V  groups,  can  be  rearranged  to  form  a  delta  group  (one  trans- 
former remaining  idle),  and  the  delta  group  will  have  the  capacity 
of  three  transformers  while  the  two  V  groups  will  add  the  capacity 
of  only  two  transformers.  The  addition  of  two  transformers, 
connected  in  V,  in  parallel  with  a  delta  group  adds  the  capacity 
of  only  one  transformer  to  the  capacity  of  the  total  group. 
Although  two  V-connected  groups  should  never  be  used  in  parallel 


Digitized  by 


Google 


TRANSFORMERS 


407 


with  a  delta  group,  they  may  be  paralleled  with  one  another  and 
in  this  case  will  give  a  greater  capacity  than  three  units  con- 
nected in  delta.  The  capacity  of  the  two  V  groups  would  be 
0.866  times  four  or  3.46  as  against  three,  the  corresponding 
rating  of  three  transformers  connected  in  delta. 

Table  XL VI  gives  the  transformer  capacities  available  with 
various  combinations  of  open  and  closed  delta  groups. 

T-T,  lAs  with  the  open-delta  arrangement,  the  T-T  con- 
nection requires  only  two  single-phase  transformers,.  Ptgr'249, 
representing  the  diagram  of  conneetionr. ,  A  is  called  the  main 
transformer  and  is  provided  with  a  50  per  cent  voltage  tap  to 


^ 


/ 


wnaAaaa/  Wsaaaaa/  u 

A  B  r 

/WS/WsAA  AAA/WVSA 

o,       6,     cj  \di  f, 


+^ 


nr 


K 


Ol 


iei 


Fig.  249. 


which  the  teaser  transformer  jfi^  is  connected.  This  transformer 
may  be  designed  for  only  86.6  per  cent  of  the  line  or  main  trans- 
former voltage,  but  generally  it  is  made  identical  with  the  main 
transformer  and  operated  at  reduced  flux  density.  It  should  be 
noted  that  although  the  teaser  operates  at  86.6  per  cent  of  line 
voltage  it  is  unnecessary  to  provide  an  86.6  per  cent  tap  as  is 
often  supposed.  On  this  account  it  is  possible  to  operate  two 
identical  transformers  connected  T-T  as  well  as  open  delta,  when 
one  transformer  of  a  delta-delta  bank  bums  out,  the  only  require- 
ment for  the  T-T  connection  being  a  50  per  cent  tap.  Although 
interlacing  is  not  required  between  halves  of  the  main  winding 
nevertheless  each  half  of  the  primary  winding  must  be  properly 
wound  with  respect  to  the  corresponding  half  of  the  secondary 


Digitized  by 


Google 


408  ELECTRICAL  EQUIPMENT 

winding.  The  three-phase  capacity  of  the  T  connection  as  is 
shown  ixi  the  table  is  the  same  as  for  the  open-delta  connection, 
that  is,  86.6  per  cent  of  single-phase  capacity,  but  on  account  of 
the  fact  that  the  teaser  operates  at  a  lower  flux  density,  the  ef- 
ficiency of  the  T  connection  is  somewhat  greater  than  In  the  open- 
delta  or  V  connection. 

Two  ordinary  transformers  may  also  be  used  with  T  connect 
tion  provided  a  50  per  cent  tap  is  available.  It  is  also  more 
economical,  to  operate  with  T  connection  than  with  V  connection, 
when  one  transformer  has  burned  out. 

The  T  connection,  as  shown  in  ^ig.  250,  ciui  also  be  used  for 
three-phase  synchronous  converters,  and  the  neutral  point  can 
readily  be  brought  out  for  Edison  three-wire  service.    The  neutral 


w 

\ 

——io  Heatnd 

Hi. 

K^ 

1 

T 

"T*L 

la 

• 

f- 

— <-i. 

^^■■^-i 

Fig.  250. 

is  then  brought  out  from  a  point  at  one-third  the  height  of  the 
teaser  winding  and  the  m.m.f.  of  the  direct  current  i  will  balance, 
as  shown  in  the  diagram. 

For  T  connection  with  ungrounded  neutral  the  voltage  stress 
is  the  same  as  for  the  delta  system,  and  with  grounded  neutral 
the  voltage  stress  between  line  and  ground  is  limited  to  58  per 
cent  of  normal.'  * 

Assuming  again  that  as  with  the  open-delta  connection  the 
two  transformers  shall  be  capable  of  supplying  a  load  equal  to 

-— =  1.5EJ,  the  Kv.A.  rating  of  the  main  transformer  must, 

therefore,  be  equal  to  1.73EI,  while  the  Kv.A.  of  the  teaser  trans- 
former only  is  equal  to  1.73/X0.866£=1.5B/.  The  two  trans- 
formers are,  however,  designed  to  carry  the  same  currents  and  are 
generally  made  identical,  so  that  the  single-phase  ratings  of  either 


Digitized  by 


Google 


TRANSFORMERS  409 

1  73 
transformer  must  also  here  be  -f^  =  1.155  or  15.5  per  cent  greater 

1.0 

than  one-half  the  group  rating. 

Phase  Transformation.  Of  the  connections  for  transforming 
one  pol3rphase  system  into  another  with  a  different  number  of 
phases,  the  following  are  the  most  commonly  used: 

Two-  or  three-phase  to  single-phase. 
Two-phase  to  six-phase.  - 
Three-phase  to  two-phase. 
Three-phase  to  six-phase. 

Two-  or  Threerphase  to  Single-phase.^  It  is  practicaUy  impos- 
sible to  transform  from  polyphase  to  single-phase  by  means  of 
static  transformation  with  balanced  conditions.  Various  schemes 
have  been  proposed  and  investigated,  but  none  of  the  combina- 
tions give  better  results  than  can  be  obtained  by  simply  using  a 
transformer  across  on  phase. 

The  reason  for  this  is  explained  by  Dr.  Steinmetz  (A.I.E.E. 
1892)  to  be  as  follows: 

'^Single-phase  power  changes  from  a  maximum  to  zero  and  back 
to  maximum  every  half  cycle,  while  polyphase  power  is  delivered 
at  a  constant  rate.  Therefore,  any  system  capable  of  transform- 
ing from  balanced  polyphase  current  to  single-phase  current  must 
be  capable  of  storing  energy  during  the  interval  of  time  when  the 
power  delivered  to  the  single-phase  side  is  less  than  the  power 
received  from  the  three-phase  side.  The  transformer  is  incapable 
of  fulfilling  this  requirement." 

Nevertheless,  it  is  desirable  to  know  the  best  method  of 
taking  single-phase  power  frpm  a  three-phase  system  and  often 
ingenious  although  compUcated  connections  are  proposed  with  the 
idea  of  more  uniformly  distributing  a  single-phase  load.  Most 
of  these  schemes  do  not  present  a  single  feature  that  is  superior 
to  the  placing  of  the  single-phase  load  directly  across  two  wires. 
When  there  is  one  feature  which  is  apparently  superior,  there  are 
generally  imdesirable  features  which  more  than  offset  it.  The 
four  schemes  shown  in  Figs.  251  to  254  are  ones  conmionly  sug- 
gested and  Table  XL VII  gives  the  characteristics  of  these  con- 
nections and  shows  that  they  are  inferior  to  straight  single-phase 

*  Three  papers  on  Single-phase  Power  Service  from  Polyphase  Sjrstems 
appeared  in  A.I.E.E.  Proceedings  for  October,  1916. 


Digitized  by 


Google 


410 


ELECTRICAL  EQUIPMENT 


transformation.  All  values  except  for  power  are  given  with 
reference  to  straight  single-phase  as  unity.  The  total  value  of 
power  delivered  is  the  same  in  all  cases.  By  straight  single-phase 
is  meant  connecting  one  transformer  between  two  wires  of  a  three- 
phase  system.  The  only  condition  under  which  there  seems  to 
be  an  advantage  is  in  schemes  1  and  3  where  it  will  be  noticed 


A.  1.155  1 

B.  2.31  I 

C.  1.155  1 


Power  Factor ». 806. 

Fig.  251. 


A,  2  I 

B,  0 

C,  21 


Power  Factor  «  .  666 

Fig.  252. 


Br 


WE- 


5  -N^ 


A,  1.155  1 

B,  2.31  I 

C,  1.155  1 


3^,  .635 
Power  Factor  "1^,  1. 

Av.  .778. 

Fig.  253. 


■^   A 


.707E-H     I 
-,7OTE ^  I 


-^ 


A,  2.23  I  3^.  .707. 

B,  .1631  Power  factor  =  1^.   .707. 

C,  .599  I  Av.  .707. 

Fig.  254. 


that  a  delta-connected  generator  has  a  maximum  current  of  0.577 
as  against  0.667  for  the  straight  single-phase.  To  offset  this,  both 
schemes  1  and  3  require  two  transformers  possessing  greater  total 
capacity  and  also  imposed  upon  the  line  a  greater  maximum 
current. 

Two-phase  to  Six-phase.  The  double-T  connection,  as  shown 
in  Fig.  255,  is  generally  used  in  cases  where  a  six-phase  synchronous 
converter  is  to  be  operated  from  a  two-phase  supply  system,  and 
where  the  two-phase  voltage  requires  some  transformation  in 
order  to  obtain  the  correct  alternating-current  voltage  for  the 
converter.  The  cost  of  double-T-connect«d  transformers  and  a 
,  standard  six-phase  rotary  converter  will  occasionally  be  less  than 
that  of  two-phase  transformers  and  a  special  two-phase  converter. 


Digitized  by 


Google 


TRANSFORMERS 
TABLE  XLVII 


411 


Capacity. 

Power- 

QBNERATORfl. 

Scheme  No. 

No. 
Trans. 

Trans. 
Each. 

Cap. 
Total. 

factor  for 
Non-induc- 
ttve  Load. 

Y- con- 
nected. 

Delta  con- 
nected. 

Cur- 
rent. 

Watu 

Cur- 
rent. 

Watts 

1 

2 

0.577 

1.55 

0.866 

0.577 
1.155 
0.577 

i 

i 
i 

0.570 

0 

0.577 

i 
0 

i 

2 

3 

0.500 

1.500 

0.666 

1.0 

0 

1.0 

i 
0 

i 

1 
} 
i 

i 
i 

i 

3 

2P 

/  1.000 
I   0.577 

P  0.635 

0.577 

i 

0.577 

i 

1.577t 

S  1.000 

1.155 

i 

0 

0 

/  SI. 000 
I       0 

Av.  0.810 

0.577 

i 

0.577 

i 

(  0.707 

4 

2P 

0.557* 
I   0.150* 

1.821*t 

P  0.707 

1.115 

0.622 

rO.644 

0.622 

S  0.707 

0.815 

0.333 

0.172 
10.471 

0.045 
0.333 

S 

/  0.707 
I   0.707 

Av.  0.707 

0.300 

0.045 

Straight 

Single-phase 

1 

1.00 

1.00 

1.00 

1.0 

0 

1.0 

i 
0 

J 

1 
i 

i 
i 
i 

*  Onebalf  of  main  has  capacity  of  0.557;  other  half  0.150;  total  capacity  computed 
on  basis  that  both  halves  are  alike  and  of  lanr®  capacity. 

t  On  basis  of  primary  capacities  when  there  is  a  difference  between  primary  and 
secondary. 

T  connection,  however,  requires  specially  designed  transformers, 
and  the  complication  of  starting  taps  and  switches  is  a  disad- 
vantage. 

The  syBtem  requires  two  transformers  of  the  same  impedance, 
each  equipped  with  two  low-voltage  windings,  connected  in  such 


Digitized  by 


Google 


412 


ELECTRICAL  EQUIPMENT 


a  way  that  they  are  displaced  180"*  from  each  other,  thus  producing 
the  six-phase  relation. 

The  voltages  are  the  same  as  for  the  T-connected  three-phase 


wVWWVW    wwwvww 

MAM 

Jo,       d, 


S-Wlro 
"d 


J 


4.WIre 
20 


>v« 


DDDODr 

1  2  S    4  5   6 
Collector  Rings 


dn    r 


J-id. 
— V J 


Fig.  255. 


system,  and  each  transformer  must  be  15  per  cent  greater  than 
half  of  the  power  required  for  the  rotary. 

The  neutral  can  also  be  brought  out  on  the  six-phase  side, 
although  this  furthermore  increases  the  complication  of  the  con- 
nection. 

Three^hcLse  to  Two-phase.  A  number  of  schemes  for  three- 
phase  to  two-phase  transformation,  and  vice  versa,  have  been 
devised,  but  the  most  commonly  used  method  is  the  T  con- 
nection for  either  balanced  or  unbalanced  service. 

'Balanced  T  or  Scott  Connection.  This  connection  is-Aown  in 
Kg.  25ft  and  requires  two  transformers  which  on  the  three-phase 
side  are  connected  in  T,  the  number  of  effective  turns  in  the  teaser 
winding  being  86.6  per  cent  of  the  number  of  turns  in  the  main 
winding.  On  the  two-phase  side  both  mains  and  teaser  windings 
are  identical  and,  as  shown  in  the  figure,  are  electrically  inde- 
pendent, when  supplying  a  two-phase,  four-wire  system.  Gen- 
erally, the  main  and  teaser  transformers  are  made  identical  for 
the  sake  of  interchangeability,  in  which  case  the  three-phase 
winding  is  provided  with  both  a  50  per  cent  and  an  86.6  per  cent 
tap,  as  shown  by  the  dotted  lines  in  Fig.  256,  so  that  when  used  as  a 
main  the  50  per  cent  tap  is  used  and  when  used  as  a  teaser  the 
86.6  per  cent  tap  is  used,  the  13.4  per  cent  winding  bemg  left  idle 


Digitized  by 


Google 


TRANSFORMERS 


413 


Each  of  the  two  halves  of  the  three-phase  winding  should  further- 
more be  distributed  over  the  entire  winding  length  of  the  core  in 
order  to  prevent  flux  distortion  and  poor  regulation.  The  T 
connection  requires  6.7  per  cent  more  copper  than  single-phase 
transformers  delivering  the  same  power  on  account  of  the  idle 


a        x\       \    b       \c  \      y  l 

wwwnaAa/    vwwVvw 


/ 

-4/ 


r 

a?  i     b 


/WWWNAAA 

a,  b, 


AAAAAAAAA 


6|  a 


Fig.  256. 

copper  in  the  teaser  and  also  on  account  of  the  fact  that  wattless 
currents  flow  in  the  three-phase  side  of  the  main  winding. 

The  neutral  of  the  three-phase  side,  which  is  one-third  the 
height  of  the  teaser  winding,  can  be  brought  out  for  four-wire 
operation  although  the  transformer  construction  is  somewhat 
complicated  thereby.  When  operating  without  the  neutral  point 
grounded  on  the  three-phase  side,  the  maximum  insulation 
strain,  if  a  permanent  ground  occurs,  is  equal  to  the  line  volt- 
age vi? 

Unbalanced  T.  This  connection  may  sometimes  be  of  use  in 
emergency  conditions  where  a  transformer  with  an  86.6  per  cent 
tap  is  not  available  and  a  teaser  transformer  of  the  same  voltage 
as  the  main  transformer  must  be  used. 

In  this  connection  two  transformers  of  exactly  the  same 
capacity  and  voltage  are  used.  The  phases,  however,  are  no 
longer  strictly  120°  apart,  and  it  is  assumed  that  the  same  con- 
nection is  used  at  each  end  of  the  line.  As  it  is  not  a  true  three- 
phase  system,  any  attempt  to  operate  in  multiple  with  a  three- 
phase  S3rstem  or  three-phase  apparatus  will  cause  serious  unbal- 
anced currents. 

The  unbalanced  T  connection  occurs  when  voltage  is  applied 


Digitized  by 


Google 


414 


ELECTRICAL  EQUIPMENT 


from  the  two-phase  side.    When  balanced  three-phase  voltages 
are  applied  the  voltages  on  the  two-phase  side  will  be  unequal. 

The  connections  and  voltage  relation  of  this  system  are  shown 
in  Fig.  257.    With  equal  currents  in  the  two-phase  system,  the 


a    U   b      \c         d\ 
VWSAA/     VWWV 


y% 


\ 


\ 
6V 


/wvw\ 

6,       c,   _   d| 


V — V — H 


Fig.  257. 

currents  in  the  three  transmission  wires  will  be  the  same  as  in  the 
coils,  namely:  a =112  amperes,  6  =  112  and  d=100,  with  the 
voltages  as  indicated  in  the  diagram. 

An  unbalancing  of  the  two-phase  distributing  network  afifects 
the  currents  in  the  three  transmission  wires,  in  that  an  increase  of 
the  load  on  phase  D  further  increases  the  unbalancing,  while,  if 
phase  E  be  loaded  in  the  neighborhood  of  15  per  cent  in  excess  of 
phase  £>,  the  transmission  line  currents  become  practically  bal- 
anced. 

With  no  neutral  the  maximum  insulation  stress  under  all  con- 
ditions arising  from  a  permanent  ground  would  be  1.12  times  F. 

Symmetrical  or  Woodbridge  Connection,  In  the  previous  two 
T-connected  methods,  the  two-phase  windings  are  electrically 
distinct.  There  are,  however,  a  number  of  schemes  in  which  the 
windings  on  the  two-phase  side  are  electrically  interconnected 
in  one  way  or  another. 

Such  a  system  of  connections  is  shown  in  Fig.  258.  It  con- 
sists of  three  windings,  one  for  each  phase.  Two  of  the  phases 
are  identical,  each  consisting  of  two  coils,  wound  for  0.577  times 
the  two-phase  line  voltage  and  ha\dng  a  current  capacity  of  0.577 
times  the  two-phase  line  current.  The  third  phase  consists  of 
three  coils,  one  being  wound  for  0.577  times  the  line  voltage  and 
the  other  two  being  identical  and  wound  for  0.212  times  the  line 


Digitized  by 


Google 


TRANSFORMERS 


415 


voltage.    The  respective  current  capacities  are  0.421,  1,  and  1 
times  the  line  current. 

One  advantage  of  this  system  is  the  fact  that  voltages  and 
currents  do  not  exceed  those  which  would  occur  in  single-phase 
operation,  giving  an  internal  power-factor  of  the  system  of  100 
per  cent,  whereas  in  the  T  connections  the  average  power-factor 
is  only  96.4  per  cent.  The  three-phase  side  may  be  connected 
either  delta  or  Y.    This  connection,  requiring  less  copper  and 


a  b\      e  d|       e  J] 

vwww   vwww    Vwww 

AAA  AM 


Ci  kii   ld„d,   Ice,.    f^X 


Fig.  258. 

being  sUghtly  more  efScient  than  the  T  connection,  is  recom- 
mended in  place  of  the  T  connection  for  three-phase  units,  pro- 
vided no  taps  are  required  on  the  two-phase  side.  If  single-phase 
imits  are  desired,  the  use  of  this  connection  becomes  doubtful 
owing  to  the  multiplicity  of  leads  and  coils  on  the  two-phase  side. 
The  connection  is  very  seldom  used,  principally  on  account  of 
the  electrical  interconnections  of  the  phases  on  the  two-phase 
side.  This  prevents  it  from  being  used  on  a  three-wire  system, 
while,  on  the  other  hand,  a  cross  between  the  two  phases  results  in 
a  short-circuit. 

Three-phase  to  Three-phase — Two-phase.  It  is  possible  by  means 
of  transformer  connection  to  derive  from  a  three-phase  primary 
circuit  a  four-wire  secondary  circuit,  three  wires  of  which  rep- 
resent a  three-phase  system  and  the  four  wires  making  a  two-phase 
system.  Prom  such  a  system  independent  three-phase  or  two- 
phase  loads  may  be  taken  simultaneously.  This  may  be  accom- 
plished by  three  single-phase  transformers  provided  with  special 


Digitized  by 


Google 


416 


ELECTRICAL  EQUIPMENT 


windings  or  by  one  three-phase  transformer,  as  shown  in  Fig.  259. 
Primary  winding  may  be  connected  either  Y  or  delta  and  is  in 
no  wise  different  from  an  ordinary  three-phase  winding.  The 
secondary,  however,  is  provided  with  15^  per  cent  coils  in  two  of 
the  phases  and  15^  per  cent  taps  in  the  other  phase  in  such  a 
manner  which  are  interconnected,  as  shown  in  Fig.  259. 


a  b\       c  d\      te  71 

VWVWV     VWWW     VWWW 


/VW\ 


M 


AAAA 

c,    c„ 


^*^/  c„ 

\ 

1  ^" 

V. 

V  — 

1 

1 

Fig.  259. 


a       [e        b    \c  d 

vwwvw    Www 

AAAMAM      (V\NV\N\ 
a,     |e,   fe.    c,  |y;        d, 


30 


It 


20 

Fig.  260. 


/ 

*  _  _ 

M    1, 

"■!    c. 

i!**' 

f V— 


This  may  also  be  accomplished  by  means  of  two  transformers 
T  connected  as  shown  in  Fig.  260. 

The  choice  between  the  two  methods  given  above  of  obtain- 
ing three-phase  and  two-phase  on  four  wires  depends  for  the  most 
part  upon  whether  the  three-phase  or  the  two-phase  load  predom- 
inates.    Where  the  three-phase  load  is  predominant,  it  is  evident 


Digitized  by 


Google 


TRANSFORMERS 


417 


that  a  connection  given  in  Fig.  260  is  superior,  but  where  the 
two-phase  load  predominates,  the  T  connection  is  preferable. 

Three-phase  to  Six-phase.  In  transforming  from  three-  to  six- 
phase,  there  are  four  different  connectionSi  which  may  be  used, 
namely: 

Diametrical. 
Double-delta. 
Double-Y. 
Double-T. 

Diametrical.  The  diametrical  connection,  as  represented  in 
Fig.  261,  is  the  most  commonly  used  of  any  three-phase  to  six- 
phase  transformations,  and  there  is  very  little  reason  for  using 


a  b 

vvwwW 


c  d\ 

vwwwv 

/WVYNAAA     /WNMAAA 
Oi     I      bi       c,      I      d, 


AVWWV 
MAAAAAA 


12    8    4    6    6 
CoUfBOtor  Ring* 


Fig.  261. 

any  other  connection  for  the  operation  of  six-phase  converters. 
It  requires  only  one  low-voltage  coil  on  each  transformer  which 
are  connected  to  diametrically  opposite  points  on  the  armature 
windings.  It  furthermore  gives  the  simplest  arrangement  of 
switches,  transformer  taps  and  connections  for  starting  six-phase 
converters  from  the  alternating  current  side,  while  on  the  other 
hand  it  is  possible  to  operate  a  six-phase  converter  at  reduced 
capacity  with  one  transformer  out  of  service,  leaving. the  other 
two  connected  across  their  respective  diameters. 

With  diametrically  connected  low-voltage  windings,  the  high- 
voltage  windings  should  preferably  be  connected  in  delta  so  as 
to  avoid  the  triple  frequency  harmonics  of  the  e.m.f.,  as  described 
under  Y-Y  connection  on  page  403.  With  regulating  pole  con- 
verters, however,  the  high-voltage  windings  must  be  connected  Y 


Digitized  by  VjOOQ IC 


418 


ELECTRICAL  EQUIPMENT 


on  account  of  the  fact  that  the  third  harmonic  voltage  is  made 
use  of  to  obtain  the  direct-current  voltage  regulation  and  in  such 
a  case  the  windings  must  be  insulated  for  double  line  voltage  to 
groimd  and  3.46  times  normal  Y-voltage  across  windings,  due  to 
the  presence  of  the  third  harmonic  e.m.f 's.  The  middle  points  of 
the  diametrical  windings  can  readily  be  connected  together  and 
brought  out  for  three-wire  Edison  service,  the  unbalanced  three- 
wire  direct  current  having  no  distorting  effect.  Arrangements 
should  then  be  made  for  opening  the  neutral  connections  during 


a  be  d       e  f 

wvwwvv   Wwwwv   vwwww 


OoUector  Ring* 


Fig.  262. 


starting  to  avoid  short  circuit.    When  used  with  regulating  pole 

converters  the  neutral  must  be  isolated. 

The  current  in  each  coil  on  the  low-voltage  side  is  equal  to 

y    output  of  transformer  in  watts  .      .i_    i     j  -    i    i        j 

/= — V,  ,  ,. 7-^ — i — r- assmnmg  the  load  is  balanced 

3  X  diametrical  voltage 

and  that  the  power-factor  is  unity. 

With  six-phase  diametrical  connection  with  common  neutral, 
one-half  the  output  can  be  taken  from  the  low-voltage  side  for 
operating  three-phase  without  change  of  diametrical  voltage. 
If  full  three-phase  output  should  be  desired,  the  coils  can  be  con- 
nected in  delta  in  which  case  the  diametrical  voltage  is  increased 
14  per  cent.  Hie  full  three-phase  output  at  1.73  times  the  dia- 
metrical voltage  may  be  obtained  by  connecting  the  coils  in  Y, 
in  which  case  the  neutral  should  be  grounded  and  if  the  high 
windings  are  Y-connected  the  system  is  subject  to  the  dangers 
of  the  third  harmonic  e.m.f's.  as  previously  explained.  It  must 
also  be  ascertained  if  the  insulation  of  the  windings  can  withstand 


Digitized  by 


Google 


TRANSFORMERS 


419 


the  increased  voltage  safely.  If  the  secondary  windings  are  made 
up  of  two  distinct  sections,  which  is  not,  however,  standard  prac- 
tice, the  connections  may  be  made  as  in  Fig.  262.  The  latter 
connection  is,  however,  somewhat  compUcated  and  when  three- 
phase  operation  with  full  output  is  desired  and  without  change  of 
voltage,  the  double-delta  connection  is  generally  preferable. 

Dovble-deUa.  For  the  double-delta  connection  two  inde- 
pendent low-voltage  coils  are  required  for  each  transformer,  as 
shown  in  Fig.  263.    The  second  set  are  all  reversed,  and  then  con- 


vwwww   wwvvvw  vwwww 

MM  MM     MM  MM     MM 


MM 
«ii   fu 


OoOeetor  Rings 


Fig.  263. 


nected  in  a  similar  manner  to  the  first  set,  so  that  the  two  deltas 
are  displaced  180°. 

The  high-voltage  windings  should  preferably  be  connected 
delta,  as  it  permits  the  system  to  be  ox)erated  with  only  two 
transformers,  in  case  one  should  be  damaged. 

The  current  in  each  coil  for  double-delta  is  equal  to  J= 

output  in  watts  ,   .  ^  .         ,  „  ,    r     .  p,« 

-T-r r. ;; — :;  and  the  ciurent  m  each  line  equals  7X1.73. 

delta  voltage  X  2  X  3 

Full  output,  three-phase  may  also  be  obtained  by  connecting 
as  shown  in  Fig.  262. 

Double-delta  connection  cannot  be  used  with  Edison  three- 
wire  service,  as  it  has  no  neutral,  and  in  such  cases  separate  auto 
transformers  would  be  required. 

Dovble-Y.  Like  the  double-delta,  this  system  requires  two 
sets  of  low-voltage  coils,  displaced  180®,  as  shown  in  Fig.  264. 

The  high-voltage  windings  may  be  either  delta-  or  Y-connected 
even  with  regulating  pole  converters,  but  in  this  case  the  two  low- 


Digitized  by 


Google 


420 


ELECTRICAL  EQUIPMENT 


voltage  neutrals  must  not  be  connected  together.  Where  the 
high-voltage  windings  are  Y-connected  the  danger  of  Y-Y  oper- 
ation should  be  considered,  and  the  neutral  should  be  grounded. 


a  b 

vwww 


RAAA     AAA  AAA     AAA 
«i,^        ^^i^i    ^uf^       «i  fi\ 


f 
VWWW 

/w\ 


&[5dddd 

12    8    4     6    8 
Collector  Rings 


Fig.  264. 


a  [cc 

naaaaAaaa/ 


blc 


aaaa 


aaaa  aaaa 

c,     dj     c„  cU 


12    3    4    6    6 
Collector  Rings 


Fig.  265. 


^/b 


/■ 


T 

> 

2 


arr 


The  current  in  each  leg  is  equal  to  7= and 

YvoltageX  1.73X2 

the  line  current  has  the  same  value. 

Double-T,    Fig.  265  represents  the  double-T  connection  for 

transforming  from  three-phase  to  six-phase.    The  low-voltage 

connections  are  similar  to  the  two-phase — six-phase  system  shown 

in  Fig.  255,  and  the  high-voltage  windings  are  connected  in  T. 


Digitized  by 


Google 


TRANSFORMERS 


421 


la  b 


/SAAAAA 
fli  bi\ 


Pigs.  262  to  265  are  the  connections  of  single-phase  trans- 
formers used  for  six-phase  operation,  and  they  do  not  apply  to 
three-phase  units. 

Parallel  Operation.  In  order  that  two  or  more  transformers 
or  groups  of  transformers  shall  operate  successfully  in  parallel  it 
is  necessary  that  they  are  connected  so  that  their  polarity  is  the 
same,  that  their  voltages  and  voltage  ratios  are  identical,  and  that 
their  impedances  in  ohms  are  inversely  proportional  to  the  ratings. 

The  polarity  expresses  the  phase  relation  between  the  high  and 
low  voltages  as  measured  at  the  terminals.  When  the  trans- 
formers are  of  the  same  manufacture,  they  usually  have  the  same 
polarity,  while  if  of  different  makes  some  may  have  the  high-  and 
low-voltage  windings  in  phase  and  others  180°  apart. 

Effect  of  Polarity  on  Parallel  Operation  It  i^  easy  to  deter- 
mine the  right  polarity  of  two  single-phase  transformers  which  are 
to  operate  in  parallel.  Kg.  266  represents  such  a  case  in  which 
all  connections  are  made  except  fci. 
If  now  the  voltage  between  fci  and  di 
is  zero  it  indicates  that  the  two  trans- 
formers have  the  same  polarity,  while 
if  the  polarities  were  opposite  the 
voltage  from  fci  to  di  would  be  the 
sum  of  the  two  transformers,  and  the 
joining  of  the  two  leads  would  cause 
a  short-circuit.  When  testing  for 
polarity  the  two  terminals  should, 
therefore,  be  joined  through  a  fuse  or  automatic  switch.  If  the 
fuse  does  not  blow,  the  connection  may  be  made  permanent, 
while,  if  the  fuse  blows  the  two  leads  of  one  transformer  must  be 
reversed. 

With  three-phase  transformer  banks  operating  in  parallel  it  is 
also  necessary  that  the  phase  relation  of  the  voltages  in  the  two 
banks  is  the  same,  both  as  to  direction  and  position.  It  is,  there- 
fore, not  possible  to  parallel  a  group  of  transformers  which  is 
connected  in  delta  on  both  high-  and  low-voltage  sides  with  a 
group  connected  in  delta  on  the  high-voltage  side  and  Y  on  the 
low-voltage  side  or  vice  versa.  On  the  other  hand,  it  is  possible 
to  parallel  a  delta-delta  connection  with  a  Y-Y  connection,  and 
also  a  delta-Y  connection  with  a  Y-delta  connection. 

Three-phase  transformer  banks  divide  themselves  into  three 


c  d 

www 

AA/WVA 


Fig.  266. 


Digitized  by 


Google 


422 


ELECTRICAL  EQUIPMENT 


groups,  depending  upon  the  angular  displacement  between  high- 
voltage  and  low-voltage  windings.  These  groups  are  given  in 
Fig.  267,  which  shows  that  the  delta-delta  connection  and  the  Y-Y 
connection  are  similar,  both  capable  of  being  connected  so  as  to 


Groap 

Angular 

Dtoidaceinent 


OroupU 

Angular 

Displacement 

180" 


A  BX  Y  A  BX  Y 

C         Y X  C       Y  X 

AV  XY 

A  B         Z  A  B       Z 

C  Z  0         Z 

A^^  X> 

A  B   X  A  B  X 

Fig.  267. 


Group  HI 
Angular 


give  an  angular  displacement  of  zero  degrees  between  high  voltage 
and  low  voltage,  or  an  angular  displacement  of  180  d^rees  between 
high  and  low  voltages.  Group  3  consists  of  the  delta-Y  or  Y-delta 
bank,  in  which  the  angular  displacement  is  30^. 

Three-phase  transformer  banks  will  not  operate  in  parallel 
unless  the  angular  displacements  between  high  and  low  voltages 
are  equal.    The  operative  parallel  connections  are  as  follows: 

TABLE  XLVIII 
Operative  Parallel  Connections 


LOW-VOLTAQK  SIDE 

HIQH-VOLTAGB  BIDE 

A 

B 

A 

B 

1 

Delta 

Delta 

Delta 

Delta 

2 

Y 

Y 

Y 

Y 

3 

Delta 

Y 

Delta 

Y 

4 

Y 

Delta 

Y 

Delta 

5 

Delta 

Delta 

Y 

Y 

6 

Delta 

Y 

Y 

Delta 

7 

Y 

Y 

Delta 

Delta 

8 

Y 

Delta 

Delta 

Y 

Digitized  by 


Google 


TRANSFORMERS 


423 


There  are  four  other  combinations  possible  for  these  two 
banks  of  transformers,  but  these  combinations  will  not  operate 
in  parallel.    These  are  as  follows: 

TABLE  XLIX 
Inoperative  Parallel  Connbctionb 


LOW-VOLTAOB  BIDE 

HXQH-yOLTAOB  SIDB 

A 

B 

A 

B 

1 

Delta 

Delta 

Delta 

Y 

2 

Delta 

Delta 

Y 

Delta 

3 

Y 

Y 

Delta 

Y 

4 

Y 

Y 

Y 

Delta 

For  example,  consider  case  No.  2 — ^low-voltage  sides  in  delta 
and  high-voltage  sides  in  Y  and  delta  respectively.  Then  as- 
suming the  low-voltage  sides  already  paralleled  and  high-voltage 
sides  open,  the  phase  diagrams  are  as  follows  where  4,  B,  C,  a, 
6,  c  represent  one    bank 

A 

A 


/r^ 


\ 


\Q 


Fig.  267a. 
Low-voltage  Side. 


Fig.  267b. 
High-voltage  Side. 


and  X,  y,  Z,  x,  y,  z  the 
second  bank.  (See  Fig. 
267a). 

Then  if  b  t^nd  y  be 
joined  on  the  low-voltage 
side,  serious  displacement 
voltages  occur  between  a 
and  X  and  c  and  z  (see 
Fig.  267b),  and  if  these  terminals  are  connected,  these  displace- 
ment voltages  will  cause  heavy  short-circuit  currents  and  destroy 
the  transformers. 

The  reversal  of  two  leads  of  either  the  high-and-low  voltage 
windings  will  reverse  the  polarity,  this  being  identical  with  re- 
versing one  winding.  Reversing  the  Une  leads  of  a  delta-  or  T- 
connected  combination  will,  however,  not  reverse  the  polarity, 
since  the  transformer  leads  themselves  must  be  changed  in  order 
to  make  the  change  in  polarity. 

With  delta-delta  connection,  the  reversal  of  one  or  two 
high-voltage  windings  will  immediately  produce  a  short-circuit 


Digitized  by 


Google 


424  ELECTRICAL  EQUIPMENT 

when  the  low-voltage  delta  is  closed  and  the  maximum  voltage 
difference  will  be  double  line  voltage. 

For  delta- Y  connection,  such  a  reversal  w'll  not  produce  a 
short  circuit  when  the  Y  is  closed,  but  the  voltages  and  phase 
relations  will  be  unequal.  The  maximum  potential  difference 
will  equal  the  line  voltage. 

A  reversal  of  one  or  two  high-voltage  windings  with  a  Y- 
delta  connection  will  immediately  produce  a  short  circuit  when 
the  delta  is  closed,  and  the  maximum  potential  difference  will 
be  double  line  voltage. 

With  Y-Y  connection  the  result  of  reversing  a  high-voltage 
coil  will  be  the  same  as  for  the  delta-Y  connection. 

Effect  of  Ratio  on  Parallel  Operation.  For  successful  paraDel 
operation,  correct  ratios  between  the  high-  and  low-voltage  wind- 
ings of  the  different  banks  is,  as  previously  mentioned,  also  essen- 
tial, otherwise  a  cross-current  will  be  established,  even  if  the  ratios 
are  only  slightly  different.  This  current  is  then  due  to  the  dif- 
ference of  the  two  voltages  divided  by  the  simi  of  the  impedances 
of  the  two  transformers,  and  its  effect  is  to  balance  the  voltages  of 
the  two  transformers  with  a  resultant  equilibrium  of  the  two 
transformers. 

To  determine  this  current,  assume  that  ei  and  zi  are  the  volt- 
age and  impedance  in  low-voltage  terms  of  one  transformer  and 
€2  and  Z2  are  corresponding  terms  of  the  second  transformer,  con- 
nected in  parallel  with  the  other.  The  circulating  current  would 
then  be 


1  = 


Zi  +  Z2 

where  21  and  Z2  are  expressed  in  ohms.    Or  expressed  in  percentage 
of  normal  current  by  the  following  formula: 

^  ,  •    Per  cent  voltage  difference,, -^^ 

Per  cent  / = 5 ^^ ^. j X 100. 

bum  of  per  cent  unpedance 

For  example,  suppose  that  the  voltage  ratios  of  two  trans- 
formers are  such  as  to  cause  a  voltage  difference  of  2  per  cent.  If 
each  transformer  furthermore  has  a  2  per  cent  impedance,  the 
circulating  current  is  equal  to 

2 
Per  cent  / = ^rro  ^  ^^  ~  ^  ^^  ^^^*' 


Digitized  by 


Google 


TRANSFORMERS  425 

which  means  that  a  current  equal  to  50  per  cent  of  normal  circulates 
between  the  transformers  in  both  high-  and  low-voltage  windings. 
It  adds  to  the  load  current  in  the  transformer  having  the  higher 
induced  voltage  and  subtracts  in  the  other,  causing  the  former  to 
carry  the  greater  load. 

The  imp)edance  Zi  can  be  found  for  the  first  transformer  by 
impressing  a  voltage  on  the  low-voltage  winding  with  the  high- 
voltage  winding  short-circuited.    The  current  is  then  read,  and  if 

E 
I  is  the  current  and  E  the  voltage,  then  zi =-y.    In  the  same  man- 
ner Z2  is  determined. 

With  three-phase  delta-delta-connected  transformers  different 
voltage  ratios  will  cause  unbalanced  voltages  and  set  up  a  circu- 
lating current  within  the  delta  in  both  the  high-  and  low-voltage 
windings.  Unbalanced  voltages  outside  the  delta  can,  however, 
not  produce  any  circulating  currents  within  the  delta,  and  un- 
balanced voltages  applied  to  a  delta-connected  transformer  bank 
cannot  be  equahzed  on  the  low-voltage  side  by  the  introduction 
of  additional  voltage  in  the  delta. 

As  with  single-phase  transformers  the  value  of  the  circulating 
cmrent  is  obtained  by  dividing  the  voltage  difference  by  the  total 
impedance  of  the  transformer  bank.  For  example,  if  three  trans- 
formers having  impedances  of  4  per  cent  are  connected  delta- 
delta,  and  one  has  a  ratio  1  per  cent  greater  than  the  other  two,  the 
resulting  circulating  ciurent  will  be 

Per  cent  /=^—^x  100=8.33  per  cent. 

When  the  load  is  taken  from  such  a  bank,  the  load  currents  and 
circulating  currents  are  superimposed,  and  the  transformer  having 
the  highest  secondary  voltage  will  carry  the  greatest  load,  as 
before. 

With  delta-Y-connected  transformers  a  slight  difference  in  the 
ratios  has  a  very  small  effect  compared  with  a  delta-delta-con- 
nected bank.  This  is  due  to  the  shifting  of  the  neutral  point, 
causing  an  equaUzation  of  the  voltages. 

Effect  of  Impedance  on  Parallel  Operation.  In  addition  to 
identical  polarities  and  voltage  ratios  a  successful  parallel  opera- 
tion of  transformers  requires  that  their  impedances  are  in  inverse 
proportion  to  the  load  which  they  are  to  carry,  so  that  the  voltage 


Digitized  by 


Google 


426  ELECTRICAL  EQUIPMENT 

drop  from  no  load  to  full  load  is  the  same  in  all  the  units,  both  id 
magnitude  and  phase. 

The  impedance  of  a  transformer  is  generally  expressed  as  the 
voltage  drop  at  normal  load  in  percentage  of  normal  voltage.  It  is 
the  resultant  of  two  components;  the  resistance  drop,  which 
depends  only  on  the  ohmic  resistance  of  the  windings  and  is  in 
phase  with  the  current,  and  the  reactance  drop,  which  depends 
on  the  magnetic  leakage  between  the  high-  and  low-tension  wind- 
ings and  is  90®  out  of  phase  with  the  current. 

Thus  per  cent  ZZ=V  (per  cent  IR)^+ (per  cent  /X)^, 

where  ZZ= total  impedance  drop; 

7/2= resistance  drop  of  high-  and  low-voltage  windings; 
/Xss  reactance  drop  of  high-  and  low-voltage  windings. 

The  value  of  per  cent  IZ  is  easily  obtained  by  short-circuiting 
one  winding  and  measuring  the  e.m.f.  which  must  be  applied  at 
the  terminals  of  the  other  winding  to  force  full-load  currents 
through  the  winding  at  normal  frequency.  The  impedance  may, 
therefore,  be  measured  directly. 

The  resistance  e.m.f.  is  equal  to  the  high-voltage  ciurent  mul- 
tiplied by  the  equivalent  resistance  of  the  transformer,  which  may 
be  obtained  by  measuring  the  resistance  of  both  the  high-  and  low- 
voltage  windings,  and,  adding  to  the  resistance  of  the  high-voltage 
windings  that  of  the  low-voltage  multipUed  by  the  square  of  the 
ratio  of  transformation. 

The  reactance  e.m.f.  may  be  calculated  from  the  known  values 
for  the  impedance  cm.f .  and  resistance  e.m.f.    Thus 

IX=V(/Z)2-(/i2)2. 

In  the  majority  of  power  transformers,  the  total  resistance 
drop  is  small  compared  to  the  reactance  drop,  in  which  case  the 
p)er  cent  impedance  drop  (per  cent  IZ)  can  be  taken  as  approx- 
imately equal  to  the  per  cent  reactance  drop  (per  cent  IX).  In 
many  lighting  transformers,  however,  where  the  reactance  is 
made  as  small  as  possible,  this  cannot  be  done  without  introducing 
considerable  error. 

The  following  formulae  may  be  used  for  finding  the  division  of 


Digitized  by 


Google 


TRANSFORMERS  427 

load  between  any  number  of  transformer  banks  operating  in  par- 
allel on  single-phase  circuits. 

/     Kv.A.     \ 

J Vper  cent  ZZ/i , 

^^""Z     Kv.A.     \       /      Kv.A.     \  ^^^ 

\p)er  cent  IZ)  \     \per  cent  IZ/2+]  ... 

/     Kv.A.     \ 

J \per  cent/Z/2 j 

^^•"/      Kv.A     \       /      Kv.A.     \  ^^"^ 

\percent/Z/i     \per  cent /Z/2+;  .  .  . 

where  /i  =  load  current  in  transformer  bank  No.  1; 
72= load  current  in  transformer  bank  No.  2; 
ZL=line  current  for  any  given  load; 

'    'rr)  ==  capacity  rating  of  bank  No.  1,  divided  by  its 

per  cent  IZ/  ^        *-      -^  ° 

per  cent  impedance; 

'  ^  '     )  =  capacity  rating  of  bank  No.  2,  divided  by  its 

per  cent/Z/2 

per  cent  imp)edance. 

The  above  formulae  are,  however,  only  correct  when  the  relative 
ratio  between  the  resistance  and  reactance  of  all  the  transformers 
are  equal.  If  not,  the  sum  of  the  individual  load  currents  will  be 
greater  than  the  current  in  the  line,  due  to  a  phase  difference 
between  the  gurrents  in  the  different  transformers.  The  error 
introduced  by  the  inequalities  in  the  values  of  this  ratio  is  gen- 
erally so  small  that  it  can  be  safely  neglected. 

For  delta-delta  connected  transformers  the  effect  of  different 
impedances  is  also  an  unequal  division  of  load  among  the  three 
transformers.  The  curves  of  Fig.  268  show  the  relation  of  current 
in  the  three  legs  of  the  delta,  assuming  two  legs  always  to  be  alike 
in  percentage  impedance  and  capacity.  The  absciss*  represent 
ratio  of  impedances  of  like  legs  to  the  odd  leg. 

Z3    Z3 


Digitized  by 


Google 


428  ELECTRICAL  EQUIPMENT 

Where  Zi,  Z2,  and  Z3,  are  the  impedances  of  the  different  legs. 

oer  cent  IZ 
But  since  Z  is  proportional  to  ^—^^ — r —  we  can  write 


(per  cent  ZZ\       /per  cent  IZ\ 
Kv.A.     /i^V     Kv.A.     /a 
(per  cent  IZ\       /per  cent  IZ\ 
Kv.A.     /3     \     Kv.A.     /a 


If  Zl=  line  current  for  any  given  balanced  load,  and  Zi,  Z2,  and  h 
are  the  leg  currents,  with  the  same  load,  the  ordinates  of  the  cur\^e 

represent  the  ratio  of  leg  current  to  Une  current  fi  =  ir  and  7-, 

respectively. 

If,  for  example,  we  have  three  transformers  connected  in  delta- 
delta,  with  capacities  and  impedances  as  follows: 

Kv.A.i  =  100,  per  cent  ZZi  =  2; 
Kv.A.2  =  100,  per  cent  ZZ2  =  2; 
Kv.A.3=  50,  per  cent  ZZ3  =  2.3; 
Line  voltage  =  1000; 


we  find  that 

and 

also 


^»=:J?=0.68; 

It,       iL 

^=0.40. 
It. 


If  /i  =  100  amp.,  the  normal  current  for  that  transformer, 

^'^=0:68=^^^*'"P- 

Z3  would  then  be  equal  to  147X0.40  =  59  amp.  or  18  per  cent 
overload  on  leg  3. 

Again,  if  we  assume  that  Z3  =  50  so  as  not  to  overload  leg  3, 
then  Zl  =  125  and  Zi  =  85  and  legs,  1  and  2  are,  therefore,  carrj'ing 
only  85  per  cent  of  their  rated  capacity.  This  means  that  without 
any  overload  on  any  of  the  three  transformers,  the  system  can 


Digitized  by 


Google 


TRANSFORMERS 


429 


cany  only  125  amp.  line  current  or  87  per  cent  of  the  rated  capacity 
of  the  three  transformers. 

At  the  point  where  r=0,  we  have  the  current  in  legs  1  and  2 
equal  to  the  line  current,  giving  the  condition  of  open  delta.    By 


1.0 

1 

— 

^ 

n 

r 

"^ 

— 

— 

— 

— 

— 

— 

r 

■— 

— 

r 

'^ 

\ 

Otr 

s 

I; 

, 

V, 

'I 

t= 

_ 

_ 

^A 

^— 

^ 

"- 

a 

s 

J 

"" 

3 

'-' 

u 

>>» 

■^ 

2 

^ 

^ 

n 

^ 

^_ 

_ 

_ 

._ 

_ 

_ 

_ 

_ 

_ 

, 

_ 

i- 

u 

- 

^ 

- 

?=* 

^ 

- 

^ 

- 

" 

- 

- 

^ 

- 

T 

= 

^ 

— 

^ 

= 

OQA 

/ 

l^ 

L 

/ 

3, 

> 

/ 

1 

0.3 

J 

f 

i 

/ 

— 1 

i 

•< 

r 

1 

» 

1, 

n 

2 

Q 

3 

ft 

4 

0) 

T— Batio  of  Impedances 
Fia.  268. 

decreasing  the  capacity  of  leg  3  to  zero,  which  is  the  same  as 
increasing  its  impedance  to  infinity,  we  have  but  two  legs  on  which 
to  carry  the  three-phase  load. 

With  delta- Y-connected  transformer  banks  a  small  difference 
in  the  per  cent  impedance  has,  as  for  the  voltage  ratios,  a  negligible 
effect.  For  example,  if  two  transformers  having  impedances  of 
6  per  cent  are  connected  in  delta- Y  with  another  transformer 
ha\dng  an  impedance  of  3  per  cent,  the  potential  of  the  neutral 
point  will  be  shifted  at  full  load  by  an  amount  approximately  equal 
to  one-third  of  (6%-3%)  or  1  per  cent  of  the  normal  voltage  of  the 
transformer. 

Mechanical  Design.  For  self-cooled  power  transformers  of 
moderate  capacity  the  tanks  are  generally  made  of  corrugated 
sheet  steel  (Fig.  269),  the  bottom  of  the  top  edges  of  which  are 
permanently  cast  into  the  base  and  the  top  rim  simultaneously 
with  the  pouring  of  the  castings,  thus  forming  a  perfectly  cast- 
welded  joint.  For  larger  sizes  tubular  tanks  are  usually  supplied. 
These  are  of  the  plain  steel-plate  construction  with  a*  number  of 
wrought-iron  tubes,  so  arranged  with  connections  at  top  and  bot- 
tom as  to  allow  a  natural  circulation  of  the  oil  between  the  tank 


Digitized  by 


Google 


430 


ELECTRICAL  EQUIPMENT 


and  the  tubes  (Fig.  270).  All  the  jointB  are  welded  and  oil-tight. 
For  the  very  largest  sizes,  where  the  tank  with  attached  radiator 
tubes  becomes  too  large  for  transportation,  a  design  shown  in 


Fig.  269.— Self-cooled  Transformer  with  Corrugated  Tank.     Outdoor  Type. 

Fig.  271  has  been  used.  It  consists  of  separate  radiator  sections 
of  welded,  fluted  steel,  which  may  be  detached  dining  trans- 
portation. 


Digitized  by 


Google 


TRANSFORMERS 


431 


For  water-cooled  transformers  the  tanks  are  mostly  of  a  heavy 
steel-plate  construction  with  all  joints  welded  (Fig.  272).  Some- 
times a  corrugated  design  is  also  used  to  increase  the  radiating 
surface. 

It  is  advisable  to  have  the  transformer  covers  tight-fitting  to 


Fig.  270. — Combination  Self-cooled,  Water-cooled  Transformer. 

prevent  entrance  of  moisture.  This  is  effectively  accomplished 
by  placing  a  gasket  between  the  tank  and  the  cover.  In  order, 
however,  to  maintain  atmospheric  pressure  in  the  air  space  above 
the  oil,  "  breathers ''  are,  as  a  rule,  used.  This  equalizes  the 
pressure  within  and  without  the  tank  and  prevents  the  precipi- 


Digitized  by 


Google 


432 


ELECTRICAL  EQUIPMENT 


tation  of  moisture  from  the  enclosed  air,  which  would  take  place, 
due  to  imequal  pressure  and  the  resulting  condensation,  if  adequate 
faciUties  for  breathing  were  not  arranged.  The  chloride-filled 
breather  is  generally  considered  the  best  type,  its  location  being 
shown  in  Fig.  276. 

The  tanks  may  also  be  completely  filled  with  oil  and  provided 


Fig.  271.— 8000-Kv.A.,   44,000-6600-Volt    Radiator  Type,  Outdoor  Trans- 
former. 


with  expansion  tanks,  thus  giving  the  extreme  protection  against 
moisture  or  the  collection  of  explosive  gases. 

Most  tanks  are  suitable  for  indoor  or  outdoor  service  if  proper 
cover  and  bushing  equipment  is  provided. 

In  order  to  facilitate  moving  it  may  sometimes  be  advisable  to 
equip  the  transformers  with  wheels  or  trucks.    If  wheels  alone  are 


Digitized  by 


Google 


TRANSFORMERS 


433 


Fig.  272. — 4000-Kv.A.-55,000-Volt  Single-phase  Water-cooled  Transformer 

desired,  they  are  usually  mounted  on  axles  attached  to  the  base  of 
the  tank.  Trucks,  on  the  other  hand,  consist  of  a  structural  steel 
frame  with  wheels  fitted  into  the  same. 

When  reasonably  pure  water  can  be  obtained,  no  trouble  is 
experienced  with  cooling  coils  in  water-cooled  transformers,  but 
if  the  water  is  imusually  impure  the  cooling  system  is  liable  to 


Digitized  by 


Google 


434  ELECTRICAL  EQUIPMENT 

give  trouble  due  to  the  pipe  coils  being  clogged  up  or  destroyed  in 
three  ways: 

1st.  Corrosion  due  to  air  in  the  water. 
2d.  Corrosion  due  to  acids  or  alkali  in  the  water. 
.    3d.  Deposit  of  solid  matter  from  the  water. 

The  special  grade  of  iron  used  in  the  manufacture  of  cooling 
coils  offers  much  greater  resistance  to  corrosion  than  ordinary 
steel  does.  On  this  account,  it  is  only  under  exceptionally  severe 
conditions  that  it  is  economical  to  take  the  extra  precaution  of 
using  copper  coils;  brass  being  considered  inferior  to  copper. 

Iron  coils  will  not  be  noticeably  corroded  by  the  air  ordinarily 
held  in  suspension  in  the  water.  If  the  cooling  water  is  taken  from 
a  supply  of  shallow  or  rapidly  moving  water,  such  water  is  likely  to 
contain  an  abnormal  amount  of  air  which  will  rapidly  attack  the 
inner  surface  of  the  cooling  coil.  When  it  is  suspected  that  the 
water  contains  acid  or  alkali  it  should  be  analyzed  and  the  results 
referred  to  the  experts  for  advice.  A  one-gallon  sample  is  neces- 
sary for  a  proper  analysis. 

When  there  is  an  excessive  quantity  of  alkali  or  earth  salts  in 
solution,  the  heating  of  the  water  will  cause  a  deposit  of  this  salt 
previously  in  solution.  Such  an  action  will,  of  course,  take  place 
regardless  of  the  material  of  the  cooling  coil  and  can  be  best 
guarded  against  by  operating  wdth  a  rapid  flow  of  water  with  its 
resulting  low  temperature  and  flushing  action.  When  the  water 
has  much  suspended  solid  matter,  that  is,  if  it  is  muddy,  it  should 
be  filtered,  or  in  less  severe  cases  protection  could  be  obtained  by  a 
rapid  flow  of  water.  The  deposit  of  such  solids  in  the  water  will 
become  more  rapid  as  the  surface  is  roughened  by  deposit  or  cor- 
rosion, due  to  the  increased  resistance  in  the  path  of  the  outer 
portion  of  the  column  of  water. 

Wrought-iron  cooling  pipe  is  ordinarily  made  of  extra  heavy 
lap-welded  inch  or  inch  and  one-half  pipe,  withstanding  a  test 
pressure  of  1000  pounds  per  square  inch.  Copper  coils,  on  the 
other  hand,  are  mostly  designed  to  withstand  a  test  pressure  of 
250  pounds  per  square  inch  and  are,  therefore,  not  as  desirable  as 
iron  coils  from  a  mechanical  standpoint. 

The  coils,  which  may  be  constructed  in  single  or  multiple 
layers,  are  placed  inside  the  upper  part  of  the  tank  and  are  usually 
bolted  to  the  same     By  means  of  a  three-way  valve  at  the  inlet 


Digitized  by 


Google 


TRANSFORMERS 


435 


(Fig.  273)  the  water  may  be  admitted  to,  shut-off  from,  or  drained 
from  the  coil,  the  draining  being  by  gravity. 

Water-flow  indicators  are  desirable  in  order  to  enable  the 
attendants  to  quickly  observe  that  the  water  is  flowing  inasjnuch 


Fig.  273. — Typical  Transformer   Installation  Showing  Water  Piping  Con- 
nections and  Three-way  Valve. 

as  most  water-cooled  transfonners  would  overheat  in  a  short  time 
if  the  water  supply  were  shut  off.  There  are  two  kinds  of  flow 
indicators  in  general  use.  The  sight-flow  indicator  and  the  check- 
valve  indicator.    The  former  is  of  the  open  type  and  consists  of  a 


Digitized  by 


Google 


436 


ELECTRICAL  EQUIPMENT 


funnel-shaped  bowl  into  which  the  water  flows  and  from  which  it 
drains  into  the  waste.  The  latter  is  constructed  on  the  check- 
valve  principle.     It  is  provided  with  a  valve  rod  working  through 


Fig.  274. — Water-cooled  Shell-type  Transformer  Removed  from  Tank. 

a  water-tight  bushing  and  acting  to  close  an  electric  circuit  which, 
in  turn,  may  light  a  lamp  or  operate  a  relay-depending  on  the 
condition  of  the  water  flow.    When  this  is  stopped  or  reduced 


Digitized  by 


Google 


TRANSFORMERS 


437 


below  a  certain  point,  the  circuit  is  broken  by  the  action  of  a 
springy  and  the  lamp  goes  out.  It  may  also  be  obtained  with  an 
indicator  for  use  on  open-circuit  signal  systems,  in  which  case  the 
signal  circuit  is  closed  when  the  water  flow  is  interrupted. 

With  the  shell-type  construction  the  core  iron  is  assembled 


Fig.  275,— Core  for  4000  Kv.A.  Core-type  Transformer. 

around  the  coils  (Fig.  274),  and  follows  the  coil  assembly  instead 
of  preceding  it  as  in  core-type  construction.  The  cores  for  the 
latter  are  two-legged  for  single-phase  (Fig.  275)  and  thr^e-legged 
for  three-phase  units.  They  are  built  up  from  sheet  laminations 
of  high-grade  non-aging  siUcon  steel,  clamped  at  top  and  bottom 


Digitized  by 


Google 


438 


ELECTRICAL  EQUIPMENT 


between  angle  irons,  and  are  insulated  from  the  windings  by  oil 
duets  and  cylindrical  insulating  tubes. 


Fig.  276. — 5000-Kv.A.   Circular   Disc-coil,   Core-type,   Three-phase   Trans- 
former, Facing  Low-voltage  Side. 

The  windings   of  shell-type   transformers  consist  of  rectan- 
gular shaped  coils,  while,  for  the  core-type  design  these  have  a 


Digitized  by 


Google 


TRANSFORMERS  439 

circular  shape  which  has  many  advantages  over  the  former,  in 
that  they  can  be  more  easily  insulated  and  supported  to  with- 
stand the  mechanical  stresses  due  to  short-circuits. 

With  shell-type  windings  the  coils  are  assembled  into  several 
primary  and  secondary  groups  so  mixed  as  to  obtain  the  proper 
compromise  between  voltage  regulation  and  a  desirable  reactance, 
the  spacings  being  furthermore  dependent  on  the  required  dielec- 
tric and  the  oil  flow  necessary  for  cooling. 

With  the  core-type  diBsign,  the  following  three  diflferent  wind- 
ing arrangements  are  in  use: 

1.  Interleaved  disc  coils,  for  low  and  moderate  voltages. 

2.  Concentric  cylinder  coils,  for  intermediate  voltages. 

3.  Concentric  disc  cylinder  coils,  for  high  voltages. 

With  the  interleaved  construction  the  coils  are  assembled 
horizontally  over  an  insulating  cylinder  around  the  core,  the 
primary  and  secondary  coils  being  interleaved  in  synmietrical 
groups  with  insulating  oil  ducts  and  barriers  between  them 
(see  Figs.  276  and  277).  They  are  usually  wound  with  rectangular 
conductor,  one  turn  per  layer.  The  whole  structure  is  securely 
braced  at  each  end  by  plates  rigidly  engaging  with  the  steel  channel 
core  clamps.  There  are  usually  four  or  more  groups  in  the  wind- 
ings, depending  upon  the  capacity,  voltage  and  the  required  react- 
ance. 

The  concentric  cyUnder  type  involves  a  construction  in  which 
all  the  coils  are  in  the  form  of  cylinders  assembled  concentrically 
around  the  core  legs,  insulated  from  each  other  and  from  the  core 
by  insulating  cylinders  (Fig.  278).  The  low-voltage  coils  are 
placed  nearest  the  core  and  may  be  wound  with  rectangular  strip 
on  edge  or  flat  depending  upon  the  number  of  turns  and  the  size  of 
conductor  required.  The  high-voltage  coils  may  be  single-  or 
double-cylinder  edge-wound  coils  or,  if  the  size  of  conductor  is 
small,  the  winding  may  be  broken  up  into  a  number  of  small  sec- 
tions and  wound  with  round  wire  in  layers. 

The  concentric  disc  cylinder  type  is  a  combination  of  the 
above,  the  high-voltage  coils  being  of  the  disc  form  and  wound 
the  same  as  the  coils  for  the  interleaved  disc  type,  while  the  low 
voltage  coils  are  cylindrical  the  same  as  in  the  concentric  cylinder 
type  (Fig.  279).  The  high-voltage  coil  is  placed  outside  and  the 
low  voltage  inside,  next  to  the  core,  cylindrical  insulations  being 


Digitized  by 


Google 


1 IP 

1             IP 

(           IS 

Fig.  277. — Interleaved  Disc  CJoil  Windings  for  Ck)re-type  Transformers. 


p 

( 1 

r__j 

"1^ 

P 

( 1 

t    —1 

J          1 

^ 

Fig.  278. — Concentric  Cylinder  Coil      Fig.  279. — Concentric  Disc-cylinder 
Windings    for    Core-type    Trans-  Coil      Windings     for     Core-type 

formers.  Transformers. 

440 


Digitized  by 


Google 


TRANSFORMEBS  441 

placed  beween  the  core  and  the  low-voltage  coil,  and  between  the 
high-  and  low-voltage  coils  the  same  as  used  for  the  other  types. 
This  construction  is  necessary  for  transformers  of  the  higher 
voltages,  requiring  a  greater  number  of  turns  and  an  increased 
amount  of  turn  insulation.  Therefore,  it  can  be  seen  that,  at  a 
given  capacity  a  point  will  be  reached  where  it  will  be  impossible 
to  use  cylindrical  edge-wound  coils  because  the  conductor  will 
become  too  thin  to  wind  on  edge,  while,  on  the  other  hand,  the 
losses  may  limit  the  use  of  such  a  construction  before  the  mechanical 
considerations.  As  the  edge-woimd  conductor  becomes  thin  on 
account  of  increased  turns  and  insulation,  the  width,  which  is  the 
thickness  of  the  cylinder,  must  be  increased  sufficiently  to  give 
the  proper  current  carrying  capacity.  This  width  is  perpen- 
dicular to  the  leakage  flux  and  eddy  ciurent  losses  are  accordingly 
set  up  in  the  conductors.  It  is  quite  possible  to  reach  a  point 
where  an  increase  in  the  width  of  the  conductor  will  give  an 
increased  total  loss  in  the  copper  on  account  of  the  eddy  current 
losses  increasing  faster  than  the  PR  loss  decreases,  due  to  a  layer 
conductor.  The  disc  coils  effectively  overcome  these  difficulties^ 
first,  because  the  width  of  the  coils  is  sufficient  to  accommodate 
the  required  turns,  and  second,  because  the  width  of  the  rect* 
angular  conductor  is  parallel  to  the  leakage  flux  and,  therefore, 
does  not  increase  the  eddy  current  loss. 

After  the  coils  are  wound  they  are  clamped  to  dimensions  and 
thoroughly  baked  and  vacuum  treated  to  insure  the  complete 
elimination  of  moisture.  Nimierous  treatments  in  insulating 
compounds  are  then  applied,  sealing  up  all  interstices  and  cement- 
ing each  coil  into  a  soUd  structure.  The  coils  are  then  subjected 
to  further  baking,  after  which  the  clamps  ai'e  removed  and  the 
proper  number  of  tapings  applied,  followed  by  the  final  series 
of  treatments  and  bakings,  after  which  the  coils  are  ready  for 
assembly. 

The  taps  are  always  placed  in  the  coils  located  in  the  central 
portions  of  the  winding  where  the  potential  strains  are  at  a  mini- 
mum. To  faciUtate  the  bringing  up  of  several  leads  from  the  t^ps, 
a  new  arrangement  is  being  used  in  modem  transformers.  It 
consists  of  multi-conductor  leads,  two  or  more  insulated  cables 
being  boimd  together  and  heavily  wrapped  with  varaished  cambric, 
forming  a  stiff  solid  structure  that  is  easUy  supported  and  well 
insulated  from  groimd  (Fig.  276).    Each  element  of  the  group 


Digitized  by 


Google 


442  ELECTRICAL  EQUIPMENT 

terminates  in  a  threaded  stud  mounted  in  a  circular  fiber  disc  with 
arrangement  for  interconnection  by  short  links.  To  prevent  the 
possibility  of  short-circuiting  sections  of  the  winding,  all  threaded 
studs,  between  which  short  circuits  could  be  made,  have  the  same 
thread  and  dimensions,  while  the  studs  to  be  connected  differ  in 
size.  The  connecting  link  is  also  fitted  with  couplings  which  differ 
in  size  from  one  another  so  that  unlike  studs  on  the  circular  disc 
may  be  coupled;  this  arrangement  rendering  harmful  connections 
impossible. 

When  the  above-described  multi-conductor  arrangement  does 
not  prove  practical  on  account  of  very  high  voltages  or  too  many 
taps,  a  terminal  connection  board,  generally  made  of  oil-treated 
maple,  can  be  used,  to  which  all  leads  are  brought,  separately 
bushed  and  provided  with  suitable  terminals  for  interconnections. 
This  board  is  normally  submerged  in  the  oil. 

The  main  leads  in  self  and  water-cooled  power  transformers 
arc  brought  out  through  insulating  bushings  in  the  cover.  Usually 
only  two  high-tension  terminals  are  brought  out  for  single-phase 
units,  while  for  three-phase  units  three  or  four  bushings  may  be 
provided,  depending  on  whether  the  neutral  is  to  be  brought  out. 
The  same  also  applies  to  the  low-tension  leads. 

The  design  of  the  leads  for  moderate  voltages  involves  no 
difficulties.  For  indoor  transformers  they  usually  consist  of  a 
metal  rod  heavily  insulated  with  several  wrappings  of  black  var- 
nished cambric,  fiber  collars  being  added  for  the  high-voltage 
ranges  to  increase  the  creeping  surface  (Fig.  272).  For  outdoor 
service  these  leads  are  covered  with  a  petticoated  porcelain 
bushing.  The  conductor  may  also  consist  of  a  flexible  cable 
pa&sing  up  through  a  tube  making  connection  between  the  line 
and  the  transformer  winding.  The  lead  proper  (flexible  cable) 
may,  therefore,  be  disconnected  from  the  line  at  the  top  of  the 
bushing  and  slid  down  through  it,  in  ca*je  it  is  desirable  to  remove 
the  cover  from  the  tank  without  disturbing  the  cone. 

For  higher  voltages,  above  70,000,  the  bushing  design  involves 
greater  difficulties,  it  being  necessary  to  carefully  equalize  the 
potential  and  keep  the  gradient  at  or  below  the  amount  which  is 
safe  for  the  weakest  point.  The  latest  type  of  compound-filled 
bushings  with  one-piece  porcelain  shells  is  undoubtedly  the  most 
satisfactory  design  brought  out  to  date  (Fig.  280).  They  consist 
of  a  single  top  and  a  single  bottom  porcelain  with  a  central  section 


Digitized  by 


Google 


TRANSFORMERS 


443 


of  metal  grounded  to  the  cover.  The  central  metal  section 
extends  below  the  oil  level  of  the  transformer  to  prevent  corona. 
This  type  is  also  provided  with  a  flexible  cable  passing  through  a 
metal  tube  extending  the  length  of  the  bushing  and  supported  at 

top  and  bottom,  connection  being  made  at  

the  top  through  a  water-tight  cap.  The 
central  metal  tube  is  surrounded  by  con- 
centric insulating  cyUnders  dividing  the  oil 
space.  The  joints  between  the  top  and 
bottom  porcelains  and  metal  sections  are 
gasketed  and  clamped,  so  that  they  are 
absolutely  oil-tight.  The  upper  porcelain  is 
surmoimted  by  a  heavy  glass  expansion 
chamber  which  is  also  used  as  a  gauge  in 
filling. 

OiL  Transformers  should  contain  suffi- 
cient oil  to  completely  immerse  the  core, 
windings  and  cooling  coil,  and  a  gauge 
should  be  attached  to  the  tank  in  a  con- 
spicuous place  to  indicate  the  oil  level, 
while  a  valve  should  be  provided  at  the 
bottom  for  drawing  oflF  the  oil. 

Transformer  oils  should  have  good  in- 
sulating properties,  a  high  flash  and  low 
viscosity,  so  that  the  heat  may  be  readily 
conducted  from  the  coils  and  core  to  the 
radiating  surfaces.  The  flash  and  burning 
points  are  second  in  importance  only  to 
viscosity,  and,  in  fact,  vary  together;  that 
is  to  say,  an  oil  having  a  high  burning-point 
compared  with  another  oil  will  probably  be 
high  in  viscosity.  It  is  this  property  of  oil 
to  resist  ignition  until  it  is  first  heated  to  a 
temperature,  known  as  its  fire  or  burning- 
point,  which  enhances  its  value  as  an  insulating  and  cooling 
medium.  At  a  temperature  somewhat  below  the  fire  or  burning- 
point  the  oil  gives  off  vapors  which,  as  they  come  from  the  surface 
of  the  oil,  may  be  ignited  in  Utile  flashes  or  puffs  of  flame.  This  is 
known  as  the  flash-point.  The  oil  will  not  support  combustion, 
however,  until  these  flashes  are  sustained  iminterruptedly,  or,  in 


Fig  .  280.— 155,000-Volt 
Compound  Filled 
Flange-clamped  Por- 
celain Bushing. 


Digitized  by 


Google 


444  ELECTRICAL  EQUIPMENT 

other  words,  until  the  buming-pomt  is  readied.  It  is,  therefore^ 
obvious  that  high  flash  and  burning-points  are  desirable  in  in- 
sulating oils  in  order  that  the  fire  risk  attendant  on  their  use 
may  be  reduced  to  a  minimum. 

Of  extreme  importance  is  also  the  percentage  of  deposit,  which 
may  be  thrown  down  from  an  oil  in  service.  Most  organic  sub- 
stances, when  exposed  to  even  moderate  temperatures,  are  sub- 
ject to  slow  chajsges,  which,  in  case  of  oils,  are  probably  due  to 
chemical  change,  such  as  oxidation  of  some  of  the  constituents, 
and  when  this  deposit  is  excessive  efficient  cooling  is  very  much 
restricted.  Somewhat  similar  to  this  deposit,  some  oils  produce  a 
jelly-like  substance,  which  forms  after  continuous  operation,  and 
in  general,  the  higher  the  temperature  the  more  rapid  these  changes 
take  place.  A  very  sUght  trace  of  the  deposit  is  in  no  degree 
harmful  and  will  ordinarily  only  be  found  under  the  most  severe 
conditions  following  a  long  period  of  service. 

Transformer  oils  must  also  be  watched  for  presence  of  injurioas 
impurities  such  as  acids,  alkalis  and  free  sulphur.  An  access  of 
acid  particularly  would  result  in  deterioration  of  insulation  and 
other  materials  of  which  the  transformer  is  constructed.  Free 
sulphur,  even  in  extremely  minute  quantities,  is  seriously  detri- 
mental to  the  windings,  the  chemical  action  on  exposed  copper 
causing  the  conductors  themselves  to  be  gradually  eaten  through. 
These  characteristics  are,  however,  very  carefully  watched  by  the 
transformer  manufacturers,  so  that  the  oils  furnished  are  ordinaril}' 
free  from  such  injurious  impurities. 

The  characteristics  of  oils  in  general  use  vary  somewhat, 
depending  on  the  type  of  transformer  as  well  as  on  the  practice 
of  the  transformer  manufacturer.  One  of  the  largest  of  these  sup- 
plies oil  of  the  following  characteristics  for  its  water-cooled  trans- 
formers: 

Flash-point 130**  C. 

Burning-point 145**  C. 

Freezing-point — 15**  C. 

Viscosity  at  40**  C 40  sec. 

This  kind  of  oil  is  also  supplied  with  oil-cooled  transformers 
and  combination  self-  and  water-cooled  transformers  where  the 
guaranteed  normal  load  temperature  rise  is  less  that  50^  C 


Digitized  by 


Google 


TRANSFORMERS  445 

Where  the  rise  is  50^  C.  and  higher,  oil  with  the  following  char- 
acteristics is  used: 

Flash-point 160**C. 

Burning-point 175°  C. 

Freezing-point — 10°  C. 

Viscosity  at  40°  C 60  sec. 

Transformers  which  may  be  operated  mider  severe  weather 
conditions,  such  as  outdoor  types,  may  also  be  supplied  with  an 
oil  having  a  freezing-point  of  —30°  C. 

The  necessary  puncture  strength  of  oils  is:  40,000  volts  punc- 
ture with  J-inch  discs  spaced  0.2  inch  apart;  or,  22,000  volts 
puncture  with  1-inch  discs  spaced  0.1  inch  apart. 

In  order  to  ascertain  the  temperature  at  which  a  transformer 
is  operating,  it  is  advisable  to  equip  them  with  thermometers  and 
these  should  be  located  in  such  a  place  that  they  can  easily  be 
read.  Different  thermometers  are  in  use,  some  being  of  the 
ordinary  mercury  type,  this  being  mostly  suppUed  with  self-cooled 
transformers  and  may  be  equipped  with  electrical  contacts  for 
connecting  to  an  alarm  circuit. 

A  thermometer  which  is  very  extensively  used  in  connection 
with  water-cooled  transformers  is  illustrated  in  Fig.  281.  It 
depends  for  its  operation  upon  the  expansion  of  mercury  in  a 
sensitive  steel  tube.  The  bulb  is  connected  to  the  indicating 
instrument  by  a  small  capillary  steel  tube,  this  tube  being  con- 
nected to  a  spring  to  which  the  indicating  pointer  is  attached 
through  a  rack  and  pinion.  The  capillary  tube  is  of  such  length 
that  the  bulb  may  be  placed  in  the  oil  at  the  hottest  part  of  the 
transformer.  Variations  of  temperature  at  the  bulb  cause  cor- 
responding contraction  or  expansion  of  the  Hquid  confined  in 
this  bulb,  and  this  is  transmitted  to  the  capillary  tube  connecting 
to  the  indicating  mechanism.  The  instrument  can  readily  be 
equipped  with  contact  points  for  connection  to  an  alarm  circuit. 

Drying  Transformers.  Transformers  shipped  assembled,  but 
not  filled  with  oil,  should  be  very  thoroughly  and  intelligently 
inspected  before  deciding  that  the  drying  may  be  omitted.  In 
every  case  a  thorough  inspection  is  necessary,  and  if  there  is  any 
evidence  of  mildew  or  moisture,  a  drjdng-out  run  is  necessary. 
Recent  improvements  in  design  and  method  of  shipping  make  it 
practicable  where  conditions  demand  it,  to  ship  transformers  with 


Digitized  by 


Google 


446  ELECTRICAL  EQUIPMENT 

such  precautionary  measures  that  drying  in  most  cases  will  be  un- 
necessary. Large  transformers  with  properly  constructed  tanks, 
provided  with  chloride  breathers  may  be  shipped  with  or  without 
oil.  Small  high-voltage  transformers  may  be  shipped  filled  with 
oil  using  chloride  breathers  where  necessary.  With  such  ship- 
ments careful  examination,  if  shipped  without  oil,  and  oil  tests  if 
shipped  oil  filled,  are  of  utmost  importance.  The  oil  samples 
should  be  taken  both  from  top  and  bottom  after  the  tank  has 


Fia.  281. — Thermometer  with  Electrical  Connections  for  Uae  on 
Water-cooled  Transformers. 

stood  for  twenty-four  hours,  the  required  puncture  strength  of  the 
oil  to  be  as  previously  given. 

Where  transformers  are  shipped  without  the  oil  in  the  tanks 
it  is  almost  invariably  necessary  to  dry  them  out  first.  This  may 
be  accomplished  in  several  ways,  of  which  the  "  external "  and 
the  "  internal  '^  heat  methods  are  mostly  used. 

The  "  external  ^'  method  requires  the  circulation  of  heated  air 
through  the  transformer  in  its  tanks.     Dry  air  is  forced  at  a  tem- 


Digitized  by 


Google 


TRANSFORMERS 


447 


perature  of  85°  C.  into  coils  and  insulation  at  the  bottom  of  the 
transformer,  allowing  same  to  escape  at  the  top.  The  quantity  of 
air  should  be  such  that  the  temperature  of  escaping  air  is  approxi- 
mately the  same  as  the  ingoing  temperature.  Various  pipes  and 
.deflectors  may  have  to  be  used  to  properly  distribute  the  air, 
and  precautions  should  be  taken  to  prevent  oil  from  running 
from  the  transformer  into  the  heater  as  it  may  cause  a  serious 
fire.  The  quantity  of  air  to  give  good  results  with  different  size 
tanks  is  as  follows: 


Diameter  in  In.  or  Equiva- 
lent Area  of  Tank. 

Cu.ft.  Air  per  Min. 

54  to  72  inclusive 

78  to  96  inclusive 

102  to  120  inclusive 

126  to  144  inclusive 

150  to  168  inclusive 

600 

900 

1200 

1500 

1800 

An  outfit  which  is  especially  adapted  for  furnishing  hot  air  for 
transformer  drying  is  shown  in  Fig.  282.  It  consists  of  an  electric 
air  heater  blower  and  air  strainer.  The  air  heater  requires  20  to 
25  Kv.A.  at  110  or  220  volts  to  operate  it,  and  the  blower  about 


Fig.  282. — Hot-air  Drying  Outfit  for  Transformers. 

2  H.P.  to  drive  it  at  normal  output.  The  air  strainer  when  in 
operation  should  be  wrapped  with  cheesecloth  to  prevent  the  dust 
from  entering  the  blower  and  being  blown  into  the  transformer. 
This  cloth  should  be  changed  from  time  to  time  as  the  dirt  accu- 
mulates on  it. 


Digitized  by 


Google 


448  ELECTRICAL  EQUIPMENT 

In  cases  where  it  is  impractical  to  apply  the  hot-air  method  for 
drying,  the  "  internal "  or  short-circuit  method  may  be  used. 
The  transformer  should  then  preferably  be  taken  out  of  the  tank 
or,  otherwise,  the  manhole  cover  should  be  removed  and  the  valve 
in  the  base  opened  to  give  as  great  a  circulation  of  air  as  possible 
under  the  conditions. 

This  method  requires  one  winding  to  be  short-circuited  and  a 
voltage  appUed  to  the  other  so  that  sufficient  current  will  flow  in 
the  windings  to  raise  the  temperature  to  approximately  70**  C. 
The  amount  of  current  necessary  to  effect  this  temperature  ranges 
between  one-fifth  and  one-third  of  the  full-load  current,  depend- 
ing upon  the  room  temperature  and  the  design  of  the  transforaier. 
The  impedance  volts  necessary  to  give  the  specified  range  in  cur- 
rent varies  from  0.4  to  1.5  per  cent  of  the  rated  voltage  of  the 
winding  to  which  the  impedance  voltage  is  appUed. 

The  temperatm^  of  the  winding  can  be  determined  by  the 
increase  in  resistance,  which  is  calculated  as  follows: 

Let  ftc  =  resistance  at  room  temperature,  or  cold  resistance; 
fc=room  or  coil  temperature  for  cold  resistance; 
Rh  =  hot  resistance ; 
iji  =  temperature  of  windings  hot; 

then 

,     RH{2SS+tc)-238Rc 
^'"^  Re  ' 

and  rise 

=  tH-tc. 

A  simple  method  for  determining  the  temperature  of  the  wind- 
ing is  to  assume  that  for  each  per  cent  increase  in  resistance  the 
temperature  rise  is  approximately  2|°  C. 

The  duration  of  the  drying  run  depends  upon  the  voltage  and 
size  of  the  transformer  and  also  upon  its  condition  as  to  moisture 
at  the  time  it  is  dried.  For  transformers  under  20,000  volts  the 
drying  should  be  continued  not  less  than  twenty-four  hours; 
20,000  to  30,000  volts,  48  hours;  between  30,000  and  40,000  volts., 
seventy-two  hours.  Higher  voltages  may  require  longer.  It  is 
obvious  that  some  consideration  must  be  given  to  the  capacity  of 
the  transformer.  Transformers  of  less  than  100  Kv.A.  may  only 
require  twenty-four  hours.  For  transformers  between  200  Kv.A. 
and  500  Kv.A.  the  process  may  be  limited  to  thirty-six  hours; 


Digitized  by 


Google 


TRANSFORMERS  449 

between  500  Kv.A.  and  1000  Kv.A.,  to  forty-eight  hours;  be- 
tween 1000  Kv.A.  and  2000  Kv.A.  to  sixty  hours;  for  all  larger 
capacities  the  process  should  be  carried  on  for  at  least  seventy-two 
hours.  In  case  there  is  no  evidence  that  the  transformer  is  unduly 
moist,  discretion  may  be  used  in  slightly  decreasing  the  limits 
given  for  the  voltage.  A  transformer  of  20,000  to  30,000  volts, 
for  instance,  having  a  capacity  of  200  Kv.A.  or  less,  may  be  dried 
in  only  twenty-four  hours.  The  limits  given  for  the  capacities, 
however,  should  be  rigidly  adhered  to,  and  in  no  case  should  the 
process  be  carried  on  for  less  than  twenty-four  hours. 

While  the  insulation  resistance  of  a  transformer  cannot  be 
relied  upon  as  a  siu-e  indication  of  its  condition  at  any  one  time,  the 
general  trend  of  megger  readings  as  a  drying  nm  proceeds  is  a 
fairly  accurate  indication  of  the  progress  of  drying.  The  drying 
process  should  be  continued  imtil  the  curve  becomes  approxi- 
mately flat  at  an  elevation  considerably  above  the  low  point  of 
the  curve.  Variation  in  temperatures  causes  wide  variation  in 
resistance,  the  values  varying  inversely.  If  the  megger  shows  a 
short  circuit,  that  is,  an  insulation  resistance  too  low  to  be  read, 
it  is  very  likely  due  to  an  excessive  amount  of  moisture.  Low 
readings  also  sometimes  indicate  the  presence  of  moist  spots  in  the 
insulation.  Widely  different  megger  readings  may  be  obtained 
on  different  transformers,  but  average  readings  should  be  approx- 
imately alike  for  transformers  of  the  same  capacity  and  design. 
Shell-type  transformers  have,  in  general,  a  lower  insulation  resist- 
ance than  core-type. 

Oil  Diying.  Oil,  whether  shipped  in  sealed  barrels  or  in 
special  tank  cars  direct  from  the  manufacturer,  may  require 
drying  at  its  destination  before  it  is  suitable  for  use  in  high- volt- 
age transformers.  All  oil  should  be  tested  before  using,  but,  if 
it  is  absolutely  necessary  to  use  a  part  of  oil  from  barrels  before 
tests  can  be  made,  the  barrels  should  be  allowed  to  settle  for  sev- 
eral hours  and  then  the  oil  pumped  from  the  top  to  within  4  inches 
of  the  bottom;  i.e.,  do  not  use  the  oil  which  settles  in  the  bottom 
until  it  can  be  tested  and  dried  if  necessary.  Oil  drums  should  be 
stored  lying  on  their  sides. 

The  best  method  for  dr3ring  and  filtering  oil  consists  of  forcing 
it  under  pressure  through  several  layers  of  blotting  paper,  which 
removes  all  moisture  and  solid  matter  held  in  suspension  in  the  oil. 
A  filter  press,  such  as  shown  in  Fig.  283,  has  been  developed  for 


Digitized  by 


Google 


450 


ELECTRICAL  EQUIPMENT 


this  purpose,  and  by  this  method  from  360  to  1200  gallons  of  oil, 
according  to  the  size  of  the  press,  can  be  treated  in  an  hour. 

The  essential  portions  of  the  filter  consist  of  a  series  of  alternate 
flat  cast-iron  plates  and  frames,  the  blotting  paper  being  placed 
between  them,  and  the  whole  clamped  tightly  by  means  of  a 
large  screw  and  lever  at  one  end.  Both  plates  and  frames  have 
large  cored  holes  in  the  lower  comers,  serving  as  inlet  and  outlet 
for  the  oil.  The  surface  of  the  plates,  except  for  a  one-half  inch 
rim  round  the  edge,  is  grooved  or  corrugated  both  vertically  and 


Fig.  283.— Method  of  Using  Oil  Dryer  and  Filter  to  Dry  OQ  in  a 
Transformer  as  Installed. 


horizontally  on  both  sides,  forming  the  checkered  or  so-called 
"  pyramid  "  surface  which  supports  the  paper  and  forms  channels 
communicating  with  the  outlet  at  the  corners.  This  form  of 
surface  is  more  eflScient  than  a  single  set  of  corrugations  or  the 
use  of  perforated  metal.  The  oil  enters  at  the  lower  left-hand 
corner  of  the  filter,  passes  through  a  series  of  cored  holes  in  the 
plates  and  frames  and  punched  holes  in  the  blotting  paper  and 
enters  and  fills  in  parallel  the  chambers  formed  by  the  frames  and 
plates.    It  then  passes  through  the  blotting  paper,  along  the 


Digitized  by 


Google 


TRANSFORMERS  451 

grooves  of  the  pyramid  surface,  to  the  lower  right-hand  comer 
of  the  plate,  and  then  through  a  series  of  small  holes  drilled  from 
the  surface  of  each  plate  to  a  cored  passageway,  similar  to  the 
inlet.  A  rotary  gear  or  multi-stage  centrifugal  pump  is  used  for 
forcing  the  oil  through  the  filters. 

One  of  the  greatest  advantages  of  this  outfit  is  that  the  treat- 
ment can  be  carried  on  while  the  transformer  is  in  operation,  and 
without  the  use  of  separate  tanks  for  the  oil,  as  seen  in  the  illus- 
tration. 

Oil  Testing.  The  sample  bottles  or  cans  should  be  thoroughly 
cleaned  and  dried  before  using,  and  it  is  generally  satisfactory 
to  rinse  very  thoroughly  with  clean,  dry  oil  and  allow  the  recep- 
tacle to  drain  for  a  few  minutes.  The  test  samples  should  be  taken 
only  after  the  oil  has  settled  for  some  time,  varying  from  eight 
hours  for  a  barrel  to  several  days  for  a  large  transformer.  Cold 
oil  is  much  slower  in  settling  and  may  hardly  settle  at  all.  Oil 
samples  from  barrels  should  be  taken  about  J  inch  from  the  bot- 
tom of  the  dnrni  and  a  brass  or  glass  "  thief  "  can  be  conveniently 
used  for  this  purpose.  The  same  method  should  be  used  for 
cleaning  this  as  is  used  for  container. 

A  compact  oil-testing  set  by  means  of  which  the  dielectric 
strength  of  oil  can  easily  be  determined  is  illustrated  in  Fig.  284. 
It  consists  of  a  testing  transformer  with  an  induction  regulator 
for  voltage  control  and  an  oil-spark  gap,  all  of  which  are  assembled 
as  a  unit.  Before  using,  the  spark  gap  should  be  cleaned  by  simply 
rinsing  with  clean,  dry  oil.  Its  terminals,  which  are  1.0  inch  in 
diameter,  should  be  adjusted  0.1  inch  apart  by  means  of  a  gauge. 
The  spark  receptacle  should  be  nearly  filled  with  the  oil  and 
allowed  to  stand  for  a  moment  to  give  bubbles  time  to  escape, 
especially  if  the  oil  is  cold.  The  rate  of  increase  of  voltage  should 
be  as  fast  as  can  be  accurately  read  on  the  voltmeter,  the  total 
time  of  application  of  voltage,  from  zero  to  breakdown  valve, 
usually  being  about  five  seconds  The  average  voltage  of  five 
tests  is  generally  taken  as  the  dielectric  strength  of  the  oil. 

When  drawing  samples  of  oil  from  the  bottom  of  transformers, 
or  large  tank,  several  quarts  should  be  drawn  off  before  taking 
sample  in  order  to  eliminate  dirt  or  water  which  may  have  accumu- 
lated in  the  valve,  connecting  pipes,  etc.  The  best  way  to  clean 
and  dry  oil  drums  is  to  rinse  them  very  thoroughly  with  five  or 
ten  gallons  of  gasolene,  benzine,  or  dry  transformer  oil.    The 


Digitized  by 


Google 


452 


ELECTRICAL  EQUIPMENT 


rinsing  operation  should  be  repeated  several  times,  using  fresh 
liquid  each  time  and  draining  the  drums  very  thoroughly  after 
each  rinsing. 

Operation.     Artificially  cooled  transformers  should  not  be  run 
continuously,   even  at  no  load,   without  the  cooling  medium. 


Fig.  284.-^0,000-Volt  OU-testing  Set. 

Therefore,  it  is  essential  to  maintain  a  proper  circulation  of  the 
cooling  system. 

K  the  water  circulation  of  water-cooled  transformers  is  for  any 
reason  stopped,  the  load  should  be  immediately  reduced  as  much 
as  possible,  and  a  close  watch  kept  on  the  temperature.  Reduce 
the  load  if  for  any  other  reason  the  oil  at  the  top,  near  the  center 
of  the  tank,  approaches  80°  C.    This  temperature  should  be  rec- 


Digitized  by 


Google 


TRANSFORMERS  463 

ognized  as  an  absolute  limit  and  must  not  be  exceeded    It  should 
be  held  only  during  an  emergency  period  of  short  duration. 

The  ingoing  cooling  water  should  never  have  a  maximum  tem- 
perature of  over  25°  C. 

Nearly  all  cooUng  water  will  in  time  cause  scale  or  sediment  to 
form  in  the  cooling  coils.  The  time  required  to  clog  up  a  coil 
depends  on  the  nature  and  amount  of  foreign  matter  in  the  water. 
The  clogging  materially  decreases  the  eflSciency  of  the  coil  and  is. 
indicated  by  a  high  oil  temperature  and  a  decreased  flow  of  water, 
load  conditions  and  water  pressure  remaining  the  same. 

The  most  frequent  cause  of  clogging  of  iron  cooling  coils  is  a 
large  quantity  of  air  in  the  water,  resulting  in  the  formation  of  a 
scaly  oxide. 

Scale  and  sediment  can  be  removed  from  cooling  coils  without 
removing  the  coils  from  the  tank.  Both  inlet  and  outlet  pipes 
should  be  disconnected  from  the  water  system  and  temporarily 
piped  to  a  point  a  number  of  feet  away  from  the  transformer, 
where  the  coil  can  be  filled  and  emptied  safely.  Especial  care 
must  be  taken  to  prevent  any  acid,  dirt  or  water  from  getting 
into  the  transformer. 

Blow  or  siphon  all  the  water  from  the  cooling  coil  and  then  fill 
it  with  a  solution  of  hydrochloric  acid,  specific  gravity  1.10. 
(Equal  parts  of  concentrated  hydrochloric  acid  and  commercially 
pure  water  will  give  this  specific  gravity.)  After  the  solution  has 
stood  in  the  coils  about  an  hour,  flush  out  thoroughly  with  clean 
water.  If  all  the  scale  is  not  removed  the  first  time,  repeat  until 
the  coil  is  clean,  using  a  new  solution  each  time.  The  number  of 
times  it  is  necessary  to  repeat  the  process  will  depend  on  the  con- 
dition of  the  coil,  though  ordinarily  one  or  two  fillings  will  be  suf- 
ficient. 

The  chemical  action  which  takes  place  is  very  noticeable 
and  often  forces  acid,  sediment,  etc.,  from  both  ends  of  the  coils; 
therefore,  it  is  well  to  leave  both  ends  open  to  prevent  abnormal 
pressure. 

When  water-cooled  transformers  have  operated  for  some  time, 
especially  if  the  operating  temperature?  are  high,  the  oil  may 
leave  a  deposit  on  the  outside  surface  of  the  cooling  coils.  Any 
deposit  decreases  the  eflSciency  of  the  coils  and  should  be  re- 
moved. This  condition  of  the  coils  is  indicated  by  higher  oil 
temperature,  water  flow  and  load  conditions  remaining  the  same 


Digitized  by 


Google 


454  ELECTRICAL  EQUIPMENT 

The  coil  should  be  examined  whenever  indications  point  to  the 
formation  of  a  deposit. 

When  water-cooled  transformers  are  idle  and  exf)osed  to  cold, 
the  water  must  be  drained  or  blown  out  of  the  cooling  coils.  In 
addition  to  draining  or  blowing  out  the  water,  the  cooling  coil 
should  be  dried  by  forcing  heated  air  through  it.  If  not  con- 
venient to  force  heated  air  through  the  coil,  enough  alcohol 
should  be  poured  into  the  coil  to  fill  the  two  bottom  turns  of  each 
section. 

During  the  first  month  of  service  of  transformers  having  a 
potential  of  40,000  volts  or  over,  samples  of  oil  should  be  drawn 
each  week  from  the  bottom  of  the  tank  and  tested.  Samples  from 
aU  transformers  should  be  drawn  and  tested  once  every  six  months. 

If  at  any  time  the  oil  should  puncture  below  the  safe  voltage 
the  filter  press  may  be  used  for  treating  it  without  taking  the 
transformer  out  of  service.  Oil  should  be  drawn  from  valve  in  the 
base,  passed  through  the  filter  press  and  returned  to  the  trans- 
former through  the  cover,  discharging  into  the  tank  diagonally 
opposite  the  valve  in  the  base  and  so  directing  the  discharge  that 
it  is  not  directly  over  the  coils  and  insulation.  Circulate  until 
the  oil  tests  satisfactorily. 

The  oil  level  in  transformers  should  be  kept  up  to  the  mark  on 
the  oil  gauge.  On  oil-cooled  transformers  with  external  cooling 
pipes,  the  oil  must  be  above  the  top  pipes  in  the  tanks  or  the  oil 
will  not  circulate  and  transformer  will  overheat. 

When  chloride  breathers  are  provided,  only  anhydrous  chlor- 
ide of  calcium  in  half-inch  lumps  or  larger  should  be  used.  The 
frequency  with  which  new  chloride  may  be  added  will  depend  on 
the  changes  in  temperature  and  the  humidity  of  the  atmosphere. 

Oil-cooled  transformers,  occasionally,  are  operated  under  con- 
ditions of  poor  ventilation,  overload,  or  over-voltage.  Any  of 
these  conditions,  or  a  combination  of  them  may  raise  the  tem- 
perature of  the  oil  abnormally  high,  causing  the  oil  to  throw  down  a 
deposit  which  forms  on  the  transformer  surfaces.  Should  the 
deposit  on  any  surface,  except  the  base,  reach  an  average  thick- 
ness of  about  J  inch,  the  oil  should  be  renewed  as  soon  as  possible. 
Before  putting  new  oil  into  the  tank  the  sediment  should  be 
removed  from  aU  surfaces  and  the  windings  cleaned  by  forcing 
dry,  clean  Transil  oil  through  all  ducts  and  against  all  surfaces 
until  all  deposit  is  removed. 


Digitized  by 


Google 


TRANSFORMERS  455 

Temperatures  should  be  read  daily  (or  more  often),  and  if  an 
oil  temperature  of  80°  or  over  for  the  self-cooled  is  indicated,  or  65° 
or  over  for  the  water-cooled,  the  transformer  must  be  cut  out  of 
service  at  once  and  the  cause  of  the  excessive  heating  looked  into. 
These  or  higher  temperatures  of  oil  may  indicate  that  the  interior 
temperature  of  the  windings  were  exceeding  the  safe  hottest  spot 
value,  this  being  limited  to  105°  C.  for  self-cooled  and  90°  C.  for 
water-cooled  transformers  as  previously  stated. 

Regardless  of  oil  temperatinre  as  indicated  by  thermometers, 
transformers  should  not  be  operated  at  overloads  not  stipulated 
by  the  specifications.  When  operatmg  water-cooled  transformers 
at  an  overload  the  amount  of  water  should  be  increased  in  propor- 
tion to  the  load.  On  account  of  the  increased  amount  of  water 
during  overload,  the  temperature  of  the  oil  will  not  rise  as  fast  as 
the  temperature  of  the  windings  and  any  of  the  causes  leading 
to  excessive  heating  will  have  more  pronounced  effect  under  these 
conditions.  Therefore,  transformers  during  overload  should  be 
watched  with  especial  care  to  see  that  the  oil  temperatures  are 
kept  well  below  the  temperature  limits  specified. 

Compartments  in  which  oil-insulated  self-cooled  transformers 
are  installed  should  be  thoroughly  ventilated.  Openings  for  cool 
air  should  be  provided  at  various  points  near  the  floor,  and  outlets 
should  be  in  or  near  the  roof,  which  should  not  be  closer  than  6  to 
10  feet  from  the  top  of  the  transformer.  The  room  temperature 
in  which  transformers  are  installed  should  not  exceed  the  tem- 
perature of  the  air  entering  the  room  by  more  than  5°,  and  pre- 
smnably,  the  entering  air  will  come  from  the  outside,  or,  at  least, 
from  a  source  not  much  warmer  than  the  outside  air. 

There  is  practically  no  danger  of  condensation  of  moisture  in 
transformers  which  have  no  chloride  breathers  if  the  oil  at  all 
time  IB  kept  10°  or  more  above  the  room  temperature.  It  is  also 
desirable,  especially  in  moist  climates,  to  keep  the  oil  in  idle 
transformers  (not  equipped  with  breathers),  slightly  warm  in 
order  to  eUmiinate  the  chance  of  the  oil  becoming  moist.  This 
may  be  accomplished  by  appl3ring  voltage  alone  for  a  few  hours 
each  day  Water-cooled  transformers  should  be  watched  to  see, 
that  the  oil  temperature  does  not  drop  below  the  limits  specified; 
and  if  it  does,  the  amount  of  water  must  be  decreased  until  the 
oil  attains  a  temperature  of  at  least  10°  above  the  surround- 
ing air. 


Digitized  by 


Google 


456 


ELECTRICAL  EQUIPMENT 


Oil-supply  System.  Many  different  schemes  are  used  in 
laying  out  the  oil-supply  system.  The  piping  should,  however, 
always  be  arranged  so  that  the  transformers  may  be  readily  and 
quickly  drained  for  inspection  and  in  case  of  emergencies.  This 
draining  also  refers  to  the  piping  itself.  Storage  tanks  should  be 
provided  for  both  filtered  and  unfiltered  oil,  and  these  are  gener- 
ally located  in  the  basement.  Sometimes  they  are  installed  in 
compartments  and  occasionally  the  tanks  are  further  imbedded  in 
sand  as  an  additional  fire  protection. 

A  flexible  oil-piping  system  for  a  transformer  installation  is 
shown  in  the  diagram  (Fig.  285).    This  system  will  allow  the  oil 


To  Sewer 


I      I 


To  oil  Swltcb  Tanks* 


Z  *Note;-    Iq  very  lar^e  InstaUatloiM 
^  where  oil  switctaes  are  not 

7  located  near  tniDsforinera,  a 

^  delivery  header  for  oil 

^  switches  Is  rocommeoded 

2 


& 


DralD  and  Storage 
Taoks  below  Floor 
Level  of  Traas- 
formers 


± 


Oil  Filter  and  Pamp 
=^  Located  on  Floor  Level  of  Tanks 


Storoffe  Tank  for 
OU  Switches 


Fia.  285. — Diagram  Showing  Method  of  Arranging  Transformer  Oil  Piping. 

to  be  circulated  from  any  transformer  to  either  tank;  from  one 
tank  to  the  other,  either  directly  or  through  the  filter  press;  and 
finally,  from  either  tank  to  the  transformers,  either  directly  or 
through  the  filter  press.    A  connection  to  the  sewer  or  tailrace 


Digitized  by 


Google 


TRANSFORMERS  457 

should  also  be  provided  for  draining  ofif  the  oil  in  case  of  emer- 
gency. The  movement  of  the  oil  may  be  accomplished  either  by 
applying  compressed  air  to  the  tanks  or  by  means  of  the  motor- 
driven  pump  of  the  filter  press  or  other  separate  pumps. 

Occasionally  an  intermediate  oil  tank  is  provided  and  installed 
on  the  main  floor  or  gallery  at  an  elevation  that  the  oil  can  be 
drawn  into  any  of  the  transformers  by  gravity.  The  oil  is  then 
pumped  from  the  storage  tanks  in  the  basement  after  being  fil- 
tered. A  motor-driven  air  compressor  and  vacuimi  pump  may 
also  be  required,  being  operated  as  a  vacuxmi  pump  for  exhausting 
the  air  from  the  transformer  cases  so  that  the  oil  may  be  drawn 
into  the  same,  or  as  a  compressor  for  pumping  in  air  in  the  inter- 
mediate storage  tank  to  assist  gravity  in  emptying  the  same. 

Cooling  Water  System.  The  design  of  the  cooling  water 
system  depends  on  the  nature  of  the  development,  i.e.,  whether 
low-head  or  high-head,  and  also  on  whether  a  suflScient  continuous 
water  supply  can  be  obtained.  This  is  not  the  case  in  many  sub- 
stations and  imder  such  conditions  it  becomes  necessary  to  pro- 
vide cooUng  ponds  and  reservoirs.  The  water  from  the  pond  is 
pumped  to  the  transformers  and  after  passing  through  the  coohng 
coils  it  is  returned  to  the  pond,  where  it  is  cooled.  This  may  be 
eflfected  either  by  a  spray  or  by  providing  a  basin  of  such  dimen- 
sions that  a  suflScient  cooling  is  obtained  by  a  radiation  of  the  heat 
from  the  water  to  the  air.  The  latter  method  is  much  superior 
to  the  former  in  which  air  is  liable  to  be  carried  along  with  the 
water,  causing  a  rapid  oxidation  of  the  iron  cooling  coils. 

For  the  generating  station  transformers  it  is  customary  to 
take  the  cooUng  water  from  the  forebay  or  from  the  penstocks. 
In  the  former  case  it  may  be  necessary  to  provide  pumps  for  con- 
vejring  the  same  through  the  cooling  coils.  For  high-head  develop- 
ments where  the  pressure  may  be  too  high  for  the  cooling  coils, 
a  reducing  valve  must  be  installed,  but  this  is,  as  a  rule,  not  neces- 
sary in  low-head  plants  or  with  iron  cooling  coils  which  can  with- 
stand a  much  higher  pressure  than  copper  coils. 

The  water  should  be  taken  from  at  least  two  separate  intakes, 
and  it  is  needless  to  say  that  it  must  be  free  from  silt  and  sus- 
pended particles.  For  this  reason  strainers  should  be. provided 
before  it  enters  the  distributing  headers,  and  these  strainers  should 
be  so  arranged  that  they  can  be  readily  removed  and  cleaned. 


Digitized  by 


Google 


458  ELECTRICAL  EQUIPMENT 


7.  CURRENT-LXMITING  REACTORS 

Purpose  of  Reactors.  Modern  generating  and  transmission 
systems  have  reached  such  magnitudes  as  to  make  it  necessary  to 
very  carefully  analyze  the  abnormal  conditions,  which  may  take 
place  during  short  circuits  on  the  system,  with  a  view  of  pro- 
viding such  means  as  may  be  required  for  protection  not  only  of 
the  apparatus  involved,  but  also  the  service  as  a  whole.  This  is 
the  function  of  a  reactor  by  means  of  which  the  flow  of  current  on  a 
short  circuit  may  be  limited  to  a  safe  value.  It  accomplishes  this 
purpose  by  reason  of  the  voltage  drop  or  back  pressure  which  it 
exerts  in  the  circuit. 

By  means  of  the  proper  installation  of  reactors  the  whole 
station,  or  even  several  stations,  may  be  operated  in  multiple  while 
at  the  same  time  the  several  sections  may  be  protected  from  each 
other  and  each  section  from  the  individual  circuits  which  it  feeds. 
Troubles  may  be  localized  or  isolated  practically  where  they  orig- 
inate without  communicating  their  disturbing  effects. 

When  a  short-circuit  occurs  on  a  system  the  voltage  will  drop, 
depending  on  the  magnitude  of  the  short  circuit  and  the  inherent 
characteristics  of  the  generators,  i.e.,  their  impedance.  A  severe 
short-circuit,  such  as  may  occur  when  there  are  no  reactors,  will 
cause  the  voltage  to  drop  to  a  low  value  in  a  few  cycles,  whereas 
on  a  less  severe  short-circuit,  the  time  taken  for  the  voltage  to 
drop  to  the  same  low  value  will  be  longer.  Synchronous  apparatus 
will  stand  a  complete  loss  of  power  for  a  few  cycles  only,  but  will 
stand  a 'reduction  of  voltage  for  a  longer  period.  It  is  important 
then  that  the  value  of  short-circuit  be  small  and  that  it  be  cleared 
in  the  shortest  possible  time.  Introducing  reactors  will  limit  the 
maximum  value  of  the  current,  and  with  the  latest  type  of  relays, 
the  time  required  for  selective  switch  action  is  very  short,  so  that 
a  trouble  can  be  locaUzed  and  cleared  before  the  apparatus  on  the 
rest  of  the  system  is  affected. 

The  protective  and  localizing  functions  of  a  reactor  are,  how- 
ever, quite  distinct.  The  former,  since  all  the  evil  effects  of  heavy 
current — excessive  mechanical  stresses,  heating,  etc.,  are  pro- 
portional to  the  square  of  the  current,  is  measured  in  terms  involv- 
ing the  square  of  the  total  reactance,  while  the  latter  is  measured 
in  terms  of  the  first  power  of  the  reactance  involved. 

The  chief  purpose  of  a  reactor  is,  therefore,  to  limit  the  flow  of 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  459 

current  into  a  short  circuit  with  a  view  to  protect  the  apparatus 
from  overheating  as  well  as  failure  from  destructive  mechanical 
forces;  also  protecting  the  system  as  a  whole  against  shut-down 
by  maintaining  the  voltage  on  part  of  the  system  while  the  short 
circuit  is  being  cleared. 

Rating.  Reactors  are  generally  spoken  of  as  introducing  a 
certain  per  cent  reactance  in  a  circuit.  This  is  the  ratio  of  the 
voltage  drop  across  the  reactor  (when  the  rated  current  of  the  cir- 
cuit at  rated  frequency  is  flowing  through  the  reactor),  to  the 
voltage  between  line  and  neutral  on  three-phase  circuits,  or  the 
voltage  between  the  lines  on  single-phase  circuits.  The  reactance 
is,  therefore,  expressed  as  being  single-phase  in  either  case. 

The  kilovolt-ampere  (Kv.A.)  rating  of  the  reactor  is  the  product 
of  the  voltage  drop  across  the  reactor  and  the  rated  current.  For 
generator,  transformer  and  feeder  reactors  the  rated  current  is 
usually  taken  as  equal  to  the  current-carrying  capacity  of  the 
apparatus,  while,  for  bus  sectionalizing  reactances,  it  is  determined 
by  the  power  which  must  be  transferred  over  the  reactor.  This  is 
very  often  chosen  so  as  to  correspond  to  the  capacity  of  one  of  the 
generators. 

Current-limiting  reactors  should  furthermore  be  designed  for 
the  maximum  load  current  they  will  have  to  carry.  Being  self- 
cooled  and  having  neither  iron  nor  oil  to  provide  thermal  storage 
they  reach  their  maximmn  temperature  very  quickly.  Therefore, 
in  cases  where  the  apparatus  or  circuits  must  carry  overloads  for 
two  hours  or  more,  this  overload  current  should  be  considered  the 
rated  current  of  the  reactor,  and  the  capacity  should  be  selected 
on  this  basis.  Under  this  assumption,  a  temperature  rise  of  86**  C. 
represents  common  practice,  the  rise  being  based  on  an  ambient 
room  temperatm^  of  40®  C. 

As  reactors,  as  a  rule,  do  not  have  an  iron  core  to  become 
magnetically  saturated,  the  reactive  drop  will  be  proportional  to 
the  current.  That  is,  if  a  circuit  having  a  5  per  cent  reactor  were 
to  be  short-circuited  at  the  reactor  terminal  on  the  load  side  and 
having  full  sustained  voltage  on  the  supply  side,  the  sustained 
current  would  be  limited  to  100-^5  or  twenty  times  normal.  It 
should  be  remembered  that  transformers  and  generators  in  cir- 
cuit with  the  reactor  also  have  definite  values  of  reactance  which, 
when  expressed  in  terms  of  the  current  of  the  circuit  (per  cent 
reactive  drop  with  normal  current  flowing)  may  be  added  directly 


Digitized  by 


Google 


460  ELECTRICAL  EQUIPMENT 

to  the  reactance  of  the  reactor  to  determine  the  total  apparatus 

reactance  of  the  circuit.    This  total  reactance,  plus  the  reactance 

of  the  line  up  to  the  point  of  short-circuit  divided  into  100,  gives 

the  approximate  short-circuit  current  (the  result  being  expressed 

in  number  of  times  normal). 

Care  must  be  exercised  in  calculating  the  possible  short-circuit 

current  of  a  system  that  the  various  per  cent  reactances  are  on  the 

same  basis,  i.e.,  on  the  same  current  value.    For  example,  if  the 

reactance  for  a  6000  Kv.A.,  three-phase  transformer  is  given  as 

6  per  cent  but  a  value  is  required  which  corresponds  to  one  of  the 

generators,  having  a  capacity  of,  say,  4000  Kv.A.,  three-phase, 

4000 
the  corresponding  value  would  then  be  ^7^7^X6=4  per  cent. 

Similarly,  it  must  also  be  remembered  that  reactance  values 
given  for  single-phase  transformers  really  refer  to  a  bank  of  three 
such  transformers.  For  example,  the  reactance  of  a  6000  Kv.A., 
single-phase  transformer  is  given  as  3  per  cent.  This,  then,  usually 
refers  to  the  full-load  current  from  a  bank  of  three  such  units,  i.e., 
18,000  Kv.A.,  so  that  if  the  reactance  were  to  be  converted  to  the 
basis  of  a  6000  Kv.A.  generator,  its  corresponding  value  would  be 

— -— X3  =  1  per  cent.    A  careful  consideration  of  the  above  is  of 

the  greatest  importance  when  reactance  values  for  generators, 
transformers  and  transmission  lines  of  dififerent  capacities  are  to 
be  combined. 

For  the  designation  of  the  rating  of  a  current  limiting  reactor 
the  following  method  is  generaUy  used: 

"  Type Frequency Kv.A Volts  Drop 

Amperes Reactor  to  give per  cent  reactive  drop 

in Kv.A volt phase  circuit." 

The  type  s3mabols  generally  used  are  CLS,  CLQ  and  CLT. 
The  meaning  of  the  symbols  is  as  follows: 
C.L. — Current-limiting  reactor. 
8. — Single-phase  (may  apply  to  any  one  reactor  of  a 
group  of  two  or  three  for  use  in  two-  or  three- 
phase  circuits). 
Q. — ^Two  phase   (two  single-phase  reactors  mounted 

together). 
T. — ^Three-phase  (three  single-phase  reactors  mounted 
together). 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  461 

ft 

Far  Example:  A  6  per  cent  reactor  in  a  60-cycle,  6600-volt, 
100-amp.,  single-phase  circuit,  means  that  the  reactor  will  have  a 
drop  of  5  per  cent,  or  330  volts,  when  the  rated  current  is  flowing. 

The  rating  of  the  reactor  will  be  as  follows: 

C.L.S.— 60  (cycles),  33  (Kv.A.),  330  (volt  drop),  100  (amperes) 
reactor — to  give  5  per  cent  reactive  drop  in  a  660 
Kv.A.,  6600  volt  single-phase  circuit. 

In  the  case  of  three-phase  circuits  the  percentage  drop  is 
always  based  on  the  voltage  between  hne  and  neutral. 

For  Example:    A  6  per  cent  reactor  in  a  three-phase  60-cycle, 

660Q-volt,  100-amp.  circuit  means  that  each  reactor  (of  the  three) 

6600 
will  have  a  drop  of  — 7^X0.05  =  191  volts  when  normal  current  is 

V3 

flowing. 

The  rating  will  then  be  as  follows: 

C.L.S.—60  (cycles),    19.1    (Kv.A.),    191    (volts    drop),    100 

(amperes)   reactor — ^to  give   5  per   cent  reactive 

drop  in  1145  Kv.A.,  6600-volt  three-phase  circuit. 

Rating  as  Affected  by  Frequency.    A  reactor  designed  for  a 

given  frequency  may  be  used  in  a  circuit  of  different  frequency,  in 

which  case  the  per  cent  reactance  is  approximately  equal  to  the 

ratio  of  the  frequency  for  which  it  is  to  be  used  to  the  frequency 

for  which  it  is  designed  times  the  per  cent  reactance  for  which 

it  is  designed. 

For  Example:  A  3J  per  cent  25-cycle  reactor  may  be  used  in  a 

40-cycle  circuit,  in  which  case  the  per  cent  reactance  is  approxi- 

40 
mately  7rrX3i  =  5.6  per  cent. 
25 

Rating  as  Affected  by  Voltage.  A  standard  reactor  can  be 
used  for  lower  voltage  circuits  than  those  for  which  it  is  designed, 
in  which  case  the  per  cent  reactance  is  increased  in  the  ratio  of  the 
voltage  for  which  it  is  designed  to  that  for  which  it  is  to  be  used. 

For  Example:  On  an  11,000-volt  three-phase  circuit  requiring 
the  introduction  of  about  3|  per  cent  reactance,  it  will  be  possible 
to  use  a  13,200  volt  3^  per  cent  reactor.  The  reactance  will  be 
^.^13,200     .^  , 

Rating  as  Affected  by  Current.  A  standard  reactor  may  be 
used  for  lower  currents  than  that  for  which  it  is  designed,  in 
which  case  the  per  cent  reactance  decreases  with  the  ratio  of  the 


Digitized  by 


Google 


462 


ELECTRICAL  EQUIPMENT 


current  for  which  it  is  to  be  used  to  the  current  for  which  it  is 

designed. 

For  Example:  A  3J-per  cent  350-amp.  reactor  may  be  used  in  a 

^  300 

300-amp.  circuit  where  it  will  insert  3^  X^t^t,  =  3  per  cent  reactance. 

350 

From  the  foregoing  it  is  seen  that  a  3J  per  cent,  25-cycle, 

13,200- volt,  360-amp.  reactor  will  introdifce  in  a  40-cycle,  11,000- 

40 
volt,    300-amp.    circuit   a   reactance   of   approximately  3JX— 

^13.200^300    __. 

Effect  of  Reactance  on  Power-factor.  Increasing  the  reac- 
tance in  the  system  results  but  in  a  slightly  lower  power-factor, 
the  curve  in  Fig.  286  showing  the  variation  of  power-factor  with 
per  cent  reactance.  It  is  to  be  noted  that  if  the  power-factor  of 
the  circuit  were  90  per  cent,  corresponding  to  a  reactance  of  44  per 
cent,  then  the  introduction  of  a  3J  per  cent  reactor  would  increase 


iUU 

E^ 

■^ 

" 

90 

^ 

^ 

'Sfc,, 

V, 

\^ 

80 

X 

\ 

Ito 

£fi0 

\ 

X 

\ 

s 

\ 

s 

\ 

§40 

V 

\ 

^80 

\ 

20 

\ 

I 

10 

I 

1 

t\ 

1 

10      :»      30     40      50      00     70     80      90     100 
Per  Cent  Reactance  of  Circuit 

Fig.  286. 


the  reactance  of  47§  per  cent  and  the  power-factor  would  be  low- 
ered to  88  per  cent.  The  introduction  of  a  slightly  larger  reactor, 
say  4.2  per  cent,  would  decrease  the  power-factor  to  practically 
the  same  amount.  On  the  other  hand,  if  the  power-factor  of  the 
circuit  were  70  per  cent,  the  introduction  of  a  3^  per  cent  reactor 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS 


463 


would  reduce  the  power-factor  to  about  66  per  cent  and  a  4.2  per 
cent  reactor  to  65.5  per  cent. 

Effect  of  Reactance  on  Regulation.  As  in  the  case  of  the  power- 
factor,  an  increase  in  the  reactance  results  in  a  slightly  poorer 
regulation,  the  effect  being  more  marked  if  the  operating  power- 
factor  is  much  below  unity.  The  curves  in  Fig.  287  show  the  vari- 
ation in  regulation  with  per  cent  reactance,  and  it  will  be  noted 


Mi 


Corv.   A-P.F.-IOOJ 
"        B-P.F.-   9I; 
-        C-P.F.-   Ml 
•       D-P.F.-   W 

I  Before  Inaertiiw  RMctoace 

(             ««                         M                               U 
(            M                        U                             .< 

IT         X                .i                    n 

J 

I 

/ 

J 

/ 

J 

J 

/ 

f 

/ 

/ 

4 

r 

/ 

f 

/ 

/ 

/ 

/ 

/ 

q 

/ 

/ 

/ 

/ 

/ 

y 

/ 

/ 

/ 

^ 

9 

/ 

i 

r 

/ 

^ 

/ 

/ 

i 

f 

/ 

/' 

/ 

/ 

B 

/ 

/ 

J 

f 

/ 

/ 

f 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

^ 

r 

/ 

/ 

/ 

/ 

/ 

y 

/ 

i 

7 

/ 

/ 

^ 

y 

/ 

J 

V 

J 

V 

A 

^^ 

_ 

u- 

£ 

^ 

B 

^ 

^ 

2       8       4       5        6        7       8 
Per  Cent  Reactance  of  Circuit 

Fig.  287. 


10 


that  with  a  90  per  cent  power-factor  the  introduction  of  a  3J  per 
cent  reactor  will  increase  the  regulation  1.6  per  cent  and  a  4.2 
per  cent  reactor  1.9  per  cent.  With  a  power-factor  of  70  per  cent 
the  increase  in  the  regulation  would  be  respectively  2.5  and  3.0 
per  cent.  However,  the  amount  by  which  the  voltage  of  the 
system  is  lowered  is  not  seriously  large  and  can  readily  be  com- 
pensated for  by  increasing  the  voltage  of  the  generators. 


Digitized  by 


Google 


464 


ELECTRICAL  EQUIPMENT 


The  above  discussion  shows  that  a  reactance  somewhat  above 
that  required  for  current  Umiting  protection  does  not  materially 
affect  the  regulation  or  the  power-factor,  and  in  many  cases  it 
may,  therefore,  be  advantageous  to  use  a  somewhat  higher  reac- 
tance than  that  which  would  be  required,  and  thereby  gain  the 
advantage  of  reduction  in  cost  which  can  be  obtained  by  using 
standard  ratings. 

Losses.  The  losses  in  reactors  are  not  a  serious  matter  but 
should,  of  course,  be  taken  into  consideration  in  laying  out  the 
system.  They  are  due  to  the  PR  and  eddy-current  losses  in  the 
conductors  and  possibly  average  5  per  cent  of  the  rating  of  the 
reactor.  In  some  cases,  however,  the  losses  may  be  somewhat 
higher  and  in  others  considerably  less. 

Assume,  for  example,  a  4  per  cent  feeder  reactor  on  a  3000-Kv.A. 
feeder,  the  three  coils  would  have  a  combined  capacity  of  120 

Kv.A.  or  40  Kv.A.  per  coil.  The 
losses  at  5  per  cent  would  equal 
about  2000  watts  per  coil  or  6  kilo- 
watts on  the  3000  Kv.A.  feeder; 
that  is  to  say,  one-fifth  of  one  per 
cent  at  the  maximum  load  of  the 
feeder  which  may  last  only  for  a 
comparatively  short  period  during 
the  day.  Since  the  losses  are 
nearly  all  copper  losses  which  go 
•down  as  the  square  of  the  current, 
at  one-half  load,  the  losses  would 
only  be  one-fourth  of  the  above. 
Bus  reactors,  on  the  other 
hand,  carry  normally  very  Uttle,  if  any,  current  and  the  losses 
under  normal  operations  are,  therefore,  negUgible. 

Inductance.  The  inductance  of  ciurent  limiting  reactors 
may  be  calculated  with  sufficient  accuracy  by  the  following  fo^ 
mula  by  Prof.  Morgan  Brooks: 


Fig.  288.— Reactance  CoU. 


L  = 


(27rriV)2 
6+1.5/+r 


Xf'Xf"X  10-^  henrys, 


m  which  (see  Fig.  288), 

r  =  mean  radius  of  coil  in  centimeters; 
6  =  axial  length  of  coil  in  centimeters; 
f  =  thickness  of  winding  in  centimeters. 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS 


465 


Both  b  and  t  include  the  thickness  of  insulation  or,  if  the  turns  are 
air  insulated,  are  equal  to  the  pitch  of  the  winding  times  the  num- 
ber of  turns.  If  there  is  only  one  turn,  the  values  are  equal  to  the 
diameter  of  the  wire. 

iV  =  total  number  of  turns  in  coil; 

F'  and  F"  are  correction  factors  depending  on  the  coil  shape; 


F'  = 


10&+13<+2r 
106+10.7<+L4r' 


F"=0.61og,o(l00+fg'). 


The  reactance,  X,  is  equal  to  2ir/L  ohms. 

Location.  Reactors  may  be  located  in  the  system  in  such  a 
way  that  they  will  not  only  reduce  the  mechanical  strains  due  to 
short  circuit,  but  will  also  practically  localize  its  eflfect  to  the  cir- 
cuit or  section  where  it  occurs.  They  may  thus  be  placed  in  the 
generator  leads,  between  the  bus-sections,  in  the  low-tension  trans- 
former leads  or  in  outgoing  low-tension  feeders.  Which  one  of  the 
above  locations  or  combinations  thereof  is  preferable  depends  upon 
a  number  of  conditions,  each  location  having  its  advantages  and 
disadvantages. 

Generator  Reactors.  With  reactors  in  the  generator  leads 
(Fig.  289)  the  current  flowing  in  the  armature  winding  of  the 
generator  is  limited,  and  this 


m^ 


M^ 


method,  therefore,  gives  protec- 
tion to  the  generator  itself.  It 
necessarily  also  limits  the  cur- 
rent that  can  flow  into  any 
short-circuit  beyond  the  react- 
ors, inasmuch  as  the  amount  of 
current  which  can  flow  is  limited 
to  what  the  generators  can  sup- 
ply. An  objection  to  generator 
reactors  is  the  fact  that  a  short-circuit  on  or  near  the  busbars 
will  cause  a  voltage  drop  on  all  the  lines  or  feeders  connected 
thereto.  If  the  short  is  severe,  the  voltage  may  drop  to  zero 
and  this,  of  course,  will  cause  all  the  synchronous  apparatus  con- 
nected to  the  system  to  drop  out  of  step.    It  is,  therefore,  evi- 


FiG.  289.— Generator  Reactors. 


Digitized  by 


Google 


466  ELECTRICAL  EQUIPMENT 

dent  that  reactors  in  the  generator  leads  offer  no  protection  to 
troubles  of  this  nature. 

In  hydro-electric  power  systems  with  slow-  or  medium-speed 
multi-polar  generators,  the  inherent  reactance  of  these  is,  as  a  rule, 
sufficiently  high  and  the  construction  such  that  the  machines  can 
safely  withstand  momentary  short-circuits,  and  generator  reactors 
are  very  seldom  used  in  hydro-electric  plants.  If  such  reactors 
are  used,  they  should  be  placed  in  the  line  leads  as  close  to  the 
generator  as  possible  and  not  in  the  neutral. 

Bus  Reactors,  These  are  very  extensively  used  in  hydro- 
electric stations  and  permit  of  an  unlimited  extension  of  the  sys- 
tem. The  bus-bars  are  divided 
[  into  sections  by  reactors  (Fig. 

A  1a  aAa  290),  and  trouble  may  thereby 


-^000^ 


6     6     6     0 


be  confined  to  the  particular  sec- 
tion on  which  the  short-circuit 
takes  place,  while  under  noraial 
operation  a  free  exchange  of  cur- 
rent may  take  place,  thereby 
retaining  the  advantage  of  par- 
FiQ.  290.— Bus  Reactor.  allel  operation.     A  shorlHiircuit 

then  can  seriously  involve  one 
bus-bar  section  only,  and  the  destructive  power  of  a  short-circuit 
is  limited  to  the  generating  capacity  of  that  one  section  plus  the 
limited  power  which  can  flow  from  the  two  adjoining  sections. 

The  voltage  of  the  section  upon  which  the  short-circuit  takes 
place  falls  to  zero  and  the  reactors  connecting  the  two  adjacent 
sections  each  thus  consume  the  total  voltage  during  the  transfer  of 
the  short-circuit  current.  Strictly  speaking,  the  transfer  does  not, 
however,  take  place  by  a  drop  of  voltage  between  the  sections, 
but  by  a  phase  displacement  between  the  voltages  of  the  bus-bar 
sections,  as  explained  later. 

Bus  reactors  aflford,  of  course,  no  protection  to  the  generators 
connected  to  the  section  on  which  the  trouble  occurs,  but  they 
give  added  protection  to  the  generators  on  the  other  sections. 

Transformer  and  Feeder  Reactors.  With  modem  high-voltage 
transmission  systems  where  the  transformers  are  connected  on  the 
unit  principle  so  as  to  form  a  part  of  the  transmission  line,  reactors 
in  the  low-tension  transformer  leads  (Fig.  291)  may  be  of  consid- 
erable value  for  protecting  against  short-circuits  in  the  lines, 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS 


467 


where  they,  of  course,  mostly  take  place.  Modem  transformers 
are,  however,  generally  built  with  a  comparatively  high  inherent 
reactance,  so  that  they  can  safely  withstand  short-circuits,  and 
reactors  are,  therefore,  very  seldom  installed  in  this  manner. 

Reactors  in  low-tension  feeders  (Fig.  292)  are,  however,  very 
common  and  have  many  advantages.  The  probability  of  a 
short-circuit  in  a  feeder  is  far  greater  than  in  any  other  part  of 
the  system,  and  the  short-circuit  current  through  a  feeder  switch 
may  be  considerable,  since  the  current  from  all  the  generators  will 
pass  through  the  same  and  possibly  also  the  current  from  other 


m^ 


OOOOOOOO 


Fig.  291. — ^Transfonner  Reactors. 


Fig.  292.— Feeder  Reactors. 


synchronous  machines  on  the  system.  By  means  of  feeder 
reactors,  however,  such  troubles  may  be  still  more  limited  than  if 
bus  reactors  were  provided,  and  it  is  merely  a  question  of  cost 
whether  such  reactors  can  be  afforded. 

Feeder  reactors,  of  course,  only  give  protection  for  those  short- 
circuits  which  occur  on  the  feeders  beyond  the  point  where  they 
are  installed,  and  do  not  give  protection  to  short-circuits  which 
occur  on  the  busbars  or  in  the  generators,  transformers  or  their 
connections. 

Stdt  System.  This  scheme  (Fig.  293)  was  proposed  by  tne 
late  Mr.  H.  G.  Stott,  of  the  Interborough  Rapid  Transit  Company 
of  New  York,  and  is  now  being  quite  extensively  used  in  connec- 
tion with  large  steam  turbine-driven  central  stations.  The 
feeders  are  grouped  and  fed  from  diflferent  bus  sections  which  are 
individually  energized  by  generators  deUvering  current  through  5 
per  cent  reactors.  The  bus  sections  are  normally  operated  sep- 
arately but  may  be  instantly  connected  by  tie  switches.    To  per- 


Digitized  by 


Google 


468 


ELECTRICAL  EQUIPMENT 


mit  this  emergency  connection,  each  generator  in  operation  is 
permanently  connected  to  a  common  S3aichronizmg  bus  through 
2  per  cent  reactors  which  keep  the  generators  in  step  and  also 
serve  the  purpose  of  bus-tie  reactors.  When  this  scheme  is  em- 
ployed with  a  bus  divided  into  several  sections  the  voltage  regular 
tion  is  much  better  when  there  is  current  exchange  than  when 


Feeders 


~^- 


^- 


>5  Per  cent 


oS  Per  cenl 


I  Generator  I— <  i         (  Generator  j__4 1  (  Generator  ]       , 


>  2  Per  cent 


>a  Per  cent 


^yoclironlxidff  Bob 
Fig.  203. — Stott  System  of  Reactor  Arrangement. 

ordinary  bus-tie  reactors  are  used.  This  is  obvious  from  the  fact 
that  to  get  the  same  protection  as  here  obtained,  5  per  cent  bus- 
tie  reactors  would  have  to  be  used  and  the  energy  exchanged 
between  two  non-adjacent  sections  would  suffer  a  large  voltage 
drop.  If  it  is  not  considered  necessary  to  protect  the  generators 
themselves  against  current  surges,  the  5  per  cent  reactors  may  be 
omitted. 

Number  of  Reactors.  The  following  is  considered  the  best 
practice  for  locating  reactors  in  various  circuits: 

(a)  For  single-phase  circuits  a  single  reactor  in  one  side  of  the 
line. 

(6)  For  tw(Hphase,  four-wire  circuits  two  reactors,  one  in  one 
side  of  the  line  of  each  phase. 

(c)  For  two-phase,  three-wire  circuits  one  reactor  in  each  of  the 
outside  lines  (as  distinguished  from  the  neutral  or  conunon  wire). 

(d)  For  three-phase  circuits  one  reactor  in  each  line. 

Size  of  Reactor.  The  selection  of  proper  reactors  for  a  system 
requires,  first  of  all,  a  complete  investigation  of  the  possible  sboit 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  469 

circuit  currents  which  are  liable  to  be  set  up  due  to  faults  in  the 
various  parts  of  the  system.  When  a  short-circuit  occurs,  the 
maximum  short-circuit  current  is  hmited  by  the  total  eflfective 
impedance  at  that  instant  in  the  generators,  transformers,  and 
transmission  lines  to  the  fault  in  question.  This  value  is,  how- 
ever, not  constant,  but  decreases  rapidly  until  a  value  limited  by 
the  synchronous  impedance  of  the  generators  is  reached  (see 
"  Synchronous  Generators,"  page  292).  A  sharp  distinction  must, 
therefore,  be  made  between  an  instantaneous  and  a  sustained 
short-circuit,  the  former  being  dependent  upon  the  instantaneous 
effective  impedance  of  the  system  and  the  latter  on  the  sustained 
effective  impedance.  Except  for  long  transmission  and  distribution 
lines,  the  resistance  is,  as  a  rule,  of  such  small  value  compared  to  the 
reactance,  that  for  all  practical  purposes  it  may  be  neglected  and 
the  calculations  based  on  reactance  only  instead  of  impedance. 

As  previously  stated,  a  severe  short-circuit  may  result  in  a 
mechanical  destruction  of  the  apparatus  or  an  overheating  of  the 
same.  The  former  is,  of  course,  chiefly  due  to  the  instantaneous 
current  rush,  while  the  sustained  short-circuit  current  ordinarily 
determines  the  thermal  effect. 

The  instantaneous  short-circuit  current  is  readily  calculated, 

being  equal  to  the  normal  current  multiplied  by  100  and  divided 

by  the  total  reactance  to  the  fault,  expressed  in  per  cent.    For 

modem   water-wheel-driven   generators   the   inherent   reactance 

varies  from  15  to  25  per  cent  and  for  transformers  from  6  to  10 

per  cent.    As  expressed  in  per  cent  it  may  be  obtained  from  the 

formula: 

_ZXKv.A, 

^       10X£2  • 

where  p= reactance  in  per  cent; 

X= single-phase  reactance  in  ohms; 
E= voltage  between  phases  in  kilovolts. 

The  reactance  in  ohms  per  mile  of  one  wire  of  a  sjrmmetrical 
three-phase  circuit  is 

Z=2ir/L=27r/r^.74  logio|+.0805)l0-3l, 

in  which  a = spacing  between  centers  of  conductors  in  inches; 
r= radius  of  conductors  in  inches. 

In  considering  the  amount  of  current  that  will  feed  into  a  short 


Digitized  by 


Google 


470  ELECTRICAL  EQUIPMENT 

circuit,  the  83aichronous  apparatus  connected  to  the  system  in  the 
form  of  load  must,  of  course,  also  be  taken  into  account,  as  on  a 
short-circuit  there  is  a  tendency  for  them  to  feed  back  into  the 
system,  due  to  the  inertia  of  their  rotating  elements.  It  is,  of 
course,  also  evident  that  strictly  "  spare  "  equipments  need  not 
be  included  in  the  calculations. 

In  dealing  with  the  effects  of  short-circuits  we  must  consider 
the  damages  which  they  may  cause  to  generators,  transfonners, 
circuit  breakers,  cables  or  bus-bars  and  against  which  protection 
must  be  provided  in  the  form  of  reactors  for  limiting  the  excessive 
currents  to  values  which  may  be  safely  withstood  by  the  apparatus. 

Generators  and  transformers  are,  as  previously  stated,  now 
designed  with  such  mechanical  rigidity  that  they  can  safely  with- 
stand the  mechanical  forces  arising  from  dead  short-circuits  across 
their  own  terminals. 

As  far  as  oil  circuit  breakers  are  concerned,  the  problem  is 
much  more  difficult  and  their  rupturing  capacity  is,  as  a  rule,  the 
limiting  feature  in  determining  the  value  of  the  permissible  short- 
circuit  current.  The  power  which  has  to  be  broken  on  a  short 
circuit  depends  naturally  on  how  quickly  the  circuit  breaker  opens 
and  also  on  the  rate  at  which  the  short-circuit  current  dies  down. 
Due  to  inertia,  it  is,  of  course,  impossible  for  a  breaker  to  open 
instantaneously  and  consequently  no  breaker  is  ever  called  on  to 
open  the  momentary  short-circuit  current  that  occurs  during  the 
first  few  cycles,  but  it  has  to  be  strong  enough  mechanically  to 
resist  the  magnetic  stresses  set  up  dining  such  a  short-circuit. 
Large  capacity  breakers  equipped  with  "  instantaneous  "  acting 
relays  can  be  made  to  open  in  about  one-quarter  of  a  second  and 
if  the  short-circuit  occurs  close  to  the  generating  station  the 
power  which  has  to  be  broken  averages  approximately  60  per 
cent  of  the  maximum  instantaneous  value.  If  the  trouble  should 
occur  at  a  considerable  distance  from  the  power-house,  the  rate 
at  which  the  short-circuit  current  dies  down  would  be  much 
slower,  so  that  the  power  which  would  have  to  be  broken  might 
be  nearly  equal  to  the  instantaneous  value,  but  due  to  the  addi- 
tional reactance  of  the  Une  this  value  will,  as  a  rule,  be  less  than 
the  above,  which,  therefore,  should  be  used  in  governing  the  cur- 
rent which  must  be  broken  under  the  worst  conditions.  For  non- 
automatic  switches  or  switches  equipped  with  definite  time  limit 
relays  with  a  setting  over  0.8  second,  the  rupturing  capacity 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  471 

corresponds  to  the  sustained  short-K^ircuit  current,  while,  for 
switches  with  inverse  time  action,  the  condition  approximating 
''instantaneous/'  as  above,  must  be  assmned.  The  maximum 
instantaneous  value  means  the  root-mean-square  value  'of  a  S3an- 
metrical  wave.  Similarly  for  the  rupturing  capacity  of  oil  circuit 
breakers,  as  tests  have  shown  that  the  wave  becomes  practically 
symmetrical  in  the  minimum  time  in  which  a  breaker  can  open. 

There  is  a  great  variety  of  oil-circuit  breakers  in  the  market 
with  rupturing  capacities  of  several  hundred  thousand  Kv.A. 
As  a  rule,  switches  with  the  higher  rating  will  be  required  near 
the  generating  station,  while  under  some  conditions  smaller 
switches  may  be  used,  for  instance,  in  substations,  where  the 
added  reactance  of  transfonners  and  lines  serve  to  reduce  the 
value  of  the  short-circuit  current. 

The  mechanical  forces  acting  between  the  conductors  of  a 

three-phase  cable  may  be  obtained  from  the  following  formula. 

It  is  assumed  that  all  three  conductors  are  equally  spaced  and 

simultaneously  short-circuited,  the  r.m.s.  current  being  equal  in 

each  phase.    Then  the  force,  Fo,  tending  to  repel  any  conductor  in 

a  direction  at  right  angles  to  a  plane  passing  through  the  other 

two  is: 

-,     4.67XPX10-7 

/Tp- pounds  per  foot,* 

where  J = r.m.s.  value  of  sine  wave  = 


V2    ' 
a = Distance  between  conductors  in  inches. 

Thus,  in  a  paper  insulated,  lead-covered  cable,  the  force  is  exerted 
on  the  over-all  wrapping  around  all  three  conductors  and  also  on 
the  lead  sheath,  and  the  tensile  strength  of  the  paper  and  lead 
must  be  sufficient  to  withstand  the  stress  thus  placed  upon  them. 
On  bus-bars  this  force  tends  to  throw  the  bars  away  from  the 
center  of  the  equilateral  triangle  of  which  each  bus  is  assimied  to 
form  one  apex,  and  produces  a  tension  or  compression  on  the  bus- 
bar clamps,  depending  on  the  location  of  the  insulators.  The 
bus-bars,  due  to  their  spacing  being  inherently  greater  than  the 
conductors  of  a  cable,  are  subject  to  a  much  lower  disruptive 
force  per  unit  length,  but,  on  the  other  hand,  since  they  are  sup- 

1  Gross,  A.IJS.E.,  Jan.,  1915. 


Digitized  by 


Google 


472  ELECTRICAL  EQUIPMENT 

ported  at  only  frequent  intervals  rather  than  continuously,  as  is  a 
cable,  the  force  on  any  support  may  become  excessive. 

The  above  refers  to  a  three-phase  short  circuit.  If,  however, 
the  short  is  between  two  of  the  conductors  instead  of  between 
all  three,  the  force  will  only  be  86.6  per  cent  of  the  three-phase 
value,  based  on  the  same  current. 

If  the  bus-bars  are  installed  in  the  same  plane,  the  force  acting 
on  the  outside  bars  is  only  86.6  per  cent  of  what  it  would  be  if  the 
bars  were  spaced  at  the  vertices  of  an  equilateral  triangle. 

The  heating  of  cables  may,  on  the  other  hand,  be  the  limiting 
feature  as  far  as  the  permissible  short-circuit  current  is  concerned, 
since  it  is  quite  possible  for  the  temperature  of  the  conductor  to 
rise  to  such  a  point  as  to  endanger  the  insulation  of  the  cable  even 
in  the  short  time  that  it  takes  an  oil  switch  to  open,  especially  if  it 
is  non-automatic  or  provided  with  a  definite  time-limit  relay. 
The  calculations  involved  in  determining  the  temperature  rise  are 
intricate  and  the  reader  is  referred  to  a  paper  by  I.  W.  Gross  in 
A.I.E.E.  Proceedings  for  January,  1915. 

In  calculating  the  short-circuit  current  let  us,  as  an  example, 
first  assume  a  system  consisting  of  four  10,000  Kv.A.  generators, 
with  10  per  cent  inherent  reactance,  operating  in  parallel  on  a 
bus.  With  a  short  circuit  in  one  of  the  step-up  transformers, 
what  would  be  the  required  instantaneous  rupturing  capacity  of 
the  low-tension  transformer  circuit  breaker? 

Since  the  four  generators  are  connected  in  i)arallel,  the  com- 
bined reactance  will  be  equal  to  -j-=2.5  per  cent  and  the  total 

10000 
short-circuit  current,  expressed  in  Kv.A.,  equal  to     ^  _    XIOQ 

2.5 

=400,000  Kv.A.  The  bus-bars  must  then  be  designed  to  with- 
stand the  mechanical  stresses  due  to  twice  this  current  on  account 
of  the  possible  unsymmetrical  nature  of  the  cmrent  wave,  while 
the  rupturing  capacity  of  the  switch  would  have  to  be  about  60 
per  cent  of  the  above,  or  240,000  Kv.A. 

As  far  as  the  generators  themselves  are  concerned,  it  has  pre- 
viously been  stated  that  those  of  modern  design  are  now  being 
designed  to  safely  withstand  short-circuits.  The  generator 
switches  under  the  worst  condition,  i.e.,  with  a  short  in  one 
of  the  generators,  would  be  called  upon  to  break  the  combined 
current  of  only  three  generators,  and  as  these  switches  as  a  rule 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  473 

are  made  non-automatic,  it  would  only  be  the  sustained  value  of 
the  current,  thus  probably  about  two  and  one-half  times  the 
normal  rating  or  75.000  Kv.A.  With  an  automatic  voltage 
regulator  holding  up  the  excitation,  this  value  would,  however, 
be  greatly  increased. 

If  the  inherent  reactance  had  been  less  than  the  above,  or  the 
capacity  of  the  generators  greater,  it  might  have  been  necessary 
to  install  external  reactors  in  the  generator  leads  to  limit  the 
short-circuit  current  which  the  switch  would  have  to  rupture,  as 
shown  in  Fig:^289.  This  is,  however,  never  done  in  hydro- 
electric stations,  and  if  such  a  condition  should  arise  the  bus  is 
generally  sectionalized  by  means  of  reactors  as  shown  in  Kg.  290 
and  as  explained  in  the  following. 

As  previously  stated,  the  purpose  of  installing  bus-bar  reactors 
is  to  limit  the  amoimt  of  current  that  can  flow  into  a  fault  in  one 
section  of  the  bus-bars,  and  so  confine  the  disturbance  to  that  part 
of  the  system  on  which  the  fault  occurs.  Bus  reactors  should 
have  a  reactance  sufficiently  high  so  that  in  case  of  a  short-circuit 
on  one  bus  section  the  voltage  of  the  adjoining  sections  is  not 
seriously  disturbed  by  the  current  flowing  from  them  over  the 
reactors  into  the  short-circuit.  On  the  other  hand,  it  is  highly 
desirable  to  operate  all  the  generators  of  the  station  in  parallel, 
and  this  necessitates  a  reactor  of  a  low  enough  reactance  to  permit 
the  interchange  current  between  the  bus  sections  to  take  carp  of 
the  required  distribution  of  the  load  along  the  bus. 

The  amoimt  of  reactance  to  be  installed  involves  a  careful 
study  of  the  layout  of  the  system.  Probably  a  value  allowing  a 
transfer  of  power  equal  to  the  capacity  of  one  generator  (one-half 
from  each  adjacent  section),  may  be  considered  sufficient.  If 
then  each  generator  had  a  short-circuit  current  of  eight  times 
normal  full-load  cmrent,  the  value  of  the  reactors  would  have  to  be 
25  per  cent,  based  on  the  full-load  current  of  one  generator,  and 
the  current  carrying  capacity  would  have  to  correspond  to  one- 
half  of  the  full-load  cmrent  of  one  generator,  this  being  the  full 
load  on  the  reactor.  The  displacement  between  the  sections  on 
the  above  assmnptions  would  be  approximately  7^*^,  a  value  at 
which  the  generators  of  the  sections  could  safely  be  maintained  in 
parallel.  As  a  fact,  this  could  be  done  safely  at  twice  this  angle 
and  they  would  probably  not  fall  out  of  step  until  the  displace- 
ment was  three  or  four  times  this  value. 


Digitized  by 


Google 


474  ELECTRICAL  EQUIPMENT 

The  number  of  sections  into  which  a  bus  should  be  divided 
depends  largely  upon  the  individual  sytem,  and  the  conditions 
under  which  it  is  expected  to  operate.  In  the  above  example, 
with  four  generators  on  the  section,  the  total  power  which  an  oil 
switch  may  be  required  to  rupture  would  be  equivalent  to  the 
short-circuit  current  of  five  generators  or  forty  times  the  full  load 
of  one  generator.  If  the  generators  were  rated,  say,  20,000  Kv.A., 
and  the  switch  equipped  with  instantaneous  relays,  the  switch 
would  have  to  rupture  40 X 20,000 X. 6 =480,000  Kv.A. 

When  dealing  with  the  subject  of  bus  reactors,  it  may  be  of 
interest  to  consider  their  action  a  little  more  fully,  and,  in  order  to 
obtain  some  idea  of  the  angular  relations  of  the  currents  and 
voltages  the  following  case  will  be  considered. 

Assmne  an  arrangement  as  illustrated  in  Fig.  294.  The  equip- 
ment consists  of  four  20,000  Kv.A.  generators,  having  a  short- 
circuit  ratio  of  eight  times  normal  full-load  current.  The  bus  is 
divided  in  two  sections  by  means  of  a  reactor  which  will  peraiit  a 
power  transfer  equivalent  to  one-half  the  capacity  of  one  generator, 
as  shown.  The  power-factor  of  the  load  is  0.8  and  it  is  assumed 
that  the  generators  are  to  carry  equal  loads  and  that  the  voltages 
of  the  two  bus  sections  A  and  B  are  kept  the  same. 

It  is  at  once  apparent  that  the  generators  on  section  A  must 
supply  10,000  Kv.A.  through  the  reactor  to  section  B,  and  in 
order  to  limit  the  amount  to  this  value,  a  25  per  cent  reactor  is 
required,  this  figure  being  based  on  the  rating  of  one  generator. 
Based  on  the  actual  transfer  energy  (one-half  the  capacity  of  one 
generator),  it  would  be  12i  per  cent;  thus,  a  total  of  1250  KvA., 
three-phase,  or  416  Kv.A.  per  single  reactor. 

The  diagram  illustrating  the  current  and  voltage  relations  may 
be  constructed  as  follows:  Draw  OA  and  OB,  representing  the 
equal  voltages  of  the  two  sections,  in  such  a  manner  that  AB, 
which  represents  the  voltage  across  the  reactor,  is  12J  per  cent  of 
OA.  Since  this  voltage  differs  in  phase  from  the  current  prac- 
tically 90*^  (neglecting  the  reactor  losses),  it  follows  that  the 
angular  position  of  the  circulating  current  is  midway  between  the 
voltages  OA  and  OB.  OC  represents  the  current  on  section  A 
lagging  approximately  37*^  (cos  <t>  =  .8)  behind  its  voltage  Oi, 
while  OD  represents  the  current  on  section  B,  this,  in  turn,  lagging 
37*"  behind  the  voltage  OB.  OC  and  OD  should  be  drawn  to  scale 
BO  that  theix  lengths  represent  the  actual  proportions  between  the 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS 


475 


loads;   i.e.,  OC  should  correspond  to  30,000  and  OD  to  50,000 
Kv.A.    CE  and  DF  now  represent  the  current  flowing  through 


80,000  EtA. 


Four  80.000  KvA,  Generators 

B        I        A 


FkG.  2^. — Arrangement  of  Bus  Reactor  and  Diagram  Showing 
Current  and  Voltage  Relations. 

the  reactor,  the  phase  position  of  these  corresponding  to  the  middle 
line  between  OA  and  OB,    The  current  of  the  generators  on  sec- 


Digitized  by 


Google 


476  ELECTRICAL  EQUIPMENT 

tion  A  is  represented  by  the  vector  OE  and  that  of  the  generatws 
on  section  B  by  OF.  It  will,  therefore,  be  noted  that  the  current 
through  the  reactor  increases  the  load  on  generator  A  and  de- 
creases that  on  B.  Similarly,  the  power-factor  of  the  load  on  A 
has  been  increased  and  that  on  B  decreased.  The  projection  OF 
on  OB  equals  the  projection  OE  on  OA  showing  that  the  energy 
deUvered  by  the  generators  on  each  section  is  equal. 

The  size  of  feeder  reactors  depends  on  the  size  of  the  feeders, 
the  relation  of  their  capacity  to  that  of  the  generators  and  the 
capacity  of  the  feeder  circuit  breakers,  i.e.,  their  safe  rupturing 
capacity.  In  general,  the  reactor  required  for  an  overhead  cir- 
cuit will  be  lees  than  for  an  underground  cable,  because  the  former 
usually  has  a  higher  reactance. 

As  an  example,  assume  a  100,000  Kv.A.  station,  the  inherent 
reactance  of  the  generators  being  such  as  to  limit  the  short- 
circuit  current  to  six  times  full-load  current.  In  case  of  a  short 
circuit  on  one  of  the  feeders  close  to  the  bus-bars,  not  less  than 
600,000  Kv.A.  would  pass  into  the  fault,  and  if  the  capacity  of 
the  feeders  were  3000  Kv.A.,  this  would  be  equal  to  two  hundred 
times  the  normal  capacity  of  the  feeders  and  the  reactance  of  the 
generators  would,  therefore,  only  be  equivalent  to  one-half  per 
cent  reactance  in  the  feeders. 

If  now  a  3  per  cent  reactor  is  placed  in  each  feeder  the  total 

reactance  will  be  equal  to  3.5  per  cent  and  the  worst  possible  short 

3000 
circuit   conditions   would   be   equivalent   to   -^-r- X 100 =86,000 

3.5 

Kv.A.,  or  28.6  times  the  normal  capacity.    The  voltage  of  the 

bus  instead  of  dropping  to  zero,  would  only  be  reduced  to  28.6X3 

or  to  approximately  86  per  cent  of  its  normal  value. 

Besides  the  above,  the  problem  must  also  be  dealt  with  from 
the  economical  point  of  view.  For  example,  the  cost  of  the 
different  types  and  sizes  of  reactors  must  be  compared,  the  space 
occupied  thereby  must  be  considered  as  well  as  the  eflFect  which 
the  introduction  of  reactors  may  have  in  permitting  less  expensive 
switches  and  apparatus  to  be  used. 

The  magnitude  and  intricate  connections  of  modem  transmis- 
sion systems  makes  the  determination  of  the  probable  short-circuit 
current  at  the  various  points  a  very  tedious  work,  and,  in  order  to 
facilitate  the  calculations  it  is  always  desirable  and  almost  neces- 
sary to  graphically  represent  the  system  in  a  diagram  with  the 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS 


477 


reactances  given  for  the  different  apparatus  and  circuits.    Those 
values  should  preferably  be  expressed  in  per  cent,  based  on  some 
nominal  capacity  such  as  the  capacity  of  the  principal  generating . 
unit  as  previously  explained.     The  procedure  of  calculation  is 
best  explained  by  an  example: 

Assume  a  network,  as  shown  in  Fig.  295.    The  various  por- 
tions of  the  circuit  have  the  per  cent  reactance  indicated,  all  based 


8ab.Sta. 


z.ost 


8ub.8W 


Fig.  295. — Typical  Connection  of  a  Simple  Transmission  System. 


on  10,000  Kv.A.     Stations  No.  1  and  2  are  generating  stations, 
each  containing  a  number  of  generators  with  a  combined  reac- 
tance, as  shown.    For  a  three-phase  short  circuit  at  A,  the  short 
circuit  Kv.A.  is  found  as  follows: 
6.2+3  =  9.2 
3.3+1.52=4.82 
1 


1 


1 


=3.16 


9.2  '  4.82 


Digitized  by 


Google 


478  ELECTRICAL  EQUIPMENT 

3.16+3.54  s  6.7  total  reactance  of  circuit  from  Station 

No.  1  to  short-circuit. 
1.43+0.75=2.18 
1 


1     .  1 


=0.685 


2.18  '  1 
0.686+1.6=2.285 
1 


1 


=  1.215 


2.285  '  2.6 

1.215+0.765=1.98 

1.98+7.4+2.5=11.88  total  reactance  of  circuit  from  Sta- 
tion No.  2  to  short-circuit. 
1 

4.29  combined    reactance   £rom    Station 


6.7  '  11.88 


No.  1  and  2  to  short-circuit. 


100 
f^X10,000=23.3XlO,000=233,000  Kv.A.  afshort-dr- 

4.A9  ., 

CUlt. 

The  proportion  of  this  furnished  by  Station  No.  1  and  Station 
No.  2  may  be  found  as  follows: 

^   :  =  .0842 


11.88 
Station  No.  1: 

1400 
m^X233,000=  149,000  Kv.A. 

Station  No.  2: 

^^X233,000=84.000  Kv.A. 

In  like  manner  the  proportion  of  this  Kv.A.  that  flows  over 
each  individual  portion  of  the  circuit  may  be  readily  determined 

In  certain  cases  a  system  may  consist  of  such  a  complex  net- 
work of  lines  so  as  to  make  the  calculations  exceedingly  difficult 
and  the  results  consequently  more  or  less  uncertain.  To  aid  in 
the  solution  of  problems  of  this  nature,  an  electrical  device  has 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  479 

been  designed  by  which  the  results  can  be  obtained  directly  and 
with  sufficient  accuracy  for  most  practical  purposes. 

It  consists  of  a  table  under  which  are  mounted  a  number  of 
rheostats  of  the  disk  type  having  the  operating  handles  pro- 
jecting through  the  top  of  the  table  as  shown  in  Fig.  296.  To 
each  handle  is  fastened  a  pointer  which  revolves  over  a  graduated 
dial  on  top  of  the  table,  the  graduations  being  in  per  cent  reac- 
tance (actually  resistance).  The  terminals  of  each  of  the  rheo- 
stats are  brought  out  to  metal  blocks,  also  fastened  to  the  top  of 
the  table.    These  blocks  contain  holes  in  which  may  be  inserted 


Fig.  296. — Device  for  Calculating  Short-circuit  Currents. 

taper  plugs  connected  together  by  flexible  leads  so  that  the  rheo- 
stats can  be  interconnected  in  any  desired  manner.  The  resistance 
of  the  rheostats  is  taken  as  representing  reactance  in  an  actual 
system,  and  a  rheostat  may  thus  be  set  for  any  value  of  equivalent 
reactance  and  plugged  into  the  network  if  desired.  Direct- 
current  at  125  volts  is  used  for  operating  the  table,  the  negative 
side  being  connected  to  ground  and  when  it  is  desired  to  place  a 
short-circuit  on  any  part  of  the  system,  that  point  is  simply  con- 
nected to  the  ground  in  such  a  manner  as  to  establish  a  short- 
circuit  through  the  rheostats  representing  the  generators  and  the, 
rheostats  representing  the  interconnected  network  of  lines.    The 


Digitized  by 


Google 


480 


ELECTRICAL  EQUIPMENT 


current  in  any  part  of  the  sytsem  can  be  read  by  means  of  an 
ammeter. 

For  a  more  complete  description  of  this  calculating  tabkthe 
reader  is  referred  to  the  "  General  Electric  Review  "  for  October, 
1916. 

Fig.  297  shows  a  complicated  network  in  which  a  number  of 
generators  feed  a  conunon  bus  at  points  separated  by  bus-bar 


AX  Kv- 


-»m.wo  Kv^ 
— nmr^ 


O 


%!f 


lOi 


1 901,000 
Kv-a. 


ll9S,000 


-v28,000  Kv^ 

— Tnnnr^— 


O 

19M 


101,000 
Kv-iu 


-».7.000  Kv-A. 

-/yyyip 


116,000 
Kv-a. 


:iL« 


-►5S,000  Kr-m, 

-nnnp 

Of 


o 


145,000 
Kr-m. 


SM,000 
Kr-m, 


3.1tf 


1»T.M 


7.1* 


-»1S,000  Kr-fc 
— ^I'lTV" 


0' 


18,000 
Kv-ft. 


SMf 


Fig.  297. — Short-circuit  Current  Calculations. 


reactors.  The  percentages  of  reactance  given  are  based  on  45,000 
Kv.A.  The  short-circuit  occurs  at  the  point  A.  The  solution  of 
this  problem  is  rather  involved,  and  it  has  been  accomplished  in 
this  case  by  means  of  the  calculating  table  described,  with  the 
results  indicated  on  the  figure. 

Single-phase  Short-circuit  Currents.  Heretofore,  we  have 
dealt  with  three-phase  or  balanced  currents.  Of  late  years  the 
tendency  has  been  more  and  more  toward  the  operation  of  systems 
with  transformers  connected  in  Y  and  neutral  groimded  on  the 
high-voltage  side.  When  a  ground  occurs  on  the  Une  a  three- 
phase  short  circuit  does  not  result  but  rather  a  single-phase  short- 
circuit.  A  brief  outline  of  the  method  used  in  handling  such 
problems  is  given  in  the  following,  and  for  a  more  detailed  study  of 


Digitized  by 


Google 


CUREENT-LIMITING  REACTORS 


481 


the  subject  the  reader  is  referred  to  an  article  in  the  ''  General 
Electric  Review  "  of  June,  1917,  by  W.W.  Lewis,  entitled  "  Short- 
Circuit  Currents  on  Grounded  Neutral  Systems." 

Referring  to  Fig.  298:  Let  G  represent  a  generator,  T\  a  trans- 
former with  high  voltage  winding  connected  in  Y  and  neutral 


Fia.  298. — Single-phase  Short  Circuits. 

grounded;  T2  a  transformer  stepping  down  the  voltage  for  the 
load  L.  The  ohmic  reactance  of  the  generator  is  represented  by 
xi;  of  the  step-down  transformer  by  X2]  of  the  grounded  trans- 
former by  z'y  of  the  portions  of  line  from  transformer  to  the  point 
A  by  T/i  and  2/2,  and  of  the  total  length  of  line  by  y,  E  is  the  normal 
hi^-tenaion  voltage.  All  reactances,  etc.,  are  expressed  in  terms 
of  their  high-voltage  equivalents. 

Assume  a  ground  at  A,  Then  currents  will  flow  as  indicated 
by  the  arrows.  The  value  of  the  current  is  expressed  by  the  fol- 
lowing equation: 


t  = 


.577g 
xi+z+yi 


or  expressed  in  per  cent  reactance  based  on  the  normal  three- 
phase  line  current  I 

100/ 


t  = 


per  cent  Izi+per  cent  ly i+per  cent  Iz' 


Now  consider  the  arrangement  of  Fig.  299,  i.e.,  ungrounded 
transformer  Ti  at  the  generating  end  and  transformer  T2  with 
groimded  neutral  at  the  load  end.  The  short-circuit  current  will 
flow,  as  indicated  by  the  arrows.  The  delta  winding  of  trans- 
former T2  serves  to  cause  equal  in-phase  currents  to  flow  in  each 


Digitized  by 


Google 


482 


ELECTRICAL  EQUIPMENT 


leg  of  the  Y.    The  voltage  drop  in  each  part  of  the  circuit  is  in 
phase  with  the  voltage  of  the  short-circuited  leg  a-i,  and  the 


^^ —    y 
Fig.  299. — Single-phase  Short-circuit  Currents 

total  voltage  drop  is  equal  to  cr-d  or  0.866B.    The  following  equa- 
tions may  be  written  from  the  figure: 

.8665 = i{xi+X2+y+2yi)  +e, 
2(e-iz)=i(y2+z); 
from  which  we  find 

.866S 


t  = 


(x.+..)+|+^^+|' 


or  expressed  in  per  cent  reactance  based  on  normal  three-phase 
line  current  I 

._ 100/ 

'   i{%ixi+7oix2)+%iy+7oiyi+7oi^ 

Based  on  these  fundamental  equations  it  is  possible  to  sdve 
problems  in  cases  involving  a  number  of  generating  stations,  a 
network  of  lines,  etc.  As  the  number  of  generating  stations 
increases,  however,  the  equations  increase  in  complexity  and  the 
solution  becomes  quite  laborious.  The  labor  is  lessened  some- 
what by  representing  the  network  by  an  equivalent  circuit  with 
the  component  parts  expressed  in  per  cent  reactance  and  solving 
either  by  the  sUde  rule  or  by  the  calculating  table. 

An  example  will  illustrate  this.  In  Fig.  300  let  d  and  Gs 
represent  generators,  Ti  and  T2  transformers  with  isolated  neutrals 
and  Ts  a  transformer  with  grounded  neutral.  The  percentages 
of  reactance  based  on  10,000  Kv.A.  100,000  volts  and  three-phase 
are  indicated. 


Digitized  by 


Google 


CURRENT-LIMITING  REACTTORS 


483 


For  a  ground  on  one  line  at  the  point  A,  giving  a  single-phase 
short-circuit,  currents  flow,  as  shown  by  the  arrows.    An  equiva- 


Fia.  300. — Calculation  of  Single-phase  Short-circuit  CuirentB. 

-IM         -3.M 


FiQ.  301. — ^Elquivalent  Short-circuit  Corresponding  to  Fig.  300. 

lent  circuit  for  Fig.  300  may  be  drawn  as  shown  in  Fig.  301. 
This  circuit  may  be  solved  as  follows: 

10+3.33+4-1-4=21.33 
12+4+10+10=36 
1  1  1 


1      .    1      .0469+. 0278     .0747 


=  13.4 


21.33  '  36 

3+2+3+7=15 

13.4+15=28.4 

100 
ti+t2=^X/=3.52X67.7=203 

0469 
n  =  ^Q;^X203  =  .628X203  =  127.5  amps. 

0278 
t2=^^X203=.372X203=75.5  amps. 


Digitized  by 


Google 


484  EILECTRICAL  EQUIPMENT 


^pikliiK\il&\v 


f^immm 


r  *  - 


fl^frinnnnfr 


o 


s 


Digitized  by 


Google 


CURRENT-LIMITING  REACTORS  485 

Mechanical.  Design.  Current-limiting  reactors  must  be  de- 
signed so  as  not  to  saturate  at  short-circuit  when  the  full-circuit 
voltage  comes  across  the  reactance,  and  for  that  reason  they  are, 
as  a  rule,  built  without  an  iron  core.  There  is,  however,  no  the- 
oretical objection  to  the  use  of  iron  and  if,  for  example,  a  reactor 
for  say  25  per  cent  were  required,  it  would  be  feasible  and  pos- 
sibly even  economical  to  provide  an  iron  core,  which,  in  such  a 
case,  would  have  to  have  a  normal  magnetic  density  of  one- 
fourth  the  saturation.  For  3  to  10  per  cent  reactors,  however,  an 
excessive  amount  of  iron  would  be  required  to  prevent  saturation 
at  short-circuits,  thus  making  an  iron  core  highly  tmeconomical. 

The  latest  construction  of  reactors  is  shown  in  Kg.  302.  It  is 
known  as  the  "  cast-in  "  type  because  of  the  fact  that  the  winding 
is  cast  and  directly  supported  in  the  concrete  structure. 

The  conductor,  which  may  consist  of  one  or  several  cables  in 
multiple,  is  wound  radially  in  conical  layers,  an  ample  factor  of 
safety  being  preserved  between  each  and  every  turn.  The  adja- 
cent layers  are  inclined  in  opposite  directions  with  ample  spaciogs 
between  the  layers,  the  spacing  varying  with  the  voltage  of  the 
circuit  and  the  munbers  of  layers  required.  Ample  spacing  is 
essential  during  short-circuit  conditions  since  there  is  almost 
always  arcing  at  the  point  of  short-circuit  which  may  set  up  high- 
frequency  disturbances.  Any  two  layers  thus  converge  toward  the 
point  where  the  interconnecting  cross-over  is  made  and  where  the 
maximum  voltage  between  the  layers  is  consequently  equal  to 
<that  between  turns. 

The  windings  are  held  rigidly  in  their  position  by  the  vertical 
coil  supports  which  are  cast  around  the  turns  after  these  have 
been  woimd  in  a  form.  The  concrete  is  thereafter  cured  under 
high  steam  pressinre  which  gives  it  a  mechanical  strength  obtained 
in  no  other  way. 

8.   SWITCHING  EQUIPMENT 

The  engineering  problems  in  connection  with  the  operation 
of  high-voltage  hydro-electric  transmission  systems  are  very 
largely  those  which  have  to  do  with  preventing  interruptions  to 
the  service  and  which  isolate  and  localize  electrical  disturbances 
before  they  can  become  of  a  general  nature.  This  resolves  itself 
not  only  into  the  general  design  of  the  apparatus  but  also  to  a 
careful  study  of  the  best  possible  arrangement  of  the  different 


Digitized  by 


Google 


486  ELECTRICAL .  EQUIPMENT 

circuits  and  the  method  of  switching.  Reliability  and  contin- 
uity of  service  are  the  main  considerations,  but  besides  this  the 
protection  of  the  apparatus  from  injury  should  not  be  lost  sight  of. 

The  switching  equipment  is  the  key  to  the  entire  system,  and 
the  first  requisite  to  decide  on  is  the  system  of  connections,  the 
diagram  of  which  should  be  worked  out  with  the  greatest  care, 
taking  into  consideration  the  various  equipments  and  the  normal, 
as  well  as  possible  abnormal,  operating  conditions  of  the  entire 
system.  The  design  of  the  control  boards  and  the  selection  and 
arrangement  of  the  oil  circuit  breakers,  bus-bars,  etc.,  depends 
greatly  on  the  system  of  connections;  in  fact,  the  design  of  the 
entire  power-station. 

In  taking  up  the  various  problems  dealing  with  the  design  of  a 
switching  equipment,  space  will  only  pennit  the  fundamental 
principles  to  be  dealt  with,  and  only  some  of  the  more  important 
apparatus  can  be  briefly  described.  It  would  be  of  Uttle  value  to 
go  into  the  minute  details  of  the  engineering  features  connected 
with  a  switching  equipment  because  the  art  changes  so  rapidly, 
and  new  and  improved  lines  of  apparatus  are  brought  on  the 
market  so  rapidly,  that  they  change  for  ahnost  every  new  impor- 
tant installation. 

System  of  Connections  and  Relay  Protection.  In  laying  out 
the  S3rstem  of  connections  and  the  protective  switching  and 
relaying  equipment  for  a  high-tension  transmission  sjrstem,  there 
are  a  number  of  general  principles  which  must  be  kept  clearly  in 
mind.  Chief  among  these  is  continuity  of  service  which  is  now  of 
prime  importance  and  this  has  been  brought  about  mainly  by 
the  steadily  increasing  demand  for  a  much  higher  standard  of 
service  than  formerly.  This,  in  turn,  involves  a  flexibiUty  in  the 
arrangement  of  the  connections  so  as  to  reduce  to  the  absolute 
minimum  the  amount  of  apparatus  which  will  be  automatically 
disconnected  in  case  of  trouble,  and  also  to  provide  for  sectional- 
izing  any  apparatus  for  inspection  and  repairs.  Besides  this,  the 
protection  of  the  apparatus  from  injury  should  be  given  careful 
study.  These  considerations  are,  however,  very  closely  con- 
nected and  must  naturally  be  treated  together.  In  this  connec- 
tion it  should  be  noted  that  the  function  of  an  automatic  selective 
switching  is  not  any  longer  correlated  to  the  idea  of  protecting 
the  apparatus  against  ordinary  overloads,  but  that  the  relays  are 
intended  to  operate  only  on  breakdowns,  although  their  setting 


Digitized  by 


Google 


SWITCfflNG  EQUIPMENT  487 

is  usually  given  in  per  cent  overload  of  the  rated  capacity  of  the 
circuit. 

The  particular  system  of  connections  to  be  used  depends 
obviously  on  the  conditions  to  be  met,  and  each  system  must 
be  studied  and  an  individual  solution  applied.  There  are,  how- 
ever, many  points  of  similarity,  and  the  solution  in  one  case  will 
serve  as  a  partial  guide,  at  least  in  others.  In  any  event,  the 
system  as  a  whole  should  be  carefully  considered  in  deciding  on 
the  connections,  and  the  conclusions  should  not  be  based  on  the 
condition  in  a  generating  station  or  a  substation  alone.  The 
characteristics  of  the  customer's  load  conditions  must  be  care- 
fully investigated  and  futiu*e  probable  loads  and  additions  pre- 
determined as  far  as  possible. 

It  is  especially  essential  to  provide  an  uninterrupted  service 
for  large  and  important  customers,  as  the  success  of  the  project 
depends  in  most  cases  entirely  on  the  ability  to  maintain  a  satis- 
factory service  for  these,  but,  on  the  other  hand,  the  smaUer  cus- 
tomers must  also,  of  course,  be  considered  and  given  the  best 
service  possible.  For  this  reason  the  power  to  important  cus- 
tomers is  often  supplied  from  two  sources,  such  as  from  two  sub- 
stations or  by  means  of  double-line  circuits,  etc.  Two  such 
sources  of  supply  are,  of  course,  the  ideal  arrangement,  in  which 
case  one  of  them  would  be  automatically  cut  out  in  case  of  trouble 
while  the  other  would  be  kept  in  operation  and  continue  to  carry 
the  load.  This,  however,  is  not  always  possible  for  every  cus- 
tomer. 

In  a  general  way  the  service  of  a  large  power  system  with  its 
transmission  and  distributing  lines  can  be  likened  to  a  combined 
express  and  local  train  service  of  a  transportation  company.  The 
transmission  lines  feeding  the  different  substations  on  the  system 
correspond  to  the  express  trains  and  must  be  absolutely  free  from 
interruption,  for  which  reason  such  lines  should  be  so  arranged 
that  any  substation  is  fed  by  two  independent  circuits.  The 
local  train  service  would,  on  the  other  hand,  correspond  to  the 
distributing  liDes,  and  any  interruptions  which  might  be  per- 
mitted to  occur,  should  be  confined  to  these  local  circuits.  Of 
course,  if  the  service  demands,  even  these  circuits  can  be  installed 
in  duplicate. 

In  a  power  transmission  system  the  chief  source  of  trouble  is 
always  the  transmission  line  and  it  can  mostly  be  traced  back  to 


Digitized  by 


Google 


488  ELECTRICAL  EQUIPMENT 

the  insulators.  This  subject  of  insulator  design  has  been  studied 
very  carefully  during  the  past  few  years  and  great  improvements 
have  been  made,  but  they  have  as  yet  an  apparent  deterioration 
causing  breakdowns  from  time  to  time.  Together  with  atmos- 
pheric disturbances  in  districts  frequented  by  Ughtning  storms,  it 
makes  the  transmission  line  a  vulnerable  part  of  the  system  and 
the  largest  percentage  of  troubles  is  caused,  thereby.  Apparatus 
troubles  are  furthermore  often  traced  directly  to  line  troubles  as  a 
secondary  cause  from  arcing  groimds,  surges,  etc. 

The  secret  of  success  in  relay  protection  is  speed.  That  is, 
the  faulty  sections  should  be  cut  out  so  rapidly  as  to  prevent  the 
synchronous  apparatus  connected  to  the  system  from  falling  out 
of  step  and  stopping.  The  time  limit  for  this  differs,  however, 
depending  on  the  stability  of  the  apparatus  and  where  the  short- 
circuit  occurs.  The  closer  ^o  the  machines,  the  shorter  the  time 
before  they  drop  out. 

The  longer  an  arcing  ground  hangs  on,  the  more  damage 
it  will  do  in  breaking  insulators  and  melting  off  the  transmission 
wires.  The  arc  is  very  small  to  begin  with,  but  increases  rapidly 
in  size  and  should  therefore  be  quickly  cleared  so  as  to  cause  as 
Uttle  damage  as  possible. 

Interruptions  can,  in  many  cases,  be  traced  to  the  customer's 
own  fault.  For  example  the  motor  breakers  may  be  set  at  such 
low-tripping  value,  that  if  the  power  of  the  system  should  mo- 
mentarily drop  off  and  come  on  again,  the  heavy  current  rush 
would  trip  the  breaker  and  disconnect  the  machine.  To  provide 
against  such  interruptions  the  breaker  need,  of  course,  only  be 
set  for  a  sufficiently  high  value.  Similarly,  with  motor  breakers 
provided  with  low-voltage  releases,  which  would  cause  the  motor 
to  be  cut  off  from  the  system  on  any  momentary  voltage  drop  unless 
provided  with  a  time-limit  device.  Such  relays  should  therefore 
be  avoided  as  far  as  possible  if  strict  continuity  of  service  is 
essential. 

The  time  in  which  a  fault  might  be  cleared  depends  naturally 
on  how  quickly  the  switches  may  disconnect  the  faulty  section. 
This  in  turn  depends  on  the  rapidity  of  the  switch  action,  and  on 
the  characteristics  of  the  relay  which  is  used  for  closing  the  trip- 
ping circuit  of  the  oil  circuit  breaker. 

Due  to  the  inertia  of  the  moving  parts  it  is,  of  course,  impos- 
sible for  a  breaker  to  open  instantaneously,  and  it  requires  approx- 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  489 

imately  one-quarter  second  for  a  large  breaker  to  open  after  the 
tripping  coils  have  been  energized.  The  time  interval  between  the 
moment  at  which  a  short-circuit  takes  place  and  the  moment  at 
which  the  tripping  circuit  of  the  breaker  is  closed  may  be  varied 
at  will  by  selecting  relays  of  different  time  settings. 

By  means  of  such  overload  relays  in  connection  with  reverse 
power,  balanced,  differential  and  pilot  wire  relays,  the  character- 
istics and  uses  of  which  are  described  in  the  section  on  "  Relays," 
it  is  possible  to  obtain  a  selective  automatic  switch  action,  which 
will  only  disconnect  the  faulty  section  of  the  system  without 
interrupting  the  remainder  thereof.  The  types  of  relays  and  their 
arrangement  to  accomplish  such  a  result  depend  entirely  on  the 
system  of  connection  and  the  conditions  to  be  met. 

The  generators  should  preferably  be  paralleled  on  a  low- 
tension  bus  and  this  should  be  arranged  so  that  it  can  be  inspected 
and  cleaned  from  time  to  time  without  shutting  down  the  sta- 
tion. With  smaller  stations  a  single  bus  may  be  sufficient  and 
by  sectionalizing  the  same  the  operation  may  be  so  arranged  that 
one  section  can  be  cleaned  when  the  units  belonging  thereto  are 
cut  out  during  Ught  load.  As  a  rule,  however,  important  stations 
should  be  provided  with  double  generator  buses  (Fig.  303)  and 
the  generators  connected  thereto  either  by  means  of  double  oil 
circuit  breakers  or  by  means  of  one  common  oil  circuit  breaker  and 
two  sets  of  disconnecting  switches,  one  for  each  bus.  Double 
oil  circuit  breakers  are  preferable  as  they  permit  the  transfer  to 
be  done  entirely  from  the  main  switchboard  and  thus  insures  a 
speedier  operation.  This  also  applies  to  the  transfer  on  the  high- 
tension  side,  and  in  this  case  it  is  even  more  important,  due  to 
the  greater  difficulty  of  manipulating  the  large  high-tension  dis- 
connecting switches.  Double  oil  circuit  breakers  further  permit 
of  inspection  and  repair  of  one  breaker,  while  the  other  is  in  service. 

The  low-tension  buses  should,  furthermore,  be  sectionalized 
if  the  capacity  of  the  station  is  large,  so  as  to  limit  the  short-cir- 
cuit current  to  a  value  which  can  safely  be  ruptured  by  the  oil- 
circuit  breakers,  as  fully  described  in  the  section  on  "  Curren1>- 
limitmg  Reactors." 

The  transformers  should  preferably  be  grouped  so  as  to 
forai  units  with  the  lines  and  with  such  an  arrangement  the 
double  low-tension  bus  is  preferable  in  order  to  obtain  the  most 
flexible  method  of  transfer. 


Digitized  by 


Google 


490 


ELECTRICAL  EQUIPMENT 


Stations  may,  however,  be  found  in  which  the  transformers 
are  grouped  with  the  generators,  as  in  generating  station  C  (Fig. 


T 


'Snb^titiOD 


FiQ.  303.— Tjrpical  System  of  Connections. 

305).    In  such  a  case  a  parelleling  bus  may  be  omitted  and  simply 
a  low-tension  transfer  bus  provided. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


491 


'SubYI 


T.   T    .^snbin: 


8«b  ymt^y^  4?j'a'"'  IT 

.'       >  ■       I 

I    '.      !    I. 


d      Q 


f     4    f     * 


^Sabn 


PJ tZ^ ?  't' 


B     ?:? 


^ 


«»^ 


f>f^f^    .g^ 


Oeneratioff 
SteUpQ 


Fig.  304. — ^Typical  System  of  Connections. 


Digitized  by 


Google 


492  ELECTRICAL  EQUIPMENT 

The  reason  for  grouping  the  traneformers  with  the  lines  is  to 
avoid  switching  on  the  higl:\-tension  side  of  the  transformers. 
Paralleling  of  the  high-tension  side  should  also  be  avoided  and  this 
appUes  especially  to  delta-connected  transformers  where  the  surges 
set  up  by  arcing  grounds  on  one  line  may  be  transmitted  to  the 
other  line.  With  transformers  having  the  high-tension  windings 
Y-connected  and  the  neutral  grotmded,  paralleling  on  the  high- 
tension  side  may  not  be  so  serious  and  may  in  certain  cases  be 
advisable  for  the  sake  of  flexibility. 

Means  should,  however,  always  be  provided  for  transfer 
on  the  high-tension  side,  and  this  may  be  done  in  various  ways 
as  indicated  in  Figs.  303  and  304,  in  the  former  case  simply  by  a 
tie  between  the  two  lines  and  in  the  latter  by  a  transfer  bus. 
Such  means  for  transfer  should  also  be  arranged  at  intervals 
along  the  transmission  lines,  perferably  at  substations  (Fig.  305), 
or  places  where  branch  lines  are  tapped  to  the  main  line. 

It  is  customary  to  make  the  generator  circuit  breakers  non- 
automatic,  but  for  very  large  and  important  imits  it  may  be 
desirable  to  protect  them  against  internal  short  circuits,  which  is 
readily  accomplished  by  means  of  differential  relays  as  described 
under  "  Rela3rs." 

The  switch  and  relay  protection  of  the  transmission  lines,  tie 
lines,  etc.,  is  very  compUcated  and  no  general  rules  can  be  given 
except  to  state  that  the  protective  features  should  be  of  such  a 
selective  nature  that  when  trouble  occurs,  the  section  involved 
should  be  inmiediately  disconnected  without  the  dropping  of 
imnecessary  load  or  power.  The  protective  devices  to  accomplish 
this  depend  entirely  on  the  conditions  involved  and  are  best 
explained  by  considering  a  few  typical  examples. 

Example  I:  This  refers  to  a  system  as  illustrated  by  the 
disgram  in  Fig.  303  and  consists  of  one  generating  station  feeding 
a  single  substation  over  two  parallel  transmission  lines. 

All  the  high-tension  Une  circuit  breakers  are  non-automatic 
and  are  only  intended  for  sectionalizing  purposes,  as  are  the  high- 
tension  tie  breakers,  which  should  be  open  under  normal  operation 
so  that  the  system  would  only  be  operated  in  paraUd  on  the  low- 
tension  side  of  the  step-up  as  well  as  the  step-down  transformers. 
The  low-tension  transformer  circuit  breakers  in  the  generating 
and  substation  respectively  should  be  of  the  automatic  tjrpe,  the 
former  being  provided  v^th  time-limit  relays  and  the  latter  with 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


493 


OeDeratlDff- Station  jL 


Fig.  306. — Typical  System  of  Connections. 


Digitized  by 


Google 


494  ELECTRICAL  EQUIPMENT 

reverse  power  relays  in  connection  with  overload  relays  which, 
in  a  system  of  this  kind,  can  be  set  for  practically  instantaneous 
action. 

The  time  settings  of  the  relays  should,  in  general,  be  so  arranged 
that  the  substation  circuit  breaker  trips  out  first  and  then  the 
breaker  in  the  generating  station,  thus  disconnecting  only  the 
faulty  line.  The  load  is  then  shifted  over  to  the  other  line  which 
remains  in  service  and  may  overload  the  transformers  of  this  hne. 
This  will  not  cause  any  danger,  as  transformers  can  readily  cany 
up  to  100  per  cent  overload  for  a  few  minutes  until  the  operator 
has  had  time  to  open  the  high-tension  circuit  breakers  of  the  faulty 
line  and  close  the  tie  circuit  breakers  and  the  low-tension  trans- 
former breakers,  thus  again  paralleling  the  transformers.  The 
overload  relays  on  the  generating  station  breakers  should,  there- 
fore, be  set  sufficiently  high  so  that  they  can  carry  the  entire  load 
without  tripping. 

The  outgoing  substation  feeder  circuit  breakers  should  be 
equipped  with  inverse  time-limit  relays  set  proportionaDy  lower 
than  the  overload  relays  in  the  generating  station.  In  a  system 
of  this  kind  the  time  element  may  be  very  short,  which  is  an 
important  item,  as  previously  mentioned.  The  substation  line 
relays  can,  therefore,  be  set  for  instantaneous  action  on  reversal 
and,  in  such  a  case,  the  generating  station  overload  relays  need 
only  be  set  for  a  second  at  the  most.  The  feeder  relays  may  also 
be  set  for  nearly  instantaneous  action  in  order  to  have  them  trip 
before  the  overload  line  relays  in  the  generating  station. 

Example  II:  This  refers  to  a  somewhat  more  complicated 
system,  as  illustrated  in  the  diagram.  Fig.  304.  It  consists  of 
one  large  generating  station  still  feeding  one  main  substation  from 
which  several  distribution  systems  are  supplied. 

The  main  substation  in  this  case  is  fed  over  three  parallel 
transmission  Unes  and  as  far  as  the  relay  protection  for  these  is 
concerned,  it  may  be  done  in  the  same  manner  as  explained  in 
Example  I,  but  interconnected  reverse  power  relajrs  may  also  be 
used  in  either  case. 

The  first  consideration  in  relaying  a  system  of  this  kind  is  to 
keep  the  power  on  the  main  substation  bus,  no  matter  what  hap- 
pens, and  in  protecting  the  various  circuits  beyond,  the  sub- 
station bus  may  be  treated  just  as  if  it  were  the  generating  sta- 
tion bus. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  495 

Substations  II  and  III  are  fed  in  tandem,  the  fonner  by  three 
parallel  transmission  lines  and  the  latter  by  only  two,  and  the 
relaying  of  these  lines  should  be  identical  with  the  main  trans- 
mission lines  with  the  exception,  of  course,  that  the  time  settings 
have  to  be  proportionally  lower.  This  clearly  demonstrates  the 
point  of  graded  time  settings,  and  it  is  evident  that  the  circuit 
breakers  furthest  away  from  the  generating  station  should  have  the 
lowest  setting  and  the  succeeding  relays  in  each  section,  counting 
towards  the  generating  station,  should  each  have  an  ^increase  in 
the  time  element  of  about  half  a  second.  This  may  put  an 
excessive  time  on  the  breaker  nearest  the  generating  station  and, 
in  that  case,  by  the  use  of  inverse  time  limit  relays  without  def- 
inite minimum  time  setting,  taking  advantage  of  both  the  time 
and  current  difference,  it  may  be  possible  to  considerably  shorten 
the  time  on  all  the  relays.  This  usually  involves  a  careful  calcu- 
lation of  the  actual  short-circuit  values  to  determine  the  required 
settings.  In  certain  cases  where  the  time  setting  of  the  relays 
nearest  the  generating  station  has  become  rather  high,  it  has  been 
the  practice  to  also  install  an  instantaneous  overload  relay  in 
parallel  with  the  time  limit  relay  on  the  circuit  breaker  nearest  the 
generating  station  and  to  set  this  relay  very  high,  the  idea  being 
that,  in  case  of  a  severe  short-circuit,  it  should  disconnect  the  cir- 
cuit inamediately.  The  use  of  such  an  arrangement  is,  however, 
questionable  as  it  often  happens  that  the  instantaneous  relay 
acts  when  it  should  not,  thus  crippling  the  entire  service  of  all  the 
sections  in  the  series. 

Substations  IV  to  VIII  are  connected  on  the  ring  system  prin- 
ciple and  the  relaying  can  be  done  in  several  ways.  One  way 
would  be  to  provide  reverse  power  and  overload  relays  on  the 
incoming  line  circuit  breakers  in  each  substation  and  inverse  time 
limit  relays  on  the  outgoing  line  circuit  breakers,  this  being,  of 
course,  on  the  assumption  that  the  power  is  being  fed  into  station 
VI  over  both  lines.  Circuit  breakers  a,  c  and  e  would  then  be  pro- 
vided with  overload  relays  only  and  /,  d  and  h  with  reverse  power 
relays  in  combination  with  overload  relays.  The  settings  of  the 
overload  relays  would  be  in  the  following  order:  a,  c,  6,  /,  d  and  6; 
a  having  the  highest  setting  and  6  the  lowest. 

Example  III:  This  illustrates  a  system  consisting  of  three 
generating  stations  feeding  a  number  of  substations,  the  con- 
nections, as  illustrated  on  the  diagram  Fig.  305,  being  on  the 


Digitized  by 


Google 


496  ELECTRICAL  EQUIPMENT 

ring  principle.  In  a  case  of  this  nature  the  current  is  liable  to  be 
fed  in  either  direction  at  any  time,  and  the  protection  would  best 
be  accomplished  by  equipping  all  the  circuit  breakers  where 
parallel  connection  is  made  by  balanced  or  interconnected  reverse 
power  relays.  Where  transformers  are  involved  this  would  be  on 
the  low-tension  side  of  these,  the  interconnection  being  between 
similar  phases  of  the  two  parallel  lines. 

Where  balanced  current  conditions  may  be  assured,  the 
relays  may  be  set  for  instantaneous  action,  otherwise  it  might  be 
necessary  to  impose  a  slight  time  delay.  In  case  one  line  should 
become  disabled  it  will  then  immediately  be  disconnected  and 
arrangements  can  be  made  whereby  the  circuit  breakers  of  the 
other  line  would  be  automatically  provided  with  time-limit  fea- 
tures by  the  opening  of  the  circuit  breakers  of  the  disabled  line. 

Oil  Circuit  Breakers.  Oil  circuit  breakers  are  nearly  always 
used  for  rupturing  alternating-current  circuits,  due  to  the  fact 
that  they  do  not  cause  any  abnormal  disturbances  in  the  circuit, 
and  because  they  confine  the  destructive  effects  of  the  arc  to 
a  small  volume.  One  of  the  distinctive  features  of  the  oil  circuit 
breaker  is  that  the  current  is  interrupted  when  the  current  which 
is  maintaining  the  arc  passes  through  zero,  at  which  point  the 
electro-magnetic  energy  is  minimum.  It  remains  so  until  the 
voltage  between  the  contacts  rises  to  a  sufficient  value  to  punc- 
ture the  oil  insulation.  When  this  takes  place  the  flow  of  cur- 
rent is  reestablished  and  flows  for  another  half  cyde  and  so  on 
imtil  sufficient  insulation  is  interposed  between  the  contacts  to 
resist  the  maximum  voltage.  This  feature  is  taken  advantage  of, 
and  modern  oil  circuit  breakers  are  designed  with  a  view  of  utiliz- 
ing the  pressinre  developed  by  the  arc  to  introduce  a  large  amount 
of  oil  between  the  contacts. 

Owing  to  the  great  range  and  the  amount  of  current,  voltage 
and  power  to  be  handled  by  oil  circuit  breakers  for  such  cir- 
cuits, various  types  have  been  designed  to  suit  different  conditions. 
For  moderate  amoimts  of  power,  where  the  size  and  cost  of  the 
breaker  is  to  be  kept  to  a  minimum,  it  is  often  possible  to  locate 
all  of  the  poles  of  the  breaker  in  one  oil  tank.  For  slightly  larger 
amounts  of  power,  each  pole  is  placed  in  a  separate  oil  tank,  but 
all  poles  are  mounted  on  the  same  frame;  for  still  greater  amounts 
of  power,  at  moderate  voltages,  each  pole  is  in  a  separate  tank, 
and  each  tank  is  in  a  separate  masonry  compartment,  while  for 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  497 

very  high  voltage  work,  each  pole  is  in  a  separate  steel  tank  of 
such  substantial  construction  as  to  be  proof  against  any  explosion 
due  to  the  effect  of  short  circuit. 

The  circuit-breaker  rating  should  be  based  on  the  maximum 
current  which  it  is  to  carry  continuously  without  overheating,  and 
a  breaker. should  therefore  be  selected  which  has  a  capacity  at 
least  equal  to  the  maximum  rating  or  the  one  or  two-hour  overload 
rating  of  the  circuit.  At  the  normal  rated-  load,  current-carrying 
parts  should  not  heat  more  than  30°  C,  above  an  ambient  tem- 
perature of  40**  C,  providing  the  connections  to  the  breaker  do 
Act  heat  to  a  greater  extent.  The  rise  on  tripping  solenoids  and 
accessory  parts  shall  not  exceed  50°  C.  The  dielectric  test  should 
be  2\  times  rated  voltage  plus  2000. 

In  selecting  the  proper  type  of  breaker  to  use  for  a  certain 
case,  it  is  not  enough  that  the  breaker  has  a  sufficient  current- 
carrying  capacity  or  that  it  is  capable  to  withstand  the  operating 
voltage.  The  amount  of  energy  or  kilovolt-amperes  which  the 
switch  may  be  called  upon  to  rupture  under  abnormal  conditions, 
such  as  a  short-circuit,  is  a  very  important  matter  and  deserves 
the  most  careful  attention. 

Based  on  its  rupturing  capacity,  the  rating  of  an  oil  circuit 
breaker  is  necessarily  more  or  less  empirical,  and  is  generally 
determined  by  exhaustive  short-circuit  tests.  It  depends  prin- 
cipally on  the  amount  of  oil  over  the  break  at  the  starting  of  the 
arc,  the  amount  of  space  above  the  oil  for  gas  expansion,  the  shape 
and  strength  of  the  oil  tank  and  its  fastenings  and  on  the  length 
and  rapidity  of  the  contact  movement. 

There  are  many  different  ways  of  rating  oil  circuit  breakers, 
but  it  appears  that  the  most  logical  way  would  be  to  base  the  rup- 
turing capacity  on  the  maximum  "  instantaneous "  kilovolt- 
amperes  which  the  switch  would  be  capable  of  rupturing.  By 
"  instantaneous ''  is  here  meant  the  elimination  of  time-limit 
relays  in  tripping.  The  problem  of  choosing  an  oil  circuit  breaker 
for  a  given  location  would  then  resolve  itself  in  determining  the 
kilovolt-amperes  that  can  be  delivered  on  short-circuit  through  the 
breaker.  This  value  depends  naturally  on  how  quickly  the  oil 
circuit  breaker  opens  and  also  on  the  rate  at  which  the  short- 
circuit  current  dies  down.  Due  to  inertia,  it  is,  of  course,  impos- 
sible for  a  breaker  to  open  instantaneously,  and  consequently  no 
breaker  is  ever  called  on  to  open  the  momentary  short-circuit 


Digitized  by 


Google 


498  ELECTRICAL  EQUIPMENT 

current  that  occurs  during  the  few  first  cycles,  but  it  has  to  be 
strong  mechanically  to  resist  the  magnetic  stresses  set  up  during 
such  a  short-circuit.  Large-capacity  breakers  equipped  with 
"  instantaneous  "  acting  relays  can  be  made  to  open  in  about 
one-quarter  second,  and  the  power  which  has  to  be  broken  under 
such  conditions  averages  under  the  worst  conditions  approxi- 
mately 60  per  cent  of  the  maximum  instantaneous  value.  For 
non-automatic  switches  or  switches  equipped  with  definite  time- 
limit  relays  with  a  setting  over  0.8  second,  the  rupturing  capacity 
corresponds  to  the  sustained  short-circuit  current,  while  for 
switches  with  inverse  time  action  the  condition  approximating 
"instantaneous,"  as  above,  must  be  assiuned.  When  speaking 
of  the  maximiun  instantaneous  value,  the  root-mean-square  value 
is  meant. 

There  is  a  great  variety  of  oil  circuit  breakers  in  the  market 
with  rupturing  capacities  of  several  hundred  thousand  Kv.A. 
As  a  rule,  switches  with  the  higher  rating  will  be  required  near  the 
generating  station,  while  under  some  conditions,  the  added  react- 
ance of  transformers  and  lines  serve  to  reduce  the  value  of  the 
short-circuit  current.  (See  also  section  on  "  Current-limiting 
Reactors.") 

Unfortimately  there  is  some  difference  in  rating  oil  circuit 
breakers,  and  it  is  very  important,  in  any  oil  circuit  breaker  nego- 
tiation, that  the  actual  meaning  of  the  guarantee  is  fully  under- 
stood. So,  for  example,  the  term  "rupturing  capacity"  has 
been  given  two  meanings;  one,  as  indicating  the  rated  Kv.A. 
capacity  in  generators  which  may  be  short-circuited  and  under 
such  conditions  opened  by  the  breaker  in  question;  the  other,  as 
indicating  the  actual  current  which  the  breaker  opens  at  the  time 
of  short-circuit,  this  capacity  generally  being  expressed  in  Kv.A. 
equivalent  to  the  actual  current  opened  at  the  normal  circuit  volt- 
age. Furthermore,  the  term  "  ultimate  breaking  capacity  "  has 
been  used  to  indicate  either  of  the  above  conditions,  and  it  can  be 
seen  inmiediately  to  what  confusion  this  difference  in  the  meaning 
of  the  guaranteed  rating  can  lead.  The  importance  of  a  clear 
understanding  of  just  what  is  meant  cannot  be  over-emphasized. 

Fig.  306  represents  a  type  of  circuit  breaker  which  is  intended 
for  use  in  small  and  moderate-capacity  stations  for  voltages  up 
to  22,000.  It  can  be  mounted  on  the  pipe  frame  supporting  the 
switchboard  panels,  on  framework  remote  from  the  panel,  or  in 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


499 


cells,  depending  on  the  ampere  capacity  or  the  voltage.  It  may 
be  operated  by  hand  from  the  switchboard  by  means  of  operating 
rods  through  a  system  of  bell 
cranks,  or  electrically  by 
means  of  a  solenoid  con- 
trolled from  the  main  switch 
board. 

The  stationary  contacts 
consist  of  copper  fingers 
flared  at  the  tips,  one  ex- 
tending so  as  to  act  as  an 
arcing  tip.  The  movable 
contact  blades  are  wedge- 
shaped,  confining  the  arc  of 
the  blade,  protecting  the 
actual  contact  surfaces  from 
the  damaging  effect  of  the 
arc. 

The  oil  vessel  is  of  heavy 
sheet  metal  hned  with  treat- 
ed laminated  wood.  Multi- 
pole  switches  of  smaller 
capacity  have  all  poles  in  one  tank  with  treated  wooden  barriers 
between  each  pole,  while  for  larger  capacities  one  tank  is  pro- 
vided for  each  pole. 

In  the  more  important  large  capacity  stations  where  it  is  of  the 
utmost  importance  to  prevent  trouble  in  any  one  circuit  or  phase 
being  conmiunicated  to  other  pants  of  a  station  or  system,  the  oil 
circuit  breakers  are  located  in  separate  compartments,  and  in  some 
cases  barriers  isolate  each  phase,  and  even  each  oil  tank  is  separated 
if  additional  safety  factors  are  desired. 

The  oil  circuit  breaker  with  the  highest  rupturing  capacity 
which  has  so  far  been  put  into  service  is  of  the  general  type  shown 
in  Fig.  307  and  its  ultimate  development  with  maximum  isolation 
in  Fig.  308. 

These  switches  are  generally  known  as  type  H,  and  are  made 
for  carrying  very  high  currents  (up  to  4000  amperes),  and  are  most 
generally  used  for  the  ordinary  generator  voltages  up  to  13,200, 
although  they  can  be  obtained  for  voltages  up  to  70,000. 

Each  pole  is  made  up  in  part,  of  two  separate  seamless  steel 


Fig.  306. — Small  and  Moderate-capac- 
ity Oil  Circuit  Breaker.  Remote 
Controlled  and  Mounted  on  Pipe 
Frame  Work. 


Digitized  by 


Google 


500 


ELECTRICAL  EQUIPMENT 


vessels,  in  each  of  which  the  circuit  is  broken  under  oil.  There 
are  thus  two  breaks  per  pole,  the  general  construction  of  the  oil 
vessel  being  apparent  from  Fig.  309.    Each  contact  consists  of  a 


Fig.  307. — High-capacity    Motor-operated    Oil    Circuit  Breaker  with  Two 
Tanks  for  Each  Phase,  and  Phases  Isolated  from  Each  Other. 

metal  rod  which  bears  against  the  inner  surface  of  four  longi- 
tudinal segments  of  a  cylinder  secured  in  position  by  helical 
springs.    This  arrangement  insures  a  heavy  and  imiform  contact 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


501 


pressure,  and  automatically  compensates  for  any  wear  of  the 
surface  of  either  the  stationary  contact  segments  or  the  contact  rod. 
When  the  arc  is  ruptured,  whatever  burning  results  takes  place 
on  the  bell  mouth  of  the  stationary  contact  segments  or  on  the 


Lik. 


Fig.  308.— High-capacity  Oil  Circuit  Breaker  with  Tanks  Arranged  in  Tandem 
and  Separated  by  Barriers. 

rounded  end  of  the  movable  contact  rods,  and  in  no  case  causes 
damage  to  the  working  contact  smfaces.  The  contacts  are  self- 
aligning  and  easily  renewable. 

For  higher-current   capacities,   however,  additional  primary 


Digitized  by 


Google 


502 


ELECTRICAL  EQUIPMENT 


contacts  are  provided.  These  carry  the  greater  part  of  the 
current  flowing  through  the  switch  and  obviate  the  necessity  of 
having  large  currents  to  pass  within  the  oil  vessel.    These  nuiin 

contacts  are  outside  the  oil  ves- 
sel but  inside  the  fireproof  com- 
partments of  the  ceU,  and  so 
placed  as  to  secure  the  maxi- 
mum radiation.  In  opening  the 
breaker  they  break  contact  be- 
fore the  contact  rod,  which  opens 
the  circuit  and  ruptures  the  con- 
sequent arc  under  oil.  The  main 
contacts  are  of  the  laminated 
brush  type  or  of  the  ordinary 
wedge-shaped  finger  type. 

To  prevent  throwing   oil,  a 
baffle  is  used  in  each  oil  vessel. 
By  the  baffle,  the  movement  im- 
parted to  the  oil  by  the  expan- 
sion of  the  gases  formed  by  the 
arc  when  the  circuit  is  opened 
imder  load  is   checked  and  di- 
verted in  such  a  manner  as  to 
allow  the  gases  to  separate  from 
the  oil  and  escape  through  the 
vent  in  the  cover  of  the  oil  ves- 
sel, while  the  oil  itself  is  forced 
back  into  the  region  of  the  break- 
ing   arc    under    pressure,    thus 
shortening  the  time  of  breaking 
the  arc,  confining  the  disturbance 
or  explosive  effect  on  short  cir- 
cuit   and     practically    eliminate 
flashes  due  to  hot  gases  and  the 
oil   from   the  oil   vessels.     The 
movement  of  the  oil  away  from 
and  towards  the  center  of  the  oil 
vessel  on  the  breaking  of  the  cir- 
cuit and  also  the  movement  of 
the  gases,  are  indicated  in  Fig.  309.     The  oil  loses  its  velocity 


Fi«.  3n9.— Oil  Vessel  for  High- 
capacity  Oil  Circuit  Breaker 
Showing  Oil  Baffle  Arrangement 
and  Contacts. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  503 

before  the  cover  of  the  oil  vessel  is  reached  and,  therefore,  its 
tendency  to  be  thrown  out  is  reduced. 

For  each  vessel  there  are  two  insulating  bushings.  The  upper 
one  is  clamped  to  the  oil  vessel  cover  and  serves  as  guide  to  the 
movable  contact  rod  and  also  insulates  the  rod  from  the  oil  vessel. 
The  bottom  bushing  is  fastened  to  the  base  supporting  the  oil 
vessel  by  means  of  a  metal  clamp  which  holds  it  in  proper  align- 
ment. Generally  these  switches  are  bottom  connected  but  can 
be  obtained  for  combination  bottom  and  back  connection. 

The  operating  mechanism  is  located  above  the  cell  structure 
and  connected  to  the  contacts  by  operating  rods  of  specially 
treated  wood.  Direct-current  motor  drive  is  reconmiended  for 
use  whenever  possible,  and  when  no  other  suitable  source  of  direct 
current  is  available,  a  storage  battery  with  motor  generator  for 
charging  may  be  installed.  (See  **  Oil  Circuit  Breaker  Bat- 
teries.") Alternating-current  motors  can  be  furnished  if  for  any 
reason  direct-current  operation  is  not  practicable.  It  should  be 
borne  in  mind,  however,  that  with  alternating-current  motor 
operation,  a  constant  source  of  alternating  current  should  be 
available  unless  it  is  agreeable  to  close  by  hand  some  oil  circuit 
breaker,  which  would  provide  the  necessary  operating  current. 

This  type  of  breaker  is,  of  course,  always  controlled  by  the 
control  switch  on  the  main  switchboard.  It  may  be  non-auto- 
matic or  automatic,  the  latter  feature  being  obtained  by  circuit- 
closing  relays,  with  the  relay  contacts  connected  in  multiple  with 
the  contacts  of  the  opening  button  of  the  control  switch.  When 
the  relsLys  operate,  they  close  a  direct-current  auxiliary  circuit 
through  the  tripping  magnet  of  the  oil  circuit  breaker  and  it 
immediately  opens. 

Fig.  310  illustrates  Si,  line  of  tank-type  oil  circuit  breakers  which 
is  used  for  stations  of  moderate  and  large  capacity  for  voltages 
from  35,000  to  110,000.  Indoor  and  outdoor  breakers  are  prac- 
tically similar.  The  only  difference  consists  of  the  addition  to  the 
indoor  breaker  of  a  few  parts  to  enable  it  to  be  serviceable  both 
from  a  mechanical  and  an  electrical  standpoint  under  all  weather 
conditions. 

A  noteworthy  advance  in  these  breakers  consists  of  mounting 
them  on  framework  and  in  the  handling  of  the  tanks  by  a  tank- 
lifting  device.  Such  a  construction,  however,  is  limited  to 
switches  below  110,000  volts.    The  Ufter  consists  of  a  detachable 


Digitized  by 


Google 


504 


ELECTRICAL  EQUIPMENT 


frame  equipped  with  shaft,  handle  worm  gear  and  winding  and 
unwinding  drums.  The  advantage  of  this  equipment  is  that  it 
allows  a  tank  to  be  removed  or  placed  in  position  without  diffi- 


FiG.  310.    Typical  35,000- volt  Oil  Circuit  Breaker  of  the  Tank-type  Cto- 
struction  Mounted  on  Framework. 


culty.  The  device  is  readily  detachable  and  can  be  moved  by  one 
man  from  one  breaker  to  another.  These  breakers  are  always 
top-connected  and  self-contained.     They  are  made  for  either 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  605 

automatic  or  non-automatic  operation,  and  may  be  closed  by  hand 
or  solenoids. 

The  automatic  breakers  are  tripped  under  overload  by  series 
trip  coils  or  secondary  relays,  the  latter  method  being  almost 
entirely  used  in  modern  installations.  The  secondary  tripping 
mechajiism  consists  of  a  system  of  toggles  and  latches  so  con- 
structed that  only  a  slight  pressure  is  needed  to  open  the  breaker. 
The  tripping  coils  may  be  energized  from  standard  current  trans- 
formers, from  bushing-type  current  transformers  or  from  a  source 
of  constant  potential,  the  current  adjustment  being  accom- 
plished by  varying  the  position  of  the  plunger  in  the  trip  coil  and 
the  inverse  time  relay  by  a  dashpot.     (See  also  "  Relays.") 

The  operating  mechanism  is  secured  to  the  cast-iron  cover  of 
the  heavy  welded  sheet-steel  tank.  There  are  two  fixed  contacts 
.  in  each  switch  element  between  which  one  phase  of  the  circuit  is 
made  and  broken  by  a  horizontal  contact  blade.  Each  contact 
blade  is  connected  to  the  operating  mechanism  by  a  specially 
treated,  hard  wooden  rod  which  passes  through  the  cover  of  the 
switch  in  an  insulating  bushing.  The  stationary  contacts  con- 
sist of  widely  flared  fingers  and  long  arcing  tips  which  also  act  as  a 
guide  to  the  entering  blade.  The  movable  contacts  are  wedge- 
shaped,  which  confines  the  arc  to  the  top  edge  of  the  blade  and 
the  flared  portion  of  the  finger  tips.  The  contacts  are  always 
smooth  and  bright  due  to  the  sliding  effect  which  they  are  sub- 
jected to  on  opening  and  closing,  and  the  arrangement  of  the 
burning  tips. 

The  design  of  the  bushings  depends  entirely  on  the  voltage  for 
which  the  switch  is  intended.  For  the  35,000-volt  size,  they  are 
made  in  one  piece  of  wet  porcelain  and  extend  from  the  terminal 
to  the  contacts  below  the  oil.  For  higher  voltages  each  bushing 
consists  of  two  porcelain  sections,  an  upper  and  a  lower,  joined 
together  by  heavy  supporting  iron  flanges,  which  also  serve  as  a 
means  of  attaching  to  the  breaker  or  for  housing  the  bushing  trans- 
formers, where  such  are  required.  For  moderate  voltages  the 
contact  rod  which  passes  through  the  bushing  is  simply  insulated 
by  an  insulating  material  and  the  bushing  filled  with  an  insulating 
compound  of  high  dielectric  strength.  For  higher  voltages,  70,000 
and  above,  the  bushings  generally  contain  a  number  of  cylinders 
of  insulating  material  concentric  with  the  conducting  tube,  the 
whole  being  filled  with  compound.    These  cylinders  in  connection 


Digitized  by 


Google 


506 


ELECTRICAL  EQUIPMENT 


Fia.  311.— 135,000- volt  Oil  Circuit  Breaker.    Front  Unit  Supported  on  Frame- 
work to  Show  Interior  Construction. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  507 

with  equalizing  shields  serve  to  evenly  distribute  the  potential 
gradient  of  the  bushing. 

For  each  pole  there  is  a  separate  oil  tank  provided  with  gas 
vents  and  oil  gauges.  Drain-cocks  may  also  be  obtained  if  desired 
and  are  to  be  recommended  for  all  large  floor-mounted  switches. 

Pig.  311  shows  a  large-capacity  tank-type  oil  circuit  breaker  for 
indoor  services  at  135,000  volts.  It  is  almost  identical  to  the 
switches  previously  described,  the  main  points  of  construction 
being  apparent  from  the  illustration.  At  the  upper  end  of  each 
bushing  is  a  combined  expansion  chamber  and  gauge  glass  which 
affords  opportunity  to  view  at  all  times  the  insulating  compound 
with  which  the  bushings  are  filled.  The  terminal  on  the  upper 
end  of  a  bushing  is  of  such  shape  that  it  can  be  used  for  attaching 
a  crane  hook  to  lift  the  bushing  out  of  or  replace  it  in  the  breaker. 

High-grade  mineral  oil  should  be  used  for  all  oil  circuit  breakers. 
It  should  have  a  high  flash  and  ignition  point  as  well  as  high 
resistance  to  carbonization. 

Relays.  Relays  may  be  defined  as  protective  devices  used  in 
connection  with  circuit  breakers  to  disconnect  any  part  or  section 
of  a  system  on  which  a  fault  occurs  but  leave  the  rest  of  the  system 
in  operation  without  being  further  affected  by  the  faulty  section. 
In  general,  a  relay  consists  of,  first,  a  coil  or  system  of  coils  con- 
nected either  directly  in  series  or  in  parallel  with  the  circuit  con- 
trolled or  to  secondaries  of  current  or  potential  transformers,  the 
current  and  potential  coils  then  being  wound  for  a  low  value, 
usually  five  amperes  for  the  current  coil  and  110  volts  for  the 
potential  coil,  although  other  values  might  be  used  if  desired. 
In  the  former  case  it  is  termed  a  primary  or  series  relay  and  in  the 
latter  a  secondary  relay.  Second,  a  relay  consists  of  a  movable 
part  such  as  a  plunger  or  a  revolving  disk,  etc.,  whose  travel  is 
controlled  by  the  relay  coils,  and  third  of  a  contact  device  which 
is  actuated  by  the  movable  part  and  which  controls  the  operating 
circuit,  such,  for  instance,  as  the  trip  coil  of  the  circuit  breaker 
to  which  it  is  connected.  Although  smaller  circuit  breakers 
may  be  opened  by  the  relay  core  striking  the  tripping  latch 
directly,  larger  breakers  are  usually  provided  with  separate 
tripping  coils,  the  cores  of  which,  when  completing  their  travel 
strike  the  latch  and  release  the  switch. 

The  impedance  of  a  relay  coil  is  relatively  small  compared  to 
that  of  an  oil  circuit  breaker  trip  coU,  and  if  a  nimiber  of  instru- 


Digitized  by 


Google 


508 


ELECTRICAL  EQUIPMENT 


ments  and  meters  are  connected  to  a  current  transformer  their 
accuracies  are  naturally  affected  by  the  total  load  imposed  on  the 
transformer  secondary,  decreasing  rapidly  as  the  load  rises  above 
a  certain  point.  Some  oil  circuit  breaker  trip  coils  have  a  high 
impedance,  and  meter  combinations  requiring  considerable  accu- 
racy consequently  should  not  be  used  in  series  with  them.  By  inter- 
posing a  relay,  which  cuts  out  the  trip  coils  except  at  the  moment 
of  trouble,  the  total  load  can  be  very  materially  reduced.  The 
relay  therefore  simply  serves  to  control  the  tripping  circuit  and 
may  be  either  circuit-closing  or  circuit-opening.  In  the  former 
case  (Fig.  312),  the  relay  contacts  are  normally  open  and  the  trip 
coils  dead,  but  at  the  moment  of  operation  contact  is  made,  thus 


Oil 
Switch 


Mi 


<>  ?  9 


O 


TVfpOon 

r 


I 


'"W"  Ground 


tioiiir 


Terminal 
i      Board 


G«ncnitOr 


Hi 


6 


I 


[i 


Fig.  312. — Connections  of  Circuit- 
closing  Relay. 


Fig.  313. — Connections  of  Circuit- 
opening  Relay. 


completing  the  circuit  and  energizing  the  trip  coil,  which  in  turn 
causes  the  switch  to  be  released.  In  the  latter  case  (Fig.  313) 
the  relay  contacts  are  normally  closed  and  the  trip  coils  de-ener- 
gized, because  the  current  will  then  take  the  path  of  least  resist- 
ance through  the  contact  blocks  and  not  through  the  comparatively 
high  impedance  path  through  the  trip  coil  winding.  When  a  shortr 
circuit  occurs  on  the  main  circuit,  the  contacts  open,  and  force  the 
current  through  the  trip  coils,  which  then  operate  and  open  the 
switch.  As  noted  from  the  diagrams,  circuit-closing  relays  require 
a  separate  source  of  power,  preferably  direct  current,  for  operat- 
ing the  trip  coil,  while  for  the  open-circuit  type  the  tripping  cur- 
rent is  obtained  from  the  secondary  of  the  current  transfcrmer. 
Circuit-closing  relays,  are,  however,  almost  exclusively  employed 
in  connection  with  the  circuit  breakers  used  on  large  power  systems 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


509 


and  circuit  opening  relays  only  in  those  cases  where  direct  current 
is  not  available.  On  account  of  the  heavy  secondary  currents 
which  are  liable  to  flow  on  severe  short-circuits  and  due  to  the 
comparatively  high  impedance  of  the  trip  coil,  which  may  tend 
to  hold  up  the  voltage,  a  considerable  arc  is  liable  to  be  set  up  when 
the  contacts  are  opened,  and  there  is  therefore  a  limit  above  which 
it  is  not  safe  to  use  circuit-opening  relays.  As  a  rule  they  should 
not  be  used  when  the  short-circuit  current  exceeds  ten  times  the 
normal  rating  of  the  current  transformer. 

There  are  a  large  number  of  different  types  of  relays,  but  only 
a  few  of  those  in  ordinary  use  on  power  transmission  systems  will 
be  considered.  Neither  will  any  detailed  description  of  their  con- 
struction be  given  as  changes  and  improvements  are  made  so  fre- 
quently that  this  would  soon  be  obsolete.  It  will  therefore  be 
the  aim  in  the  following  to  merely  deal  with  their  fundamental 
principles  and  characteristics. 

Overload  Relays.  These  may  be  instantaneous,  definite  time 
limit  and  inverse  time  limit.    With  instantaneous  relays,  the 


0«*isr 


Bkffik 


Fig.  314. — Instantaneous  Overload  Plunger- type  Relay. 


contact  device  will  operate  immediately  and  close  the  tripping 
circuit  of  the  breaker  when  the  abnormal  conditions  which  the 


Digitized  by 


Google 


610 


ELECTRICAL  EQUIPMENT 


relay  is  to  take  care  of  make  their  appearance  and  start  the  moving 
part  of  the  relay.  With  definite  time-limit  relays  there  is,  as  the 
name  implies,  a  definite  time  delay  imposed  between  these  two 
moments,  independent  of  the  magnitude  of  the  disturbance,  and 
the  time  hmit  therefore  becomes  practicaQy  constant  for  any 
given  setting.  With  inverse  time-Umit  relays  the  time  delay  is 
inversely  proportional  to  the  magnitude  of  the  disturbance,  so 

that  with  a  heavy  short-circuit  it 
will  be  practically  instantaneous 
for  any  time  setting,  while  on  a 
Ught  overload  the  time  may  be 
several  seconds,  depending  on  the 
setting. 

For  instantaneous  overload 
relays  the  plunger  tj'pe  (Fig.  314) 
is  considered  the  best.  It  simply 
consists  of  a  core  or  plunger 
which  is  movable  within  a  sole- 
noid. When  a  suflScient  amount 
of  current  is  passed  through  the 
winding  the  core  is  pulled  up  and 
causes  the  cone-shape9  disc  at 
the  top  to  bridge  the  gap  between 
the  contacts.  The  position  of 
the  plunger  with  respect  to  the 
coil  is  adjustable,  the  lower  its 
position  the  more  current  is  re- 
quired   to   pull   it    into  dosing 

Fig.  315.-Double.poleBeUow8  Type  P^^^^^^^  ^^^  ^^  adjusting  its 
Inverse  Time  Limit  Overload  position  it  may  be  set  to  take  any 
Relay.  predetermined   strength  of  cur- 

rent within  the  range  of  the  coil. 
Inverse  time-limit  relays  may  be  either  of  the  bellows  tynpe 
or  the  induction  type.  The  former  (Fig.  315)  is  similar  to  the 
instantaneous  type  to  which  a  compressible  leather  bellows  has 
been  interposed  between  the  moving  part  and  the  contact  device. 
When  the  relay  is  not  operating,  the  bellows  is  fully  extended  and 
the  moving  core  presses  against  the  same  and  tends  to  force  the 
air  through  an  aperture.  The  air  must  be  driven  out  of  the  bel- 
lowes  and  the  bellows  compressed  completely  before  contact  can  be 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


511 


made.  The  rapidity  with  which  the  air  escapes,  that  is,  the  time 
intervening  between  the  start  of  the  moving  part  and  the  com- 
pletion of  contact,  is  a  function  of  the  power  behind  the  com- 
pression moving  part,  which  in  turn  depends  on  the  magnitude 
of  the  electrical  force  actuating  the  relay  coil.  The  size  of  the  hole 
through  which  the  air  escapes  can  be  varied  so  that  different  time 
elements  may  be  obtained  for  disturbing  forces  of  the  same  mag- 
nitude, and  different  time  curves  for  the  same  range  of  disturb^ 
ance. 

In  the  induction  type  overload  relay  (Fig.  316),  the  actuating 
forces  are  due  to  the  interaction  of  induced  currents  in  a  moving 
metal  element  with  the  induc- 
ing magnetic  field.  A  lami- 
nated iron  core  is  surrounded 
by  one  or  more  windings,  and 
in  the  air  gap  of  the  core  is 
pivoted  the  moving  element, 
usually  a  Ught  aluminum 
disc.  When  current  is  passed 
through  the  main  windings, 
eddy  currents  are  induced  in 
the  disc  which  tends  to  rotate 
and  close  the  contacts  after  a 
predetermined  angle  of  motion. 
The  retarding  force  is  pro- 
duced by  having  the  same  disc 
pass  between  the  poles  of 
permanent  magnets,  in  which 
case  the  eddy  currents  induced  by  these  will  retard  the  motion. 

The  relays  are  designed  for  use  in  the  secondary  circuit 
of  current  transformers,  and  the  normal  rating,  or  continuous 
current-carrying  capacity,  is  5  amperes.  Taps  are  provided  in 
the  relay  winding,  and  by  inserting  a  metal  plug  in  a  current 
tap  plate,  settings  4,  5,  6,  8  and  10  amperes  may  be  obtained, 
these  figiu*es  representing  the  lowest  current  values  required  to 
close  the  relay  contacts.  Any  tap  setting,  multiphed  by  the 
ratio  of  the  current  transformers,  gives  the  corresponding  primary 
or  line  current. 

A  time-current  index  plate  is  provided  as  a  guide  for  deter- 
mining the  settings  of  the  relay,  and  the  current  values  are 


Fig. 


316. — Induction   Type   Overload 
Time  Limit  Relay. 


Digitized  by 


Google 


512 


ELECTRICAL  EQUIPMENT 


indicated  by  the  figures  1.5,  2,  3,  5,  etc.,  in  the  "  Times  current 
tap  setting "  column.  These  figures  can  be  translated  into 
amperes  by  multiplying  them  by  the  current  tap  setting  which 
is  to  be  used.  Time  settings  are  made  by  a  lever  which  changes 
the  length  of  travel  of  the  disc,  the  time  scale  being  at  the  bottom 
of  the  index  plate.  Therefore,  with  the  time  lever  set  over  a 
definite  graduation  mark,  the  values  given  in  the  correspond- 
ingly marked  column  are  the  approximate  time  delays,  in  seconds, 
which  will  be  obtained  at  the  current  values  opposite  in  the 


0  1  i  3  4  5  6  7  8  0  10  20  80  40 

Current,  in  Multiples  of  Current  Tap  Setting 
Fig.  317. — Induction  Type  Time  Limit  Relay  Characteristics. 


"  times  current  tap  setting "  columns.  In  general,  the  time 
delay  values  should  be  chosen  at  a  current  value  approximating 
the  short-circuit  current  of  the  line,  and  the  proper  setting  of 
the  time  lever  for  a  given  time  delay  may  be  determined  by 
referring  to  the  table  on  the  time  current  index  plate.  First 
determine  which  factor  in  the  "times  current  tap  setting" 
coluLmn  represents  the  current  at  which  this  time  delay  is 
desired.  The  position  of  the  time  lever  can  then  be  found  by 
an  inspection  of  the  row  of  time  delay  values  opposite  this 
factor. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


513 


Fig.  317  shows  a  number  of  time-current  characteristic  curves 
of  this  relay  and  the  constantly  decreasing  time  as  the  current 
increases  should  be  noted.  The  curves  consist  of  an  inverse  time 
portion  up  to  approximately  20  times  the  minimum  current,  set- 
ting, blended  into  a  definite  time  portion  instead  of  converging. 

The  above  type  of  relay  may  also  be  used  where  a  definite 
time  action  is  required.  Otherwise  a  bellows  type  relay  may 
be  used  in  which  the  moving  part  starts  immediately  when  the 
tripping  value  is  reached  and  compresses  a  spring,  and  this  in  turn 
actuates  the  diaphragm  and  the  contact  device,  the  time  required 
by  the  spring  for  this  operation  being  entirely  independent  of  the 
magnitude  of  the  disturbance,  but  dependent  only  on  the  stored- 
up  energy  of  the  spring  and  the  setting  of  the  air-escape  hole.     To 


ii 


[^ 


rVft 


u 


Tbree-Phase 

Inductive  Load  In 

Power  StatioD 


Three-Phase 
Ungrounded 


Tbree-Pbaae 
Grounded  Neutral 


Fig.   318. — Circuit-closing    Overload    Relay   Connections   Showing   Use   of 
Single-,  Two-  or  Three-pole  Units. 

obviate  inaccuracies  due  to  slow  closing  it  is  advisable  to  combine 
this  relay  with  an  instantaneous  one.  No  mechanical  action  would 
then  be  exerted  on  the  spring  until  the  disturbance  had  risen  to  a 
value  sufficiently  large  to  operate  the  instaneous  relay  and  to 
throw  the  definite  time  limit  relay  into  circuit.  Where  direct 
current  is  available,  the  coil  of  the  instantaneous  relay  should  be 
connected  to  the  main  A.C.  circuit  and  the  definite  time  limit 
relay,  having  a  D.C.  potential  coil,  connected  to  the  contact  device 
of  the  instantaneous  relay,  and  tripping  in  turn  the  circuit-dis- 
connecting device.  Where  no  direct  current  is  available,  a  cir- 
cuit-opening instantaneous  relay  in  combination  with  a  definite 
time  limit  relay  with  A.C.  coil  is  required,  so  that  the  definite  time- 
limit  relay  is  not  connected  in  until  the  disturbance  has  reached 
a  value  sufficiently  high  to  operate  the  instantaneous  relay. 


Digitized  by 


Google 


514  ELECTRICAL  EQUIPMENT 

Bellows-type  relays  axe  very  rugged  and  are  extensively  used 
for  ordinary  service,  while  the  induction  relay  is  desirable  where 
extreme  accuracy  is  required  such  as  to  insure  selective  switch  ac- 
tions on  compUcated  networks. 

Overload  relays  are  usually  made  single-pole,  but  one,  two  or 
three  relays  may  be  combined  as  the  conditions  may  demand, 
the  usual  practice  being  shown  by  the  connection  diagrams  in 
Fig.  318.  Single-pole  relays  may  be  used  on  single-phase  and 
balanced  three-phase  circuits;  double-pole  relays  on  ungrounded 
three-phase  circuits  and  two-phase  circuits  which  are  not  inter- 
connected; triple-pole  relays  on  three-phase  grounded  neutral 
and  interconnected  two-phase  circuits. 

Reverse  Power  Relays.  These  operate  on  a  reversal  of  the 
energy  in  the  circuit  to  which  they  are  connected.  They  may 
be  either  of  the  dynamometer  type  or  the  induction  type. 

The  dynamometer  type  (Fig.  319)  consists  of  a  potential  coil 
pivoted  in  the  center  of  a  current  coil  in  such  a  manner  as  to 
obtain  dynamometer  action,  the  two  coils  being  mounted  in  a 
magnet  frame.  The  pivot  which  supports  the  potential  coil  also 
supports  the  movable  contact,  and  when  the  flow  of  power  is  in 
normal  direction  or  at  no  load,  the  contact  lever  is  held  against 
a  stop  by  a  spring.  Upon  reversal  of  power  the  potential  coil 
tends  to  turn  and  throws  the  contact  lever  against  a  station- 
ary contact,  completing  the  tripping  circuit  of  the  oil  circuit 
breaker. 

The  dynamometer  type  of  relay  is  generally  built  in  single- 
pole  imits  which  may  be  combined  in  the  same  manner  as  overload 
relays,  for  the  protection  of  polyphase  circuits,  as  previously 
described.  Figs  320  and  321  show  the  connections  for  a  relay 
of  this  type  as  used  on  three-phase  nongrounded  and  grounded 
circuits.  Three  potential  transformers  are  shown  for  the  latter 
case,  but  two  may  be  used  if  the  volt-ampere  load  permits. 

Reverse  power  relays  are  in  themselves  always  instantaneous 
and  for  time  action  they  must  be  combined  with  overload  relays 
with  such  features.  An  overload  relay  is  always  recommended 
for  the  induction  type  reverse  power  relay,  even  for  instantaneous 
isLction  due  to  its  sensitiveness.  This  overload  relay,  although  not 
necessary,  is  nevertheless  also  recommended  with  the  dynamom- 
eter type.  When  used  in  connection  with  overload  relays  the  con- 
tacts of  both  relays  are  connected  in  series  so  that  both  must 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


515 


operate  before  the  breaker  will  be  tripped.  Any  t3rpe  of  overload 
relay  can  be  used,  although  the  plimger  type  is  recommended  when 
instantaneous  action  is  desired.  Otherwise  the  induction  t3rpe 
may  equally  well  be  used. 

The  induction  type  reverse  power  relay  is  based  on  the  prin- 


PiG.  319. — Single-pole  Dynamometer-type  Reveree-power  Relay. 

ciple  of  the  wattmeter,  in  which  a  disc  or  rotating  element  is  actu- 
ated by  both  current  and  voltage  windings.  The  torque  generated 
is  proportional  to  the  instantaneous  products  of  the  ciurent  and 
voltage,  i.e.,  the  watts. 

The  relay  shown  in  Fig.  322  is  the  polyphase  type  and  the 
arrangement  of  the  driving  elements  on  a  common  shaft  has  sev- 
eral advantages.    There  are  three  separate  driving  elements,  each 


Digitized  by 


Google 


616 


ELECTRICAL  EQUIPMENT 


TemlBftl 


Fig.    320. — Connections     for     Dynamometer-type    Reverse-power    Relay. 
Three-phase  Ungrounded  Circuit. 


+  - 


Fig.   321. — Connections    for     Dynamomet>er-type    Reverse-power    Relay. 
Three-phase  Circuit  with  Grounded  Neutral. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


517 


having  a  current  coil  and  a  potential  coil  used  for  both  quarter- 
SLnd  three-phase  circuits.  The  third  element  is  required  for  delta 
or  ungrounded  Y  circuits  in  order  that  each  phase  may  be  properlj' 
represented  in  every  short-circuit.  If  two  elements  were  used 
many  single-phase  troubles 
would  involve  only  one  of  these 
elements  and  the  benefit  of  poly- 
phase action  would  be  lost. 
Although  only  one  element  may 
be  involved  in  case  of  a  groimd 
on  a  grounded  Y  circuit,  the 
voltage  triangle  will  not  have 
become  so  badly  distorted  as 
when  a  single-phase  line  to  line 
short  exists.  For  delta  or  un- 
groimded  Y  circuits  two  cur- 
rent and  two  potential  trans- 
formers are  sufficient.  The 
third  ciu'rent  coil  carries  the  re- 
sultant current  of  the  two  cur- 
rent transformers  and  the  third 
potential  coil  is  connected  across 
the  open  delta  of  the  two  poten- 
tial transformers.  These  elements  all  operate  through  one  shaft 
to  control  one  set  of  contacts.  In  this  three-element  relay,  two 
discs  are  used,  the  upper  one  of  which  is  driven  by  one  element 
and  the  lower  by  two  elements,  one  in  front  and  one  in  back. 

The  polyphase  construction  makes  the  action  of  the  relay  more 
reliable  than  could  be  obtained  by  means  of  three  single-phase 
relays  because  of  the  fact  that  any  incorrect  tendency  on  the  part 
of  one  phase  is  balanced  by  a  similar  but  opposite  incorrect  ten- 
dency on  some  other  phase.  The  incorrect  tendencies  being 
balanced  out,  the  true  net  power  direction  will  not  be  over- 
powered. 

The  polyphase  relay  should  not  be  used  on  systems  having  the 
neutral  grounded,  except  after  proper  investigation,  imless  two 
or  more  parallel  lines  are  involved  and  the  relays  are  inter- 
connected in  a  balanced  group.  In  such  case  the  power  currents 
are  balanced  out  and  the  fault  current  controls  the  operation  of  the 
relay. 


Fig.  322. — Polyphase  Induction-type 
Reverse-power  Relay.  Cover  and 
Register  Removed. 


Digitized  by 


Google 


518 


ELECTRICAL  EQUIPMENT 


Figs.  323  and  324  give  the  connections  for  this  type  of  relay 
both  for  ungrounded  and  grounded  three-phase  circuits.  Three 
potential  transformers  to  be  used  in  the  latter  case  if  thevoltr 
ampere  load  is  too  great  for  only  two. 

Interconnected  Reverse  Power  Relays.    For  two  or  more  parallel 


Fig.  323. — ^Connections  for  Poljrphase  Induction-type  Reverse-power  Relay 
for  Ungrounded  Three-phase  Circuits. 


tie  lines,  over  which  energy  may  normally  be  fed  in  either  direction, 
reverse  power  relays  with  interconnected  current  coils  may  be 
used  at  each  end  of  the  tie  Unes.  The  interconnection  of  the  cur- 
rent coils  is  such  that  the  influence  of  each  circuit  on  its  relay  will 
be  completely  overcome  by  the  other  circuit  so  long  as  conditions 
are  normal.     If  a  short  should  occur  in  one  line,  the  unbalanced 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


519 


condition  will  result  in  the  isolation  of  that  line  without  affecting 
any  other. 

In  the  diagram  (Fig.  325),  the  solid  arrows  indicate  the  rela- 
tive directions  and  intensities  of  the  energies  in  the  various  parts 
of  such  parallel  lines  when  in  normal  operation  and  with  power 


Fig.  324. — Connections  for  Polyphase  Induction-type  Reverse-power  Relay 
for  Grounded  Three-phase  Circuiis. 

being  fed  from  Station  A  to  Station  B.  Should  power  be  reversed 
and  fed  from  B  to  A,  then  all  solid  arrows  would  be  inverted.  In 
either  case  it  will  be  noted  that  the  current  coils  of  all  relays  oppose 
each  other.  There  will  be  no  tendency  to  operate  under  these 
conditions  no  matter  how  much  current  may  be  carried  by  the  tie 
line. 


Digitized  by 


Google 


520 


ELECTRICAL  EQXnPMENT 


Consider^  for  example,  that  one  of  the  two  lines  is  shorted  at 
S,  near  Station  A.  The  dotted  arrows  then  indicate  the  changes 
that  take  place.  Power  flows  out  from  Station  B  over  both  lines. 
The  weaker  influence  of  line  No.  1  tends  to  prevent  any  action 
of  these  relays  but  it  may  be  sufficiently  overpowered  by  the  heav- 
ier current  in  line  No.  2,  in  which  case  the  relay  26  wiD  operate. 


TieUae'l 


Diiccfioa  of  OBfTcat 
10  Operate  Bahy 


StaUon^B" 

Fig.  325.— iSimplified  Connection  Diagram  of  Interconnected  Reverse-power 

Relays. 


At  the  same  time  any  force  exerted  in  the  relay  16  will  simply 
oppose  the  closing  of  its  contacts.  The  same  is  true  of  relay  la. 
Consequently  neither  oil  switch  of  line  No.  1  will  be  disturbed. 

The  effect  in  relay  2a,  however,  is  very  different.  Here  the 
currents  are  both  in  the  proper  direction  to  operate  the  relay. 
This  relay,  therefore,  trips  its  oil  switch  inunediately,  and,  return- 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


521 


ing  to  relay  2b,  it  will  be  seen  that  the  opening  of  oil  switch  2a 
will  have  resulted  in  the  reversal  of  the  current  in  line  No.  1.  If 
the  relay  26  has  not  operated  previously,  it  cannot  fail  to  do  so 
now.  Had  the  short-circuit  occurred  at  some  othei-  point,  the 
energy  intensities  and  directions,  and,  consequently,  the  order  of 
the  operations,  would  have  been  somewhat  changed  from  those 
outlined  above,  but,  in  any  event,  the  final  outcome  would  have 
been  the  isolation  of  the  injured  line 
without  afiFecting  its  companion. 

It  may  be  observed  that  with  line 
No.  2  cut  off,  the  counteracting  in- 
fluence in  the  relays  of  line  No.  1  is 
removed.  Under  these  conditions  a 
short-circuit  outside  of  the  tie  line 
might  result  in  the  opening  of  the  one 
remaining  circuit.  This  difficulty 
may  be  overcome  by  the  use  of  auxili- 
ary switches  connected  so  as  to  render 
the  second  line  nonautomatic  follow- 
ing the  opening  of  the  oil  switch  in 
the  faulty  line,  or  better  still,  to 
automatically  insert  instead,  time 
Umit  overload  relays. 

Balanced  Relays.  These  are  in- 
tended for  protecting  parallel  cir- 
cuits against  faults  which  would 
materially  unbalance  the  currents  in 
these  parallel  lines.  In  the  case  of 
parallel  outgoing  lines,  when  a  short- 
circuit  occurs  on  one  line,  the  current 
in  that  line  will  become  greater  than 
in  the  others,  and  by  reason  of  this  difference  the  circuit  breaker 
of  that  line  will  be  opened.  So  long  as  no  fault  exists  on  any 
line,  no  relay  will  tend  to  trip,  therefore,  no  amount  of  balanced 
overload  on  the  lines  would  open  any  circuit  breaker.  Balanced 
rela3rs  operate  on  current  alone,  and  should  be  used  on  the 
power  end  of  the  circuits  only. 

Splitrconductor  Relays,  This  system  consists  in  splitting  each 
conductor  into  two  parts  and  using  a  relay  which  operates  when- 
ever the  current  in  the  two  halves  becomes  unbalanced.     The 


■Tripping 
'  Battery 


Fig.    325a.  —  Split-conductor 
Method  of  Relay  Protection. 


Digitized  by 


Google 


522 


ELECTRICAI.  EQUIPMENT 


diagram  (Fig.  325a),  illustrates  the  connections  for  one  conduc- 
tor. It  involves  a  standard  overload  relay  but  a  special  current 
transformer.  This  has  three  windings;  two  primary  to  which  the 
two  halves  of  the  split  conductor  are  connected,  and  one  secondary 
connected  to  the  relay  which  controls  the  circuit  breaker  trip  coil. 
Under  normal  operation  the  current  divides  equally  between  the 
two  parallel  paths  and  in  each  transformer  the  magnetizing  efifect 
of  the  two  primary  coils  are  equal  and  opposite.  The  transformer, 
therefore,  oflFers  no  impedance  to  the  current  flow  and  the  sec- 
ondary windings  and  relays  are  unaffected.  If  a  fault  develops 
in  one  of  the  two  parallel  conductors,  however,  it  is  evident  that 
the  balance  between  the  two  primary  transformer  windings  will 
be  upset,  thus  producing  a  magnetizing  effect  on  the  secondai^' 
windings,  exciting  the  relays  and  tripping  the  circuit  breakers. 

Differential  Relays.    These  are  intended  for  the  protection  of 
generators,  transformers,  etc.,  from  internal  short  circuits  and 


A.C.  Oonantor 


Tkmnaforman 


Oircuit  BrMtker 


Olrcalt  Breaker 


ft  •  utrcau  on^m&t 


Fig.  326.— Differential  Relay  Con-        Fig.  327.— Diflferential   Relay  Con- 
nection for  Generator  Protection.  nection  for  Transformer  Protection. 


operate  always  instantaneously.  They  are  of  the  ordinary  plunger 
type  and  may  be  provided  with  one  or  two  coils,  one  generally  being 
used  for  generator  protection  and  two  for  protecting  transformers, 
as  shown  in  Figs.  326  and  327. 

When  one  current  coil  is  used,  the  secondaries  of  the  current 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  623 

transformers  are  connected  in  series  in  the  circuit  containing  the 
relay  coil  and,  in  such  a  manner,  that,  imder  normal  conditions,  the 
current  would  simply  circulate  in  the  secondary  circuit  and  not 
enter  the  relay  coil  due  to  its  higher  impedance.  If,  however, 
trouble  should  occur  in  the  generator,  there  would  be  a  reversal  of 
current  through  the  ciurent  transformer  nearest  the  oil  circuit 
breaker,  and  the  two  secondary  currents  would  naturally  oppose 
each  other,  in  which  case  both  would  take  the  path  through  the 
relay  coil.  This  would,  therefore,  receive  the  resultant  of  both 
currents  and  trip  out  the  oil  circuit  breaker  and  disconnect  the 
faulty  generator  imit  from  the  system. 

If  it  should  so  happen  that  the  two  current  transformer  prima- 
ries differ  from  that  of  the  power  transformer,  which  may  easily 
occur  when  tap  connections  are  changed,  the  secondary  ciurents 
in  the  two  current  transformers  would  not  be  equal.  This  would 
mean  that  there  would  be  a  resultant  current  or  flux  in  the  relay 
which  would  be  equivalent  to  that  difference,  and  satisfactory 
operation  would  be  affected  to  some  extent.  It  is,  therefore, 
important  that,  with  normal  load  on  the  power  transformer,  the 
unbalanced  current,  that  is,  the  difference  between  the  secondary 
currents  in  the  current  transformers  connected  to  the  two  sides 
of  the  power  transformer  should  be  zero.  Otherwise  two  coils 
should  be  used,  as  shown  in  Fig.  327.  These  are  wound  on  the 
same  core,  the  coils  being  connected  separately  to  current  trans- 
formers in  the  primaries  and  secondaries  of  the  power  trans- 
former. Normally  the  coils  oppose  each  other,  with  resultant 
zero  flux  in  the  relay  core.  When  a  winding  of  the  power  trans- 
former is  short-circuited,  the  other  Unes  in  parallel  feed  back  into 
the  short,  reversing  the  direction  of  one  coil  so  that  the  flux  in 
the  core  becomes  cumulative  and  the  relay  operates.  "When 
used  in  connection  with  generators  the  neutral  point  must  be 
opened  for  the  insertion  of  cmrent  transformers,  as  shown. 

Pilot  Wire  Relays.  For  a  single  tie  line,  over  which  energy 
may  normally  be  fed  in  either  direction,  reverse  power  relays  at 
each  end  of  the  circuit  connected  by  means  of  pilot  wires,  will  open 
both  ends  of  such  a  line  whenever  trouble  exists  on  that  line,  and 
under  no  other  conditions.  Energy  may  flow  in  either  direction 
so  long  as  the  energy  in  the  two  ends  of  the  line  shall  flow  in  the 
same  direction.  These  relays  are  equipped  with  double-throw 
contacts,  the  construction  of  the  relays  being  such  that  so  long  as 


Digitized  by 


Google 


524 


ELECTRICAL  EQUIPMENT 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  525 

energy  flows  in  the  same  direction  in  the  two  ends  of  the  line,  all 
the  contacts  of  the  relays  connected  to  the  tie  line  will  take  a 
uniform  position.  If  the  direction  of  energy  should  change 
over  the  entire  line,  both  contacts  would  simultaneously  reverse, 
bringing  them  once  more  to  a  uniform  position.  Under  these 
circumstances,  the  circuit  of  the  low-voltage  trip  (see  Fig.  328), 
will  be  imbroken  and  the  tripping  circuit  will  consequently  be 
kept  open.  A  slight  time  delay  is  provided  for  the  overload 
relays  simply  to  insure  sufficient  delay  to  allow  all  relay  contacts 
to  swing  to  their  proper  position  on  the  occurrence  of  a  normal 
reversal  of  energy  in  the  tie  line.  If,  however,  trouble  should 
occur  between  the  stations,  power  would  be  fed  into  the  line  from 
each  end,  and,  as  a  consequence,  the  relay  contacts  on  one  end  of 
the  line  will  remain  at  one  side  while  the  relay  contacts  at  the 
other  end  of  the  circuit  will  be  thrown  to  the  opposite  side.  This 
will  result  in  opening  the  circuit  of  the  time-limit,  low-voltage 
rdays,  and  the  falling  of  the  low-voltage  relay  plungers  will  close 
the  oil  switch  tripping  circuits  at  each  end  of  the  line  and  isolate 
the  circuit. 

High-iensicn  Series  Relays.  These  are,  in  general,  of  the  same 
principle  as  the  ordinary  plunger  type  relay.  They  are  chiefly 
used  with  high-tension  oil  circuit  breakers  for  overload  protection 
where  current  transformers  are  not  installed  or  warranted,  and 
may  be  either  of  the  instantaneous  or  inverse  time-limit  type. 
The  coil  is  connected  directly  in  series  with  the  Une  and  moimted 
on  a  post-type  insulator,  the  size  of  which  depends  on  the  voltage. 
The  plunger  of  the  relay  is  by  means  of  a  long  wooden  rod  con- 
nected to  a  circuit-closing  switch  which  can  be  mounted  on  any 
vertical  flat  surface  below  the  location  of  the  relay  coil. 

Over-^oUage  Relays.  These  may  be  either  instantaneous  or 
time  limit  and  are  similar  in  construction  to  overload  plunger- 
type  relays,  differing  only  in  that  potential  windings  are  sub- 
stituted for  the  current  coils.  They  may  be  used  to  protect  gen- 
erators, transformers  or  other  power  apparatus  against  damage  due 
to  abnormal  voltages.  For  this  purpose  the  relay  should  be  con- 
nected so  as  either  to  open  up  the  field  circuit  of  the  alternators 
or  introduce  into  each  field  circuit  a  sufficient  resistance  to  insure  a 
reasonably  low  potential  on  the  system. 

The  conditions  most  frequently  responsible  for  a  dangerous 
rise  in  potential  is  the  loss  of  load  on  a  power-station  while  the 


Digitized  by 


Google 


526  ELECTRICAL  EQUIPMENT 

generators  are  operating  under  considerable  excitation.  The 
abnormal  voltage  is,  therefore,  usually  accomplished  by  a  de- 
creased current.     To  guard  against  the  possibility  of  opening  the 


Nl/ 


Potcntl«l  iff ^  |«wltch  which  BhoHfl 

WfoJSer        ^  h*"  flew  -^ 


—ir 


/  BoDTCt  ^[ToopeaiiureoaoC 

rto  opening  coll  of  I  ^  ^^ fl«id'«witeh  or  of 

->jficM-fwilchorof  Potential    ?[^  \       f>|«^ 


Pot. 
Relaj 


ntial 
uormer 


Poteni 
Tranut 

: $IteU7 

<Lo^ 


r>\  •witch  which  ahorta 


Current 
Tranftformer 


Fig.  329. — Simplified  Diagram  of  Over-vcltage  Relay  Connections. 

field  circuit  under  any  condition,  other  than  the  loss  of  load,  a 
circuit  opening  overload  relay  or  circuit  closing  underload  relay 
may  be  connected  to  the  line  with  its  contacts  in  series  with  those 
oif  the  over-voltage  relay.     Fig.  329  shows  the  connections. 

Low-voltage  Relays.  These  are  of  the  circuit-opening  plunger- 
type  provided  with  a  potential  winding  regularly  wound  for  use 
on  the  110-volt  secondaries  of  potential  transformers. 

In  operation,  so  long  as  the  potential  is  about  normal,  the 
plunger  is  held  up,  causing  the  contacts  to  remain  open.  When  the 
potential  falls  below  one-half  normal,  the  plunger  is  released  and 
the  circuit  closed.  In  some  cases  the  plunger  must  be  pushed  up 
by  hand,  after  potential  has  been  apphed.  Usually,  however, 
coils  are  used  which  will  automatically  raise  the  plunger  when 
normal  voltage  is  restored. 

Underload  Relays,  These  are  made  with  circuit-closing  con- 
tacts for  instantaneous  operation  and  are  similar  to  low-voltage 
relays  with  the  difference  that  current  coils  are  substituted  instead 
of  potential  coils. 

Trip-free  Relay.  This  is  a  safety  device  intended  for  use  with 
electrically  controlled  circuit  breakers,  in  that  it  prevents  them 
from  being  held  closed  on  overloads.  To  accomplish  this,  the  trip- 
free  relay  is  simply  added  to  the  standard  control  wiring.  After 
the  breaker  comes  out  on  overload  it  cannot  be  thrown  in  again 
until  the  closing  contacts  of  the  control  switch  have  been  allowed 
to  retm-n  to  the  open  position.  The  diagram  in  Fig.  330  illustrates 
the  connections. 

Signal  Relays.  These  are  used  for  indicating  to  the  attendant 
the  automatic  opening  of  circuit  breakers.    When  these  are  closed 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


527 


Operating  Bme»  ' 


FomII 


Aaziliaryl 
Switch  ? 


JHB 


CloslBjT  >   Switch  f  ? 
Coll     S  J-  ^ 


Oontrol 
Relay 


rollL 


"^Slf 


I 


to  nearect  opcratiaff 
pMitive  BtM  depending 
on  whether  Relay  is 
moanted  near  the 
Gircnii  Breaker  or  on 
the  Panel 


=SD'"^Trlp  Free 


A  Cloeed  when  Oil  Switch  is  Closed. 
B  Cloeed  when  Oil  Switch  is  Open. 


Fig.  330. — Connections  for  Trip-free  Relay. 


Jo  Cloelnff 
Contact 


a? 


Relay 


►  «  , .  __  To  Corrent 


Transfermer 
in  Main  Line 


Red  Lamp 


.^Controlling  Switch 


^ 


To  Opening 
Contpct 


Oreen  Lamp 


lii  VoU  Operating 


To  Clodng 
Contact 


[tUelay 


Pose  _^ 

n 


.  To  Cnrrent 
Tranafermer 
in  Main  Line 


Red  Lamp 


^  Controlling  Switch 


^ 


Green  Lamp 


To  Opening 
Contact 


Bell 


Battery 


L 


rC 


Relay  Bos  ^ 


Fig.  331. — Connections  for  Bell-alarm  Relay. 


Digitized  by 


Google 


528 


ELECTRICAL  EQUIPMENT 


by  hand  and  opened  either  by  hand  or  by  some  automatic  tripping 
arrangement,  a  circuit-closing  auxiliary  switch  for  closing  the  alarm 
circuit  is  so  moimted  on  the  operating  mechanism  that  when  the 
circuit  breaker  is  opened  by  the  hand-closing  mechanism,  the  aux- 
iliary switch  does  not  operate.  But  if  the  tripping  is  affected  by 
the  automatic  mechanism,  the  auxiliary  switch  will  close  and 
throw  in  circuit  the  alarm  device. 

On  electrically  operated  circuit  breakers  no  arrangement  of  a 
mechanically  operated  auxiliary  switch,  which  will  allow  it  to 

distinguish  between  nonautomatic 
and  automatic  opening,  can  be  con- 
ventionally made.  Consequently,  to 
inform  the  operator  of  automatic 
opening,  there  is  used  generally  a 
bell  alarm  relay  with  its  operating 
coil  connected  in  the  power  supply  of 
the  circuit-breaker  tripping  coils, 
(Kg.  331).  The  operation  of  the 
relay  is  not  affected  by  the  control 
switch  circuits,  and  is  energized  only 
when  current  passes  through  the 
tripping  circuit  contacts  of  one  or 
more  of  the  protective  relays. 

Whenever  a   circuit   breaker   is 
automatically  tripped,  the  relay  coil 
is  energized  for  an  instant  through 
the  circuit  of  the  overload  trip.    As 
it  may  be  necessary  to  ring  an  alarm 
bell  for  some  time  to  attract  the 
operator's  attention  to  the  fact  that 
a  device  has  been  opened  automatic- 
ally, the  relay  plunger  is  notched  so 
that  it  remains  up  in  the  closed  posi- 
tion  until   pulled   down    by  hand, 
which  shuts  off  the  alarm  bell  by 
opening  the  bell-alarm  circuit. 
Control  Relays.    These  are  used  in  connection  with  the  con- 
trol switches  for  electrically  operated  oil  circuit  breakers,  etc. 
Since  these  control  switches,  as  a  rule,  are  not  constructed  to  open 
a  current  of  sufficient  capacity  to  operate  the  closing  coil  of  the 


^Pio.  332. — Solenoid  Control 
Relay. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  529 

solenoid,  for  example,  it  is  necessary  to  use  a  control  relay  with  its 
operating  coil  connected  across  the  closing  contacts  of  the  con- 
trol switch  and  the  relay  contacts  in  series  with  the  solenoid  closing 
coil.  This  relay  is  illustrated  in  Fig.  332  and  the  connections  in 
Fig.  351. 

Switchboards.  The  switchboard  of  the  modern  large  power 
station  is,  strictly  speaking,  not  a  switchboard  in  the  original  sense 
of  the  word.  While  for  small  stations  the  entire  instrument  and 
switch  equipment  may  be  mounted  directly  on  the  board,  for 
large  stations  the  oil  circuit  breakers  and  bus-bars  are  always 
mounted  at  some  distance  from  the  same,  the  location  being  deter- 
mined by  convenience  of  wiring  and  safety.  In  such  a  case  the 
switchboard  is  rather  a  control  board  and  contains  only  the  con- 
trol switches,  instruments  and  the  various  other  auxiliary  devices 
such  as  indicating  lamps,  plugs  and  receptacles  for  measuring 
the  voltage  and  for  synchronizing,  etc. 

The  design  of  a  switchboard  involves  a  careful  consideration 
of  the  apparatus  to  be  controlled,  the  system  of  connections, 
arrangement  of  cables  and  other  wiring,  and  on  the  general  design 
of  the  station.  The  various  apparatus  on  the  board  should  be 
arranged  so  as  to  faciUtate  the  operation,  and  for  this  reason  the 
board  is  always  divided  up  in  panels  corresponding  to  the  machin- 
ery or  circuits  which  are  to  be  controlled.  The  exciter  and  the 
regulator  panels  are  generally  located  at  one  end,  then  the  generator 
panels,  station  panel,  transformer  and  outgoing  line  panels  in  order 
mentioned.  This  arrangement  may,  of  course,  be  different  so  as 
to  more  closely  correspond  to  the  arrangement  of  the  apparatus. 
Blank  panels  should  preferably  be  provided  for  future  machinery 
from  the  beginning.  The  expense  of  such  panels  is  very  httle  and 
it  facilitates  the  addition  of  instrument  equipments  for  future 
units  considerably.  In  such  a  case  it  will  only  be  necessary 
to  remove  the  blank  panels,  have  the  necessary  instruments  and 
wiring  moimted  thereon,  then  replace  them  on  the  framework 
and  make  the  necessary  remaining  connections,  thus  causing  the 
least  disturbance  to  the  rest  of  the  equipment. 

Pipe  framework  is  now  almost  universally  used  for  support- 
ing the  panels  on  account  of  neatness  and  simphcity .  The  material 
of  the  panels  may  be  slate  or  marble.  Where  live  parts  are 
mounted  indirectly  thereon,  slate  should  not  be  used  if  the  voltage 
is  higher  than  1200,  and  marble  is  limited  to  about  3300.    Natural 


Digitized  by 


Google 


530 


ELECTRICAL  EQUIPMENT 


black  slate  is  best  suited  for  switchboard  work,  as  it  is  not  easily 
marred  or  stained  and  can  readily  be  matched  when  making 
extensions. 

The  small  wiring  on  the  back  of  the  panels  should  be  done 
neatly  and  regularly  to  facilitate  tracing  of  connections,  and  it 
should  be  arranged  in  a  manner  best  suited  for  connection  to  the 
control  wires  coming  to  the  board. 

The  back  and  ends  of  the  board  may  be  closed  by  a  wire  and 


Fig.  333. — ^Arrangement  of  2300-volt  Switchboard  with  Switches  Mounted 
on  the  Pipe  Work  Supporting  the  Panda. 

grille-work  screen  to  prevent  tampering  with  the  apparatus 
back  of  the  panels,  while,  on  the  other  hand,  they  greatly  enhance 
the  appearance  of  the  installation.  Switchboards  provided  with 
these  screens  comply  with  the  most  stringent  rulings  of  safety 
first  regulations  since  the  screens  afford  complete  protection 
against  accidental  contact  with  live  parts  by  operators  and 
others. 

Switchboards  may  be  classified  according  to  the  style  of  con- 


Digitized  by  VjOOQ  IC 


SWITCHING  EQUIPMENT 


531 


struction  or  according  to  the  manner  in  which  the  oil  circuit 
breakers  are  mounted  and  controlled.  Based  on  design  we 
have: 

1.  Vertical  panel  boards. 

2.  Benchboards. 

And,  according  to  method  of  control: 

1.  Self-contained  boards. 

2.  Mechanically  remote-control  boards. 

3.  Electrically  (remote-control  boards. 


Shovld  b«  sunk  in  floor  nlxght  \ 
ly  if  Oil  Switch  Can  i«  to  be  " 
rtmoved  with  Switch  open 


I! 


FiQ.  334.-rArrangBment  of  2300-volt  Switchboard  with  Mechanically  Remote- 
control  Switches  Mounted  on  Open  Pipe  Work. 

The  self-contained  switchboard  is  always  of  the  vertical  type, 
Fig.  333,  and  has  all  the  apparatus,  including  the  oil  circuit 
breakers  mounted  near  the  panels. 

The  mechanically  remote-control  board  is  also  of  the  vertical 


Digitized  by 


Google 


532 


ELECTRICAL  EQUIPMENT 


panel  type  Fig.  334,  but  the  oil  switches  and  bus-bars  are 
mounted  on  a  pipe  or  other  structure  somewhat  to  the  rear  of 
the  panels,  the  switches  being  operated  by  handles,  located  on 
the  front  of  the  panels,  through  the  medium  of  mechanical  con- 
necting rods. 

The  electrically  remote-control  switchboard  may  be  either  of 
the  vertical  panel  tjrpe  or  of  the  bench-board  type,  depending  on 


Fig.  335. — ^IVpical  Vertical-type  Switchboard  with  Hand-operated  Ofl  Cir- 
cuit Breakers.     Front  View. 

the  conditions  to  be  met.  The  oil  circuit  breakers  and  the  bus- 
bars are  installed  in  the  most  convenient  place  in  the  station,  oft^n 
at  a  considerable  distance  from  the  board  The  breakers  are 
then  operated  by  means  of  solenoids  or  motors,  which  in  turn  are 
controlled  from  the  switchboard. 

The  proper  type  of  switchboard  to  be  selected  depends  on  the 
apparatus  involved,  particularly  the  oil  circuit  breakers  and  the 
bus-bars,  and  these  in  turn  on  the  power  to  be  handled,  the  voltage, 
operating  features,  spacfe  available,  etc. 

With  stations  of  large  capacity  and  high  transmission  poten- 
tials, requiring  a  heavy  switching  equipment,  manual  control  is 
practically  impossible,  partly  from  mechanical  reasons  and  partly 
on  account  of  the  increased  space  factor  required  by  the  breakers, 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


533 


buses,  etc.,  and  recourse  was  had  to  the  methods  of  remote  con- 
trcd. 

Commencing  with  the  manually  operated  remote-controlled 
switches  equipped  with  rods  and  bell  cranks,  good  practice  finally 
recognized  the  desirabiUty  of  employing  solenoid  or  motor- 
operated  breakers  controlled  from  a  central  point.    This  arrange- 


FiG.  336.— Rear  View  of  Switchboard  Shown  in  Fig.  335. 

ment  permitted  the  location  of  the  control  board  without  reference 
to  the  location  of  the  breakers  or  the  apparatus  which  they  con- 
trol. Absolute  isolation  of  the  high-tension  equipment  may  thus 
be  secured,  thereby  largely  eUminating  the  personal  hazard  and 
danger  of  accidental  contact  and  making  possible  the  use  of  the 
Qunimum  amount  of  high-tension  busses  inside  the  station. 


Digitized  by 


Google 


534 


ELECTRICAL  EQUIPMENT 


It  is  difficult  to  give  any  accurate  recommendation  as  to  where 
the  dividing  line  should  be  between  the  different  arrangements. 
In  general,  it  may,  however,  be  said  that  those  shown  in  Figs.  333 
and  334  can  be  used  for  voltages  up  to  6600  and  station  capacities 
not  exceeding  5000  kilowatts.  For  higher  capacities  and  voltages 
it  is  advisable  to  mount  the  oil  circuit  breakers  in  compartments. 
In  fact,  most  high  capacity  switches  for  moderate  voltages  are 


Fig.  337. — Typical  Vertical  Type  Switchboard  with  Electrically  Operated 
Remote-control  Oil  Circuit  Breakers. 


made  for  cell  mounting,  but  above  22,000  volts  they  are,  as  a 
rule,  of  the  open  design. 

Figs.  335  and  336  show  the  front  and  rear  views  of  a  typical 
switchboard  of  the  vertical  panel  type  with  hand-operated  oil 
circuit  breakers  mounted  at  the  rear  of  the  panels.  Fig.  337 
shows  a  similar  board  for  electrically  remote  control  circuit 
breakers. 

It  is  often  found  in  a  large  and  complex  installation  that  if  all 
the  instruments  and  apparatus  were  located  on  a  vertical  switch- 
board, its  dimensions  would  be  too  great  for  convenient  opera- 
tion, and  many  appliances  such  as  control  switches,  sjTichronizing 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  635 

and  potential  receptacles  could  not  aU  be  accommodated  in  a 
position  most  convenient  for  the  operator.  To  overcome  these 
diflSiculties  the  benchboard  has  been  introduced.  In  this  manner 
the  useful  surface  has  been  increased  by  an  amount  almost  equal 
to  the  top  of  the  bench,  the  latter  oflFering  an  excellent  position  for 
control  apparatus,  bringing  it  within  distinct  view  and  convenient 
reach  of  the  operator. 

Another  advantage  is  also  incidentally  obtained  by  reason  of 
the  greater  distances  between  the  instnunents  and  the  operator, 
which  enables  him  to  observe  a  greater  nimiber  of  instruments 
from  any  point  while  manipulating  the  control  apparatus.  A 
further  advantage  may  be  taken  of  this  condition  by  increasing 
the  height  of  the  instnunent  section,  if  desirable,  in  order  to 
allow  room  for  more  instruments,  which  may  be  read  without 
difficulty. 

Figs.  338  to  341  show  different  types  of  bench  boards  in  use 
and  the  relative  locations  of  the  different  pieces  of  apparatus. 
Which  type  should  be  used  depends  entirely  upon  the  apparatus 
involved  and  on  the  local  conditions.  It  is  thus  often  foimd  that 
a  bench  board  of  a  certain  design  will  give  the  best  result  for  con- 
trolling the  machines,  while  a  vertical  panel  board  will  be  more 
feasible  for  feeder  circuits.  When  separating  the  boards  the 
nmnber  of  operators  required  should  always  be  considered. 

.  Pedestal  control  boards  are  occasionally  used,  but  there  seems 
to  be  no  real  advantage  in  splitting  up  the  equipment  to  such  an 
extent.  Figs.  342  and  343  illustrate  two  typical  bench  board 
designs,  and  Fig.  344  shows  the  control  room  of  the  Mississippi 
River  Power  Company  at  Keokuk.  The  operation  in  this  sta- 
tion is  completely  controlled  by  a  chief  dispatcher,  who  is  in  tele- 
phonic commimication  with  all  parts  of  the  system.  A  special 
desk  is  provided  for  him,  on  which  is  moimted  the  telephone 
switchboard,  while,  in  front  of  this  desk  a  miniature  arcnshaped 
switchboard  is  installed  which  contains  a  set  of  mimic  bus-bars 
showing  by  means  of  small  indicating  lights  the  open  or  closed 
position  of  all  the  breakers  in  the  station.  It  also  contains 
graphic  voltmeters  and  ammeters  for  recording  the  voltage  on 
each  bus  section  and  the  current  in  each  of  the  outgoing  lines. 

The  main  control  switchboard  is  divided  into  sections  corre- 
sponding to  the  bus  sections,  with  an  additional  section  for  the 
auxiliary  equipment.     The  arrangement  of  these  boards  is  at 


Digitized  by 


Google 


536 


ELECTRICAL  EQUIPMENT 


Met«r« 


Pok  and  Sjmehr.  Receptaeles^ 
Oontrol  SwitcbM  &,  L&m^ 
Mimic  BiMM. 


nd.Rhe<Mtat 
Control 


ReUj* 


Removftbla 
OriUPnnola 


Fig.  338. — A  Simple  Type  of  Combination  Control  Board  and  Instrument 
Board  Showing  the  Locations  Best  Suited  for  the  Various  Pieces  of 
Apparatus. 


T^|ti>ffti]k»r« 


En^biu  %\gna\% 


CtoDtrol  HwEtrhpi  iV  Lamm, 


Fid  RU*Hi*!4* 
Conlrol 


laiifumirali 


luiaf* 
fiirtMa^  ttd 

laiM-RiqalliS 


Fig.  339. — An  Enlargement  on  the  Arrangement  Shown  in  Fig.  338,  which 
Meets  the  Demand  of  Greater  Working  Surface  by  the  Addition  of 
Rear  Panels. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


537 


Pot.  and  Synch.  Receptacles 
Oontrol  Switdies  6l  Lamp 
Mimic  Buses 


na.  Rheostat 
Control 


Instnimonts 


Meters  and 
Graphic  Instruments 


Relays 


Testing  Links 
and  Switches 

for  Relays  and 
Instruments 


Fig.  340. — Control  Board  with  Independent  Instrument  Board.    This  arrange- 
ment offers  more  useful  surface  than  does  that  of  Fig.  338. 


ments 

r 

View  of 

/ 

Removahia 
Grill  Panels 


Pot.  and  Syneh«  Receptacles^ 
Oontrol  Switches  &  Lamgi         ^ 
Mimic  Buses 


Fid.  Rheostat 
Oontrol 


Relays 


Gallery  Rail 
O 


Removable    C 
GriU  Panels 


Gallery 


FiQ.  341. — A  Gallery  Type  of  Bench-board  which  Permits  the  Operator  View- 
ing the  Machines  through  the  Board. 


Digitized  by 


Google 


538 


ELECTRICAL  EQUIPMENT 


the  present  time  in  the  form  of  an  L,  although  ultimately  it  will  be 
in  the  form  of  a  U  with  the  dispatcher  board  in  the  center. 

Diffused  illumination  in  the  control  room  is  provided  by  means 
of  a  skyUght,  which  forms  the  entire  ceiling.  In  order  to  prevent 
glare  on  the  instruments  it  also  became  necessary  to  provide  amber- 
colored  glass  in  the  windows.  At  night  a  diffused  illumination  is 
accomplished  by  tungsten  lamps,  which  are  mounted  back  of  the 
skylight  panes. 

Instrument  Equipment.  The  instrument  and  meter  equip- 
ment for  any  particular  installation  should  be  chosen  with  the  idea 


Fig.  342. — Typical  Benchboard  of  the  Continuous  Type. 

of  getting  something  which  is  satisfactory  from  an  engineering 
standpoint,  at  the  same  time  keeping  in  mind  its  cost  in  proportion 
to  that  of  the  total  installation,  and  also  considering  the  class  of 
attendants  who  will  operate  the  board.  It  is  not  good  economy 
to  invest  in  an  elaborate  set  of  instruments  when  the  man  who 
operates  the  plant  does  not  understand  their  use.  In  the  large 
installations,  where  more  intelligent  help  is  employed,  the  efficiency 
of  the  plant  can  be  greatly  improved  by  the  use  of  instruments 
which  are  understood,  but  which  would  be  more  than  useless  in 
the  hands  of  the  unskilled  attendant. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


539 


Obviously  it  is  difiScult  to  establish  exact  dividing  lines  which 
will  cover  all  conditions.  The  tables  given  in  the  following  give 
the  instrument  equipment  recommended  for  use  on  the  circuits 
eniunerated.    Special  operating  conditions  and  requirements  will 


Fig.  343.--Typical  Benchboard  of  the  Gallery  Type. 

often  demand  different  measuring  apparatus  than  that  given,  but 
the  table  will,  in  all  cases,  serve  as  a  guide  in  choosing  a  suitable 
equipment. 

Instruments  of  each  different  function  are  valuable  under 
certain  conditions  or  to  aid  in  accomplishing  certain  results.    To 


Digitized  by 


Google 


540 


ELECTRICAL  EQUIPMENT 


assist  in  the  choice  of  these  and  to  explain  the  advantage  gained 
by  using  each  particular  instrument  the  information  in  the  follow- 
ing paragraphs  will  be  found  useful. 

For  Direct-ctirrent  Installations.  Direct-current  Ammeters. 
(1)  On  machines  of  all  kinds  heating  is  the  factor  which  deter- 
mines the  load  which  can  be  carried  safely  assuming  the  voltage 
normal.    Ammeters  give  an  indication  of  the  heating  of  circuits 


FlQ.  844. — Mississippi    River  Power  Company.    Chief   Operator's  Room 
Showing  Control  Boards  and  Switchboards. 

in  which  they  are  connected  and  consequently  are  indispensable 
for  machine  circuits. 

(2)  They  show  the  division  of  load  between  machines. 

(3)  On  feeder  circuits  they  indicate  which  feeders  are  over- 
loading the  machines,  and  also  furnish  a  means  for  indicating 
the  gradual  growth  or  decline  in  the  demands  made  upon  the  gen- 
erating apparatus  by  any  particular  feeder,  thus  giving  a  warning 
that  the  capacity  of  the  apparatus  must  be  changed,  or  the 
feeder  load  rearranged. 

Direct-current  Voltmeters.  (1)  They  show  that  machines 
are  being  operated  at  a  voltage  not  too  high  to  damage  their 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  541 

insulation,  or  to  damage  apparatus  for  which  the  machines  fur- 
nish power. 

(2)  They  are  required  when  paralleling  machines,  which  must 
be  of  the  correct  polarity  and  at  very  nearly  the  same  voltage 
in  order  to  enable  throwing  them  together  with  the  least  disturb* 
ance. 

(3)  They  can  be  used  as  ground-detecting  devices  by  making 
proper  connections  to  the  system. 

Curve-Drawing  Instruments,  (1)  They  give  a  permanent 
record  of  the  running  conditions  of  the  circuits  in  which  they  are 
connected  without  the  loss  of  time  and  possible  chance  of  error 
which  occur  when  such  records  are  computed  from  the  readings 
of  indicating  instruments.  Showing,  as  they  do,  the  distribution 
of  the  load  for  every  hour  of  the  day  throughout  the  year,  they 
place  in  the  hands  of  the  management  very  valuable  information 
which  forms  the  basis  for  future  extensions  or  improvements  of 
service  and  load  distribution. 

For  Altematmg-current  Installations.  AUemating<urrenl 
Ammeters.  (1)  They  give  an  indication  of  the  heating  of  the 
armature  of  the  machine.  This  is  a  thing  which  the  indicating 
wattmeters  will  not  do  because  of  the  fact  that  it  measiu-es  only 
the  energy  component  while  the  ammeter  measures  the  reactive 
as  well  as  the  energy  component  of  the  current,  both  of  which 
produce  heating. 

(2)  In  case  machines  in  multiple  are  running  at  the  same  power- 
factor  anmieters  show  the  division  of  load. 

(3)  On  feeder  circuits,  ammeters  indicate  which  feeders  are 
overloading  the  machine. 

(4)  On  overhead-transmission  lines  the  use  of  three  anameters, 
one  in  each  phase  gives  an  indication  of  trouble  on  the  lines,  such 
as  grotmding. 

AUemating-currerU  Voltmeters.  (1)  They  show  that  machines 
are  being  operated  at  a  voltage  not  too  high  to  damage  the  in- 
sulation, or  to  damage  apparatus  for  which  the  machines  furnish 
power. 

(2)  They  are  valuable  when  paralleling  machines  which  must 
be  at  very  nearly  the  same  voltage  in  order  to  enable  throwing 
them  together  with  the  least  disturbance  to  the  system. 

(3)  They  can  be  used  as  ground-detecting  devices  by  making 
proper  connections  to  the  system. 


Digitized  by 


Google 


542  ELECTRICAL  EQUIPMENT 

(4)  The  compensated  type  or  ordinary  type  with  line  drop 
compensator  is  useful  to  indicate  at  the  power  station  the  voltage 
at  any  predetermined  point  of  a  feeder. 

Direct-current  Field  Ammeters,  (1)  They  give  an  indication 
of  the  heating  in  the  fields  of  machines. 

(2)  They  assist  in  locating  trouble  in  a  machine.  For  in- 
stance, in  case  the  alternating  current  voltmeter  on  a  generator, 
which  is  supposedly  operating  normally,  shows  that  there  is  no 
voltage  generated,  a  glance  at  the  field  ammeter  may  show  no 
reading,  in  which  case  it  is  evident  immediately  that  the  field 
circuit  is  broken  or  the  exciter  system  in  trouble. 

(3)  They  give  an  indication  of  cross  currents  in  generators. 
For  instance,  consider  a  generator  panel  containing  nmin  alter- 
nating-current ammeter,  power-factor  indicator,  voltmeter,  and 
field  ammeter.  If  the  machine  is  up  to  speed,  the  amount  of 
field  current  in  excess  of  normal  which  is  required  at  a  given  power- 
factor  to  hold  normal  voltage,  shows  proportionately  the  amount 
of  cross  current. 

(4)  They  are  of  great  value  in  the  fields  of  synchronous 
motors,  because  for  any  given  load  and  power-factor  the  armature 
current  is  a  minimum  for  a  certain  value  of  the  field  current  for 
which  the  field  can  be  adjusted  with  the  aid  of  the  field  ammeter. 

Indicating  Wattmeters,  (1)  They  show  the  actual  power  in  a 
circuit  no  matter  what  the  power-factor  since  they  measure  the 
energy  but  not  the  reactive  component.  This  makes  them  valu- 
able in  the  circuits  of  alternating-current  machines  operated  in 
multiple  since  they  show  the  division  of  load  between  machines, 
something  which  ammeters  alone  do  not  indicate,  except  when 
machines  are  operated  at  exactly  the  same  power-factor  and 
voltage. 

(2)  In  the  absence  of  curve-drawing  instnmients,  they  fur- 
nish a  means  for  obtaining  the  load  curve  of  a  station. 

(3)  They  indicate  reversal  of  power  in  a  circuit  which  an 
ammeter  will  not  do. 

Power-factor  Indicators,  (1)  It  is  a  well-understood  fact  that 
it  is  most  economical  to  operate  power  plants  at  as  high  a  power- 
factor,  as  possible  in  order  to  get  maximum  output  from  the 
machines.  The  power-factor  indicator  is  very  useful  in  telling 
directly  what  this  power-factor  is.  Proper  wiring  arrangements 
can  be  made  to  use  only  one  instrument  per  board,  plugging  it  to 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  643 

different  circuits.  In  this  way  the  circuits  of  poor  power-factor 
can  be  discovered  and  steps  taken  to  improve  conditions  if  con- 
sidered desirable.  Where  synchronous  condensers  are  used  for 
power-factor  correction,  the  power-factor  indicator  connected  to 
the  bus  or  circuit  to  be  corrected,  becomes  particularly  valuable. 

(2)  Generators  in  multiple  will  operate  at  maximum  output 
when  they  are  all  running  at  the  same  power-factor,  reducmg 
cross  currents  to  a  minimum.  The  power-factor  indicator  affords 
the  easiest  means  of  making  this  adjustment,  since  it  shows  the 
power-factor  of  each  machine  at  a  glance  without  the  necessity 
of  computing  this  from  the  readings  of  other  instruments. 

(3)  The  reading  of  a  power-factor  indicator  in  connection 
with  that  of  an  ammeter  and  voltmeter  makes  it  possible  to  readily 
figure  the  kilowatt  output  of  a  machine  without  the  use  of  an 
indicating  wattmeter. 

Reactive  VoUrampere  Indicators.  (1)  They  measure  the  idle 
or  reactive  portion  of  the  power  and  are  the  only  instruments 
which  do  so  directly. 

(2)  In  connection  with  the  reading  of  an  indicating  wattmeter 
the  readings  of  the  reactive  volt-ampere  indicator  give  an  easy 
means  for  figuring  the  power-factor. 

(3)  They  are  considered  in  some  cases  more  valuable  than 
power-factor  indicators  since  they  given  an  actual  quantitative 
reading  in  kilovolt-amperes  while  the  power-factor  indicator  gives 
a  reading  in  per  cent  only.  This  fact  can  readily  be  seen  from  an 
inspection  of  the  following  simple  formula: 

T>         r    X  True  watts 

Power-factor =- 


Apparent  watts' 


(Where  the  apparent  watts  is  the  vector  sum  of  the  true  watts 
and  the  reactive  watts.)  The  reading  of  a  power-factor  indicator 
gives  no  actual  indication  of  magnitude  of  the  idle  current  which 
cause  heating.  For  instance,  at  light  load  a  power-factor  of  0.7 
or  0.8  would  be  no  cause  for  alarm,  while  at  full  load  or  overload 
it  might  mean  serious  heating  due  to  idle  currents.  This  is  espe- 
cially true  on  synchronous  converters,  where  on  account  of  the 
rectifying  action  of  such  machines,  the  cross-section  of  copper  is 
made  smaller  than  in  a  generator  of  the  same  capacity. 

Frequency  Indicators.     (1)  Machines  operate  most  econom- 


Digitized  by 


Google 


544  ELECTRICAL  EQUIPMENT 

ically  at  the  frequency  for  which  they  are  designed,  which  makes 
the  use  of  the  frequency  indicators  evident. 

(2)  They  are  valuable  when  synchronizing  machines,  since 
they  can  be  connected  on  the  incoming  machine  and  indicate  its 
speed,  showing  whether  it  is  too  high  or  too  low.  However, 
where  a  s3nQchronism  indicator  is  installed  they  are  not  required 
for  this  piu-pose,  since  this  instrument  shows  whether  the  speed  of 
the  incoming  machine  is  high  or  low. 

Synchronism  Indicator,  (1)  The  synchronism  indicator  affords 
the  quickest  and  safest  means  for  paralleling  machines,  since  it 
shows  when  the  machines  are  in  step  and  in  phase,  indicating  by 
the  position  of  the  needle  the  difference  in  the  phase  relations 
between  the  machines,  and  telling  whether  the  incoming  machine 
is  running  too  fast  or  too  slow.  It  is  superior  to  synchronizing 
with  lamps,  because  the  latter  give  no  indication  of  the  relative 
speed  of  the  incoming  machine.  The  lamps  will  indicate  when  the 
machines  are  of  the  same  frequency,  but  the  phase  relations  can 
be  judged  only  by  the  brilliancy  of  the  light. 

When  synchronizing  with  lamps  dark,  the  phase  relation  of  the 
machines  will  be  shown  by  the  brilliancy  of  the  light  to  a  point 
where  the  machines  are  approximately  45*^  out  of  phase,  below 
which  point  there  will  not  be  sufficient  voltage  across  the  lamp  to 
make  it  glow.  Again,  in  case  there  is  an  inopportune  failure  of 
the  lamp,  the  operator  might  be  misled  iad  throw  the  machines 
together  when  out  of  phase  with  possible  disastrous  results. 

When  synchronizing  with  lamps  bright,  it  is  difficult  to  deter- 
mine, after  watching  the  lamps  for  some  time,  at  just  what  instant 
they  are  burning  at  full  briUiancy,  and,  therefore,  at  just  what 
instant  the  machines  are  in  synchronism. 

Synchronizing  on  high-tension  lines,  while  often  desirable, 
has  been  out  of  the  question  because  of  the  excessive  cost  and 
space  required  for  installing  the  necessary  potential  transformers 
for  a  secondary  synchronism  indicator.  A  glow  synchronism 
indicator  is  now  available  for  this  purpose  on  circuits  of  13,200 
volts  and  above.  The  new  indicator  depends  for  its  operation 
upon  the  principle  of  electrostatic  discharge  in  a  vacuum. 

The  instnunent  case  resembles  the  ordinary  round  pattern 
switchboard  instrument.  Inside  the  case  are  receptacles  for  hold- 
ing the  special  glowers  which  project  through  holes  in  the  cover. 
Connections  from  the  line  to  the  device  are  made  through  con- 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  546 

densers,  which  consist  of  siispension  insulators  having  an  insula^ 
tion  equal  to  that  used  on  the  line.  Normally  the  glowers  have 
the  appearance  of  ordinkry  spherical  frosted  incandescent  lamp 
bulbs.  When,  however,  there  is  a  proper  difference  of  potential 
across  their  terminals  they  will  glow  with  a  reddish  hue.  When 
the  lines  are  not  in  synchronism,  the  glowers  will  light  up  in 
succession,  showing  the  relative  direction  of  rotation  and  indi- 
cating whether  the  incoming  machine  is  running  fast  or  slow. 
When  i^mchronism  is  reached  there  will  be  no  rotating  effect,  and 
one  glower  will  be  dark  while  the  other  two  will  glow  at  about 
half  brilliancy. 

Electrostaiic  Ground  Detectors.  (1)  They  give  a  constant 
indication  of  the  condition  of  the  system  with  respect  to  grounds 
which,  if  not  detected  immediately,  often  result  in  very  serious 
burnouts  or  voltage  disturbances. 

(2)  They  are  superior  to  any  system  of  ground  detecting  which 
necessitates  the  plugging  of  potential  transformers  and  lamps  or 
voltmeters  to  different  phases  of  a  polyphase  system;  first,  because 
the  polyphase  electrostatic  ground  detector  shows,  at  a  glance, 
whether  there  is  a  ground  on  any  phase,  while  with  the  other 
scheme  it  is  necessary  to  plug  the  primary  side  of  the  transformer 
to  the  different  phases  before  the  test  is  completed;  and,  second, 
because  the  electrostatic  ground  detector  is  supplied  with  a  scale 
for  reading  the  severity  of  the  ground  while  with  lamps  only  an 
approximate  indication  is  obtained  ordinarily,  and  for  high 
resistance  grounds  no  indication  whatever,  since  the  ordinary  125- 
volt  carbon  lamp  will  not  glow  at  much  less  than  25  volts  across 
its  terminals. 

Temperature  Indicators.  (1)  It  is  of  great  value  to  know  the 
temperatitfe  of  certain  parts  of  generator  and  transformer  wind- 
ings that  are  inaccessible  for  thermometer  measiu*ements.  An 
instnunent  known  as  the  temperature  indicator  has  been  pro- 
duced to  determine  these  temperatures.  Copper  coils  of  known 
resistance  are  placed  in  the  parts  whose  temperature  it  is  desired 
to  know.  The  changes  in  resistance  are  shown  on  the  scale  of 
the  indicator,  which  is  marked  in  degrees  Centigrade  correspond- 
ing to  the  change  in  resistance.  The  instrument  itself  is  a  differ- 
ential voltmeter  with  three  terminals.  The  connections  are  such 
that  one  of  the  moving  coil  windings  is  in  series  with  a  resistance 
coil  which  has  a  zero  temperatm-e  coefficient  and  a  resistance  equal 


Digitized  by 


Google 


546  ELECTRICAL  EQUIPMENT 

to  that  of  the  copper  temperature  coil,  and  the  other  winding  is  in 
series  with  the  copper  temperature  coil.  When  the  temperature 
of  the  copper  coil  rises,  the  current  in  that  branch  of  the  circuit 
decreases  and  causes  a  corresponding  deflection  toward  a  higher 
temperature  on  the  scale  of  the  instrument.  The  reverse  is  the 
case  when  the  temperature  falls. 

Curve-drawing  Instruments.  (1)  They  give  a  permanent 
record  of  the  running  conditions  of  the  circuits  in  which  they  are 
connected  without  the  loss  of  time  and  possible  chance  of  error 
which  occiu"  when  such  records  are  computed  from  the  readings 
of  indicating  instruments.  Showing,  as  they  do,  the  distribution 
of  the  load  for  every  hour  of  the  day  throughout  the  year,  they 
place  in  the  hands  of  the  management  very  valuable  information 
which  forms  the  basis  for  future  extensions  or  improvements  of 
service  and  load  distribution. 

The  following  tables  give  the  instrument  equipment  usuaBy 
employed  for  use  on  the  circuits  enumerated.  In  giving  these, 
each  circuit  is  considered  a  complete  unit  in  itself.  A  combina- 
tion of  two  imits  does  not  mean  that  all  instruments  listed  for  each 
separately  will  be  used  on  the  combination.  For  instance,  where 
a  generator  and  transformer  are  permanently  connected  together 
and  operated  as  a  unit,  there  is  no  necessity' for  using  an  ammeter 
in  the  transformer  circuit,  since  it  would  simply  dupUcate  the 
reading  of  the  generator  ammeter.  Other  similar  cases  are  numer' 
ous,  such  as  combined  generator  and  feeder  circuit,  combined 
transformer  and  feeder  circuit,  etc.  Special  operating  conditions 
and  requirements  will  often  demand  different  measuring  apparatus 
than  that  given,  but  the  tables  will  at  least  serve  as  a  guide  in 
choosing  a  suitable  equipment  in  all  cases.  The  small  letters 
in  the  tables  refer  to  the  notes  following  the  tables. 

Ctirrent  and  Potential  Transformers.  When  the  voltage  or 
current  of  the  circuit  to  which  the  instruments  are  to  be  con- 
nected exceeds  a  certain  limit  above  which  primary  instruments 
are  not  built,  potential  and  current  transformers  are  employed, 
the  instnunent  coils  being  operated  from  the  secondaries  of  these 
transformers.  As  a  matter  of  safety  to  the  operator,  secondary 
instnunents  are  recommended  for  all  circuits  in  excess  of  650 
volts. 

Since  the  normal  rating  of  the  secondary  of  current  transform- 
ers is  5  amperes,  secondary  current  coils  are  ordinarily  wound 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


547 


TABLE  L 
Direct  Curreki! 


Name  and  Number  op  Instruments  Used. 

Circuit  Measured. 

^«^BAK^%A*V      A:WA\0Vm9%AM  V*X*  • 

Ammeter. 

Voltmeter.  * 

Two-wire  generator 

1 

1  per  switchboard  plugged  to 
each  generator 

Two-wire  exciter  gen- 

1 

(d) 

erator 

Brush  arc  generator 

1    (Plugged  to  read 
each  machine  cir- 
cuit) 

None  required 

Two-wire  feeder 

1(a) 

None  required  ordinarily 

Railway  feeder 

1 

Plug  to  station  voltmeter  to 
read  troUey  voltage 

Two-wire  battery  t 

1  (Zero  center) 

1  plugged  to  read  battery  and 
bus  voltage 

Two-wire   synchron- 

1 

1  per  switchboard  plugged  to 

ous  converter 

each  machine 

Two-wire  motor 

Ub) 

None  required 

Three-wire  generator 

2  (One  in  positive  and 

1  per  switchboard  plugged  to 

one    in    negative 

read  voltage  between  outside 

lead) 

wires  of  each  machine 

Three-wire  feeder 

2  (One  in  positive  and 
one     in     negative 
lead) 

None  required  ordinarily 

Three-wire  synchron- 

2 (One  in  positive  and 

1  per  switchboard  plugged  to 

ous  converter 

one     in     negative 

read  voltage  between  outside 

lead) 

wires  of  each  machine 

Three-wire  balancer 

1  (Zero  center)  (con- 

1 plugged  to  each  machine  of 

nected  in  neutral) 

the  balancer  set 

(a)  On  multiple-circuit  feeder  panels  controlling  feeders  of  small  capacity,  am- 
meters are  usually  omitted. 

'  (b)  On  small  motors,  ammeters  are  usually  not  furnished. 

(d)  Where  there  are  only  two  exciters  operating  in  parallel,  one  voltmeter  is  used 
on  each  exciter  equipment.  Where  there  are  three  or  more  exciters,  two  voltmeters 
are  employed  and  mounted  together  on  a  swinging  bracket  at  the  end  of  the  board, 
usually  on  the  same  bracket  containing  the  alternating  current  voltmeters  and  syn- 
chronism indicator.  One  is  connected  to  the  bus  and  the  other  is  arranged  to  be 
plugged  to  any  machine  to  read  voltage  at  any  time.  In  many  instances  exciters  are 
direct  connected  or  belted  to  the  alternating-current  machines,  the  fields  of  which  they 
excite,  and  are  not  operated  in  parallel,  no  separate  panels  being  furnished  to  control 
them.  In  such  cases  no  measuring  instruments  are  furnished,  the  field  ammeter  of 
the  alternating  current  machine  taking  the  place  of  the  exciter  ammeter,  while 
there  is  ordinarily  no  use  of  the  voltmeter. 

*  Where  the  different  types  of  circuits  given  in  the  first  column  occur  in  the  same 
board,  only  one  voltmeter  need  be  supplied,  providing  the  scale  is  suitable  for  the  volt- 
age of  all  circuits  to  be  measured. 

t  Due  to  the  large  number  of  methods  of  connecting  batteries,  no  definite  instru- 
ment equipment  can  be  listed  to  apply  to  all  cases.  The  above  represents  a  simple 
equipment  for  measuring  charging  and  discharge  current  and  voltages  as  indicated. 


Digitized  by 


Google 


548 


ELECTRICAL  EQUIPMENT 


U    U   U  tm   a 

«  «  o  o  g 

ggggl 

O  O  O  O  9 

lllp 

£fiS£S 

CO  ec^  *  J 

AAAAo 
coeoMoU 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


549 


43     n 

II 


.a.a  ja 
o  o  o 

«  B   CO 

»«     H      IM 

III 


»&& 
»^» 


CO        CO  CO  CO 

s  sss 


X3 

5 


I 

2 


3     3    5S3        3 

cc^ooticoeoco^^N 


H 
t 

5  o 

n 

•I 
I* 

llHi 


sll^-g 


.  o 


5i 


i? 


«  «  «   0)  ' 


llllllllfl 

coeoeococo'*'**^'^ 

siiiiiiiii 

a  Cki  o.  A  a  o.  A  ?  A  a 

coeoeocoeocoeoSc^lM 


%% 


?-   B  « 


'    «  C  d  O    0*0 


J?; 


5^ 


§    s 
2    s 


I 


s?:- 


III    \l 


-as  SB  s3 
Igfaia 

,  -bI  «?  •? 


sa    a 


Digitized  by 


Google 


660  ELECTRICAL  EQUIPMENT 

for  this  capacity.  When,  with  a  certain  capacity  of  current 
transformer  determined  by  the  load  of  the  circuit,  the  scale  of  the 
instriunent  would  be  too  large  to  allow  a  good  reading  at  light  loads, 
4-ampere  windings  may  be  used,  the  scale  then  being  about  80 
per  cent  of  that  corresponding  to  that  used  with  the  6-ampere 
winding.  Secondary  potential  coils  for  all  instruments  except 
voltmeters  are  ordinarily  wound  for  110  volts,  the  voltage  of  the 
secondary  side  of  standard  potential  transformers. 

Instruments  may  be  operated  from  the  same  cmrent  trans- 
formers which  are  used  with  the  oil  circuit-breaker  trip  coils  or 
relays,  providing  the  volt-ampere  load  is  such  that  the  accuracy 
of  the  instrument  and  transformer  combination  comes  within 
certain  set  limits.  Wattmeters,  however,  should  not  be  connected 
to  the  same  cm-rent  transformers  which  are  used  with  differential 
or  reverse  power  relays  or  with  compensated  voltmeters  (indi- 
cating or  contact-making)  or  Une-drop  compensators. 

The  same  potential  transformers  can  be  used  for  operating 
instruments  and  potential  coils  of  relays,  low-voltage  release  or 
other  apparatus  as  long  as  the  rated  secondary  volt-ampere  load  of 
the  transformer  is  not  exceeded.  This  load  and  its  power-factor 
must  be  clearly  distinguished  from  the  load  and  power-factor  of 
the  main  circuit  which  are  measured  by  the  measuring  outfit 
of  which  the  instrument  transformer  is  a  part. 

The  term  "  equivalent  secondary  connected  load  "  is  used  in 
connection  with  a  circuit  to  denote  the  volt-ampere  load  carried 
by  the  secondary  of  an  instriunent  transformer  when  this  load 
differs  from  the  result  of  combining  the  volt-amperes  of  the  separate 
devices  in  series  or  in  multiple  because  the  secondary  is  inter- 
connected with  other  instrument  transformer  secondaries.  The 
power-factor  of  the  equivalent  secondary  load  of  a  current  trans- 
former imder  these  conditions  is  also  affected  by  the  intercon- 
nection. 

The  volt-ampere  of  the  various  secondarj^  devices,  such  as 
indicating  instruments,  meters,  relays,  etc.,  varies  considerably 
and  should  be  obtained  from  the  manufacturer. 

The  secondaries  and  cases  or  frames  of  current  transformers 
should  be  grounded  whenever  possible.  The  switchboard  wiring 
should  be  carefully  considered  to  see  if  this  can  be  done  without 
interfering  with  the  proper  operation  of  the  instruments  connected 
to  the  transformers.    The  grounding  of  the  cases  serves  the  double 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  651 

purpose  of  protecting  the  switchboard  attendant  SLnd  freeing  the 
instruments  from  the  effects  of  electrostatic  charges  which  might 
otherwise  collect  on  the  cases  and  cause  errors. 

The  primary  of  current  transformers  should  never  be  left  in  the 
line  with  the  secondary  open-circuited,  as  this  will  set  up  a  heavy 
flux  through  the  core,  over-saturating  the  iron  and  causing  it  to 
greatly  overheat.  If  for  any  reason,  therefore,  it  becomes  neces- 
sary to  remove  the  meter  or  any  current-carrying  device  from  the 
secondary  circuit  of  a  current  transformer,  the  secondary  should 
be  short-circuited  by  a  wire  or  some  other  means.        -      . 

Potential  transformers  are  used  to  insulate  the  meters  from  the 
high  potential  circuit  as  well  as  to  do  away  with  a  large  amount 
of  resistance  in  series  with  the  meters  which  would  be  necessary 
if  the  meters  were  connected  directly  to  the  high-potential  cir- 
cuits. Except  in  special  cases,  they  are  generally  prptected  by 
fuses  inserted  in  the  primary  leads. 

The  connections  for  the  multitude  of  instruments,  meters, 
relays,  etc.,  with  their  citfrent  and  potential  transformers  which 
are  used  in  the  modem  power  station  are  very  intricate.  While 
for  individual  equipments  such  connections  may  be  standardized, 
the  combinations  used  in  a  large  station  are  generally  such  as  to 
make  the  connections  more  or  less  special  in  order  to  give  the  best 
results.  Individual  diagrams  are  as  a  rule  contained  in  the 
bulletins  issued  by  the  various  manufacturers,  and  the  making 
up  of  the  main  wiring  diagram  for  any  important  installation 
should  be  left  to  the  manufacturer  supplying  the  switchboard. 
A  typical  diagram  of  connections  for  an  individual  exciter,  an 
A.C.  generator  and  an  outgoing  feeder  is  shown  in  Fig.  345  as  an 
example. 

Key  to  Symbols 

A.  =  Ammeter. 

B.A.S.  =  Bell-alarm  switch. 

C.T.      =  Current  transformer. 

F.  =Fuse. 

F.A.      =  Field  anmieter. 

F.S.       =  Field  switch. 

G.C.S.  =  Governor-control  switch. 

K.S.      =  Knife  switch. 

L.S.       =  Limit  switch  (included  with  governor  motor). 

O.S.       =  Oil  switch. 


Digitized  by  CjOOQ IC 


562  ELECTRICAL  EQUIPMENT 

P.I.W.  *=Pol3rphafie  indicating  wattmeter. 
P.R.W.= Polyphase  watthour  meter. 


P.R. 

= Potential  receptacle. 

P.P. 

= Potential  plug. 

P.T. 

==  Potential  transformer. 

Rheo. 

«  Rheostat. 

S. 

«Shmit. 

S.R. 

= Synchronizing  receptacle. 

S.P. 

s  Synchronizing  plugs. 

T.B. 

-Terminal  board  for  secondary  leads  from  current  and 

potential  transformers. 

T.C. 

=Trip  coil  on  oil  switch. 

V. 

«  Voltmeter. 

Bzdter  and  Field  ControL  For  the  electrical  control  of 
exciter  circuits  it  is  usual  to  omit  fuses  or  other  overload  devices  in 
order  to  prevent  any  interruption  in  the  supply  of  field  current  to 
the  alternating-current  generators,  thereby  insimng  continuous 
operation,  which,  in  most  stations,  is  an  essential  feature  and 
is  of  more  importance  than  protection  of  the  exciters  from  dam- 
age. Also  as  an  insm-ance  against  injury  to  the  alternating- 
current  generator  field  windings.  When  trouble  occurs  in  the 
exciting  sjrstem  and  ope  is  the  overload  devices  on  all  the  exciters 
connected,  the  generator  field  circuits  are  broken  at  points  where 
no  discharge  resistances  are  interposed  and  the  generator  field 
windings  are  consequently  liable  to  pimcture  by  the  high-induced 
voltage  to  which  they  are  subjected.  If  overload  protection  is 
insisted  upon,  it  is  recommended  that  the  overload  devices, 
fuses  or  circuit  breakers,  be  based  on  double  the  normal  capacity 
of  the  exciter  so  as  to  open  only  in  case  of  very  serious  trouble. 

For  large  plants  having  a  number  of  exciters  in  parallel  and 
where  the  expense  involved  is  of  secondary  consideration,  it  is 
customary  to  provide  reverse-current  circuit  breakers  without 
any  overload  attachment.  The  reverse-current  device  serves  to 
disconnect  a  defective  exciter  while  the  remaining  exciters  con- 
tinue in  service. 

Circuits  for  motors  driving  exciters  are  usually  considered  as 
feeder  circuits  and  overload  protection  is  accordingly  recom- 
mended for  the  motor.  A  time-limit  device  is  preferable  for  this 
overload  feature,  and,  if  an  instantaneous  device  is  used,  it  should 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


553 


be  set  very  high.  When  operating  conditions  make  it  necessary, 
the  overload  feature  can  be  very  readily  disconnected.  With 
motor-driven  exciters  operating  in  parallel,  it  is  also  advisable  to 
equip  the  exciter  circuits  with  reverse-current  circuit  breakers,  so 


as  to  prevent  any  set  which  might  be  disconnected  from  the  bus 
on  the  motor  side  to  continue  to  operate  by  its  exciter  running  as  a 
motor  and  taking  power  from  the  exciter  bus.  The  D.C.  breaker 
could,  of  course,  also  be  provided  with  a  shunt  trip  arrangement 


Digitized  by 


Google 


554  ELECTKICAL  EQUIPMENT 

■whereby  the  opening  of  the  A.C  oil  circuit  breaker  would  in  turn 
trip  the  D.C.  breaker. 

For  small  and  medium  size  installations  the  field  switches  are 
usually  of  the  ordinary  knife  switch  type  mounted  directly  on  the 
main  switchboard.  For  large  installations  it  is,  however,  com- 
mon practice  to  .employ  solenoid-operated  carbon-break  circuit 
^•[breakers.  These  are  often  mounted  on  panels  near  their  respective 
exciters  so  as  to  reduce  the  length  of  connections  to  a  minimum, 
and  controlled  from  the  main  board. 

Occasionally  a  separate  direct-current  switchboard  is  provided 
and  located  at  some  convenient  place  near  the  exciters.  On  this 
board  is  then  mounted  all  the  exciter  and  field  switches  as  well  as 
other  low-voltage  switches  and  circuit  breakers  for  the  various 
station  circuits. 

Field  switches  for  disconnecting  the  individual  fields  of  the 
A.C.  generators  should  always  be  provided.  These  switches  are 
known  as  "  field  discharge  switches  "  because  their  design  is  such 
that  when  they  are  opened  a  discharge  resistance  is  automatically 
inserted  in  series  with  the  field  circuit.  If  this  should  be  suddenly 
broken,  an  excessively  high  potential  may  be  induced  in  the  field 
winding  which  might  puncture  its  insulation.  By  inserting  a 
resistance  in  the  circuit,  the  e.m.f.  induced  in  the  field  coils  by  the 
dying  magnetic  flux  produces  a  current  through  this  resistance; 
thus,  the  energy  stored  up  in  the  magnetic  field,  when  the  cur- 
rent was  compelled  to  increase  against  the  induced  counter  e.m.f., 
is  now  discharged  in  this  resistance  where  it  appears  as  heat. 
The  construction  of  the  switch  is  such  that  in  opening  the  same 
the  resistance  circuit  is  closed  before  the  field  is  disconnected 
from  the  exciter  or  field  bus,  while,  in  closing  the  switch  the 
tesistance  circuit  is  opened  before  the  field  is  connected  to  the 
.  exciter.  By  this  means  all  destructive  arcing  is  also  avoided, 
JFor  the  field  can  never  be  broken  without  shunting  it  through 
•the  discharge  resistance.     Certain  types  of  switches  are,  on  the 

•  other  hand,  provided  with  a  stop  so  that  they  cannot  be  com- 
pletely opened  until  this  has  been  withdrawn,  thus  giving  the 
induced  field  energy  time  to  be  dissipated  through  the  discharge 

•  dip  to'  the  discharge  resistance  before  the  circuit  is  broken. 

'  -  Field  switches  may  be  either  hand  operated  or  solenoid  oper- 
ated, similar  to  the  exciter  switches.  In  the  former  case  they  may 
be  identical  to  ordinary  knife  switches,  to  which  discharge  clips 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  .  555 

have  been  added,  and  mounted  on  the  front  of  the  panel.  It  is 
becoming  very  general  practice,  however,  to  mount  the  hve  part 
back  of  the  switchboard  and  operate  it  by  a  handle  from  the  front 
of  the  board.  This  type  of  field  switch  is  regarded  as  a  "  safety  . 
first  "  device  of  great  importance  and  is  to  be  recommended  in  all 
cases.  The  switchboard  attendant  cannot  come  in  contact  with 
the  live  parts  or  arc  when  operating,  and  instruments  and  other 
adjacent  equipment  are  safe  from  damage  by  burning  which 
occasionally  happens  with  the  front-of-board  type. 

With  benchboard  equipments  and  with  large  capacity  vertical 
switchboards  where  remote  control  is  desirable,  solenoid-operated 
field  switches  are  often  employed.  While  controlled  from  the 
main  board,  they  may  be  located  at  the  most  convenient  pointy 
for  example,  near  the  generators  or  on  the  exciter  board.  They 
are  similar  in  construction  to  the  non-automatic,  self-contained, 
solenoid-operated,  air  circuit  breaker  with  the  addition  of  a  dis- 
charge switch  (Fig.  346). 

Solenoid-operated  field  switches  for  A.C.  generators  and  for 
synchronous  motors  started,  as  is  usual  with  motors  of  250-volt 
excitation,  with  the  field  short-circuited,  should  be  double-pole 
with  common  closing  and  common  opening  coil.  No  provision 
is  made  for  automatically  interrupting  the  discharge  circuit 
after  the  switch  opens,  although  the  discharge  blade  can  be  ope- 
rated by  hand.  Where  economy  is  of  importance,  it  is  sometimes 
customary  with  A.C.  generators  to  provide  one  single  pole  solenoid- 
operated  field  switch  for  one  pole  and  ordinary  knife  switch  for 
the  other,  the  former  being  remote-controlled  from  the  main  board 
while  the  latter  is  hand-operated. 

With  sjmchronous  motors  Started  from  the  A.C.  side  with 
field  open  as  is  usual  with  motors  of  125-volt  excitation,  solenoid- 
operated  field  switches  are  made  ordinarily  of  two  single-pole 
elements  with  independent  opening  and  independent  closing 
coils.  Both  poles  close  simultaneously  and  connect  the  discharge 
resistance  across  the  field;  but  one  pole  precedes  the  other  a  short 
time  in  opening.  When  the  other  pole  opens,  the  discharge  cir- 
cuit is  interrupted. 

Occasionally  the  field  switch  has  been  used  to  cut  the  voltiage^ 
ofiF  a  machine  in  case  of  trouble  and  this  is  becoming  more  and> 
more  a  general  practice.  The  switch  is  then  equipped  with  a  shunt  - 
trip  and  an  overload  relay  is  installed  in  the  main  circuit,  in  wjiich- 


Digitized  by 


Google 


556 


ELECTRICAL  EQUIPMENT 


case  an  overload  in  the  latter  will  cause  the  field  switch  to  trip, 
thus  killing  the  voltage  of  the  generator. 

The  operating  mechanism  of  field  rheostats  depends  on  their 
size  which  in  tium  governs  their  location.    The  smallest 


Fia.  346.— ^lenoid-operated  Field  Switch. 

up  to  about  25  amperes,  can  usually  be  moimted  directly  back 
of  the  board,  and  it  is  only  necessary  to  extend  the  shaft  of  the 
rheostat  and  connect  it  directly  to  the  handwheel  on  the  front  d 
the  panel.    Concentric  handwheel    mechanisms  are  also    very 


Digitized  by 


Google 


SWITCmNG  EQUIPMENT 


557 


common,  one  of  the  wheels  being  for  the  exciter  field  rheostat 
and  the  other  for  the  main  generator  field  rheostat.  Such  arrange* 
ments  permit  of  quite  a  saving  in  the  space  required. 

For  larger  sizes  it  becomes  necessary  to  mount  the  rheostats 
remote  from  the  switchboard,  in  the  basement  or  otherwise. 
The  operating  mechanism  may  then  consist  of  a  sprocket-wheel 
chain  drive,  operated  by  a  handwheel  on  the  front  of  the  board, 
or  it  may  be  electrical,  either  in  the  form  of  ratchets  or  motors 
controlled  from  the  main  board. 
A  typical  arrangement  of  a 
sprocket-wheel  chain  drive  is 
shown  in  Fig.  347,  but  it  is,  of 
course,  evident  that  the  rheostat 
proper  can  be  located  in  many 
different  positions  than  what  is 
shown.  This  class  of  control  is 
generally  limited  to  rheostat 
capacities  of  up  to  about  350 
amperes. 

In  many  installations  it  is, 
however,  not  possible  to  locate 
the  rheostat  so  that  the  dial 
switch  can  be  operated  by  means 
of  chain  drive  from  a  hand- 
wheel  on  the  panel.  For  such 
conditions  the  rheostat  can  be 
equipped  with  an  electrically 
operated  ratchet  switch  (Fig. 
348),  which  can  readily  be  con- 
trolled from  the  main  board,  and 
the  rheostat  proper  can  be  located  in  any  part  of  the  station. 
The  capacity  is  limited  to  the  same  as  the  chain-operated  type, 
i.e.,  about  350  amperes,  and  the  operation  is  as  follows: 

The  switch  arm  is  carried  around  by  pawls  which  engage 
the  knurled  rim  of  a  wheel  to  which  the  switch  arm  is  rigidly 
fastened.  These  pawls  are  controlled  by  a  core  actuated  in  com- 
mon by  the  solenoids  AA.  When  the  solenoids  are  de-energiaed 
the  pawls  are  disengaged  and  in  their  normal  position  rest  equi- 
distsint  from  the  solenoids.  To  cut  resistance  into  the  field,  it 
is  necessary  to  close  to  the  left  the  single-pole  switch  B.    This 


Fig.  347.— Sprocket-wheel  Chain 
Drive  for  Field  Rheostats. 


Digitized  by  VjOOQ IC 


558 


ELECTRICAL  EQUIPMENT 


energizes  the  left-hand  solenbidi  engages  the  leflrhand  paT'I  and 
movea  the  dial  switch  in  a  clockwise  direction.  When  the  solenoid 
core  has  reached  its  extreme  point  of  travel,  the  winding  of  the 
solenoid'  is  automatically  open-circuited  by  the  small  switch  C, 
and  the  pawl  is  immediately  pulled  to  its  neutral  position  by  a 
Spritig,  automatically  closing  the  circuit  of  the  solenoid  switch  by 
tjhe  small  switch  C«  The  same  cycle  of  operation  is  then  repeated 
until  the  switch  B  is  opened.  If  it  be  desired  to  cut  resistance 
out  of  the  field  circuit  the  single-pole  switch  Bis  closed  to  the  right 
when  the  same  cycle  of  operation  is  performed  and  the  dial  switch 


Looking  at  Face  of  Switch 


Fio.  348. — Connections  of  Solenoid-operated  Ratchet-driven  Field  Rheostat 

Switch. 

ijaoves  in  a  (Soimter-clockwise  instead  of  a  clockwise  direction. 
Each  end  of  the  switch  dial  is  provided  with  a  limit  switch,  D, 
which  is  automatically  operated  by  the  switch  arm  to  open  the 
circuit  of  the  solenoid  when  the  resistance  is  entirely  cut  in  or 
out.  The  purpose  of  the  limit  switch,  D,  is  simply  to  protect  the 
apparatus  in  case  the  controlling  circuit  is  left  closed  when  the 
difij  swdtch  has  reached  its  extreme  point  of  travel  in  either 
direction. 

For  circuits  above  300  to  350  amperes  the  motor-operated  type 
of  rheostat  (Fig.  349)  is  the  most  practical,  as  the  heavy  contact 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


559 


on  the  dial  switch  is  not  easily  overcome  with  the  solenoid  or  hand- 
wheel  control.  The  motor  is  of  the  series  type  with  a  field  wind- 
ing enabling  the  dial  switch  to  be  operated  in  either  direction 
by  the  control  switch  on  the  main  board.  As  with  the  ratchet- 
driven  type,  each  end  of  the  switch  dial  is  provided  with  a  limit 


?f  *  3¥^ 


Fig.  349. — Electrically  Operated  Motor-driven  Rheostat 

switch  which  is  automatically  operated  by  the  switch  arm  to  open 
the  motor  circuit. 

Voltmeter  and  Synchronizing  Receptacles.  These  are  devices 
which  provide  a  ready  means  for  connecting  a  voltmeter  to  any 
machine  or  any  phase  of  the  same  and  thus  reduce  the  number 


Digitized  by 


Google 


560 


ELECTRICAL  EQUIPMENT 


of  instruments  required.  Also,  for  making  the  necessary  connec- 
tions at  the  time  of  synchronizing.  The  contact  elements  are  of 
brass  and  come  through  the  panel  to  the  front,  but  are  counter- 
sunk in  a  hard  rubber  escutcheon  plate,  which  makes  accidental 
contact  very  unUkely.    The  plugs  have  brass  contacts  supported 

by  a  hard  rubber  shield,  which 
also  serves  as  a  protection  to 
the  hand. 

As  will  be  noted  from  the 
diagram  of  connections  (Fig. 
345),  eight-point  voltmeter 
receptacles  are  provided  for 
the  A.C.  generator  so  that 
the  voltage  across  all  the  three 
phases  can  be  read  in  turn 
when  the  plug  is  inserted. 

With  the  synchronizing 
scheme,  as  shown  in  Fig.  345, 
the  synchronizing  is  actually 
done  between  the  machines. 
For  this  reason  two  plugs  are 
required,  one  of  which  is  in- 
serted in  the  receptacles  of  one 
of  the  machines  which  is  run- 
ning and  the  other  in  the  re- 
ceptacles of  the  machine 
which  is  to  be  started  and 
sjrnchronized. 

Ammeter  Transfer  Recep- 
tacles. These  are  for  reading 
the  current  in  any  of  three 
phases  on  one  ammeter  by 
changing  the  connections  from 
the  front  of  the  panel.  Each  unit  of  a  group  consists  of  a  brass 
plug  switch  receptacle  with  fiber  insulation,  with  contacts  back 
of  the  panel  and  with  a  molded  bushing  on  the  front.  For  read- 
ing the  current,  the  transfer  plug  is  inserted  in  rotation  in  each 
of  the  three  receptacles  of  a  group.  Between  such  readings  the 
plug  can  be  left  inserted  in  one  receptacle,  thus  giving  a  continu- 
ous indication  on  that  phase. 


Fig.  350. — Automatic  Throw-over 
Switch. 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  561 

Throw-over  Switches.  A  sudden  failure  of  the  source  of 
power  for  the  Ughting  system  in  the  power  station  is  a  more  or 
less  frequent  and  troublesome  occurrence.  To  take  care  of  such 
an  emergency  and  facilitate  the  re-estabUshment  of  normal  con- 
ditions where  apparatus  may  have  been  shut  down  due  to  the  fail- 
lu-e  of  power,  a  switch  for  automatically  throwing  the  lights  to  an 
auxiliary  or  reserve  source  becomes  very  handy.  The  switch 
shown  in  Fig.  350  accomplishes  this  result.  The  device  consists 
of  a  special  double-throw  switch  held  closed  by  a  latch  on  one 
throw  against  a  pair  of  springs. 

To  close  the  hghting  circuit  with  the  normal  source  of  power  in 
operation,  the  switch  is  thrown  in  the  lower  set  of  contacts  and 
latched  in  the  closed  position  by  hand.  When  a  failure  of  the 
source  occurs,  a  low-voltage  release  is  caused  to  drop  its  armature, 
tripping  the  latch  free  from  the  crossbar  above  it.  The  springs 
on  the  hinge  cUps  of  the  switch  then  quickly  force  the  switch  into 
the  upper  set  of  contacts,  which  are  connected  to  %he  reserve 
source  of  power.  At  the  same  time  an  auxiliary  switch  at  the 
top  is  thrown  into  contact,  causing  a  bell  or  other  indicator  to 
operate  to  attract  the  station  attendant's  notice.  After  the 
resumption  of  normal  conditions,  the  switch  must  be  thrown  by 
hand  into  the  lower  contacts  and  latched. 

Calibrating  Terminals.  A  quick  and  convenient  method  of 
making  connections  for  caUbrating  instrmnents,  etc.,  is  very 
desirable,  and  this  has  led  to  a  very  general  use  of  providing 
calibrating  terminals  on  all  important  switchboards.  These  may 
be  mounted  either  on  the  front  or  back  of  the  panels,  the  choice 
being  governed  by  the  conditions.  For  example,  where  it  is 
difficult  to  carry  on  such  tests  on  the  back  of  a  board,  the  ter- 
minals may  readily  be  mounted  on  the  front,  while  if  there  is 
plenty  of  room  in  the  rear,  it  may  be  advantageous  to  locate  the 
caUbrating  terminals  there  in  order  to  utilize  the  space  on  the 
front  otherwise. 

The  terminals  for  the  ciu-rent  transformer  connections  should 
be  such,  that  the  testing  instrument  can  be  connected  in  the  cir- 
cuit without  breaking  the  continuity  of  the  circuit,  as  explained 
under  "  Current  Transformers." 

Control  Switches.  Remote  electrically  operated  oil  or  air 
circuit  breakers  are  controlled  by  small  double-throw  control 
switches,    usually   mounted   on   the   main   switchboard.    How- 


Digitized  by 


Google 


562 


ELECTRICAL  EQUIPMENT 


Olodng  _  ^ 


OpermtiogBoaes 


-mA    Auxiliarj 

,^-Yy^    Switch 


Red  L*inp 


Cl<M!Dff_ 

Contact 

Openiiiff_ 

ConlAct 


ll 


^ 


jOpening 
CoU 


RMistaaoe  for 
-  lue  onlj  on 
269  V.  or  0UO  V, 


ever,  since  the  energizing  current  of  the  operating  mechanism  may 
be  considerable,  such  as  for  motor-operated  breakers  or  for  the 

closing  coil  of  solenoid  oper- 
ated breakers,  it  is  not  custom- 
ary to  rely  on  the  control 
switch  for  breaking  this  cur- 
rent, and  an  intermediate  con- 
trol relay  (Fig.  332)  is  pro- 
vided for  this  purpose.  The 
operating  coil  of  this  control 
relay  is  then  connected  across 
the  closing  contacts  of  the 
control  switch  and  the  relay 
contacts  in  series  with  the 
motor  circuit  or  the  solenoid 
closing  coil  (Mg.  351). 

Control    switches    should 

always  be  designed   so  that 

all  connections  may  be  made 

on  the  back  of  the  panel,  and 

80  as  to  render  it  impossible  to  operate  by  accidentally  leaning 

against  the  switch.    This  is  accomplished  in  the  "  pull-button  " 

type,  which  has  the  contacts  on  the  back  of  the  panel,  with  pull 


.Control 


Switch 


Green  Lamp 


Opemting  Bn— 


Fig.  351. — Connection  for  Control 
Switch  for  Direct-current  Solenoid- 
control  Circuits. 


Fig.  352.— Pull-button  Control  Switch. 

rods  brought  through  the  panel  to  the  handles  on  the  front  (Fig. 
352).    The  switch  returns  to  the  open  position  by  reason  of  a 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  563 

spring  and  both  throws  (closing  and  opening),  are  interlocked.  It 
is  provided  with  a  mechanical  device  to  indicate  which  throw  was 
last  closed  and,  in  addition,  with  red  and  green  bulPs-eye  lamps  to 
indicate  the  actual  position  of  the  circuit  breaker.  The  necessary 
auxiliary  switches  for  these  lamps  are  provided  with  the  breaker. 

Mimic  Buses.  It  is  sometimes  customary  to  place  on  the 
switchboard  copper  connections,  known  as  mimic  buses,  rep- 
resenting the  main  connections  of  the  station.  These  are  often 
desirable  as  they  keep  before  the  operator  the  whole  arrange- 
ment of  the  circuits,  enabling  him  to  see  at  a  glance  what  is  the 
proper  switch  to  open  or  close.  On  the  other  hand,  their  use  may 
sometimes  cause  either  a  crowded  or  uns3anmetrical  arrange- 
ment. 

Figs.  337,  342,  or  344,  illustrate  the  use  of  such  mimic  buses. 

Bus  and  Switch  Structures.  As  previously  stated,  bus-bars  or 
electrically  operated  oil  circuit  breakers  are  not  necessarily  placed 
near  the  controlling  switchboard,  but  should  be  placed  with  con- 
venience to  connections  and  safety  from  fire  and  in  handling. 

Isolating  barriers  or  compartments  are  recommended  for 
voltages  up  to  15,000  where  the  capacity  is  above,  say,  5000  Kw. 
in  order  to  prevent  any  destructive  effects  of  short-circuits  from 
spreading  and  involving  the  entire  bus  structure. 

Furthermore,  the  compartments  act  as  a  guard  against  anyone 
touching  the  exposed  parts  of  the  buses  and  breakers  and  gives  a 
certain  amount  of  fimsh  and  completeness  to  the  station.  The 
cost  of  the  cell  structure  is  not  of  great  consideration  and  is  only  a 
small  percentage  of  the  total  cost  of  the  station. 

For  higher  voltages  the  currents  naturally  become  corre- 
spondingly less,  minimizing  the  destructive  effects  of  short-circuits, 
and,  on  the  other  hand,  the  spacings  required  are  greater  so  that 
open  work  generally  becomes  preferable. 

Various  materials  have  been  used  for  bus  and  oil  circuit 
breaker  compartments,  namely,  brick,  concrete,  soapstone  and 
slate,  and  sometimes  a  combination  of  brick  with  one  of  the  other 
materials.  Brick  compartments  are  the  cheapest  and  if  properly 
made  give  the  best  appearance.  The  use  of  common  brick  is, 
however,  not  recommended  because  most  of  the  walls  are  four 
inches  thick  and  the  sizes  of  the  brick  vary  so,  while,  on  the  other 
hand,  the  bonds  are  so  large  that  a  neat  job  cannot  generally  be 
obtained.    Inasmuch  as  the  cost  of  laying  the  brick  is  about  75 


Digitized  by 


Google 


664  ELECTRICAL  EQUIPMENT 

per  cent  of  the  total  cost|  very  little  is  added  by  substituting  a 
face  brick.  With  this  type  of  construction  the  compartment 
shelves  are  generally  made  of  concrete  or  soapstone,  from  2  to  3 
inches  thick,  depending  on  the  size  of  the  compartment. 

Concrete,  although  more  costly,  has  gained  in  favor  over 
brick  work,  and  therefore  the  majority  of  bus  and  switch  compart- 
ments nowadays  are  built  of  concrete,  especially  for  the  laiiger  sta^ 
tions.  In  some  cases  complete  forms  are  made,  usually  of  wood, 
and  the  whole  compartment  poured,  giving  a  very  substantial 
construction.  It  is  more  often  the  case,  however,  that  concrete 
slabs  are  used,  set  in  cement. 

The  general  dimensions  of  bus  and  switch  compartments  are 
determined  by  the  minimum  distance  allowable  between  conductors 
and  ground  (see  table  LII,  page  627),  the  brick  or  concrete  being 
considered  as  ground.  The  switching  apparatus  also  governs  to 
a  great  extent  the  dimensions  of  the  compartment,  although  even 
here  it  is  generally  a  matter  of  ground  distance  in  the  apparatus. 
For  mechanical  reasons  and  accessibility  the  distances  are  gen- 
erally increased  somewhat;  this  also  to  guard  against  joints, 
clamps  or  bolts  acting  as  spillways  at  times  of  abnormal  voltage 
rises  on  the  system.  Low-voltage  compartments,  where  relatively 
heavy  copper  is  used,  should  have  proportionally  more  liberal  dis- 
tances than  those  for  equal  capacities  but  of  higher  voltages,  with 
connections  of  smaller  size. 

Removable  doors  are  recommended  for  all  openings  of  compart- 
ments to  prevent  accidental  contact  with  live  parts,  and  in  the 
case  of  oil  circuit  breakers,  to  prevent  the  scattering  of  oil  should 
it  be  forced  out  of  the  oil  vessels.  Compartment  doors  should 
be  made  of  light,  fireproof  material  and  swimg  from  the  top  to 
allow  free  movement  in  case  of  explosion  in  the  compartment. 
Asbestos  lumber  with  a  light  wood  frame  has  proved  to  be  the  most 
satisfactory  construction  for  compartment  doors.  Compartment 
doors  should  be  considered  as  ground,  that  is,  in  respect  to  all  live 
parts. 

The  arrangement  of  switch  and  bus  structures  varies  consider- 
ably, depending  not  only  on  the  S3rstem  of  connections,  but  also 
on  the  different  designs  of  the  circuit  breakers.  It  is  therefore 
impossible  to  give  any  definite  recommendations  that  wiD  meet 
all  conditions.  In  addition  to  the  illustrations  shown  in  the  sec- 
tion on  "  Arrangement  of  Apparatus,"  page  175,  Pigs.  363  to  357 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  565 

are  given,  which  show  some  typical  arrangements  which  are  self- 
explanatory. 

In  la3ring  out  the  structure  attention  should  also  be  given  to 
the  current  and  potential  transformers.  The  latter  with  their 
fuses  require  considerable  space  for  higher  voltages  and  have  to  be 
installed  in  certain  positions.  This  refers  especially  to  oil-cooled 
transformers  and  expulsion  fuses,  so  that  if  in  the  preliminary 
design  these  points  are  not  taken  into  consideration  considerable 
diflSculty  may  be  encountered  in  finding  suitable  accommodation 
for  them.  When  current  and  potential  transformers  are  installed 
in  separate  compartments,  holes  should  be  left  in  the  partition 
walls  to  accommodate  conduits  for  the  secondaries  between 
phases,  and  in  case  of  potential  transformers  porcelain  bushings 
should  be  provided  for  the  primaries. 

For  voltages  above  15,000  the  circmt  breakers  are,  as  a  rule, 
of  the  top-connected  tank  construction  and  compartments  are 
entirely  omitted,  especially  for  the  higher  voltages.  The  conduct- 
ors must  necessarily  be  spaced  farther  apart  and  at  a  consider* 
able  distance  from  the  floor,  so  as  to  be  out  of  reach.  Different 
arrangements  are  used  for  nearly  every  new  station,  as  seen  from 
the  illustrations.  Figs.  93  to  101. 

The  busbars  are  an  important  part  of  the  installation,  carrying- 
the  whole  energy  of  the  plant  in  a  confined  space.  The  material 
is  usually  copper  and  the  conductors  may  be  either  cylindrical 
rods  or  tubes  or  rectangular  bars.  The  former  are  generally 
used  for  the  high-tension  ouses  and  connections,  but  the  latter  are 
essential  for  lower  voltages  where  large  currents  are  to  be  carried, 
necessitating  a  larger  cross-section.  In  such  cases  the  bus  is 
laminated,  i.e.,  it  consists  of  a  number  of  bars  arranged  side  by 
side  with  ventilating  ducts  between.  This  insures  a  large  radiat- 
ing surface,  while  at  the  same  time  this  construction  permits  a 
tapering  of  the  bus  so  as  to  utilize  the  material  to  the  best  advan- 
tage. Additional  bars  may  also  readily  be  added  in  case  the 
capacity  needs  to  be  increased  in  the  future. 

The  buses  as  well  as  the  connections  to  the  oil  circuit  breakers, 
etc.,  should  be  so  proportioned  as  not  to  attain  an  excessive  tem- 
perature rise  under  the  maximum  current  which  they  are  intended 
to  carry.  For  direct-current  work  the  features  affecting  the  tem- 
perature rise  are  the  size  of  the  bar,  the  number  of  laminations, 
spacing  of  laminations,  spacing  between  poles,  whether  the  bars 


Digitized  by 


Google 


566 


ELECTRICAL  EQUIPMENT 


axe  run  flat  or  on  edge,  and  whether  open  or  enclosed  in  compart- 
ments. For  alternating-current  work  the  heating  in  addition 
depends  on  the  skin-effect  and  the 
inherent  reactance  of  individual 
laminations  and  phases. 

The  permissible  heating  will  de- 
pend on  the  fact  whether  these  bus- 
bars are  simple  uninterrupted  car- 
riers of  electricity  from  one  end  to 
another,  or  whether  connections 
are  taken  off  the  bus  at  certain 


Fig.  353. 


Fig.  354. 


Typical  Low-tension  High-capacity  Switch  and  Bus  Structures. 

points  to  circuit  breakers,  etc.  In  the  latter  case  the  heating  of 
the  bus-bars  or  of  the  whole  combination  from  bus  to  circuit 
breaker  must  be  kept  at  a  low  enough  figure  so  that  the  total 
temperature  rise  is  below  the  temperature  rise  permitted  for  the 
breaker,  which  generally  is  30°  C.  The  connection  bars  should, 
therefore,  in  such  cases  be  so  proportioned  as  not  to  develop  a 


Digitized  by 


Google 


I 


m;^^Mi^^¥zm^i^''^^^^:i 


^ 


Fia.  355. 


Fig.  356.  Fia.  357. 

Typical  Low-tension  High-capacity  Switch  and  Bus  Structures. 

567 


Digitized  by 


Google 


568 


ELECTRICAL  EQUIPMENT 


temperature  rise  in  excess  of  this  value  and  the  bus-bars  not  in 
excess  of  35°  C.  above  the  ambient  temperature. 

The  curves  in  Fig.  358,  which  have  been  derived  from  a  large 
number  of  actual  tests,  show  how  the  current  density  in  amperes 
per  square  inch,  based  on  a  30°  C.  rise,  will  vary  in  accordance  with 
the  number  and  width  of  lamination.  The  bars  are  I  inch  thick 
and  run  on  edge,  and  the  spacing  between  the  laminations  is  also 
i  inch  and  between  the  centers  of  the  phases  8  inches. 

The  great  variations  in  the  density  for  the  dififerent  conditions 


Kninber  of  Lamlnatloitt 
1      2      3       4'     6      « 


1300 

i 

i 

i 

OBOO 

\ 

\ 

1100 

^ 

J 

Ban 

v\ 

l\ 

1000 

^ 

^» 

•" 

w 

^ 

\u 

V 

V 

\ 

^ 

;bu 

\\ 

I 

^ 

<> 

•^ 

W 

■ :: 

Y 

Vu 

::t 

Bui 

no 

^ 

X 

'^ 

V 

\   \ 

\ 

) 

W 

•"J 

" 

] 

Kn 

^ 

s 

\v 

\ 

R^ 

<:$ 

\ 

^ 

IfM 

s 

^ 

^ 

bii 

SOO 

s 

N 

[>liec 

iCu 

rrenl 

All 

25 

Ung 
Cyc 

Cor 
lea 

eat 

Alt 

ernat 

:ycl« 

3nrr4 

ent 

Fig.  358. — Permissible  Amperes  per  Square  Inch  in  Copper  Connections. 


Installed  in  Open  Air  on  Edge. 
i"  Spacing  between  Laminations. 
Laminations  i"  thick. 


8"  Spacing  between  phases 
30*^  C.  Temperature  Rim. 


is  apparent  from  the  curves.  An  increase  in  the  spacing  between 
laminations  from  J  inch  to  J  inch  will  naturally  increase  the  ven- 
tilation, and  thereby  the  permissible  cm'rent  which  can  be  carried 
at  30°  C.  rise,  at  least  on  direct-current.  For  several  lamina- 
tions, run  flat,  that  is,  with  their  width  parallel  to  the  floor,  the 
heating  will  be  at  least  25  per  cent  greater  than  when  the  bars 
are  run  on  edge.  Furthermore,  consideration  must  be  given  to 
the  fact  that  the  ventilation  of  buses  in  compartments  is  not  as 
good  as  in  the  open,  and  for  this  reason  it  will  generally  be  advis- 
able to  limit  the  temperature  rise  for  such  conditions  to  a  figure 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


56d 


iBaraperpluae 


i 


i  Bars  per  phase 


somewhat  below  the  permissible  temperature  rise  of  buses  in  the 
open. 

Skin-effect  can  best  be  taken  care  of  by  arranging  the  bus-bars 
so  as  to  simulate  a  cylinder  or  tube,  and  this  is  done  by  running 
the  laminations  as  much  as  possible  in  pairs,  as  shown  in  Fig.  350. 
The  distance  between  the  pairs  should_  then 
be  as  great  as  the  space  of  the  bus-bar  com- 
partments will  permit. 

With  the  bars  run  flat  in  the  compart- 
ments, the  connections  can,  as  a  rule,  be 
made  easier,  but,  as  previously  stated,  the 
ventilation  becomes  poorer  than  if  run  on 
edge.  On  the  other  hand,  installing  them 
on  edge  gives  a  more  substantial  construc- 
tion in  that  it  increases  their  strength  and 
abiUty  to  withstand  short-circuit  stresses. 

With  alternating  current  bus-bars  run 
flat,  the  reactance  of  the  laminations  in 
the  outside  phases  varies  quite  consider- 
ably, this  effect  being  more  noticeable  the 
less  the  distance  between  phase  centers. 
The  effect  of  this  difference  of  the  inductive 
reactance  in  the  bars,  due  to  the  different 
distance  between  the  middle  phase  and  the 
individual  laminations,  will  cause  the  lami- 
nation nearest  the  middle  pha^  to  develop 
the  least  reactance,  and  the  lamination 
farthest  away  from  the  middle  phase  to  develop  the  highest 
reactance.  Therefore,  the  lamination  nearest  the  middle  phase 
will  carry  the  highest  ciu-rent  and  the  bar  farthest  away  from  the 
middle  phase  the  lowest  current.  If  the  bus-bars  are  placed  on 
edge  this  difference  of  inductive  reactance  in  the  laminations 
disappears,  and  the  only  effects  to  be  looked  out  for  on  A.C. 
bus-bars  is  then  the  matter  of  ventilation  and  skin-effect. 

Both  the  buses  and  the  connections  should  be  securely  sup- 
ported and  the  insulators  should  be  bolted  or  clamped  to  the  wall 
or  slab  and  not  cemented,  since  this  construction  causes  consider- 
able inconvenience  when  it  becomes  necessary  to  exchange  an 
insulator.  Several  different  lines  of  bus-bar  supports  are  now  on 
the  market,  two  typical  types  being  illustrated  in  Figs.  360  and 


Q  Bars  per  phase 

Fig.  359.— Method  of 
Paiiing  Bus-bars  to 
Reduce  Skin  Effect. 


Digitized  by 


Google 


670 


ELECTRICAL  EQUIPMENT 


Fig.  360.— Bus  Insulator; 
Bus  Laid  Flat. 


Fig.  361. — Bus  Insulator; 
Bus  on  Edge. 


361.  The  former  is  for 
mounting  the  buses  on  edge 
and  the  latter  on  the  side  or 
flat. 

In  stations  of  large  capac- 
ity precautions  should  be 
taken  in  supporting  the 
buses  in  the  compartments, 
due  to  the  great  stresses 
which  are  exerted  imder 
short-circuit  conditions.  This 
subject  is  dealt  with  in 
detail  in  the  section  on 
"Current-limiting  Reactors," 
page  458.  Fig.  362  shows 
the  design  for  a  support  to 
be  used  under  such  condi- 
tions. It  consists  of  two 
porcelain  insulators,  fitted 
loosely  into  the  horizontal 
compartment  barriers,  as 
shown.  Two  alloy  clamps  of 
similar  design,  held  apart  by 
four  brass  pillars  fitting 
loosely  into  holes  in  the 
clamps,  form  the  support  for 
the  bars.  The  top  damp 
has  a  threaded  stud  extend- 
ing into  a  hollow  in  the  top 
insulator.  By  tightening  the 
nut  on  this  stud  against  the 
top  insulator,  the  whole  sup- 
port is  held  firmly  in  place. 
By  loosening  this  nut  to  the 
limit  of  its  travel  against  the 
top  clamp,  it  is  possible  to 
lift  the  top  clamp  for  the  re- 
ception of  new  laminations 
of  bus  or  to  remove  the  top 
insulator,    there    being    just 


Digitized  by  CjOOQ IC 


SWITCHING  EQUIPMENT 


571 


enough  play  to  permit  it  to  clear  the  top  stud.  Subsequently 
the  remaining  parts  of  the  support  can  be  easily  removed  for 
repair  or  inspection.  The  individual  laminations  of  the  bus  are 
separated  by  fillers,  and  the 
number  of  laminations  can  be 
varied  at  will  by  using  pillars 
of  the  proper  length. 

The  bus  supports  should 
be  located  near  openings  in 
the  compartments  so  as  to  be 
accessible  for  cleaning  and 
inspection  (Fig.  363).  This 
also  refers  to  all  the  clamped 
joints  between  the  buses  and 
the  connections. 

For  very  high  voltages  the 
buses  generally  consist  of 
round  copper  rods  or  tubing, 
the  sizes  given  in  Table  LVI, 
page  638,  being  quite  com- 
mon. These  buses  are  gener- 
ally supported  from  the  roof 
trusses  by  suspension  insula- 
tors and  the  connections  on 
post-type  insulators  mounted 
on  the  walls  (Fig.  364). 

For  long  buses,  provision 
must  also  be  made  for  expan- 
sion and  contraction  due  to 
temperature  changes.  The 
diagram  in  Fig.  365  gives  the 
linear  expansion  of  copper 
buses,  the  values  being  based 
on  an  installation  tempera- 
ture of  25°  C.  =  75°  F.  The  actual  expansion  over  any  tem- 
peratiu^  range  on  the  chart  is  the  algebraic  sum  of  the  expansion 
values  shown  for  the  temperature  limiting  range.  The  chart 
has  been  corrected  for  variations  in  the  coeflScient,  and  the  actual 
temperatiu'es  should,  therefore,  be  used. 

The  problem  of  bringing  a  high-tension  wire  out  of  a  building 


Fig.  362. — Bus-bar  Support  for  Large 
Capadties  in  Compartments. 


Digitized  by 


Google 


572  ELECTRICAL  EQUIPMENT 

is  similar  to  bringing  one  out  of  a  transformer.  It  is  usually 
best  to  bring  the  high-tension  conductors  out  through  the  roof, 
although  in  some  cases  a  wall  outlet  may  be  advantageous.  No 
fixed  rule  can  be  made  in  this  respect  since  the  method  depends  on 
the  particular  layout,  arrangement  of  buses,  disconnecting  switches, 
and  lightning  arresters.     For  pressures  of  100,000  and  higher,  the 


Fig.  363. — Low-tension  Bus  Compartments. 

weight  of  the  outlet  bushings  and  their  great  size  as  well  as  the 
required  ground  clearance  from  steel  must  be  taken  into  consid- 
eration when  designing  the  roof.  Figs.  366  and  367  show  two 
typical  designs  of  line  entrances. 

Owing  to  the  cost  of  providing  suitable  buildings  for  trans- 
formers and  switching  equipments  operating  at  very  high  poten- 
tials, the  question  of  placing  this  apparatus  outdoors  is  one  that  is 
receiving  a  great  deal  of  attention.    Numerous  transformer  and 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


573 


switching  stations  of  this  kind  are  in  successful  operation,  and, 
while  the  practice  has  only  been  in  connection  with  a  few  generat- 
ing stations,  the  results  obtained  from  these  installations  have 
clearly  demonstrated  the  practicabihty  of  such  a  design.  Notable 
among  such  systems  is  that  of  the  Utah  Light  &  Power  Company. 
The  high-tension  buses  and  connections  together  with  the  dis- 
connecting switches,  choke  coils,  lightning-arrester  horn  gaps,  etc., 
are  generally  mounted  on  steel  structures  or  trusses  supported  on 
towers,  the  layout  being  governed  by  the  equipment  and  the 
method  of  control  which  has  been  adopted.  The  line  wires 
should  be  securely  anchored  before  entering  the  station  structure 


Fig.  364. — ^Tjrpical  High-voltage  Bus  and  Switch  Structure. 

and  no  unnecessary  strains  should  be  permitted  in  the  wires  inside 
the  structure.  Consideration  should  be  given  to  deflections 
resulting  from  different  pulls  on  the  connections  and  also  to  un- 
equal settlement  of  supporting  towers,  which  may  readily  cause 
excessive  stresses  and  insulator  breakages,  resulting  in  service 
interruptions.  The  spacing  of  all  the  conductors,  as  well  as  that 
of  apparatus  should  be  liberal  but  not  large. 

The  oil  circuit  breakers  and  transformers  are  generally  located 
on  the  ground,  the  oil  circuit  breakers  being  placed  below  the  dis- 
connecting switches.  It  is  often  desirable  to  provide  some  sort 
of  housing  or  roofing  for  partially  protecting  the  oil  circuit  break- 


Digitized  by 


Google 


674 


ELECTRICAL  EQUIPMENT 


ers,  and  where  low-tension  switching  equipments  and  attendance 
are  required  a  small  building  must  necessarily  be  installed.  Such 
a  building  can  then  contain  also  a  repair  shop,  storage-battery 
equipment  for  operating  the  oil  switches,  etc.  The  transformers 
should  be  placed  on  concrete  foundations  of  a  sufficient  height  to 
be  clear  of  water,  and  the  stations  should  further  be  well  paved 


i 
.f 

.6 

II 

.6 

// 
*     .4 

II 

(, 

•S  .0 
-/ 

25** 

77 

o 

r 

/ 

■/ 

- 

1 

1 

/ 

— 

/ 

r 

// 

lA 

i/ 

1 

'A 

/ 

- 

1 

/ 

^ 

y 

/ 

■ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

/ 

! 

/ 

/ 

/ 

/ 

/ 

/ 

4 

-.fl" 

-41 

/ 

/ 

/ 

/ 

V 

/ 

/ 

/ 

1 

/ 

/ 

/ 

-20" -10"    0'    10°  gomes'*)"  40"   m"   eo°  7o*'77'80'  90**  100"  no*  lar  iso*  i4or 

Temperature  Degrees 
Fig.  365. — Linear  Expansion  of  Copper  Bus-bans.^ 

and  drained  around  the  apparatus.  Transfer  tracks  with  a 
truck  will  also  be  found  very  convenient  when  moving  the  appa- 
ratus. Cement  walks  should  be  laid  on  that  portion  of  the  ground 
where  the  operator  is  most  apt  to  pass  in  his  inspection  trips  and 
work  about  the  place.  The  oil  piping  to  the  transformers  and 
» By  courtesy  of  General  Devices  and  Fittingis  Company. 


Digitized  by 


Google 


•  SWITCHING  EQUIPMENT 


575 


Fig.  366.--Typical  Wall  Entrance  for  Moderate  Voltage. 


Fig.  367.— Typical  Roof  Entrance  for  High  Voltage. 


Digitized  by 


Google 


576 


ELECTRICAL  EQUIPMENT 


•°a"*"»»nj:        9nri  jirasai^ 


Digitized  by  CjOOQ IC 


SWITCHING  EQUIPMENT 


577 


switches,  and  the  water  piping,  if  water-cooled  transformers  are 
provided,  should  be  so  arranged  that  connections  can  be  made  or 
broken  for  any  unit  without  disturbing  the  operation  of  the  other. 

Figs.  368  to  370  illustrate  typical  outdoor  arrangements, 
and  Fig.  371  shows  how  the  low-tension  leads  can  be  brought  from 
the  building  through  tunnels  to  the  outdoor  structure.  The 
leads  shown  in  the  illustration  come  from  the  low-tension  terminals 
of  a  transformer  located  above. 

Disconnecting  Switches.  In  all  high-tension  circuits  it  is 
customary  to  install  knife-type  disconnecting  switches  for  iso- 
lating oil  circuit  breakers,  feeders,  etc.,  and  for  making  various 


Fig.  369.— 66,000-volt  Outdoor  Substation.  ; 

'i 

connections  that  do  not  have  to  be  opened  under  load.  For' 
voltages  of  2500  or  less,  these  disconnecting  switches  are  mounted  [ 
directly  on  a  base  of  marble  or  similar  material,  while  for  higher 
voltages  post  insulators  of  various  kinds  mounted  on  pipe  work 
or  steel  bases  are  used  to  support  the  switch  jaws.  Up  to  33,000 
volts,  these  disconnecting  switches  are  made  for  either  front 
connection  or  rear  connection  or  both.  For  higher  voltages  they 
are  invariably  made  for  front  connection  only,  and  in  order  to 
insure  rigidity  and  prevent  oscillations  where  the  blade  becomes 
very  long,  as  for  switches  of  the  higher  voltages,  the  blades  may 
be  of  a  truss  design  (Fig.  372). 


Digitized  by 


Google 


578 


ELECTRICAL  EQUIPMENT 


Disconnecting  switches  are  usually  operated  by  means  of  an 
insulating  rod  or  switch  hook  which  is  made  of  selected  material 
especially  treated  for  the  purpose  and  capable  of  safely  with- 
standing the  operating  voltage.  For  medium  voltages,  holes  are 
provided  in  the  ends  of  the  switch  blades  for  the  insertion  of  the 
hook,  but  for  higher  voltages  where  the  length  of  the  handle  may 


Fig.  370.— 110,000-volt  Outdoor  Transformer  and  Switching  Station. 

be  up  to  15  feet  or  more,  it  becomes  difficult  to  insert  the  hook 
and  this  is  provided  on  the  switch  blade  instead,  as  will  be  noted 
from  the  illustration.  Sometimes  means  are  provided  for  ground- 
ing the  handle  when  in  use. 

When  disconnecting  switches  are  so  mounted  that  the  blade 
forms  the  portion  of  a  loop,  the  switch  may  be  thrown  open  by 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  579 

the  magnetic  repulsion  suddenly  set  up  by  a  large  rush  of  current 
consequent  upon  a  heavy  overload  or  short-circuit.  This  with 
very  few,  if  any  exceptions,  results  in  damage  to  the  switch, 
caused  by  its  opening  under  heavy  load.  To  obviate  such  possible 
results,  disconnecting  switches  should  be  provided  with  safety 
locks  which  hold  the  switch  blade  in  a  closed  position  until  opened 
by  the  operator.    The  catch  is  closed  automatically  when  the 


Fig.  371. — Showing  Method  of  Bringing  Low-tension  Leads  from  Outdoor 
Transformers  to  Building  through  a  Tunnel. 


switch  is  closed,  and  it  may  be  of  a  design  so  as  to  serve  in  addition 
as  a  guide  for  the  blade  in  closing. 

The  ordinary  high-voltage  knife-blade  disconnecting  switch, 
operated  by  a  hook  on  the  end  of  a  long  rod,  necessitates  an  amount 
of  space  of  the  operator  directly  below  the  switch  and  perpendicu- 
lar to  its  base,  depending  both  upon  the  length  of  the  blade  and  of 
the  rod  used  to  open  and  close  it. 

Where  the  space  is  restricted  this  design  may  therefore  not 
be  the  best  suitable  and  a  switch  as  shown  in  Fig.  373  has  been 
developed  for  such  conditions.     It  is  operated  from  directly  below 


Digitized  by 


Google 


580 


ELECTRICAL  EQUIPMENT 


by  a  disconnecting  switch  hook.  There  is  not  needed  the  room 
which  would  otherwiae  have  been  necessary  for  the  operator  to 
use  the  switch  hook  at  the  considerable  an^le  required. 


FiQ.  372.— 110,000-volt  Disconnecting  Switch  with  Safety  Catoh  and  Opening 

Device. 

The  insulators,  insulator  caps,  and  terminals  are  standard. 
The  blade  is  a  copper  rod  with  a  cast  eye  fastened  on  one  end  and 
a  readily  renewable  solid  brass  contact  tip  on  the  other.    The  sta- 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


581 


tionary  contacts  are  the  same  as  those  used  on  H-type  oil  circuit 
breakers. 

When' the  switch  is  opened  a  flange  near  the  tip  of  the  blade 
prevents  the  blade  from  dropping  below  the  upper  part  of  the  lower 
stationary  contact.  A  wide 
flare  on  the  lower  end  of 
the  upper  contact  leads  the 
blade  into  place  when  the 
switch  is  being  closed. 

After  the  blade  is  closed 
a  slight  turn  to  the  right  or 
left  by  the  operating  rod 
locks  the  blade  in  position 
and  prevents  it  from  open- 
ing except  when  desired. 

Sometimes  the  discon- 
necting switches  are  wired 
up  to  indicating  lamps 
mounted  on  the  control 
switchboard.  These  lamps 
are  then  inserted  in  the 
miniature  bus-connections 
and  will  show  to  the  oper- 
ator whether  the  switches 
are  in  the  open  or  closed 
position. 

The  switch  shown  in 
Fig.  374  is  for  use  on 
heavy  outdoor  service.  All 
the  three  poles  are  operated 
simultaneously  by  a  lever 


Fig. 


373. — Special  Disconnecting  Switch 
for  Restricted  Quarters. 


or  handle  which  can  be  located  at  any  height  from  the  ground 
and  locked  in  either  open  or  closed  position.  It  is  of  the  single- 
break  type,  equipped  with  a  horn-type  arc  deflector  on  the  sta- 
tionary contact.  The  shape  and  location  of  the  horn  in  conjunc- 
tion with  the  upward  movement  of  the  switch  blade  definitely 
confines  the  arc  on  rupturing  the  exciting  current  of  a  line  to  the 
horn  and  blade  and  quickly  ruptures  the  arc  without  short-cir- 
cuiting the  Une  or  involving  adjacent  apparatus.  In  operating 
the  disconnecting  switch  the  blades  move  in  a  vertical  plane 


Digitized  by 


Google 


582 


ELECTRICAL  EQUIPMENT 


describing  an  arc  90°  to  go  to  the  full  open  position.  When  the 
switch  opens  an  arc,  the  arc  is  drawn  upward  on  the  arc  deflector 
and  the  end  of  the  switch  blade. 


Fry^7¥ 


L, 


Fig.  374. — 110,000- volt,  Three-pole,  Single-throw,  Disconnecting  switcL 

The  construction  of  the  switch  blade  is  such  that  any  snow 
or  ice  that  has  collected  on  stationary  contact  or  contact  parts  of 
the  switch  are  readily  removed  either  on  opening  or  closing  the 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  583 

switch.    The  operating  mechanism  can  be  thoroughly  grounded 
to  prevent  any  danger  to  the  operator. 

A  suspension-type  switch  for  mounting  directly  in  a  trans- 
mission Une  at  the  point  of  support  of  a  tower  is  shown  in  Fig.  375. 
The  blades  are  suspended  underneath  a  string  of  strain  insulators 


^ 


Fig.  375.— 90,000-volt  Outdoor  Disconnecting  Switch  with  Strain  Insulators. 

and  open  downward.  The  end  of  the  switch  with  its  T-shaped 
casting  is  supported  from  the  suspension  insulators,  and  the  L- 
shaped  casting  on  the  opposite  end  is  connected  directly  to  the 
span  and  is  dependent  on  this  to  support  it  in  an  approximately 
horizontal  position.  The  blade  guide  serves  also  as  a  safety 
catch  to  hold  the  blade  closed. 

Signal  Systems.  In  large  power  stations  it  becomes  essential 
to  provide  some  means  of  communication  between  the  switch- 
board operator  and  the  machine  attendants,  and  different  systems 
of  illuminated  dials,  bells  or  whistles  are  used.  It  is  important 
that  this  apparatus  should  be  located  in  a  position  most  convenient 
to  the  operators,  so  as  to  save  time  and  avoid  possible  errors  at 
critical  moments.  Direct  visual  signals  between  these  persons 
are  practically  impossible,  without  a  moving  or  turning  by  the 
switchboard  operator  from  his  position  before  the  instnunent 
and  control  apparatus.  This  should  not  be  expected  of  him,  as  it 
would  mean  relocating  himself  with  reference  to  the  switchboard 
equipment  for  every  signal  received  or  sent. 

In  stations  of  moderate  size  it  may  be  sufficient  to  install  one 
conmion  large  illuminated  sign  which  is  visible  from  any  place  in 
the  station.     It  contains  the  unit  numbers  and  the  most  important 


Digitized  by 


Google 


5S4 


ELECTRICAL  EQUIPMENT 


signals  such  as  "  start,"  "  stop,"  "  stand-by,"  etc.,  and  is  con- 
controlled  from  the  switchboard,  a  whistle  being  used  for  calling 
the  operator's  attention  to  the  signals.  Sometimes  provision 
is  also  made  for  answering  or  returning  the  signals  to  the  switch- 
board. 

Possibly  the  most  satisfactory  and  most  generally  used  signal 
system  is  the  individual  push-button  equipment,  shown  in  Fig. 
376.  It  consists  of  an  individual  stand  for  each  machine  unit  with 
the  signals  mounted  thereon,  as  shown.    Similar  signal  equip- 


FiG.  376. — Individual  Push-button  Signal  Equipment  with  Stand  for  One 

Machine. 

ments  are  also  provided  on  the  respective  machine  panels  on  the 
benchboard,  the  two  corresponding  equipments  being  connected 
together  electrically.  The  signals  consist  of  colored  glass  win- 
dows with  white  letters  illuminated  by  small  lamps  behind.  Oppo- 
site each  signal  is  a  three-way  push-button  switch,  and  a  gong  is 
installed  near  each  machine  and  also  at  the  switchboard.  Pushing 
a  button,  for  example,  at  the  switchboard  rings  the  gong  at  the 
machine  to  which  the  signal  is  sent  simultaneously  illuminating 
the  particular  signal  which  was  sent  at  both  places.    The  gong 


Digitized  by 


Google 


SWITCHING  EQUIPMENT 


585 


keeps  on  ringing  and  the  signal  remains  illuminated  until  the 
machine  operator  acknowledges  the  signal  by  pressing  the  cor- 
responding button  on  his  equipment.  The  connection  diagram 
for  a  small  equipment  of  this  type  is  shown  in  Fig.  377. 

It  is,  of  course,  not  necessary  to  install  the  signals  near  the 
machines  on  pedestals.  They  are  often  located  on  the  nearby 
wall  where  they  can  easily  be  seen,  and  occasionally  various 
colored  lamps  are  installed  at  the  side  of  the  respective  signals  so 
that  they  can  be  read  more  quickly  and  distinctly  from  a  distance. 
One  company,  for  example,  uses  a  blue  light  beside  the  "  stand- 


L-.- 


kl  kl  1^1 


R«layaad 


•  '       *  *  '    aa}cl|Gong> — *  *  '         i 


ki    ki    kl: 


<^ 


Swil 


ja 


itab«« 

mpaj 


MAchlne  Stand 


Fig.  377. — Connection  Diagram  of  Two  Signal  Equipments  with  Three  Signals 

by  "  signal,  a  red  for  the  "  fast,"  a  green  for  the  "  slow  "  and 
white  for  all  the  others. 

What  the  signals  should  read  depends,  of  course,  to  some 
extent  on  the  local  operating  conditions.  The  following  are, 
however,  very  conmion:  "Stand-by,"  "start,"  "fast,"  "slow," 
"  stop,"  and  "  O.K."  These  are  used  in  the  power-house  of  the 
Pennsylvania  Water  and  Power  Company,  their  meaning  being  as 
follows: 

"  Standi  ":  Stand  near  governor  and  await  fmiiher  orders. 
Correct  any  apparent   governor  trouble.    Trouble  impending. 


Digitized  by 


Google 


586  ELECTRICAL  EQUIPMENT 

The  "  Stand-by  "  signal  is  to  be  used  during  the  cutting  out  of 
units,  tests,  Ughtning  storm,  or  other  expected  troubles. 

''  Start ":  Start  unit  at  once  on  hand  control. 

*'  Start  Fast ":  (Combination  signal).  Start  unit  as  quickly 
as  possible. 

<<  Faa  ":  If  unit  is  not  on  the  bus,  increase  speed.  If  unit  is 
on  the  bus,  increase  gate  opening  gradually.  If  the  signal  is 
flickered,  increase  rapidly. 

"  Stop  ":  Shut  down  unit  at  once. 

"  O.K.  ":  Unit  on  bus.  Engage  governor-control  motor  gear. 
Conditions  normal.  Further  attention  not  needed.  Cancels 
"  Start "  or  "  Fast "  signal.  The  "  O.K."  signal  is  also  used  when 
unit  has  come  to  rest  and  field  has  been  taken  off. 

The  whistle  used  in  this  power  station  is  electrically  controlled 
from  the  switchboard  and  is  operated  by  compressed  air  at  300 
pounds  pressure.  It  is  located  at  one  end  of  the  power-house  and 
is  loud  enough  to  be  heard  over  the  noise  of  the  machinery  in  all 
parts  of  the  building,  and  can  be  heard  outside  the  building  for 
quite  a  distance.  It  is  used  principally  for  calling  persons  con- 
nected with  the  operation,  the  code  being  as  follows: 

Attention  to  signals  — 

Assistant  operator 

Machine  man  —    — 

lightning  storm  on 


"  On  hearing  this  signal  a  special  arc  extinguisher  observer  will 
report  to  operator." 

Hydraulic  floorman 

Hold  frequency 

This  is  an  emergency  signal  to  be  used  in  case  the  station  is 
swamped  or  running  away.  "  If  the  station  is  swamped,  force  all 
machines  to  fuU  gate  opening;  if  running  away,  close  all  hand- 
control  machines  until  frequency  returns  to  n(»inal.  If  governor 
system  has  failed,  governor  machines  must  be  changed  over  to 
hand-control  and  regulated  imtil  frequency  returns  to  normal. 
Pumpman  must  make  every  effort  to  hold  pressure  on  governor 
and  handrcontrol  systems,  starting  pumps  and  taking  any  other 
necessary  steps.  Extra  men,  unless  otherwise  detailed,  to  report 
to  floorman  on  governor  floor." 

Emergency  stand  by 

*'  Serious    general    emergency    existing    or    impending.    All 

Digitized  by  VjOOQ IC 


SWITCHING  EQUIPMENT 


587 


attendants  stand  by.  Extra  men  report  to  floorman  or  operator, 
unless  otherwise  detailed.  Chief  and  assistant  chief  operators 
proceed  to  benchboard,  maintenance  men  report  to  chief  operator." 

There  is  another  emergency  whistle  located  on  the  roof  of  the 
building,  for  the  purpose  of  calling  assistance  during  operating 
emergencies  and  for  calling  the 
operating  heads  and  company 
physician  in  case  they  could  not 
be  located  by  telephone.  This 
whistle  can  be  heard  a  distance 
of  five  or  six  miles. 

A  novel  signal  system  is  used 
by  the  Mississippi  River 'Power 
Company  in  its  station  at  Keo- 
kuk. In  general  it  consists  of 
transmitting  and  receiving  dials 
with  the  signal  words  plainly 
marked  thereon.  A  pointer  on 
the  receiving  dial  is  electrically 
connected  to  follow  the  position 
of  a  handle  on  the  transmitting 
dial.  Fig.  378  illustrates  a  ped- 
estal containing  a  transmitter 
(lower  dial)  and  a  receiver 
(upper  dial).  One  pedestal  is 
located  in  front  of  each  genera- 
tor in  the  generator  room  (Fig. 
3),  and  a  similar  equipment, 
although  without  the  pedestal, 
is  located  on  each  generator 
panel  in  the  control  room. 

A  diagram  of  connection  of 
the  apparatus,  which  is  known 
as  position  indicators,  is  given  in 
Fig.  379.     Each  complete  equip* 

ment  consists,  as  said,  of  two  machines,  a  transmitter  and  a  re- 
ceiver, connected  as  shown  and  resembUng  in  design  small  induc- 
tion motors.  The  stators  are  provided  with  an  ordinary  closed 
winding,  three  equidistant  points  being  permanently  connected 
t<^ether.    The  rotors  are  bipolar,  connected  in  multiple  and  ener- 


Fia.  378. — Signal  Equiptnent  at 
Mississippi  River  Power  Com- 
pany. Generator  Room  Ped- 
estal. 


Digitized  by 


Google 


588 


ELECTRICAL  EQUIPMENT 


Fig.  379. — Diagram  of  Connections  of 
Position  Indicator. 


gized  from  a  25-cycle,  125  volt,  sincle-phase  source;  the  stator 
being  energized  by  inductions  from  the  rotors. 

The  movement  of  the  transmitter  rotor,  which  is  mechanically 
operated  by  a  handle,  induces  voltage  in  the  stator  winding. 
This  voltage  is  transmitted  by  the  three-phase  tie  to  the  stator 

^  of  the  receiver  and  dupU- 

^v^^cs^^^  Z  cj^tes  in  it  the  same  polar- 
ity  and  voltage  conditions 
developed  in  the  transmit- 
ter stator,  but  in  the  reverse 
direction.  The  rotor  of  the 
receiver  is  energized  in  the 
ekme  direction  as  that  of 
the  transmitter,  and  conse- 
quently reacted  upon  by 
tiie  polarized  stator  until 
their  magnetic  axes  coincide 
and  the  rotors  of  the  transmitter  and  receiver  are  in  the  same 
relative  position.  With  the  rotors  thus,  no  current  flows  between 
the  stators.  Any  difference  in  the  position  of  the  transmitter 
and  receiver  rotors  causes  a  flow  of  current  and  resultant  torque 
which  moves  the  receiver  rotor  and  dial  pointer  to  the  same  rela- 
tive position  as  that  of  the  transmitter.  On  both  the  pedestals 
and  benchboard,  at  each  side  of  the  transmitter  handle,  are  located 
double  push-button  switches  which  are  employed  for  operating 
signal  lamps,  whistles  and  bells. 

The  method  of  signaling  is  as  follows:  When  the  switchboard 
operator  desires  to  send  a  signal  he  turns  the  handle  of  the  trans- 
mitter until  its  dial  indicates  the  signal  he  wishes  to  send.  This 
signal  will  be  indicated  on  the  dial  of  the  receiver  in  the  generator 
room.  He  then  pushes  the  button  on  the  right  of  the  handle. 
This  Ughts  a  lamp  on  the  generator  (Fig.  3)  and  blovjs  a  whistle 
in  the  generator  room  to  attract  the  attention  of  the  man  in  chai^ge 
of  the  particular  machine.  As  soon  as  the  attendant  has  read 
the  signal  on  his  receiver,  he  will  turn  the  handle  of  the  transmitter 
on  the  pedestal  to  the  same  signal.  He  will  then  push  the  button 
at  the  right  of  the  handle,  which  will  extinguish  the  lamp  and 
cut  out  the  whistle.  Next  he  will  push  the  button  at  the  left  of 
the  handle,  which  operation  will  light  a  lamp  in  the  switchboard 
room  and  also  ring  a  signal  bell  indicating  to  the  switchboard 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  589 

man  that  the  generator  attendant  has  received  the  signal  and  also 
just  what  signal  he  received.  The  switchboard  operator,  after 
having  seen  this  returned  signal,  will  push  the  button  at  the  left 
of  the  transmitter  handle,  which  will  extinguish  the  lamp  and  cut 
out  the  signal  bell.  This  completes  the  cycle  of  sending  and 
receiving  a  signal. 

The  system  is  identical  to  that  used  on  the  Panama  Canal  to 
indicate  the  position  of  the  lock  machinery. 

The  signal  i^tem  in  any  important  station  is  alwa3rs  sup» 
plemented  by  a  multiplenstation  interconmiunicating  telephone 
system.  This  is  used  when  special  orders  or  instructions  are  to  be 
given. 

Multi-recorder.  The  multi-recorder  is  a  device  for  recording 
on  a  strip  of  paper  the  exact  time  of  the  occurrence  of  any  elec- 
trical phenomena  and  is  appUcable  in  central  stations  for  record- 
ing switching  operations,  line  surges  and  other  disturbances 
beyond  the  control  of  the  operator.  In  case  of  accidents  such  a 
record  is  of  particular  value  because  it  enables  the  engineer  to  know 
where  and  when  the  trouble  started  and  how  the  switching  was 
done. 

The  recorder  consists  essentially  of  a  number  of  stamps  operated 
by  a  clockwork  and  printing  the  time,  within  fraction  of  seconds, 
of  the  event  to  which  they  are  relayed.  A  description  of  this  device 
is  given  by  Prof.  E.  E.  F.  Creighton  in  the  A.I.E.E  Transactions, 
1912,  page  825. 

Oil  Circuit  Breaker  Batteries.  The  operation  of  remote-con- 
trol oil  circuit  breakers,  field  switches,  field  rheostats,  signal  lights, 
etc.,  necessitate  an  absolutely  reliable  source  of  energy  which 
should  be  entirely  independent  of  the  regular  distribution  circuits 
and  held  in  reserve  exclusively  for  this  purpose. 

It  is  therefore  usual  to  ftistall  a  motor-generator  set  consisting 
of  an  induction  motor  driven  by  power  from  the  A.C.  circuit, 
direct  connected  to  a  direct-cxurent  generator.  In  order,  however, 
to  insure  continuity  of  service  in  case  of  an  interruption  in  the 
supply  of  current  from  this  machine,  whether  due  to  failure  of 
the  power  supply  on  the  A.C.  circuit  or  to  some  derangement 
in  the  machine  itself,  it  is  standard  practice  to  install  a  storage 
battery,  which  is  normally  kept  floating  across  the  terminals  of 
the  direct-ciurent  machine.  This  motor  generator  is  kept  run- 
ning continuously  except  for  such  brief  periods  of  time  when  it  may 


Digitized  by 


Google 


590  ELECTRICAL  EQUIPMENT 

be  necessary  to  shut  it  down  for  inspection  or  repairs,  and  under 
normal  conditions  it  carries  the  steady  load  due  to  the  signal  lamps, 
and  supplies  a  small  amount  of  charging  current  to  the  batteiy 
in  order  to  keep  it  fully  charged  at  all  times  and  ready  for  ser\'ice. 
This  direct-current  machine  is  of  the  shunt-woimd  type  having 
a  decidedly  drooping  characteristic,  so  that  when  a  heavy  demand 
occurs,  due  to  the  opening  or  closing  of  oil  switches,  ^tc,  the  load 
is  divided  between  the  machine  and  the  battery,  and  the  machine 
itself  is  thus  protected  against  excessive  momentary  overload. 

The  normal  voltage  of  the  control  circuit  is  approximately 
125  volts,  but  the  D.C.  generator  is  designed  for  the  maximum 
charging  voltage  of  the  battery,  which  may  rise  to  about  2.80 
volts  per  cell.  The  ampere  capacity  of  the  generator  should  be 
equal  to  the  normal  charging  rate  of  the  battery  plus  the  current 
required  for  the  signal  lamps. 

It  will  be  noted  from  the  above  that  imder  ordinary  conditions 
of  operation  the  battery  does  very  Uttle  work,  and  the  maximum 
demand  upon  it  occurs  only  when  it  is  necessary  to  open  or  dose 
a  number  of  switches  simultaneously  at  a  time  when  the  motor 
generator  set  is  inoperative. 

The  ampere  capacity  of  the  battery  is  determined  by  ascer- 
taining the  maximum  possible  demand  due  to  the  simultaneous 
operation  of  as  many  of  the  remote-control  devices  as  are  Uable 
to  be  operated  at  once,  and,  selecting  a  battery  of  sufficient  size 
to  supply  this  current  for  the  period  of  time  necessary  without 
dropping  in  voltage  below  a  certain  permissible  minimum.  The 
number  of  cells  is  usually  fixed  at  60,  and  for  this  number  a 
floating  voltage  of  about  127  volts  is  suitable. 

Standard  remote-control  apparatus  is  usually  designed  to 
operate  over  a  comparatively  wide  range  of  voltage  variation, 
owing  to  the  fact  that  such  apparatus*  is  in  some  cases  operated 
from  an  exciter  circuit  whose  voltage  is  varied  by  automatic 
regulators.  In  order  to  provide  ample  margin  of  safety,  a  mini- 
miun  voltage  of  90  is  usually  fixed  for  the  battery  when  carrying 
its  maximum  load.  This  is  equivalent  to  1.5  volts  per  cell.  A 
properly  designed  storage  battery  equipped  with  low-resistance 
intercell  connections  and  provided  with  conductors  of  ample 
capacity  for  connecting  to  the  switchboard  may  be  discharged  at 
five  times  the  one-hour  rate  (twenty  times  the  eight-hour  rate) 
for  a  period  of  one  minute  without  dropping  below  the  limiting 


Digitized  by 


Google 


SWITCHING  EQUIPMENT  591 

voltage  of  1.5  per  cell  above  mentioned.  Oil  switch  batteries  are 
frequently,  therefore,  designed  to  work  at  five  times  the  one-hour 
rate  when  the  maximum  possible  load  is  to  be  carried  with  the 
motor  generator  set  shut  down.  In  order  to  determine  the 
maximum  possible  load,  it  is  usual  to  figure  that  not  more  than 
two  remote-control  switches  will  be  closed  at  one  time,  and  not 
more  than  one-half  of  the  total  number  of  automatic  switches 
will  be  tripped  simultaneously.  When  more  than  twenty  oil 
switches  are  installed,  it  is  considered  safe  to  figure  on  not  more 
than  one-third  of  the  total  number  of  automatic  switches  being 
tripped  at  the  same  time.  The  duration  of  any  single  switching 
operation  is  but  a  fraction  of  a  minute,  and  a  battery  subjected 
to  intermittent  discharges  at  high  rates  recuperates  rapidly  during 
the  intervals  of  rest,  so  that  a  battery  figured,  as  above,  will  easily 
handle  as  many  successive  operations  as  are  Uable  to  be  required. 
The  current  required  for  the  operation  of  oil  circuit  breakers,  etc., 
varies  with  the  size  and  make,  and  should  be  obtained  from  the 
respective  manufacturers. 

In  some  cases  an  emergency  station  lighting  circuit  may  be 
arranged  for  connection  to  the  oil  switch  battery  in  case  of  com- 
plete interruption  of  other  sources  of  Ught.  To  provide  for  this, 
a  battery  of  greater  ampere-hour  capacity  may  be  required  than 
that  determined  by  the  oil  switch  service  alone. 

In  order  to  permit  giving  the  battery  a  charge  to  maximum 
voltage  by  raising  the  voltage  of  the  generator  without  subjecting 
the  signal  lamps  and  remote-control  apparatus  to  this  high  voltage 
a  tap  is  taken  from  the  battery  to  the  switchboard  by  means  of 
which  a  group  of  10  cells  may  be  cut  out.  At  the  beginning  of 
charge  the  entire  60  cells  are  connected  to  the  dynamo,  whose 
voltage  is  raised  suflSciently  to  dehver  the  charging  current, 
while  50  cells  are  connected  across  the  control  circuit.  The  cur- 
rent required  for  the  signal  lamps  under  these  conditions  passes 
through  the  end  cell  group  in  addition  to  the  charging  current  of 
the  main  battery,  and  the  charging  of  the  end  cells  is,  therefore, 
completed  before  that  of  the  main  battery.  The  end  cell  group 
is  then  cut  out  and  the  charging  of  the  remaining  50  cells  is  com- 
pleted. The  maximum  voltage  of  these  50  cells  at  the  end  of 
charge  will  be  nearly  140  volts.  The  signal  lamps  are  designed 
to  stand  this  voltage  for  a  short  time,  and  the  standard  remote 
control  apparatus  will  operate  satisfactorily  at  this  voltage. 


Digitized  by 


Google 


592 


ELECTRICAL  EQUIPMENT 


In  Fig.  380  is  shown  the  diagram  of  connections  for  this  scheme. 
The  negative  bus  is  divided  into  two  sections  and  two  single-pole, 
double-thi-ow  knife  switches,  A  and  B,  are  provided,  one  connected 
to  each  section  of  the  negative  bus.  When  A  is  thrown  down  the 
60  cells  of  battery  are  connected  across  the  generator  terminals, 


12B  Volt  Opsnttiac  Bu>^ 


Motor-Generfttor  Bet 
Nonnal  (Floatioff) : 
CharsrcfiOGelb: 
CharveM  Cells; 
Motoc  Q«D«nitor  Set  a 

Fig.  380. — Diagram  of  Connection  for  Floating  Oil  Circuit-breaker  Battery. 

and  when  B  is  thrown  down  the  two  sections  of  bus  are  connected 
together  and  current  is  furnished  to  the  control  circuit  by  the 
dynamo  with  the  battery  floating  in  parallel.  This  is  the  normal 
position  of  these  switches. 

At  the  beginning  of  the  charge,  switch  B  is  thrown  up  con- 
necting the  control  circuit  across  50  cells,  and  the  voltage  of  the 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  593 

dynamo  which  is  still  connected  across  60  cells,  is  raised  until  the 
desired  charging  current  is  obtained.  When  the 'end  cell  group  is 
fully  charged,  as  indicated  by  free  gassing  and  maximum  specific 
gravity,  switch  A  is  thrown  up,  cutting  out  the  end  cell  group 
and  the  charging  of  the  main  battery  is  completed. 

An  overload  circuit  breaker  is  provided  in  the  positive  lead 
from  the  generator.  In  some  cases  a  reverse-current  trip  has  been 
provided  for  this  circuit  breaker;  but  this  is  usually  omitted, 
owing  to  the  fact  that  a  momentary  variation  of  frequency  on  the 
83rstem  might  lower  the  speed  of  the  motor  generator  set  and  reverse 
the  ciurrent,  thus  tripping  the  circuit  breaker  unnecessarily.  A 
momentary  reversal  of  current  through  the  generator  would 
usually  be  quite  harmless. 

In  the  battery  leads. fuses  are  inserted  rather  than  circuit 
breakers,  as  it  is  not  desired  to  have  the  battery  circuit  open 
except  under  extreme  conditions,  such  as  short-circuit  in  the  con- 
trol S3rstem. 

When  the  battery  is  kept  continually  floating  at  practically 
constant  voltage  across  the  D.C.  operating  bus,  and  another 
source  of  current,  such  as  a  motor  generator  set,  is  provided  to 
supply  the  steady  load  of  signal  lamps,  etc.,  so  that  the  battery 
work  is  limited  to  occasional  momentary  discharges  when  the  oil 
switches  are  operated  or  to  such  sustained  discharges  as  may  be 
called  for  in  case  the  normal  source  of  current  should  fail — in  other 
words,  where  the  conditions  call  for  strictly  emergency  stand-by 
service  from  the  battery — the  Exide  or  similar  type  of  battery  in 
glass  jars  is  recommended,  this  being  the*  same  type  that  is  now 
generally  used  for  stand-by  service  in  the  large  central  station 
lighting  systems.  Where  a  method  of  operation  is  adopted  in 
which  the  battery  is  discharged  continuously  on  the  bus  until 
nearly  exhausted  and  then  recharged,  thus  involving  repeated 
cycles  of  charge  and  discharge,  the  Manchester  type  of  plate  or 
similar  is  recommended,  the  Exide  plate  being  only  reconmiended 
for  use  on  floating  batteries  at  approximately  constant  voltage 
and  discharging  only  under  temporary  emergency  conditions. 

9.   OVER-VOLTAGE  PROTECTION 

Classification  of  Over-voltages.  High-voltage  disturbances 
may  be  divided  into  two  broad  classes.  First,  that  covering 
actual  high  voltages  in  which  the  excess  voltage  exists  between  the 


Digitized  by  VjOOQ IC 


594  ELECTRICAL  EQUIPMENT 

phase  conductors  or  between  the  phase  conductors  and  ground. 
Second,  that  covering  locaUzed  high  voltages  in  which  the  excessive 
potential  difference  exists  between  two  points  along  the  same  con- 
ductor. In  these  cases  the  "  conductor  "  is  supposed  to  include 
the  line  wires  as  well  as  the  generator  and  transformer  windings. 

To  the  first  class  belong  those  distiu*bances  which  are  caused 
by  overspeeds,  poor  regulation  and  resonance,  while  the  nature 
of  disturbances  caused  by  switching,  arcing  -grounds,  and  Ught- 
ning  may  be  such  that  they  may  belong  to  either  class.  Where 
the  impulses  or  traveling  waves  set  up  are  of  comparatively  low 
frequency  and  consequently  of  sloping  wave  front,  the  disturbance 
can,  however,  genertdly  be  classed  with  the  former,  and  when  of 
high  frequency  and  steep  wave  front  with  the  latter. 

Excessive  over-voltages  are  very  apt  to  occur  when  water- 
wheel-driven  generators  run  away,  especially  if  they  are  provided 
with  direct-connected  exciters.  Actual  experience  has  thus 
demonstrated  that  under  such  conditions  the  generator  and  trans- 
mission voltages  may  reach  three  times  their  normal  value,  which 
of  course  subjects  the  apparatus  to  unreasonable  strains.  To 
provide  against  this,  automatic  brake  equipments  are  provided 
or  else  high  voltage  cut-out  relays  which  automatically  insert 
resistances  in  the  exciter  fields  if  the  voltage  exceeds  a  certain  pre- 
determined value. 

In  the  design  of  modern  long-distance  transmission  lines  it 
is  generally  the  regulation,  or  the  variation  in  voltage  which  occurs 
when  the  load  is  thrown  on  or  off,  that  is  the  governing  factor 
rather  than  the  energy  loss.  Not  only  may  the  voltage  drop 
under  load  be  quite  large,  especially  when  the  load  has  a  low  power- 
factor,  but  with  the  high-transmission  voltages  now  in  use  the 
capacity  effect  of  the  lines  becomes  very  high,  which  in  turn  ma^^ 
result  in  a  considerable  voltage  rise  at  the  substation  at  light 
loads.  This  is  now  one  of  the  chief  arguments  against  isolated 
delta  connection  for  long-distance  high-tension  lines.  It  was 
formerly  claimed  that  such  a  system  could  be  temporarily 
operated  with  one  line  grounded.  Recent  experiences  on  laiige 
systems,  however,  indicate  that  this  is  not  feasible,  as  in  the 
event  of  a  ground  the  charging  cm-rent,  which  is  a  function  of 
the  voltage  from  wire  to  neutral,  will  be  increased  because  the 
natural  is  shifted  from  the  center  of  the  delta  to  one  comer. 
This  increase  will  be  about  73  per  cent  and  will  of  course  in 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  695 

turn  cause  an  additional  voltage  rise  at  no  load,  which  is  not 
pennissible. 

The  voltage  rise  caused  by  the  charging  current  in  a  long  line 
may  cause  a  breakdown  of  the  air  nearest  the  line  conductor  and 
cause  corona  which  may  seriously  increase  the  transmission  losses. 
They  may  also  unduly  strain  other  insulations  on  the  system  and 
afifect  the  operation  of  the  Ughtning  arresters,  the  normal  voltage 
range  of  which  should  be  kept  within  reasonable  limits  for  satis- 
factory operation.  On  the  other  hand  it  is  well  known  how  the 
operation  of  motors  is  affected  by  voltage  variations  and  that 
the  life  of  lamps  is  seriously  reduced  if  the  voltage  is  too  high, 
not  to  speak  of  the  unpleasantness  of  a  variation  in  the  intensity 
of  the  illumination,  which  of  course  accompanies  a  fluctuation  in 
the  voltage. 

From  the  above  it  is  imperative  that  the  regulation  of  a  modem 
system  be  kept  within  certain  permissible  limits,  and  with  high- 
voltage  systems  this  is  most  readily  accomplished  by  installing 
synchronous  condensers  with  automatic  voltage  regulators  in  the 
substation.  As  previously  stated,  the  large-capacity  currents 
of  long-distance  Unes  cause  a  rise  of  voltage  from  the  generator  to 
receiver  at  Ught  load,  while  at  full  load  the  lagging  current  taken 
by  the  load  will  cause  a  drop  of  voltage  from  generator  to  receiver. 
It  is,  therefore,  evident  that  the  voltage  may  be  kept  constant  or 
within  certain  limits,  at  the  receiving  end,  if  a  synchronous  con- 
denser is  installed  there,  and  its  field  adjusted  so  as  to  make  it 
take  a  lagging  current  at  no  load  and  a  leading  current  at  full  load; 
in  the  first  case  to  oflfset  the  effect  of  the  line  capacity  and  in  the 
second  to  offset  the  surplus  lagging  load  current. 

Resonance  must  also  be  guarded  against,  as  it  can  give  rise 
to  large  currents  which  may  open  the  circuit  protecting  devices 
and  interrupt  the  service,  or  the  potential  may  be  raised  to  a  value 
at  which  the  installation  of  the  system  is  broken  down.  In  an 
electric  circuit  the  inductive  reactance  and  the  capacity  reactance 
oppose  each  other.  If  of  equal  value  they  neutraUze  each  other, 
in  which  case  the  resistance  of  the  circuit  limits  the  value  of  the 
current.  This  may,  therefore,  reach  very  high  values  and  when 
passing  through  the  inductance  and  capacity  the  voltage  at  these 
would  in  turn  be  very  high. 

To  illustrate  this  further;  assume  a  circuit  having  a  resistance 
of  say  60  ohms  and  a  capacity  reactance  of  1000  ohms,  then  thQ 


Digitized  by 


Google 


596  ELECTRICAL  EQUIPMENT 

total  impedance  would  be  equal  to  \/5^+ToO(^  =  1000  ohms 

approximately.    With   100,000  volts  impressed  on  this   circuit 

100  000 
the  current  flow  would  be        '      =100.    If  now  in  addition 

lUUU 

the  circuit  contains  an  inductive  reactance  of  1000  ohms,  it  is 
evident  that  this  entirely  neutralizes  the  capacity  reactance  and 
that  the  current  is  only  limited  by  the  50-ohm  resistance,  thus 

in  this  case  equal  to  — ^ — =2000  amperes.    With  this  current 
oU 

flowing  the  voltage  across  either  the  inductance  or  capacity  be- 
comes equal  to  2000X1000=2,000,000  volts,  which  of  course 
would  be  far  beyond  destruction.  Of  course,  this  extreme  con- 
dition does  not  apply  to  an  ordinary  transmission  line  where  the  re- 
sistance, inductance  and  capacitance  is  distributed,  but  destruc- 
tive voltages  may  be  set  up  where  inductance  and  capacitance 
is  concentrated. 

Fortimately,  the  characteristics  of  transmission  systems  are 
such  that  their  inductive  reactance  is  not  large  enough  to  neu- 
tralize the  capacity  reactance  at  the  fundamental  generator 
frequency.  Since,  however,  the  inductive  reactance  increases 
and  the  capacity  reactance  decreases  proportionally  to  frequency, 
the  two  reactances  come  nearer  together  for  high  frequencies, 
such  as  for  the  high  harmonics  of  the  generator  wave.  These 
may,  therefore,  be  the  cause  of  resonance  rise  of  voltage  between 
the  line  capacity  and  circuit  inductance.  With  modem  alterna- 
tors, however,  the  higher  harmonics  are  generally  so  small  that 
there  is  not  much  danger  from  resonance. 

Abnormal  voltages  can  also  be  caused  by  traveling  waves 
which  are  set  up  when  the  equiUbrium  of  an  electric  circuit  is 
disturbed.  Such  disturbances  may  originate  in  the  circuit  itself 
as  by  switching  or  they  may  be  due  to  external  causes,  such  as 
atmospheric  lightning  phenomena. 

When  an  electric  circuit  is  connected  to  a  generator  or  other 
source  of  energy,  a  wave  of  voltage  and  current  shoots  out  along 
the  line  with  a  very  high  velocity  and  charges  the  same.  If  the 
maximum  value  of  the  voltage  is  e  and  the  maximum  value  of  Uie 
current  i,  the  wave  possesses  per  unit  length  an  electrostatic 

energy  of  -^  watt  seconds  and  an  electro-magnetic  energy  of  — 

watt  seconds,  C  being  the  capacity  in  farads  and  L  the  inductance 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION 


697 


in  henrys  per  unit  length  (cm.),  of  the  circuit.  These  two  quan- 
tities are  equal  or  -—  =  — -  and  the  relation  between  the  voltage 
and  ciurent  at  a  certain  point  of  the  traveling  wave  is,  therefore. 


'=\&' 


\§^ 


is  termed  the  ''  natural  impedance  "  of  the  circuit,  and  is  of 

great  value  in  the  study  of  transient  phenomena. 

If  the  line  is  open-circuited  at  the  farthey  end,  it  is  obvious 
that  when  the  wave  reaches  this  point  it  cannot  flow  any  fmiiher, 


-M- 


■^« 


Fio.  381. — Reflection  of  a  Travel- 
ing Wave  at  the  Open-circuited 
End  of  a  Line. 


Fig.  382.— Reflection  of  a  Travel- 
ing Wave  at  the  Short-circuited 
End  of  a  Line. 


but  is  reflected,  the  voltage  and  current  of  the  reflected  wave 
being  of  the  same  values  as  in  the  original  waves  because  the 
energy  remains  constant.  The  total  current  of  the  incoming  and 
reflected  wave  must,  however,  be  zero  on  account  of  the  open- 
circuited  line,  and  the  whole  energy  is,  therefore,  stored  at  this 
point  in  the  electrostatic  field.  The  reflected  current  wave  must 
therefore  be  reversed  and  its  value  equal  —  t,  while  the  value  of  the 
voltage  wave  at  the  end  of  the  Une  where  the  original  and  reflected 
waves  overlap  is,  therefore,  equal  to  26,  as  shown  in  Fig.  381. 

When  the  end  of  the  line. is  short-circuited,  however,  the  con- 
ditions are  entirely  reversed.  In  that  case  the  voltage  at  this 
point  must  be  zero,  and  all  the  energy  is  stored  in  the  electro- 


Digitized  by 


Google 


ELECTRICAL  EQUIPMENT 


magnetic  field,  the  value  of  the  total  current  at  the  end  of  the  line 

being  equal  to  2i,  Fig.  382. 

The  wave  travels  twice  forth  and  back  over  the  entire  length 
vyz^/VyvyyvwM g:-  of  the  line,  after  which  the 


'&^M 


mm^mm;^mm^Mmm 


t-O 


conditions  return  to  the 
same  state  as  at  the  begin- 
ning, Fig.  383.  It  will,  how- 
ever, continue  to  oscillate 
forth  and  back  until 
damped  out  by  the  resist- 
ance and  leakage  of  the 
line,  after  which  it  assumes 
a  stationary  condition  with 
a  charge  corresponding  to 
the  voltage  of  the  generator. 
The  wave  length,  or 
rather  the  distance  which 
the  wave  front  travels  in 
completing  the  above  cycle, 
is  obviously  equal  to  four 
times  the  length  of  the  line, 
and  the  frequency  of  the 
oscillation  is 

V  _       1 

where  I  is  the  length  of  the 
hne,    and    v    or  — ==:  the 

Vie 

velocity  at  which  electric 
energy  travels  through  a 
circuit    whose    inductance 

Fia.  383.-One  Complete  Oscillation  of  a  f^^^      capacity     p^     unit 
Traveling  Wave  Set  Up  when  Switching  length  are  L  and  C.    This 

velocity  for  overhead  hnes 
is  equal  to  the  velocity  of 
light,  or  188,000  miles  per  second.  The  waves  in  the  above  illus- 
trations are  shown  of  a  rectangular  form  which  could  only  be 
the  case  if  the  generators  had  no  resistance  or  inductance.  Ordi- 
narily, however,  they  are  of  a  more  or  less  sloping  character. 


mmmmmm^- 


±^ 


\^^^^^:^^^^^::mm^^^^-^^ 


^^>S^^^:>^:>:<Sc^^^^ 


in  an  Open-circuited  Line. 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  599 

In  the  above  it  was  assumed  that  the  end  of  the  line  was  either 
open-  or  short-circuited.  If  a  non-inductive  resistance,  R,  is 
connected  across  the  end  of  a  line,  the  voltage  of  the  reflected 
wave,  and  thus  the  total  voltage  at  this  point,  necessarily  depends 
on  the  value  of  this  resistance.  When  fi=  oo  it  naturdly  resem- 
bles an  open-circuit  in  which  case  the  maximum  voltage  is  equal 
to  double  the  normal  value,  while  if  R—0,  or  negligible,  thus 

resembling  a  short-circuit,  the  voltage  is  zero.     With  R=^-^ 


4 


there  is  no  reflected  wave  at  all.    If  R  >y]p  there  is  a  partial 

reflection  with  reversal  of  current,  while,  if  R<-v/pj  there  is  a 

partial  reflection  with  reversal  of  voltage.  With  an  inductive 
receiving  circuit,  this  acts  in  the  first  instant  as  a  resistance  of 
infinite  value,  and  voltage  reaches  double  value,  while  a  con- 
denser under  similar  conditions  would  act  as  a  short-circuit,  and 
the  voltage  would  be  zero. 

From  the  preceding  it  follows  that  when  a  dead  high-tension 
transmission  line  is  to  be  energized  the  best  practice  to  follow 
would  be  to  switch  the  Une  onto  the  dead  transformers  first  by 
means  of  the  high-tension  switch  and  then  energize  the  com- 
bination of  line  and  transformers  by  closing  the  low-tension 
switch  to  the  generating  source,  this  sequence  of  closing  the 
switches  will  obviate  the  high-tension  surges  and,  consequently, 
minimize  the  danger  of  insulation  breakdown. 

It  is  also  of  greatest  importance  to  consider  the  changes  which 
take  place  at  a  transition  point  between  two  circuits  of  different 
characteristics,  when  a  traveling  wave  passes  from  one  to  the 
other,  such  as,  for  example,  where  an  undergroimd  circuit  joins  an 
overhead,  or  where  a  transmission  line  is  connected  to  a  trans- 
former. 

Assume  that  a  traveUng  wave  with  the  voltage  e  and  the  cur- 
rent i  approaches  from  a  circuit  having  a  natural  impedance 


^"4. 

"S.- 


and  enters  a  second  circuit  with  a  natural  impedance  of 

j^  .    Part  of  the  wave  will  then  be  reflected  and  part 
transmitted.    It  is  also  evident  that  at  the  transition  point  the 


Digitized  by 


Google 


600  ELECTRICAL  EQUIPMENT 

potential  will  be  the  sum  of  the  incoming  and  reflected  waves, 
while  the  current  will  be  represented  by  the  difference  of  the  two 
waves  since  they  travel  in  opposite  direction.  If  we  thus  denote 
the  voltage  and  current  of  the  reflected  wave  by  62  and  12  and  of 
the  transmitted  wave  by  ei  and  ti,  we  get  the  following  relation 
at  the  transition  point. 

€+62  =  €1; 

but 

€ 

e2 
The  amplitude  of  the  transmitted  voltage  wave  is,  therefore, 

and  of  the  reflected  voltage  wave 
Similarly  we  get  for  the  current 


2Z2 
and 


''^z^W 


.      Z2  —  Z1. 

''-z:+z^' 

If,  therefore,  Z2  has  a  higher  value  than  Zi,  it  foDows  that  the 
voltage  of  the  traveling  wave  is  transmitted  to  the  second  circuit 
at  an  increased  amplitude  and  vice  versa.  A  traveling  wave 
originating  in  an  underground  cable  will,  therefore,  enter  an  over- 
head circuit  with  an  increase  in  voltage,  while  a  wave  originaUng 
in  an  overhead  circuit  will  pass  into  a  cable  system  with  a  lower 
voltage. 

These  relations  between  the  reflected  and  transmitted  waves  to 
the  incoming  wave  are,  however,  only  applicable  to  cases  where 


Digitized  by 


Google 


OVER..VOLTAGE  PROTECTION 


601 


rAAAAAAAA^— I 


z, 


z. 


the  wave  in  passing  the  transition  point  continues  its  travel  in  the 
form  of  a  wave;  that  is,  in  case  we  have  distributed  inductance 
and  capacity  on  both  sides  of  the  transition  point.  If,  on  the 
other  hand,  resistance,  inductance  and  capacity  are  concentrated 
at  the  transition  point,  the  conditions  become  entirely  dififerent, 
and  it  has  been  suggested  that  such  a  scheme  should  be  used  for 
protecting  transformers  and  machinery  against  the  traveling 
waves  entering  from  the  line.  The  use  of  inductance  and  capacity 
has  been  advocated  for  some  time,  and  both  have  the  properties 
of  changing  the  wave  front  of  the  transmitted  wave  so  that  it 
begins  with  zero  and  rises  gradually  to  its  full  value.  The  reflected 
wave,  however,  will  have  a  rectangular  or  steep  wave  front,  sim- 
ilar to  the  incoming  wave. 

The  energy  of  the  incoming  wave  is  naturally  also  split  up  in 
two  parts,  corresponding  to  the  transmitted  and  reflected  waves, 
but  there  is  no  reduction  in  the 
total  energy.  This  has  led  to  the 
suggestion  by  Gino  Campos  to 
use  a  resistance  shunted  across 
an  inductance  (see  Fig.  384).  In 
addition  to  considerably  smooth- 
ing out  the  wave  front  of  the 
transmitted  wave,  if  causes  some 
of  the  electro-magnetic  energy  to 
be  dissipated.  The  inductance 
forces  a  wave  with  steep  front 
to  pass  through  the  resistance, 
in  voltage  and  gives  the  transmitted  wave  a  lower  value  than  the 
incoming,  while  on  the  other  hand  part  of  the  energy  of  the  wave 
r  is  dissipated   into   heat.       The 

working  current,  however,  passes 
through  the  inductance  with  a 
negligible  drop.  This  combina- 
tion is  connected  in  series  with 
the  line,  as  shown. 

Another  combination  consist- 
ing of  a  resistance  in  series  with  a 
condenser  or  capacitance,  but  connected  between  the  line  wires 
or  between  the  line  wires  and  ground  is  shown  in  Fig.  385.  Both 
of  these  devices  or  combinations  are  particularly  effective  as 


L 

Fig.  384. — Protective  Device,  Con- 
sisting of  an  Inductance  Shunted 
by  a  Resistance'.  This  combina- 
tion is  for  Series  Ck)nnection  in  a 
Circuit. 

This,  in  turn,  results  in  a  drop 


■AWvAA- 


FiG.  385. — Protective  Device,  Con- 
sisting of  a  Capacitance  in  Series 
with  a  Resistance.  This  combi- 
nation is  used  in  shunt  with  a 
circuit. 


Digitized  by 


Google 


602 


ELECTRICAL  EQUIPMENT 


protective   devices   as    they 
dissipate  the  energy  of  high- 
frequency  waves.    They  are, 
therefore,   generally    termed 
"  high-frequency  absorbers/' 
Fig.  386  shows  how  the 
reflection  and  transmission  of 
a  traveling  wave  takes  place 
in  a  particular  case  with  in- 
ductance and  resistance  con- 
centrated at  the   transition 
point.    The  ampUtude  of  the 
waves  as  well  as  their  wave 
fronts  are,  of  course,  dep)en- 
dent  on  the  natural  impe- 
dancesof  the  circuits  on  either 
side  of  the  transition  point, 
as  well  as  on  the  value  of  the 
inductance     and    resistance 
concentrated  at  this  point. 
The    calculations 
are,  however,  of  a 
rather    intricate 
nature   and    be- 
yond the  scope  of 
this  book.    It  is 
seen,      however, 
that  with  a   pro- 
tective device  of 
this     kind,    both 
the     trsmsmitted 
and       reflected 
waves  have  steep 
fronts  although  of 
less  height    than 
the  original  wave. 
This   has  led   to 
Fig.   386.— The   Reflection   and   Transmission   of  a  the  suggestion  of 
Traveling  Wave  with  Concentrated  Inductance  and  adding     a     con- 
Resistance  at  the  Transmission  Point.  denser  to  Campos' 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  603 

combination,  in  which  case  the  voltage  at  the  front  of  both  the 
reflected  and  transmitted  waves  would  be  zero.  Both  these 
devices  are  patented. 

The  above  has  dealt  with  the  excess  voltages  which  could 
occur  when  a  line  is  connected  to  a  source  of  energy.  Dangerous 
voltages  are,  however,  also  liable  to  be  set  up  when  a  loaded  or 
short-circuited  line  is  suddenly  broken.  In  this  case  the  voltage 
rise  depends  on  the  value  of  the  interrupted  current,  and  the  rapid- 
ity with  which  the  circuit  is  broken,  and  again  on  the  natural 
impedance  of  the  circuit. 

It  was  previously  shown  that  the  energy  of  a  circuit  was  stored 
in  both  the  magnetic  and  dielectric  fields,  corresponding  to  the 
current  and  voltage  values.  At  a  certain  instant,  therefore,  the 
two  stored  quantities  are  equal,  while  if  the  current  is  zero  all  the 
energy  must,  of  course,  be  stored  in  the  dielectric  field  and  vice 
versa.    We  thus  had: 

2        2  ' 
and  the  relation  between  voltage  and  current 


^=W- 


For  transmission  work  the  ratio 

-^  =  138  log  —  ohms, 

C  T 

and  this  value  generally  falls  between  400  and  200  ohms.  For 
transformers,  however,  it  is  considerably  higher,  being  around 
3000,  while  an  tmderground  cable  has  a  much  lower  natiural  im- 
pedance than  an  overhead  circuit. 

For  example,  if  in  a  circuit  having  a  natural  impedance  of 
400  ohms,  a  current  with  a  maximum  value  of  200  amperes  is 
suddenly  broken,  the  surge  pressure  cannot  exceed  200X400= 
80,000  volts,  because  this  is  the  maximum  value  of  the  voltage 
wave  which  is  necessary  for  storing  in  the  dielectric  field  the  whole 
amount  of  energy  which  was  previously  stored  in  the  electro- 
magnetic field. 

Traveling  waves  similar  to  the  above  are  also  set  up  by  atmos- 
pheric lightning  phenomena.  The  gradual  accumulation  of  static 
charge  on  a  line  from  the  neighboring  atmosphere  increases  its 


Digitized  by 


Google 


604  ELECTRICAL  EQUIPMENT 

potential  with  respect  to  the  earth,  which  may  ultimately  become 
so  great  as  to  puncture  the  insulators.  Suppose  now  that  there 
is  a  Ughtning  discharge  between  cloud  and  cloud  or  between 
cloud  and  ground.  This  is  followed  immediately  by  a  redistri- 
bution of  the  electrostatic  field,  and  a  general  equalization  of 
potential  occurs.  The  static  charge  so  set  free  moves  along  the 
line  as  an  impulse  or  traveling  wave.  Such  waves  may  have  a 
potential  many  times  greater  than  that  caused  by  switching,  and 
they  may  have  a  very  steep  wave  front  and  thus  produce  high 
potential  differences  between  points  along  the  conductor,  such  as 
across  individual  transformer  coils  or  group  of  coils. 

Several  forms  of  protective  devices  of  more  or  less  value  have 
been  devised  to  guard  against  abnormal  voltage  conditions.  Of 
these  the  aluminum-cell  electrolytic  Ughtning  arrester  possesses 
ideal  characteristics  against  such  high-voltage  disturbances, 
where  the  excess  voltage  occurs  between  the  phase  conductors  or 
between  the  phase  conductors  and  ground.  The  films  of  the 
arrester  introduce  a  barrier  to  the  normal  potential  of  the  system, 
but  allow  the  energy  of  an  abnormal  disturbance  to  discharge 
readily.  The  arrester  is  generally  used  in  connection  with  choke 
coils,  the  function  of  which  is  to  retard  and  reflect  the  incoming 
waves  sufficiently  to  allow  the  arrester  to  better  perform  its  duty. 

Overhead  ground  wires  are  also  very  generally  used  to  protect 
transmission  lines  against  excessive  static  charges,  the  cost  of 
high-voltage  lightning  arresters  making  their  installation  along 
the  line  impractical. 

The  nature  of  high-frequency  disturbances  is  a  comparatively 
recent  discovery,  and  the  means  and  methods  for  preventing 
them  and  protecting  against  them  is  still  being  studied  and  inves- 
tigated. The  greater  damage  caused  by  such  high-frequency 
disturbances  has  occurred  in  high-voltage  transformers,  as  would 
naturally  be  expected.  The  best  protection  against  them,  there- 
fore, is  to  insulate  heavily  the  individual  coil  groups,  while 
inductances  and  energy-absorbing  devices  may,  as  stated,  have  to 
be  relied  upon  for  further  protection. 

Lightning  Arresters.  Aluminum-cell  electrolytic  lightning 
arresters  are  nowadays  used  almost  entirely  for  Ughtning  pro- 
tection of  high-voltage  transmission  systems.  This  type  of 
arrester  has  an  enormous  discharge  capacity,  and  its  general 
characteristics  are  well  known.     The  arrester,  however,  is  not  a 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  605 

universal  protector  against  all  kinds  of  interruptions.  For 
example,  while  it  meets  the  usual,  and  most  of  the  tmusual,  needs 
in  protection  against  disruptive  potentials  from  lightning,  an 
arrester  located  in  the  station  cannot,  and  is  not  expected  to,  pro- 
tect an  insulator  out  on  the  line  from  a  lightning  flash.  Neither 
is  it  designed  to  protect  against  surges  of  comparatively  low 
potential. 

The  design  is  based  on  the  characteristics  of  a  oeU  consisting 
of  two  aluminum  plates,  on  which  has  been  formed  a  film  of  hydrox- 
ide of  aluminum,  immersed  in  a  suitable  electrolyte.  This  film  is 
formed  on  the  aluminum  plates  by  a  series  of  chemical  and  electro- 
chemical treatments  at  the  factory.  Up  to  a  certain  critical 
voltage  this  hydroxide  film  has  the  property  of  insulating,  or 
rather  opposing,  the  flow  of  ciurent  and  is,  therefore,  closely 
analogous  to  a  counter  electro-motive  force.  Up  to  this  critical 
voltage  only  a  small  leakage  and  charging  current  can  flow,  but, 
during  any  rise  above  this  voltage  the  current  flow  through  the 
cell  is  limited  only  by  the  actual  resistance  of  the  electrolyte, 
which  is  very  low. 

The  action  is  comparable  to  that  of  the  well-known  safety 
valve  of  a  steam  boiler  by  which  the  steam  is  confined  until  the 
pressure  rises  to  a  given  value,  at  which  point  the  valve  opens  and 
releases  the  excess  pressure.  This  action  of  the  aliuninum  cell  is 
also  closely  analogous  to  that  of  a  storage  battery  on  direct-cur- 
rent. Up  to  about  two  volts  per  cell,  the  storage  battery,  when 
charged,  opposes  an  equal  counter  electro-motive  force,  shutting 
off  the  flow  of  current;  but  for  voltages  above  this  value  the  cur- 
rent is  limited  only  by  the  internal  resistance  of  the  cell.  This 
characteristic  makes  the  aluminum'  cell  ideal  as  a  means  of  dis- 
charging abnormal  potentials  or  surges  in  electric  circuits.  It 
practically  prevents  the  flow  of  current  at  operating  voltages, 
but  instantly  short-circuits  such  abnormal  portion  of  a  potential 
wave,  or  surge,  as  would  be  dangerous  to  the  insulation  of  the 
system. 

A  volt-ampere  characteristic  curve  of  the  aluminum  ceD  on 
alternating-current  is  shown  in  Fig.  387,  and  it  should  be  noted 
that  the  critical  alternating-current  voltage  is  between  336  and 
360  volts.  This  curve  gives  the  discharge  rate  only  up  to  5  am- 
peres in  order  to  better  illustrate  the  normal  and  critical  voltage 
points.     Above  this  value  the  discharge   rate  depends  almost 


Digitized  by 


Google 


606 


ELECTRICAL  EQUIPMENT 


entirely  upon  the  internal  resistance  of  the  electrolyte,  for  exam- 
ple, at  double  the  nonnal  operating  voltage,  or  600  volts  per  cell, 
the  current  discharge  is  about  600  amperes  for  a  brief  time. 

When  a  cell  is  connected  permanently  to  the  circuit,  two  con- 
ditions of  voltage  are  involved,  which  may  be  distinguished  as  the 
temporary  critical  voltage  and  the  permanent  critical  voltage. 
For  example,  if  each  cell  has  300  volts  appUed  to  it  constantly, 
and  the  voltage  is  suddenly  raised  to,  say,  326  volts,  there  will  be 
a  considerable  rush  of  current  until  the  film  thickness  has  been  in- 
creased to  withstand  the  extra  26  volts;    this  usually  requiring 


*00|  1  1  1  1  r  T  1  1  1 

;::;:e;e;I::"4:-!-t 

,    . ,  ,  1 

.....  .^_^=t^ 

=  T=" 

— r  ^^  ^      •    - 

/ 

Hz    :    ""       ,_^ 

r 

^             1 

r    t     *"                       n~ 

1 

sW     3--    *  - 

x^  iT"ii" 

+  T  '^ 

, 

&        IE    ~ 

«ifln     X--        -H 

£      13     -1-  - 

1                                     1_ 

J 

1 

80    7          ■ 

~r 

/                    —1 

I~ 

--    -  ::     ::^ 

Amperes 
Fig.  387. — ^Volt-ampere  Characteristic  Curve  of  an  Aluminum  Cell  on  Alter- 
nating Current. 

several  seconds.  In  this  case  325  volts  is  the  temporary  critical 
value  of  the  cell.  Similar  action  will  occur  at  any  potential  up  to 
about  the  permanent  critical  voltage,  or  the  voltage  at  which  the 
film  cannot  fmi^her  thicken  and  therefore  allows  a  free  flow  of 
current.  If  the  voltage  is  agam  reduced  to  300,  the  excess  thick- 
ness of  the  film  will  be  gradually  dissolved,  and  if  it  varies  period- 
ically between  two  values,  each  of  which  is  less  than  the  perma- 
nent critical  value,  the  temporary  critical  voltage  will  be  higher. 
This  feature  is  of  great  importance  as  it  provides  a  means  of  dis- 
charging abnormal  surges  the  instant  the  pressing  rises  above  the 
impressed  value. 

The  number  of  cells  for  a  circuit  is  so  chosen  that  the  maxi- 
miun  voltage  per  cell  will  be  approximately  300  volts,  or  always 
less  than  the  permanent  critical  voltage. 

Besides  the  valve  action  already  described  there  is  another 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  607 

characteristic  of  the  cell  of  great  importance.  The  thin  insulating 
film  of  aluminum  hydroxide  between  the  conducting  aluminum 
and  the  conducting  electrolyte  acts  as  a  dielectric,  and  the  cell, 
therefore,  is  an  electrostatic  condenser.  Due  to  this  capacity, 
however,  aluminum  arresters  cannot  be  connected  permanently 
to  the  circuits  and  horn  gaps  are,  therefore,  inserted  in  series  with 
the  connections. 

Another  characteristic  of  the  aluminum  ceU  is  the  dissolution 
of  a  part  of  the  film  when  the  plates  stand  in  the  electrol3rte  and 
the  cell  is  disconnected  from  the  circuit.  The  film  is  composed  of 
two  parts;  one  part  is  hard  and  insoluble,  and  apparently  acts  as 
a  skeleton  to  hold  the  more  soluble  part.  The  action  of  the  cell 
seems  to  indicate  that  the  soluble  part  of  the  film  is  composed  of 
gases  in  a  liquid  form.  When  a  cell  which  has  stood  for  some  time 
disconnected  is  reconnected  to  the  circuit,  there  is  a  momentary 
rush  of  current  which  re-forms  the  part  of  the  film  which  has  dis- 
solved. This  current  rush  will  have  increasing  values  as  the  inter- 
vals of  rest  of  the  cell  are  made  greater.  If  the  cell  has  stood  dis- 
connected from  the  circuit  for  some  time,  especially  in  a  warm 
cUmate,  there  is  a  possibility  that  the  initial  current  rush  will  be 
sufllcient  to  open  the  circuit  breakers  or  oil  switches.  This  cur- 
rent rush  also  raised  the  temperature  of  the  cell,  and  if  this  tem- 
perature rise  is  great  it  is  objectionable.  When  the  cells  do  not 
stand  for  more  than  a  day,  however,  the  film  dissolution  and  initial 
current  rush  are  negligible.  Suitable  means,  as  later  described, 
are  provided  with  the  arresters  for  throwing  them  directly  on  the 
line  and  charging  them  by  a  very  simple  operation,  and  thus  the 
film  may  be  always  kept  in  good  condition. 

The  aluminum  Ughtning  arresters  for  alternating-current  cir- 
cuits from  1000  to  155,000  volts  consist  essentially  of  inverted 
aluminum  cones  arranged  in  stacks  and  insulated  from  one  another 
(Kg.  388).  An  electrolyte  partially  fills  the  space  between  adjacent 
cones,  so  forming  aluminum  cells  connected  in  series.  The  stacks 
of  cones  with  the  electrolyte  between  them  are  then  inmiersed  in 
a  tank  of  oil.  The  electrolyte  being  heavier  than  the  oil  remains 
between  the  aluminum  cones.  Between  the  stack  of  cones  and 
the  steel  tank,  tubes  of  insulating  material  are  placed.  These 
improve  the  circulation  of  the  oil  and  increase  the  insulation 
between  the  live  parts.  The  oil  improves  the  insulation  between 
cones,  prevents  evaporation  of  the  solution  and,  due  to  its  heat- 


Digitized  by 


Google 


608 


KLECTRICAL  EQUIPMENT 


To  Sphere  -  Horn  OafJ 


Cast  dr^ffC9r*!^wr, 


£  ^ft^i^  ^hm^Se^ts 


ff^ufaty 


Fig.  388.— Section  through  Tank  of  130,000-volt  Aluminum-cell  Lightning 

Arrester. 


Digitized  by 


Google 


OVERrVOLTAGE  PROTECTION  609 

absorbing  capacity,  enables  the  arresters  to  discharge  continuously 
for  long  periods,  a  very  valuable  feature  of  these  arresters.  The 
tanks  are  of  steel  with  welded  seams. 

The  location  and  arrangement  of  aluminum  lightning  arrester 
installation  depend  greatly  upon  the  station  layout.  In  general 
the  arrester  should  be  installed  as  near  as  possible  to  the  apparatus 
or  station  to  be  protected.  The  ideal  arrangement  would  be  to 
have  the  tanks  and  horn  gaps  installed  as  a  complete  imit  just 
inside  the  station.  For  lower  voltage  equipments  this  is  feasible, 
as  the  arcing  at  the  gaps  is  not  severe  even  in  abnormal  cases. 
Above  27,000  volts,  this  practice  is  usually  questionable  and  it  is 
reconmiended  that  the  horn  gaps  be  installed  outside  the  building, 
with  leads  tapping  the  line  near  its  entrance  to  the  station,  the 
object,  of  course,  being  to  isolate  any  arc  from  the  station  appara- 
tus. The  tanks,  cones  and  transfer  device  may  be  installed  inside 
of  the  station  in  a  suitable  compartment.  This  requires  the  use 
of  an  extra  set  of  either  wall  or  roof  entrance  bushings  in  addition 
to  those  used  for  the  hne  entrance  leads. 

Wherever  horn  gaps  are  mounted  inside  the  building  sufficient 
clearance  should  be  allowed  over  them.  The  exact  distance  to 
be  allowed  depends  upon  the  voltage  and  the  nature  of  the  material 
or  apparatus  under  which  the  horns  are  installed.  If  there  are 
cables,  wires,  buses,  or  any  material  which  would  be  damaged 
by  fire,  considerable  distance  should  be  allowed.  On  the  other 
hand,  if  there  are  only  concrete  and  iron  beams  of  the  floor  or 
roof  a  much  smaller  clearance  is  permissible.  Normally  there  is 
no  appreciable  arc  at  the  gaps,  but  in  abnormal  cases  where  the 
film  has  been  allowed  to  get  out  of  order,  the  arc  might  be  of 
considerable  size.  Where  there  are  no  buses  or  inflammable 
apparatus,  the  following  are  the  minimum  clearances  from  the 
tops  of  horns  to  be  allowed: 

Feet. 

Up  to  16,100  volts 3 

16,100  to  37,900  volts 4 

37,900  to  75,000  volts 6 

Above  75,000  volts,  the  horn  gaps  should  never  be  placed 
indoors. 

The  horn  gaps  for  arresters  for  27,000  volts  and  above  are 
supported  on  a  pipe  framework  which  is  so  designed  as  to  permit 
mounting  op  either  wooden  or  steel  towers,  or,  if  desirable,  on  the 


Digitized  by 


Google 


610 


ELECTRICAL  EQUIPMENT 


roof  of  the  station  or  on  suitable  brackets  on  the  outside  wall  of  the 
station.  They  should  be  so  located  that  the  pipe  and  lever,  by 
which  they  are  operated,  can  be  brought  down  in  a  place  convenient 
for  the  operator  and  if  possible  where  he  can  observe  the  arcing 
at  the  horns  during  discharge. 

With  lightning  arrester  equipments  for  higher  voltages  there 
is,  however,  a  growing  tendency  to  install  the  entire  equipment 
outdoors  (Fig.  389).  Any  objection  to  installing  arrester  tanks 
out  of  doors  comes,  of  coiu*se,  from  the  increased  liabiUty  of  freez- 
ing the  electrolyte  in  cold  weather  and  the  abnormal  film  dis- 


FiQ.  389. — Outdoor  Iiightning  Arrester  Installation. 

solution  when  exposed  to  the  sun  on  hot  da3rs.  This,  therefore, 
has  a  bearing  on  the  electrol3rte  which  should  be  used.  These 
are  two  kinds,  both  of  which  freeze  at  about  20°  F.  One  will, 
however,  better  withstand  severe  winter  temperatures  and  the 
other  excessive  sununer  temperatures.  Should,  therefore,  for 
example,  the  operating  temperature  during  summer  exceed  100°  F., 
with  freezing  temperature  in  winter,  it  would  be  preferable  to 
use  the  electrolyte  for  the  warm  weather  and  provide  means 
to  prevent  freezing  during  the  winter  months.  The  electroljie 
may  not  be  injured  by  freezing,  but  when  frozen  the  internal 
resistance  of  the  arrester  is  considerably  increased  and  hence 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION 


611 


its  discharge  rate  is  materially  lowered.  Where  warm  climatic 
conditions  prevail  the  arrester  should  be  in  as  cool  a  place  as  pos- 
sible and  protected  from  the  direct  rays  of  the  sim  (Fig.  390). 
A  high  initial  temperature  will  reduce  the  available  heat  storage 
capacity  of  an  arrester  and  its  abiUty  to  care  for  long  continuous 
discharges.  A  high  operating  temperature  also  increases  the  rate 
of  dissolution  of  the  films  which  would  necessitate  more  frequent 
charging.  In  some  cases  it  may  be  found  advisable  to  charge 
two  or  more  times  a  day.  When  operating  imder  conditions  of 
high  temperature  any  failure  to  periodically  charge  the  arrester 


Fio.  390. — Oatdoor  Ughtning  Arrester  Installation  Showing  Protecting  Shield 

against  Sun. 

increases  the  liabiUty  of  damage  from  a  heavy  charging  cur- 
rent. 

Only  arresters  of  the  outdoor  type,  with  special  bushings  and 
covers,  should  be  installed  out  of  doors.  Care  must  be  taken  to 
see  that  the  bushings  and  covers  are  correctly  assembled  to  be 
water-tight.  The  arresters  may  be  mounted  either  on  a  platform 
between  poles  or  on  a  platform  near  the  ground  and  surrounded 
by  a  fence.  The  position  of  the  arresters  should  preferably  be 
such  that  their  operation  can  be  observed  by  the  station  attendant. 
While  installing  arresters  out  of  doors  care  must  be  taken  not  to 
let  the  wooden  and  fiber  parts  of  the  cone  stack  become  wet  in 
case  of  rain  and  to  keep  dust  from  the  cones  and  electrol3rte. 


Digitized  by 


Google 


612  '  ELECTRICAL  EQUIPMENT 

The  wiring  connections  of  lightning  arresters  are  important. 
The  discharge  circuit  should  contain  minimum  impedance  and 
hence  must  furnish  the  shortest  and  most  direct  path  from  Une 
to  groimd.  The  most  severe  disturbances  which  an  arrester  is 
called  upon  to  handle  are  of  high  frequencies,  and  it  is  therefore 
imperative  to  eliminate  all  necessary  inductance.  The  features 
favorable  for  low  inductance  are  short  length  of  conductor,  large 
radius  bends  and  large  surface  of  conductor.  Copper  tubing  is 
strongly  recommended  for  wiring  high-voltage  arresters.  It  has 
the  advantage  over  either  copper  strip  or  solid  conductors  in  that 
it  is  easily  supported,  requires  fewer  insulators,  and  is  therefore 
cheaper  to  install. 

In  all  lightning  arrester  installations,  good,  permanent,  low- 
resistance  grounds  are  essential  for  the  satisfactory  operation  of 
the  arresters.  Poor  grounds  cause  loss  in  protection  with  an 
ultimate  loss  in  apparatus.  It  has  been  customary  to  ground  a 
lightning  arrester  by  means  of  a  large  metal  plate  buried  in  a  bed 
of  charcoal  at  a  depth  of  6  or  8  feet  in  the  earth.  A  more  satis- 
factory method  of  making  a  ground  is  to  drive  a  mmiber  of  1-inch 
iron  pipes  6  or  8  feet  into  the  earth  about  the  station,  connecting 
all  these  pipes  together  by  means  of  a  copper  wire,  or,  preferably, 
by  a  thin  copper  strip.  A  quantity  of  salt  should  be  placed 
around  each  pipe  under  the  surface  of  the  earth  and  the  ground 
thoroughly  moistened  with  water.  It  is  advisable  to  connect 
these  earth  pipes  to  the  iron  framework  of  the  station,  and  also 
to  any  water  mains,  metal  flumes,  or  trolley  rails  that  are  avail- 
able. For  the  usual  size  station  the  following  recommendation  is 
made:  place  three  earth  pipes  equally  spaced  near  each  outside 
wall,  making  twelve  altogether,  and  place  three  extra  pipes 
spaced  about  6  feet  apart  at  a  point  nearest  the  arrester. 

Where  plates  are  placed  in  streams  of  running  water,  they 
should  be  buried  in  the  mud  along  the  bank  in  preference  to  laying 
them  in  the  stream.  Streams  with  rocky  bottoms  are  to  be 
avoided.  Whenever  plates  are  placed  at  any  distance  from  the 
arrester  it  is  necessary  also  to  drive  a  pipe  in  the  earth  directly 
beneath  the  arrester,  thus  making  the  ground  connections  as  short 
as  possible.  Earth  plates  at  a  distance  cannot  be  depended  upon. 
Long  ground  wires  in  a  station  can  not  be  depended  upon  imless  a 
lead  is  carried  to  the  multiple  earth  pipes  described  above.  As  it 
is  advisable  to  occasionally  examine  the  underground  connections 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  613 

to  see  that  they  are  in  proper  condition,  it  is  well  to  keep  on  file 
exact  plans  of  the  location  of  ground  plates,  ground  wires  and 
pipes,  with  a  brief  description,  so  that  the  data  may  be  readily 
referred  to.  From  time  to  time  the  resistance  of  these  ground 
connections  should  be  measured  to  determine  their  condition. 
This  is  very  easily  done  when  pipe  grounds  are  installed,  as  the 
resistance  of  one  pipe  can  be  accurately  determined  when  three  or 
more  pipes  are  used.  For  example:  If  there  are  three  pipes, 
namely,  Z,  7,  and  Z,  and  the  resistance  of  X+Y=20  ohms,  as 
measured  by  a  voltmeter,  the  resistance  of  X+Z  =  15  ohms,  and 
the  resistance  of  Y+Z  =  20  ohms,  then,  by  solving  the  equations: 

X+Y=20; 

X+Z=15 

Y—Z=  5  subtracting; 

y-z=  5 

Y+Z  =  20 


2F      =25  adding 

y=12iohms 

Z  =  20-12^  =  7iohms 

-X:  =  15-7i  =  7iohms. 

The  resistance  of  a  single  pipe  ground  in  good  condition  has  an 
average  value  of  about  15  ohms.  A  more  approximate  method 
of  keeping  accotmt  of  the  condition  of  the  earth  connections  is  to 
divide  the  earth  pipes  into  two  groups  and  connect  each  group  to 
the  110-volt  Ughting  circuit  with  an  ammeter  in  series.  If  there  is 
a  flow  of  about  20  amperes  the  conditions  are  satisfactory  pro- 
vided the  earth  pipes  are  properly  distributed  around  the  station. 
Aluminum  cell  arresters  for  non-grounded  as  well  as  grounded 
circuits  above  7250  volts  consist  of  four  units,  each  containing  a 
single  or  a  double  stack  of  cells  depending  on  the  voltage.  Three 
of  the  units  have  one  terminal  connected  to  the  circuit,  the  other 
being  connected  together;  the  fourth  unit  is  inserted  between  this 
multiple  connection  and  ground.  This  gives  the  same  protection 
between  line  and  line  as  between  line  and  groimd.  A  transfer 
device  is  provided  for  interchanging  the  ground  unit  with  one  of 
the  line  imits  during  the  charging  operation  so  that  the  films  of  all 
the  cells  will  be  formed  to  the  same  value. 


Digitized  by 


Google 


614  ELECTRICAL  EQUIPMENT 

It  was  previously  stated  that  it  is  necessary  to  charge  the  cells 
from  time  to  time  to  prevent  the  dissolution  and  consequent  rush 
of  dynamic  current  which  would  otherwise  occur  when  the  arrester 
discharges.  The  charging  operation  consists  simply  in  simul- 
taneously closing  the  three  horn  gaps  and  holding  them  closed  for  a 
period  of  five  seconds,  the  full  line  potential  thus  applied  across 
the  line  cells  causing  a  small  charging  current  to  flow  and  reform 
the  films  to  their  normal  condition.  Thereafter,  with  the  horn 
gaps  open  in  their  normal  position,  the  position  of  the  transfer 
device  is  reversed  and  the  horn  gaps  again  closed  for  five  seconds 
and  returned  to  normal  position.  The  complete  charging  opera- 
tion takes  but  a  few  seconds  and  should  be  performed  daily  (ff 
even  oftener  should  conditions  so  demand. 

Most  arresters  are  now  provided  with  charging  resistances  so 
as  to  minimize  the  oscillations  set  up  by  the  chaining  and  their 
harmful  effects  on  nearby  telephone  lines,  at  the  same  time  also 
greatly  increasing  the  life  of  the  cones  and  the  electrol3rtes.  An 
auxiliary  horn  gap,  fitted  with  a  charging  contact,  and  in  series 
with  the  resistance  is  installed  above  and  in  parallel  with  the  main 
gap  (see  Fig.  391).  At  the  time  of  charging  the  contact  bridge 
the  auxiliary  gap  and  charges  the  cells  through  the  resistance, 
the  cmrent  flow  being  limited  to  a  modei-ate  value. 

The  charging  current  taken  by  an  aluminum  cell  arrester  is 
the  best  means  of  indicating  its  condition,  and  the  value  may 
readily  be  ascertained  by  a  device  known  as  a  charging-current 
indicator.  An  arrester  in  good  condition  has  a  charging  current 
of  approximately  0.26  ampere  on  25-cycle  circuits,  0.30  ampere  on 
40-cycle,  and  0.40  ampere  on  60-cycle  circuits.  Should  these 
values  be  doubled,  the  arrester  must  be  charged  more  frequently 
and  the  current  carefully  measured  until  it  comes  down  to  normal. 
It  is  only  when  this  additional  charging  fails  to  reduce  the  charging 
current  that  an  inspection  of  the  ceDs  is  necessary.  The  essential 
parts  of  the  charging-ciurent  indicator  are  an  ammeter  mounted 
on  a  specially  constructed  switch  stick  and  a  set  of  jacks.  These 
jacks  are  so  connected  in  the  arrester  circuit  that  when  the  amme- 
ter switch  stick  is  inserted  in  them  and  the  horn  gaps  shortrcir- 
cuited,  the  charging  current  flows  through  the  meter. 

Most  modem  arresters  have  their  horn  gaps  provided  with 
spheres  which  greatly  decrease  the  dielectric  spark  lag,  espedaDy 
for  voltages  with  steep  wave  fronts.    The  arrangement  shown 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION 


616 


Fig.  391. — Lightning  Arrester  Sphere   and  Horn  Gaps  with  Charging 

Resistance 


Digitized  by  VjOOQ IC 


616 


ELECTRICAL  EQUIPMENT 


in  Fig.  391  provides  three  gaps  which  may  be  so  set  as  to  provide 
three  paths  for  the  discharge.  All  low-frequency  discharges 
would  form  corona  and  ionize  the  gap  between  the  horns,  passing 
across  the  same  and  to  ground  through  the  resistance  and  the 
cells  while  a  high-frequency  discharge  would  pass  through  the 
upper  of  the  two  sphere  gaps  and  similarly  to  ground.  The 
lower  sphere  gap  has  a  wider  setting  than  the  upper  sphere  gap, 
but  if  the  quantity  of  the  discharge  is  too  great  to  be  dissipated 
through  the  upper  paths,  the  dischai^  automatically  shunts  to 


#- 


Fig.  392.— 90,000- volt  (^hoke  Coil  for  Station  Service. 

the  main  gap,  where  it  is  not  impeded  by  the  resistances,  and  goes 
directly  through  the  cells  to  ground.  The  resistance  is  of  low 
value  and  consequently  all  but  the  heaviest  discharges  are  taken 
care  of  by  the  auxiliary  paths. 

A  knowledge  of  all  discharges  is  of  immense  value  to  operating 
engineers  in  studjdng  conditions  of  abnormal  voltage  on  trans- 
mission and  cable  systems.  For  this  purpose  a  discharge  recorder 
has  been  developed,  which  will  register  the  time  and  nature  of  dis- 
charges through  an  arrester.  This  recorder  consists  of  four  spark 
gaps  so  arranged  that  the  discharges  between  lines  or  between 
lines  and  ground  pass  through  the  gaps.      The  spark  gaps  are 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  617 

assembled  with  a  clock-operated  drum  in  such  a  maimer  that  a 
continuous  record  is  obtained,  showing  all  discharges  by  means  of 
punctures  in  a  moving  roll  of  paper.  This  paper  passes  through 
the  gaps  at  a  rate  of  about  1  inch  per  horn*,  which  gives  an  accurate 
record  of  the  time  and  duration  of  each  discharge.  Besides  being 
valuable  in  recording  discharges  due  to  abnormal  voltages  on  a 
system,  the  discharge  recorder  is  of  value  in  indicating  and 
recording  the  daily  charging  of  the  Ughtning  arresters.  With  such 
a  recorder  it  can  be  told  whether  the  arresters  are  or  are  not 
being  properly  charged  by  the  station  operator;  and  besides  the 
puncture  gives  some  indication  of  the  condition  of  the  arrester. 

Except  in  underground  cable  systems,  choke  coils  should 
always  be  installed  in  the  circuit  between  the  Ughtning  arrester 
and  the  apparatus  to  be  protected,  thus  holding  back  incoming 


Fig.  393. — Strain-type  Suspension  Choke  Coil  for  Station  or  Outdoor  Service. 

impulse  from  the  latter  until  the  lightning  arrester  discharges  to 
earth. 

Choke  coils  are  built  either  according  to  a  stationary  or  sus- 
pension design.  Of  the  former,  the  hour-glass  type  (Fig.  392) 
is  the  most  satisfactory,  in  that  it  avoids  the  necessity  of  supports 
between  the  turns,  so  that  high-frequency  disturbances  in  ground 
are  prevented  from  passing  across  the  turns.  The  air  insulation 
between  the  turns  is  also  preferable,  so  that  in  case  of  impulses 
with  extremely  steep  wave  fronts,  causing  arcing  between  turns, 
they  will  re-insulate  themselves. 

Suspension  choke  coils  (Fig.  393)  can  usually  be  incorporated 
with  the  other  high-tension  wiring,  thus  saving  a  number  of  ex- 
pensive insulators,  for  ,which  reason  they  in  many  instances 
may  prove  preferable. 

Fig.  394  shows  a  thunderstorm  map  for  the  years  1604-1913, 
as  prepared  by  the  U.  S.  Weather  Bureau. 


Digitized  by 


Google 


618 


ELECTRICAL  EQUIPMENT 


Arcing  Ground  Suppressor.    Arcing  ground   suppressors   as 
well  as  short-circuit  suppressors,  described  in  the  next  section, 


are  used  for  protecting  Une  insulators  against  arcs  and  the  con- 
sequent vicious  surges  accompanying  such  accidental  arcs,  which 
generally  follow  after  Ughtning  discharges. 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION 


619 


The  arcing  ground  suppressor,  as  described  in  the  following, 
is  intended  to  be  used  with  non-grounded  systems.     It  is  abo 


Fig.  395. — Elementary  Diagram  of  Arcing  Ground  Suppressor. 

limited  to  steel  tower  lines,  as  on  a  wood-pole  line  the  resistance 
of  the  pole  is  liable  to  prevent  suflScient  current  flowing  to  ground 
to  reduce  the  potential  sufficiently  to  operate  the  relay. 

The  arcing  ground  suppressor,  as  generally  built,  consists 
of  three  single-pole,  independent,  motor-operated  oil  switches, 
electrically  and  mechanically 
interlocked,  to  prevent  more 
than  one  operating  at  the 
same  time.  Each  switch  is 
connected  to  ground  on  one 
side  and  to  the  line  on  the 
other.  The  suppressor  is  con- 
trolled by  a  phase-selecting 
relay,  which  remains  inactive 
while  the  system  is  balanced, 
but  when  unbalanced,  due  to  a 
ground  on  one  phase,  it  oper- 
ates the  corresponding  phase 
of  the  suppressor,  which,  in 
turn,  grounds  the  same  phase 
of  the  line,  thus  shunting  the 
current  and  extinguishing  the 
arc.  The  switch  is  then  auto- 
matically opened  and  will  remain  so  provided  that  the  ground  was 
only  temporary,  such  as  an  insulator  spilling  over.     If  the  ground 


Fig.  396.— Phase-selecting  Relay  for 
Arcing  Ground  Suppressor. 


Digitized  by 


Google 


620  ELECTRICAL  EQUIPMENT 

is  of  a  pennanent  nature;  such  as  caused  by  the  puncture  of  an 
insulator,  the  switch  will  immediately  close  a  second  time  and  be 
locked  in  the  closed  position  until  opened  by  hand  after  the  ground 
has  been  removed.  Should,  however,  the  switch  stay  open  for  a 
fraction  of  a  second  after  the  first  stroke,  the  "second  stroke 
device  "  would  become  inoperative,  as  it  will  only  come  into  action 
when  the  switch  starts  to  close  the  second  time  immediately  after 
the  first  time.  To  prevent  the  possible  operation  of  the  suppres- 
sor in  cases  of  short-circuits,  an  overload  relay  may  be  provided 
which  opens  the  control  circuit  of  the  suppressor. 

Fig.  395  shows  an  elementary  diagram  of  an  arcing  ground 
suppressor  and  Fig.  396  the  phase  selecting  relay  for  the  same. 

Short-circuit  Suppressor.  This  device  operates  on  the  same 
principle  as  the  arcing  ground  suppressor,  but  it  is  intended  for  use 
on  grounded  systems  where  any  arc  to  ground  would  form  a  short- 
circuit.  The  suppressor  is  connected  between  each  line  wire  and 
ground,  and  consists  of  a  fuse  in  series  with  a  gap  which  is  instantly 
closed  when  a  short-circuit,  caused  by  an  arc-over  or  ground, 
occurs.  The  arc  is  thus  shunted  until  the  fuse  blows  which  gives 
sufficient  time  to  allow  the  arc  to  extinguish  itself.  For  a  single- 
phase  short-circuit  two  of  the  fuses  will  blow  and  for  a  three-phase 
short-circuit  all  three  fuses.  If  the  trouble  does  not  clear  itself 
or  if  there  is  a  dead  ground,  of  course,  the  main  oil  circuit  breaker 
will  finally  disconnect  the  entire  circuit  as  usual. 

Protection  of  Telephone  Lines.  Telephone  lines  paralleling 
high-tension  power  transmission  Unes  are  subjected  to  influences 
which  may  under  certain  conditions  interfere  with  the  proper 
transmission  of  speech.  This  interfering  influence  is  in  all  cases 
due  to  the  static  induction  from  the  high-tension  transmission 
line.  Under  normal  operating  conditions,  that  is,  with  fairly  well- 
balanced  three-phase  circuits,  this  influence  will  be  sUght,  but  with 
abnormal  operating  conditions  on  the  transmission  line  the  effect 
created  on  a  telephone  line  may  increase  to  such  an  extent  as  to 
become  destructive.  In  addition  to  these  influences  the  tele- 
phone Une  is  subjected  to  disturbances  occasioned  by  lightning 
discharges,  which,  however,  are  very  similar  in  character  to  the 
effects  created  by  abnormal  conditions  on  the  transmission  line, 
that  is,  during  the  time  of  switching  with  unbalanced  phases  or 
arcing  grounds,  etc. 

Under  normal  operating  conditions  the  effect  of  the  static 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  621 

induction  upon  the  two  wires  of  the  telephone  line  is  practically 
the  same,  with  the  result  that  the  two  wires  wiU  assume  a  certain 
potential  with  regard  to  earth.  With  a  well  insulated  and  properly 
transposed  metallic  line,  the  potentials  of  each  wire  against  ground 
will  be  nearly  alike,  and  hence  there  will  be  no  difference  of  poten- 
tial between  the  two  wires  themselves.  In  telephone  work, 
however,  even  the  smallest  difference  of  potential  between  the 
wires  will  create  a  flow  of  current  through  the  telephone  receiver. 
This  current,  being  alternating,  produces  a  noise  in  the  receiver 
which  may  be  loud  enough  to  make  talking  impossible.  The 
higher  the  voltage  of  a  transmission  line  and  the  closer  the  tele- 
phone line  is  located  to  the  same,  the  more  prominent  will  be  the 
noise  in  the  telephone,  with  slightly  unbalanced  telephone  lines. 
As  this  disturbing  current  is  due  to  a  difference  of  potential,  it  is 
obvious  that  the  noise  in  the  receiver  is  in  a  measure  independent 
of  the  absolute  value  of  the  voltage  on  each  Une  to  ground,  and 
that  it  cannot  be  eUminated  unless  the  voltage  on  both  wires  be 
made  exactly  alike.  This  condition,  which  is  termed  '^  bal- 
anced," is  realized  by  properly  insulating  and  transposing  the 
telephone  lines.  The  larger  the  number  of  transpositions  per 
mile,  the  more  will  the  potential  on  the  wires  be  equalized  and  the 
better  the  insulation  of  the  Unes,  the  less  will  there  be  a  chance  for 
a  leak  to  ground,  causing  a  drop  of  potential  on  that  particular 
wire,  with  a  subsequent  result  of  unbalancing  the  line  and  ren- 
dering it  noisy. 

From  the  above,  it  will  be  seen  that  as  far  as  the  noise  on  the 
line  is  concerned  it  can  be  kept  down  within  any  limits,  provided 
the  telephone  line  is  properly  transposed  and  substantially  insu- 
lated. On  the  other  hand,  it  will  be  seen  that  the  existing  poten- 
tial between  telephone  lines  and  ground,  by  reaching  high  values 
may  not  necessarily  impair  the  transmission  of  speech,  but  will 
seriously  strain  the  insulation  of  the  instruments  and  make  the 
use  of  the  same  by  the  operators  dangerous. 

-  Various  schemes  and  devices  have  been  developed  for  the  pro- 
tection of  telephone  lines  with  more  or  less  satisfactory  results. 
The  proper  protective  equipment  to  be  used  depends  entirely  on 
the  arrangement  of  the  lines  and  the  abnormal  conditions  against 
which  it  is  requirecl  to  protect. 

For  lightning  disturbances  only,  the  standard  vacumn  gap 
gives  the  best  and  most  reliable  discharge  path  for  these  poten- 


Digitized  by 


Google 


622  ELECTRICAL  EQUIPMENT 

tials  to  ground.  On  the  other  hand,  where  there  are  induced 
potentials  in  the  telephone  line  either  between  lines  or  from  lines 
to  ground,  that  is  either  due  to  electro-magnetic  or  electrostatic 
induction,  a  multi-gap  arrester,  using  knurled  cylinders  for  the 
electrodes,  is  used  between  lines  and  ground.  This  is  to  avoid 
continual  grounding  of  the  telephone  lines  through  the  low  break- 
down path  of  the  vacuum  arrester  due  to  the  induced  potential  to 
ground  which  may  be  of  quite  high  value.  The  vacuum  gap  is 
put  across  the  telephone  lines  where  the  induced  potentials  can 
be  controlled  by  careful  transposition.  Here  the  vacuum  arrester 
holds  the  voltage  across  the  telephone  apparatus  to  a  value  below 
its  breakdown. 

Where  there  is  any  possibility  of  induction  troubles  and  this 
may  occm*  up  to  one-quarter  or  one-half  mile  away  from  the 
power  circuit  under  abnormal  conditions,  the  telephone  line  insu- 
lating transformer  is  of  prime  importance.  This  provides  an 
insulation  barrier  of  25,000  volts  test  between  the  telephone 
instruments  and  the  lines.  On  the  line  side  of  these  transformers, 
which  should  be  used  at  every  telephone  station,  are  installed 
the  combined  multi-gap  and  vacuum-gap  unit  which  hold  the 
voltages  to  ground  and  between  lines  to  moderate  values.  In 
series  with  this  in  the  telephone  lines  are  fused  switches  for  cut- 
ting of!  the  apparatus  in  case  of  heavy  continued  dischai^ges 
through  the  gaps,  caused  by  induced  potentials  or  crosses.  They 
can  also  be  operated  as  straight  switches  to  cut  off  the  station  in 
any  emergency. 

As  a  fiu-ther  protection  in  case  of  induced  potentials  particu- 
larly for  potentials  to  earth,  the  drainage  coil  or  bleeding  coil  can 
be  used.  These  should  be  few  in  niunber,  usually  two,  as  too 
many  will  seriously  affect  the  operation  of  the  telephone  circuits. 
These  coils  give  a  high  impedance  path  across  the  telephone  line 
thus  shunting  the  high-frequency  talking  currents,  but  provide  at 
the  same  time  a  low  impedance  path  for  the  flow  of  equal  currents 
from  both  lines  to  groimd  at  the  center  of  the  coil.  These  coils, 
where  used,  should  be  protected  by  cut-outs  to  guard  against 
burn-out  from  heavy  currents  under  abnormal  conditions  on  the 
power  line. 

With  the  addition  of  possible  crosses  with  the  power  line  the 
only  additional  feature  to  the  above  scheme  is  the  double-pole 
horn  gap  which  serves  as  an  auxiliary  protection  to  the  telephone 


Digitized  by 


Google 


OVER-VOLTAGE  PROTECTION  623 

line  insulation  until  the  phone  or  power  lines  burn  oflF.  Where 
there  is  a  cross  but  no  paralleling,  it  is  only  necessary  to  use  the 
fused  switch  on  either  side  of  the  cross  to  isolate  this  section  in 
case  of  a  break. 

Prom  the  standpoint  of  protection,  telephone  circuits  can  be 
classified  as  follows; 

Class  1.  Telephone  circuits  which  do  not  cross  or  parallel 
power  lines. 

Class  2.  Telephone  circuits  which  cross  but  do  not  parallel 
power  lines. 

Class  3.    Telephone  circuits  which  parallel  power  lines  but 
are  not  on  the  same  towers  or  poles  and  do  not  cross  power  Knes. 
Class  4.     Telephone  circuits  which  are  on  the  towers  or  poles 
with  the  power  lines. 

This  classification  covers  every  possible  case,  from  a  telephone 
line  far  removed  from  the  power  circuit  to  one  mounted  on  the 
transmission  towers  themselves.  Classes  3  and  4  are  the  most 
conunon.  The  sources  of  trouble  vary  from  lightning  only  in 
Class  1,  to  lightning,  crosses,  and  induction  in  Class  4. 

The  reconmiendations  for  the  protection  of  the  telephone 
circuit  according  to  the  classification  of  the  circuit  into  which  it 
falls  are  as  follows: 

Class  1.  Telephone  circuits  which  do  not  parallel  or  cross 
power  lines. 

Disturbances:    Lightning. 

Recommendations:    Vacuum-tube  lightning  arresters  from 
each  Une  to  groimd  at  all  telephone  stations. 
Class  2.    Telephone  circuits  which  cross  but  do  not  parallel 
power  lines. 

Disturbances:    These  circuits  are  subject  to  Ughtning  dis- 
turbances and  to  contact  with  high-voltage  power  lines 
through  broken  wires,  etc.  •  They  are  not  subject,  to  any 
extent,  to  electro-magnetic  or  electrostatic  induction. 
Recommendations : 

1.  Combined  double-pole  fused  switch  and  vacuum-tube 

Ughtning  arrester  in  series  with  the  main  telephone 
line  on  both  sides  of  crossing  at  nearest  telephone 
stations. 

2.  Combined  vacuum-tube  and  air-gap  lightning  arresters 

at  all  other  stations. 


Digitized  by 


Google 


624  ELECTRICAL  EQUIPMENT 

C^ass  3.    Telephone  circuits  which  parallel  power  lines,  but 
are  not  on  the  same  towers  or  poles  and  do  not  cross  power  lines. 
Disturbances:  These  circuits  are  subject  to  lightning  disturb- 
ances, and  electro-magnetic  and  electrostatic  induction. 
They  are  not  subject  to  contact  with  the  power  lines. 
Recommendations : 

1.  Insulating  transformers  at  all  telephone  stations. 

2.  C!ombined  double-pole  fused  switch  and  vacuum-tube 

Ughtning  arrester  at  all  telephone  stations  on  the  line 
side  of  the  insulating  transformer. 

3.  Drainage  coils,  preferably  one  at  each  end  of  line. 

A  diagram  of  connections  for  the  apparatus  used  on 
this  class  of  telephone  circuits  is  shown  in  Fig.  397.    The 

with  ExpuUlon  Pum*  *i»«i«"u«i  vf 


To 
Tslephond 
ApjMiratus 


.Vacoum  ■ 

Gap  Adjustable 


n 


Tal«phona  Una  Inaalaliajr  Transfocmar 


Fig.  397. — Diagram  of  Connections  for  Protective  Apparatus  Recommended 
for  Telephone  Lines,  Classes  3  and  4. 

double-pole  horn  gap  shown  on  the  diagram  is  not  used 
on  this  class  of  circuit,  but  on  circuits  coming  under 
Class  4. 
Class  4.    Telephone  circuits  which  are  carried  on  the  towers 
or  poles  with  the  power  lines. 

Distm-bances:    These  circuits  are  subject  to  lightning  dis- 
turbances, electrostatic  and   electro-magnetic  induction, 
and  to  crosses  with  the  power  lines. 
Recommendations : 

1.  Insulating  transformers  at  all  telephone  stations. 

2.  Combined  double-pole  fused  switch  and  vacuum-tube 

lightning  arrester  at  all  telephone  stations  on  the 
line  side  of  the  insulating  transformer. 

3.  Double-pole  horn  gap  across  line  at  each  station  on 

hne  side  of  all  other  apparatus  for  the  protection  of 


Digitized  by 


Google 


STATION  WIRING  625 

insulators  on  telephone  circuit  in  case  of  crosses  with 
power  lines  after  series  fuses  are  blown. 
4.  Drainage  coils  installed  with  fuses  at  each  end  of 
line;  possibly  an  additional  coil  at  the  middle  if 
the  voltage  to  ground  is  not  held  to  a  safe  value 
by  two  coils. 

10.  STATION  WnUNG 

Experience  has  shown  that  in  a  great  number  of  instances 
the  shut-down  of  power  plants  has  been  caused  by  a  defective 
installation  of  the  station  wiring.  The  design  and  construction 
of  the  cabling  and  wiring  siystem  of  a  station  is,  however,  of  equal 
importance  to  the  rest  of  the  equipment. 

It  is  obvious  that  the  main  electrical  conductors  should  be  of 
such  a  character  and  so  installed  as  to  minimize  as  far  as  possible 
any  trouble  from  short-circuits  or  grounds,  and  particularly  to 
confine  such  distiu-bances,  in  event  of  its  occurrence,  to  the  cir- 
cuit affected.  It  is  Ukewise  apparent  that  such  buses  or  circuits 
on  which  a  short  would  mean  a  complete  station  interruption 
should  be  stUl  better  insulated  and  protected. 

The  general  practice  of  not  providing  automatic  protection 
on  the  excitation  system  makes  it  essential  to  properly  install 
all  the  exciter  field  circuits  and  to  provide  sufficient  insulation  to 
care  for  the  high  inductive  voltage  inherent  to  field  circuits.  The 
safety  of  the  instrument  and  control  system  wiring  should  f mlher- 
more  not  be  neglected,  because  in  the  event  of  trouble  the  main 
circuits  may  become  involved  through  the  accidental  operation 
of  an  oil  switch  or  the  failure  of  a  switch  to  open  on  an  outside 
short-circuit.  Every  cable  and  wire  should,  therefore,  have  a 
definite  place  provided  for  it  in  advance,  just  as  much  as  any 
other  piece  of  machinery,  and  wires  carrjdng  currents  of  different 
voltages  should,  as  far  as  possible,  be  kept  apart  from  each  other. 

Insulation.  The  principal  materials  used  for  cable  insulation 
are:  rubber  compound,  saturated  paper,  and  varnished  cambric. 
Rubber  insulation  is  commonly  used  on  low-voltage  cables  of 
small  size— «ay  up  to  600  volts  and  No.  0000  B.  &  S.  For  larger 
sizes  and  higher  voltages,  either  paper  or  varnished  cambric  in- 
sulation may  be  used.  The  latter  is  very  much  less  hydroscopic 
than  paper  insulation.  In  fact,  while  not  offered  as  being  water- 
proof in  itself  without  a  lead  sheath,  it  is  nevertheless  sufficiently 


Digitized  by 


Google 


626  ELECTRICAL  EQUIPMENT 

moisture  resisting  to  be  largely  used  in  braided  form  in  relatively 
dry  places.  In  lead-covered  form,  there  is  little  likelihood  of  an 
appreciable  amount  of  moisture  being  absorbed  at  the  ends  of  the 
cable  while  open  for  the  purpose  of  jointing  or  terminating.  This 
type  of  cable  is  likewise  mechanically  stronger  and  less  likely  to 
have  the  insulation  injured  during  installation. 

Of  two  cables — ^the  one  insulated  with  paper  and  the  other 
insulated  with  varnished  cloth — each  properly  proportioned  to 
stand  the  working  pressure  and  the  same  factory  tests,  if  each 
is  installed  by  the  same  installation  gang  and  under  the  same  con- 
ditions, that  insulated  with  varnished  cloth  will  have  the  greater 
factor  of  safety  after  installations  for  the  reason  just  mentioned, 
that  it  is  less  Ukely  to  be  injured  by  bending  and  less  likely  to 
absorb  moisture  while  the  ends  are  open.  It,  therefore,  does  not 
require  so  much  skill  ir.  handling  and  jointing.  Varnished  cloth 
insulation  Ukewise  has  the  characteristic  of  being  better  able 
safely  to  withstand,  temporarily,  higher  voltage  surges  without 
injury  than  either  rubber  or  paper  insulation. 

When  cables  are  run  exposed  the  insulation  should  be  pro- 
tected by  a  good  fireproof  covering  of  asbestos  so  that  in  case  of  a 
short-circuit  the  trouble  will  not  be  communicated  to  adjacent  cu*- 
cuits.  When  run  in  conduit  or  ducts  this  type  of  covering  absorbs 
moisture  and  the  weatherproof  covering  should  be  substituted; 
as  a  fact,  a  lead  covering  is  usually  required  for  damp  places. 

All  lead-covered  cables  should  be  provided  with  endbells  for 
preventing  moisture  from  entering  the  cable  at  the  ends.  These 
endbells  and  terminals  may  be  designed  for  either  horizontal 
or  inverted  positions  and  for  convenient  connections  to  the 
machine  terminals  or  busbars. 

Open  Wiring.  If  the  number  of  cables  in  close  proximity 
does  not  make  the  run  too  congested  or  hazardous,  it  may  be  per- 
missible to  use  wires  or  cables  insulated  for  full  potentiaJ,  rigidly 
supported  on  insulators  also  good  for  full-working  potential. 
This  arrangement  gives  double  protection,  since  either  the  insu- 
lation or  the  insulators  afford  sufficient  protection  in  case  one 
should  fail.  On  the  other  hand,  the  runs,  being  exposed,  are  under 
constant  observation.  Where  the  eonductor  does  not  exceed  No. 
0000  B.  &  S.  size,  it  should  be  solid  and  not  stranded,  the  former, 
of  course,  being  more  rigid.  Where  the  amount  of  current  to  be 
carried  is  large  copper  bars  are  used.    This  is  usually  the  case  for 


Digitized  by 


Google 


STATION  WIRING 


627 


bus-bars.  They  are  seldom  insulated  because  the  addition  of  insu- 
lation on  a  group  of  bars  greatly  reduces  their  carrying  capacity 
by  stopping  the  air  circulation  between  the  laminations. 

Where  the  voltage  exceeds  13,200  bare  conductors  consisting 
of  solid  wire,  copper  tubing  or  iron  pipe  are  generally  employed. 
The  use  of  tubing  or  pipe  makes  it  possible  to  reduce  the  niunber 
of  expensive  insulators  for  supporting  it.  To  insulate  such  high- 
voltage  conductors  is  expensive  and  quite  unnecessary  because 
when  properly  installed  they  are  widdy  spaced  and  kept  well 
away  from  the  floor. 

Table  LII  gives  dimensions  for  the  spacing  of  rigid  conductors. 
These  values  are  based  on  striking  distances  between  points,  and 
are  for  guidance  in  determining  proper  distances  between  con- 
ductors and  for  general  construction  work. 

TABLE  LII 
Spacing  of  Rigid  Conductobs 


DlMBNniONB 

IN  Inches. 

Outdoors. 

iLdoors. 

To  Ground. 

Between  Live 
Pa*ts. 

To  Ground. 

Between  Live 
Parts. 

2,000  to     3,500 

3i 

4 

3 

3J 

3,501  to     7,500 

6J 

6 

4i 

5i 

7,501  to    15,000 

9 

10 

7 

9 

15,001  to    25,000 

14 

151 

lOJ 

14 

25,001  to    37,000 

19J 

22 

14i 

19i 

37,001  to   60,000 

254 

29 

19 

25J 

50,001  to   73,000 

36 

41 

27 

36 

73,001  to   95,000 

47 

53 

34J 

47 

95,001  to  115,000 

56 

64 

41 

56 

115,001  to  135,000 

66 

75 

48 

66 

135,001  to  155,000 

75 

86 

55 

75 

155,001  to  175,000 

85 

97 

62 

85 

175,001  to  195,000 

94 

108 

69 

94 

Correction  for  Altitude 
Sea  level  to  1000  feet — Use  table. 

1000  to  3000  feet — Add  10  per  cent  to  spacing  in  table. 
3001  to  5000  feet — Add  20  per  cent  to  spacing  in  table. 
5001  to  7000  feet — Add  30  per  cent  to  spacing  in  table. 
7001  to  9000  feet — ^Add  40  per  cent  to  spacing  in  table. 


Digitized  by 


Google 


628  ELECTRICAL  EQUIPMENT 

Cable  should  be  supported  every  foiu*  feet  in  vertical  runs  and 
every  three  feet  in  horizontal  runs,  while  for  tubing  the  distance 
between  the  insulators  may  be  increased  to  about  10  feet.  When 
dealing  with  large  conductors  carrying  heavy  currents,  care 
should  be  taken,  as  explained  under  the  section  of  "  Current  Lim- 
iting Reactors,"  to  rigidly  support  them  so  that  they  will  not  be 
torn  from  their  supports  when  severe  short-circuits  occur. 

Cables  in  Ducts  or  Conduits.  It  is  not  always  convenient 
or  desirable  to  run  all  of  the  conductors  exposed  for  several  reasons. 
There  may  be  no  suitable  place  to  support  such  cables.  The 
congestion  may  be  so  great  that  it  would  be  hazardous  in  other 
respects.  They  may  be  subject  to  mechanical  injury.  They 
may  be  in  a  bad  location  from  a  "  safety  first  "  standpoint.  If 
therefore,  for  any  of  the  above  reasons  it  is  undesirable  to  run 
conductors  exposed,  then  they  may  be  run  in  conduit  or  ducts  and 
ma>  be  provided  with  a  protecting  weatherproof  braid  or  lead 
sheath  as  the  occasion  demands.  It  should  be  borne  in  mind  that 
if  the  lead  sheath  is  omitted  the  conduit  or  ducts  should  be  thor- 
oughly drained  to  some  pit  so  that  water  cannot  remain  in  them. 

Iron  conduit  should  not  be  employed  on  alternating  currents 
unless  all  conductors  of  the  circuit  are  in  the  same  conduit.  The 
general  practice  is  to  use  iron  conduit  up  to  about  two  inches  in 
diameter,  above  which  fiber  conduit  is  generally  used. 

This  type  of  conduit  is  formed  in  cylindrical  shape  from  fiber 
or  wood  pulp  under  pressure.  The  pulp  is  thoroughly  saturated 
with  a  bituminous  compound  so  as  to  kill  any  vegetable  matter  or 
bacteria  which  would  tend  to  promote  decay. 

It  has  been  found  that  the  majority  of  all  initial  cable  troubles 
are  directly  traceable  to  some  injury  to  the  lead  casing  when  being 
drawn  into  the  duct,  due  to  the  roughness  of  the  walls,  and  the 
cement  which  has  seeped  throu^  the  joint  and  formed  cutting 
edges  after  hardening.  Cable  troubles  are  also  due  to  stray  cur- 
rents leaking  through  the  joints,  as  a  result  of  improper  installa- 
tion and  the  impossibility  of  securing  proper  alignment.  These 
objections,  however,  are  eliminated  by  the  use  of  fiber  conduit, 
due  to  the  smooth  interior  and  water-tight  joints.  UnUke  joining 
tile  conduit,  the  connection  made  with  fiber  conduit  is  ideal, 
affording  perfect  alignment,  without  the  use  of  mandrels  or  dowel- 
pins,  and  not  having  to  use  cement,  mortar  or  burlap  at  the  joints. 
It  is  also  true  that  fiber  conduit  is  impervious  to  moisture,  gases, 


Digitized  by 


Google 


STATION  WIRING  629 

acids,  or  other  corrosive  elements;  thus,  water,  gas  and  stray  cur- 
rents cannot  reach  the  cable  protected  by  this  material.  It  is  a 
good  non-conductor,  doing  away  entirely  with  the  trouble  with 
stray  currents,  and  it  is  also  an  absolute  prevention  against 
electrolysis,  which  destroys  many  cables,  gas  and  water  pipes 
during  each  year. 

Control  and  instrument  wiring  and  field  and  exciter  circuits 
are  invariably  run  in  iron  conduit;  first,  because  they  are  so 
numerous  and  their  directions  varied,  and  second,  because  of 
their  small  size  they  require  protection  against  mechanical  injury. 
The  cheapest  and  least  conspicuous  place  of  installment  is  in  the 
concrete  floors. 

The  practice  of  choosing  a  conduit  having  an  inside  diameter 
at  least  30  per  cent  greater  than  the  outside  diameter  of  the  cable 
will  give  good  results,  and  Table  LIII  also  gives  the  size  of  con- 
duit recommended  for  different  sizes  of  conductors.  All  con- 
ductors of  cables  for  duct  service  should  be  stranded  to  facilitate 
installation. 

In  laying  out  a  conduit  job,  first  ciscertain  the  size  and  nimiber 
of  wires  required,  then  take  the  sizes  of  conduit  from  Table  LIII. 
One-half  inch  is  usually  used  for  branch  conduits  and  is  the  small- 
est size  permitted  by  the  National  Electric  Code.  In  running 
several  conduits  together,  a  pull-box  will  be  found  more  economical 
than  elbows  for  making  turns,  as  one  pull-box  will  take  the  place 
of  several  elbows.  Do  not  pull  wires  through  conduits  with  a 
block  and  tackle,  as  it  will  not  only  injure  the  insulation,  but 
wedge  the  wires  in  such  shape  that  they  cannot  be  removed  readily 
if  desired.  Be  careful  to  ream  out  the  end  when  conduit  is  cut, 
as  the  bur  may  otherwise  cut  through  the  insulation.  Conduits 
should  be  securely  fastened  to  walls  and  ceiling  by  use  of  pipe 
straps  or  hooks.  Plug  all  exposed  ends  of  conduit  in  new  buildings 
to  prevent  plaster  and  dirt  from  falling  into  it. 

Single  vs.  Multiple  Conductors.  Low-voltage  cables  for 
direct-current  service,  such  as  exciter  and  field  leads,  are  as  a  rule 
of  the  single-conductor  type.  This,  however,  does  not  refer  to 
control  and  instrument  wiring  for  which  multi-conductors  with  as 
many  as  a  dozen  conductors  are  used.  These  are  as  a  rule  of 
different-colored  braids  so  as  to  facilitate  identification  during 
installation. 

Whether  single-  or  multiple-conductor  cables  should  be  used  for 


Digitized  by 


Google 


630 


ELECTRICAL  EQUIPMENT 


TABLE  LIII 
Conduit  Sizes  for  Different  Size  Wires 


Am- 

Size or  Pipe. 

Am- 

Sizi 

E  OF  Pipe. 

No. 

Circular 

peres, 

Circular 

peres. 

B.  &S. 

Mils. 

Rub- 

1- 

2- 

3- 

Mils. 

Rub- 

1- 

2- 

3- 

ber. 

Wlre. 

Wlre. 

Wlre. 

ber. 

Wire. 

Wire. 

Wire. 

18 

1,020 

3 

1 

600,000 

390 

2 

2 

31 

16 

2,583 

6 

1 

660,000 

420 

2 

31 

14 

4,107 

12 

J 

600,000 

460 

2 

31 

12 

6,630 

17 

i 

660,000 

476 

2 

31 

10 

10,380 

24 

1 

700,000 

500 

2 

31 

8 

16,610 

33 

1 

760,000 

626 

2 

31 

6 

26,260 

46 

li 

800,000 

660 

2 

31 

5 

33,100 

64 

* 

11 

860,000 

675 

2i 

4 

4 

41,740 

66 

li 

900,000 

600 

21 

4 

4} 

3 

62,630 

76 

U 

960,000 

626 

2i 

4 

41 

2 

66,370 

90 

2 

1,000,000 

660 

2i 

4 

4i 

1 

83,690 

107 

2 

1,100,000 

690 

2i 

4 

5 

0 

106,600 

127 

2 

2 

1,200,000 

730 

2J 

4 

5 

2.0 

133,100 

160 

2 

2 

1,300,000 

770 

2i 

41 

5 

3.0 

167,800 

177 

2 

2J 

1,400,000 

810 

3 

41 

6 

4.0 

211,600 

210 

2 

2i 

1,600,000 

860 

3 

5 

6 

200,000 

200 

2 

2i 

1,600,000 

890 

3 

5 

6 

260,000 

236 

2i 

2i 

1,700,000 

930 

3 

6 

6 

300,000 

270 

2i 

3 

1,800,000 

970 

3 

6 

7 

360,000 

300 

2i 

3 

1,900,000 

1010 

3 

6 

7 

400,000 

330 

1§ 

3 

3 

2,000,000 

1060 

3 

6 

7 

450,000 

380 

2 

3 

3i 

the  alternating  main  conductors  depends  on  the  size,  length  of 
run  and  whether  they  are  lead  covered  or  not.  When  lead  cov- 
ering on  cables  is  required,  multiple-conductor  cables  are  always 
preferable,  since  the  eddy  currents  in  the  lead  sheaths  of  the  single- 
conductor  cables  increase  the  energy  loss.  In  fact,  single-con- 
ductor, lead-covered  cables  should  not  be  used  in  large  sizes  on 
alternating-current  circuits  without  careful  consideration. 

With  high-voltage,  single-conductor,  lead-covered  cables,  static 
discharges  may  take  place  through  the  insulation  to  the  lead, 
which  rapidly  injures  the  insulation  and  a  breakdown  soon  fol- 
lows. If  the  cable  is  not  lead-covered  a  static  discharge  may  take 
place  to  the  duct,  this  also  having  a  tendency  to  break  down  the 


Digitized  by 


Google 


STATION  WIRING  631 

insulation  in  time.  In  multiple-conductor  cables  this  action  does 
not  occur,  the  static  activity  being  neutralized. 

Single-conductor  cables  are  made  in  sizes  up  to  2,000,000  CM. 
and  three-conductor  cables  up  to  500,000  CM. 

General  Practice.  The  following  is  a  general  sunamaiy  of 
prevailing  practice  covering  the  kind  of  conductors  and  the  manner 
in  which  they  are  installed  in  a  station. 

Bare  Grounded  Conductors,  Bars,  tubing,  cable,  wire:  Used 
for  all  kinds  of  ground  connections  or  ground  return  circuits. 

Bare  Conductors  on  Insulators,  Bars,  tubing,  wire:  Generally 
employed  for  circuits  above  13,200  volts. 

Insulated  Conductors  on  Insulators,  Wire,  cable,  rods;  Used 
for  all  circuits  up  to  13,200  volts  when  not  housed  in  compartments 
or  conduits. 

Insulated  Conductors  in  Iron  Conduit,  Cable:  Employed 
for  voltages  up  to  1200  volts  generally  for  small-capacity  circuits 
where  size  of  conduit  does  not  exceed  2  inches. 

Insulated  Conductors  in  Clay  or  Fiber  Ducts.  Cable:  May  be 
used  for  large  capacity  circuits  for  voltages  up  to  13,000  provided 
ducts  are  maintained  free  from  moisture. 

Leaded  Conductors  in  Ducts  or  Conduits.  Cable:  Used  for 
voltages  up  to  13,200  when  ducts  or  conduits  are  subject  to  mois- 
ture. 

For  convenience  of  reference,  station  wiring  may  also  be  classi- 
fied as  follows: 

1.  Exciter  and  field  wiring. 

2.  A.C  generator  and  low-tension  transformer  wiring. 

3.  Control  and  instrument  wiring. 

4.  High-tension  wiring. 

Exciter  and  Field  Wiring.  These  leads  consist,  as  a  rule, 
of  single-conductor  rubber-covered  cables  with  a  double  weather- 
proof braid  (or  tap)e  and  braid),  although  for  sizes  larger  than  No. 
0000  B.  4&  S.  the  insulation  may  be  varnished  cambric.  Because 
of  the  inductive  discharge  in  field  circuits,  causing  an  excessive  rise 
in  potential  when  opening  the  circuit,  it  is  important  that  a  liberal 
margin  of  safety  is  allowed  in  the  insulation.  For  damp  locations 
lead-covered  cables  may  be  required.  These  leads  are  mostly 
installed  in  iron  conduits. 

Generator  and  Transformer  Wiring.  For  this  wiring  varnished 
cambric  insulation  is,  as  previously  stated,  preferable,  the  thick- 


Digitized  by 


Google 


632  ELECTRICAL  EQUIPMENT 

ness  of  the  insulation  varying  with  the  generator  voltage.  Fcnr 
absolutely  dry  locations  a  good  weatherproof  braid  may  well 
serve  as  a  mechanical  protection  against  abrasion,  but  the  ducts 
should  nevertheless  be  provided  with  drains  so  that  the  cables  will 
under  no  circumstances  lay  in  water  which  may  be  accumulated 
from  condensation.  For  damp  localities,  lead-covered  cables 
should  always  be  used,  and  to  be  on  the  safe  side  the  use  of  such 
cables  is  always  to  be  recommended.  Endbells  are  always  re- 
quired for  such  cables. 

.Exposed  main  wiring  is  generally  considered  out  of  date,  but, 
if  used,  the  cables  should  be  well  supported  and  guarded  and  per- 
fectly covered  with  a  fireproof  covering  to  prevent  a  fire  from 
spreading  from  one  circuit  to  another.  The  installation  of  the 
cables  in  ducts  or  conducts  is  much  to  be  preferred. 

Fiber  ducts  should  be  used  for  all  alternating-current  cables, 
although  iron  conduit  is  permissible  if  all  conductors  of  one  cir- 
cuit are  run  in  the  same  conduit.  With  single-conductor,  lead- 
covered  cables,  and  preferably  also  for  multi-conductor,  fiber 
conduits  should  be  used. 

Whether  single-  or  three-conductor  cables  are  to  be  used  de- 
pends on  the  size,  the  length  of  run  and  the  loss  in  the  lead  sheatL 
Single-conductor  cables  are,  as  stated  before,  made  in  much  larger 
sizes  than  three-conductor  and  have,  of  coimie,  a  greater  radiating 
capacity,  but  on  the  other  hand,  especially  for  long  runs,  it  is  found 
that  three-conductor  cables  will  be  more  economical,  especially 
for  lead-covered  cables.  This  is  evident  when  one  considers  that 
three  lead  sheaths,  each,  however,  somewhat  smaller,  will  be 
required  as  compared  to  one.  On  the  other  hand,  the  eddy-cur- 
rent losses  in  the  lead  sheath  for  a  single-conductor  cable  is  not 
negligible,  while  with  a  multi-conductor  cable  they  are  entirely 
neutralized.  Lead  sheaths  are  as  a  rule  groimded  at  one  end  to 
get  rid  of  accumulation  of  static  electricity  and  a  ground  of  the 
lead  sheath  at  the  other  end  of  the  cable  can  very  easily  occur 
without  being  noticed,  resulting  with  single-conductor  cables  io 
circulating  currents  in  the  lead  sheath.  These  cmrents  are  only 
hmited  by  the  resistance  of  the  lead  and  the  losses  caused  thereby 
may  be  quite  considerable.  Of  course,  where  the  size  is  such  that 
two  or  more  conductors  per  phase  are  required  it  is  posBible  to 
"  nest "  the  conductors  so  as  to  neutralize  the  inductive  effects. 

In  selecting  cables  for  generator  leads,  a  larger  factor  of  safety 


Digitized  by 


Google 


STATION  WIRING  633 

should  be  allowed  than  for  ordinary  cable  practice.  Since  such 
leads  are  not  usually  protected  by  any  automatic  circuit  breakers, 
it  is  good  practice  to  select  a  cable  for  this  purpose  with  an  insu- 
lation thickness  50  per  cent  greater  than  the  normal  working 
voltage  of  the  generator. 

Control  and  Instrument  Wiring.  Under  this  class  would  be 
grouped  the  control  circuits  for  oil  switches,  rheostat  and  governor 
motors,  etc.,  secondaries  of  current  and  potential  transformers 
and  all  other  similar  conductors.  These  conductors  are  always 
of  a  very  flexible  rubber-covered  weatherproof  multi-conductor 
type,  installed  in  iron  conduit.  Occasionally  where  the  location 
is  very  damp  a  lead  covering  may  be  desirable.  With  this  cable 
it  is  possible  to  pull  it  through  a  conduit  some  100  feet  in  length 
with  four  standard  conduit  bends  in  the  run. 

The  best  practice  is  to  lay  the  conduits  in  the  floor  and  let 
them  terminate  as  near  the  switchboard  sill  as  convenient.  Fre- 
quently the  ends  of  the  conduits  are  bent  to  point  upwards  and 
cut  to  extend  just  a  short  distance  above  the  finished  floor.  This 
often  necessitates  a  number  of  visible  crossings  of  the  leads  where 
the  conduits  cannot  be  run  to  the  desired  point.  To  obtain  a 
neater  construction,  a  pull-box  with  cover  can  be  provided  in  the 
floor  along  the  back  of  the  board,  and  the  conduits  arranged  so  as 
to  terminate  in  the  walls  of  the  box.  Provision  is  then  made  for 
bringing  the  leads  from  this  box  to  the  desired  point  at  the  bottom 
of  the  board,  the  necessaiy  splices  and  crossings  being  made  in  the 
box. 

High-tension  Wiring.  For  circuits  above  13,200  volts,  bare 
conductors  are  generally  used  because  of  the  increased  cost  of 
ordinary  insulation  for  such  high  voltages,  and  because  such  con- 
ductors are  necessarily  spaced  far  apart  and  generally  located  at  a 
considerable  distance  from  the  floor.  They  are,  therefore,  rigidly 
mounted  on  insulators  and  carefully  guarded. 

Size  of  Cables.  (Current-canying  Capacity.)  For  the  com- 
paratively short  nms  encountered  in  power  stations  the  size  of 
the  conductors  is  generally  governed  by  the  permissible  current- 
carrying  capacity  and  this,  in  turn,  is  determined  within  practical 
limits  by  the  maximum  temperature  which  the  insulation  sur- 
rounding it  will  withstand.  First,  the  temperature  must  not  be 
high  enough  to  cause  too  rapid  a  rate  of  deterioration  of  the  insu- 
lation.   This  temperature  is,  roughly,  86°  C.  for  saturated  paper, 


Digitized  by 


Google 


634  ELECTRICAL  EQUIPMENT 

75°  C.  for  cambric,  and  60°  C.  for  rubber.  Second,  the  tem- 
perature must  not  be  high  enough  to  decrease  the  puncturing 
resistance  of  the  insulation  below  safe  limits.  This  temperature 
varies  with  the  normal  working  e.m.f.  of  the  circuit.  Based  on 
these  two  considerations,  it  is  reconunended  that  the  maximum 
operating  temperatures  of  the  conductors  of  insulated  cables  be 
limited  to  the  values  given  below: 

Heating  and  Temperature  of  Cables  {Standardization  Ruks  of 
the  A  J.E,E,) .  The  maximum  safe-limiting  temperature  in  degrees 
C.  at  the  surface  of  the  conductor. in  a  cable  shall  be: 

For  impregnated  paper  insulation  (85--B); 

For  varnished  cambric  insulation  (75--B); 

For  rubber  compound  insulation  (60-0.25-B); 
where  E  represents  the  eflfective  operating  e.m.f.  in  kilovolts 
between  conductors. 

Thus,  at  a  working  pressure  of  6.6  Kv.,  the  mRYiTnnm  safe- 
limiting  temperature  at  the  surface  of  the  conductor  or  conduc- 
tors in  a  cable  would  be: 

For  impregnated  paper  insulation 78.4°  C. 

For  varnished  cambric  insulation 68.4°  C. 

For  rubber  compound  insulation 58.35°  C. 

The  actual  maximmn  safe  continuous-current  load  for  any 
given  cable  is  determined  primarily  by  the  temperature  of  the 
siUTOimding  medium  and  the  rate  of  radiation.  This  current 
value  is  greater  with  direct  than  with  alternating-currents,  and 
decreases  with  increasing  frequency,  being  less  for  a  frequency  of 
60  cycles  than  for  25  cycles.  This  difference  in  carrying  capacity 
for  direct-  and  alternating-current  is  of  sUght  practical  importance 
for  conductors  less  than  500,000  cir.  mils  in  area,  at  commercial 
frequencies,  i.e.,  25  and  60  cycles. 

Furthermore,  owing  to  the  fact  that  alternating-current  flowing 
in  large  cables  has  greater  density  on  the  surface  of  the  conductor 
than  in  the  center,  so-called  skin  eflfect,  an  ordinary  cable  will  not 
carry  as  many  amperes  alternating-current  with  the  same  tem- 
perature rise  as  it  will  direct-current.  To  overcome  this,  it  has 
in  the  past  been  conunon  practice  on  single-conductor  cables, 
700,000  cir.  mils  and  larger  for  60  cycles  and  1,000,000  cir.  mils  and 
larger  for  25  cycles,  to  make  up  the  cable  in  annular  form,  using  a 
non-conducting  core  (usually  fiber),  and   stranding  the  copper 


Digitized  by 


Google 


STATION  WIRING  635 

wires  around  this.  The  annular  form  thus  increases  the  canying 
capacity  by  utiUzing  more  of  the.  copper  and  there  is  a  further 
increase  in  the  capacity  due  to  the  larger  radiating  surface.  In 
view  of  this  fact  that  the  rope  core  cable  has  a  greater  carrying 
capacity  due  to  its  increased  radiating  surface  it  could  advan- 
tageously be  adopted  for  all  cables,  direct-current  or  alternating- 
current,  for  sizes  700,000  cir.  mils  and  above. 

It  is  apparent  from  the  above  that  the  carrying  capacity  of  a 
cable  depends  on  so  many  factors  that  no  table  can  be  given  which 
applies  to  all  conditions,  and  considerable  care  should  be  exer- 
cised in  selecting  the  size  if  it  is  necessary  to  economize.  Tables 
LIV  and  LV  will,  however,  serve  as  a  guide  for  determining  the 
safe  current-carrying  capacity  imder  three  assumed  conditions, 
X,  Y,  and  Z.  Condition  X  is  such  as  to  require  the  maximum-size 
cable  while  condition  Z  is  the  most  favorable  requiring  the  mini- 
mum size. 

The  use  of  these  tables  is  best  illustrated  by  a  couple  of  exam- 
ples: 

Assume  that  it  is  desired  to  find  the  safe  size  of  a  single-con- 
ductor, varnished  cambric,  insulated  cable,  installed  in  duct,  the 
operating  voltage  being  6600  volts  and  the  continuous  current  to 
be  carried  1000  amperes. 

Referring  to  the  first  column  in  Table  LIV  we  must  use  the 
next  higher  current  values  or  1075,  and  it  is  seen  that  the  cable  may 
have  a  size  from  1,250,000  CM.  to  2,000,000  CM.,  depending  on 
the  operating  condition.  Then  going  to  Table  LV  we  find  in  the 
eighth  line  from  the  top  (corresponding  to  our  case)  that  two  con- 
ditions, Y  and  X,  are  given,  the  former  being  limited  to  a  1,000,000 
CM.  cable  and  the  latter  to  a  2,000,000  CM.  By  comparing  the 
results  from  the  two  tables  it  is  apparent  at  once  that  the  Z  con- 
dition is  out  of  the  question  entirely  and  furthermore  that  the  Y 
condition,  corresponding  to  1,500,000  CM.,  also  gives  too  small  a 
value  as  this  condition  was  Umited  to  a  1,000,000  CM.  cable. 
The  size  must,  therefore,  correspond  to  condition  X  or  2,000,000 
CM. 

As  another  example,  assume  that  a  750-volt  varnished  cambric, 
insulated  cable  in  conduit  is  to  carry  175  amperes.  What  size  is 
required? 

Referring  again  to  Table  LIV  we  have  three  different  sizes  to 
choose  from,  4/0,  2/0  and  1/0.    From  Table  LV,  sixth  line  from 


Digitized  by 


Google 


636 


ELECTRICAL  EQUIPMENT 


TABLE  UV 

Cxtrbent-Carbtinq  Capacitt  of  Cables 

(Continuous  Rated  Apparatus) 


Maximum  Ampere 

Capacity 

Condition  Z. 

Condition  Y, 

Condition  X. 

Permissible. 

' 

25 

#10 

*9 

%% 

35 

»S 

«8 

«6 

50 

#6 

#6 

«4 

70 

#6 

#4 

%2 

110 

#4 

«2 

1/0 

130 

#2 

1/0 

2/0 

175 

1/0 

2/0 

4/0 

225 

2/0 

4/0 

300,000 

290 

4/0 

300,000 

400,000 

360 

300,000 

400,000 

500,000 

450 

400,000 

500,000 

600,000 

550 

500,000 

600,000 

760,000 

675 

600,000 

750,000 

1,000,000 

775 

750,000 

1,000,000 

1,250,000 

900 

1,000,000 

1,250,000 

1,600,000 

1075 

1,250,000 

1,500.000 

2,000,000 

Cables  in  Multiple 


1300 

2-  750000 

2-  750000 

2-1OO0000 

1500 

2-  750000  ,, 

2-1000000 

2-1250000 

1750 

2-1000000 

2-1250000 

2-1500000 

2100 

2-1250000 

2-1500000 

2-2000000 

2600 

3-1000000 

3-1250000 

3-1500000 

3100 

3-1250000 

3-1500000 

3-2000000 

4200 

4-1250000 

4-1500000 

4-2000000 

5200 

5-1250000 

5-1500000 

5-2000000 

6200 

6-1250000 

fr-1500000 

6-2000000 

the  top,  we  see  that  this  case  also  involves  all  three  operating 
conditioiis  and  that  the  limit  of  the  Z  condition  is  a  4/0  cable,  so 
that  it  will  be  safe  to  use  a  1/0  cable  for  our  case. 

Suppose,  on  the  other  hand,  that  the  current  to  be  carried  had 
been  675  amperes.  This  would  have  come  within  the  limit  of  the 
Y  condition  and  the  required  size  of  the  cable  would  be  750,000 
CM 


Digitized  by 


Google 


STATION  WIRING 


637 


TABLE  LV 
Classification  of  Conditions  X,  F,  and  Z 


To 

To 

To 

To  4/0 

500.000 

1.000.000 

2.000.000 

IncluBive. 

CM. 

CM. 

CM. 

Inclusive. 

Inclusive. 

Inclusive. 

SiKGLB  Conductor: 

In  free  air: 

Up  to 

750  V. 

z 

Z 

Z 

Z 

V.C.  and  paper 

3,500  V. 

z 

Y 

Y 

Y 

7,500  V. 

Y 

Y 

Y 

Y 

15,000  V. 

Y 

Y 

Y 

Y 

Rubber 

.     750  V. 

Y 

Y 

Y 

Y 

In  ducts: 

760  V. 

Z 

Y 

Y 

X 

V.C.  and  paper 

3,600  V. 
7,500  V. 

Y 
Y 

Y 
Y 

Y 
Y 

X 
X 

15,000  V. 

Y 

Y 

Y 

X 

Rubber 

.     750  V. 

Y 

Y 

X 

X 

Three  Conductor: 

In  free  air: 

760  V. 

Y 

Y 

V.C.  or  paper.. 

3,500  V. 
7,500  V. 

Y 
Y 

Y 
Y 

15,000  V. 

Y 

Y 

Rubber 

..    750  V. 

X 

X 

Inducts: 

'p 

750  V. 

Y 

X 

V.C.  or  paper.. 

3,500  V. 
7,600  V. 

Y 
Y 

X 
X 

15,000  V. 

Y 

X 

Rubber 

.      750  V. 

Y 

X 

Single-conductor  lead-covered  cables  above  600,000  CM., 
25  cycles  and  3/0,60  cycles,  should  only  be  used  after  special  con- 
sideration is  given  to  the  leadnsheath  current,  and  multiplied 
single-conductor  cables  on  60-cycle  circuits  shall  be  suitably 
arranged  to  eliminate  initial  induction  and  thus  balance  the 
reactance  and  apportion  the  current  carried  in  each  conductor. 

For  secondary  instrument  current  wiring,  where  the  watts 
loss  in  the  secondary  leads  must  be  kept  within  certain  limits, 


Digitized  by 


Google 


638 


ELECTRICAL  EQUIPMENT 


so  as  to  deduct  as  little  as  possible  from  the  permissible  instru- 
ment load  on  the  transformer,  it  is  the  recommended  practice  to 
make  runs  up  to  75  feet  of  19/25  multi-conductor  cable,  corre- 
sponding in  conductivity  to  a  No.  12  B.  &  S.  wire.  For  runs  of 
from  75  to  150  feet,  19/22  cable,  corresponding  in  conductivity 
to  No.  10  B.  &  S.  wire,  should  be  used  for  mechanical  reasons 
as  well  as  for  increased  conductivity.  For  potential  and  control 
wiring,  19/25  cable  may  be  used  in  practically  all  instances.  The 
above  distances  refer  to  llOrVolt  circuits  and  for  220  volts  they 
can,  of  course,  be  doubled.  In  general,  the  size  of  control  leads 
must  also  be  determined  from  the  standpoint  of  voltage  drop,  the 

TABLE  LVI 
Size  and  Ampere  Capacity  of  Copper  Tubing 


Maximum  Con- 
tinuous Ampere 
Capacity. 

Outside 

Diameter, 

Inchea- 

Inside 

Diameter, 

Indies. 

150 
300 
500 

i 
a 

.776 
1.084 

p)ermissible  drop  depending  on  the  minimum  voltage  required  for 
the  apparatus  in  question.  This  is  generally  stipulated  by  the 
manufacturers. 

Instrument  transformer  secondaries  should  be  permanently 
grounded.  Where  secondaries  cannot  be  grounded  at  any  point, 
as  for  instance  in  the  case  of  instruments  and  meters  which  have 
secondary  current  and  primary  potential  coila,  the  secondary 
wiring  must  be  insulated  and  installed  to  safely  withstand  primary 
potential.  One  common  ground  bus,  not  less  than  No.  4  B.  &  S., 
should  be  run  across  the  back  of  the  switchboard,  to  which  appa- 
ratus mounted  on  the  LJ^ritchboard  intended  for  grounding  should 
be  connected.  The  switchboard  pipe  framework,  except  when 
insulated,  should  be  connected  to  this  ground  bus,  one  connec- 
tion being  made  for  every  three  pipe  joints  in  series. 

Steel  work  supporting  high-potential  switching  equipment 
should  be  carefully  grounded  at  several  points  so  as  to  prevent 
the  possibility  of  high  voltage  occurring  between  sections  of  the 


Digitized  by 


Google 


STATION  WIRING  639 

steel  work.  No  ground  connection  for  this  service  should  be  of 
less  than  No.  6  B.  &  S.  flexible  cable. 

For  open  high-tension  wiring  utilizing  bare  conductors,  the 
size  depends  on  the  current  to  be  carried  as  well  as  the  heat- 
radiating  conditions.  For  very  large  alternating  currents,  such  as 
in  low-tension  bus-bars  of  large  size,  the  skin-effect  may  be  appre- 
ciable, requiring  a  low  current  density.  As  a  rule,  this  may  vary 
anywhere  from  as  low  as  300  to  400  amperes  per  square  inch  to 
1500  amperes  per  square  inch,  depending  on  the  conditions.  This 
is  dealt  with  more  fully  under  the  section  on  "Bus-bars,"  page  565. 

For  very  high-voltage  work  using  copper  tubing  the  sizes  given 
in  Table  LVI  are  quite  common. 

Corona  Limit  of  Voltage.  Attention  must  also  be  given  to 
the  possibility  of  tjie  formation  of  corona  when  the  size  of  high- 
tension  conductors  is  determined.  Table  LVII  gives  the  highest 
safe  three-phase  voltage  for  any  given  size  of  wire.  The  values 
are  based  on  sea  level  but  may  be  corrected  for  other  altitudes 
by  the  correction  factors  given  in  table  LVIIL 

Economical  Considerations.  In  determining  the  size  of  a 
conductor  the  economical  side  of  the  problem  should  not  be  lost 
sight  of,  although  it  may  not  be  of  such  great  importance  for  the 
station  wiring  as  for  the  distribution  or  transmission  system. 
The  most  economical  area  is  that  for  which  the  annual  outlay 
equals  the  annual  cost  of  the  energy  loss,  and  according  to  this  rule, 
the  cheaper  the  power,  the  less  should  be  the  capital  outlay  for 
the  ccMiductors,  thus  allowing  a  smaller  size  to  be  used  and  a  corre- 
spondingly increased  loss.  In  general  the  cost  of  ducts,  insulators 
and  supports  may  be  considered  as  not  affected  by  the  variation 
in  size,  but  that  the  outlay  is  only  affected  by  the  comparative 
cost  of  the  cable  itself. 

Voltage  Drop.  In  a  continuous-current  circuit,  the  drop  at 
the  terminals  of  a  circuit  with  resistance  R  and  traversed  by 
a  current  /  ampere,  is  /X/2  volts.  Likewise  in  an  alternating- 
current  circuit  the  drop  in  voltage  of  a  circuit  with  an  impe- 
dance Z,  traversed  by  a  current  of  /  effective  amperes,  is/ XZ 
volts. 

The  voltage  drop  in  alternating  current  circuits,  therefore, 
depends  on  both  the  resistance  and  reactance,  but  with  wires  close 
together,  as  in  conduit  work,  the  reactance  will  generally  be  small. 
The  drop  should  be  calculated  for  the  given  power-factor,  load. 


Digitized  by 


Google 


640  ELECTRICAL  EQUIPMENT 

and  corresponding  current,  and  the  following  approximate  fonnula 
may  be  used. 

Voltd  drop  per  wire=/i2  cos  <I>+IX  sin  ^, 

where   /  =  current  per  wire  in  amperes; 
JB= resistance  in  ohms  per  wire; 
X= reactance  in  ohms  per  wire; 
Cos  <^= power-factor  of  load. 

Volts  drop  of  two-phase  circuit = 2  X  (volts  drop  per  wire). 

Volts  drop  of  three-phase  circuit  =  1.73  X  (volts  drop  per 
wire). 

Resistance  as  well^  as  reactance  values  for  single-conductor 
cables  are  given  in  Table  LIX.  The  values  afe  for  2000  feet  of 
wire,  i.e.,  for  each  wire  of  a  circuit  of  that  length,  and  apply  equally 
well  to  bare  or  lead-covered  cables  as  the  insulation  or  lead  cov- 
ering has  practically  no  effect  on  the  self-induction. 

Table  LX  gives  reactance  and  impedance  values  for  one  mile 
three-conductor  cables.  Unlike  the  reactance  values  given  in 
Table  LIX,  which  were  single-phase,  these  values  are  three-phase, 
i.e.,  by  multiplying  them  by  the  current  the  drop  in  the  full-line 
voltage  (not  voltage  to  neutral)  is  obtained  directly.  In  calculat- 
ing the  values  a  2  per  cent  allowance  for  spiral  of  strands  and  a 
2  per  cent  allowance  for  spiral  of  conductors  has  been  made.  All 
the  results  are  based  on  a  cable  one  mile  long  but  can,  of  course,  be 
obtained  for  any  shorter  distance  by  reducing  the  figures  given  in 
direct  proportion.  Similarly,  the  values  correspond  to  a  fre- 
quency of  60  cycles.  For  any  other  frequency,  the  values  given 
must  be  multiplied  by  that  frequency  and  the  result  divided  by  60. 


Digitized  by 


Google 


STATION  WIRING 


641 


TABLE  LVII 

Corona  Limit  of  Voltaqb 

Kiloyolt8  between  Lines  Three-phase  Cables 

SEA  LEVEL 


size  B.  A  8. 

Diameter  in 
Inches. 

Spacing  Fkbt. 

or  Cm. 

8 

10 

12 

14 

16 

20 

0 

00 

000 

0000 

250,000 
300,000 
350,000 

400,000 
450,000 
600,000 

800,000 

0.374 
0.420 
0.470 
0.530 

0.590 
0.620 
0.679 

0.728 
0.770 
0.818 

1.034 

95 
104 
114 
125 

138 

98 
108 
118 
130 

144 
151 
161 

171 
178 
188 

102 
111 
121 
135 

149 
156 
166 

176 
184 
194 

234 

104 
114 
124 
138 

152 
161 
170 

180 
190 
199 

241 

106 
117 
127 
141 

156 
165 
175 

185 
194 
205 

244 

109 
121 
132 
146 

161 
171 
180 

192 
200 
210 

256 

To  find  the  voltafce  at  any  altitude  multiply  the  voltaf^e  found  above  by  the  a 
corresponding  to  the  altitude,  as  given  in  Table  LVIII. 

For  oingle-plufle  or  two-phase  find  the  three-phase  volts  above  and  multiply  by 
1.10. 

TA3LE  LVm 
Altitude  Correction  Factor  S 


Altitude.  Feet. 

a 

Altitude.  Feet. 

i 

0 

1.00 

5,000 

0.82 

500 

0.98 

6,000 

0.79 

1000 

0.96 

7,000 

0.77 

1600 

0.94 

8,000 

0.74 

2000 

0.92 

9,000 

0.71 

2500 

0.91 

10,000 

0.68 

3000 

0.89 

12,000 

0.63 

4000 

0.86 

14,000 

0.58 

Digitized  by 


Google 


642 


ELECTRICAL  EQUIPMENT 


a 


3 


CG 


|a 

MO 

og 

H 

!l 

H 

I 


55 


CO  CO  CO  CO 


CO  CO  COO 


CCNOHiH 

CO  CONN 


00t««0<4| 


«0  0»NiQ 
N^r-tO 

NCM.'IN 


'•'CON^ 
■^CON-H 
CIC9MN 


C0C4OO 


eoNvHO 

NfHOOa 


dodo 


C0IO9N 
O900Q0 


«*eo«^ 


•^0900 
^.^OO 


OOO 


ookooco 

ONCOO 


)ae4aocot*e4Ud^ 


oooooooo 


Odt^COt^NlCrl 

>«cio«e40ie4e4c>i 


oooooooo 


iQO^oicoaociih. 

t«<DOiCiO'««1'CO 


OOOOOOOO 


t*0»0'-i«0  0"^0» 
lO»O"^"*C0C0N^ 


oooooooo 


^-OCONW^^C 

—  ^  N  N  cs  e> 


■^CCCONC 
NNC4CIC 


oooooooo 


h>  10  040 

Nr-I^C 

e<<o«04e 


>o»t^coo 

INt^O»«< 

)Oa»a»ai 


OOOOOOOO 


ooftd>o>cio^' 


oooooooo 


oooooooo 


oooooooo 


oooooooo 


oooooooo 


<«c*i«^«o?oa»«o 


s:sss 


oooooooo 


^u^OtQoa^aoco 
•HOOOflOobr^r* 

'^'hOOOOOO 


oooooooo 


00  04 1^  coo  04  «0 


ooooooo 


oddddddd 


04004QOeOOOOQO 
CO«Oi04h.i-it^04 
04  01 04  CO  CO  <^^  US 


oooooooo 


CO  t^'^  to  CO  ox 
•o-^r^  rococo  04 

Ol  04  04  04  04  04  04 


ooooooo 


04t^r-i«oeoooo 

"*  CO  CO  04  0101.-^ 
04  04  0404040404 


ooooooo 


Ob^OOCOOt^iO 
NN^^-iOO 
0404  0404  0104  04 


ooooooo 


o>oa»io  0400  0 

PH  O  0>  Oi  Oi  OO  00 

04  04»-»i-«r-»^r-» 


ooooooo 


ooooooo 


00  04  OOO  CO  04  >0 
0eO4t«04  9«OC0 
t«t»00»0i0i0 


ooooooo 


o-^oc^^ooic 
« 10 -^  <«< -^  CO  ro 


ooooooo 


ooooooo 


ooooooo 


04  04^r-iOOO 


ooooooo 


ooooooo 


^WO^OiOO*^ 
OOB'^OOCO'-^ 

oooaOxt<>t«o 
ooooooo 


ooooooo 


sssss 

90000 


ooooooo 


500»OC0^»HC0 


ooooo^^ 


lO^cooia 

04  04^^0 
04  04  04  0)04 


iC^«0  049 
04  0404  04*-! 


«OOiOtO 


^0»t*OI^ 

r-HCi-«O0»C 

ic^^eoco 


^004  00  to 
^009g 


S0040010 
OOOOOO 

-^oooo 


r^roo>oo4 
«oor^t*r* 

ooooo 


ooooo 


ti«o-^oor* 

00C0OI004 
OtOtOWMC 

ooooo 
ooooo 


Ot-i-«C0»O 

^•ooooooo 

O«0C0*^O 

ooooo 
ooooo 


040040^ 
kOQ0>-4O)C0 
.-•Ol'^iOO 


5o4iot-5 


o 

S; 

o 


o 


o 
d 


"-^s 


Ȥ 

'So 

S" 
•  8 

«i 

oc© 

III 


o 
d 


o 
d 


o 
d 


|c|l 

J  III 

lllt 

"    1  =  0 

^  £f  I 

g  ^^ 
I  =!-• 

£  iC£» 


Q2 


Digitized  by 


Google 


STATION  WIRING 


643 


TABLE  LX 

Approximate  Reactance  and  Impedance  of  Three  Conductor  Cables 

PER  Mile 
60  Cycles 


Thickness  of  Insulation. 

Size. 

2/32  by  2/32  In. 

3/32  by  3/32  In. 

4/32  by  4/32  In. 

A 

B 

A 

B 

A 

B 

6 
4 
2 

1 

1/0 
2/0 
3/0 
4/0 

250,000 
300,000 
350,000 

400,000 
450,000 
500,000 

0.307 
0.288 
0.272 
0.264 

0.260 
0.253 
0.247 
0.243 

0.239 
0.236 
0.233 

0.231 
0.229 
0.228 

3.843 
2.423 
1.546 
1.232 

0.988 
0.798 
0.648 
0.519 

0.470 
0.410 
0.370 

0.342 
0.320 
0.304 

0.345 
0.323 
0.302 
0.292 

0.282 
0.276 
0.268 
0.263 

0.257 
0.252 
0.248 

0.246 
0.243 
0.241 

3.845 
2.427 
1.552 
1.238 

0.993 
0.806 
0.656 
0.544 

0.478 
0.421 
0.380 

0.352 
0.330 
0.314 

0  .^79  •  ? 
0  351 

0  :.<2s 
0.:ii5 

'  0.304 
0.297 
0.287 
0.279 

0.273 
0.267 
0.262 

0.259 
0.256 
0.254 

h    n/r 

ja  f .  o-K> 

*  2  4..'. 

1  oo7 

1  2V1 

1.000 
0.815 
0.665 
0.553 

0.488 
0.430 
0.390 

0.361 
0.340 
0.325 

Thickness  op  Insulation. 

Size. 

6/32  by  5/32  In. 

13/64  by  13/64 

8/32  by  8/32  In. 

A 

B 

A 

B 

A 

B 

6 
4 
2 

1 

1/0 
2/0 
3/0 
4/0 

250,000 
300,000 
350,000 

400,000 
450,000 
500,000 

0.407 
0.376 
0.351 
0.337 

0.325 
0.315 
0.304 
0.295 

0.288 
0.281 
0.276 

0.272 
0.268 
0.266 

3.852 
2.435 
1.562 
1.250 

1.006 
0.822 
0.673 
0.561 

0.496 
0.438 
0.396 

0.371 
0.349 
0.334 

0.443 
0.410 
0.381 
0.365 

0.352 
0.340 
0.328 
0.317 

0.310 
0.301 
0.296 

0.291 
0.287 
0.283 

3.855 
2.440 
1.570 
1.258 

1.013 
0.830 
0.685 
0.572 

0.510 
0.452 
0.413 

0.384 
0.364 
0.348 

0.473 
0.439 
0.407 
0.390 

0.375 
0.360 
0.350 
0.338 

0.330 
0.320 
0.313 

0.308 
0.301 
0.298 

3.860 
2.446 
1.576 
1.265 

1.023 
0.840 
0.695 
0.585 

0.522 
0.465 
0.426 

0.398 
0.375 
0.360 

A — The  three-phase  reactance  of  a  cable  1  mile  long. 

B — The  three-phase  impedance  of  a  cable  1  mile  long. 

Note. — Of  the  two  figures  given  for  the  insulation — for  example  6/32  by  5/32 — 
one  is  the  insulation  thickness  around  each  conductor  and  the  other  the  thickness  of 
the  insulation  belt  around  the  three  conductors.  The  former  only  is  of  importance 
as  far  as  the  reactance  value  is  concerned  as  it  determines  the  distance  between  the 
conductors. 


Digitized  by 


Google 


CHAPTER  X 

ECONOMICAL  ASPECTS 

PRELIMINARY  CONSIDERATIONS 

Like  every  other  commercial  imdertaking,  the  promotion  of 
a  hyro-electric  development  involves  a  very  careful  preliminary 
investigation,  as  upon  this  will  largely  be  based  the  success  of 
obtaining  financial  support  for  the  enterprise.  Such  investiga- 
tions should  be  considered  from  the  engineering  as  well  as  the 
commercial  side,  and  the  man  to  whom  this  responsible  task  is 
entrusted  should  have  a  sound  and  conservative  judgment  in 
analyzing  such  propositions. 

This  appUes  to  small  developments  as  weU  as  large  ones,  and 
possibly  more  so  to  the  former  because  an  error  which  would  be  of 
minor  importance  in  a  large  plant  may  involve  serious  financial 
consequences  in  a  small  one. 

No  two  streams  offer  quite  the  same  problem  of  power  devel- 
opment, and  a  multitude  of  conditions  must,  therefore,  in  every 
case  be  investigated.  These  involve  a  complete  and  most  effi- 
cient study  of  the  watershed,  rainfall  and  hydrographic  data  for 
determining  the  available  stream-flow  and  the  storage  possibilities. 
Estimates  of  the  probable  market  for  the  power  and  the  planning 
of  the  development  as  to  type  and  size,  so  that  the  total  annual 
cost,  including  fixed  charges,  to  deliver  the  necessary  power,  will 
not  exceed  the  amoimt  the  available  customers  can  afford  to  pay, 
the  rates  generally  being  governed  by  the  cost  of  competing 
power  generated  from  fuel. 

The  location  of  the  development  should  be  such  that  it  will 
insure  the  most  economical  results.  Usually  this  is  when  the 
maximum  head  is  utiUzed,  but  considerations  must  also  be  given 
to  the  land  which  may  be  overflowed  by  so  increasing  the  head. 
A  study  of  the  watershed  may,  furthermore,  show  that  several 
developments  of  a  smaller  size  will  give  better  economy  than  one 
large  plant,  and  that  in  this  manner  the  entire  system  may  be 
served  in  such  a  way  that  the  power  from  the  new  developments 

644 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  645 

will  form  a  more  economical  addition  to  that  which  may  ahready 
be  supplied  by  other  plants;  in  other  words,  that  the  load  factor 
will  be  such  as  to  improve  the  load  factor  of  the  other  plants  and 
of  the  system  in  general. 

As  a  rule,  it  does  seldom  pay  to  develop  a  stream  for  the  max- 
imum stream-flow,  and  the  question  always  arises  as  to  how  much 
above  the  minimum  stream-flow  the  plant  should  be  built  out  for. 
This  also  involves  the  problem  of  providing  for  water  storage,  if 
such  is  feasible,  or  for  auxiliary  steam  power. 

The  cost  estimates  should  be  made  with  the  greatest  care  to 
leave  undone  no  amount  of  work  or  experiment  which  will  serve 
to  make  certain  the  ground  upon  which  the  estimates  are  made. 
After  having  estimated  liberally  for  all  known  requirements,  it  is 
well  to  provide  additionally  a  substantial  sum  of  money  and  so 
arrange  the  finances  that,  if,  contrary  to  expectations,  the  esti- 
mates should  be  exceeded,  sufficient  funds  remain  in  the  treasury 
for  completing  the  development,  as  nothing  is  so  discouraging, 
and  in  many  cases  so  disastrous,  as  a  reorganization  of  the  under- 
taking at  its  very  beginning. 

Every  feature  of  the  proposition  should,  of  course,  be  investi- 
gated from  the  legal  point  of  view.  This  involves  the  real  estate 
flowage  rights,  rights  of  way,  rights  of  occupying  public  high- 
ways, etc.  Such  matters  must  be  carefully  attended  to  from  the 
beginning. 

A  very  complete  general  guide  for  the  compilation  of  wat^ 
power  reports  and  field  data  has  been  prepared  by  Mr.  J.  T.  John- 
ston, Hydraulic  Engineer  of  the  Water  Power  Department  of 
the  Dominion  6f  Canada,  and  is  contained  in  its  1914  Annual 
Report.  This  guide  is  of  such  completeness  and  usefulness  that 
it  is  reprinted  in  the  following  in  full. 

GENERAL  GUIDE  FOR  THE   COMPILATION   OF  WATER  POWER 
REPORTS  AND  THE  SECURING  OF  FIELD  DATA^ 

The  increasing  number  of  inspections  and  field  investigations 
on  the  part  of  the  field  engineers  of  the  Dominion  Water  Power 
Branch,  has  rendered  desirable  the  preparation  of  a  uniform  guide 
upon  which  may  be  based  the  various  reports  forwarded  to  head 

*  From  the  1914  Annual  Report,  Dominion  Water  Power  Branch  Depart- 
ment of  Interior,  Canada. 


Digitized  by 


Google 


646  ECONOMICAL  ASPECTS 

ofiBce,  in  order  that,  so  far  as  possible,  their  form  may  be  stand- 
ardized. 

It  is  also  considered  that  a  guide  of  this  description  can  be 
used  to  advantage  by  the  engineer  when  making  his  field  investi- 
gations into  the  projects  under  examination.  A  careful  study  in 
the  field  outlined  herein,  will,  as  a  rule,  prevent  the  overlooking 
of  important  data  which  should  be  secured  on  the  ground. 

The  guide  is,  therefore,  submitted  for  a  dual  purpose;  first, 
for  use  as  a  framework  for  the  standardization  of  the  test  of  power 
reports  submitted  by  field  engineers,  and  second,  for  use  by 
engineers  while  in  the  field  as  a  general  memorandum  of  the 
various  features  calling  for  attention  and  field  study. 

Field  investigations  vary  in  character,  the  majority  dealing 
with  the  following  conditions:  (1)  Applications  for  water-power 
privil^es,  such  applications  being  unaccompanied  by  detailed 
data  as  to  the  site  of  Ptream.  (2)  Applications  for  water-power 
privileges  accompanied  by  fairly  well-developed  plans,  setting  out 
the  general  scheme  of  development.  (3)  First-hand  investiga- 
tion of  entirely  new  sites  or  series  of  sites,  for  the  purpose  of  study- 
ing power,  storage  and  conservation  features. 

In  preparing  the  following  instructions,  the  above  has  been 
kept  in  view,  and  the  outline  hereunder  is  intended  to  serve  as  a 
general  guide,  only  such  portions  being  utilized  as  are  directly 
applicable  to  the  class  of  report  under  preparation.  It  is  not 
intended  that  these  instructionts  should  limit  a  repnirt  solely  to  the 
ground  covered  herein ;  much  must  be  left  to  the  discretion  of  the 
engineer  in  charge  of  the  investigations.  The  points  briefly  dealt 
with  represent,  however,  the  general  important  features  which 
require  investigation  and  discussion,  in  order  that  the  ground  may 
be  completely  covered. 

Summary  op  Principal  Divisions 

A  brief  summary  of  the  sections  and  subheadings  follows: 
Further  details  of  the  ground  to  be  covered  under  each  section 
are  given  later. 

I.  Sources  of  data  used  in  report. 

(1)  Why  investigated  and  scope  of  investigation. 

(2)  Personal  examination — route  followed  and  time  con- 

sumed. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  647 

(3)  Run-oflf  records  from  departmental  stream  measure- 

ment offices. 

(4)  Maps. 

(5)  Existing  reports. 

(6)  Miscellaneous. 

II.  Summary  of  report. 

III.  General  introductory. 

Description,  including  location  as  to  province,  river, 
cities,  township,  range  and  section. 

IV.  Water  Supply. 

(1)  General  description  of  drainage  area. 

(2)  Actual  records  if  available,  showing  maximum,  min- 

imum, and  mean  discharge  for  each  month,  also 
absolute  minimum  for  year.  Measurements  on 
ground  if  foregoing  are  not  available. 

(3)  Rainfall,  temperature,  evaporation. 

(4)  Storage,  already  developed  and  effect  of  same. 

(5)  Storage  possibilities — 

(a)  Location  of  reservoir  site  or  sites. 
(6)  Height  of  dam  and  class  of  dam  suitable, 
(c)  Capacity  of  reservoirs  and  extent  of  adjacent 
drainage  basin. 

(6)  Prior  water  rights  above  and  below  power  site — 

water  supply,  irrigation  or  power. 

(7)  Ice  conditions,  during  winter  months  and  in  spring 

flood  (frazil,  anchor  and  floating  ice), 
(a)  Under  present  conditions  on  river. 
(6)  After  construction  of  plant. 

(c)  Without  storage. 

(d)  With  storage. 

V.  Description  of  existing  Pcmer  Development  on  the  River. 

VI.  Detailed  Work  at  Site  irwestigated. 

(1)  Scope  of  the  inspection  at  the  site. 

(2)  Accessibility  of  site  and  transportation  problems. 

(3)  Detailed  information  and  plans  of  site — 

(a)  Contour  plan  of  site. 
(6)  Cross  section, 
(c)  Profiles. 


Digitized  by 


Google 


648  ECONOMICAL  ASPECTS 

(4)  Foundation  conditions. 

(5)  Flooding  and  pondage. 

(6)  Existing  interests. 

VII.  Possible  Power  Developed. 

(1)  Horse-power  at  wheel  shaft  without  storage — 

(a)  At  minimum  flow. 

(6)  For  the  nine  high  months. 

(2)  Horse-power  at  wheel  shaft  with  storage.    Dis- 

cuss utilization  of  local  pondage  at  site  for  peak 


VIII.  Estimates. 

Cost  of  power  developed — capital  and  annual. 
Cost  of  storage — capital  and  annual. 

IX.  Market  for  Power. 

(1)  Present. 

(2)  Future. 

(3)  Length  of  transmission  lines,  etc. 

X.  Suggestions  and  Recommendations. 

XI.  Appendices,  r^, 

(1)  Plans  pertinent  to  the  actual  sites  investigated. 

(2)  Photographs. 

(3)  Run-off  records. 

(4)  Gauge  records. 

(5)  Reports. 

(6)  Maps  and  plans  of  existing  power  plants  and  struc- 

tures, etc. 

Details  as  to  the  Foregoing  Sections 
I,  Sources  of  Data  used  in  Report 

This  section  should  set  out  the  basis  and  authority  on  which  the 
investigation  was  instituted,  outline  the  scope  of  the  same,  and  the 
organization  by  means  of  which  the  field  data  were  obtained. 

It  is  also  intended  to  summarize  the  sources  of  information 
upon  which  the  subject  matter  of  the  report  is  founded,  and  to 
set  out  in  full  the  degree  of  thoroughness  with  which  the  inves- 
tigation has  been  carried  on. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  649 

II.  Summary  of  Report 

All  the  essential  features  of  the  report  should  be  brought  to- 
gether here,  in  a  brief  statement  forming  a  concise  smnmary  of 
the  whole,  tabulation  of  results  being  made  where  possible. 

III.  General  Introductory 

This  section  should  cover  the  general  features  of  the  situation 
being  investigated.  This  involves  a  general  description  of  the 
river  and  its  characteristics,  and  of  the  basin  as  a  whole,  touching 
on  drainage  area,  somrce,  direction,  drop,  falls,  rapids,  banks, 
river  bed,  tributaries,  lakes,  muskegs,  swamps,  forest,  cultivation 
along  banks,  settlements,  glaciers,  general  topographical  and 
geological  features,  etc.,  and  giving  the  definite  location  of  the 
site  under  study. 

IV.  Water  Supply 

(1)  General  Description   of  Drainage   Area. — Under   general 
description  of  the  drainage  area  those  features  should  be  deaH' 
with  which  are  of  direct  importance  to  the  question  of  the  water 
supply,  such  as  probabiUty  of  sudden  floods,  influence  of  the  sea- 
sons, etc. 

(2)  Run-off  Records. — If  the  site  inspected  is  situated  on  one 
of  the  rivers  covered  by  any  of  the  systematic  stream  measurement 
work  carried  on  by  the  department,  the  existing  records  should 
be  utiUzed  as  a  basis  upon  which  the  run-off  may  be  <iiscussed. 
A  smnmary  of  the  essential  features  of  the  discharge  co>»Ting 
high,  low  and  mean  flow,  etc.,  should  be  inserted,  while  tlie  reitorda 
in  their  complete  form  should  be  attached  as  appendices  in  Section 
XI  of  the  report.  Where  no  records  have  been  taken  on  the  river, 
estimates  or  measurements  of  the  flow  at  the  time  of  the  inspection 
should  be  made,  either  by  meter  or  by  whatever  method  of  stream 
measin-ement  is  most  applicable  or  convenient.  From  this,  in 
conjunction  with  high-water  marks  in  evidence  and  from  the  tes- 
timony of  local  inhabitants  as  to  extreme  low-  and  high-water 
conditions,  and  from  a  study  of  the  run-off  conditions  of  streams 
in  the  vicinity,  as  careful  an  estimate  as  is  possible  should  be  made 
of  the  extreme  low-  and  high-water  conditions  on  the  river,  also 
of  the  average  low  and  high  flows  which  may  be  exp)ected.  With 
these  data,  the  months  and  seasons  in  which  the  above  conditions 
are  usually  in  evidence,  must  be  given. 


Digitized  by 


Google 


650  ECONOMICAL  ASPECTS 

(3)  Rainfally  Temperature  and  Evaporation. — The  maximuin, 
minimum  and  mean  annual  rainfall  as  recorded  at  the  nearest 
stations  maintained  by  the  Meteorological  Service  should  be 
discussed,  being  utilized  in  estimating  the  run-off  if  stream-flow 
records  are  not  to  hand.  Temperatiu^  and  evaporation  records,  if 
available,  should  also  be  fully  considered. 

(4)  Storage  Already  Developed. — If  storage  is  already  in  opera- 
tion in  the  river  basin  above  the  site,  a  full  discussion  of  the  same 
is  required  under  the  heads  of  location;  owners  and  operators; 
date  of  installation;  area  and  volmne  of  reservoir  and  of  tributary 
drainage  basin;  description  and  condition  of  dam  and  structures; 
effect  on  natmral  run-off  conditions,  actual  experience  since  being 
placed  in  operation  covering  date,  time  of  filling  and  emptying 
reservoir;  gauge  records  if  available  (to  be  attached  in  full  in 
appendix);  method  of  control;  photographs,  comments,  etc.,  etc. 
Copies  of  plans  of  structures  are  to  be  secmred  if  possible. 

(5)  Storage  Possibilities. — ^The  question  of  storage  possibilities 
and  locations  on  the  upper  waters  should  be  covered  as  thoroughly 
as  the  conditions  of  the  inspection,  and  the  detailed  instructions 
issued  therewith,  may  be  required.  If  a  visit  is  made  to  any  lakes 
m  the  upper  basins,  the  general  elevation  of  the  banks  of  the  same 
relative  to  the  water  service  should  be  recorded,  with  notes  as  to 
what  flooding  would  result  if  the  lakes  were  raised  to  various 
definite  limits.  When  the  reservoir  is  in  a  surveyed  district  the 
approximate  land  flooded  should  be  given  in  sections  and  quarter 

A I  •  i'.  outlets  all  the  conditions  affecting  the  construction  of  a 
dam,  and  tlie  type  of  structmre  advisable,  are  required.  This  will 
include  foundation  conditions;  height  and  character  of  banks;  a 
section  across  the  river  at  the  point  selected  for  the  dam  carried 
sufficiently  far  up  the  banks  to  cover  all  possible  limits  to  which  it 
may  be  advisable  to  hold  the  lake  surface. 

A  profile  should  be  secured  of  the  water  surface  from  the  lake 
outlet  to  the  dam  site.  Should  there  be  a  possibility  of  securing 
storage  by  means  of  dredging  or  otherwise  clearing  the  outlet,  a 
profile  should  be  obtained  of  the  water  surface,  and,  if  possible, 
of  the  river  bed  from  the  lake  to  a  sufficient  distance  below  the 
dam  site;  any  other  field  information  necessary  to  determine  what 
is  involved  in  the  construction  of  a  dam  and  in  the  operation  of  a 
storage  reservoir  is  also  required. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  651 

When  circumstances  render  it  inadvisable  to  visit  the  upper 
waters  of  the  basin  for  the  purpose  of  personal  inspection,  a  review 
of  the  storage  situation,  as  far  as  it  can  be  gathered  from  existing 
maps  and  from  local  information,  should  be  included. 

The  surface  area  and  capacity  of  all  storage  reservoirs  consid- 
eredy  t(^ether  with  the  area  of  the  drainage  basins  adjacent  to  the 
same  and  their  sufficiency  to  fill  the  reservoirs,  should  be  fully 
covered;  the  beneficial  effect  of  such  storage  on  the  flow  of  the 
river  should  be  discussed. 

(6)  Prior  Water  Rights. — Any  existing  or  projected  schemes  of 
municipal  water  supply,  irrigation  or  water  power,  which  have 
diverted  or  may  in  the  future  permanently  divert  a  portion  of 
that  river  flow,  thus  reducing  the  water  available  at  the  site, 
should  be  investigated  and  reported  on. 

(7)  Ice  Conditions, — The  general  conditions  in  winter  along  the 
river  as  a  whole,  covering  time  of  freeze  up,  conditions  in  mid- 
winter, and  time  and  manner  of  break  up  in  the  spring,  should  be 
secured  from  whatever  local  sources  may  be  available,  or,  if  pos- 
sible, from  personal  observation.  The  question  of  anchor  and 
frazil  ice  under  present  conditions  should  be  considered  carefully, 
also  that  of  ice  jams  in  the  spring,  both  above  and  below  the  site. 
The  possible  formation  of  ice  jams  below  the  site  and  the  conse- 
quent effect  on  the  tail-water  and  floor  elevation  of  the  power- 
house, should  be  particularly  noted. 

The  frazil  and  anchor-ice  conditions,  to  be  anticipated  at  the 
site  after  the  construction  of  the  plant,  should  be  discussed.  In 
this  connection  a  careful  study  covering  the  winter  conditions  and 
troubles  experienced  in  the  operation  of  any  existing  plants  on  the 
river,  together  with  methods  of  remedying  the  same,  is  advisable. 

The  probable  effects  on  ice  conditions  of  the  development  of 
storage  for  the  purpose  of  increasing  the  wint^  flow,  should  also 
be  covered. 

V.  Description  of  Existing  Power  Plants 

Existing  power  developments  along  the  river  should  be  dealt 
with  under  the  following  general  heads:  Ownership  of  plant  and 
when  constructed;  description  of  layout  and  structures  (dam, 
intake,  penstocks,  timnels,  canal,  forebay,  power-house,  founda- 
tions, transmission,  substations,  etc.),  and  present  conditions  of 
the  same;   head  at  different  seasons;   installation  (electrical  and 


Digitized  by 


Google 


652  ECONOMICAL  ASPECTS 

hydraulic  machinery  in  detail) ;  auxiliary  power,  power-ioad  and 
power-factor,  daily  load  curves  if  possible,  use  of  power,  market 
for  power,  present  and  future;  special  features,  etc.,  comments 
and  photographs.  Plans  of  plant  to  be  secured  if  possible  and 
attached  to  appendix. 

VI.  Detailed  Work  at  Site  Investigated 

(1)  Scope  of  the  Inspection  at  the  Site, — If  a  definite  and  well- 
defined  project  be  investigated,  the  engineer  making  the  inspec- 
tion should  study  the  general  scheme  carefully  in  the  light  of  his 
personal  inspection  of  the  ground,  and  should  record  his  opinion 
as  to  the  engineering  and  economic  feasibihty  of  the  same,  pointing 
out  whatever  weaknesses  may  be  apparent,  and  recommending 
whatever  changes  in  design,  layout  or  scheme  of  development  he 
may  consider  advisable. 

When  no  definite  scheme  of  development  has  been  proposed, 
the  inspecting  engineer  is  expected  to  outline  the  most  feasible 
scheme  which  his  study  on  the  ground  may  suggest,  setting  out  the 
head  available  and  method  of  securing  the  same.  He  should  also 
gather  all  information  and  field  data  which  may  be  essential  to  its 
proper  consideration  and  to  getting  out  the  estimates.  A  layout 
of  his  scheme,  together  with  all  pertinent  data,  should  be  plotted 
on  the  contour  plan  of  the  site. 

Arrangements  should  be  made  on  the  ground  for  the  installa- 
tion and  continued  reading  of  gauges  at  all  points  where  the  record 
of  the  same  is  advisable. 

Numerous  photographs  illustrating  the  site  are  required. 

(2)  Accessibility  of  Site. — Secure  all  data  with  reference  to 
accessibihty  of  the  site.  This  includes  the  distance  to  the  nearest 
railway  line;  the  ease  or  difficulty  of  building  a  spur  line  to  the  site 
should  the  size  of  the  development  warrant  it;  the  condition  of 
any  roads  in  the  vicinity  and  their  suitability  for  heavy  transport; 
the  length  of  new  road  that  may  be  required;  the  use  which  can 
be  made  of  water  transportation  as  a  means  of  access.  In  brief, 
the  best  means  of  connecting  the  site  with  existing  lines  of  trafSc, 
should  be  covered. 

(3)  Detailed  Information  and  Plans  at  Site. — (a)  Contour  Plan, 
— Enough  rough  instrument  work  must  be  done  to  permit  of  plot- 
ting a  fairly  accurate  contoiu"  plan  of  the  whole  vicinity  covered  by 
the  proposed  layout.     These  contours  should  extend  above  the 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  653 

highest  elevation  to  which  there  is  any  possibility  of  raising  the 
head-waters  of  the  proposed  plant.  Sufficient  notes  should  be 
taken  to  plot  on  the  said  plan,  with  the  elevations,  any  rock  out- 
crops which  may  be  in  evidence.  Should  the  rock  outcrop  along 
both  banks  of  the  river,  the  continuous  line  of  demarcation  be- 
tween the  rock  and  the  overlying  material  should  be  plotted,  vnlh 
the  devaliana,  along  both  shore  lines.  The  plan  should  also  indi- 
cate all  other  classes  of  material,  such  as  clay,  gravel,  sand,  loam, 
etc.,  which  may  be  in  evidence  together  with  notes  as  to  whether 
the  site  is  wooded,  cleared  or  cultivated,  etc. 

Water  levels  (together  with  date  of  taking,  and  river-flow,  if 
possible)  should  be  recorded  and  plotted  on  this  plan  at  all  im- 
portant points,  such  as  the  brink  and  foot  of  falls  and  rapids, 
marking  the  limits  of  the  still  water  above  and  below.  All  eddies 
and  back  waters  should  be  marked  and  the  elevation  and  date 
recorded.  The  general  line  of  the  brink  and  foot  of  any  falls 
which  will  be  involved  in  a  proposed  scheme  of  development  should 
be  secured  and  tied  in  to  the  plan.  The  high-  and  low-water  levels 
to  be  expected  in  the  tail-water  of  the  projected  power  station 
are  of  particular  importance.  Maximum  high-water  marks  along 
the  shore  should  be  carefully  noted. 

All  natural  features  of  which  advantage  might  be  taken  in 
la3dng  out  a  power-plant  should  be  fully  shown  on  the  plans  and 
discussed  in  the  report. 

(6)  Cross-section, — ^A  cross-section  of  the  river  bed  and  both 
banks  along  the  line  of  the  proposed  dam,  and  sections  of  any 
alternative  sites  which  may  present  themselves  to  the  engineer  on 
the  ground,  should  be  secured  and  plotted.  Sections  when  plotted 
should  indicate  the  character  of  the  ground  surface  and  river  bed 
and  of  foundation  conditions,  either  in  evidence  or  assumed, 
throughout. 

(c)  Profiles. — A  profile  of  the  river  surface  from  the  upstream 
limit  of  the  new  pond  created  by  the  plant  is  desirable,  but  is  not 
essential  should  the  circumstances  of  the  inspection  render  the 
securing  of  the  same  inadvisable.  In  all  cases,  however,  a  profile 
of  the  river  surface  and,  if  possible,  of  the  river  bed,  from  a  point 
up  stream  from  the  dam,  to  below  the  tail-race  of  the  px)wer-plant 
is  required.  A  profile  section  through  the  dam,  intake,  head- 
race (or  pipeline,  as  the  case  may  be),  power-plant,  and  tail-race, 
showing  such  governing  elevations  as,  head-water,  crest  of  dam, 


Digitized  by 


Google 


654  ECONOMICAL  ASPECTS 

floor  of  generator  room,  tail-water,  etc.,  should  also  be  obtamed 
in  the  best  manner  which  circumstances  may  dictate. 
Profiles  of  any  pipe  or  canal  lines  are  also  required. 

(4)  Foundation  Conditions. — Full  note  should  be  made  of  the 
natiu-al  conditions  of  the  ground  and  river  bed  at  the  proposed 
dam  and  power-house  site.  If  there  is  rock  in  sight  a  full  state- 
ment of  its  character,  weathering  qualities,  etc.,  is  required.  If 
no  rock  is  in  evidence  as  careful  an  investigation  of  the  existing 
conditions  as  circumstances  permit  is  required. 

(5)  Flooding  and  Pondage, — The  direct  flooding  which  will  be 
caused  by  the  construction  of  the  proposed  or  any  feasible  plant 
at  the  site  should  be  determined  approximately  either  by  inspec- 
tion or  if  necessary  by  rough  instrument  work.  If  the  land  has 
been  surveyed  the  flooded  portion  can  be  listed  by  sections  and 
quarter  sections. 

The  utiUzation  of  this  local  pondage  in  connection  with  peak 
loads  at  the  project  plant  should  receive  general  consideration. 

(6)  Existing  Interests, — All  existing  interests,  such  as  bridges, 
trails,  roads,  railways,  buildings,  etc.,  that  may  be  affected  by  the 
construction  of  the  plant  and  by  the  consequent  flooding,  should 
be  fully  reported  on.  The  question  of  the  logging  and  fishing 
interests  on  the  river  should  be  discussed  in  considerable  detail. 

VII.  Possibk  Power  Developed 

The  question  of  power  possible  of  development  should  be  dis- 
cussed from  the  standpoints  of,  first,  no  storage  available,  and 
second,  storage  available.  Under  the  first  head  the  power  avail- 
able at  minimum  flow  and  the  power  which  might  be  developed 
during  the  eight  or  nine  months  not  included  in  the  extreme  low- 
water  season  should  be  covered. 

Under  both  headings  the  beneficial  utilization  of  the  local 
pondage  for  peak  loads  and  the  consequent  increased  power  out- 
put should  be  dealt  with. 

VIII.  Estimates 

Approximate  estimates  of  the  capital  and  annual  operating 
costs  of  the  proposed  scheme  of  development  and  the  basis  on 
which  these  are  made  should  accompany  the  report,  together  with 
similar  estimates  of  the  cost  of  any  proposed  storage  reservoirs. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  655 

IX.  Market  for  Power 

This  will  involve  as  thorough  an  investigation  as  the  circum- 
stances warrant  of  the  present  and  future  power  market  in  the 
surrounding  municipalities  and  district.  Possibilities  for  the  local 
use  of  power  at  the  site  and  in  the  immediate  vicinity  are  also  to 
be  covered.  With  the  question  of  power  market,  the  question 
of  distance  of  transmission  necessary  to  reach  the  same  requires 
careful  consideration. 

X.  Suggestions  and  Recommendations 

Suggestions,  comments  or  recommendations  with  reference  to 
the  foregoing  and  the  writer's  opinion  as  to  the  questions  at  issue 
should  be  set  out  in  full.  The  location  of  suitable  metering  sta- 
tions for  the  continuous  record  of  the  river-flow  at  vital  points 
should  be  covered  in  these  recommendations.  The  question  of 
sources  of  power  other  than  water,  in  the  vicinity  and  their  pos- 
sible more  economic  development  is,  at  times,  most  important. 
All  recommendations  should  be  set  out  definitely  and  concisely. 

XI.  Appendices 

(1)  Plans. — (a)  A  general  plan  (a  section  of  pubHshed  map  is 
desirable)  showing  the  location  of  the  power  and  storage  sites 
with  reference  to  centers  of  population.  (6)  A  general  plan  (a 
section  of  pubhshed  map)  showing  the  whole  drainage  basin  above 
t!ie  power  site,  together  with  storage  reservoirs,  (c)  Contour 
plans  of  the  sites  of  power  plants  and'  storage  dams,  (d)  Cross- 
sections  along  dam  sites,  (e)  Profiles  of  reach  of  river  affected  and 
of  pipe  and  canal  fines.  (/)  Any  other  plans  warranted  by  the 
nature  of  the  investigation. 

All  plans,  sections,  and  profiles,  etc.,  should  be  suitably  num- 
bered, and  should  be  referred  to  in  the  text  by  these  numbers 
whenever  necessary.  A  complete  list  of  the  above  plans,  giving 
numbers  and  description,  should  be  included  in  the  table  of 
contents  of  the  report. 

(2)  Photographs, — A  set  of  all  the  views  taken  to  illustrate  tlie 
different  features  of  the  report  should  be  mounted  and  included. 
Where  these  views  deal  with  power-plant  and  storage-dam  layouts, 
they  should  be  accompanied  by  a  sketch  plan  showing  the  point 
from  which  each  is  taken  and  the  direction  the  camera  faced.  The 
fihns  should  be  numbered,  dated  and  titled,  in  order  that  all  prints 


Digitized  by 


Google 


656  ECONOMICAL  ASPECTS 

may  be  immediately  recognized.  A  complete  list  of  the  photo- 
graphs, giving  numbers,  date  and  subject  should  be  included  in 
the  table  of  contents  of  the  report. 

(3)  Run  of  Records, — All  tabulated  records  and  plotted  curves 
which  may  have  been  seemed. 

(4)  Gauge  Records. — Copies  of  all  gauge  records  which  are  of 
interest  in  connection  with  the  power  or  storage  features  of  the 
report. 

(5)  Reports. — Copies  of  any  existing  reports  which  may  have 
been  made  with  reference  to  power  development  on  the  river. 

(6)  Maps. — ^Any  maps  which  may  usefully  illustrate  the  report, 
and  any  plans  which  may  have  been  obtained  covering  existing 
power-plants,  storage  works,  bridges,  etc.,  etc. 

Investigation  and  Inspection  of  a  Series  of  Sites 

Frequently  the  investigation  of  a  river  involves  the  considera- 
tion and  detailed  inspection  of  a  series  of  power  sites.  In  such 
cases,  the  report  covering  the  work  should  follow  the  foregoing 
guide,  with  the  following  sUght  changes. 

It  will  be  noted  in  the  foregoing,  that  Sections  I  to  V  can  be 
applied  as  they  stand,  to  the  compilation  of  a  report  on  a  series 
of  sites.  Sections  VI  to  VIII  are  directly  applicable  to  each  indi- 
vidual site;  Section  IX  is  applicable  to  individual  sites  or  to 
groups  as  conditions  may  warrant,  and  Sections  X  and  XI  are 
applicable  as  they  stand  to  the  ending  up  of  the  report.  In  pre- 
paring a  report  on  a  series  of  sites,  the  only  alteration  advised 
in  the  foregoing  guide  is  that  under  Section  VI,  each  site  is  treated 
as  a  unit  and  completely  covered  according  to  the  outline  in  Sec- 
tions VI  to  IX.  The  new  Sections  VII  and  VIII  will  correspond 
to  X  and  XI  in  the  foregoing  synopsis. 

Following  is  the  outhne  for  a  report  covering  a  series  of  inves- 
tigated sites,  with  the  necessary  alterations: 

I.  Sources  of  Data  Used  in  Report. 

(1)  Why  investigated  and  scope  of  investigation. 

(2)  Personal  examination,  route  followed  and  time  con- 

sumed. 

(3)  Run-ofif  records  from  departmental  stream  measure- 

ment offices. 

(4)  Maps. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  667 

(5)  Existing  reports. 

(6)  Miscellaneous. 

II.  Summary  of  Report, 

Concise  statement  of  results  of  investigations  covering  all 
essential  features  of  the  report.  Tabulation  of  results  as  to  power 
and  storage. 

III.  Generallntrodudory. 

Description,  including  location  as  to  province,  river,  cities, 
township,  range  and  section. 

IV.  Water  Supply. 

(1)  General  description  of  drainage  area. 

(2)  Actual  record  if  available  showing  maximum,  min- 

imum and  mean  discharge  for  each  month,  also 
absolute  minimum  for  year.  Measurements  on 
ground  if  foregoing  are  not  available. 

(3)  Rainfall,  temperature,  evaporation. 

(4)  Storage  already  developed  and  effect  of  same. 

(5)  Storage  possibilities. 

(a)  Location  of  reservoir  sjte  or  sites. 

(b)  Height  of  dam  and  class  of  dam  suitable. 

(c)  Capacity  of  reservoirs  and  extent  of  adjacent 

drainage  basin. 

(6)  Prior  water  rights  above  and  below  power  site; 

water  supply,  irrigation  or  power. 

(7)  Ice  conditions  during  winter  months  and  in  spring 

flood  (frazil,  anchor  and  floating  ice). 

(a)  Under  present  conditions  on  river. 

(b)  After  construction  of  plant. 

(c)  Without  storage. 

(d)  With  storage. 

V.  Description  of  Existing  Power  Developments  on  the  River. 

VI.  Sites  Investigated. 

(a)  Detailed  work  at  each  site  investigated, 

(1)  Scope  of  the  inspection  at  the  site. 

(2)  Accessibility  of  site  and  transportation  prob- 

lems. 

(3)  Detailed  information  and  plans  at  site, — 

(a)  Contour  plan  of  site. 


Digitized  by 


Google 


658  ECONOMICAL  ASPECTS 

(6)  Cross-sections, 
(c)   Profiles. 

(4)  Foundation  conditions. 

(5)  Flooding  and  pondage. 

(6)  Existing  interests. 
(6)  Possible  Power  Developed, 

(1)  Horse-power  at  wheel  shaft  without  storage, — 

(a)  At  minimum  flow. 

(6)  For  the  nine  high  months. 

(2)  Horse-power  at  wheel  shaft  with  storage.    Dis- 

cuss utilization  of  local  pondage  at  site  for 
peak  loads. 

(c)  Estimates. 

Cost  of  power  developed,  capital  and  annual. 
Cost  of  storage,  capital  and  annual. 

(d)  Market  for  Power. 

(1)  Present. 

(2)  Future. 

(3)  Length  of  transmission  lines,  etc. 
{e)  Recapitulation. 

Comprehensive  discussion  of  the  foregoing  data  as 
to  the  individual  sites,  and  a  consideration  of 
the  same  as  a  whole  or  in  groups,  as  local  condi- 
tions may  warrant. 

VH.  Suggestions  and  Recommendations. 

Vni.  Appendices. 

(1)  Plans  pertinent  to  the  actual  sites  investigated. 

(2)  Photographs. 

(3)  Run-oflf  records. 

(4)  Gauge  records. 

(5)  Reports. 

(6)  Maps  and  plans  of  existing  power  plants  and 

structm-es,  etc. 

The  details  of  the  data  to  be  covered  in  each  section  are  in  the 
main  as  previously  outlined  in  connection  with  the  report  on  an 
individual  site.    A  careful  study  of  these  details  is  desirable. 

In  section  VI  each  site  investigated  should  be  completely  cov- 
ered under  the  headings,  a,  6,  and  c,  before  discussion  on  a  second 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  659 

site  is  commenced.  The  market  for  power  under  the  heading  d 
should  be  discussed  with  each  individual  site  or  with  groups  of 
sites  as  general  conditions  may  warrant.  Plans  and  photographs 
should  be  suitably  numbered  in  order  that  they  can  be  referred  to, 
when  necessary,  in  the  text. 

Attached  as  appendices  to  this  Guide  are  reproductions  of  the 
loose-leaf  forms,  R-11  to  R-22,  used  in  the  field  by  the  engineers 
of  the  Water  Power  Branch.  The  great  flexibiUty  of  the  loose-laef 
S}'stem  is  claimed  to  be  of  outstanding  advantage  to  the  rapid  and 
efficient  carrying  on  of  the  survey  work,  more  especially  on  those 
investigations  where  the  results  have  been  plotted  into  final 
shape  in  the  field.  The  loose  leaves  generally  lend  themselves 
most  readily  to  a  simple  filing  system  in  which  the  records  of 
the  survey  are  properly  grouped,  and  are  at  all  time  available  for 
ready  reference. 


Digitized  by 


Google 


660 
R-11. 


ECONOMICAL  ASPECTS 


O 
Diary  op. 


o 


Water  Power  Branch,  Dept.  of  the  Interior, 
Ottawa 


Day  of 19 

Day  of 19 

Day  of 19 

Digitized  by 


Google 


COMPILATION  OF  WATER  POWSR  REPORTS 


661 


o 

R-12                                  Return  to 
WATER  POWER  BRANCH,  DEPT.  OF  THE  I> 

o 

Valuable 
[TERIOR,  OTTAWA 

1 
1 

1 

a 

) 

s 

i 

< 

i 

>- 

■♦3 

gn 

K* 

P 

p 

s 

• 

S 

1 

2 
o 

Digitized  by 


Google 


ECONOMICAL  ASPECTS 


o  o 

R-13  Return  to  Valuable 

WATER  POWER  BRANCH,  DEPT.  OF  THE  INTERIOR,  OTTAWA 


c 

a 

z 

< 
> 

1 1 
1  i 

i 

1 

i 

> 

L 

c 

s 

0 

< 
> 

1 

1  s 

i 

P 

1 

1 

1 

1  ' 

►- 

6 

0 

< 



i 

as 

h 
P 
S 

< 

2 
o 



1' 

ii 

i 

1 

i 

i 

Digitized  by 


Google 


COMPILATION  OF  WATER  POWEE  REPORTS 


663 


o                             o 

R-14                                 Return  to                                   Valuable 
WATER  POWER  BRANCH,  DEPT.  OF  THE  INTERIOR,  OTTAWA 

i 
& 

Page  No 
19. 

la 

§ 

H 

1 

File  No 
strument 

» 

i     : 

+ 

2 
o 

Digitized  by 


Google 


664 


ECONOMICAL  ASPECTS 


O  O 

R-15  Return  to  Valuable 

WATER  POWER  BRANCH,  DEPT.  OF  THE  INTERIOR,  OTTAWA 


9 


I 


9 


g 

z 

o 

< 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  665 

O  O 

R-16  Return  to  Valuable 

WATER  POWER  BRANCH,  DEPT.  OF  THE  INTERIOR,  OTTAWA 

LEVEL  NOTES 

Stream 

Locality 

Party Date 19 


station. 

B.  8. 

Ht.  Inst. 

F.  8. 

Elevation. 

Remark!. 



No. 


.of Sheeta Comp.  by Chk.  by. 


Digitized  by 


Google 


666  ECONOMICAL  ASPECTS 

Form  R-17 — ^Fbont 

o  o 

R-17.  Return  to 

WATER  POWER  BRANCH,  DEPARTMENT  OF  INTERIOR,  OTTAWA 

DESCRIPTION  OP  RIVER  STATION 

^ • {^]" 

near PostOflSce,  Prov.  of 

Established 191....,  by 

Name  of  observer P.  0.  address 

pay  S occupation distance time  of  daily  observation. . . . 

Location  of  station  with  respect  to  towns,  bridges,  highways,  railroads, 
tributaries,  islands,  falls,  dams,  etc 


Description  and  location  of  the  gauge,  also  relative  to  the  measuring 
station.    If  chain  gauge,  give  length  from  end  of  weight  to  the  marker 


Description  of  the  equipment  from  which  measurements  are  made. 


Location  and  description  of  initial  point  for  soundings. 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  667 

Form  R-17— Back 

Channel  above  the  station:  straight  or  curved  for  about feet, 

water  swift,  sluggish,  etc 

Channel  below  the  station:  straight  or  curved  for  about ,  feet 

water  s\?if t,  sluggish,  etc 

Right  bank:  high,  rocky  or  low,  liable  to  overflow,  dean  or  wooded,  etc. 
Left  bank:  high,  rocky  or  low,  liable  to  overflow,  clean  or  wooded,  etc. 

Bed  of  the  stream:   rocky,  gravel,  sandy,  clean  or  vegetation,  shifting 

Number  of  channels  at  low  and  high  water,  approximate  depth  of  water, 
etc 

Note  any  condition  which  may  affect  the  measurement,  etc 


Bench  marks:  Describe  fully,  give  elevation  above  zero  of  the  gauge 
and  above  sea  level  or  other  datum,  if  possible;  make  sketch  bringing  out 
the  principal  features 


Take  sufficient  soundings  to  develop  a  crossHsection  of  stream  bed  and,  by 
use  of  level,  develop  banks  to  above  high-water  mark.  Refer  all  elevations 
to  gauge  datum. 

Make  a  sketch  plan  on  cross-section  paper,  showing  the  relative  location 
of  the  station,  gauge  bench  marks,  tributaries,  towns,  etc. 

o  o 


Digitized  by 


Google 


668 


ECONOMICAL  ASPECTS 


Form  R-18— Fbont 
R-18  O  Return  to 

WATER  POWER  BRANCH 

Dept.  op  Interior,  Ottawa 

Current  Meter  Notes — Ice  Cover 

A.M. 


o 


Valuable 


Date. 


.19. 


P.M. 


Stream. 


Party Locality 

Meter  No Gauge  height,  beg end mean. 

Total  area Mean  velocity Discharge. . 


Odseuvatiokh. 

COlit»UTATIC}?^S. 

II 

Thick- 

NGaH 

Depth 
rFNDER  IcK. 

i 

B 
B 

> 

Vkuocity* 

d 

£ 

-< 

4 

1 

o 

5 

si 
be 

4 

1 

•  •••■• 

No of Sheets,  Comp.  by Chk.  by Make  notes  on  back 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  669 

Form  R-IS—Back 

Weights  used 

Wind 

Method  of  supervision  of  meter  (single  wire  or  cable) 

Stay  wire  used  or  not  used 

Point  of  measurement  with  reference  to  gauge  (i.e.,  distance  above  or  below) 

Length  of  gauge  chain  checked  and  found  to  be ft.  and  corrected  to ft. 

Ckmdition  of  gauge  and  equipment  at  river  station 

Repairs  necessary 

Remarks: 


o  o 


Digitized  by 


Google 


670 


ECONOMICAL  ASPECTS 


R-19 


FoBM  R-19— Front 
Return  to 


Water  Power  Q     Branch,  Department  of  the  Interior     Q 
Current  Meter  Notes 


Valuable 
Ottawa 


Date. 


.191. 


A.M. 

'p.m. 


Stream. 


Party Locality 

Meter  No. Gauge  height,  beg end mean 

Total  area Mean  velocity Discharge. 


Obbervationb. 

COMPUTATIOKB. 

B  c 

SI 

5   B 

i 

Q 

i 
1 

si 

I 

1 

a 

1 

1 
1 
1 

i 

< 

1 

i 

1 

♦a 
< 

si 

Diacfasise. 

- 

No of Sheets,  Comp.  by Chk.  by Make  notes  on  back 


Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  671 

Form  R-19— Back 

Weights  used 

Measurements  by  reading,  from  cable,  bridge  or  boat 

Wind 

Method  of  supervision  of  meter  (single  wire  or  cable) 

Stay  wire  used  or  not  used y 

Point  of  measurement  with  reference  to  gauge  (is  distant  above  or  below) . . . 

Length  of  gauge  chain  checked  and  foimd  to  be ft.  and  corrected  to ft. 

Ck>ndition  of  gauge  and  equipment  at  station 

Repairs  necessary 


Remabeb: 


o  o 


Digitized  by 


Google 


672 


ECONOMICAL  ASPECTS 


o                            o 

R-20                                              Return  to 
WATER  POWER  BRANCH,  DEFT.  OF  THE  INTERIOR,  OTTAWA 

No 

.191.... 

6 

.9 

1 

1 

> 

)        : 

z 

• 

.1 

i 
1 

1 

i 

1 

Surve 
Locat 
Photc 

Roll 
Holder. 

Digitized  by 


Google 


COMPILATION  OF  WATER  POWER  REPORTS  673 

R-21 

o  o 

DEPARTMENT  OF  THE  INTERIOR,  OTTAWA 

water  power  branch 

Discharge  Measurebcent  Notes 

Date ,  191 . .  No.  of  Meas 

River  at 

Width Area Mean  Vel Cor.  M.  G.  H 

Party Disch 

Gauge,  checked  with  level  and  found 

Measurement  began  at Measurement  ended  at 


Date  rated 

Method  of  meas 

No.  meas.  pts Coef 

Av.  width  sec Av.  depth . . 

G.  Ht.  change  (rate  per  hr.) 


First  reading  of  gauge ft.  at 

Gauge ft.  at  sta ft.  at 

Gauge ft.  at  sta ft.  at 

Gauge ft.  at  sta ft.  at 

Last  reading  of  gauge ft.  at 

Meter  No %  error  by rating  table . . 

Meas.  from  cable,  bridge,  boat,  wading;  Meas.  at ft.  above,  below  gauge 

If  not  at  regular  section  note  location  and  conditions 

Method  of  suspension Stay  wire Approx.  dist.  to  W.  S 

Arrangements  of  weights  and  meter;  top  hole ;  middle  hole ; 

bottom  hole 

Gauge  inspected,  found ;  Cable  inspected,  found 

Distance  apart  of  measuring  points  verified  with  steel  tape  and  found 

Wind upstr.,  downstr.,  across.    Angle  of  current 

Observer  seen  and  book  inspected 

Examine  station  locality  and  report  any  abnormal  conditions  which  might 
change  relation  of  G.  Ht.  to  disch.,  e.g.,  change  of  control;  ice  or  debris  on 

control;  back  water  from;  condition  of  station  equipment 

Sheet  No.  1  of sheets.  If  insufficient  space,  use  back  of  sheet. 


Digitized  by 


Google 


674  ECONOMICAL  ASPECTS 

R-22  O  O 

Return  to 

WATER  POWER  BRANCH,  DEPARTMENT  OF  INTERIOR,  OTTAWA 

Gauge  Record 

Station  No 

River  at 

Old  Gauge 

Location 


Zero 191 Elev . 

Kind  of  gauge. Length. . . 

New  Gauge 

Location 


Established 191 ...  .by. . . 

Zero 191 Elev. 

Kind  of  gaugq 

Reading  from ft.  to. . 

Gauge  reader Address.  .  . 

Time  of  observation 

Reason  for  change 


Remarks. 


Engineer 


Digitized  by 


Google 


AMOUNT  OF  ENERGY  AVAILABLE  675 

AMOUNT  OF  ENERGT  AVAILABLE 

The  two  principal  factors  which  enter  into  the  detennination 
of  the  available  energy  of  a  stream  are  the  fall  or  head  and  the 
quantity  of  water  flowing. 

The  head  is  usually  limited  by  the  cost  of  the  overflowed  lands, 
and  the  fall  may  be  either  naturally  concentrated  at  one  point  in  a 
cascade  or  it  may  be  artificially  concentrated,  for  the  purpose  of 
development,  by  combining  the  fall  of  several  cascades  or  a  series 
of  rapids.  This  may  be  accomplished  by  either  of  two  methods: 
First,  by  building  a  dam  at  the  downstream  end  of  the  rapids  to 
impound  the  water  so  that  the  entire  fall  is  concentrated  at  the 
dam;  or,  second,  by  building  a  dam  at  the  upstream  end  of  the 
rapids  and  conducting  the  water  through  a  closed  pipe  to  the  lower 
end  of  the  rapids,  where  the  resulting  head  and  pressure  will  be 
exactly  the  same  as  in  the  first  instance.  A  variation  in  the  latter 
method  consists  in  diverting  the  water  from  the  natural  channel 
at  the  head  of  the  rapids  and  carrying  it  to  a  canal  or  flume,  on  a 
slight  down  grade,  along  the  side  of  a  hill  to  a  suitable  point,  and 
there  erect  a  forebay  from  which  the  water  is  turned  into  pen- 
stocks which  run  directly  down  the  slope  to  the  stream,  where  the 
power-house  may  be  located.  The  latter  method,  involving  the 
construction  of  an  open  canal  or  flume,  is  open  to  the  objection 
that  trouble  may  be  experienced  from  the  accumulation  of  ice 
in  the  winter  time.  The  first  two  methods  described  are  the 
most  common. 

The  second  quantity  to  be  determined  was  the  water  flowing 
in  the  stream  per  unit  of  time,  usually  expressed  in  cubic  feet  per 
second,  but  for  low-head  developments  the  two  factors  of  head- 
and  stream-flow  are,  as  a  rule,  inseparable,  as  the  head  fluctuates 
considerably  with  the  different  stages  of  the  stream. 

To  be  of  value  the  stream-flow  data  should  extend  over  a 
period  of  several  years  (fifteen  to  twenty)  in  order  that  the  min- 
imimi  as  well  as  the  maximum  flows  which  may  be  expected,  and 
their  duration,  may  be  known,  and  while  the  average  flow  charac- 
teristics are  of  interest  they  are  not  of  very  great  value. 

The  United  States  Geological  Survey  and  various  states  have, 
for  many  years,  carried  on  a  systematic  stream-flow  measure- 
ments, and  data  are  now  available  for  streams  in  nearly  all  sections 
of  the  coimtry.    There  are,  however,  a  large  number  of  streams, 


Digitized  by 


Google 


676 


ECONOMICAL  ASPECTS 


especially  the  smaller  ones,  where  few,  if  any,  discharge  measure- 
ments have  been  made,  and  in  such  cases  it  is  necessary  to  base 
estimates  of  discharge  on  the  records  at  other  stations  in  the  same 
precipitation  belt  and  watershed,  and  data  of  other  systems  of 
similar  nature  may  be  also  used.  Rainfall  data  are  also  useful  as  a 
check  on  flow  estimates  and  they  also  show  years  of  high  and  low 
water,  but  care  should  be  exercised  in  their  use. 

The  daily  and  seasonal   distribution  of  stream-flow  is  best 
shown  graphically  in  the  hydrographic  curves,  as  fully  explained 


14Q,000n 

130,000- 

I'JO.OOO  - 

JUOJJOO- 

°  lOft,nQO  - 


I 


911,1101)- 

aiMiDO  - 

70.1JOO 

eo.uoo- 

»t,t)0O  - 
i  0,000 - 
wm  - 


\ 

\ 

\ 

y 

\ 

\ 

^ 

1l 

' 

\ 

\ 

^ 

H 

V 

V 

N 

V 

^ 

s 

N 

V 

\ 

s 

S 

S 

\ 

\ 

N 

\ 

b_ 

s 

i- 

s 

s 

\ 

S 

<^. 

\ 

^ 

[s 

\ 

\ 

\ 

Kfc, 

^ 

T. 

. 

^ 

s 

1 ^ 

h4i 

K 
ft 

i 

& 

n 

-^ 

Yi 

< 

< 

E 

u 

^ 

m 

^ 

^ 

^ 

^ 

- 

^ 

-' 

V 

?c 

- 

■^ 

^ 

^ 

H 

u 

" 

\     1 

^ 

z 

100 


IV)  ^uo 


£^ 


300 


35U3(S 


Fig.  398. — Stream-flow-Duration  Curves. 


under  the  section  of  Stream  Flow,  and  by  comparing  a  niunber  of 
such  hydrographs  the  dryest  year,  i.e.,  the  year  with  the  minimum 
flow,  can  readily  be  ascertained. 

For  convenience  in  making  a  scientific  analysis  and  study,  the 
stream-flows,  instead  of  being  arranged  chronologically  as  in  the 
hydrographs,  may  be  arranged  according  to  magnitude,  as  in  Fig. 
398.  The  discharge  is  plotted  as  ordinate  and  the  corresponding 
number  of  days  during  which  the  respective  discharge  has  oc- 
curred as  abscissas.  Instead  of  recording  the  time  in  days  it 
may  also  be  given  in  percentage  of  the  entire  year,  and  horse- 
power values  may  be  substituted  for  the  discharge  by  making 
allowance  for  any  possible  variation  in  the  head  at  the  different 
discharges. 


Digitized  by 


Google 


POWER  DEMAND  677 


POWER  DEMAND 

The  market  for  electric  power  is  of  a  most  widely  distributed 
character  and  will  always  continue  to  grow  with  the  growth  of 
the  community  in  which  it  is  located.  On  the  other  hand,  there 
are  many  instances  in  which  a  hydro-electric  development  will 
create  its  own  market  by  inducing  a  number  of  industries  to 
locate  in  its  inmiediate  vicinity,  such  as  at  Niagara  Falls,  etc. 

Whether  a  market  can  be  foimd  for  the  power  which  is  to  be 
developed  and  the  price  at  which  this  power  may  be  disposed  of 
are  two  of  the  first  questions  to  be  investigated.  This  involves  a 
close  canvass  of  the  present  power  consumption  for  both  public 
and  private  use,  the  character  of  the  p>ower  demand  as  to  periods 
of  day  and  season,  present  and  future  competition,  present  rates, 
and  the  cost  at  which  p>ower  can  locally  be  generated  from  fuel. 
From  these  investigations  it  is  p>ossible  to  anive  at  a  fairly  close 
estimate  of  the  required  capacity,  load  factor  and  value  of  the 
service,  and  future  considerations  should  be  based  thereon.  In 
the  absence  of  the  above  information  a  fairly  close  estimate  of 
the  revenue  may  be  made  by  comparing  the  possibiUties  of  the 
community  to  be  served  with  those  of  similar  places  already 
developed. 

A  typical  power  market  has  three  main  divisions,  namely, 
lighting,  manufacturing,  and  traction.  If  the  greatest  demand 
from  each  source  came  at  a  time  different  from  that  of  the  others, 
the  total  demand  would  be  so  distributed  as  greatly  to  reduce  the 
required  maximmn  capacity  of  the  power  plant.  As  a  matter  of 
fact,  however,  the  demand  from  no  one  of  these  sources  is  imif orm, 
and,  furthermore,  there  is  more  or  less  overlapping  of  these  de- 
mands. The  demand  for  manufactiuing  piu*poses  is  very  nearly 
uniform  and,  except  for  a  few  industries  and  in  exceptional  cases, 
falls  between  7  o'clock  in  the  forenoon  and  7  o'clock  in  the  after- 
noon. Practically  all  the  demand  for  Ughting  is  at  night,  chiefly 
in  the  evening.  The  period  of  traction  demand  is  longer  than 
that  for  either  manufacturing  or  lighting,  and  embraces  prac- 
tically the  entire  periods  of  both. 

TTie  period  of  lowest  combined  demand  is  normally  between 
the  hoiurs  of  midnight  and  4  o'clock  in  the  morning.  Traction 
demand  begins  in  earnest  about  6  o'clock  and  is  immediately 
followed  by  the  manufacturing  demand.    The  forenoon  period 


Digitized  by 


Google 


678  ECONOMICAL  ASPECTS 

of  active  demand  is  from  6  o'clock  to  noon.  In  the  middle  of  the 
day  manufacturing  establishments  cease  operations  for  an  hour 
or  less  and  resume  again  about  1  o'clock,  thus  restoring  the 
demand  to  the  level  of  the  forenoon.  Between  4  o'clock  and  7 
o'clock  in  the  afternoon  there  is  a  distinct  overlapping  of  the  three 
demands.  It  is  during  these  hours,  especially  in  winter,  that 
practically  all  the  lights  are  turned  on,  manufacturing  concerns 
have  not  yet  stopped  for  the  day  and  street  cars  are  carrying, 
perhaps,  their  heaviest  loads.  It  is  during  this  period  that  the 
highest  demand  of  the  twenty-four  hours  is  reached. 

There  is  also  a  seasonal  fluctuation  in  a  typical  power  market. 
The  demand  in  winter  is  usually  greater  than  in  summer  and  the 
daily  fluctuation  is  likewise  greater.  The  increased  demand  grows 
out  of  the  increased  requirements  for  Ughting  and  in  some  cases 
for  traction.  The  greater  fluctuation  is  mainly  due  to  the  fact 
that  between  the  hours  of  4  o'clock  and  8  o'clock  in  the  after- 
noon more  power  is  required  for  Ught  in  winter  than  in  sunmier. 

LOAD  AND  DIVERSIT7  FACTOR 

The  load  factor  of  a  plant  or  system  is  the  ratio  of  the  average 
to  the  maximmn  p)ower  during  a  certain  period  of  time.  The 
average  load  may  thus  be  taken  over  a  period  of  one  year,  one 
month  or  one  day,  while  the  maximum  load  must  necessarily  be 
limited  to  very  short  periods,  depending  on  the  overload  capacity 
of  the  water  wheel  or  the  generator.  In  other  words,  it  is  the 
ratio  of  the  actual  station  output  to  the  maximum  possible  output 
with  continuous  service. 

It  is  also  a  measiure  of  the  extent  to  which  the  necessary  total 
investment  is  being  utilized,  as  a  plant  with  yearly  load  factor  of 
50  per  cent  is  turning  out  just  double  the  energy  of  another  plant 
of  the  same  maximum  load  and  with  a  load  factor  of  25  peY  cent. 
This  means  that,  while  the  fixed  charges  of  both  plants  are  the 
same,  the  gross  income  of  the  plant  with  50  per  cent  load  factor 
should  be  nearly  twice  as  great  as  that  of  the  other.  The  impor- 
tance of  a  good  load  factor  is  thus  apparent,  and  everything  that 
will  improve  this  factor  should  be  sought. 

The  nature  of  the  load  as  measured  by  the  load  factor  forms 
necessarily  also  a  very  important  element  in  determining  the 
value  of  water  power  as  compared  to  steam  power.  For  load 
factors  below  50  per  cent  the  former  often  turns  out  to  be  the 


Digitized  by 


Google 


LOAD  AND  DIVERSITY  FACTOR  679 

cheaper,  but  as  the  load  factor  increases  above  this  value  water 
power  may  show  up  to  the  better  advantage.  This  is  evident 
from  the  fact  that  the  cost  of  hydro-electric  power  is  made  \ip 
chiefly  by  the  fixed  charges  and  is  very  Uttle  dependent  on  the 
operating  charges  and  the  amoimt  of  power  used. 

There  is  an  enormous  variety  of  uses  to  which  electricity  is 
applied,  the  yearly  load  factors  of  which  also  vary  widely,  as 
shown  in  Table  LXI. 

The  yearly  load  factor  for  any  class  of  service  is  determined 
largely  by  the  seasons,  the  habits  of  the  people,  and  other  con- 
ditions which  ordinarily  do  not  change  very  materially.  Im- 
provement in  the  load  factor  must,  therefore,  be  obtained  largely 
by  combining  different  classes  of  service,  the  maximmn  demands 
of  which  occur  at  different  times  of  the  day  or  of  the  year.  Also, 
the  larger  the  niunber  of  customers  in  any  class  the  better  will  be 
the  load  factor. 

A  recognition  of  the  importance  of  the  diversity  factor  has 
imdoubtedly  the  most  marked  effect  in  increasing  the  load  factor 
and  thereby  the  economy  of  production.  This  factor  is  the  ratio 
between  the  sum  of  the  maximum  demands  of  various  classes  of 
service  to  the  actual  simultaneous  maximum  demand,  and  the 
more  non-coincident  these  peak  services  are,  the  greater  will  be 
this  factor. 

The  chief  means  ot  improving  the  load  factor  has  been  the 
addition  of  industrial  load.  In  the  early  days  of  electric  Ughting 
companies,  the  load  factors  were  very  low,  due  to  the  absence  of 
day  load.  To-day  many  central  stations  sell  far  more  energy 
for  power  than  for  Ught,  and  this  is  naturally  distributed  over  a 
longer  part  of  the  twenty-foiu*  hoima.  The  power  load,  also,  not 
being  simultaneous  with  the  Ughting  load  to  any  great  extent, 
still  further  improves  the  load  factor.  Residence  load  has  gen- 
erally been  characterized  by  a  poor  load  factor,  but  by  the  use  of 
day-load  devices  such  as  flat  irons,  cooking  devices,  fans,  heating 
apparatus,  vacuum  cleaners,  etc.,  a  much  improved  result  is 
obtained. 

Tlie  problem  of  combining  electric  railway  loads  and  central 
station  loads  on  one  system  has  received  increasing  attention  in 
recent  years,  and  in  some  cities  of  this  country  great  strides  have 
been  made  toward  effecting  such  combinations  successfully. 
Kg.  399  thus  shows  a  typical  load  curve  for  a  large  city. 


Digitized  by 


Google 


680 


ECONOMICAL  ASPECTS 


There  are  a  number  of  industries  which  offer  ideal  loads  for 
large  hydro-electric  companies;  such  as  mining,  electro-chemical 
wdrk,  irrigation  and  farming,  while  much  is  expected  from  the 


[ 

I 

HOOO 

A 

\ 

18^000 

A 

/ 

\ 

/\ 

* 

1 

u 

\ 

12.000 

\ 

J 

\ 

V 

\ 

aa.acn 

• 

\ 

/ 

\ 

ao.ooo 

^\ 

1 

\ 

MOooQ 

I 

\ 

/ 

\ 

11 

\ 

\ 

/ 

\ 

\ 

1 

\ 

C7 

r^ 

\ 

i 

\ 

18,000 

\ 

/ 

/ 
/ 

V 

\ 
\ 
\ 

1' 

^ 

\ 

1 

1        / 

r 

^ 

\ 

12,000 

v 

A 

/ 

/        / 

/       1 
f 

\ 

\ 

\ 

''l 

1 

0,000 

\ 

\ 

\J' 

/ 

0 

1824        6        8       10       12       2i68l0      1S 
A.M«  P»l&* 

Fig.  399. — Typical  Daily  Load  Curve  for  Large  City  Service. 


railroad  field  when  the  time  has  arrived  for  the  economical  elec- 
trification of  our  trunk  hnes. 

Table  LXII  gives  statistics  for  1916  on  the  outputs,  peak  load, 
and  load  factors  of  a  number  of  the  largest  generating  systems  in 
this  country  and  Canada,  and  Table  LXIII  gives  the  power  re- 
quired for  certain  manufacturing  industries,  as  based  on  the  1909 
U.  S.  Census  Report. 


Digitized  by 


Google 


LOAD  AND  DIVERSITY  FACTOR  681 

TABLE  LXI 

Load  Factorb 

small  and  medium  lighting  customers 

Per  Cent. 

BuildingiSy  public 17.6 

Churches 12.4 

Clubs 9.6 

Flats 6.9 

Halls  (public) 6.9 

Hotels 24.4 

Offices  (business) 9.2 

Offices  (professional) 6.7 

Residences 7.8 

Restaurants 23.4 

Shops  (bakery) 13.1 

Shops  (tailor) 8.4 

Schools 7.2 

Stores  (dry  goods) 8.2 

Stores  (cigar) 16.8 

Stores  (drug) 19.3 

Stores  (grocery) 10.3 


LARGER  POWER  AND  LIGHTING   CUSTOMERS 

Per  Cent. 

Bakeries 12 

Blacksmith  shops 15 

Breweries 45 

Boots  and  shoes 25 

Bottling  works 10 

Candy  manufacturing 18 

Clothing  manufacturing 15 

Department  stores 30 

Furniture  manufacturing 28 

Foimdries 15 

Ice  cream  manufacturing 20 

Ice  making 30 

Laundries 20 

Machine  shops 20 

Newspapers 18 

Packing  houses 30 

Railroad  depots 50 

Tanneries 20 

TextUe  mills 20 


Digitized  by 


Google 


682 


ECONOMICAL  ASPECTS 


TABLE  LXII 

Data  on  Output  and  Load  Factor  of  Largest  Generating  Stbtems  in 

America 
(From  Electrical  World,  April,  7  1917) 


System. 


Commonwealth  Edison  Company.' 

Niagara  Falls  Power  Company 

Ontario  Power  Company  of  Niagara  Falls 

New  York  Edison  Company  &  United 

Electric  Light  &  Power  Company .... 

Montana  Power  Company 

Pacific  Gas  &  Electric  Company 

Hydraulic  Power  Company 

Toronto  Power  Compaiw 

Public  Service  Electric  Co.  of  N.  J 

Detroit  Edison  Company 

Tennessee  Power  Company 

Shawinigan  Water  &  Power  Company. . . 

Duquesne  Li^t  Company 

Philadelphia  Electric  Company 

Penninrlvania  Water  &  Power  Company.. 

Utah  Power  &  Light  Company 

Great  Western  Power  Company 

Mississippi  River  Power  Company 

Pacific  Lii^t  &  Power  Corporation 

Puget  Sound  TYaction,  Light  &  Power  Co. 

Cleveland  Electric  Illuminating  Co 

Electric  Company  of  Missouri 1 

Union  Electnc  Light  &  Power  Co / 

Commonwealth  Power,  Ry.  &  Light  Co .  . 
Southern  California  Edison  Company. . . 

Buffalo  General  Electric  Company 

New  England  Power  Company 

Edison  Elec.  Illuminating  Co.  of  Boston. 
Edison  Elec.  Illuminating  Co.  of  Bklyn. 

Wisconsin  Edison  Company \ 

Milwaukee  Elec.  Railway  &  light  Co.  / 
Portland  Railway,  Light  &  Power  Co .  . . 
Sierra  &  San  Francisco  Power  Company. 

Alabama  Power  Company 

Georgia  RaOway  &  Power  Company .... 
Minneapolis  General  Electric  Company. 

Great  Northern  Power  Company 

Washington  Water  Power  Company .... 
Adirondack  Electric  Power  Corporation . 
Rochester  Railway  &  Light  Company. .  . 

Toledo  Railways  &  Light  Company 

Virginia  Railways  &  Power  Company. . . 
Southern  Sierras  Power  Company. ...  1 
Nevada-California  Power  Company. . .  / 

Potomac  Electric  Power  Company 

Empire  District  Electric  Company 

Southwestern  Power  &  Light  Company. . 


1916 


Peak 

Yearly 

Yearty 

Load, 
Kw. 

Output. 
Kw.-hr. 

Factor. 
Per  Cent. 

369,740 

1,341,964,000 

43.20 

143,360 

1,015,525,680 

80.64 

123,900 

942,221,900 

86.80 

254,824 

856,385,319 

28.30 

149,740 

867,940,326 

84.50 

141,008 

768,304,907 

62.20 

89,275 

717,079,320 

91.50 

129,000 

660,873,579 

58.40 

174,000 

608,018,729 

39.82 

130,200 

546,925,300 

48.70 

81,650 

483,354,162 

67.00 

108,000 

478,540,000 

50.00 

101,000 

463,537,660 

52.30 

142,260 

444,785,884 

35.6 

77,000 

417,837,600 

61.8 

68,894 

412,726,000 

67.8 

74,100 

408,391,067 

62.65 

82,400 

393,400,000 

54.3 

82,765 

367,308,731 

51.76 

77.030 

353,697,263 

51.8 

84,999 

340,670,721 

45.8 

88,544 

333,964,652 
315,964,337 

43.1 

66,'93d 

299,950,513 

56.04 

65,500 

299,306.640 

57.00 

64,000 

246,000,000 

44.00 

80,539 

238.557,144 

33.72 

67,200 

233,452,500 

38.1 

64,170 

218,421,711 

39.00 

47,335 

194,146,555 
191,620,000 

46.5 

46,600 

184,345,360 
172,000,000 

51.07 

43,640 

171,672,890 

44.9 

38,200 

163,807,560 

48.8 

30,440 

162,825,400 

60.80 

41,575 

151,128,310 

41.40 

40,250 

146,069,428 

41.00 

36,428 

134,842,360 

42.2 

33.900 

132,275,000 

44.54 

22,400 

131,084,265 

66.5 

38,600 

122,158,818 

36.1 

26,900 

119,280.363 

49.7 

25,600 

95,740,000 

43.0 

Digitized  by 


Google 


PRIMARY  AND  SECONDARY  POWER 


683 


TABLE  LXIII 

Power  Rbquirsd  for  Manufacturing 

Based  on  1909  U.  S.  Census 


Horse-power 

Required  per 

SI 000  Product. 


Horse-power 

Used  per  Person 

Engaged  in 

Industry. 


Agricultural  implements 

Automobiles 

Boots  and  shoes 

Brick  and  tile 

Cement 

Chemicals 

Copper,  tin  and  sheet-iron  products. . 

Cotton  goods 

Electrical  machinery. 

Fertilizers 

Flour  and  grist-mill  products 

Foundry  and  machine  shops 

Manufactured  ice 

Iron  and  steel,  blast  furnaces 

Iron  and  steel,  rolling  mills 

Leather,  tanned,  curried  and  finished 

Lumber  and  timber 

Paper  and  wood  pulp 

Printing  and  publishing 

Packing  houses 

Copper  smelting  and  refining 

Woolen,  worsted  and  felt  goods 

Total,  all  industries 


0.69 
0.30 
0.19 
3.68 
5.90 
1.78 
0.31 
2.07 
0.72 
0.62 
0.97 
0.71 
7.40 
3.00 
2.13 
0.45 
2.46 
4.88 
0.40 
0.15 
0.42 
0.83 

0.91 


1.67 
0.89 
0.45 
4.00 

12.60 
7.50 
0.72 
3.35 
1.50 
2.95 

12.90 
1.41 

15.05 

27.30 
8.06 
2.21 
3.62 

16.05 
0.77 
1.92 
9.41 
2.06 

2.45 


PRIMARY  AND  SECONDARY  POWER 

Many  companies  make  two  classes  of  contracts  for  power, 
known  as  primary  and  secondary.  Under  the  terms  of  primary 
power  it  guarantees  to  supply  the  amount  of  p)Ower  contracted 
for  continuously  throughout  the  year,  and  it  is  evident  that 
the  maximum  amoimt  of  such  power  is  limited  by  the  minimum 
stream-flow  and  can  only  be  safely  increased  by  providing  water 
storage  or  steam  auxiliaries  to  augment  the  shortage  during 
low-water  periods. 

The  minimum  flow  of  the  stream  to  be  used  may  be  the  abso- 


Digitized  by 


Google 


684  ECONOMICAL  ASPECTS 

lute  minimum,  the  minimum  of  the  average  year,  the  average 
minimum,  or  some  other  value  of  low  discharge  of  the  stream. 
The  selection  of  the  particular  value  to  be  used  depends  upon  the 
degree  of  insurance  of  the  continuity  of  the  supply  that  is  justified 
by  the  conditions.  The  added  cost  of  the  insurance  of  the  supply 
should  be  equated  to  the  losses,  direct  and  indirect,  sustained  by 
failure  of  the  supply.  If  it  is  planned  to  seciu^  absolute  con- 
tinuity, in  so  far  as  stream-flow  is  concerned,  it  will  be  necessaiy 
'to  use  the  absolute  minimum  of  the  stream  and  to  use  it  in  con- 
nection with  the  maximum  load  that  can  occur  upon  any  day 
when  the  stream-flow  may  be  lowest.  This  d^ree  of  insurance  is 
seldom  necessary;  usually  it  will  be  sufficient  to  use  the  stream- 
flow  which  can  be  depended  on  for,  say  four  years  out  of  five; 
in  other  words,  to  eliminate  the  extraordinary  low  discharge  which 
will  occur  once  in  every  five  to  ten  years.  But  on  this  point,  as 
in  all  others  in  connection  with  the  matter,  the  decision  depends 
upon  the  experience  and  judgment  of  the  engineer,  and  no  hard- 
and-fast  rule  can  be  laid  down.  One  kind  of  load  demands  the 
highest  degree  of  insurance,  whereas  loads  of  a  different  character 
may  be  satisfactorily  served  with  a  less  degree  of  insurance. 

Secondary  power  is  that  amoimt  which  is  being  developed 
above  the  primary,  and  which  is  only  available  for  a  certain  time 
of  the  year,  such  as  eight  or  ten  months.  The  continuity  of  this 
power  is,  as  a  rule,  not  guaranteed,  and  the  right  is  reserved  to 
cut  off  such  supply  upon  reasonable  notice.  The  rates  for  sec- 
ondary power  are,  therefore,  much  lower  than  for  primary  power. 
>  The  question  of  the  sale  of  secondary  p>ower  has  yet  not 
reached  the  proportions  to  which  it  is  entitled,  but  there  is  ever>' 
reason  to  beUeve  that  by  careful  planning  of  certain  industries 
quite  a  large  amoimt  of  secondary  p)ower  could  be  very  econom- 
ically utilized. 

The  question  as  to  what  extent  a  power  site  should  be  devel- 
oped depends  necessarily  upon  the  market  conditions  for  the  two 
classes  of  service.  It  needs  no  argument  to  prove  that  where 
power  can  be  sold  at  a  high  price  and  conditions  are  favorable,  the 
development  can  be  carried  to  a  higher  point  of  stream-flow  than 
where  the  opposite  conditions  prevail.  Over-development,  how- 
ever, may  entail  fixed  charges  which  will  make  the  earning  of  a 
surplus  only  a  speculative  possibility  of  the  distant  future.  On 
the  other  hand,  the  present  demand  and  its  probable  future 


Digitized  by 


Google 


WATER  STORAGE  685 

increase  may  both  be  done  justice  by  the  correct  solution  of  this 
factor.  As  a  rule,  however,  the  development  of  a  p>ower  site 
usuaUy  also  involves  the  consideration  of  an  auxiliary  power 
source,  such  as  a  storage  reservoir  or  a  steam  plant. 

If  the  secondary  power  can  be  sold  without  an  auxiliary  steam 
plant,  the  amount  of  secondary  power  which  may  be  developed 
economically  depends  only  upon  whether  or  not  the  price  received 
for  such  power  will  cover  interest  and  profit  on  the  investment; 
that  is,  the  investment  which  is  over  and  above  that  for  develop- 
ing primary  power.  If  a  steam  plant  has  to  be  maintained  the 
amount  of  secondary  power  to  be  developed  depends  also  upon 
the  cost  of  the  steam  power. 

WATER  STORAGE 

In  order  to  increase  the  capacity  of  a  hydro-electric  plant 
at  times  of  low  water,  the  question  of  storage  is  one  of  vit^  im- 
portance, and  the  extent  to  which  the  irregular  stream-flow  can 
be  equalized  depends  upon  the  quantity  of  storage.  It  is  also 
obvious  that  no  considerable  amount  of  money  can  be  judiciously 
expended  in  the  construction  of  storage  reservoirs  imder  average 
conditions  unless  the  head  available  at  the  plant  is  considerable, 
and  this  question  must  be  largely  determined  by  local  conditions 
surroimding  each  individual  development. 

Water-storage  problems  are  most  readily  solved  graphically 
by  means  of  "  mass-curves,'^  and  the  most  economical  solution  is 
fixed  by  balancing  the  value  in  the  increase  in  output  as  against 
the  cost  of  securing  the  same.  From  the  mass-curve,  the  available 
water  for  power  is  obtained  and  this,  under  given  net  heads  will 
determine  the  p>ower  available. 

The  application  of  the  "  flow-smnmation  "  or  "  mass-curve  " 
to  problems  of  water  storage  is  clearly  explained  by  Mr.  E.  C. 
Jansen  in  the  Engineering  News  for  December  25,  1913,  as  follows: 

"  To  plot  the  stream-flow  for  any  period  of  time,  the  mean 
daily  discharges  in  any  convenient  unit  are  added  day  by  day  and 
plotted  as  ordinates,  the  units  of  time  being  represented  by 
abscissas,  so  that  the  siun  total  or  ordinate  on  any  date  repre- 
sents the  total  quantity  of  water  which  has  flowed  past  the  gauging 
station  up  to  that  date  (see  curve  ABODE,  Fig.  400).  Second- 
feet  (cubic  feet  per  second)  are  now  most  commonly  used  as  units 
of  flow  and,  when  the  mean  daily  discharges  are  expressed  as  such, 


Digitized  by 


Google 


686 


ECONOMICAL  ASPECTS 


the  summation  of  them  results  in  convenient  units  of  day-eecond- 
feet  or  a  second-foot  flowing  for  twenty-four  hours  (1.98  acre-feet) 
as  in  Fig.  400.  As  the  length  of  the  ordinates  shows  the  increase 
or  decrease  of  the  twenty-four-hour  flow,  it  will  readily  be  seen 
that  the  slope  of  the  curve  represents  the  rate  of  flow  and  that  a 
uniform  flow  is  represented  by  a  straight  line  as  FG." 


Fig.  400. — Flow-Summation  Curve. 

The  inclination  of  a  tangent  to  the  curve  at  any  point  indicates 
the  rate  of  flow  at  that  particular  time,  and  when  the  tangent  is 
parallel  to  the  abscissas  it  illustrates  that  the  flow  at  that  time  wHl 
just  balance  the  losses  caused  by  evap>orationy  seepage,  etc.,  while 
a  negative  incUnation  of  the  curve  shows  that  a  loss  from  the 
reservoir  is  taking  place. 

''Assume,  for  example,  that  FO  represents  a  regulated  or 


Digitized  by 


Google 


WATER  STORAGE  687 

uniform  rate  of  flow  of  30  second-feet,  then,  by  applying  this 
slope  as  a  tangent  to  the  summation  cm^e  at  C,  it  is  observed 
that  the  stream  from  about  October  1st  began  to  .discharge  less 
than  this  flow  and  did  not  rise  above  the  same  until  the  beginning 
of  April  at  D.  The  flow  can  be  readily  interpreted  in  this  way  by 
drawing  a  datum  and  different  slopes  or  tangents  on  a  piece  of 
tracing  linen  and  applying  this  at  any  point  on  the  curve." 

Having  a  certain  reservoir  capacity  and  the  mean  daily  dis- 
charges of  a  stream  for  a  period  of  years,  the  method  of  utiUzing 
the  summation  curve  for  finding  the  maximum  regulated  flow 
which  can  be  obtained  for  power  purposes,  is  explained  in  the  fol- 
lowing paragraphs. 

"  ABODE  represents  the  stream-flow  in  day-second-feet 
(usually  a  period  of  minimum  run-off  when  water  power  is  con- 
templated) ;  FG  is  the  desired  regulated  flow  and  H  is  the  capacity 
of  the  reservoir  ift  day-second-feet.  Starting  with  a  full  reservoir 
on  or  about  October  10,  1895  (the  smaller  units  of  time  are  pur- 
posely omitted),  the  sunmiation  curve  shows  that  the  stream-flow 
is  below  the  desired  regulated  flow  ABi,  parallel  to  FG,  and  that 
the  ordinates  JK,  LM,  etc.,  represent  the  amounts  of  storage 
required  to  maintain  the  regulation.  Plotting  these  ordinates 
below  the  high-water  level  of  the  reservoir  in  the  storage  diagram 
as  JiKi,  L\Mi,  etc.,  the  storage  curve  HiJ\L\  is  obtained,  showing 
the  behavior  of  the  reservoir  during  the  uniform  rate  of  dis- 
charge for  power  purposes.  At  5,  about  April  10,  1896,  the  sum- 
mation curve  shows  that  the  stream-flow  is  above  the  desired 
regulated  flow;  consequently,  the  ordinates  NO,  PQ,  etc.,  show 
the  amount  of  water  which  can  be  stored  and  these  ordinates  are 
plotted  as  ATiOi,  PiQi,  etc.,  for  the  remaining  portion  of  the  stor- 
age curve  imtil  the  reservoir  fills  about  June  1.  By  continuing 
the  plotting  of  these  ordinates  RS,  TU,  etc.,  as  RiSi,  TiUi,  etc., 
in  the  storage  diagram,  the  cmre  SiUiWi  is  obtained,  showing 
the  quantity  of  water  which  passes  over  the  reservoir  spillway. 
This  process  is  then  repeated  and  in  this  way  is  ascertained  the 
behavior  of  the  reservoir  from  year  to  year  while  a  continuous 
draft  is  being  made  on  it.  The  ordinate  X,  showing  the  water 
left  in  the  reservoir  at  the  end  of  the  drawing  period,  enables  one 
to  experiment  with  differently  regulated  flows  to  ascertain  just 
how  much  draft  the  reservoir  can  stand.  Frequently  two  or  three 
exceptionally  dry  years  in  succession  in  a  long  period  of  obser- 


Digitized  by 


Google 


688  ECONOMICAL  ASPECTS 

vation  will  tax  the  reservoir  capacity  to  its  limit  and  settle  the 
question  conclusively  as  to  the  maximum  regulated  flow  obtaio- 
able." 

Having  the  mean  daily  discharges  of  a  stream,  it  miay  also 
be  required  to  find  how  large  a  reservoir  is  required  to  obtain  a 
maximum  regulated  flow.  This  may  also  be  obtained  from  Fig. 
400.  By  drawing  a  line  from  B  to  D  the  maximum  regulated 
flow  utilizing  all  the  water  is  foimd,  and  the  ordinate  V2W  rep- 
resents the  capacity  of  the  reservoir  in  day-second-feet,  which 
would  be  required  to  effect  this. 

The  above  method  is  suitable  for  determining  the  power  pos- 
sibihties  of  a  given  development  when  one  or  two  power-houses 
with  accompanying  reservoirs  are  involved.  When  a  large  num- 
ber of  related  power-houses  and  reservoirs  are  involved,  this 
method  of  using  the  mass  curve  of  discharge  becomes  very  long 
and  tedious.  Also,  it  is  only  approximate,  giving  as  a  result  uni- 
form flow  of  water,  not  uniform  power,  and  it  fails  to  take  into 
account  regulative  effect  on  the  power  output  of  the  power-houses 
situated  on  the  upper  sections  of  the  watershed.  To  solve  these 
more  intricate  problems,  a  method  of  determination  has  been 
proposed  by  Mr.  L.  A.  Whitsit,  and  is  described  in  the  Engineering 
News  for  September  11,  1913.* 

The  utihzation  of  stored  water  so  as  to  absolutely  insure  a 
fixed  minimmn  flow  in  all  years,  while,  perhaps,  best  for  streams 
whose  power  is  not  developed  up  to  the  limit,  leads  to  a  very 
uneconomical  use  of  the  reservoirs  on  streams  which  already  are 
highly  developed  as  to  power.  As  a  condition  of  high  ratio  of 
development  exists  on  many  streams  where  storage  would  be  most 
desirable  and  valuable,  and  as  this  condition  will  become  more 
and  more  pronounced  on  all  power  streams,  it  is  apparent  that  the 
subject  of  this  basis  of  figuring  the  power  benefits  is  of  importance 
in  securing  a  proper  view  of  the  relation  of  water  storage  to  water 
development. 

The  conditions  may  be  such  that  when  the  method  of  regu- 
lating for  a  minimum  steady  flow  of  water  is  appUed,  it  has  been 
found,  for  example,  that  the  storage  capacity  would  have  been 
used  to  its  full  extent  only  once  in  ten  years.  Dming  six  of  the 
ten  years  it  would  not  have  been  used  at  all,  and  during  two  years 

^  See  also  Engineering  News,  Aug.  24,  1916. 


Digitized  by 


Google 


WATER  STORAGE  689 

only  about  one-half  of  the  capacity  would  have  been  used.  It  is, 
therefore,  evident  that  capital  if  invested  for  use  only  once  in  ten 
years  must  when  it  is  used  yield  a  very  large  return.  Such  a 
method  of  management  of  a  storage  reservoir  would  call  forth 
just  criticism  when  it  was  discovered  that  after  money  had  been 
spent  for  the  auxiliary  power  during  the  low-water  season,  the  stor- 
age reservoir  remains  full  of  water.  This  has  led  the  Water 
Supply  Commission  of  the  State  of  New  York  to  deduce  a  new 
method  of  computation,  which  is  based  on  an  average  rate  of 
release  of  stored  waters,  so  that  while  the  assurance  of  a  certain 
minimum  flow  would  not  be  unduly  sacrificed,  the  entire  volume 
of  stored  water  could  be  used  practically  every  year.  This 
method,  which  has  been  termed  the  "  utility  "  method  to  dis- 
tinguish it  from  the  *'  insurance  "  method,  has  been  based  on  a 
knowledge  of  the  conditions  of  the  larger  streams  of  the  State, 
where  the  developments  can  be  run  at  full  capacity  up  to  about 
60  per  cent  of  the  time  reckoned  over  a  long  period  of  time,  and  it 
assumes  that  there  is  always  sufficient  demand  for  power  to 
absorb  any  additions  and  render  further  development  after 
regulation  as  desirable  as  before. 

Figs.  401  and  402  represent  graphically  the  results  of  an  inves- 
tigation for  the  regulation  of  the  Genesee  River  by  providing  a 
storage  reservoir  having  a  capacity  of  13.5  billion  cubic  feet. 

The  stream-flows  are  arranged  according  to  magnitude,  and 
result  in  the  curve  marked  "  Natural  Flow  of  River."  Although 
the  vertical  scale  is  given  in  horse-power,  the  power  is  propor- 
tional to  the  stream-flow  as  long  as  the  head  is  not  affected,  and 
the  curve  would  not  be  changed  in  any  respect  if  stream-flow 
instead  of  power  were  used.  In  order  to  plot  the  "  Regulated 
Flow  "  curve  the  mass  curve,  as  previousl;^  explained,  is  used,  and 
the  regulated  flows  are  also  arranged  according  to  magnitude 
and  the  values  plotted  as  for  the  natural  flow. 

The  results  were  based  on  a  "  present "  wheel  installation  of 
29,200  horse-power,  and  by  referring  to  diagram.  Fig.  401,  it  will 
be  seen  that  one-fifth  of  all  the  water  power  with  regulated 
flow  and  present  wheel  capacity  will  be  derived  from  the  stored 
water,  shown  by  the  vertically  sectioned  area.  Without  regula- 
tion the  present  installation  can  be  operated  at  its  full  capacity 
for  only  58  per  cent  of  the  time  and  diminishes  to  a  minimum  of 
about  7500  horse-power.    Similarly  the  amount  of  energy  neces- 


Digitized  by 


Google 


ECONOMICAL  ASPECTS 


sary  on  the  average  from  auxiliary  power  is  shown  by  the  hori- 
zontally sectioned  area.  It  amounts  to  approximately  3000 
horse-power,  which  thus  is  required  to  maintain  continuously  the 
full  power  output  equal  to  the  present  wheel  capacity. 

The  diagram  in  Fig.  402  indicates  what  will  be  the  limit  of 


111        jo       k:u        4i>        ^'1 
Percentage  of  Time 


Fig.  401. — Power-percentQ.ge   of   Time   Curves   of   the  Genesee   River   at 

Rochester,  N.  Y. 


^u  4L»  ^U         m  iU  MJ         iHJi         i(Mi 

Percentage  of  Time 


Fig.  402. — ^Power-percentage   of   Time   Curves  of  the  Genesee  River  at 

Rochester,  N.  Y. 

economic  development,  it  being  near  the  point  where  the  regulated- 
flow  curve  takes  the  sharp  downward  bend.  As  the  installation 
capacity  increases  above  that  amount,  the  percentage  of  time 
during  which  further  capacity  can  be  .used,  diminishes  rapidly. 


Digitized  by 


Google 


WATER  STORAGE  691 

The  economic  limit  of  capacity  for  the  particular  development  in 
question,  for  a  steady  twenty-four-hour  j)Ower  after  regulation,  is 
thus  seen  to  be  approximately  40,000  horse-j)Ower,  based  on  a' 
228-foot  head.  Such  a  development  would  run  twenty-four 
hour?  per  day  58  per  cent  of  the  time,  or  seven  months  per  year 
on  the  average.  The  energy  furnished  by  the  natural  flow  each 
year  would  be  29,000  horse-power-years,  from  stored  water  8400 
horse-power-years,  and  from  the  auxiliary  source  3100  horse- 
power-years. 

The  diagrams  also  bring  out  the  fact  that  full  economic  advan- 
tages of  the  stream  cannot  be  secured  even  after  regulation  without 
auxiliary  power.  They  also  show  that  a  small  auxiliary  plant 
will  render  more  additional  energy  available  from  the  stream-flow 
after  regulation  than  the  same  amount  of  auxiliary  capacity  would 
render  available  before  regulation;  i.e.,  after  regulation  auxiUary 
pow^er  is  more  essential  to  the  best  economic  results  than  before 
regulation. 

All  the  above  has  been  based  on  a  steady  twenty-four-hour  use 
of  power;  i.e.,  a  load  factor  of  100  per  cent.  The  general  con- 
clusions are  not,  however,  affected  by  a  smaller  load  factor,  and 
where  there  is  pondage  a  low  load  factor  simply  permits  a  larger 
economic  installation.  Thus,  in  the  above  case,  with  a  load  factor 
of  62  per  cent  the  economic  development  would  be  about  64,000 
horse-power. 

A  point  in  connection  with  water-storage  problems  which  is  not 
always  realized  is,  that  while  a  given  quantity  of  water  in  stor- 
age will  raise  the  minimum  flow  of  the  stream  a  certain  definite 
amount,  a  further  addition  of  that  same  quantity  of  storage,  when 
put  into  the  stream,  will  not  raise  the  minimum  flow  by  any- 
thing like  the  first  quantity,  because  its  use  will  have  to  be  distrib- 
uted throughout  a  longer  period  of  time  in  the  year.  Therefore, 
as  storage  reservoirs  continue  to  be  built  out,  the  increment  in 
the  minimum  flow  becomes  less  and  less,  which  means  that  as  the 
development  of  storage  reservoirs  progresses,  the  economical 
outlay  per  unit  of  storage  becomes  less  and  less,  and  the  time 
comes  when  it  becomes  cheaper  to  increase  the  minimum  flow 
by  means  of  an  auxiliary  steam  plant. 

This  may  be  illustrated  by  the  diagram,  Fig.  403,  which  rep- 
resents a  typical  hydrograph  or  river-flow  curve.  It  will  be  noted 
that  the  minimum  flow  shown  by  this  curve  is  470  cu.  sec.  ft. 


Digitized  by 


Google 


692 


ECONOMICAL  ASPECTS 


The  introduction  of  100  sq.  mi.  ft.  of  stored  water  will  raise  the 
minimum  flow  to  1100  cu.  sec.  ft.,  a  difference  of  630  cu.  sec.  ft. 
If  now  further  stored  water  in  units  of  100  sq.  mi.  ft.  is  intro- 
duced, the  figure  clearly  shows  the  decreasing  amount  by  which 
the  minimum  flow  is  increased.  It  is,  however,  to  be  distinctly 
understood  that  it  applies  solely  to  the  minimum  rate  of  stream- 


1     1    1 

ll 

Annn 

EFFECT  OF  STORED  WATER      • 
fiOOO  --                                                                                                                                               — ^ 

°^                                                                                                              ON                              1 

MINIMUM  FLOW  OF  STREAM      1 

^4000     / ! L-      _      1 

1          ,      J.         ji    1 L.iL- 

._  'Win                                                                                                        -            - .  - .             _        ♦  ^              — 

^3000           -           ^                       -     -           -     — 

1  ._....^ +-V---I ^- 

(thlOOSj.Mi.Ft.           I                                        A 

/-'    'j./j'r*/           2n«                    '                      hT                                                        ^^^  ^^ '' ^^v    i^f''> 

^^f^.^^^J'               lit                     1                        B                      _|                            1             ^'^''^^  ^'^^^^    )   j  "^ 

"^©L                                                           t^'piti!-:^ 

liDimpm  Flow  W  ihoul  Storage                            |             ;   '         , 

irr   1    iTTT     M!  '   1 

Mar.     Apr.      May     June     July      Aug.     Sept.     Oct.     Not.     Dec.     Jan.     Feb. 

P'iG.  403. — ^Effect  of  Stored  Water  on  Minimum  Flow. 

flow  and  does  not  mean  a  proportionately  lower  volume  of  water 
available  for  j)Ower  production. 

This  decrease  in  minimum  flow  increment  is  shown  by  the  curve 
Fig.  404,  which  carries  the  stored  water  up  to  800  sq.  mi.  ft., 
resulting,  in  this  particular  instance,  in  the  minimum  flow  of  2000 
cu.  sec.  ft.,  as  against  470  cu.  sec.  ft.  without  storage. 

AUXILIARY  STATIONS 

In  the  previous  section  it  was  shown  that  the  full  economic 
advantage  of  the  stream,  even  with  storage  regulation,  cannot  be 
secured  without  a  source  of  auxiliary  power.    Such  auxiliary 


Digitized  by 


Google 


AUXILIARY  STATIONS 


693 


stations  may  be  divided  into  four  classes  according  to  their  utiliza- 
tion, although,  in  reality,  they  may  not  differ  essentially,  as  some 
stations  may  serve  two  or  three  different  purposes  simultaneously. 


2800 

flram 

1800 

^ 

^ 

"" 

1000 

; 

1.00 

^1200 

^ 

c 

V 

^ 

o 

^   800 

600 

400 

/ 

/ 

INCREASE  OF  MINIMUM  FLOW 

WITH 

ADDITIONAL  STORAGE 

/ 

' 

0 

100 


aoo 


900  400  500 

Square  Mile  Foot  of  Storage 


000 


700 


800 


Fig.  404. — Increase  of  Minimum  Flow  with  Additional  Storage. 

Class  I.  Stand-iy  Stations,  which  are  intended  to  take  care  of  the 
load  in  case  of  a  breakdown  to  the  hydro-electric  machinery 
or  the  transmission  lines. 
Class  II.  LovMJoaier  Stations,  which  are  intended  to  supplement 

the  load  during  low-water  periods. 
Class  III.  Peak-load  Stations,  which  are  intended  to  carry  peak 

loads. 
Class  IV.  Base4oad  Stations,  which  are  intended  to  operate  con- 
tinuously, the  water  power  being  supplemental  to  the 
steam  power. 
Prime  Movers.    There  are  four  kinds  of  prime  movers  which 
may  be  used  for  auxiliary  stations;  the  steam  turbine,  the  steam 
engine,  the  gas  engine  and  the  oil  engine.     Of  these,  the  steam- 
turbine  is  used  almost  exclusively,  but  the  question  of  deciding 
on  the  most  economical  and  practical  equipment  is,  naturally,  a 
problem  which  involves  a  study  of  each  case  individually. 


Digitized  by 


Google 


694  ECONOMIACL  ASPECTS 

The  auxiliary  p)Ower  can  either  be  secured  by  op)eratiDg  old 
steam  plants  of  the  power  customers  which  have  been  shut  down 
by  pm-chase  of  power  from  a  water-power  company  or  by  con- 
structing new  steam-turbine  plants  as  part  of  the  water-power 
system. 

Stand-by  Stations.  Emergency  reserve  stations  are,  as  a  rule, 
more  necessary  in  the  early  days  of  a  hydro-electric  develop- 
ment than  after  the  operating  conditions  become  settled.  They 
are,  however,  essential  in  order  to  provide  against  possible  inter- 
ruptions to  the  service  and  contract  provisions  are  often  such 
as  to  make  their  installation  almost  imperative. 

The  size  of  such  stations  is  naturally  governed  by  the  load 
which  must  be  maintained  under  all  conditions.  Their  location 
should  be  close  to  important  distributing  centers  so  as  to  be  use- 
ful in  case  of  breakdown  of  the  transmission  lines.  For  large  and 
extended  systems  it  may  be  advisable  to  provide  two  or  more 
distributed  stations  rather  than  one  of  the  combined  capacity. 

A  quick  start  is  an  essential  requirement  of  an  emergency 
stand-by  station.  It  is,  however,  not  customary  to  have  all 
the  boilers  under  fire  to  take  over  the  load  inmiediately  in  case 
of  an  interruption.  Some  of  the  boilers  are,  as  a  rule,  kept 
imder  banked  fires  part  of  the  time  to  secure  the  most  important 
load,  and  the  turbines  are  operated  as  synchronous  condensers 
to  improve  the  power-factor  of  the  entire  transmission  system, 
which  may  carry  a  large  inductive  load. 

Under  these  circumstances  it  is  particularly  easy  to  respond 
to  sudden  load  demands  because  the  imit  is  already  up  to  speed 
and  in  synchronism,  the  turbine  is  kept  warmed  up,  and  only  a 
change  in  the  field  excitation  is  necessary  to  place  the  unit  on  the 
line,  which  takes  only  a  few  minutes  at  the  most.  When  storms 
are  approaching,  the  entire  reserve  equipment  should  be  made 
ready  to  respond  immediately  to  any  emergency  that  may  arise. 

The  first  cost  of  the  station  should  be  low,  while  efficiency  is 
not  such  an  important  item.  Consideration  should,  however,  be 
given  to  the  possibility  that  it  may  later  be  used  under  other 
operating  conditions  requiring  the  highest  efficiency.  It  is, 
therefore,  often  advisable  to  make  provision  in  the  design  from  the 
beginning  so  that  economizers  and  other  labor-saving  devices  may 
be  installed  at  a  future  date,  should  conditions  so  demand.  With 
large  steam-turbine  units  it  is,  however,  practical  to  obtain  the 


Digitized  by 


Google 


AUXILIARY  STATIONS  695 

moet  efficient  unit  at  practically  the  same  cost  as  one  of  poorer 
efficiency.  A  less  boiler  capacity  is,  of  course,  needed  with  a 
higher  turbine  efficiency  and  consequently  a  plant  of  high  effi- 
ciency can,  as  a  rule,  be  built  at  practically  the  same  cost  as  one 
of  lower  efficiency. 

Low-water  Stations.  The  function  of  the  auxiliary  plant, 
when  used  as  supplemental  capacity  during  low-water  periods  is 
similar  to  that  of  the  storage  reservoir.  It  converts  at  least  a 
part  of  the  secondary  power,  which  would  be  available  only  part 
of  the  year,  into  primary  power  available  at  all  times,  thus  increas- 
ing its  sale  value.  It  is  also  of  value  in  making  up  shortage  of 
water  power  from  loss  of  head  during  high  back-water  caused  by 
floods.  Enough  pondage  can  usually  be  provided  to  insure  that 
daily  fluctuations  can  be  taken  care  of,  even  though  the  peak  load 
is  somewhat  in  excess  of  the  power  corresponding  to  the  minimiun 
stream-flow.  This,  of  course,  necessitates  that  the  average  or 
integrated  load  over  the  twenty-four-hour  period  must  be  within 
the  energy  available  from  the  minimum  stream-flow. 

The  problem,  therefore,  really  resolves  itself  into  two  ques- 
tions: First,  in  the  case  of  a  plant  already  in  operation,  to  what  an 
extent  shall  an  auxiliary  supply  be  provided  to  convert  the  variable 
power  supply  into  a  continuous  supply?  Second,  in  case  of  a 
new  development,  for  what  capacity  shall  it  be  bmlt? 

Both  cases  involve  a  study  of  the  stream-flow  and  the  load 
conditions,  the  first  cost  and  annual  operating  charges  for  the 
hydro-electric  plant  of  different  capacities  as  well  as  the  cor- 
responding charges  for  auxiliary  plants  of  the  required  capacities. 
In  the  first  case  the  cost  of  the  auxiliary  supply  for  various  degrees 
of  insurance  is  determined  and  compared  with  the  increased  earn- 
ings obtained  by  converting  the  secondary  j)ower  into  primary. 
In  the  second  case  the  problem  may  be  considered  from  several 
different  points  of  view.  So,  for  example,  one  may  start  out  with 
the  assumption  that  the  total  cost  per  kilowatt  and  year  shall  be  a 
minimiun,  or,  if  all  the  power  produced  can  be  sold  in  the  market 
at  a  certain  price,  it  should  be  investigated  at  what  plant  capacity 
the  profit  becomes  a  maximum.  In  the  case  of  a  new  develop- 
ment, the  cost  per  kilowatt  decreases  as  the  capacity  increases, 
and  an  increase  in  the  annual  cost  per  kilowatt  of  the  auxiliary 
plant  is  accompanied  by  a  decrease  in  the  annual  cost  of  the 
hydraulic  plant    A  point  may,  therefore,  be  reached  at  which 


Digitized  by 


Google 


696  ECONOMICAL  ASPECTS 

the  sum  of  the  two  is  a  minimum,  and  this  would  fix  the  most 
economical  capacity  of  the  development  and,  hence,  the  point  of 
greatest  profit  for  a  given  market  price  of  energy.  The  entire 
problem  of  determining  the  economical  capacity  of  a  combined 
hydro-electric  and  steam-power  plant  is  very  complicated.  An 
excellent  treatise  on  this  subject,  offering  a  new  method  of  solution, 
was  presented  by  Dr.  C.  T.  Hutchinson  before  the  A.I.E.E., 
February,  1914,  and  the  reader  is  referred  to  the  same  for  further 
information.^ 

The  size  of  the  auxiliary  station  is  determined  by  the  differ- 
ence between  the  demand  curve  and  the  stream-flow  curve, 
except  where  storage  is  available,  in  which  the  stream-flow  as 
affected  by  the  same  should  be  used. 

In  order  to  obtain  the  best  results,  the  method  of  operation 
also  deserves  a  careful  consideration.  In  this  connection  R.  C. 
Muir  in  the  General  Electric  Review  for  June,  1913,  makes  the  fol- 
lowing recommendations:  "  In  order  to  get  the  best  economy  out 
of  the  steam  station  it  must  operate  at  practically  a  constant  load 
corresponding  to  full  load  on  one  or  more  units.  In  order  to  get 
the  best  economy  out  of  the  water-power  station  with  the  water 
available  during  low-water  periods,  the  highest  water  level  attain- 
able— ^in  other  words  the  maximum  head — ^must  be  maintained  at 
all  times. 

"  It  is  impossible  to  conform  to  both  of  these  requirements, 
especially  where  the  minimum  stream-flow  capacity  and  the 
steam-station  capacity  combined  are  not  suflScient  to  cany  the 
•  peak  load.  In  this  case  the  steam  plant  can  be  operated  at  prac- 
tically a  constant  load,  using  the  water  power  during  the  peaks 
and  storing  water  during  the  balance  of  the  time.  With  high- 
head  plants  the  head  gained  by  storage  is  not  of  importance; 
so  that  the  steam  plant  can  be  operated  most  economically  on 
constant  load,  allowing  the  water  power  to  take  the  peaks.  With 
low-head  plants  having  considerable  storage  capacity  both  plants 
can  be  operated  advantageously  dm-ing  the  low-water  period. 
Here  again  the  water  power  should  carry  the  peaks,  and  the  steam- 
plant  operated  at  constant  load'  over  a  suflScient  part  of  the  day 
so  that  the  water  level  will  not  be  materially  affected.  This 
method  of  operation  will  prove  much  more  economical,  both  as 
regards  fuel  used  and  labor  required,  than  the  method  of  carrying 

» See  also  an  article  by  H.  S.  Putnam,  A.I.E.E.  June,  1917. 


Digitized  by 


Google 


AUXILIARY  STATIONS  697 

heavy  loads  on  the  steam  plant  during  the  peaks,  thereby  requir- 
ing more  boilers  and  machines  in  service  and,  consequently,  more 
fuel  and  operators. 

"  The  term  '  peaks '  is  intended  to  cover  heavy  load  periods 
of  the  daily  load  curve,  and  not  the  momentary  load  fluctuations. 
Assuming  equal  governor  or  speed  regulation  and  equal,  fly  wheel 
effects,  these  momentary  load  fluctuations  are  divided  between 
the  stations  in  proportion  to  the  total  capacities  of  the  generators 
operating  in  each  station.  The  flywheel  effect  of  the  steam  tur- 
bine is  usually  the  larger  and  the  steam  turbine  governor  is  the 
more  sensitive.  The  steam  turbine  station  will,  therefore,  ordi- 
narily take  more  of  the  momentary  fluctuation  than  its  propor- 
tionate capacity  in  operation. 

"  Some  fuel  can  be  saved  in  developments  of  this  kind  by 
carefully  observing  the  rainfall  within  the  drainage  area  of  the 
stream  developed.  In  case  of  racinfall  within  this  area  the  steam 
plant  can  be  shut  down  inmiediately  and  all  the  load  taken  over  by 
the  hydraulic  plant  at  the  expense  of  reducing  the  level  of  the 
reservoir.  The  increased  stream-flow  will  again  fill  the  reservoir. 
Rainfall  at  the  head  waters  of  a  large  stream  would  not  materially 
increase  the  stream-flow  at  the  development  for  some  time;  and, 
consequently,  a  considerable  saving  in  fuel  would  thus  be  effected. 
During  the  dry  season,  water  flowing  over  the  dam  means  fuel 
wasted^  and,  therefore,  if  enough  reliance  could  be  placed  in 
weather  forecasts  to  anticipate  rainfalls,  the  steam  plant  could  be 
shut  down  in  time,  so  that  the  reservoir  level  would  be  reduced 
suflSciently  to  take  care  of  the  increased  flow  without  wasting  any 
more  over  the  dam  than  necessary." 

Peak-load  Stations.  The  function  of  the  auxiliary  plant  used 
to  carry  the  daily  peaks  of  load  on  the  system  is  similar  to  that  of 
pondage  above  the  water-power  plant,  increasing  the  operating 
load  factor  and,  consequently,  the  output  from  water.  In  the 
case  of  the  supplemental  plant,  the  first  cost  and  relative  economy 
of  generation  must  be  governed  by  the  proportion  of  the  total 
output  of  the  system  to  be  carried  by  the  auxiliary  plant,  i.e., 
the  higher  the  percentage  carried  by  the  auxiliary  plant,  the 
more  important  becomes  the  economy  of  generation  and  the  less 
imj)ortant  the  first  cost  and  resulting  fixed  charges. 

Base-load  Stations.  Where  the  conditions  are  such  that  the 
average  power  demand  exceeds  the  capacity  of  the  hydraulic 


Digitized  by 


Google 


698  ECONOMICAL  ASPECTS 

plant  it  is  usually  preferable  to  operate  the  auxiliary  steam  plant 
continuously,  the  water  power  then  being  supplemental  to  the 
steam  power.  Low  operating  costs  are  essential  for  this  type  of 
plant  and,  as  far  as  the  operation  is  concerned,  the  reconmienda- 
tions  given  for  the  low-water  plants  also  apply  in  this  case. 

INTERCONNECTION  OF  SYSTEMS 

The  interconnection  of  hydro-electric  transmission  systems 
is  also  a  step  in  the  right  direction,  as  demonstrated  in  our  South- 
em  States,  where  not  less  than  seven  large  systems  are  tied  to- 
gether, furnishing  power  to  each  other  on  an  "  interchange  "  con- 
tract basis.  The  advantages  of  this  are  obvious.  The  peak  loads 
of  the  different  systems  may  not  coincide,  the  minimum  stream- 
flow  may  occur  at  different  times  on  the  different  watersheds, 
conunon  steam  reserve  stations  may  be  used,  and,  in  general, 
the  operation  may  be  so  improved  that  a  most  efficient  and  re- 
Uable  service  can  be  rendered  to  the  customers  of  all  the  systems 
so  tied  together. 

In  some  cases  groups  of  established  syBtems  although  located 
in  vastly  different  locaUties  may  be  brought  together  under  one 
holding  company,  and  to  the  creation  of  such  companies  may, 
in  many  instances,  be  attributed  the  high-class  service  and  finan- 
cial success  of  our  small  and  medium-size  light  and  power  systems. 
The  economies  due  to  a  central  management,  the  benefit»of  the 
beK^t  technical  and  expert  advice  applied  even  to  the  smallest 
central  station,  the  cumulative  effect  of  active,  up-to-date  new- 
buaness  campaigns  at  every  j)oint,  all  have  contributed  to  an 
unpi.')ved  and  cheaper  service  to  the  consumer,  and  without  the 
faciUti3s  of  such  a  control  they  could  exist  only  in  the  larger  com- 
mimitieb.  Another  very  important  advantage  is  the  great  prob- 
lem of  financing  all  these  undertakings  and  providing  funds  for 
extensions  to  meet  the  ever-growing  demand  of  the  pubUe  for 
electric  service.  It  is  possibly  in  providing  ready  financial  facil- 
ities for  these  purposes  that  the  holding  company  performs  its 
most  important  function. 

In  order  to  give  the  people  the  best  service  and  the  lowest  rates 
all  public  utiUties  must,  of  necessity,  be  natural  monopolies,  and 
the  pubUc-service  regulation  is  a  recognition  by  the  State  of  the 
essentially  monopolistic  character  of  these  enterprises.  The 
favorable  showing  of  virtual  monopoUes  in  reducing  the  cost  of 


Digitized  by 


Google 


INVESTIGATION  OF  AN  ENTERPRISE  699 

electric  power  is  due  mainly  to  a  reduction  in  the  capital  expenses, 
lower  operating  costs,  and  in  no  less  degree  to  the  reduced  risk 
to  the  investor.  By  effective  safeguards  and  a  well-considered 
policy  of  public  control  the  electric  securities  have  become  one  of 
the  most  desirable  investments,  and  there  is  every  indication  that 
efficient  public-service  regulation  will  make  possible  even  further 
reductions  in  the  cost  of  electric-power  production  of  public-service 
utilities. 

mVESTIGATION  OF  AN  ENTERPRISE 

The  following  points  cover  broadly  the  important  items  upon 
which  an  investor  must  have  information  in  order  to  judge  in- 
telligently of  an  oflfering  to  finance  an  enterprise,  and  for  a  more 
complete  treatise  of  the  subject  the  reader  is  referred  to  Francis 
Cooper's  book,  *'  Financing  an  Enterprise." 

I.  Nature  of  Enterprise. 

1.  Is  the  basis  of  the  enterprise  sound? 

2.  Is  the  business  or  undertaking  profitable  elsewhere? 

3.  What  competition  or  opposition  will  be  met? 

4.  What  peculiar  advantages  does  it  enjoy  over  these 

others? 

5.  Can  it  be  conducted  profitably  under  existing  condi- 

tions? 

II.  Plan  of  Organization. 

1.  In  what  state  organized? 

2.  What  is  the  capitalization? 

3.  Is  the  capitalization  reasonable? 

4.  Has  the  stock  been  issued  in  whole  or  in  part  and,  if 

so,  for  what? 

5.  Is  the  stock  offered  for  sale  full-paid  and  non-assessable? 

6.  Has  any  of  the  stock  preferences? 

7.  Is  any  of  the  stock  unissued  or  held  in  the  treasury? 

8.  Who  has  stock  control? 

9.  Are  the  rights  of  smaller  stockholders  protected? 

10.  Are  there  any  imusual  features  in  charter  or  by-laws? 

m.  Present  Condition  of  ErUerprise. 
As  to  Property: 

1.  What  properties  or  rights  are  controlled? 

2.  What  is  their  value  and  how  estimated? 


Digitized  by 


Google 


700  ECONOMICAL  ASPECTS 

3.  Are  these  properties  or  rights  owned,  or  held  under 

lease^  license,  grant,  option  or  otherwise? 

4.  If  owned,  are  titles  perfect? 

5.  Are  there  any  incumbrances  on  the  properties  or 

rights? 

6.  If  not  owned,  are  the  holding  papers  in  due  form? 

7.  If  not  owned,  are  the  terms  of  holding  reasonable, 

satisfactory  and  safe? 

8.  In  event  of  liquidation,  what  would  be  worth  of 

property? 

As  to  Operation: 

1.  What  operations  have  been  or  are  now  carried? 

2.  What  have  been  the  results? 

3.  What  difficulties,  if  any,  have  been  encountered? 

4.  What  is  demand  for  the  product  or  operation  of 

the  enterprise? 

5.  What  is  present  status  of  the  enterprise? 

6.  Are  proper  books  kept? 

As  to  Finance: 

1.  What  are  the  present  assets  and  their  actual  value? 

2.  What  debts,  claims,  fees,  rents,  royalties  or  other 

payments  or  obligations  are  now  due  or  are  to 
be  met  and  carried? 

3.  From  what  resom^ces  are  these  to  be  met? 

4.  Who  handles  the  moneys  and  under  what  safe- 

guards? 

5.  What  are  or  will  be  the  running  expenses,  salaries, 

etc.? 

IV.  Management. 
Directors: 

1.  How  many  members  in  the  board? 

2.  Who  are  these  members? 

3.  What  is  their  past  record  and  present  business 

status? 

4.  Who  are  the  active  members  of  the  board? 

5.  Who,  if  any,  are  inactive? 

6.  Are  meetings  regularly  held  and  attended? 

7.  Who  compose  the  Executive  Committee,  if  any, 

and  what  are  its  powers? 


Digitized  by 


Google 


INVESTIGATION  OF  AN  ENTERPRISE  701 

8.  Are  the  directors  stockholders  and,  if  so,  to  a 
material  amount? 

Officers: 

1.  Who  are  the  officers? 

2.  What  are  their  previous  records? 

3.  What  are  their  special  present  qualifications? 

4.  Are  they  able  to  work  together  without  friction? 

5.  What  compensation  do  they  receive  or  are  they 

to  receive? 

6.  Are  they  interested  in  the  enterprise  beyond  their 

salaries? 

V.  Plan  of  Operation. 

1.  What  is  the  general  plan  of  operation? 

2.  What  special  reasons,  if  any,  led  to  its  adoption? 

VI.  Disposition  of  Money  Asked  for. 

1.  Does  the  money  from  sale  of  stock  go  into  the 

treasury  of  the  company? 

2.  If  any  does  not  go  into  the  treasury,  to  whom  does  it 

go,  and  for  what  purpose? 

3.  Of  money  going  into  the  treasury,  what  proportion 

goes  into  active  development  and  operation? 

4.  What  part  goes  to  pay  off  existing  debts,  obligations 

and  claims? 

5.  What  part,  if  any,  goes  to  pay  for  promotion  expenses, 

conmiissions,  etc.? 

6.  How  is  the  development  and  operating  money  to  be 

applied? 

7.  Is  the  amount  asked  for  sufficient  to  accomplish  the 

desired  results? 

8.  Will  it  place  the  company  on  a  self-supporting  or 

profitable  basis? 

VII.  The  Proposition. 

1.  Is  the  general  proposition  a  fair  one? 

2.  Is  the  price  of  stock  or  bonds  reasonable? 

3.  How  do  these  prices  compare  with  any  former  prices? 

4.  If  common  stock  is  offered,  do  preferred  stock,  bonds 

or  other  profit-sharing  obligations  take  precedence 
and  to  what  amount? 


Digitized  by 


Google 


702  ECONOMICAL  ASPECTS 

5.  What  reserve  of  profits  will  be  retained  before  divi- 

dends are  to  be  declared? 

6.  If  preferred  stock  is  offered,  is  it  cumulative,  does  it 

vote,  when  is  it  redeemable,  and  at  what  price, 
what  sinking  fimd  provision  is  made  for  redemption 
and  are  any  peculiar  provisions  attached?  Do  any 
bonds  or  other  obligations  take  precedence  of  the 
preferred  stock? 

7.  If  bonds  are  offered,  what  interest  is  paid,  and  when 

and  where;  uj)on  what  property  are  they  secured 
and  when  and  how  are  they  paid;  is  the  trustee 
or  trust  company  of  repute;  imder  what  conditions 
are  the  bonds  foreclosable;  when,  and  how  are  they 
or  may  they  be  redeemed;  are  there  any  other 
securities  taking  precedence,  and  are  there  any 
peculiar  provisions  in  deed  of  trust? 

VIII.  General. 

1.  What  is  the  previous  history  of  the  enterprise  or 

the  property  or  undeiiiaking  on  which  it  is  based? 

2.  If  inventions  enter  prominently,  what  is  the  pre- 

vious record  of  the.  inventor? 

3.  By  whom  are  the  statements  made  and  is  the  party 

making  them  reliable? 

4.  Are  there  any  contracts  or  obligations  not  now  effec- 

tive by  which  the  enterprise  will  subsequently  be 
affected? 

COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 

The  cost  of  water  power  depends  upon  a  great  variety  of 
factors,  the  essential  feature  of  the  design  of  the  plant  being  to 
keep  the  cost  within  reasonable  limits,  so  that  the  fixed  charges, 
which  constitute  by  far  the  greatest  part  of  the  power  cost,  shall  not 
be  excessive.  The  allowable  cost  of  a  water  power  can  obviously 
not  be  more  than  the  cost  of  producing  the  same  amount  of  power 
by  some  other  means,  usually  steam.  The  cost  of  generating  the 
power  should,  furthermore,  not  be  confused  with  the  cost  of  power 
deUvered.  Besides  the  cost  of  producing  the  power  in  the  gen- 
erating station  comes  the  expenses  involved  in  distributing  the 
same  to  the  customers,  which  often  amount  to  several  times  that 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  703 

of  the  former,  especially  with  hydraulic  developments  where  the 
power  must  be  transmitted  for  great  distances  at  high  voltages 
to  the  market  center  and  there  stepped  down  to  a  moderate  dis- 
tributing voltage  and  again  at  the  point  of  utilization  to  the  volt- 
age required  for  the  power  or  lighting  load.  It  is  the  costs  of 
these  transformations,  transmission  and  distributions,  besides  the 
general  expense,  which  makes  the  cost  of  power  to  the  customer 
so  much  higher  than  the  cost  of  actually  producing  the  power  at 
the  generating  station  bus-bars. 

The  cost  of  the  plant  varies  through  the  widest  possible  limits, 
depending  on  its  location  as  regards  facilities  for  construction  and 
for  transmission,  the  quantity  of  water  and  regularity  of  flow, 
the  total  head,  conditions  of  the  labor  market,  both  as  to  quality 
and  supply,  etc. 

There  are  usually  more  elements  of  chance  and  more  unknown 
factors  in  a  hydraulic  development  than  in  a  steam  plant,  and 
these  facts  should  be  taken  into  consideration  and  properly  cared 
for  in  making  up  the  cost  estimate.  In  many  instances  cost 
figures  must  be  obtained  from  similar  work  under  similar  condi- 
tions, and  the  dependence  to  be  placed  on  the  source  of  informa- 
tion must  be  duly  considered  and  weighed.  Each  case  must  be 
carefully  examined  and  studied  from  the  conditions  bearing  directly 
upon  it  and  the  deductions  made  accordingly.  For  a  very  com- 
plete classification  of  the  construction  and  operating  accounts  the 
reader  is  referred  to  the  report  by  the  N.E.L.A.  Accounting 
Committee  for  1914. 

The  total  cost  of  a  hydro-electric  plant  may  be  properly  divided 
into  three  parts,  viz. : 

1.  Development  expenses. 

2.  Physical  costs. 

3.  Overhead  charges. 

Development  Expenses.  These  include  all  of  the  preUminary 
expenses  incidental  to  the  building  up  of  the  project  and  which 
are  not  directly  involved  in  the  actual  construction  of  the  prop- 
erty. They  include  expenditures  on  account  of  promotion,  in- 
corporation and  organization,  condemnation  and  other  legal 
expenses  as  well  as  cost  of  surveys,  expert  estimates,  etc. 

The  cost  of  securing  money  is  also  an  important  item  in  the 
development  of  a  property.    By  this  is  not  meant  the  interest 


Digitized  by 


Google 


704  ECONOMICAL  ASPECTS 

and  dividends  which  are  paid  on  the  securities  of  the  company  to 
the  stockholder  and  bondholder  and  which  are  essential  to  make 
future  issues  marketable,  but  we  are  dealing  with  the  actual  costs 
to  the  utility  of  placing  its  securities  in  the  hands  of  the  public. 
This  cost  of  securing  the  money  should  be  distinguished  from  pro- 
moters' services  and  from  bond  discount.  The  latter  is  an 
adjustment  between  the  amount  paid  by  the  public  for  the  bond 
and  its  face  value,  due  to  the  difference  of  the  interest  rate  of  the 
bond  and  the  interest  rate  prevailing  at  the  time  of  the  sale  of  the 
bond,  and  it  may  occur  a  number  of  times  during  the  life  of  the 
corporation.  The  cost  of  securing  money  is  a  very  different  thing, 
and  only  comes  once — when  the  original  capital  is  acquired.  That 
such  costs  are  legitimate  and  must  be  recc^nized  cannot  fairly 
be  denied.  The  existence  of  numerous  banking  and  brokerage 
houses  specializing  in  public-utility  securities  shows  that  it  costs 
to  secure  money  just  as  to  purchase  generators,  cable,  land  or 
any  of  the  tangible  construction  elements  of  a  property. 

The  losses  inciured  in  the  sale  of  securities,  that  is,  brokerage 
and  discoimts,  should,  of  course,  also  be  included. 

The  development  expenses  will  sometimes  amount  to  as  high 
as  20  per  cent  of  the  cost  of  the  physical  plant,  depending,  of 
course,  on  the  attractiveness  of  the  undertaking  and  the  rate  at 
which  the  securities  can  be  disposed. 

Physical  Costs.  Thes^  should  cover  the  actual  costs  of  con- 
structing the  plant,  including  material,  apparatus  and  labor. 
The  cost  of  each  unit  of  the  plant  elements  in  its  final  position  is 
composed  not  only  of  its  first  cost  but  of  all  other  items  of  expense 
which  are  necessarily  involved.  These  may  be  any  or  all  of  the 
following:  Freight,  storehouse  cost,  inspection,  assembling  or 
fitting,  transportation  from  storehouse  to  work  and  distribution, 
labor  of  placing  element  in  position,  transportation  of  men  and 
tools  to  work,  lost  time  of  men  during  travel  or  wet  weather, 
losses  on  tools  and  material.  After  the  cost  has  been  estimated 
as  closely  as  possible  it  has  become  an  accepted  rule  to  add  a 
general  percentage  of  the  same  to  cover  contingencies,  omissions 
and  errors.  This  percentage  is  frequently  estimated  as  10  per 
cent  and  sometimes  higher,  depending  on  the  imcertainties 
involved  in  the  proposition. 

The  physical  equipment  includes: 
Land  and  water  rights. 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  705 

Hydraulic  construction: 
Dam,  intake,  forebay,  water  conductors,  etc. 

Generating  station: 
Building,  hydraulic  and  electric  equipments,  etc. 

Transmission  lines. 

Substations. 

Distributing  system. 

Auxiliary  steam  station. 
Overhead  Charges.  Besides  the  above  expenses  for  the  de- 
velopment and  actual  construction  of  the  physical  plant,  there 
are  others  which  must  be  considered  as  a  part  of  the  total  cost  of 
any  complete  development.  These  are  termed  overhead  charges 
and  are  as  follows: 

Engineering  and  superintendence. 

Organization. 

Legal  expenses. 

Taxes  and  insurance. 

Interest  during  construction. 

Working  capital. 
Engineering  and  superintendence  should  cover  all  costs  of 
architecture  and  engineering.  This  includes  all  designs  and 
drafting,  plans  and  supervision  of  construction,  as  well  as  all 
other  items  which  properly  come  under  this  department.  They 
vary  from  3  to  5  per  cent  of  the  construction  cost. 

Organization  expenses  should  cover  the  cost  of  organization 
and  administration  for  construction,  including  general  office 
expenses.    They  generally  amount  to  from  3  to  5  per  cent. 

Legal  expenses  incurred  during  the  construction  period  should 
be  distinguished  from  those  included  under  development  expenses. 
They  should  cover  only  such  legal  work  which  may  be  necessary 
in  obtaining  such  rights  as  may  be  needed  to  carry  out  the  con- 
struction. 

Taxes  must  be  paid  on  the  property  from  the  time  of  purchase, 
usually  months  or  even  years  before  the  development  is  com- 
pleted. Likewise  insurance  must  be  paid  and  should  include 
not  only  fire  insurance,  but  casualty  insurance,  covering  both 
employees  and  public  liability.  The  estimate  of  these  expenses 
can  be  accurately  made  from  prevailing  rates.  Taxes  amount  to 
about  one-half  of  1  per  cent  and  insurance  about  the  same  amount. 
Interest  during  construction  accruing  on  the  idle  capital,  rep- 


Digitized  by 


Google 


706  ECONOMICAL  ASPECTS 

resented  either  by  cash  or  plant,  must  be  included  in  the  estimate. 
The  length  of  time  for  which  to  compute  the  same  will  naturally 
vary  with  the  time  required  for  the  construction,  but  as  a  rule  it  is 
figured  at  the  full  annual  rate  for  one-half  the  construction  period. 

Working  capital  of  a  reasonable  amount  must,  of  course,  be 
provided  for  carrying  on  the  business  and  must  be  considered  as  a 
part  of  the  property. 

From  the  above  it  is  seen  that  the  overhead  charges  form  a  large 
percentage  of  the  cost  of  developing  a  system  and  it  may  approx- 
imately be  taken  as  from  20  to  30  per  cent  of  the  physical  cost. 

Cost  data  on  hydro-electric  plants  are  scarce,  and  when  ob- 
tained the  greatest  caution  must  be  exercised  in  using  them  for 
estimating  other  projects.  They  are  greatly  affected  by  local 
conditions,  as,  for  example,  the  nature  of  the  soil  in  determining 
the  cost  of  excavation,  the  price  paid  for  labor,  freight  and  trans- 
portation charges,  market  value  of  raw  and  other  material,  appa- 
ratus, etc. 

In  order,  however,  to  give  the  reader  an  approximate  idea  of 
the  costs  involved,  the  following  figures  are  given.  They  are 
based  both  on  actual  costs  and  on  estimates  under  normal  con- 
ditions, but  the  authors  wish  again  to  repeat  their  caution  as 
to  a  careful  discrimination  of  then:  use. 

Estimated  Cost  of  600  Kw.  Hydro-electric  Power  Station 

It  is  proposed  to  install  two  units,  each  comprising  a  500-H.P. 
turbine  operating  under  a  60-foot  head  and  driving  a  300-Kw. 
generator.  Two  separately  driven  exciter  units  and  complete 
switching  equipment,  but  no  step-up  transformers.  The  dam  is 
already  provided  and  is  not  included  in  the  estimate. 

Penstock  and  flume,  including  headworks,  connections,  tunnel,  etc.  $22,500 

Regulating  tank,  including  housing 1,500 

Power  station;  foundation  and  buildings  complete  with  interior  work 

and  fittings 9,800 

Staff  house  and  miscellaneous 3,000 

Equipment  in  power-house,  consisting  of  two  500-H.P.  turbines  with 

governors,  generators,  exciters,  switching,  equipment,  etc 30,200 

Total $67,000 

Add  for  contingencies,  engineering,  supervision  and  inspection,  12 

per  cent,  say $8,000 

Grand  total $76,000 


Digitized  by 


Google 


.COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 


707 


Annual  Cost  op  Opbkation 
Overhead  charges: 

Yearly  installment  of  principal  and  interest.    Debenture 

to  be  retired  in  thirty  years  at  5  per  cent S4,875 

Maintenance  account,  being  an  amount  set  aside  yearly 
against  major  repairs,  renewals  and  reasonable  ex- 
tensions, 2}  per  cent 1,875 

$6,760 

Operating  charges: 

Salary,  superintendent  and  general  office  expenses $2,000 

Wages  of  operators  at  power  station 2,200 

Supplies  and  minor  repairs 900 

$5,100 

Total  annual  cost $11,850 

Or  approximately  $20  per  Kw.-year. 

Municipal  Hydro-electric  Plant  of  City  op  Sturgis,  Mich. 
Capacity,  1100  Kw. 
This  development  consists  of  a  hollow  reinforced  concrete 
spillway  dam,  308  feet  long  and  24  feet  high.  This  spillway  con- 
nects with  an  earth  embankment  400  feet  long  and  24  feet  high. 
The  power-house  contains  two  550-Kw.  2300-volt  generators 
driven  by  two  844-H.P.  turbines,  and  a  40-Kw.  exciter  driven  by 
a  67-H.P.  turbine.  The  head  is  22  feet.  Six  200-Kw.  oil-cooled 
transformers  for  stepping  up  the  voltage  to  22,000  are  provided, 
also  complete  switching  equipment  and  Ughtning  arresters.  The 
ultimate  development  will  include  two  additional  generating  units 
and  one  additional  exciter. 

Cost  Data  Based  on  Ultimate  Development 


Items. 


Total  Cost. 


Cost  per 

H.P.  at 

Wheel  Shaft. 


Cost  per  Kw. 
at  Switch- 
board. 


Power-house  and  machinery 

Spillway 

Tailrace 

Embankment 

Bridge  changes 

Transmission  line 

Real  estate 

Substation  and  incidentals . 

Totals 


$110,000 
22,000 
20,000 
8,000 
8,000 
20,000 
50,000 
12,000 


$32.56 
6.50 
5.93 
2.36 
2.36 
5.93 
14.81 
.3.55 


$45.90 
9.16 
8.36 
3.33 
3.33 
8.37 
20.90 
5.01 


$250,000 


$74.00 


$104.36 


Digitized  by 


Google 


708  ECONOMICAL  ASPECTS 

Actual  Cost  of  a  4800-H.P.  Development  Operating  tjndeb 

90  Feet  Head 

This  plant  was  designed  to  utilize  the  water  flowing  from  a 
large  storage  reservoir  primarily  built  for  domestic  and  industrial 
service.  It  comprises  four  48-inch  cast-iron  penstocks  dischai^g 
into  four  1200-H.P.  horizontal  turbines,  each  direct-connected  to 
a  1000-Kv.A.  (800-Kw.  0.8  P.F.),  60-cycle,  13,200-volt  generator 
operating  at  a  speed  of  400  R.P.M.  The  exciter  equipment  con- 
sists of  two  60-Kw.  turbine-driven  units. 

The  first  cost  of  the  installation  was  $227,474,  itemized  as 
follows: 

Station  building $113,786 

Foundations  of  turbines  and  generators 7,883 

Total  station  cost $121,769 

Turbines  and  generators $70,574 

Labor  and  materials 5,043 

Penstocks  and  valves 1,375 

Venturi  meters 6,212 

Traveling  crane 2,500 

Total  equipment $99,704 

Lightning  arresters  and  outgoing  line  equipment 6,001 

Total $227,474 

Per  H.P $47.50 

Per  Kw 71.00 

Fixed  Charqes  and  Operating  Expenses  (Yearly) 

Labor,  1  electrical  engineer,  1  operator,  2  helpers,  1  helper  part  time  $5,531 

Fuel  for  heating  building 86 

Repairs  and  appliances 354 

Oil  and  waste 87 

Small  supplies 262 

Taxes 2,675 

Interest  at  6  per  cent 11,374 

Depreciation,  station  and  machinery,  4  per  cent 4,475 

Depreciation  on  transmission  equipment,  2  per  cent 120 

Total $24,964 

Daily  output  in  kilowatt-hours 18,000 

Total  cost  per  kilowatt-hour 0.46  cent 


Digitized  by 


Google 


^COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  709 

Estimated   Cost  op   a  6000-H.P.  Development  Operating 

UNDER  A  27-foot  HeAD 

This  development  is  assumed  to  comprise  two  3000-H.P. 
vertical-shaft  turbines  driving  two  2600-Kv.A.  (2000-Kw.,  0.8 
P.F.)  2300-volt  generators  operating  at  a  speed  in  the  neigh- 
borhood of  75  to  80  R.P.M.  Two  three-phase  transformer  units 
of  capacities  corresponding  to  the  generators  are  provided,  the 
high-tension  trajismission  voltage  being  33,000.  Provision  is 
also  made  in  the  building  for  future  installation  of  a  third  gener- 
ator as  well  as  a  transformer  unit.  It  is  intended  that  this  plant 
is  to  be  erected  in  connection  with  an  existing  dam  on  a  navigable 
stream,  thus  doing  away  with  the  necessity  of  any  pipe  Une  or 
similar  structures  to  carry  the  water  to  the  power-house;  neither 
do  the  figures  include  any  allowance  for  dam  or  spillway. 

Cost  Estimate 

Electrical  equipment $80,000 

Delivery  and  erection 7,500 

$87,500 

Hydraulic  equipment 55,000 

Delivery  and  erection 5,000 

60,000 

50-ton  crane,  oil  and  water  piping  and  misc.  equipment 
in  place 8,500  8,500 

Concrete  foundations,  hydraulic  tubes,  headrace,  etc . . .      55,000 

Building,  exclusive  of  foundation 32,000 

Excavation 6,000 

93,000 

5  miles  double-circuit  line  on  steel  towers 35,000  35,000 

Contingencies  (10  per  cent) 28,400 

Interest  and  insurance  during  construction 15,000 

Engineering  and  superintendence 20,000 

Total $347,400 

.     •         Per  H.P $58.00 

Per  Kw 87.00 

Estimated  Cost  of  a  6000-Kw.  Hydro-electric  Development 
Operating  under  a  47-Foot  Head 

This  development  contemplates  a  hollow  reinforced  con- 
crete dam,  465  feet  long  and  about  55  feet  high,  including  spiUing 
and  sluiceways.  An  intake  stnicture  with  controlling  devices  is 
to  be  provided  in  connection  with  the  dam  and  the  water  is  from 


Digitized  by 


Google 


710  ECONOMICAL  ASPECTS 

there  to  be  led  through  an  open  concrete-lined  canal,  2600  feet 
long  and  with  a  cross-sectional  area  of  525  square  feet,  to  a  fore- 
bay.  The  forebay  is  divided  in  three  sections  provided  with 
gates  and  trash  racks,  and  there  will  be  three  penstocks,  10  feet 
6  inches  in  diameter  and  265  feet  long. 

The  power-house  equipment  comprises  three  3500-H.P.  tur- 
bines with  governors,  driving  three  2000  Kw.  generators  with 
direct-connected  exciters.  Provision  is  also  made  for  trans- 
formers, switching  equipment  and  necessary  station  auxiliaries, 
such  as  crane,  etc. 

Estimated  Cost  of  Plant 

Main  dam  and  headworks $313,660 

Canal,  including  lining 62,000 

Forebay 23,000 

Penstocks 35,750 

Power-house 61,000 

Machinery: 

Turbines  and  governors 42,000 

Generators  and  exciters 52,000 

Transformers  and  switching  apparatus 36,000 

Total $626,410 

Engineering  and  contingencies $d4,690 

$720,100 

Interest  during  construction 28,000 

Grand  total $748,100 

The  total  capital  cost  of  the  plant,  including  the  proportion  of 
the  cost  of  creation  of  storage,  also  the  proportion  of  the  cost  of  a 
duplicate  transmission  line,  and  proportion  of  a  transformer  sta- 
tion and  equipment  is: 

Capital  cost  of  plant $748,100.00 

Transmission  lines  and  station  equipment 64,700.00 

Storage 103,000.00 

Total  capital  cost $916,800.00 

Annual  charges: 

1.  Interest  on  capital  invested  assuming  financing  is  done 

on  bonds  at  6  per  cent  sold  at  par $54,900.00 

2.  Sinking  fund  to  retire  bonds  in  thirty  years  reinvested 

at  4  per  cent,  say  1}  per  cent 16,050.00 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 


711 


3.  Depreciation  on  plant  adjusted  between  general  works 

and  equipment  to  provide  for  major  repairs  and  re- 
newals    $13,700.00 

4.  Operation    and    maintenance,  including    management, 

superintendence,  wages  for  operators  of  plant,  trans- 
mission line,  receiving  station,  storage  regulation, 
minor  repairs,  supplies,  and  upkeep,  etc 20,650.00 

Total  annual  charges $105,300.00 

Cost  per  Kw.  year,  delivered 17.50 

Cost  op  the  Minidoka  Power  Station  of  United  States 
Reclamation  Service 

Capacity,  7000  Kv.A. 

The  power-house  is  a  reinforced  concrete  structure  with  steel 
roof  trusses  and  purlins  covered  by  matched  liunber  and  galvanized 
corrugated  iron.  It  measures  149  feet  in  length,  50  feet  in  width 
and  90  feet  in  height.  It  contains  five  2000  H.P.  main  turbines 
operating  imder  a  head  of  46  feet,  driving  five  1400-Kv.A.  2200- 
volt  generators  at  a  speed  of  200  R.P.M.  There  are  also  two  180- 
H.P.  turbine-driven  exciters  and  each  main  generator  is  directly 
connected  to  a  three-phase  transformer,  stepping  up  the  voltage 
to  33,000.  Complete  switching  and  Ughtning-arrester  equip- 
ment is  included  in  the  estimate,  but  no  allowance  is  made  for  the 
dam,  this  forming  part  of  the  irrigation  system. 


Cost  of  Power-house 


Total  Cost. 

Cost  per  Kw. 

Building.      .       , 

$82,000 
73,000 
83,000 
26,200 
55,500 
60,000 
7,300 
23,200 
11,100 
15,000 

$11.70 

Hydraulic  machinery 

10.40 

T*^lf?ctric  machinery 

11.80 

^^ight  and  hauling, , . 

3.75 

Erection 

7.90 

Tailrace 

8.60 

Roads  and  telephone  lines 

1.10 

Camo  and  oermanent  Quarters 

3.35 

Engineering  and  incidentals 

1.55 

Administration  charges,  etc 

2.15 

Total 

$436,300 

$62.30 

Digitized  by 


Google 


712 


ECONOMICAL  ASPECTS 
Annual  Cost  op  Operation 


Item. 

Operation: 

Labor 

Supplies , 

Repairs: 

Labor , 

Supplies  and  material 

Superintendence,  clerical,  camp,  etc. 
General  expense  and  administration 

Operating  expense 


Expense  per  Year. 


$5,700 
950 

900 

300 

1,700 

450 

$10,000 


A  depreciation  of  5  per  cent  ($21,800)  has  also  been  charged  to 
this  development.  No  taxes  or  interest  is  charged,  the  under- 
taking being  done  by  the  Government.  Assuming  7  per  cent 
for  interest  and  taxes  the  total  operating  expenses  would  amount 
to  $62,000.  A  total  of  about  15,000,000  Kw.  hr.  were  delivered 
during  one  year,  thus  corresponding  to  a  cost  of  $0.0041  per 
Kw.hr. 

Actual  Cost  op  20,000-Kv.A.  Hydro-electric  Power  Devel- 
opment OP  THE  City  op  Tacoma,  Washington 

This  development  comprises  a  concrete  dam  approximately 
45  feet  high  and  a  spillway  of  260  feet.  Intake,  racks,  regulating 
gates  and  a  settling  channel,  the  latter  being  780  feet  long,  40 
feet  wide  and  20  feet  deep.  From  the  settling  basin  the  water  is 
carried  through  an  8  X  8-foot  tunnel,  10,000  feet  long,  to  a  reg- 
ulating reservoir  approximately  500X500  feet,  having  a  capacity 
of  about  3,000,000  cubic  feet  available  for  use  during  peak  loads. 
Each  main  turbine  has  a  separate  riveted-steel  penstock  about 
780  feet  long  and  ranging  in  size  from  78  inches  at  the  top  to  48 
inches  at  the  gate  valves  in  front  of  the  turbines.  The  two 
exciter  wheels  are  supplied  from  one  24-inch  pipe  which  divides 
in  the  generator  room. 

The  power-house  consists  of  three  buildings  of  the  common 
wall  type  of  construction  of  concrete  and  brick,  with  galvanized- 
iron  roof  supported  by  steel  roof  trusses.  There  are  four  8000- 
H.P.  horizontal  main  turbines  operating  under  a  415-foot  effective 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  713 

head  at  460  R.P.M.,  driving  four  5000  Kv.A.  three-phase,  60- 
cycle,  6600-volt  generators.  There  are  also  two  300-H.P.,  400- 
R.P.M.  turbines,  two  200-Kw.  125  volts  exciters,  and  twelve 
1667-Kw.-  gVoVo-volt  water-cooled  step-up  transformers  arranged 
in  four  banks.  Also  the  necessary  switching  and  Ughting  arrester 
equipment. 

The  entire  cost  of  the  development  was  as  follows: 

Genebatino  Plant 

Water  rights $30,000.00 

Hydro-power  plant,  land 168,696.50 

Building  fixtures  and  grounds 208,621.33 

Dam,  intake,  fliunes,  reservoirs,  penstocks 1,156,728.24 

Equipment 200,640.66 

Total $1,764,686.73 

Substation 

Equipment $85,577.20 

Building,  fixtures  and  land 110,619.40 

Total $196,196.60 

Tbansmibsion 

Land $66,226.65 

Equipment 118,193.23 

Simdry 2.89 

"■ 

Total $184,422.77 

Genebal  Expenditubes 
(During  Construction  of  Plant) 

OflSce  fumitiwe  and  fixtures $2,993.91 

Engineering  and  superintendence 95,866.87 

Injuries  and  damages 85.00 

Interest '. . .  83,860.47 

Miscellaneous 26,872.00 

Total $209,678.25 

Grand  total $2,354,984.35 

Cost  op  Hydro-electric  Plants 

E.  V.  Pannell  in  Electrical  News  for  February  15,  1917,  gives 
the  following  capital  cost  of  four  undertakings,  that  of  the  fifth 
being  estimated.  The  costs  are  separated  in  five  items,  which, 
for  comparison  are  also  shown  graphically  in  Fig.  405,  p.  716. 


Digitized  by 


Google 


714 


ECONOMICAL  ASPECTS 


Cost  op  City  op  Seattle  Municipal  Hydro-electric  Plant 

(Journal  Electricity,  Power  and  Gas,  July  18,  1914) 

Genbral  Costs 


DiviBlon  of  Plant. 

Cost. 

Cost  per  Kw.  on 

Basis  of  15.500  Kw. 

Capacity. 

Wood  crib  dam 

$61,863.80 
299,471.59 
354,387.44 
232,629.62 
242,096.21 
95,550.32 

$1,285,998.98 

$3.99 

Penstocks 

19.32 

Power  station 

22  86 

Transmission  lines 

15.01 

City  substations 

15.62 

6.16 

Total  generating  system 

$82.96 

Detail  Costs 


Division  of  Plant. 


Capacity, 
Kw. 


Cost. 


Unit  Cost, 
per  Kw. 


Wood  crib  dam 

Penstocks,  combined 

No.  1  Penstock,  complete 

15,407  ft.  49  in.  wood  stave  pipe,  com- 
plete in  place 

1,061  ft.  48  in.  steel  pipe,  308,000  lbs., 

complete 

16,468  lineal  ft.  grading  and  filling. . . 

No.  2  penstock,  complete 

15,865  ft.  68  in.  wood  stave  pipe,  com- 
plete  

1902  ft.  48  in.  steel  pipe,  with  Y-oon- 
nection,   valves  ana  cross-over  to 

smaUer  pipe 

Two  36-in.  standpipes,  65  and  70  ft. 

high 

16,816  lineal  ft.  grading  and  filling. . . 

Cedar  Falls  generating  station,  total 

Power-house   buildiings,    station,  switch 
house,  transformer  house  and  freight 

shed 

Emplovees'  cottages 

Two  8bOO-H.P.  turbines  with  hydraulic 
valves,  governors  and  relief  valves,  com- 
plete in  place 

Two  2400-H.P.  Pelton  wheels,  with  valves 

and  governors,  complete 

Two  5000-Kw.  generators,  oomplete  in 

place 

Two  1750-Kw.  generators,  complete  in 

place 

Two  75-Kw.  exciters,  with  Pelton  wheels 

and  governor 

One  150-Kw.  exciter  with  Girard  wheel . . 

Switchboard,  complete 

2300-volt  wiring,  busses  and  switches . . . 

Nine   150a-Kw.,  60,000-volt  transform- 

ers,  in  place 


9,000 

11,000 

3,600 


7,400 


13,500 


10,000 
3,500 

10,000 

3,500 

150 

150 

13,500 

13,500 

13,500 


$  61,863.80 

299,471.59 

84,475.79 

33,044.16 

14,386.01 

37,045.62 

214,995.80 

131,561.78 


19,587.27 

2,316.31 

61,530.44 

354,387.44 


47,829.77 
10,386.82 


53,296.55 

28,200.00 

39,422.00 

23,782.00 

5,383.00 

4,500.00 

11,042.45 

30,348.29 

74,649.17 


$  6.87 
27.23 
23.40 


29.00 


26.30 


5.33 

8.05 

3.94 

6.50 

35.80 

30.00 

.82 

2.25 

5.54 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 


716 


CoflT  OF  CiTT  OF  Sbattlb  Mxtkicipal  Htdbo-elbctbic  Plant — CanHnued 


Division  of  Plant. 


Capacity, 
Kw. 


Cost. 


Unit  Cost, 
per  Kw. 


60,000-volt  wiring  and  switches 
Transmifflion  lines,  total 

No.  1  transmission  line,  total. . . 

Right  of  way  for  both  lines 

1515  poles  and  anns,  in  place. . 

4605  insulators 

117,500  lbs.  No.  2  copper  wire. . 

Two  telephone  lines;  one  of  No.  10  cop- 
per, one  of  No.  14  iron,  on  power  line 
poles,  oomi>lete 

No.  2  transmission  line 

732  poles,  with  arms,  in  place 

2256  insulators,  in  place 

374,700  lb.  No.  4-0  stranded  copper  wire 

Telephone  line  A  in.  plow  steel  cable,  on 


and 


power  line  poles 

Linemen's  cottages,  incomplete 

City  substations,  total 

Main  substation,  Seventh  avenue 
Yesler  Way,  total 

Substation  building 

60,000-volt  switches  and  wiring 

Ei[i^t  1500-Kw.  50,000-volt  transformers 
m  place 

15,000-volt   and   2500-volt  wiring   and 
switches 

Station  switchboard 

Twelve   2500-volt  feeder  regulators  on 
commercial  circuits 

500-Kw.  direct-current  motor  generator 
set 

Twelve  100-lamp  constant-current  trans- 
formers with  switches  and  wiriqg 

500-ampere  hour,  500-volt  storage  bat- 
tery   

60-Kw.  motor  generator 

Four  outlying  substations 

Seven  15,000  to  2500-volt  transformers, 
total  3300  Kw 

Five  constant-current  transformers,  com- 
plete  

Three  2500-volt  feeder  regulators 

Station  wiring  and  switches 

Four  buildings,  corrugated  iron 

Lake  Union  Auxiliary  Station 

Buildingcomplete 

2500-HJ*.  Pelton-Francis   water  wheel 
with  governor  and  valves,  complete. .  . 

1500-Kw.,  2500-volt  alternator  with  ex- 
citer, complete 

Station,  wiring,  switches  and  switchboard 

3400  ft.  40-in.  steel  penstock,  complete, 
400,000  lbs 

Special    tie-line,   2600-volt,    two-phase, 
819,000  cm.  aluminum,  complete 


40,000 
40,000 
13,000 


27,000 


12,000 
12,000 


40,000 

12,000 

12,000 
12,000 

600 

500 

720 

500 

60 

3,300 

3,300 

300 

150 

3,300 


1,900 
1,900 

1,900 

1,900 
1,900 

1,900 


25,547.79 
232,629.62 
119,012.72 
40,490.39 
21,584.04 
19,938.29 
28,480.24 


8,519.76 

112,889.99 

21,943.69 

7,699.37 

72,044.53 

4,475.49 

726.91 

242,096.21 

216,063.89 

30,081.26 

7,250.00 

56,350.00 

46,155.83 
17,500.00 

14,750.00 

15,500.00 

15,250.00 

11,576.80 

1,650.00 

26,032.32 

15,582.26 

4,925.00 

3,330.00 

545.06 

1,650.00 

95,550.32 

10,044.45 

8,914.82 

10,675.85 
8,150.25 

41,456.51 

16,308.44 


.64 
5.82 
9.18 


4.18 


20.17 

18.00 

.18 

4.69 

3.85 
1.46 

24.58 

31.00 

21.20 

23.15 

27.50 

7.90 

4.73 

16.42 

22.20 

.16 

50.30 
5.27 

4.80 

5.62 
4.30 

21.80 


Digitized  by 


Google 


716 


ECONOMICAL  ASPECTS 


The  different  items  cover: 

1.  Dam  and  forebay,  including  connecting  flumes  or  tunnels 
and  all  preliminary  de-watering,  excavation,  concrete,  masonzy 
and  sluicing. 

2.  Penstocks  and  valves. 

3.  Generating  machinery,  including  turbines  with  governors 


DamAForabAy 

Penatocka     Machinerx 

Baildlnffs 

Enff'g.  iBtcTMt  EKk         1 

a 

i-i 

a 

ToUl 
BUM 

A      1           88 

1  «   1 

0 

0 

I-. 

6U0 

B     1     «      1 

a 

H 

IIUO 

c    1        »        1 

1  -» 1 

1  «  1 

D    |u.i  1 

i-'i 

|.M| 

S 

mm 

1  ««  1 

i-i 

H 

«l 

l»l 

100.00 

E     1         «         1 

Fia.  405. — Diagram  Showing  Cost  in  Dollars  per  Kw.  of  Modem  Hydro- 
Electric  Plants. 


Plant  A 

60.000  kw. 

600  ft.  head 

Plant  B 

18.000  kw. 

90  ft.  head 

Plant  C 

30.000  kw. 

164  ft.  head 

Plant  D 

2.500  kw. 

60  ft.  head 

Plant  E  (est.) 

30.000  kw. 

100  ft.  head 

and  regulating  gates,  generators  including  exciters,  transformers, 
switch  gear. 

4.  Building  for  power-house,  switch-house,  tailrace,  etc. 

6.  Engineering,  interest,  contingencies. 

ESTIBIATES  OF  CoST  OF  HyDRO-ELECTRIC  DEVELOPMENTS 

Pages  717  to  723  contain,  in  considerable  detail,  the  cost 
of  construction  and  operation  of  several  water-power  projects  as 
contained  in  Bulletin  5,  prepared  by*  the  State  Engineer's  OflSce 
of  Oregon. 

The  unit  prices  used  in  the  estimates  of  cost  were  determined  as 
follows: 

Concrete.  Proportions  for  massive  concrete:  One  part  Port- 
land cement,  two  and  one-half  parts  sand,  five  parts  broken 
stone  of  size  corresponding  to  gravel,  and  two  and  one-half  parts 
broken  stone  corresponding  to  cobblestone  size.  For  canal  lining 
and  other  thin  concrete  the  larger  size  will  not  be  used. 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 


717 


Material. 

Price,  F.O.B. 
Portland. 

Local  Freight. 

Railway  and 

Wagon  and 

Storage. 

Total. 

Cementi  per  barrel 

SI. 60 
25.00 

S   .60 

6.00 

$2.20 

Lumber,  per  thoiisand 

Sftnd.  Twr  cubic  viirrl 

31.00 
1  50 

Broken  stone,  per  cubic  yard. . . 
(Crushed  on  the  job) 

1.50 

Estimate  op  Cost  peb  Cubic  Yard  of  Concrete 


For  What  Used. 

Cement. 

Sand. 

Stone. 

Forms. 

Labor. 

Total. 

Canal  lining 

Forebay,  etc 

$3.00 
3.00 

$.70 
.70 

$1.40 
1.40 

$1.90 
1.90 

$3.00 
3.00 

$10.00 
10.00 

Dams 
The  estimated  cost  of  concrete  varied  with  volume  as  follows: 

More  than  200,000  cubic  yards $  6.00 

100,000  to  200,000  cubic  yards 6.50 

50,000  to  100,000  cubic  yards 7.00 

25,000  to  50,000  cubic  yards 8.00 

10,000  to  25,000  cubic  yards 9.00 

Under  10,000  cubic  yards 10.00 

Rock  Excavation 
I>am  foundations,  not  including  estimate  for  cofferdam,  per  cubic  yard  $1 .25 

Canals  and  forebayti,  per  cubic  yard 1 .25 

Tunnels,  etc.,  per  cubic  yard $8.00  to  15.00 

Steel  Work 
Trash  racks  (Bessemer-steel  rails) : 

Factory  price,  per  pound $0 .  01} 

Freight,  per  pound .  . , Oil 

Fabrication  and  placing 02 

Total $  .05 

Pipe  work  for  penstocks: 

Factory  price,  plate,  per  pound $  .01} 

Freight 01} 

Fabrication  and  placing,  per  pound 03}  to  .03} 

Total $.06i  CO  .07 

Note.    See  page   724  for   unit   prices   on  Hydraulic   and   Electrical 
Equipment. 


Digitized  by 


Google 


718  ECONOMICAL  ASPECTS 

Oak  Springs  Power  Site 

Estimate  of  Cost: 

Power  head 32  ft. 

Flow  used  for  estimate 3,700  c.  f.  8. 

Brake  horsepower  (80  per  cent  eff.) 10,824  (8100  Ew.) 

Dam: 
Total  height,  50  feet;    length  of  crest.  480  feet; 
length  of  spillway,  200  feet. 

Masonry,  15.310  cubic  yards,  at  $9.00 $137.790 .00 

Excavation,  6443  cubic  yards,  at  $1.25 8.054 .00 

Cofferdam 70.000 .00 

Incidentals  and  special  foundation  contingencies ...  34,156 .00 

$250,000.00 

Forebay.  etc.: 

Excavation,  12,000  cubic  yards,  at  $1.25 15,000 .00 

Concrete  walls,  1500  cubic  yards  at  $10.00 15,000 .00 

Trash  raclcs,  12,000  pounds  steel,  at  5c 600 .00 

Stop  logs 400.00 

81.000.00 

Headgates,  Penstocks,  etc.: 

10  sliding  headgates,  set  in  place,  at  $750.00 7,500.00 

10  hydraulic  relief  valves,  in  place,  at  $1.200.00 . . .        12,000.00 
800  feet  ^inch   steel  penstock.  12  feet  diameter, 

530  pounds,  per  foot  at  5}c.,  $34.45 27,560 .00 

47.000.00 

Power-house  and  draft  tubes: 

Power-house,  reinforced  concrete,  8100  Kw.,  at  $5.00 

per  Kw 40,500.00         40,500.00 

Summation $368,560.00 

Engineering  and  contingencies,  25  per  cent 92,140.00 

Interest  during  construction,  5  per  cent  approx 25.300.00 

$486,000.00 
Hydro-electrical  machinery: 

10  horizontal  water-wheel  units,  1085  H.P.,  in  place. 

speed  200  R.P.M..  at  $10.000.00 100.000 .00 

10  750-Kw.  generators.  200  R.P.M.  at  $8.00 60.000.00 

Exciter  turbines  and  exciters,  in  place,  at  80c.  per 

Kw 6.480 .00 

Transformers,  at  $4.00  Kw 32.400 .00 

Switchboard  and  accessories,  cables,  etc.,  at  $2.25 

per  Kw 18,225 .00 

Traveling  crane,  30-ton 9,(X)0 .00 

Quarters,  water  supply,  etc 20,000 .00 

Summation 246.105 .00 

Engineemg  and  contingencies,  25  per  cent 61.525 .00 

Interest  during  construction,  approx 6,370.00 

314.000.00 

Summation 800,000.00 

Railway,  realignment.  8  miles,  at  $50.000.00 400,000.00 

Total. construction  cost $1,200,000.00 

Total  amount  of  power.  E.H.P.,  10.824. 

Construction  cost,  per  E.H.P 110 .87 

Assumed  right  of  way  cost,  per  E.H.P 5 .00 

Cost  of  development,  per  E.H.P $115.87 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS 


719 


LocKiT  Power  Site 

Estimate  of  cost: 

Power  head 70  feet 

Flow  used  for  estimate 4,500  c.  f.  s. 

Brake  horse-power  (80  per  cent  eff.) 28,630  (21,500  Kw.) 


Dam: 

Total  height,  94  feet;    length  of  crest,  720  feet; 
length  of  spillway,  225  feet. 

Masonry,  56.014  cubic  yards,  at  $7.00 $392,008 .00 

Excavation,  15,533  cubic  yards,  at  $1.25 19,417 .00 

Cofferdam 50,000 .00 

Trash  racks,  30,0(X)  pounds  steel,  at  5c 1,500.00 

Incidentals  and  special  foundation  contingencies .  .  56,985 .00 

Headgates,  penstocks,  etc.: 

10  sUdlng  headgates,  set  in  place,  at  $750.00 7,500.00 

10  hydraulic  relief  valves.  In  place,  at  $1.200.00 . .  12,000.00 

1,650  feet  ^inch  steel  penstock,  11  feet  diameter. . 

600  pounds  per  foot  at  6ic.,  $32.50 53,625.00 

1450  feet  i^inch  steel  penstock,  10  feet  diameter, 

450  pounds,  per  foot  at  7c.,  $31.50 45,675 .00 

Power-house  and  draft  tubes; 
Power-house,  reinforced  concrete,  21,500  Kw.,  at 

$5.00  per  Kw 107,500.00 

Summation 

Engineering  and  contingencies,  25  per  cent 

Interest  during  construction,  i  of  3  years,  at  4  per  cent, 
6  per  cent  approz 

Hydro-electrical  Machinery: 

10  horizontal  water  wheel  units,  2860  H.P.,  In 

place,  speed  360  R.P.M.,  at  $15,000 150,000.00 

10  2500-Kw.  generators,  350  R.P.M.,  at  $7.00  per 

Kw 175,000.00 

Exciter  turbines  and  exciters.  In  place,  at  80c. 

per  Kw 17,200.00 

Transformers,  21,500  Kw.,  at  $4.00  per  Kw 86,000.00 

Switchboard  and  accessories,  cables,  etc.,  at  $2.23 

pcrKw 48,000.00 

Traveling  crane,  30-ton 9,000 .00 

Quarters,  water  supply,  etc 20,000. 00 

Summation 505,200 .00 

Engineering  and  contingencies,  20  per  cent 101,000 .00 

Interest  during  construction,  3  per  cent  approx 18,800.00 

Total  construction  cost 

Total  amount  of  power,  E.H.P.,  28,630. 

Construction  cost,  per  E.H.P 56 .41 

Assumed  right  of  way  cost,  per  E.H.P 10 .00 

Cost  of  development,  per  E.H.P 


$520,000.00 


118,800.00 


107,500.00 

$746,300.00 
186,575.00 

57,125.00 
$990,000.00 


625.000.00 
$1,615,000.00 


$66.41 


Digitized  by 


Google 


720  ECONOMICAL  ASPECTS 

Mecca  Power  Site 

Estimate  of  cost: 

Power  head 90  feet 

Flow  used  for  estimate. 3,400  c.  f .  s. 

Brake  horse-power  (80  per  cent  eflf.) 27.760  (20.750  Kw.) 

Dam: 

Total  height,  110  feet;   length  of  crest.  650  feet; 
length  of  spillway,  160  feet. 

Masonry.  64.787  cubic  yards,  at  $7.00 S453.509 .00 

Excavation,  10,020  cubic  yards,  at  S1.25 13.650.00 

Cofferdam 40,000 .00 

Incidentals  and  special  foundation  contingencies .  .  22.841 .00 

$530.000. 00 

Forebay,  etc.: 

Trash  racks,  12,000  pounds  steel,  at  5c 600 .00 

Stop  logs 400 .00 

1,000.00 

Headgates,  penstocks,  etc.: 

8  sliding  headgates.  set  in  place,  at  $900 .00 7,200 .  00 

8  hydraulic  relief  valves,  in  place,  at  $1,200.00 .  .  .  9,600.00 

1400  feet  A-inch  steel  penstock,  12  feet  diameter, 

530  pounds  per  foot  at  O^c,  $34.45 48.230.00 

600  feet  A-inch  steel  penstock,  11  feet  diameter, 

615  pounds,  per  foot  at  7c.,  $43.05 25,830 .00 

90.86000 

Power-house  and  draft  tubes: 

Power-house,  reinforced  concrete,  20,750  Kw.,  at 

$5.00  per  Kw 103,750 .00        103.750  O^J 

Summation $725,610.00 

Engineering  and  contingencies,  25  per  cent 181.402.00 

Interest  during  construction,  6  per  cent  approx 62,988.00 

$970,000.00 
Hydro-electrical  machinery: 
8  horizontal  water  wheel  units,  in  place,  3470  H.P., 

speed  400  R.P.M.,  at  $10,400.00 83,200 .00 

8  2500- Kw.  generators.  400  R.P.M..  at  $6.00  per  Kw.  120.000 . 00 
Exciter  turbines  and  exciters,  in  place,  at  80  c.  per 

Kw 16.600.00 

Transformers,  at  $4.00  per  Kw 83,000 .00 

Switchboard  and  accessories,  cables,  etc.,  at  $2.25 

per  Kw 46.687 .00 

Traveling  crane,  40-ton 15,000 .00 

Quarters,  water  supply,  etc 20,000 .00 

Summation $384.487 .00 

Engineering  and  contingencies.  20  per  cent 76,895 .00 

Interest  during  construction.  2  per  cent  approx 8.618.00 

470.000 .  00 

Summation $1,440,000.00 

Railway  realigned,  6  miles  at  $50,000 300.000  00 

Total  construction  cost $1.740,00 .000 

Total  amount  of  power,  E.H.P.,  27,760. 

Construction  cost,  per  E.H.P 62.68 

Assumed  right  of  way  cost,  per  E.H.P 5.00 

Cost  of  development,  per  E.H.P $67.68 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  721 

White  Horse  Rapids  Power  Site 

Batimate  of  Cost: 

Power  head 138  feet 

Flow  used  for  estimate 3,700  c.  f.  s. 

Brake  horse-power  (80  per  cent  eff.) 47,200  (35,100  Kw.) 

Dam: 

Total  height,  122  feet;   total  length  of  crest.  440 
feet;  length  of  spillway,  160  feet. 

Masonry.  56.762  cubic  yards,  at  S7.00 $397,334 .00 

Excavation,  13,771  cubic  yards,  at  $1.25 17,214 .00 

CofTerdam 40,000 .00 

Incidentals  and  special  foundation  contingencies.         55.452.00 

$610,000.00 

Forebay.  etc.: 

Excavation,  10,000  cubic  yards,  at  $1.25 12,500 .00 

Concrete  walls,  2500  cubic  yards,  at  $10.00 25,000.00 

Trash  racks,  120,000  pounds  steel,  at  5c 600 .00 

Stop  logs 400 .00 

38,500.00 

Diversion  line: 

Canal  excavation,  260.000  cubic  yards,  at  $1.25. . .        325,000.00 
Canal  lining.  4400  cubic  yards,  at  $10.00 44.000 .00 

369,000 .  00 

Headgates.  penstocks,  etc. : 

10  sliding  headgates,  set  in  place,  at  $900.00 9,000 .00 

10  hydraulic  relief  valves,  in  place,  at  $1,200.00 . . .  12,000.00 

1100  feet  ^inch  steel  penstock,  11  feet  diameter, 

500  pounds  per  foot  at  6}c.,  $32.50 35,750.00 

600  feet  i-inch  steel  penstock,   10  feet  diameter, 

495  pounds  per  foot  at  7c.,  per  foot  $34.65 20.790.00 

77,640.00 

Power-house  and  draft  tubes: 

Power-house,  reinforced  concrete,  35,100  Kw.,  at 

$5.00  per  Kw.  (made  the  same  as  Frieda) .  . 176,000.00        176.000 .00 

Summation $1,171,040.00 

Engineering  and  contingencies,  25  per  cent 292,760 .00 

Interest  during  construction,  i  of  2  years,  4  per 

cent  approx 60.200 .00 

$1,524,000.00 
Hydro-electrical  machinery: 

10  horizontal  water  wheel  units,  in  place,  4720  H.P., 

speed  450,  at  $22,000 220,000.00 

10  3500  Kw.  generators.  450  R.P.  M.,  at  $5.00 . . .  175,000.00 
Exciter  turbines  and  exciters,  in  place,  at  80c.  per 

Kw 28.080.00 

Transformers,  at  $4.00  per  Kw 140,400.00 

Switchboard  and  accessories,  cables,  etc.,  at  $2.25 

per  Kw 78,975.00 

Traveling  crane,  40-ton 15,000 .00 

Quarters,  water  supply,  etc 20.000 .00 

Summation $677.455 .00 

Engineering  and  contingencies,  20  per  cent 135,491 .0() 

Interest  during  construction,  20  per  cent  approx 16,054 .00 

829,000.00 

Railway  realigned,  9  miles,  at  $50.000 .00 450.000 .00 

Total  construction  cost $2,803,000 .00 

Total  amount  of  power.  E.H.P.,  47.200. 

Construction  cost,  per  E.H.P 59 .38 

Assumed  right  of  way  cost,  per  E.H.P 5 .00 

Cost  of  development  per  E.H.P $64 .38 


Digitized  by 


Google 


722 


ECONOMICAL  ASPECTS 


Metouus  Power  Site 

Estiznate  of  cost: 

Power  head 210  feet 

Flow  used  for  estimate 3.400  c.  f.  i. 

Brake  horse-power  (80  per  cent  eO.) 64,060  (48,700  Ew.) 

Dam: 

Total  height,  236  feet;   length  of  crest.  420  feet: 
length  of  spillway.  125  feet. 

Masonry,  183.000  cubic  yards,  at  S6.50 $1,189,500.00 

Excavation,  37,570  cubic  yards,  at  SI. 25 46.062 .00 

Cofferdam 75,000.00 

Wagon  roads 25,000.00 

Incidentals  and  si;>ecial  foundation  contingencies. .        164,;38.00 

$1,500,500.00 

Forebay.  etc.: 

Excavation,  8000  cubic  yards,  at  $1.25 10.000 .00 

Concrete  walls,  1500  cubic  yards,  at  $10.00 15.000 .00 

Trash  racks,  20.000  pounds  steel,  at  5c 1.000 .00 

26.000 .  00 

Diversion  line: 

Tunnel  excavation  and  lining.  300  feet  by  15  feet. 

by  20  feet,  at  $150.00 45.000.00 

Headgates,  penstocks,  etc.: 

10  sUding  headgates,  set  in  place,  at  $900.00 9,000 .00 

10  hydraulic  relief  valves  in  place,  at  $1.200.00..  12.000.00 

500  feet  A-iuch  steel  penstock,  12  feet  diameter, 

530  pounds  per  foot,  at  Ofc,  $34.45 17,225 .00 

500  feet  A-inch  steel  penstock,  10  feet  diameter, 

565  pounds  per  foot  at  7c.,  $39.55 19,775 .00 

58,000.00 

Power-house  and  draft  tubes: 

Power-house,  reinforced  concrete,  48,700  Ew..  at 
$5.00  per  Ew 243.500 .00       243.500  00 

Summation $1,873,000.00 

Engineering  and  contingencies,  25  per  cent 468,250.00 

Interest  during  construction,  8  per  cent 208,750  00 

$2,550,000.00 
Hydro-electrical  machinery: 

10  horizontal  water  wheel  units,  in  place,  6496  H.P., 

speed  400  R.P.M.,  at  $24.000 .00 240,000 .00 

10  5000-Ew.  generators,  400  R.P.M.,  at  $5.00  per 

Ew 250,000.00 

Exciter  turbines  and  exciters,   in  place,  at  82c. 

per  Ew 40,000.00 

Transformers,  at  $4.00  per  Ew 194,800 .00 

Switchboard  and  accessories,  cables,  etc.,  at    $2.25 

per  Ew 109,575.00 

Traveling  crane,  40-ton 15,000.00 

Quarters,  water  supply,  etc 20.000 .00 

Summation $869,375 .00 

Engineering  and  contingencies,  20  per  cent 173,875 .00 

Interest  during  construction,  30  per  cent  approx 36,750 .00 

1.080.00000 

Total  construction  cost $3,630,000.00 

Total  amount  of  power,  E.H.P.,  64,960. 

Construction  cost,  per  E.II.P 55 .88 

Assumed  right  of  way  cost,  per  E.H.P 5.00 

Cost  of  development,  per  E.H.P 160. S8 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  723 

Jefferson  Creek  Power  Site 

Estimate  of  cost: 

Power  head 400  feet 

Flow  used  for  estimate 1,000  c.  f.  s. 

Brake  hone-power  (80  per  cent  eff.) 36,363  (27,100  Kw.) 

Dam: 

Total  height,  20  feet ;  length  of  crest.  00  feet ;  length 
of  spUIway.  80  feet. 

Masonry.  1000  cubic  yards,  at  $10.00 $10.000 .00 

Excavation,  300  cubic  yards,  at  $2.00 600 .00 

Cofferdam 800 .00 

Incidentals  and  foundation  contingencies 8.600 .00 

$20,000.00 

Forebay.  etc.: 

Excavation,  concrete  walls,  trash  racks,  etc 25.000 .00 

Diversion  line: 

Canal  excavation  and  lining,  8  feet  by  30  feet  by 

41.000  feet,  at  $30.00 1,230,000.00 

Headgates,  penstocks,  etc.: 

4  sliding  headgates.  set  in  place,  at  $900.00 3,600.00 

4  hydraulic  relief  valves,  in  place,  at  $1,200.00. .  .  .  4.800.00 

1000  feet  A-inch  steel  penstock.  10  feet  diameter, 

450  pounds,  per  foot  at  6ic.,  $29.25 29.250 .00 

1000  feet  i-inch  steel  penstock.  9  feet  diameter. 

440  pounds,  per  foot  at  6}c..  $28.60 28,600 .00 

1000  feet,  ^inch  steel  penstock,  8  feet  diameter, 

500  pounds  per  foot  at  7c.,  $35.00 35.000 .00 

101,250.00 

Power-house  and  draft  tubes: 

Power-house,  reinforced  concrete,  27,100  H.P.,  at 

$5.00  per  Kw 136.500 .00 

Summation $1,511,750.00 

Engineering  and  contingencies,  25  per  cent 377,938 .00 

Interest  during  construction,  i  of  2  years,  at  4  per 

cent  approx 80,812 .00 

$1,970,000.00 
Hydro-electrical  machinery: 
4  horizontal  water-wheel  units,  in  place,  9091  H.P., 

speed  360  R.P.M.,  at  $31.000.00 124.000.00 

4  7000-Kw.  generators.  365  R.P.M.,  at  $5.00  per  Kw.  140,000 .00 
Exciter  turbines  and  exciters,  in  place,  at  80c.  per 

Kw 21,680.00 

Transformers,  at  $4.00  per  Kw 108,400 .00 

Switchboard  and  accessories,  cables,  etc..  at  $2.25 

per  Kw 60,975.00 

Traveling  crane.  40-ton 15.000 .  00 

Quarters,  water  supply,  etc 20,000 .00 

Summation $490,055 .  00 

Engineering  and  contingencies,  20  per  cent 98,011 .00 

Interest  during  construction,  2  per  cent  approx.  .  .  11,934.00 

600.000.00 

Total  construction  cost $2,570,000.00 

Total  amount  of  power,  E.H.P.,  36,363. 

Construction  cost,  per  E.H.P 70.67 

Assumed  right  of  way  cost,  per  E.H.P 5.00 

Cost  of  development,  per  E.H.P $75.67 


Digitized  by 


Google 


724  ECONOMICAL  ASPECTS 

Hydraulic  Equipment.  Horizontal  turbine  water  wheels  in 
pairs.  Estimate  based  on  figures  obtained  from  two  independent 
manufacturers.  Prices  include  freight  charges  and  cost  of  installa- 
tion.    Relief  valves  are  estimated  separately. 

Electrical  Equipment.  Prices  on  electrical  equipment  are 
based  upon  estimates  of  manufacturers  of  electrical  machinery,  and 
are  as  follows: 

Generators,  all  of  the  3-phase,  2300-volt,  60-cycle  type,  per  Kw.  output: 

For  heads  of  under  40  feet 18.00 

For  heads  of  under  40  to  80  feet 7.00 

For  heads  of  80  to  120  feet. 6.00 

For  heads  of  120  feet 5.00 

Exciter  turbines  and  exciters,  per  Kw.  output,  whole  plant 80 

Switchboard  and  accessories,  cables,  etc.,  per  Kw.  output,  whole 

plant 2.25 

Transformers,  oil  insulated  and  water  cooled,  2,300-60,000  volts, 

per  Kw.  output,  whole  plant 4.00 


Cost  of  Georgia  Railway  and  Power  Company's  Develop- 
ment AT  Tallulah  FallS;  Ga. 

(A.I.E.E.,  October  11,  1915) 

The  development  consists  essentially  of  an  artificial  reservoir 
of  a  capacity  of  1,400,000,000  cubic  feet  formed  by^  two  reinforced 
concrete  buttress  dams  located  near  near  Mathis,  Ga.,  seven 
miles  from  the  diverting  dam  and  intake  at  Tallulah  Falls;  an 
artificial  reservoir  at  Tallulah  Falls  having  an  available  pondage 
of  63,000,000  cubic  feet  formed  by  a  cyclopean  masonry  dam  of 
the  gravity  type  located  some  60  feet  below  the  tunnel  intake; 
a  tunnel  with  a  cross-sectional  area  of  151  square  feet  6666  feet 
long  leading  from  the  intake  at  the  Tallulah  reservoir  to  the 
surge  or  pressure  tank  at  the  top  of  the  gorge  immediately  above 
the  power-house;  five  steel  penstocks  5  feet  in  diameter,  each  of 
which  serves  a  17,000-H.P.  Francis  type  water  turbine  in  the 
power-house.  Five  three-phase,  60-cycle  6600-volt,  vertical 
generators  are  direct-connected  to  these  water  wheels. 

The  electrical  energy  from  these  machines  is  stepped  up  from 
6600  volts  to  110,000  volts  for  transmission  by  five  banks  of  three 
3333  Kw.  single-phase  static  transformers  of  the  water-cooled 
type  and  is  transmitted  over  two  outgoing  lines. 

Reservoir,    The  reservoir  covers  834  acres,  most  of  which  was 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  725 

heavily  timbered  prior  to  the  construction  period.  It  was  cleared 
of  timber,  brush  and  other  debris  before  the  impounding  began, 
at  a  cost  of  $21  per  acre,  represented  by  $8.35  for  cutting  and 
$12.65  for  gathering  and  burning. 

Reservoir  Dams.  There  are  two  reinforced  buttressed  dams, 
the  largest  is  660  feet  in  length,  93  feet  high  to  the  crest  of  the 
spillway  and  114  feet  to  the  top  walkway.  The  other  dam  is 
much  smaller.  The  quantities  involved  in  the  construction  of 
these  two  dams  were  2,200,000  pounds  of  steel  reinforcing,  and 
38,000  cubic  yards  of  concrete. 

The  following  figures  give  the  cost  per  cubic  yard  of  these  two 
dams: 

Quarry $1,611 

Crushing  and  mixing 818 

Freight  and  engine  service 1 .110 

Placing  concrete 744 

Reinforcement 1.4i7 

Placing  reinforcement 823 

Labor. 3.746 

Cement 2.777 

Sand 126 

Plant,  erecting  and  maintenance 1 .496 

Small  tools  and  supphes 1 .  123 

Lumber 1 .034 

Miscellaneous  expenditures 1 .617 

Superintendence  and  overhead 1 .443 

Total $19,916 


Di  erting  Dam.  This  dam  is  of  the  gravity  type  built  of 
Cyclopean  masonry,  heavy  stone  forming  a  Uttle  over  one-third  of 
the  mass.  The  dam  is  110  feet  high  from  the  stream  stratum  and 
has  a  length  of  426  feet.  The  spillway  section  is  280  feet  in  length, 
made  up  of  ten  28-foot  openings  between  concrete  piers.  There 
was  used  in  this  dam  39,000  cubic  yards  of  concrete  which  was 
placed  by  the  contractors  at  $4.80  per  cubic  yard,  the  actual  cost 
possibly  being  about  $3.70  per  cubic  yard.  The  cost  of  bridge 
piers  and  flashboards  is  additional.  The  contract  price  for  the 
excavation  work  was  $1.50  per  cubic  yard. 

Intake.  The  intake  is  a  self-contained  reinforced  structure 
divided  by  partitions  into  five  sections.  The  construction  involved 
about  7000  cubic  yards  of  excavation,  mostly  rock,  and  2670 


Digitized  by 


Google 


726  ECONOMICAL  ASPECTS 

cubic  yards  of  concrete.    The  detailed  cost  of  excavation  and  cod- 
Crete  for  the  intake  was  as  follows: 

Excavation:                                                             Per  Cubic  TanL 

Lumber $0,974 

Explosives 0.0G5 

Miscellaneous  supplies 0. 123 

Transportation 0.071 

Liability  insurance 0 .049 

Removing  debris 0.235 

Total $1,517 

Concrete: 

Labor $3 .  902 

Cement 1.982 

Lumber 0.794 

Freight: 0.042 

Transportation 0.203 

Liability  insurance 0. 136 

Erection  of  plant 0.400 

Crusher 1 .  280 

Miscellaneous  supplies 0.205 

Removing  debris 0.086 

Total $9,030 

Tunnel  The  tunnel  is  6666  feet  long,  and  has  a  net  area  of 
151  square  feet  inside  the  concrete  lining.  About  75  per  cent  of 
the  tunnel  was  driven  by  the  top-heading  method  and  for  the 
remainder  the  lower  heading  or  stopping  method,  which  proved 
to  be  much  cheaper.  The  total  excavations  amounted  to  56,000 
cubic  yards. 

The  unit  cost  of  excavating  39,831  yards  of  this  tunnel  was  as 
follows: 

Per  Cubic  Yard. 

Labor $3,833 

Explosives 0.604 

Lubricants 0.019 

Piping 0.026 

Drill  repairs 0.172 

Miscellaneous  supplies 0.237 

Freight 0.087 

Transportation 0.247 

Liability  insurance 0. 181 

Miscellaneous  charges 0.066 

Depreciation  on  equipment 0. 150 

Power 0.306 

Total $5,928 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  727 

The  concrete  lining  of  the  tunnel  called  for  the  placing  of 
18,966  cubic  yards  of  concrete,  the  unit  cost  of  the  lining  being: 

Labor S5.061 

Cement 1 .970 

Misoellaneous  materials 0. 405 

Lumber 0. 136 

Freight 0.065 

Transportation 0 .  155 

Liability  insurance 0 .  165 

Royalty  on  mixers 0.413 

Miscellaneous  cost 0 .245 

Crushing  stone 1 .  991 

Quarrying  stone 0.858 

Plasterers 0.202 

Cleaning  tunnel 0.376 

Total $12,042 

The  entire  tunnel  was  grouted  with  grout  consisting  of  one 
part  cement  to  one  and  one-half  parts  sand.  The  cost  of  the 
grouting  was  $1,436  per  cubic  yard  of  concrete  Uning,  made  up  of 
the  following  unit  figures: 


Item. 


Labor 

Cement 

Transportation 

Liability  insurance 

Miscellaneous  supplies . 


Cost  per  Linear 
Foot  of  Tunnel. 


$2,209 
1.649 
0.001 
0.065 
0.155 

$4,079 


The  following  figures  give  the  approximate  total  cost  of  the 
tunnel  per  linear  foot: 

Excavation $44.44 

Concrete  lining 34 .  20 

Grouting 4.08 

Adits  and  shafts 1 .91 

Compressor  plants,  spur  tracks  and  operation 8.99 

Steel  forms 2.94 

Total $96.56 


Digitized  by 


Google 


728  ECONOMICAL  ASPECTS 

Forebay,  The  forebay  is  a  reinforced  concrete  structure, 
30X70  feet  and  95  feet  deep.  The  excavation  involved  some 
4750  cubic  yards  of  rock  and  the  thickness  of  the  concrete  in  the 
walls  of  the  tank  varied  from  3  to  6  feet.  Some  700  tons  of  steel 
reinforcement  were  used. 

The  cost  of  the  rock  excavation  at  the  forebay  was  as  foUows: 

Per  Cubic  Yard. 

Labor SI  .620 

Explosives 0. 106 

Transportation 0.089 

Liability  insurance 0.067 

MisceUaneous  supplies 0 .  246 

Miscellaneous  expenses 0.038 

$2,166 

The  concrete  Uninjt  of  the  forebay  shows  the  following  unit 
figures: 

Per  Cubic  Yard. 

Labor $1 .680 

Cement 1 .920 

Lumber 0.117 

Freight 0.012 

Transportation 0.013 

Liability  insurance 0.049 

Miscellaneous  expenses 0 .  178 

Crushing  stone 1 .569 

MisceUaneous  supplies 0.033 

$5,571 

The  above  figures  represent  the  unit  cost  of  the  concrete 
below  elevation  1500.  The  thickness  of  concrete  was  so  small 
and  the  amount  of  reinforcement  so  great  in  comparison  with  the 
concrete  below  elevation  1500  that  no  unit  copper  yard  was  made. 
The  concrete  used  above  elevation  1500  cost  $1,925  per  superficial 
square  foot  surface  one  side. 

Power-plant  Builiing.  The  power-plant  buildings  are  con- 
structed with  a  concrete  substructure  and  a  structural  steel  frame 
work  enclosed  with  full  brick  walls  as  a  superstructure.  The  gen- 
erator building  is  186  feet  long,  42  feet  3  inches  wide  and  49  feet 
high  above  generator  floor.  The  switch-house  is  277  feet  long, 
46  feet  wide  and  103  feet  high. 

There  are  five  vertical  reaction  turbines  operating  under  an 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  729 

effective  head  of  580  feet  at  a  speed  of  614  R.P.M.,  driving  five 
12,000-Kv.A.  6600-volt  generators  with  direct-connected  exciters. 
There  are  also  five  transformer  banks  each  consisting  of  three 
3333-Kv.A.  single-phase  transformers  for  stepping  up  the  voltage 
to  110,000. 

The  unit  cost  of  the  power-house  buildings  and  installed  equip- 
ment is  given  in  the  following,  the  cost  of  the  hydraulic  and 
electrical  equipment  being  based  on  the  installed  capacity  of 
50,000  Kw.  on  the  original  rating  and  that  of  the  buildings  and 
other  equipment  on  60,000  Kw.,  the  ultimate  capacity  of  the 
original  rating. 

Per  Kw.  Capacity. 

Buildings  and  foundations: 

Rock  excavation $0,428 

Concreting  foundations  and  substructure 2 .  114 

Structural  steel 0.622 

Handling  and  unloading 0.030 

Erecting 0. 109 

Brick,  sand  and  cement 0.460 

Handling,  mixing  and  laying 7 0.960 

Windows  and  doors 0 .  176 

Handling  and  erecting 0.003 

Tile  roofing 0. 115 

Concrete  tile  floors 0.400 

Miscellaneous  material 0. 186 

Miscellaneous  labor  and  transportation  of  men 0 .  234 

Painting 0. 124 

Plumbing 0.063 

Building  inspection 0 .  142 

Tailrace: 

Rock  excavation .  .  ^ 0 .  197 

Cribbing .' 0.017 

Concreting  tailrace  walls 0 .  242 

Total $6,612 

Equipment: 

Hydraulic  equipment $6,682 

Handling  and  erecting 0. 463 

Electrical  equipment  and  erection 6. 236 

Auxiliary  equipment 0 .999 

Handling  and  erecting  auxiliary  equipment 0 .  106 

High-  and  low-tension  switch  and  bus  structure ...  0 .  445 

Water  and  oil  piping  system 0.244 

Total  equipment $16,074 


Digitized  by  VjOOQ IC 


730  ECONOMICAL  ASPECTS 

Grouping  the  above  items  under  a  more  condensed  form,  we 
have: 

Tailrace $0,456 

Buildings — substructure 2 .  542 

BuUdings — superstructures 3 .  372 

Buildings — inspection 0. 142 

Total  equipment 15.074 

Cost  per  Kw.  capacity $21 .586 

In  addition  to  this  cost  there  is  a  certain  proportion  of  the 
temporary  compressor  plant,  spur  tracks,  general  tool  and  utility 
equipment,  etc.,  amounting  to  $1,178,  which  should  be  charg^to 
this  power-plant  construction,  making  the  total  cost  of  the  power- 
plant  buildings  and  equipment  $22,764  per  kilowatt  capacity. 

As  the  foregoing  costs  do  not,  in  some  instances,  give  the  cost 
of  completed  structure  under  the  various  headings,  the  following 
table  will  supply  the  construction  cost  per  kilowatt  capacity  of  the 
entire  power  production  plant,  including  reservoirs,  dams,  all 
hydraulic  conduits,  power  plant  and  equipment,  and  including 
temporary  construction  plant,  such  as  compressor  plants,  water 
system,  spur  tracks,  etc. 

Mathis  dams  and  reservoirs $17. 104 

Intake  dam  and  bridge 4.660 

Intake 1.102 

Tunnel 12.379 

Forebay 2.395 

Penstock  tunnels  and  portal 0.694 

Penstocks  and  foundations 5 .  568 

Power  plant  and  equipment 22.764 

Total  construction  cost  power  production 
plant  per  kilowatt $66,666 

The  following  gives  the  percentage  relation  of  various  ex- 
penses on  the  development  as  a  whole,  which  might  be  applicable 
to  any  other  development,  and  therefore  does  not  include  the 
cost  of  land  or  property  expense:  ^^      ^^ 

General  construction  expenditure 75 .575 

General  engineering  expense 3 .078 

General  legal  expense 1 .891 

Interest,  bonds  and  advances  during  construction.  11 .315 

General  overhead  expense 1 .  773 

General  contract  expense 6.368 

Total 100.000 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  731 

Estimate  of  72,000-Kw.  Generating  Station  at  Great  Falls, 
Potomac  River,  for  Supplying  Light  and  Power  for  the 
Use  of  United  States  and  the  District  of  Columbia 

(From  H.  R.  Document  No.  1400) 

This  proposed  project  provides  for  a  dam  across  the  Potomac 
River  at  Great  Falls,  creating  a  lake  or  reservoir  of  some  3000  acres 
area  and  an  operating  head  of  111  feet.  The  dam  is  in  two 
parts,  a  spillway  dam  and  an  intake  dam.  The  former  is  of 
the  arched  type,  somewhat  similar  to  the  spillway  section  of  the 
Gatun  dam,  at  Panama,  and  comprises  eighteen  openings  sepa- 
rated by  piers  and  provided  with  Stoney  gates.  A  gatehouse  is 
arranged  for  on  top  of  the  intake  dam,  from  which  nine  pen- 
stocks convey  the  water  to  the  tm*bines.  These  are  of  riveted 
steel  from  f  to  f -inch  thickness,  the  inside  diameter  being  13  feet 
and  the  length  140  feet. 

When  completed  the  equipment  will  comprise  nine  12,500- 
H.P.  single-runner  vertical  turbines  operating  at  150  R.P.M. 
under  a  head  of  111  feet.  These  will  drive  nine  10,000-Kv.A. 
(8000  Kw.  .8  P.F.)  3-phase,  60-cycle,  13,209-volt  generators,  with 
direct-connected  exciters.  Provision  is  further  made  for  com- 
plete switching  equipment  and  station  auxiUaries. 

The  allowance  in  the  original  estimate  for  relocating  the  Ches- 
apeake and  Ohio  Canal  has  been  omitted  in  the  following: 

Estimated  Ck>ST 
Spillway  dam: 

Piers,  superior  concrete,  7540  cubic  yards,  at  $9.00 $67,000 

Piers  concrete,  27,800  cubic  yards,  at  $8.00 222,000 

Water-flow  guides,  concrete,  1850  cubic  yards  at  $8 15,000 

Dam,  superior  reinforced  concrete,  37,400  cubic  yards,  at  $9 337,000 

Dam,  Cyclopean  superior  concrete,  36,200  cubic  yards,  at  $5.50*. .  200,000 

Dam,  Cyclopean  concrete,  233,550  cubic  yards,  at  $4.50  * 1,050,000 

Total  masonry $1,891,000 

Excavation,  rock,  115,400  cubic  yards  at  $2.50 289,000 

Stoney  gates,  18,  erected,  weight  1,162,000  pounds,  at  $0.08 130,000 

.  Stoney  gates,  fittings  and  machinery,  etc.,  18  sets  at  $6500 117,000 

Floating  caisson 5,000 

Foot  bridge,  erected,  weight  833,000  pounds,  at  $0.08 65,000 

Railing,  2850  feet,  at  $1.75 5,000 

Total  spillway  dam $2,502,000 


Digitized  by 


Google 


732  ECONOMICAL  ASPECTS 

Intake  Dam  and  Power-house 

Power-house  superstructure,  2,200,000  cubic  feet  at  15c $330,000 

Power-house,  substracture,  2,000,000  cubic  feet,  at  17c 340,000 

Intake  house,  superstructure,  750,000  cubic  feet,  at  15c 113,000 

Intake  house,  substructure,  471,000  cubic  feet,  at  17c 80,000 

Cranes  and  raih-oad  track 15,000 

Turbines,  erected,  9,  at  $51,000 459,000 

Central  lubrication  system 27,000 

Electrical  units,  9,  and  switchboard  etc.,  at  $90,000 810,000 

Intake  dam,  cyclopean  concrete,  107,700  cubic  yards,  at  $4.50  * .  485,000 
Excavation,  intake  dam,  power-house,  and  tailraoe,  475,000  cubic 

yards,  at  $2.50 1,187,500 

Penstocks,  10  erected,  1,350,000  pounds,  at  8c 108,000 

Rack  bars,  10  sets,  9350  square  feet,  at  $1.75 16,500 

Head  gates,  2 5,000 

3  pumps  and  their  motors,  erected 20,000 

Force  main,  laid,  300,000  pounds,  at  14c 42,000 

Shore  wasteway 25,000 

Road  and  branch  railroad 100,000 

Total  intake  dam  and  power-house $4,163,000 

Note. — Prices  marked  thua  (*)  are  reduced  by  reason  of  part  cost  being  borne 
by  rock  excavation. 

Summary 

Spillway  dam $2,502,000 

Intake  dam  and  power-house 4,163,000 

Land  and  water  rights 1,500.000 

Engineering  and  contingencies 585,000 

Total $8,750,000 

Cost  of  P©wer 

Estimate  for  319.4  millions  kilowatt-hours  annual  output, or  100,000  HP. 

effective  peak  load. 

Operation: 

Administration  and  labor $60,000 

Maintenance  and  supplies 20,000 

S8O000 

Depreciation,  headworks  and  power-house:  •^^ 

1  per  cent  on  masonry $4,910,500  $49,105 

2  per  cent  on  steel  work 438,500  8,770 

3  per  cent  on  machinery 1,316,000  39,480 

Fixed  charges:                                                                »«•«««'«»  *^'^ 

Interest,  3  per  cent,  sinking  fund  3  per  cent,  or  6  per  cent  on 

above  $8,750,000 $525,000 

Total $702,355 

Or  2.2  mills  per  kilowatt-hour  of  output. 


Digitized  by 


Google 


CX)ST  OF  HYDRO-ELECTRIC  POWER  PLANTS  733 

Estimated  Cost  op  200,000  and  300,000  Horse-power  Hydro- 
electric Developments 

From  Bulletin  No.  5.  State  Engineer's  Office.  Oregon 
Head:    200  feet  minimum. 

ESTIHATB   OF   CoST,    200,000   ELECTRICAL   HOR8E-POWER   INSTALLATION 

River  diversion: 

Temporary  diversion  channel: 

Excavation  above  elevation  885, 500,000  cubic  yards, 

at  Sl.OO $500,000.00 

Excavation   below   elevation   885,    100.000   cubic 

yards,  at  S1.50 150,000.00 

Concrete,  lining,  10,000  cubic  yards,  at  S8.00 80,000.00 

Concrete  walls,  5000  cubic  yards,  at  S8.00 40,000.00 

Concrete,  miscellaneous,  2000  cubic  yards,  at  SIO.  20,000.00 

Steel  reinforcing.  100  tons,  at  SIOO.OO 10,000 .00 

$800,000.00 

Cofferdams,  earth  and  rock  fill:  ■ 

Upper  cofferdam,  300,000  cubic  yards,  at  Sl.OO .  .  .        300.000.00        ' 
Lower  cofferdam   (first  structure),   100,000  cubic 

yards,  at  Sl.OO 100,000 .00 

Lower   cofferdam    (replacing    structure),    100,000 

cubic  yards,  at  Sl.OO 100,000 .00 

500,000.00 

Extraordinary  contingency  (insurance  allowance) . . .  500,000 .  00 

Main  dam: 

Excavation  below  elevation  885,  350,000  cubic  yards, 

at  S2.00 700,000 .00 

Excavation  above  elevation  885.  50,000  cubic  yards, 

at  Sl.OO 50,000.00 

Concrete.  760,000  cubic  yards,  at  S6.00 4,560,000 .00 

Movable  dam  crest 190.000 .00 

6,500,000.00 

Forebay  and  penstocks: 

Tunnel  excavation,  32.000  cubic  yards,  at  S5.00 160,000.00 

Tunnel  lining  concrete,  6000  cubic  yards,  at  S8.00. .  .  40,000.00 

Open  excavation,  200,000  cubic  yards,  at  Sl.OO 200.000.00 

Forebay  walls,  concrete,  20,000  cubic  yards,  at  S7.00  140,000.00 

Penstock  cradles,  concrete.  5000  cubic  yards,  at  S8.00  40,000 .00 

Gates,  trash  racks,  etc 200,000, 00 

Penstocks,  1500  tons,  at  S140.00 210,000 .00 

Reinforcing  steel,  100  tons,  at  SIOO.OO 10,000.00 

1,000,000.00 

Power  and  transformer  house: 

Excavation,  60,000  cubic  yards,  at  Sl.OO 60,000.00 

Concrete,  20.000  cubic  yards,  at  S8.00 160,000.00 

Concrete,  5000  cubic  yards,  at  S12.00 60.000.00 

Reinforcing  steel,  100  tons,  at  SIOO.OO 10.000 .00 

Roof,  crane,  etc 60,000 .00 

350,000 .  00 

Right  of  way  (assumed) 150.000 .00 

Summation  of  above  items 8,800,000 .00 

Engineering  and  contingencies,  26  per  cent 2,200,000.00 

Interest  during  construction,  2 i  years  at  4  to  10  per  cent  1 ,100,000 .  00 

$12,100,000.00 

Hydro-electric  equipment: 
Turbines,  venerators,  exciters,  and  governors,  7  units, 

26,000  Kw.  each,  at  S200.000 .00 1,400,000 .00 

Switchboard,  plant  wiring,  etc 200,000.00 

Transformers.  1.50,000  Kw 600,000 .00 

Freight,  erection  and  installation 400.000 .00 

Summation  of  above  items S2,500,000 .00 

Engineering  and  contingencies,  20  per  cent 600,000 .00 

Interest,  }  yr.  at  4  per  cent  say  3  J  per  cent 100,000 .00 

3,100,000.00 

Total  for  project SI 6,200,000 .00 

200,000  E.H.P.  continuous  development  at  S76.00  per 
E.H.P. 


Digitized  by 


Google 


734  ECONOMICAL  ASPECTS 

Estimate  of  Cost,  100,000  Electrical  Horse-power  Additional  Power 

Forebay,  penstocks,  power-house  and  tailrace: 
Additional,  including  25  per  cent  for  engineer- 
ing and  contingencies  and  10  per  cent 
interest  during  construction $1,000,000.00 

Additional  equipment,  including  20  per  cent  for 
engineering  and  contingencies,  and  3 
per  cent  for  interest,  100,000  H.P 1,500,000.00 


Summation '  $2,500,000.00 

This  is  the  total  additional  cost  to  supply  100,000  horse-power 
additional  power  during  the  part  of  the  time  for  which 
tHe  flow  of  the  river  is  in  excess  of  15,000 
second-feet. 

Estimated  cost  of  storage  to  maintain  a  mini- 
mum flow  of  15,000  second-feet,  500,000 
acre-feet 2,000,000.00 


Total  additional $4,500,000.00 

Total  for  200,000  H.P.  project  (preceding  esti- 
mate)   15,200,000.00 


Total  for  project $19,700,000.00 

300,000  E.H.P.  at  approximately  $66.00  per  E.H.P. 

Estimate  of  Annual  Cost 

For  200,000  Electrical  Horse-power  Continuous  Development  . 

This  estimate  has  been  made  on  the  basis  of  the  following  assumptioDs: 
Interest  rate  4  per  cent,  assumed  life  of  dams,  forebay,  substructure  of  power- 
house, and  tailrace,  fifty  years.  Assumed  life  of  movable  crest  gates,  trash 
racks,  penstocks,  superstructure  of  power-house  and  equipment  fifteen  years. 
Annual  replacement  fund,  for  fifty-year  life  portion,  $10,800,000 

at  i  per  cent $72,000.00 

For  fifteen-year  hfe  portion,  $4,400,000  at  5  per  cent 220,000.00 

Aimual  interest,  $15,200,000.00,  at  4  per  cent 608,000.00 

Annual  maintenance  and  repairs . 60,000.00 

Attendance  and  administration 80,000.00 


Total  annual  cost,  200,000  E.H.P.  development.  . .  $1,040,000.00 
Aimual  cost  per  E.H.P.  on  basis  of  100  per  cent  load 

factor $5.20 

Additional  25  per  cent 1 .  30 


Aimual  cost,  if  only  80  per  cei^t  of  the  power  is  used    $6.50 

These  costs  are  based  upon  utilization  of  the  power  immediately  upon 
completion  of  the  project. 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  735 

Estimate  of  Annual  Cost 

For  300,000  Electrical  Horse-power  Continuous  Development 

Based  upon  similar  assumptions  to  those  for  the  200,000  E.H.P.  development. 
'  Annual  replacement  fund,  for  fifty-year  life  portion,  $13,200,000 

at  J  per  cent $88,000.00 

For  fifteen-year  life  portion,  $6,500,000 .  00,  at  5  per  cent .  325,000 .  00 

Annual  interest,  $19,700,000.00,  at  4  per  cent 788,000.00 

Annual  maintenance  and  repairs 90,000.00 

Attendance  and  administration 119,000.00 

Total  annual  cost $1,410,000.00 

Annual  cost  per  E.H.P.  of  base  load $4.70 

Additional  25  per  cent 1 .  20 

Annual  cost  if  only  80  per  cent  of  the  power  is  used. .  $5 .90 


Estimated  Cost  op  Proposed  Columbia  River  Project 
Capacity: 

480,000  horse-power 12  months  per  year 

600,000  horse-power 11  months  per  year 

700,000  horse-power 10  months  per  year 

800,000  horse-power 8  months  per  year 

The  following  cost  estimate  on  this  proposed  extensive  devel- 
opment is  taken  from  an  article  by  Mr.  L.  F.  Harza  in  the  Journal 
for  Electricity,  Power  and  Gas,  for  March  18, 1916,  €md  the  readers 
interested  in  this  unusual  development  are  referred  to  the  long 
series  of  articles  appearing  in  said  journal  during  1915  and 
1916. 

Contingent  Margin,  The  total  cost  of  each  item  as  given  in 
the  estimates  which  follow  all  include  a  margin  of  25  per  cent  to 
cover  engineering,  administration  during  construction,  and  con- 
tingencies in  addition  to  the  amoimts  obtained  by  applying  the 
foregoing  unit  prices,  except  in  the  case  of  the  generating  machin- 
ery; in  this  case  only  15  per  cent  was  allowed,  as  these  estimates 
are  based  upon  the  higher  of  two  or  more  actual  quotations  in 
nearly  all  cases,  and  the  manufacturer  himself  would  furnish  the 
engineering  talent  except  for  erection,  which  item  has  been  in- 
cluded in  the  estimate. 


Digitized  by 


Google 


736  ECONOMICAL  ASPECTS 

ESTIMATE   OF   CAPITAL   COST 
Dam  for  closing  present  channel: 

Scheme  A  +26  per  cent $3,325,000 

Scheme  B  +25  per  cent 2.288,000 

Scheme  C  +26  per  cent 3,344,000 

Scheme  D  +26  per  cent 3,056,000 

Scheme  £  +25  per  cent 3,485,000 

Scheme  F  +25  per  cent 3,419.000 

Use  for  estimate .  - $3,350,000 

Controlling  dam: 

Camere  type  of  dam;~approximato  quantities  as  de- 
signed for  81  feet  controlled  depth. 
26,000  tons  structural  steel. 
4,000  tons  cast  steel. 
230,000  cubic  yards  of  concrete. 
1  traveling  gantry  crane. 

Estimated  cost,  reduced  25  per  cent,  for  67  feet  con- 
trolled depth  plus  26  per  cent  contingent  fund $3,861,000 

Tainter-gate  type  of  dam,  approximate  quantities  as 

designed  for  81  feet  controlled  depth. 
41,600  tons  structural  steel. 
21,800  tons  cast  steel. 
480  tons  steel  cable. 
312,460  cubic  yards  concrete. 

Estimated  cost,  reduced  26  per  cent,  for  67  feet  con- 
trolled depth  plus  26  per  cent  contingent  fund 8,837,000 

Use  for  estimate  of  controlling  dam 8.837.000 

Flood  channel: 

Approximate  quantities: 

2,078,000  cubic  yards  rock  excavation,  above  elevation 

84.0  (sill  of  flood  gates)  plus  25  per  cent 2,078,000 

Diversion  channel: 

Approximate  quantities: 

1,243,000  cubic  yards  rock  excavation  for  diversion 

channel  below  elevation  84.0. 
140,500  cubic  yards  concrete. 
810,000  F.B.M.  timber  for  cribs. 

8,000  cubic  yards  rock  fill  in  cribs. 
Estimated  cost  of  diversion  channel  and  closure  of  same 

plus  25  per  cent 2,872,000 

Ice  and  drift  sluice,  Oregon  side: 
Approximate  quantities: 
252,000  cubic  yards  rock  excavation. 
28,300  cubic  yards  concrete. 

320  tons  structural  steel  rollers. 
Estimated  cost  plus  25  per  cent 452,000 

Wing  walls  for  rock  fill  dam: 
Approximate  quantities: 
42,600  cubic  yards  concrete  plus  25  per  cent 266,000 

Main  floating  boom  and  piers: 
Approximate  quantities: 
11.394,000  f.b.m.  of  timber. 

1,055  tons  of  rods  and  drift  pins. 
3,000  cubic  yards  concrete. 
46,000  cubic  yards  rock  fill  in  piers. 
Estimated  cost  plus  25  per  cent '. . .  493.000 


Digitized  by 


Google 


(X)ST  OF  HYDRO-ELECTRIC  POWER  PLANTS  737 

Power  canal: 

Approximate  quantities: 
4.229,000  cubic  yards  rock  excavation. 
136.000  cubic  yards  rubble  walls. 
17.960  cubic  yards  concrete  lining. 
1,000,000  cubic  yards  sand  excavation. 
Two  floating  booms. 
22.000  cubie  yards  concrete. 

1 10  tons  structural-steel  roller  dama. 

Estimated  cost  plus  25  per  cent $5,394,000 

Jetty  at  intake  to  power  canal: 
Approximate  quantities: 
4.430,000  f.b.m.  of  timber. 
665.000  pounds  rods  and  drift  pins. 

2,470  cubic  yards  reinforced  concrete. 
164.000  cubic  yards  rock  fill. 

73,000  cubic  yards  sand  excavation 

Estimated  cost  plus  25  per  cent 285,000* 

Rebuilding  Five  Mile  Lock: 

Raising  walls  and  gates  and  building  draw  span,  plus 

25  per  cent 106,000 

Foiebay  and  power-house  substructure: 
Approximate  quantities 
1,584,000  cubic  yards  dry  rock  excavation. 
137.500  cubic  yards  rock  excavation  for  removal  of  cofferdam 
429,250  cubic  yards  concrete. 
5,000,000  pounds  steel  reinforcement. 
3.500.000  pounds  structural  steel  for  penstock  gates. 
2,300.000  pounds  cast  steel  for  penstock  gates. 
1,024,000  pounds  steel  trash  racks. 

24  filler  gates  and  drain  gates. 
$375,000  for  cofferdamming  and  pumping. 
Estimated  cost  plus  25  per  cent 6.852,000 

Power-house  superstructure: 

76  feet  by  1670  feet  station  building. 

Fishway. 

Tunnel  through  building  for  railroad. 

Steel  bridges  for  spanning  forebay  and  tailrace. 

Estimated  cost  plus  25  per  cent 1,475.000 

Power-house  machinery: 

23  vertical  shaft  35,000-Kw.  (50,000  Kv.A.)     25-cycle. 
11,000-volt,  75-R.P.M.,  3-pha8e  generators,  including 
stator  and  rotor,  but  not  shaft  or  bearings. 

28  mechanically  driven  exciters.  500  Kw.  each;  switch- 
b'>ard,  low-tension  oil  switches,  busbars,  and  all  mis- 
cellaneous electrical  equipment. 

88  50,000-H.P.  vertical  shaft  76  R.P.M.  turbine  units, 
including  shaft  and  oil  bearings,  governors,  and  oil 
system. 

2  250-ton  traveling  cranes  in  power  house  and  2  50-ton 
traveling  gantry  cranes  serving  penstock  gates;  mis- 
cellaneous small  equipment. 

Estimated  cost  plus  15  per  cent 12,353,000 

Reconstruction  of  railroads: 

Total  estimated  cost  plus  25  per  cent 687,000 

Other  property  damage 904,000 


Total  physical  cost $45,404,000 

Add  for  interest  during  one-half  of  five-year  construc- 
tion period  at  4  per  cent  equals  10  per  cent 4,540.000 

Total  estimated  capital  cost $49,944,000 

Use  for  total  capital  cost 50,000.000 


Digitized  by 


Google 


738  ECONOMICAL  ASPECTS 

Annual  Cost  of  Generating  Primary  Power.  The  following 
items  are  independent  of  the  interest  rate  on  capital  investment: 

Depreciation — ^Reserve  fund  assumed  to 
earn  2  per  cent  interest  and  sufficient  to 
replace  all  depreciable  parts  every  fifteen 
years,  and  to  refund  the  cost  of  all  nearly 
permanent  structures,  rock  excavation, 
concrete,  etc.,  every  fifty  years  (average 
value  3  per  cent) $1,500,000.00 

Maintenance  and  repairs — For  maintenance 
and  repairs  on  the  turbine  units,  in  addi- 
tion to  depreciation  fund,  per  annum. . .      $112,800.00 

Maintenance  and  repairs  to  generators  and 
electrical  equipment,  1}  per  cent 74,000.00 

Repairs  to  movable  dam 50,000 .00 

Painting,  average  of  one  coat  per  annum, 
43,700  tons  of  exposed  steel  (total  in  use) 
at  $1  per  ton 43,700.00 

Operating  suction  dredge  to  prevent  possible 
accimiulation  of  sand  bar  at  canal  intake, 
300  days,  $100  per  day 30,000.00 

Maintenance  of  building,  replacing  roof 
every  five  years  plus  50  per  cent  for  other 
repairs 2,400.00 

Contingent  maintenance  and  repair  expense         50,000 .  00 

Total  for  maintenance  and  repairs 362,900.00 

Attendance  and  administration 100,000.00 

Total  annual  expense  exclusive  of  interest  $1,962,900  00 

The  rate  of  interest  to  be  paid  on  the  capital  investment  will 
depend  largely  upon  the  basis  of  financing.  To  show  the  relation 
of  this  to  the  annual  cost  of  power,  interest  rates  of  3  and  4  per  cent 
have  been  assumed  as  representing  public  development  under  dif- 
ferent conditions.  There  has  also  been  assumed  a  rate  of  6  per 
cent  on  securities  originally  discounted  10  per  cent,  plus  1  per 
cent  taxes,  this  basis  being  intended  to  represent  approximately 
the  cost  under  corporate  financing.  The  results  are  as  follows: 
No  sinking  fimd  has  been  provided,  as  it  is  not  properly  chargeable 
to  the  cost  of  generation.  The  depreciation  or  amortization  fund 
would  provide  for  keeping  the  project  permanently  in  first-class 
operating  condition.  A  water-power  property  is  of  such  unques- 
tionably permanent  value  as  to  make  it  imnecessary  to  recover 
the  principal  in  a  short  time  as  with  many  industrial  enterprises 


Digitized  by 


Google 


COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  739 

which  are  subject  at  any  time  to  the  necessity  of  coxnpiete  Uquida- 
tion  due  to  unforeseen  competition.  In  the  case  of  corporate 
finance,  especially,  a  sinking  fund  might,  however,  assist  in  secur- 
ing easier  terms  in  marketing  the  securities,  but  in  any  event  is 
amply  covered  by  the  25  per  cent  contingent  fund.  A  50-year 
sinking  fimd  drawing  2  per  cent  interest  would  involve  an  annual 
expense  of  $1.20  per  continuous  electrical  horse-power. 

Three  per  cent  basis: 

Depreciation,  maintenance  and  repairs  as  above . .  $1,962,900 .  00 

3  per  cent  interest  on  $50,000,000 1,500,000.00 

Total  annual  charges $3,462,000.00 

Annual  cost  per  peak  electrical  horse-power  year 

of  base  load  (480,000  H.P.) $7.22 

Add  25  per  cent 1 .80 

Use $9.02 

Four  per  cent  basis: 

Depreciation,  maintenance  and  repairs  as  before.  $1,962,900.00 

4  per  cent  interest  on  $50,000,000 2,000,000.00 

Total  annual  charges $3,962,900.00 

Per  peak  horse-power  year 8. 27 

Add  25  per  cent 2.07 

Use $10 .  34 

Six  per  cent  basis: 

Depreciation,  etc.,  as  before $1,962,900.00 

Add  for  6  per  cent  on  securities  originally  sold  at 

10  per  cent  discount,  equivalent  to  6.67  per  cent  3,340,000 .  00 

^dd  for  taxes  1  per  cent 500,000.00 

Total  annual  charges $5,802,900.00 

Cost  of  power  on  usual  basis  of  private  enterprise 

per  peak  horse-power  per  year 12 .  10 

Add  25  per  cent 3.03 

Use $15 .  13 

Cost  of  Generation  Contingent  upon  Sale  of  Surplus  Power.  If 
the  sale  of  the  surplus  power  is  to  be  assumed,  then  an  additional 
item  of  depeciation  should  be  added  to  provide  for  the  possibility 
of  severe  runner  erosion  for  the  low-head  units  when  operating  at 


Digitized  by 


Google 


740  ECONOMICAL  ASPECTS 

heads  above  80  feet.  The  value  of  one  runner  including  freight 
and  erection  would  be  about  $27,000. 

About  seven  low-head  units  are  required  to  operate  at  80  foot- 
head  to  produce  800,000  H.P.  with  a  decreasing  number  at  the 
higher  heads  where  the  erosion  would  be  most  severe.  If  we 
assimie  to  replace  all  seven  runners  every  three  years,  the  annual 
additional  charge  would  be  $63,000,  say  $75,000.  This  item  is 
very  small  compared  with  the  additional  profit  which  the  surplus 
power  should  bring. 

It  might  be  assumed  roughly  that  eleven  months'  surplus 
power  be  worth  80  per  cent  of  the  value  of  continuous  power,  ten 
months'  power  60  per  cent  and  eight  months'  power  30  per  cent. 

If  the  various  prices  now  be  weighted  according  to  the  amount 
available,  and  using  the  price  of  primary  power  as  unity,  there  will 
result: 

480,000X1.00=480,000 
120,000  X  .80=  96,000 
100,000  X  .60=  60,000 
100,000  X    .30=  30,000 


800,000  actual  or  666,000  weighted  power 

The  quotient  of  these,  totals  or  0.8333,  now  represents  the  aver- 
age unit  value  of  all  power,  as  a  proportion  of  the  value  of  primary 
power,  and  666,000  represents  the  equivalent  primary  power  to 
produce  the  same  income.  If  all  power  were  to  be  sold  at  prices 
bearing  the  above  ratio  to  each  other,  the  actual  costs  of  pro- 
duction of  primary  power  would  then  be  obtained  by  first  adding 
$75,000  to  the  annual  charges  and  then  dividing  by  666,000. 

Based  upon  3  per  cent  interest: 

Former  annual  charge $3,462,900.00 

Add  for  runner  depreciation 75,000.00 

Total $3,537,900.00 

Add  25  per  cent 884,000.00 

Use $4,421,900,00 

Cost  per  peak  primary- horse-power $6.63 

Cost  per  11  mo.  surplus  H.P 5 .  30 

Cost  per  10  mo.  surplus  H.P 4.00 

Cost  per    8  mo.  surplus  H.P 2.00 


Digitized  by 


Google 


'  COST  OF  HYDRO-ELECTRIC  POWER  PLANTS  741 

Based  upon  4  per  cent  interest: 

Former  annual  charge $3,962,900.00 

Add  for  runner  depreciation 75,000.00 

Total $4,037,900.00 

Add  25  per  cent 1,009,500.00 

Use $5,047,400.00 

Cost  per  primary  horse-power 7.58 

Cost  per  11  mo.  surplus  H.P 6.06 

Cost  per  10  mo.  surplus  H.P. 4 .  55 

Cost  per   8  mo.  surplus. 2.27 

Based  upon  6  per  cent  interest — on  securities  sold  at  90: 

Former  annual  charge $5,802,900.00 

Add  for  runner  depreciation 75,000,00 

Total $5,877,900.00 

Add  25  per  cent 1,469,500.00 

Use $7,347,400.00 

Cost  per  primary  horse-power $11 .  02 

Cost  per  11  mo.  surplus  H.P 8.82 

Cost  per  10  mo.  surplus  H.P 6.62 

Cost  per   8  mo.  surplus  H.P.. 3.31 

The  computations  for  the  capital  cost  and  cost  of  power  for  the 
case  in  which  a  period  of  ten  years  was  allowed  for  building  up  the 
load,  were  made  by  starting  with  the  initial  investment  necessary 
to  deliver  one-tenth  of  the  power,  and  then  progressively  adding 
for  each  year  the  deficit,  or  difference  between  interest  on  the  pre- 
viously accumulated  investment,  operating  expenses,  etc.,  and 
the  earnings  of  the  year  in  question,  to  the  investment  of  the  pre- 
vious year.  It  was  necessary  first  to  assume  a  price  of  power  and 
after  computing  the  transactions  of  the  ten-year  period,  to  then 
correct  this  assumption  by  a  process  of  successive  approximations 
until  an  assiunption  was  made  which  provided  the  desired  25  per 
cent  margin  at  the  end  of  the  ten-year  period. 


Digitized  by 


Google 


742  ECONOMICAL  ASPECTS 

COST  OF  POWER » 

The  cost  of  hydro-electric  power  can  be  considered  as  made 
up  of  two  parts:  The  fixed  charges  and  the  operating  expenses. 
These,  in  turn  are  made  up  as  follows: 

Fixed  Charges: 

Interest  on  investment. 

Taxes  and  insurcknce. 

Depreciation. 
Operating  Expenses: 

General  administration. 

Labor. 

SuppUes. 

Maintenance  and  repairs. 

In  estimating  the  cost  of  power  a  thorough  distinction  must,  as 
previously  stated,  be  made  between  the  cost  of  the  same  at  the 
generating  station  bus-bars  and  the  cost  when  delivered  to  the 
customer.  In  the  former  case  the  cost  should  be  based  on  only 
such  portions  of  the  charges  and  expenses  which  are  applicable 
to  the  generating  station,  while  in  order  to  obtain  the  cost  of  power 
delivered,  the  total  expenses  must,  of  course,  be  considered. 

The  rate  of  interest  on  the  investment  varies  cmd  depends  on  the 
risk  involved.  In  risky  undertakings  the  rates  of  interest  are 
higher  than  where  greater  safety  obtains,  and  if  money  put  into 
new  enterprises  involving  risk  of  loss  were  not  allowed  to  earn  any 
more  than  a  normal  rate  of  interest,  it  would  be  poor  policy  for 
the  inventor  to  put  his  money  in  such  undertakings.  Bonds, 
therefore,  should  draw  the  lowest  rate  of  interest  because,  as  a 
rule,  they  are  safe,  being  secured  by  a  mortgage  on  the  property. 
So,  for  example,  many  government  bonds  draw  only  an  interest  of 
3  per  cent  because  there  is  no  risk  involved.  The  rate  on  public 
service  bonds,  on  the  other  hand,  is  higher,  averaging  about  5  per 
cent,  but,  of  course,  when  they  are  sold  at  a  discount  the  actual 
interest  earned  by  the  investor  is  greater.  The  interest  on  the  stock, 
however,  which  cannot  be  declared  until  the  bond  interest  has 
been  paid,  should  be  enough  higher  than  the  normal  interest  to  com- 
pensate for  the  lesser  security.  A  rate  at  least  2  per  cent  higher 
than  prevailing  bank  rates  seems  justifiable  and  conrniissions  are 
frequently  approving  rates  of  retiu^n  of  7  per  cent  and  8  per  cent. 
1  See  previous  section  for  actual  and  estimated  costs. 


Digitized  by 


Google 


COST  OF  POWER 


743 


The  second  item  under  the  fixed  charges  is  taxes  and  insurance. 
The  amount  necessarily  depends  on  the  rates  available,  but,  for 
.estimating  purposes  it  is  common  practice  to  allow  ^  per  cent  of 
the  physical  cost  for  each,  making  a  total  of  1  per  cent. 

Depreciation  is  the  loss  in  value  which  occurs  during  the  period 
which  the  property  is  in  service,  either  due  to  wear  and  tear  or 
obsolescence,  and  a  certain  sum  of  money  must  be  set  aside 
annually  for  renewing  this  property.  There  are  different  methods 
of  providing  for  depreciation,  but  the  sinking  fund  or  annuity 

TABLE  LXIV 


Property. 

Dams,  masonry 

Pipe  lines,  iron 

Pipe  lines,  wood-stone 

Power-house  building,  fire-proof. . 

Water-wheels 

Generators 

Transformers 

Switching  equipment 

Miscellaneous  auxiliaries 

Transmission  lines,  steel  towers .  . 
Transmission  lines,  wood  poles. . . 

Underground  cable  system 

Service  transformers 


Total  Life.  Years. 


50 
30-40 
15-25 
50-75 

20 

20 

20 
12-15 

10 
25-30 

15 
20-25 

15 


method  is  best  applicable  to  public  utility  properties.  It  pro- 
vides for  setting  aside  each  year  a  sum  that,  invested  in  a  certain 
rate  of  interest  compounded  annually,  it  will  equal  the  cost  of  the 
property,  less  its  scrap  value,  at  the  end  of  its  assumed  life.  Thus, 
if  a  certain  portion  of  a  plant  costs  S10,000  and  has  a  life  of  ten 
years,  with  a  scrap  value  of  10  per  cent  or  $1000,  and  it  is  desired 
to  set  aside  such  a  sum  that,  at  5  per  cent  interest  compounded 
annually,  will  accumulate  an  amount  equal  to  the  cost,  less  the 
scrap  value,  at  the  end  of  the  life  period,  it  will  then  be  found,  by 
referring  to  an  annuity  table,  that  $9000X0.0795  or  $715.50 
annually  will  produce  the  required  amount.  As  the  life,  as  well 
as  the  scrap  value  of  the  different  elements  varies  to  a  consider- 
able extent,  the  depreciation  should  be  figured  separately  for  each 
item,  and  thereafter  averaged. 

The  useful  life  of  the  plant  apparatus  or  equipments  is  purely 


Digitized  by 


Google 


744 


ECONOMICAL  ASPECTS 


a  speculative  matter,  and  past  experience,  knowledge  of  the  art 
and  careful  judgment  must  be  exercised  in  arriving  at  the  prob- 
able life  of  apparatus  and  property.    See  Table  LXIV. 

The  operating  exp)enses,  which  include  labor,  repairs,  main- 
tenance and  supplies,  will  vary  with  the  amount  of  power  manu- 
factured, that  is,  the  load  factor.  They  are,  however,  by  no  means, 
proportional  to  it  and  form  a  much  smaller  part  of  the  total  cost 
than  with  steam  stations,  where  the  fuel  expenses  come  in  and 
where  both  labor,  repair  and  supply  items  are  much  higher. 
Based  on  a  50  per  cent  load  factor,  the  operating  expenses  may 
range  anywhere  from  0.3c.  per  Kw.-hr.  for  a  small  station  to 
0.02c.  or  less  for  a  large  station.  Some  approximate  representa- 
tive values  are  giyen  in  the  following: 

TABLE  LXV 
Operating  Expenses  of  Hydro-electric  Stations 


station  Capacity 
inKw. 

Operating  Expenses 
In  Cents  per  Kw.-hr. 

station  Capacity 
In  Kw. 

operating  ExpeuM 
in  Cents  per  Kw.-hr. 

2,500 

5,000 

10,000 

15,000 

0.1 
0.08 
0.06 
0.05 

25,000 

50,000 

75,000 

100,000 

0.04 
0.03 
0.02 
0.015 

TABLE  LXVI 

Approximate  Cost  of  Steam  Turbine  Stations  and  Power 

(Based  on  Coal  at  S3.00  per  Ton) 


Cost  per  Kw.-hr.  in  Cents. 

Capacity  of  Station 

Cost  of  Station 

Load  Factor. 

in  Kw. 

per  Kw. 

60  Per  Cent. 

76  Per  Cent, 

500 

S95.00 

1.02 

0.86 

1,000 

80.00 

0.88 

0.74 

2,000 

75.00 

0.77 

0.63 

3,000 

70.00 

0.69 

0.58 

4,000 

65.00 

0.66 

0.54 

5,000 

62.50 

0.62 

0,51 

7,500 

60.00 

0.57 

0,48 

10,000 

57.50 

0.53 

0.45 

15,000 

55.00 

0.51 

0.43 

25,000 

50.00 

0.48 

0.41 

Digitized  by 


Google 


COST  OF  POWER 


745 


Digitized  by 


Google 


CHAPTER  XI 
ORGANIZATION  AND  OPERATION 

Management.  The  measure  of  financial  success  attained  in  a 
hydro-electric  development  is  to  the  greatest  extent  measured 
by  the  skill  and  judgment  of  its  management.  The  department 
heads  should  study  the  men  whom  they  employ  and  also  the  prob- 
lem of  handling  them  to  the  best  advantage.  It  should  be  the 
object  of  the  department  chiefs  so  to  dispose  both  men  and 
material  that  their  possibiUties  will  be  best  realized.  An  ade- 
quate system  of  records  should  be  kept  showing  what  the  several 
departments  are  doing,  and  promptness  and  completeness  in  this 
respect  should  be  insisted  upon.  Regular  meetings  between  the 
department  heads  and  their  men  is  advisable,  and  many  com- 
panies have  inaugiu^ted  suggestive  systems  by  which  suggestions 
for  the  improvement  of  the  operating  and  service  conditions  are 
invited,  prizes  being  given  at  regular  intervals  for  the  best  sug- 
gestions received. 

The  organization  of  a  hydro-electric  company  naturally  varies 
considerably,  not  only  depending  on  the  size  of  the  system,  but 
also  on  the  nature  of  the  same.  An  idea  of  the  extensive  force 
required  by  a  large  company  such  as  that  of  the  Great  Western 
Power  Company,  is  obtained  by  referring  to  the  chart  given  in 
Fig.  406. 

Operating  Force.  The  selection  and  maintenance  of  an 
efficient  and  reliable  operating  force  is  also  essential,  as  upon  the 
same  dej)ends  the  quaUty  of  service  rendered.  Most  modem 
systems  of  any  size  have  a  method  of  operation  which  corresponds 
to  that  of  a  train  dispatcher  on  steam  railroads,  and  where  many 
different  plants  are  attached  to  the  same  network,  this  becomes 
practically  necessary.  A  load  dispatcher  is  located  at  some  con- 
venient point,  which  often  is  not  at  a  power-house,  and  is  phced 
in  charge  of  the  whole  system  and  personally  directs  every  opera- 
tion in  all  stations.  He  is  in  telephone  communication  \\nth  all 
operators  and  keeps  a  record  of  the  changes  and  connections  made 

746 


Digitized  by 


Google 


OPERATING  FORCE  747 

in  each  part  of  the  system  by  means  of  a  system  of  pins  and 
markers  on  a  large  map  or  plan  of  the  circuits  and  apparatus  of 
the  plant.  He  receives  at  regular  intervals  readings  of  loads, 
water  conditions,  etc.,  which  he  marks  down  on  the  record  sheet 
before  him,  and  from  these  records  and  recording  instruments  in 
his  office  he  is  able  to  keep  close  watch  on  the  conditions  and 
make  changes  in  load  generation,  voltage,  frequency,  gate  open- 
ings, etc.,  in  order  to  obtain  the  most  satisfactory  and  efficient 
operation. 

The  real  value  of  a  load  dispatcher  looms  up  under  abnormal 
or  trouble  conditions.  When  trouble  affects  the  system  it  is 
instantly  apparent  on  the  recording  instruments.  The  system 
operator  inmiediately  gets  into  communication  with  the  station 
affected  and  in  case  of  transmission  hne  trouble  learns  what 
switches  have  opened  and  then,  if  possible,  gives  orders  to  cut 
over  to  duplicate  lines.  The  faulty  hne  receives  one  or  two 
trials,  either  at  full  voltage  or  by  bringing  the  voltage  up  slowly 
on  separate  generators.  If  the  short  or  trouble  still  shows  up  on 
the  Une  ammeters,  the  hne  is  cut  up  into  sections,  according  to 
the  judgment  of  the  system  operator,  and  tried  until  the  faulty  • 
section  is  located.  Patrolmen  and  repair  men,  who  are  on  con- 
stant call,  then  receive  directions  for  making  the  repairs.  In  the 
case  of  trouble  on  the  distribution  system,  as,  for  instance,  where  a 
feeder  will  not  stay  in  owing  to  a  short  on  the  line,  it  is  imme- 
diately reported  and  turned  over  to  the  line  department,  which 
looks  after  the  repairing  of  the  line.  In  case  of  trouble  with  the 
underground  system,  the  system  operator  supervises  the  locating 
and  disconnecting  of  the  faulty  feeder  and  then  notifies  the  under- 
ground department,  whose  business  it  is  to  repair  the  trouble. 
In  case  of  trouble  of  a  serious  nature,  the  heads  of  the  depart- 
ments affected  are  notified  and  take  active  charge  of  the  situation. 

The  organization  of  the  operating  force  of  a  hydro-electric 
generating  station  is  necessarily  less  complicated  than  in  a  steam 
station.  It  is  determined  largely  by  the  location  and  the  arrange- 
ment, and  there  are  so  many  different  conditions  in  such  systems 
that  it  is  impossible  to  reconmiend  any  exact  form  of  organization, 
as  really  no  two  can  be  quite  alike.  If  the  station  is  not  too  large 
it  is  desired  to  have  the  hydrauhc  superintendent  report  to  the 
station  superintendent,  but  if  the  development  is  of  such  a  mag- 
nitude as  to  require  the  entire  time  of  a  superintendent  for  each 


Digitized  by 


Google 


748  ORGANIZATION  AND  OPERATION 

of  the  departments  under  consideration,  a  position  is  warranted 
for  a  man  to  whom  both  electric  and  hydraulic  superintendents 
will  report,  thus  still  bringing  the  responsibility  of  operation  of  the 
two  departments  under  one  head. 

As  a  general  rule,  for  the  same  capacity  installed,  a  plant 
having  horizontal  units  can  get  along  with  a  smaller  force  than  one 
using  vertical  units.  It  is  a  general  practice  to  maintain  one  man 
at  all  times  on  each  of  the  different  levels  or  floors  of  the  power- 
house, such  as  the  switchboard  gallery,  the  main  floor  and  the 
basement,  where  with  vertical  units  the  turbines  proper,  as  well 
as  the  oil  pumps  and  other  auxiliaries  are  located.  The  man  in 
the  basement  could,  in  all  probabiUty,  be  dispensed  with  in  plants 
using  horizontal  units.  In  addition  to  these  men  a  chief  operator 
should  be  provided  for  each  shift,  whose  duties  should  carry  him 
to  all  parts  of  the  building.  For  a  very  large  station  the  above 
force  may  be  entirely  inadequate,  cmd  for  small  plants  the  force 
may  be  reduced. 

The  switching  operations  are  determined  by  the  general  method 
of  operation.  It  is  desirable  to  eliminate  all  high-tension  switch- 
ing under  load,  due  to  the  fact  that  such  switching  may  set  up 
surges  which  may  discharge  into  the  transformers  and  cause 
resonance,  resulting  in  internal  disturbances  in  the  same.  When 
a  line  is  to  be  cut  into  service,  the  high-tension  switches  in  the 
main  and  substation  should  be  closed  first,  then  the  low-tension 
transformer  switch  in  the  generating  station  should  be  closed, 
energizing  the  transformers  and  the  Une,  after  which  the  low- 
tension  transformer  switch  in  the  substation  is  closed  cmd  the  load 
picked  up.  In  case  it  becomes  necessary  to  open  a  high-tension 
switch  in  a  loaded  line,  the  circuit  should,  if  possible,  first  be 
parallel  with  another  before  opening  the  switch.  If,  on  the  other 
hand,  transformers  are  to  be  paralleled  on  both  high-  and  low- 
tension  sides,  the  low-tension  switch  should  be  closed  first,  assum- 
ing that  the  low-tension  bus  is  energized.  Similarly,  in  cutting 
out  the  transformer  the  low-tension  switch  should  be  opened  last. 

Operating  Records.  One  of  the  essential  things  in  connection 
with  the  operation  of  hydro-electric  generating  stations  is  the 
keeping  of  acciu'ate  records.  Record  sheets  should  contain  only 
the  most  important  readings,  as  with  complicated  forms  the 
attendant  generally  reahzes  that  a  large  number  of  the  readings 
are  of  no  importance  and  for  this  reason  he  becomes  very  lax  in 


Digitized  by 


Google 


OPERATING   RECORDS 


749 


his  attention  to  the  readings  in  general,  and  as  a  consequence 
the  important  ones  may  suffer.    The  following  description  applies 


. 

t 

1 

\ 

I 

^ 

J 
i 

] 

\ 

;' 

t[ 

^ 

^ 

I   ' 

i 

: 

g 

» 

T 

1 

1 

\ 

j 

1 

^ 

1 

■1 

i 

i\ 

!' 

i 

i 

1 

1 

1 

2 

0 

- 

9 

fa 

9 

- 

3 

£ 

- 

0 

' 

□ 

0 

A 

-X. 

4 

ul 

"  -       -  ^  ** 

W 

1 

V 

s 

1 

' ' 

Si 

[ 

T1 
1 

CL 

z 

\ 

, 

J 

3 

^,1 

0 
>- 

Si 

t 

E 

* 

\ 

- 

** 

x> 

O 

1 

4—  — 

' 

1 

- 

1 

0 

< 

[T 

^  0        Z  ,  H 

« 

1 

V 

1 

1 

- 

1 

ll 

r 

o 

1- 
< 

1:1 
< 

ll 

t 

X 

H 
C 

* 

uJ 

* 

« 

" 

u 

a. 

E           J 

r   UJ   qp 

< 

^ 

d 

2 

d 

z 

c 

Z 

1 

i 

a. 

-r* 

^ 

CO 

•f* 

Lt 

«d 

t-*     (3& 

^    ' 

M        94 

l^run  HI 
lnjUHOJiMinii 

O 

13^1 

W 

Gl 

iwrt 

N 

VV 

^ 

G 

U^li^ 

X: 

1 

I 


I 


o 


to  €m  actual  record  sheet  for  a  medium-size  station  (Fig.  407), 
which  has  been  found  to  give  satisfactory  results.  The  sheet  is  of 
the  size  of  ordinary  letter  paper  and  is  ruled  for  hourly  records  of 


Digitized  by  CjOOQ IC 


750  ORGANIZATION  AND  OPERATION 

''  Water,"  /'  Main  Units,"  "  Cycles,"  "  Power-factor,"  "  Excit- 
ers,"  "  Transformers  and  Floodgates."  These  items  are  listed 
vertically  and  the  sheet  is  divided  into  24  vertical  columns,  one 
for  each  hour.  At  the  top  are  given  the  "  Forebay  "  readings  and 
"  Tailrace "  readings,  the  difference  between  which  gives  the 
"  Effective  Head."  Immediately  below  are  listed  the  indicated 
kilowatts  and  per  cent  gate  opening  of  each  generator  in  service, 
following  which  are  given  the  "  Total  Indicated  Kilowatts " 
and  "  Total  Per  Cent  Gate."  The  total  kilowatt  hom^  during  each 
hour,  as  read  from  the  watt-hour  meters,  is  plotted  as  a  block 
curve  extending  across  the  face  of  the  sheet. 

This  serves  as  a  better  record  for  the  actual  station  output 
than  the  indicated  kilowatts.  It  has  been  found  necessar>', 
however,  to  follow  the  indicated  kilowatts  to  serve  as  a  check 
on  the  efficiency  and  condition  of  the  units  in  general,  from  time 
to  time,  as  well  as  to  determine  what  capacity  would  be  required 
for  short  interval  peaks.  The  station  voltage  is  also  plotted  as  a 
block  curve  across  the  face  of  the  sheet. 

The  exciters  from  an  individual  group,  and  for  each  exciter 
the  voltage  current  and  per  cent  gate  opening  are  recorded. 

Transformer  records  are  Umited  to  the  temperatures.  These 
are  taken  hourly,  at  which  time  the  oil  elevation  is  noted  but  not 
recorded.  If  the  transformer  is  not  in  service  the  column  in 
which  the  temperature  is  listed  is  left  blank.  If  in  service  the 
temperature  is  taken  and  recorded. 

Under  the  item,  "  Floodgates  "  the  total  opem'ng  of  the  flood- 
gates in  feet  is  recorded,  rather  than  each  one  separately.  This 
record  is  maintained  daily,  the  flow  of  the  river  at  each  of  the 
stations  being  followed  very  closely. 

At  the  bottom  of  the  sheet  appear  the  daily  readings  of  the 
various  generator  and  feeder  watt-hour  meters  taken  at  mid- 
night of  each  twenty-four  hours.  The  following  items  are  also 
recorded  at  the  bottom  of  the  sheet:  "  Total  Generated,"  or  the 
total  output  of  the  station  for  twenty-four  hours;  the  "  Maximum 
Hour  Time,"  or  the  maximum  kw.-hr.  of  any  particular  hour 
during  the  day;  the  "  Maximum  Kw.  Time,"  or  the  maximum 
indicated  kilowatts  at  any  particular  instant;  the  "  Average 
Load,"  obtained  by  dividing  the  total  kilowatt  hours  generated  by 
24;  the  "Load  Factor,"  obtained  by  dividing  the  "Average 
Load  "  by  the  "  Max.  Kw.  Time";   the  "  Average  Flow  of  the 


Digitized  by 


Google 


Digitized  by 


Google 


- 

DAILY  ^ 

GE^ 

GENERATOR 

H     .                          GFNERATOR 

GENFRAl 

A4»l 

Ktto- 
wattt 

HV 

i  Ih  I'EKT 

"i 

KlhH 

watts 

»L 

oW 

A.*«l 
Abbhw 

iah>- 

watta 

HUSlI 

Tin*        AapMM 

Ptm- 

^ 

TMal 

Pr«M- 

ore 

V*M. 

Total 

^Mj 

M'd't 

!    i 

■ 

1 

• 

1            l1 

t 

10 

1 

u 

I 

!• 

' 

u 

ITd't 

Ar«rftg« 

ToUl 
Arerftge 

Total 

Avcrace  T 

INTEGRATING 

1        II 

OINEIIATOIIt 

II 

"^ 

Omi. 

1.    Oei 

1.             a«a. 

.Om. 

-55=s- 

*fiS?* 

II         .Cii««t 

■^' 

1 

If 

^ 

H 

K,.H.||                      1 

1 

GENE 

LOAD  DATA 

II                          PRECIPITATION                          ||                      STREi 

PMkLoad                          Kw. 

H  luinfon               1 

iBohee 

-l-M. 

L(MdFa«Cor                         f 

UsMwiftn 

.. 

H  »■•■« 

ItTbUleqiaTalentTAiafal 

•  • 

H   ap.SL 

11 

II 

"^ 

No.lii«.Wh6ebo|wmto4 

.._      _  Wheel 

Wbeel               Wkeell              Wkeelj              Wheel  ||Vo. 

1                 1                 ll<v« 

** 

TIME  AND  NATURE  OF  SWITCH!* 

■ 

SHIFT 

.       III*^*"!!!"! 

,lLlo        IL 

Itt  Operator 

tod       « 

Srd 

4th               4. 

Fl 


Digitized  by 


Google 


LOG 


—   191_. 


GENERATOR        j 

GENERATOR         I 

ENTIRE  STATION         | 

Time 

A«»l 

Ktlo- 

WfttU 

HD 

piN 

ftif 

AMioI 
AapwM 

Kilo- 
watt* 

HE.* 

DIN 

6ET 

Aotud 

Kilo- 
watta 

Power 
Factor 

Kmpma 

Preaa- 
ure 

VMf 

Total 

PrM6- 

ure 

Vam. 

Total 

M'd't 

10 

11 

Noon 

10 

11 

Wi*t 

Total 

Total 

Peak 

Total 
Average 

Aver 

nge 



Arei 

*«• 

R  READINGS 


TO  LOCAL  DWTRIBUTINO  6UB  STATION      J 

STATION  USE 

ClKQit 

Cfrmiit 

Circuit 

to  OlfMto  M«i«« 

*-±r' 

Cireuit 

circuit 

Total 

3 

lATA 


JCHARQE                      1 

WHEEL  DISCHARGE 

AIR  TEMPERATURE;  HUMIDITY 

.IL 

A.M. 

P.M. 

1 

Max.  Temp. 

Deg.P. 

tim^m 

Max.    X8.rl 

bX 

Mai. 

XS.F 

Min.      •• 

Min.       •• 

ap.8. 

Min. 

.. 

Av. 

•• 

At. 

Av. 

•• 

Humiditj 

J< 

n 

Average  liead  in  ftMt 
on  all  WatorwheeU 

WeatheR 

B^UNUSUAL  OCCURRANCES 


To  face  page  761. 

Digitized  by  CjOOQ IC 


Digitized  by 


Google 


OPERATING  AND  MAINTENANCE  INSTRUCTIONS      751 

Hiver  in  Cubic  Feet  per  Second,"  calculated  each  day  and  con- 
verted into  *'  Available  Capacity  of  River,"  which  is  shown  in 
kw.-hr.;  the  "Available  Capacity  of  Power  House,"  shown  in 
kw.-hr.,  and  determined  by  calculating  the  capacity  of  the  ma- 
chines under  the  average  head  for  twenty-four  hours;  the  "  Kw.- 
hr.  Lost,"  or  the  diflference  between  what  was  actually  generated 
by  the  machines  and  what  could  have  been  secured  from  the  river 
during  the  same  number  of  hours. 

Any  important  notes  of  operation  are  entered  on  the  back 
of  each  day's  log  sheet.  These  notes,  together  with  certain  rec- 
ords for  log  sheets,  are  also  entered  each  day  in  a  log  book  kept 
on  the  operator's  desk  at  all  times  for  reference  purposes.  Weather 
conditions  and  temperatures  are  recorded  four  times  daily,  at 
midnight,  6  a.m.,  noon,  and  6  p.m.  A  rain  guage  is  provided  on 
the  roof  of  the  station  from  which  records  of  precipitation  covering 
each  twenty-four  hours  are  obtained. 

A  record  form  of  a  large  power  gfsrstem  in  the  West  is  shown 
in  Fig.  408. 

Operating  and  Maintenance  Instructions.  Several  of  the 
larger  hydro-electric  companies  have  developed  very  successful 
systems  of  systematizing  the  operating  details  and  properly 
training  the  operating  force,  thus  obtaining  a  considerable  im- 
provement over  the  methods  ordinarily  in  use.  A  description  of 
the  practice  by  one  of  the  larger  hydro-electric  companies,  as 
given  in  the  1917  Report  of  the  N.E.L.A.  Committee  on  Prime 
Movers,  should  therefore  be  of  interest. 

Operating  Instructions.  The  operation  of  the  plant  is  covered 
by  instructions  which  express  in  writing  not  oply  what  must 
be  done  in  the  case  of  certain  emergencies,  but  also  describe  how 
the  plant  must  be  run  imder  normal  conditions. 

These  "  Permanent  Instructions,"  as  they  are  called,  are 
divided  as  follows: 

General  Station  Rules,  etc. 

Safety  Rules. 

Electrical  Operation — Normal  and  Abnormal. 

Hydraulic  Operation — ^Normal  and  Abnormal. 

Duties  of  Operating  Men. 

Record  and  Forms. 

Electrical  Maintenance. 

Hydraulic  Maintenance. 


Digitized  by 


Google 


752  ORGANIZATION  AND  OPERATION 

The  "  General  Station  Rules  "  govern  the  employees  as  a 
body  and  are  concerned  with  such  things  as  wages,  hours,  leaves 
of  absence,  vacations,  and  miscellaneous  matters  regarding  the 
conduct  of  the  men  in  the  stations. 

Under  "Safety  Rules"  come  the  usual  regulations  provid- 
ing for  the  safety  of  the  men  working  aroimd  electrical  and 
mechanical  equipment.  Safety  Rules  also  include  rules  for  the 
"  Hold-OflF "  system,  by  which  the  men  are  protected  while 
working  on  apparatus. 

Under  the  "  Electrical  Operating  Instructions "  are  two 
divisions — normal  and  abnormal.  The  normal  instructions  deal 
with  every-day  conditions,  and  their  aim  is  to  specify  how  the 
apparatus  shall  be  handled,  what  the  connections  shall  be,  and 
how  the  various  other  routine  operations  of  the  station  shall  be 
performed.  The  abnormal  instructions  are  developed  from  cases 
of  trouble  that  have  been  experienced  in  the  station,  and  such  as 
might  occur.  They  include  general  instructions  on  handling 
trouble,  instructions  on  various  line  complications  and  on  gen- 
erator, transformer,  bus  and  oil  switch  trouble.  They  also  in- 
clude the  handling  of  the  station  during  lightning  storms  and 
low-water  season  operation,  when  particular  attention  must  be 
paid  to  efficiency,  as  well  as  instructions  for  the  flood  season, 
and  ice  and  sleet. 

The  "  Hydraulic  Operating "  instructions  are  similwly 
divided  into  normal  and  abnormal. 

The  section  on  "  Records  and  Forms  "  includes  instructions 
on  the  use  of  the  various  forms,  such  as  log  sheets,  graphic  meter 
records,  and  also  on  record  and  tabulation  work.  The  section 
on  "Duties"  specifies  the  particular  duties  of  each  operating 
attendant.  The  "  Electrical  and  Hydraulic  Maintenance " 
instructions  cover  such  matters  as  the  cleaning,  inspection  and 
repair  of  apparatus. 

These  instructions  have  been  found  very  valuable  in  crystal- 
lizing the  operation  of  the  plant,  making  it  more  automatic  and 
independent  of  the  personal  element.  They  have  also  made  it 
considerably  easier  for  those  in  charge  to  break  in  and  instruct 
new  men;  under  them  all  operators  tend  to  do  given  things  in 
the  same  way,  a  way  which  has  been  determined  by  study  and 
experiment  to  be  the  best  way.  An  attendant  can  be  transferred 
from  one  shift  to  another  without  having  to  learn  new  methods. 


Digitized  by 


Google 


OPERATING  AND  MAINTENANCE  INSTRUCTIONS      753 

He  will  know  that  all  operations,  such  as  the  starting  and  stop- 
ping of  generators,  handling  of  switches,  etc.,  will  be  carried 
on  exactly  the  same  as  on  any  other  shift. 

A  good  example  of  the  result  of  study  and  system  in  oper- 
ating methods  is  the  comparatively  simple  matter  of  starting 
up  a  generator.  Before  the  instructions  were  put  into  effect  the 
time  for  starting  a  unit  would  vary  from  IJ  to  3  or  4  minutes, 
depending  upon  the  individual  operators  and  hydraulic  attend- 
ants. A  study  was  made  of  the  various  operations  and  the  time 
taken  to  start  a  generator,  and  it  was  found  that  by  having  the 
several  attendants  do  their  work  independently,  without  waiting 
for  one  another  and  without  waiting  for  verbal  instructions, 
operations  could  be  performed  simultaneously  which  were  for- 
merly done  successively.  It  had  been  the  practice  for  the  gov- 
ernor man  to  make  an  imspection  of  the  unit  and  for  the  opera- 
tor to  try  out  the  oil  switch,  before  the  disconnectors  were 
closed.  These  unnecessary  precautions  were  eliminated  by  in- 
sisting that  every  unit  and  oil  switch,  in  fact  every  part  of  the 
equipment,  be  ready  for  immediate  service  at  any  moment,  unless 
it  was  covered  by  a  *'  hold-off  "  tag.  The  operation  of  starting 
the  unit  on  the  governor  also  took  time,  and  it  is  now  the  practice 
to  start  the  unit  on  hand  control.  The  best  way  of  manipulating 
the  gates  to  get  the  unit  to  accelerate  more  rapidly  was  observed, 
and  the  governor  attendants  instructed  accordingly. 

It  has  also  been  made  the  practice  to  always  start  the  units 
quickly.  The  normal  time  now  taken  to  start  a  unit  is  about 
sixty  seconds.  The  record  time  on  a  stop-watch  drill  test  was 
forty-one  seconds,  while  in  an  actual  emergency  due  to  the  loss  of 
a  steam  turbine  unit  on  the  system,  and  resulting  in  frequency 
disturbance,  a  unit  was  paralleled  and  frequency  brought  to  nor- 
mal in  thirty-five  seconds  after  the  disturbance  occurred.  In 
another  case  two  units  were  paralleled  and  frequency  brought  to 
normal  IJ  minutes  after  the  trouble. 

An  important  feature  of  this  quick  starting  is  that  it  must 
accelerate  very  quickly  at  first  and  pass  through  the  synchronous 
point  very  slowly.  While  the  unit  is  accelerating  the  operator 
must  send  his  assistant  to  close  the  disconnecting  switches  and 
have  his  synchronizing  and  voltmeter  plugs  in  position  before 
the  unit  comes  up  to  speed,  so  that  as  soon  as  the  speed  passes 
through  the  synchronous  point  he  gets  his  "  shot."    If  he  misses 


Digitized  by 


Google 


754  ORGANIZATION  AND  OPERATION 

the  first  "  shot "  there  will  be  a  delay  of  from  fifteen  to  twenty 
seconds  in  bringing  the  speed  back  again,  hence  it  is  important 
that  the  governor  man  manipulate  the  speed  properly  and  be  ready 
to  take  the  first  shot  when  it  presents  itself. 

Another  point  is  to  have  the  field  rheostat  in  the  proper 
position  for  normal  voltage,  so  that  no  time  is  lost  in  manipu- 
lating the  voltage.  In  cases  of  serious  emergency  where  there 
have  ahready  been  interruptions  to  service  or  serious  fluctuations 
of  voltage,  or  where  the  hydro-electric  plant  has  separated  from 
the  steam  plant,  the  operator  is  instructed  to  parallel  without  the 
use  of  the  synchroscope,  in  order  to  save  time.  In  this  case  he 
opens  the  field  of  the  incoming  generator  while  closing  its  oil 
switch  and  inmiediately  closes  the  field  afterwards.  Under  the 
special  conditions  of  high  reactance  of  the  tmits  employed  in  the 
plant  described,  this  results  in  a  5  to  8  per  cent  fluctuation  in 
voltage  in  case  the  incoming  unit  (of  approximately  10,000-kw. 
capacity)  is  20  per  cent  less  than  the  capacity  already  tied  in  on 
the  bus. 

Maintenance.  The  first  task  was  to  get  up  a  machinery  index 
wherein  is  listed  the  station  apparatus.  A  letter  size  sheet,  or 
several  of  them,  are  devoted  to  each  piece  of  apparatus  and  upon 
these  sheets  are  noted  data  or  reference  directions  in  regard  to  the 
apparatus,  also  references  to  a  machinery  repair  log  book,  where 
may  be  obtained  detailed  information  with  regard  to  the  repair 
history  of  the  piece  of  apparatus. 

In  regard  to  the  maintenance  of  the  station,  the  operating 
attendants  do  a  large  amount  of  this  work  and  practically  all  of 
.the  inspection.  Instructions  for  cleaning  and  inspection  have 
been  very  carefully  drawn  up  and  the  operating  men  instructed 
in  the  proper  care  of  the  apparatus.  Every  piece  of  apparatus 
in  the  station  has  been  considered  individually  and  it  has  been 
determined  just  how  often  it  needs  to  be  inspected  and  how 
thorough  an  inspection  is  needed.  All  the  equipment  is  tabulated 
on  charts,  which  show  the  periodicity  of  the  inspections  and 
provide  spaces  which  are  to  be  filled  in  with  the  date  and  initials 
of  the  attendant  who  made  the  inspection.  These  charts  are 
posted  on  the  wall  in  a  conspicuous  place  and  make  an  excellent 
graphical  record  of  the  status  of  the  inspections  of  the  entire 
station  up  to  date.  Any  delayed  inspections  are,  naturally, 
inquired  into. 


Digitized  by  VjOOQ IC 


OPERATING  AND  MAINTENANCE  INSTRUCTIONS      755 

In  addition  to  the  current  inspections  by  the  operating  men 
chere  is  also  a  more  thorough  inspection  made  as  often  as  may 
be  necessary,  but  at  less  frequent  intervals,  by  the  electrical 
inspector  and  hydraulic  inspector.  The  inspection  work  for  these 
men  also  is  laid  out  on  schedule  drawn  up  in  the  form  of  charts 
and  the  date  of  inspection  similarly  noted.  This  gfystem  of  keep- 
ing track  of  inspections  has  been  the  result  of  much  experi- 
menting and  investigation  of  the  methods  of  other  companies. 
The  card  index  system,  which  is  ordinarily  used,  does  not  have 
the  advantage  of  immediate  accessibiUty  and  becomes  very  bulky 
when  each  individual  piece  of  apparatus  in  the  station  is  included. 
An  ordinary  manifold  note  book  is  used  for  trouble  reports;  the 
original  goes  to  the  office  to  note  that  the  inspection  was  made 
and  later  is  placed  on  file.  If  the  apparatus  is  found  to  be  out  of 
order  a  "  Trouble  Report "  is  made  out  on  a  regular  form,  space 
being  provided  for  the  report  of  the  man  who  is  to  remedy  the 
trouble,  and  also  for  further  report  or  remarks  from  the  Elec- 
trical or  Hydraulic  Inspector.  In  these  remarks  the  inspector  is 
supposed  to  give  assurance  that  the  trouble  will  not  occur  again, 
or  state  what  is  necessary  to  be  done  to  prevent  its  reoccurrence. 
These  reports  are  filed  and  later  become  valuable  in  eliminating 
troublesome  features  of  design,  when  new  apparatus  is  to  be 
designed  or  purchased. 

Assignment  of  Apparatus.  Another  thing  which  facilitates  the 
inspection  and  cleanliness  of  the  apparatus  is  the  assignment  of 
every  particular  piece  of  apparatus  in  the  station  to  some  particu- 
lar person.  Each  attendant  has  his  own  particular  apparatus 
for  which  he  is  responsible,  which  he  must  keep  clean  and  in  proper 
operating  condition.  When  defects  occur  in  this  equipment  he 
will  either  remedy  them  himself  or  report  them  on  a  "  Trouble 
Report."  If  the  apparatus  is  in  bad  condition  it  is  this  man  whose 
attention  is  called  to  it,  and  if  it  is  kept  in  exceptionally  good 
condition  it  is  he  who  receives  the  credit.  An  attendant  who  is 
inclined  to  be  delinquent  in  attention  to  his  apparatus  soon  finds 
that  his  equipment  compares  unfavorably  with  the  adjacent 
equipment  and  will  naturally  remedy  it  without  its  having  to  be 
brought  to  his  attention  by  his  superior. 

Exposed  tool  boards  are  mounted  at  different  points  in  the 
station  so  that  attendants  have  available  all  they  need  in  the  way 
of  tools  for  making  such  repairs  as  they  are  able  to  take  care  of 


Digitized  by 


Google 


756  ORGANIZATION  AND  OPERATION 

without  the  assistance  of  the  regular  maintenance  department. 
By  being  permitted  to  repair  their  own  apparatus  the  operating 
attendants  become  more  familiar  with  its  details  and  learn  better 
how  to  operate  it  and  take  operating  care  of  it,  and  are  given  an 
interesting  occupation,  in  addition  to  saving  the  time  of  the 
maintenance  men  in  attending  to  minor  repairs. 

The  aim  is  to  substitute  preventive  maintenance  for  breaking 
down  repairs.  The  result  of  this  inspection  and  maintenance 
system  has  been  that  the  apparatus  is  kept  in  better  condition, 
and  this  has  been  accomplished  with  the  minimum  of  attention 
on  the  part  of  superiors,  as  the  system  is  more  or  less  automatic 
in  its  workings.  At  the  same  time,  the  reports  and  schedules 
give  the  superior  very  definite  knowledge  of  the  condition  of  his 
equipment.  All  this  work  being  laid  out  before  the  man  in  the 
form  of  instructions  reUeves  the  superior  of  continually  correcting 
new  men  and  instructing  them  in  how  things  are  supposed  to 
be  done.  It  also  eliminates  dependence  on  word-of-mouth 
transmission  of  instructions  from  one  man  to  another. 

This  systemization  tends  to  minimize  the  possibility  of 
neglect  of  maintenance  work  on  apparatus,  and  by  scheduling 
the  work  as  to  time  and  making  necessary  the  planning  of  the 
work  to  get  it  through  in  that  period,  there  is  less  time  lost  in 
doing  the  maintenance  jobs  or  between  jobs,  and  the  maintenance 
or  operating  shifts  are  thus  able  to  turn  out  more  work  and  better 
work. 


Digitized  by 


Google 


APPENDIX  I 


REFERENCES    TO    DESCRIPTIONS    OF    AMERICAN 
HYDRO-ELECTRIC   POWER   SYSTEMS 


Abbreviation  op  Titlfs  of  Periodicals 

American  Institute  of  Electrical  Engineers A.I.E.E. 

American  Institute  of  Mining  Engineers A.I.M.E. 

American  Society  of  Civil  Engineers A.S.C.E. 

American  Society  of  Mechanical  Engineers A.S.M.E. 

Canadian  Electrical  News Can.  Elec.  NewB 

Canadian  Engineer Can.  Engr. 

Electrical  Age Elec.  Age 

Electrical  Engineering  (formerly  Southern  Electrician) Elec.  Engng. 

Electrical  Record Elec.  Rec. 

Electrical  Review  &  Western  Electrician Elec.  Rev. 

Electrical  World Elec.  Wld. 

Electric  Journal Elec.  Jour. 

Engineering  News Eng.  News 

Engineering  Record Eng.  Rec. 

General  Electric  Review Gen.  Elec.  Rev. 

Journal  of  Electricity,  Power  &  Gas Jour,  of  Elec. 

Power Power 

Southern  Electrician So.  Elctn. 

Western  Engineering West.  Engng. 

Alabama  Power  Company: 

Elec.  Wld September  13,  1913 

Elec.  Engng January,  1914 

Eng.  Rec April  4,  1914 

Elec.  Wld May  30,  1914 

Power August  4,  1914 

A.S.C.E September,  1914 

Elec.  Engng March,  1915 

Elec.  Engng June,  1915 

Gen.  Elec.  Rev June,  1916 

Albany  Power  &  Manufacturing  Company,  Georgia: 

Elec.  Wld June  16,  1906 

American  River  Electric  Company,  California: 

Jour,  of  Elec February,  1904 

757 


Digitized  by 


Google 


758  APPENDIX  I 

Anglo-Newfoundknd  Develoiuneiit  Company: 

Eng.  Rec July  22, 1911 

Animas  Power  &  Water  Company,  Colorado: 

Eng.  News January  4,  1906 

Eng.  Rec April  14,  1906 

Elec.  Rev April  21,  1906 

Anthony  Falls  Water  Power  Company,  Minnesota: 

Eng.  Rec May  29,  1909 

Appalachian  Power  Company,  West  T^rginia: 

So.  Electn November,  1912 

Elec.  Wld November  30,  1912 

Power March  25,  1913 

Apple  River  Power  Company,  Wisconsin: 

Elec.  Wld December  8,  1800 

Eng.  News October  12,  1905 

Arkansas  Valley  Railway,  Light  &  Power  Company,  Colorado: 
Jour,  of  Elec June  5,  1915 

Arizona  Power  Company: 

Elec.  Wld .August  18,  1910 

Elec.  Wld. .August  25,  1910 

Eng.  Rec August  20,  1910 

Jour,  of  Elec June  5,  1915 

Athens  Railway  &  Electric  Company,  Georgia: 

Power March  26,  1912 

Atlanta  Water  &  Electric  Power  Company,  Georgia: 

Eng.  Rec April  23,  1904 

Elec.  Wld December  31,  1904 

Auglaize  Power  Company,  Ohio: 

Power February  20,  1912 

Elec.  Wld November  1,  1913 

Eng.  Rec March  7,  1914 

Attgusta-Aiken  Railway  &  Electric  Company,  South  Carolina: 
Elec.  Engng April,  1914 

An  Sable  Electric  Company,  Michigan: 

Elec.  Wld April  13,  1912 

Elec.  Wd April  20,  1912 

Eng.  News May  16,  1912 

Austin  Power  Development,  Texas: 

Elec.  Rev May  22,  1915 

Elec.  Wld June  5,  1915 

Eng.  News June  10,  1915 

Bangor  Power  Company,  Maine: 

Elec.  Rec May,  1914 

Bar  Harbor  &  Union  River  Power  Company,  Maine: 

Elec.  Rec May,  1914 


Digitized  by 


Google 


APPENDIX  I  759 

Bear  Lake  Power  Company,  Idaho: 

Jour,  of  Elec May  17, 1913 

Belton  Power  Company,  South  Carolina: 

Elec.  Wld December  15,  1906 

Eng.  Rec December  15,  1906 

Bend  Water,  Light  &  Power  Company,  Oregon: 

Jour,  of  Elec October  11,  1913 

Black  Hills  Traction  Company,  South  Dakota: 

Eng.  Rec Nov.  16,  1907 

Black  River  Falls  Municipal  Development,  Wisconsui: 

Elec.  Wld June  1,  1911 

Blue  Earth  Hydro-Electric  Development,  Minnesota: 

Eng.  Rec August  26,  1911 

Braden  Copper  Company,  Chile: 

Eng.  Rec September  28,  1912 

Eng.  News May  22,  1913 

British  Columbia  Electric  Railway  Company: 

Jour,  of  Elec June  6,  1915 

British  Railway,  Light  &  Power  Company,  Oregon: 

Eng.  Rec .' March  2,  1912 

Elec.  Wld July  13,  1912 

BuU  Run  Hydro-Electric  Development,  Oregon: 

Eng.  Rec January  18,  1913 

Burlington  Light  &  Power  Company,  Vermont: 

Elec.  Wld Sept.  12, 1914 

Calgary  Power  Company,  Canada: 

Elec.  Wld December  23,  1911 

Eng.  Rec February  7,  1914 

Elec.  Wld April  11,  1914 

Jour,  of  Elec December  18,  1916 

Eng.  Rec January  15,  1916 

Eng.  Rec January  22,  1916 

Power March  14,  1916 

California  Gas  &  Electric  Company: 

See  Pacific  Gas  &  Electric  Company. 

California-Oregon  Power  Company: 

Jour,  of  Elec February  22,  1913 

Eng.  Rec June  7,  1913 

Jour,  of  Elec June  5,  1915 

Canadian  Electric  Light  Company: 

Elec.  Wld June  15,  1901 

Can.  Elec.  News June,  1902 

Eng.  News May  7,  1903 


Digitized  by 


Google 


760  APPENDIX  I 

Cimadkn  Light  &  Power  ComiMmy: 

Can.  Mec.  News September,  1911 

Eng.  Rec April  6,  1912 

Elec.  Wld August  3,  1912 

Caaadian-Niegiuw  Power  Company: 

Can.  Engr November,  19Q2 

Elec.  Rev January  3,  1903 

Elec.  Wld January  7,  1905 

Elec.  Rev December  2,  1905 

Can.  Elec.  News September,  1907 

A.S.C.E August,  1908 

Elec.  Jour June,  1914 

Carolina  Power  &  Light  Company: 

Elec.  Wld May  30,  1914 

Carp  River  Hydro-Electric  Development,  Michigan: 

Eng.  Rec November  23, 1912 

Cedars  Rapids  Mfg.  &  Power  Company,  Canada: 

Can.  Elec.  News February  1,  1913 

Eng.  Rec October  25,  1913 

Can.  Elec.  News June  15, 1914 

Eng.  Rec July  18,  1914 

Eng.  Rec July  25, 1914 

Elec.  Wld February  13, 1916 

Eng.  News March  25,  1915 

Eng.  News April  1,  1915 

Can.  Elec.  News March  1, 1916 

Can.  Elec.  News March  15, 1916 

Gen.  Elec.  Rev June,  1916 

Can.  Engr Feb.  15,  1917 

Central  Colorado  Power  Company: 

Eng.  Rec June  25,  1910 

Elec.  Wld June  23,  1910 

Elec.  Wld June  30,  1910 

Elec.  Wld July  14,  1910 

Eng.  Rec July  30,  1910 

A.I.E.E June,  1911 

Elec.  Wld October  7, 1911 

Elec.  Wld June  1,  1912 

Jour,  of  Elec June  5,  1915 

Central  Georgia  Power  Company: 

Eng.  Rec AprU  17,  1909 

Eng.  Rec May  14,  1910 

Elec.  Wld April  27,  1911 

So.  Electn May,  1911 

Elec.  Rev July  22,  1911 

Elec.  Wld. September  16, 1911 


Digitized  by 


Google 


APPENDIX  I  761 

Hec.  Wld Januaiy  26, 1913 

Elec.  Wld May  30, 1914 

Central  Maine  Power  Company: 

Elec.  Wld December  15, 1019 

Elec.  Rec June,  1914 

Elec.  Wld June  3, 1916 

Cerro  de  Pasco  Mining  Company,  Peru: 

Eng.  &  Mining  World April  3,  1915 

Chasm  Power  Company,  New  Tork: 

Elec.  Wld November  21, 1903 

Chattanooga  &  Tennessee  River  Power  Company: 

Eng.  Rec February  15,  1913 

Elec.  Engng August,  1913 

Elec.  Wld November  15,  1913 

Elec.  Rev November  22,  1913 

Power December  2,  1913 

Elec.  Wld May  30,  1914 

Gen.  Elec.  Rev August,  1916 

Eng.  Rec September  23, 1916 

Chile  Exploration  Company: 

Elec.  Wld January  2,  1915 

Elec.  Wld January  9,  1915 

Chippewa  &  Flambeau  Improvement  Company,  Wisconsin: 

Eng.  Rec April  5,  1913 

Chippewa  Railway,  Light  &  Power  Company,  ^sconsin: 

Elec.  Wld December  22,  1910 

Chittenden  Power  Company,  Vermont: 

Elec.  Wld December  2,  1905 

Eng.  Rec December  9,  1905 

Cia  Docas  de  Santos,  Brazil: 

Elec.  Wld March  16,  1912 

Gen.  Elec.  Rev October,  1912 

Cleveland  Cliffs  Iron  Company,  Michigan: 

Eng.  &  Mining  Journal February,  1912 

Elec.  Engng May,  1913 

Coast  Counties  Gas  &  Electric  Company,  California: 

Jour,  of  Elec June  5,  1915 

Coast  Valley  Gas  &  Electric  Company,  California: 

Jour,  of  Elec June  5,  1915 

Cobalt  Power  Company,  Canada: 

Can.  Elec.  News September,  1910 

Eng.  Rec March  4,  1910 

Cohoes  Company,  New  Tork: 

Eng.  Rec March  20,  1915 

Eng.  Rec March  27,  1915 

Elec,  Wld March  20,  1916 


Digitized  by 


Google 


782  APPENDIX  I 

Gen.  See.  Rev! .May,  1916 

Eng.  Rec May  29,  1915 

Power Oct.  6,  1915 

Colorado  River  Power  Company,  Texas: 

So.  Electn June,  1911 

Elec.  Wld June  8,  1911 

CoUierBtflle  Hydro-Electric  Development,  New  Tork: 

Eng.  Rec March  7,  1908 

Colorado  Springs  Light,  Heat  &  Power  Company: 

Elec.  Wld October  14, 1911 

Coltsmtms  Mills  Company,  South  Carolina: 

Eng.  Rec August  29, 1903 

Cohsmtms  Power  Company,  Georgia: 

Eng.  Rec January  16,  1904 

Elec.  Engng June,  1913 

Elec.  Wld May  30,  1914 

Concord  Electric  Company,  New  Hampshire: 

Power May  25,  1909 

Connecticut  Power  Company: 

Elec.  Wld March  3, 1917 

Connecticut  River  Power  Company: 

Eng.  Rec March  27,  1909 

Eng.  Rec. April  3,  1909 

Elec.  Wld September  9,  1909 

Power September  28, 1909 

Elec.  Wld May  25,  1911 

Power July  25,  1911 

Gen.  Elec.  Rev October,  1911 

Gen.  Elec.  Rev May,  1916 

Connecticut  River  Transmission  Company: 

Elec.  Wld October  4,  1913 

Elec.  Wld October  11,  1913 

Gen.  Elec.  Rev June,  1914 

Consumers  Power  Company,  Minnesota: 

Elec.  Wld April  13,  1911 

Elec.  Wld June  22,  1911 

Eng.  Rec August  26,  1911 

Coquitlam-Buntzen  Hydro-Electric  Power  Development,  Canada: 

Eng.  Rec September  21, 1912 

Elec.  Wld July  24,  1915 

Consolidated  Lighting  Company  of  Montpelier,  Vermont: 

Elec.  Wld January  4,  1908 

Elec.  Rev February  1,  1908 

Cumbeiland  County  Power  &  Light  Company,  Maine: 

Elec.  Rec November,  1914 

Elec.  Wld February  27,  1915 


Digitized  by 


Google 


APPENDIX  I  763 

Dakota  Power  Company: 

Elec.  Wld August  5,  1911 

Dan  River  Power  &  Mfg.  Company,  ^Hrginia: 

Eng.  Rec September  3,  1904 

Desert  Power  &  Water  Company,  Arizona: 

Jour,  of  Elec June  5,  1915 

Diana  Paper  Company,  New  Tork: 

Eng.  Rec January  6,  1912 

Dominion  Power  &  Transmission  Company,  Canada: 

Elec.  Wld March  9,  1911 

Eagle  River  Electric  Power  Company,  Oregon: 

Jour,  of  Elec June  5,  1915 

East  Creek  Electric  Light  &  Power  Company,  New  Tork: 

Eng.  Rec May  27,  1911 

Elec.  Wld August  31,  1912 

Gen.  Elec.  Rev September,  1912 

Eastern  Oregon  Light  &  Power  Company: 

Jour,  of  Elec June  5,  1915 

Edison  Sault  Electric  Company,  Michigan: 

Eng.  Rec November  2,  1907 

Electric  Development  Company  of  Ontario: 

Eng.  News November  9,  1905 

Elec.  Rev July  28,  1906 

City  of  EUensbnrg,  Washington: 

Jour,  of  Elec January  2,  1916 

Empire  District  Electric  Company,  Kansas: 

So.  Electn June,  1911 

Empire  Gas  &  Electric  Co.,  Waterloo,  N.  T.: 

Power March  28,  1916 

Empire  State  Power  Company,  New  Tork: 

Eng.  Rec August  10,  1901 

Erindale  Power  Company,  Canada: 

Elec.  Wld May  4,  1911 

Eugene  Municipal  Power  Development,  Oregon: 

Elec.  Wld May  17,  1913 

Fall  Mountain  Electric  Company,  Vermont: 

Elec.  Rec October,  1914 

Garvins  Falls  Power  Development,  New  Hampshire: 

Elec.  Wld January  17,  1903 

Eng.  Rec January  24,  1903 

Eng.  Rec May  28,  1904 

Elec.  Wld May  28,  1904 

Elec.  Wld June  4,  1904 


Digitized  by 


Google 


764  APPENDIX  I 

Geofgk-CaroUiia  Power  Company: 

Power January  20,  1914 

Elec.  Engng April,  1914 

Elec.  Engng August,  1915 

Elec.  Wld November,  20,  1915 

Elec.  Wld November  27,  1915 

Geoigia  Railway  &  Power  Company: 

So.  Elctn September,  1912 

Elec.  Rec April  12,  1913 

Elec.  Wld December  20.  1913 

Elec.  Wld December  27,  1913 

Power January  27,  1914 

Eng.  NewB January  29,  1914 

Eng.  Rec March  21,  1914 

Eng.  Rec March  28,  1914 

Eng.  News April  16,  1914 

Elec.  Engng May,  1914 

Elec.  Engng June,  1914 

Elec.  Engng July,  1914 

Elec.  Wld May  30,  1914 

Gen.  Elec.  Rev June,  1914 

Gen.  Elec.  Rev July,  1914 

A.T.E.E October,  1916 

Elec.  Wld Jan.  8,  1916 

Grand  Falls  Power  Company,  New  Brunswick: 

Elec.  Wld March  4,  1909 

Grand  Rapids-Mtiskegon  Power  Company,  Michigan: 

Elec.  Wld November  3,  1906 

Eng.  Rec October  19,  1907 

Elec.  Wld February  4,  1909 

Gnmgeville  Electric  lA0it  &  Power  Company,  Idaho: 

Jour,  of  Elec June  5, 1915 

Great  Falls  Power  Company,  Montana: 

Eng.  Rec Mareh  12,  1910 

Gen.  Elec.  Rev April,  1910 

Gen.  Elec.  Rev May,  1910 

Elec.  Wld July  6,  1912 

Great  Falls  Power  Development,  New  Jersey: 

Eng.  Rec February  22,  1913 

Great  Northern  Paper  Company,  Maine: 

Power February  9,  1909 

Great  Northern  Power  Company,  Minnesota: 

Elec.  Wld July  28,  1906 

Eng.  Rec September  7,  1907 

Eng.  Rec September  14,  1907 

Eng.  News December  26, 1907 


Digitized  by 


Google 


APPENDIX  I  765 

Power August  11, 1908 

Elec.  Wld March  25,  1916 

Great  Northern  Railway  Company,  Washington: 

Eng.  Rec • October  30,  1909 

Gen.  Elec.  Rev August,  1910 

Jour,  of  Elec January  2,  1915 

Great  Western  Power  Company,  California: 

Elec.  Wld August  26,  1909 

Elec.  Wld September  16, 1909 

Jour,  of  Elec April  9,  1910 

Eng.  Rec July  16,  1910 

So.  Mectn February,  1913 

Elec.  Wld May  29,  1915 

Jour,  of  Elec June  5,  1915 

Greenville-Carolina  Power  Company: 

Eng.  Rec October  6,  1906 

Guanajuato  Power  &  Electric  Company,  Mexico: 

Elec.  Wld August  6,  1904 

Elec.  Wld August  13,  1904 

Elec.  Wld August  20,  1914 

Hamilton  Cataract  Power,  Light  &  Traction  Company,  Canada: 

Elec.  Rev September  2, 1905 

Hanford  Irrigation  &  Power  Company,  Washington: 

Elec.  Rev .July  29,  1911 

Jour,  of  Elec January  2,  1915 

Hannawa  Water  Power  Company,  New  Tork: 

Elec.  Wld April  21,  1906 

Hartford  Electric  Light  Company,  Connecticttt: 

Elec.  Wld March  8, 1902 

Holton  Power  Company,  California: 

Jour,  of  Elec June  5,  1915 

Holyoke  Water  Power  Company,  Massachusetts: 

Elec.  Wld September  16,  1906 

Eng.  Rec September  15,  1906 

Homestake  Mining  Company,  South  Dakota: 

Jour,  of  Elec August  29,  1914 

Hortonia  Power  Company,  Vermont: 

Elec.  Rev August  12,  1916 

Hudson  River  Water  Power  Company,  New  Tork: 

Eng.  Rec March  8,  1902 

A.LM.E February,  1903 

Eng.  News June  18,  1903 

Eng.  Rec June  27,  1903 

Elec.  Wld June  27,  1903 

Elec.  Wld October  24,  1903 


Digitized  by 


Google 


766  APPENDIX  I 

Elec.  Wld May  14,  1904 

Elec.  Wld June  11,  1904 

Eng.  Rec March  4,  1905 

Haronum  Company,  Canada: 

^g.  News July  18,  1907 

Eng.  Rec August  3,  1907 

Hydro-Electric  Commission  of  Ontario,  Canada: 

Can.  Elec.  News November,  1910 

Elec.  Rev December  31,  1910 

Elec.  Wld January  6,  1912 

Elec.  Wld January  13,  1912 

Elec.  Wld January  20,  1912 

Elec.  Wld January  27,  1912 

Can.  Engr July  25,  1912 

Can.  Elec.  News April  15,  1914 

Eng.  Rec June  7,  1917 

Conmiission  Reports 

Idaho  Consolidated  Power  Company: 

Elec.  Rev June  2,  1906 

Idaho  Falls  Mwiicipal  Development: 

Elec.  Wld Sept.  21, 1902 

Idaho  Power  &  Light  Company: 

Jour,  of  Elec * June  5,  1915 

Elec.  Wld Aug.  19, 1916 

Indiana  &  Michigan  Electric  Company: 

Elec.  Rev March  4,  1911 

Inland  Portland  Cement  Company,  Washington: 

Jour,  of  Elec January  2,  1915 

Ironwood  &  Bessemer  Railway  &  Light  Company,  Michigan: 

Eng.  Rec February  28,  1914 

Isthmian  Canal  Commission,  Gatun,  Panama: 

Gen.  Elec.  Rev July,  1914 

Jim  Creek  Water,  Light  &  Power  Company,  Washmgton: 

Joiir.  of  Elec January  2,  1915 

Juniata  Water  &  Water  Power  Company,  Pennsylvania: 

Elec.  Wld December  22,  1906 

Elec.  Wld January  20,  1910 

Kakabeka  Falls  Development,  Canada: 

Elec.  Wld January  26,  1907 

Can.  Elec.  News September,  1907 

Kaministquia  Power  Company,  Canada: 

Elec.  Wld January  26,  1907 

Can.  Elec.  News February  15,  1916 


Digitized  by 


Google 


APPENDIX  I  767 

Kenora  Hydraulic  Power  Developmenty  Canada: 

Eng.  Rec July  18,  1908 

Lachine  Rapids,  Hydraislic  &  Land  Company,  Canada: 

Elec.  Rev November  21,  1903 

La  Crosse  Water  Power  Company,  Wisconsin: 

Eng.  Rec May  30,  1908 

Elec.  Wld March  31,  1910 

Elec.  Wld October  14,  1911 

Lanrentian  Power  Company,  Canada: 

Can.  Engr October  22,  1914 

Can.  Elec.  News August  1,  1916 

Can.  Elec.  News June  15,  1917 

Lewiston  &  Aobum  Company,  Maine: 

Elec.  Wld September  20,  1902 

Elec.  Wld April  8,  1906 

Lewiston-Clarkson  Improvement  Company,  Washington: 

Elec.  Wld March  28,  1914 

Jour,  of  Elec June  5, 1915 

Los  Angeles  Aqueduct: 

Eng.  News March  24,  1910 

Power September  26,  1911 

Elec.  Wld February  10,  1912 

Eng.  Rec February  3,  1912 

Eng.  Rec August  23,  1913 

Eng.  Rec November  1,  1913 

Jour,  of  Elec June  24,  1916 

Madison  River  Power  Company,  Montana: 

Elec.  Wld December  30,  1909 

Maumee  Valley  Electric  Company,  New  York: 

Elec.  Wld March  2,  1911 

Medina  Irrigation  Company,  Texas: 

Eng.  Rec October  18,  1913 

Menominee  &  Marinette  Light  &  Traction  Company: 

Eng.  Rec January  14,  1911 

Elec.  Wld January  17,  1914 

Metropolitan  Water  Works,  Massachusetts: 

Power November  26,  1912 

Mexican  Light  &  Power  Company: 

Eng.  Rec June  9,  1906 

A.S.C.E January,  1907 

Mexico  Northern  Power  Company: 

Elec.  Wld July  25,  1914 

Michigan  Power  Company: 

Elec.  Rev December  25,  1909 


Digitized  by 


Google 


768  APPENDIX  1 

MIchwiein  Power  Company,  Mexico: 

Eag.  Ree August  27, 1910 

Eng.  Rec Januaiy  21, 1911 

MJimeApolis  General  Electric  Company,  Minnesota: 

Eng.  Rec March  3,  1906 

Eng.  Rec June  29, 1907 

Eng.  Rec July  6,  1907 

Elec.  Wld. July  6,  1907 

Elec.  Wld September  7, 1907 

Mississippi  River  Power  Company,  Iowa: 

Eng.  Rec August  5,  1911 

Elec.  Wld August  6, 1911 

Elec.  Wld October  28, 1911 

Elec.  Wld May  31,  1912 

Elec.  Wld Sept.  7,  1912 

Eng.  Rec November  16,  1912 

So.  Electn JanXiaiy,  1913 

Elec.  Wld April  26,  1913 

Elec.  Wld • May  31,  1913 

Eng.  Rec July  26,  1913 

Power August  5,  1913 

Elec.  Rev Sept.  15,  1913 

Eng.  News November  13,  1913 

Gen.  Elec.  Rev February,  1914 

Gen.  Elec.  Rev April,  1914 

Elec.  Engng August,  1914 

Eng.  Rec September  18, 1915 

Eng.  Rec October  9, 1915 

Missouri  River  Power  Company,  Montana: 

A.S.C.E 1903 

Mokawk  Hydro-Electric  Company,  New  Toric: 

Elec.  Wld October  7,  1911 

Eng.  Rec November  26, 1911 

Montana  Power  Company: 

Jour,  of  Elec June  5, 1915 

Elec.  Wld June  12,  1915 

Gen.  Elec.  Rev November,  1916 

Elec.  Rev July  14,  1917 

Mount  Hood  Railway  &  Power  Company,  Oregon: 

Elec.  Wld March  22,  1913 

Elec.  Wld March  29,  1913 

Mt  Whitney  Power  &  Electric  Company,  California: 

Jour,  of  Elec December  27,  1913 

Jour,  of  Elec June  5,  1915 

Elec.  Wld June  24,  1916 


Digitized  by 


Google 


APPENDIX  I  769 

» 

Keyada-CaliforoiA  Power  Company: 

Eng.  Rec March  6,  1909 

Elec.  Wld October  17,  1914 

Elec.  Wld October  24,  1914 

Hec.  Wld October  31,  1914 

Elec.  Wld , November  7,  1914 

Elec.  Wld November  14,  1914 

Elec.  Wld November  21,  1914 

Elec.  Wld November  28,  1914 

Elec.  Wld December  5, 1914 

Elec.  Wld December  12,  1914 

Elec.  Wld December  19,  1914 

Jour,  of  Elec June  5,  1915 

Nevada  Power  &  Milling  Company,  California: 

Elec.  Wld June  30,  1906 

Eng.  Rec June  30,  1906 

NevadaValleys  Power  Company,  California: 

Jourrof  Elec June  6, 1916 

New  England  Fish  Company,  Alaska: 

Elec.  Wld January  13,  1910 

New  England  Power  Company,  Masaachnsetts: 

Elec.  Wld December  16,  1911 

Elec.  Wld December  28,  1912 

Power February  25,  1913 

New  River  Light  &  Power  Co.,  North  Carolina: 

Elec.  Wld .June  24, 1916 

Now  Milford  Power  Company,  Connectioat: 

Elec.  Wld February  13,  1904 

Niagara  Falls  Hydro-Electric  Power  &  Mfg.  Company: 

Elec.  Wld :;...# November  25,  1905 

Niagara  Falls  Power  Company: 

Eng.  Rec February  16,  1901 

A.I.E.E June,  1902 

Eng.  News October  2,  1902 

Elec.  Rev Sept.  12,  1903 

Eng.  Rec November  21,  1903 

Eng.  Rec October  18, 1913 

Nipissing  Power  Company,  Canada: 

Elec.  Wld "•"• August  12,  1911 

North  Carolina  Electric  Power  Company: 

Elec.  Wld February  17,  1912 

So.  Electn April,  1912 

North  Coast  Power  Company: 

Jour,  of  Elec June  16,  1917 


Digitized  by 


Google 


770  APPENDIX  I 

Northern  California  Power  Company: 

Elec.  Wld September  10,  1904 

Elec.  Wld September  17,  1904 

Elec.  Wld ; September  24,  1904 

Elec.  Wld October  1,  1904 

Jour,  of  Elec August  6,  1910 

Jour,  of  Elec November  4,  1911 

Elec.  Wld February  3,  1912 

Elec.  Wld.. . : May  29,  1915 

Jour,  of  Elec June  5,  1915 

Northern  Colorado  Power  Company: 

Elec.  Wld September  30, 1911 

Northern  Hydro-Electric  Power  Company,  Wisconsin: 

Elec.  Wld November  24,  1910 

Northern  Idaho  &  Montana  Power  Company: 

Elec.  Rev March  18,  1911 

Jour,  of  Elec June  5,  1915 

Jour,  of  Elec June  5,  1915 

Northern  Illinois  Light  &  Traction  Company: 

Elec.  Wld September  29,  1910 

Eng.  Rec February  24,  1912 

Elec.  Wld February  24,  1912 

North  Mountain  Power  Company,  California: 

Jour,  of  Elec February,  1905 

Northern  Ohio  Traction  &  Light  Company: 

Elec.  Wld August  22,  1914 

Northern  Ontario  Light  &  Power  Company,  Canada: 

Can.  Elec.  News 1914 

North  Washington  Power  &  Reduction  Company,  Washington: 

Jour,  of  Elec January  2,  1915 

Northwestern  Electric  Company,  Oregon: 

Elec.  Wld August  9,  1913 

Eng.  Rec October  11,  1913 

West.  Engng November,  1913 

Jour,  of  Elec December  6,  1913 

Jour,  of  Elec January  2,  1915 

Jour,  of  Elec June  5,  1915 

Olympia  Light  &  Power  Company,  Washington: 

Jour,  of  Elec January  2,  1915 

Olympic  Power  Company,  Washington: 

Jour,  of  Elec January  2,  1915 

Jour,  of  Elec June  5,  1915 

Jour  of  Elec October  9,  1915 

Jour,  of  Elec October  16,  1915 


Digitized  by 


Google 


APPENDIX  I  771 

Ontario  Power  Company,  Canada: 

A.I.E.E June,  1906 

Elec.  Wld August  26,  1906 

Elec.  Wld Sept.  2,  1906 

Elec.  Wld Sept.  9, 1906 

Elec.  News November  9,  1905 

Can.  Elec  News November,  1910 

Elec.  Rev December  31,  1910 

Can.  Elec.  News April  15,  1914 

Oregon  Power  Company: 

Jour  of  Elec June  5,  1915 

Oro  Electric  Corporation,  California: 

Elec.  Wld May  29,  1915 

Jour,  of  Elec June  5,  1915 

Ozark  Power  &  Water  Company,  Missouri: 

Eng.  Rec August  2,  1913 

Elec.  Engng November,  1913 

Gen.  Elec.  Rev September,  1914 

Pacific  Coast  Power  Company,  Washington: 

Eng.  News April  11,  1912 

Eng.  News April  18,  1912 

Eng.  Rec April  13,  1912 

Jour,  of  Elec April  13,  1912 

Pacific  Gas  &  Electric  Company,  California: 

Eng.  News August  10,  1905 

Jour,  of  Elec , .  .April  30,  1910 

Power May  21,  1912 

Elec.  Wld June  1,  1912 

Jour,  of  Elec ' March  15,  1913 

Elec.  Wld November  22,  1913 

Jour,  of  Elec December  13,  1913 

Eng.  News December  11,  1913 

Elec.  Wld May  29,  1915 

Jour,  of  Elec June  6,  1915 

Elec.  Jour June,  1915 

Jour,  of  Elec September  30,  1916 

Jour,  of  Elec March  15,  1917 

Jour,  of  Elec April  15,  1917 

Jour,  of  Elec May  1,  1916 

Elec.  Wld June  2,  1917 

Jour,  of  Elec June  1,  2917 

Jour,  of  Elec June  15,  1917 

Pacific  Light  &  Power  Company,  California: 

Jour,  of  Elec November  12,  1910 

Elec.  Wld December  30,  1911 

Jour,  of  Elec February  24,  1912 


Digitized  by 


Google 


772  APPENDIX  I 

Elec.  Wld September  7, 1912 

Eng.  Rec iSeptember  14,  1912 

Eng.  News November  7,  1912 

Elec.  Wld January  3,  1914 

Elec.  Wld January  10,  1914 

Eng.  Rec January  10,  1914 

Gen.  Elec.  Rev August,  1914 

Elec.  Wld December  9,  1916 

Pacific  Power  &  Light  Company,  Oregon: 

Elec.  Wld September  14, 1912 

Jour,  of  Elec September  14,  1912 

Jour,  of  Elec January  2,  1915 

Jour  of  Elec June  6, 1915 

Palmer  Mills  Development,  Massachiisetts: 

Eng.  Rec June  24, 1911 

Palmer  Mountain  Tnnnel  &  Power  Company,  Washington: 

Elec.  Wld July  21,  1906 

Park  Dam  Conqumy,  Iowa: 

Eng.  Rec July  6,  1912 

Parr  Shoals  Power  Company,  South  Carolina: 

Power January  20,  1914 

Elec.  Engng April,  1914 

Elec.  Age October,  1915 

Patapsco  Electric  &  Mfg.  Company,  Maryland: 

Elec.  Wld August  3, 1907 

Penn  Iron  Mining  Company,  Michigan: 

Iron  Age July  16, 1908 

Peninsular  Power  Company,  Wisconsin: 

Eng.  Rec May  24, 1913 

A.I.M.E February,  1915 

Elec.  Jour July,  1916 

Pennsylvania  Water  &  Power  Company: 

Eng.  News September  12, 1907 

Eng.  Rec September  21, 1907 

Elec.  Rev February  27, 1909 

Eng.  Rec May  28,  1910 

Elec.  Wld October  20, 1910 

Elec.  Wld May  4,  1911 

Elec.  Wld August  24,  1912 

Eng.  Rec. December  7, 1912 

Power May  13,  1913 

Eng.  Rec March  20,  1915 

A.I.E.E May,  1916 

Philadelphia  Hydro-Electric  Company,  Pennsylvania: 

Elec.  Wld December  29,  1910 


Digitized  by 


Google 


APPENDIX  I  773 

Pfabftnka  Rivei  Power  Derelopment,  Braxll: 

Elec.  Wld November  14,  1908 

Pikes  Peak  Hydro-Blectric  Company,  Colontdo: 

Eng.  Rec July  19, 1902 

Elec.  Wld July  26,  1902 

Eng.  News January  1,  1903 

Eng.  Rec May  19,  1906 

Elec.  Wld May  26,  1906 

Ffttsford  Power  Company,  Vermont: 

Power. April  13,  1915 

Elec.  Wld May  22,  1916 

Eng.  News July  1, 1916 

Phmfledge  River  Power  DeTelopment»  Canada: 

Eng.  Rec September  20,  1913 

Eng.  Rec September  27, 1913 

Eng.  News October  23,  1913 

Portfand  Electric  Company,  Maine: 

Elec.  Wld October  12, 1907 

Portland  Railway,  Light  &  Power  Company,  Oregon: 

Eng.  News June  27,  1907 

Elec.  Wld April  11,  1908 

Elec.  Wld April  18,  1908 

Elec.  Wld April  25,  1908 

Elec.  Wld December  23,  1911 

Elec.  Wld July  13, 1912 

Jour  of  Elec April  6, 1912 

Jour,  of  Elec January  4,  1913 

Jour,  of  Elec January  2,  1915 

Jour,  of  Elec June  5, 1915 

Preaompscot  Electric  Company,  Maine: 

Eng.  Rec November  2,  1912 

Pueblo  and  Suburban  Traction  &  Ligliting  Company,  Colorado: 

Elec.  Rev January  26, 1907 

Paget  Sound  Ttaction,  Liglit  &  Power  Company,  Washington: 

Elec.  Wld October  1,  1904 

Elec.  Wld October  8, 1904 

Eng.  Rec March  17,  1910 

Eng.  Rec January  13,  1912 

Jour,  of  Elec April  13,  1912 

Eng.  Rec April  13,  1912 

Eng.  News April  11,  1912 

Eng.  News April  18,  1912 

Elec.  Wld June  1,  1912 

Jour,  of  Elec June  1,  1912 

Jour,  of  Elec January  2,  1915 

Jour  of  Elec June  5,  1915 


Digitized  by 


Google 


774  APPENDIX  I 

Quebec-Jacqttes-Cartier  Electric  Company,  Canada: 

Elec.  Wld June  9,  1900 

Can.  Elec.  News June,  1902 

Quebec  Railway,  Light  &  Power  Company,  Canada: 

Can.  Engr January,  1902 

Can.  Elec.  News June,  1902 

Raven  Lake  Portland  Cement  Company,  Canada: 

Elec.  Wld April  2, 1904 

Raystown  Water  Power  Company,  Pennsylvania: 

Eng.  Rec June  28,  1913 

Rio  de  Janeiro  Tramway,  Light  ft  Power  Company,  Brazil: 

Elec.  Wld May  13,  1909 

Elec.  Wld August  12,  1909 

Elec.  Wld : .  .April  26,  1913 

Elec.  Rev January  5, 1907 

Roaring  Fork  Electric  Light  &  Power  Company,  Colorado: 

Elec.  Rev Jan.  6,  1907 

Rochester  Railway  &  Light  Company,  New  York: 

Elec.  Wld January  14,  1909 

Elec.  Wld January  28,  1909 

Elec.  Wld February  18,  1909 

Power November  14, 1916 

Rock  Creek  Power  &  Transmission  Company,  Oregon: 

Elec.  Wld September  3,  1904 

Rockingham  Power  Company,  North  Carolina: 

Elec.  Rev March  14,  1909 

Eng.  Rec April  4,  1908 

Rock  River  Hydro-Electric  Development,  Illinois: 

Elec.  Wld October  26,  1912 

Rocky  Ford  Milling  &  Power  Company,  Kansas: 

Elec.  Wld November  3,  1910 

Rogue  River  Electric  Company,  Oregon: 

Jour,  of  Elec June  5,  1909 

Rumford  Falls  Power  Company,  Maine: 

Elec.  Wld January  9,  1915 

Rttsselville  Water  &  Light  Company,  Arkansas: 

Elec.  Wld May  26,  1910 

Salmon  River  Power  Company,  New  Tork: 

Eng.  Rec October  11, 1913 

Eng.  Rec June  13,  1914 

Elec.  Wld June  13,  1914 

Elec.  Wld June  20,  1914 

Power March  9,  1915 


Digitized  by 


Google 


APPENDIX  I  775 

Saoitary  District  of  Chicago: 

Elec.  Rev February  8,  1908 

Elec.  Wld January  19,  1911 

Elec.  Rev December  5,  1914 

Elec.  Wld May  27,  1916 

San  Joaquin  Light  &  Power  Company,  California: 

Jour,  of  Elec November  28,  1908 

Eng.  Rec February  4,  1911 

Eng.  Rec February  11,  1911 

Jour,  of  Elec May  11,  1912 

Power July  16,  1912 

Jour,  of  Elec June  5,  1915 

Jour,  of  Elec June  10,  1916 

Sao  Paulo  Tramway,  Light  &  Power  Company,  Brazil: 

Can.  Elec.  News May,  1906 

Sault  Ste.  Marie  Water  Power  Development: 

Eng.  News September  25,  1902 

Elec.  Wld September  27,  1902 

Eng.  Rec ' February  21,  1903 

A.aC.E February,  1905 

Eng.  Rec November  2,  1907 

Schenectady  Power  Company,  New  York: 

Elec.  Rev March  27,  1909 

Gen.  Elec.  Rev April,  1909 

Elec.  Wld May  20,  1909 

Eng.  Rec July  24,  1909 

Seattte  Municipal  Power  Development,  Washington: 

Elec.  Wld February  27,  1904 

Elec.  Wld June  1,  1912 

Jour,  of  Elec .July  27,  1912 

Jour,  of  Elec May  3,  1913 

Jour,  of  Elec .January  2,  1915 

Elec.  Rev June  9,  1917 

Seattle-Tacoma  Power  Company,  Washington: 

Eng.  Rec March  17,  1910 

Eng.  Rec January  13,  1912 

Sewalls  Falls  Development,  New  Hampshire: 

Power .May  25,  1909 

Shawinigan  Water  &  Power  Company,  Canada: 

Can.  Engr April,  1901 

Elec.  Wld February  8,  1902 

Can.  Engr May,  1902 

Can.  Elec.  News December,  1904 

Elec.  Wld May  4,  1912 

Elec.  Wld May  11,  1912 


Digitized  by 


Google 


776  APPENDIX  I 

Sierra  Pacific  Company,  Nevada: 

See  Truckee  River  Gen.  Electr.  Company: 

Sierra  San  Francisco  Power  Company,  California: 

Jour,  of  Elec August  21,  1909 

Jour,  of  Elec September  4,  1909 

Jour,  of  Elec February  3,  1912 

Elec.  Wld. . . . May  29, 1915 

Jour,  of  Elec June  5, 1915 

Similkameen  Power  Con^iany,  Washington: 

Jour,  of  Elec June  5, 1915 

Sioux  Falls  Light  &  Power  Company,  South  Dakota: 

Elec.  Wld April  22,  1909 

Power June  22, 1909 

So.  Electn November,  1911 

Snell  Hydro-Electric  Development,  New  Tork: 

Eng.  Rec February  17, 1912 

Snow  Mountain  Water  &  Power  Company,  California: 

Elec.  Wld May  29,  1915 

Jour,  of  Elec June  5, 1915 

Sooth  bend  Electric  Company,  Michigan: 

Elec.  Wld May  30, 1903 

Southern  Aluminum  Company,  North  Carolina: 

Eng.  News June  11, 1914 

Southern  California  Edison  Company: 

Elec.  Wld February  26, 1905 

Elec.  Wld March  4,  1905 

Elec.  Wld March  11,  1906 

Elec.  Wld March  25,  1905 

Elec.  Wld April  8,  1905 

Elec.  Wld August  10,  1907 

Elec.  Wld August  17,  1907 

Elec.  Wld August  24,  1907 

Elec.  Wld August  31,  1907 

Eng.  News December  24, 1908 

Elec.  Rev March  26, 1911 

Elec.  Rev April  1,  1911 

Elec.  Rev April  8,  1911 

Power September  5, 1911 

Jour,  of  Elec June  5,  1915 

Elec.  Wld Dec.  9,  1916 

Southern  Indiana  Power  Company: 

Elec.  Wld May  18,  1911 

Eng.  Rec February  10, 1912 

Southern  Power  Company,  North  Carolina: 

Elec.  Wld July  23,  1904 

Elec.  Wld May  25,  1907 


Digitized  by 


Google 


APPENDIX  I  777 

Eng.  Rec May  18, 1907 

Eng.  Rec May  26,  1907 

Eng.  Rec June  1,  1907 

Elec.  Jour December,  1907 

A.I,E.E June,  1908 

Power January  6,  1909 

Eng.  Rec April  3,  1909 

Gen.  Elec.  Rev December,  1909 

Elec.  Wld March  24,  1910 

Eng.  Rec April  2,  1910 

Mec.  Jour April,  1911 

Elec.  Rev May  6,  1911 

Elec.  Wld July  1,  1911 

Elec.  Wld Sept.  16,  1911 

Elec.  Wld May  30,  1914 

Else.  Wld March  27,  1916 

Power March  27,  1917 

Eng.  Rec Feb.  10,  1917 

Eng.  News Feb.  16,  1917 

Soofhem  Sierras  Power  Company,  Nevada: 

Elec.  Wld August  10,  1912 

Jour,  of  Elec July  5,  1913 

Jour,  of  Elec July  12,  1913 

Elec.  Wld August  2,  1913 

Elec.  Wld October  17,  1914 

Elec.  Wld October  24,  1914 

Elec.  Wld October  31,  1914 

Elec.  Wld November  7,  1914 

Elec.  Wld November  14,  1914 

Elec.  Wld November  21,  1914 

Elec.  Wld November  28,  1914 

Elec.  Wld December  6,  1914 

Mec.  Wld December  12,  1914 

Elec.  Wld December  19,  1914 

Jour,  of  Elec June  6,  1915 

Elec.  Wld April  14,  1917 

Soufhem  Utah  Power  Company: 

Jour,  of  Elec June  6,  1915 

Soufhem  Wisconsin  Power  Company: 

Elec.  Rev August  28,  1909 

Eng.  Rec September  4,  1909 

Eng.  Rec September  18,  1909 

Elec.  Wld September  23,  1909 

Spokane  &  Inland  Empire  R.  R.  Company: 

Eng.  Rec July  20,  1907 

Eng.  Rec October  10,  1908 


Digitized  by 


Google 


778  APPENDIX  I 

Elec.  Wld October  10,  1908 

Jour,  of  Elec January  2, 1915 

Spooner  Municipal  Hydro-Electric  Development,  Wisconsin: 

Eng.  Rec August  29,  1908 

Spring  River  Power  Company,  Kansas: 

Elec.  Rev November  18, 1905 

St.  Anthony  Falls  Water  Power  Company,  Minnesota: 

Eng.  Rec. May  29,  1909 

St  Croix  River  Power  Company,  Wisconsin: 

A.I.E.E November,  1900 

Elec.  Rev April,  1914 

St  Lawrence  River  Power  Company,  Canada: 

Eng.  Rec November  3,  1900 

Eng.  News February  21,  1901 

Elec.  Rev July  27,  1901 

St  Paul  Gas  &  Electric  Company,  Wisconsin: 

Elec.  Rev April,  1914 

City  of  Sturgis  Municipal  Hydro-Electric  Development,  Michigan: 

Elec.  Wld August  25,  1910 

Eng.  Rec March  2,  1912 

Superior  Portland  Cement  Company,  Washington: 

Eng.  Rec August  22,  1908 

Jour,  of  Elec January  2,  1915 

Tacoma  Municipal  Power  Development,  Washington: 

Eng.  News March  17,  1910 

Jour,  of  Elec March  1,  1913 

Elec.  Wld August  2,  1913 

Jour,  of  Elec January  2, 1915 

Tallassee  Power  Co.,  North  Carolina: 

Elec.  Wld Nov.  25,  1916 

Telluride  Power  Company,  Utah: 

(See  also  Utah  Power  &  Light  Company). 

Eng.  Rec March  14,  1908 

Eng.  Rec March  26,  1910 

Elec.  Wld November  18,  1911 

Elec.  Wld November  25,  1911 

Elec.  Wld December  9,  1911 

Elec.  Wld December  16,  1911 

Elec.  Wld December  23, 1911 

Tennessee  Power  Company: 

Eng.  Rec June  22,  1912 

Elec.  Engng April,  1913 

Elec.  Engng February,  1914 

Elec.  Engng March,  1914 

Power March  17,  1914 


Digitized  by 


Google 


APPENDIX  I  779 

Eng.  Rec April  18, 1914 

Eng.  Rec May  16,  1914 

Elec.  Wld May  30,  1914 

Elec.  Age August,  1916 

Towaliga  Falls  Power  Company,  Georgia: 

Eng.  Rec March  9, 1907 

Toronto  &  Niagara  Power  Company: 

Eng.  Rec February  13,  1904 

Can.  Engr July,  1904 

Eng.  Rec October  8,  1904 

Elec.  Wld January  7,  1905 

Eng.  Rec April  8,  1905 

Elec.  Rev December  2, 1905 

Trinity  Gold  Mining  &  Refining  Company,  California:  . 

Jour  of  Elec May  27, 1911 

T^ckee  River  General  Electric  Company,  Nevada: 

Elec.  Rev September  22  1906 

Jour,  of  Elec November  30,  1912 

Jour,  of  Elec June  5,  1915 

Tamers  Falls  Power  &  Electric  Company,  Massachusetts: 

Elec.  Rec September,  1914 

Gen.  Elec.  Rev March,  1917 

Elec.  Wld April  21,  1917 

Elec.  Rev February  17,  1917 

Twin  Falls  Hydro-Electric  Development,  Michigan: 

Eng.  Rec May  24,  1913 

Eng.  Rec May  31, 1913 

Uncas  Power  Company,  Connecticut: 

Eng.  Rec November  21,  1908 

Elec.  Wld October  28,  1909 

United  Missouri  River  Power  Company: 

Eng.  Rec August  13,  1  10 

Eng.  News October  20,  1910 

United  States  Reclamation  Service,  Boise,  Idaho: 

Eng.  Rec August  24,  1912 

Power May  4,  1915 

United  States  Reclamation  Service,  Snake  River,  Minidoka,  Idaho: 

Eng.  Rec January  8,  1910 

Eng.  Rec February  19,  1910 

Elec.  Rev May  13,  1911 

Elec.  Rev May  20,  1911 

Elec.  Wld December  30,  1911 

United  States  Redamation  Service,  Salt  River,  Arizona: 

Eng.  Rea December  31,  1910 

Elec.  Wld. , March  30,  1911 


Digitized  by 


Google 


780  APPENDIX  I 

A.LE.E April,  1911 

Elec.  Rev December  30,  1911 

Eng.  Rec January  1,  1916 

Elec.  Jour December,  1916 

Utah  County  JAfjbt  &  Power  Company: 

Eng.  Rec May  9,  1 908 

Utah  Light  &  Railway  ComjMuiy: 

Eng.  Rec April  2, 1910 

Utah  Power  &  Lififht  Company: 

(See  also  Telluride  Power  Company). 

Jour,  of  Elec May  8,  1914 

Elec.  Rev October  24, 1914 

Elec.  Wld June  5,  1915 

Jour,  of  Elec. . .- June  5,  1915 

Elec.  Wld May  27,  1916 

Utah  Sugar  Company: 

Elec.  Wld June  18,  1904 

Elec.  Wld June  25,  1904 

Eng.  News April  13,  1906 

Utica  Gas  &  Electric  Company,  New  Tork: 

Elec.  Wld May  19,  1906 

Elec.  Rev February  23, 1907 

Vancouver  Island  Power  Company,  Canada: 

Elec.  Wld October  12,  1912 

Elec.  Wld October  19,  1912 

Eng.  Rec October  19,  1912 

Power November  9,  1916 

Elec.  Rev August  28,  1916 

Elec:  Age 9 February,  1916 

Vancouver  Power  Company,  Canada: 

Eng.  Kec July  13,  1907 

Eng.  Rec September  21, 1912 

Elec.  Wld July  24,  1916 

Ventura  Power  Company,  California: 

Jour,  of  Elec June  6, 1915 

Vermont  Marble  Company: 

Elec.  Wld July  29, 1911 

Virginia  Power  Company,  West  Virginia: 

Elec.  Wld July  31,  1916 

Warren  Hydro-Electric  &  Gas  Company,  Ohio: 

Elec.  Wld Septemoer  2, 1909 

Washhigton-Oregon  Corporation: 

Jour,  of  Elec ,,««... .  Januaiy  2, 1916 


Digitized  by 


Google 


APPENDIX  I  781 

Washington  Water  Power  Company: 

Elec.  Wld May  23,  1908 

Mac.  Wld May  30,  1908 

Eng.  Rec May  25,  1912 

Elec.  Wld June  22,  1912 

Elec.  Wld June  29,  1912 

Jour,  of  Elec April  18,  1914 

Jour,  of  Elec April  25,  1914 

Elec.  Wld May  2,  1914 

Jour,  of  Elec September  5,  1914 

Eng.  Rec September  19,  1914 

Jour,  of  Elec January  2,  1916 

Jour,  of  Elec. June  5,  1916 

Elec.  Rev June  16,  1917 

Watab  Pulp  &  Paper  Company,  Minnesota: 

Elec.  Rev February  22,  1908 

Watauga  Power  Company,  Tennessee: 

Eng.  Rec November  11,  1911 

Power March  26,  1912 

Wateree  Power  Co.,  South  Carolina: 

Eng.  Rec February  10,  1917 

Weber  &  Davies  Counties  Company,  Utah: 

Elec.  Wld December  7,  1912 

Eng.  Rec December  14,  1912 

Welland  Canal  Power  Development,  Canada: 

Elec.  Wld January  21,  1906 

Elec.  Wld January  28,  1906 

Wenatchee  Valley  Gas  &  Electric  Company,  Washington: 

Jour,  of  Elec January  2,  1916 

Jour,  of  Elec Jime  6,  1916 

Western  Canada  Power  Company: 

Eng.  Rec February  26,  1911 

Elec.  Wld July  20,  1912 

Elec.  Wld September  7,  1912 

Jour,  of  Elec August  30,  1913 

Jour,  of  Elec January  2,  1916 

Jour,  of  Elec June  6, 1916 

Western  Colorado  Power  Company,  Colorado: 

Jour,  of  Elec June  6,  1915 

Western  States  Gas  &  Electric  Company,  California: 

Elec.  Wld May  29,  1916 

Jour,  of  Elec June  6,  1916 

West  Kootenay  Power  &  Light  Company,  Canada: 

Eng.  Rec October  5,  1907 

Elec.  Wld July  27,  1912 

Jour,  of  Elec. .  .  i June  5,  1915 


Digitized  by 


Google 


782  APPENDIX  I 

Whatcom  County  Railway  &  Light  Company,  Washington: 

Elec.  Wld July  20,  1912 

White  River  Power  Company,  Wisconsin: 
'    Elec.  Wld May  4,  1911 

Winnipeg  General  Power  Company,  Canada: 

Elec.  Wld June  23,  1906 

Winnipeg  Municipal  Hydro-Electric  Works,  Canada: 

Can.  Elec.  News November,  1906 

Eng.  Rec October  9,  1909 

Elec.  Rev December  2,  1911 

Eng.  News July  4,  1912 

Can.  Elec.  News June  1,  1915 

Wisconsin-Minnesota  Light  and  Power  Company: 

Elec.  Rev March  31,  1917 

Elec.  Wld February  6,  1916 

Wisconsin  River  Power  Company: 

Elec.  Rev May  19,  1917 

Yadkin  River  Power  Company,  North  Carolina: 

So.  Electn March,  1913 

York  Haven  Water  &  Power  Company,  Pennsylvania: 

Elec.  Wld March  2,  1907 


Digitized  by 


Google 


APPENDIX  II 


783 


Q 


I 


I 


O 
l^ 

H 

O 

I 

O 

S 

I 
QQ 

2 

s 

;z; 


o 

H 

< 

Q 
O 

;z; 
g 


J:  .2  ^  S  ^  .. 


d 

s 

a 

.2 

K 

o 

J: 

09 

< 

1 

& 

z 

fl 

^ 

o 

8 

g 

■S 

s 

i 

« 

:i 

6 


3 


fita 


1-^ 

go. 


•i    8 

S.9  3 


H  S 


ITS     O 

re  ij< 


s 


I 


f5 

'Is 
■5  S 


Digitized  by 


Google 


784 


APPENDIX  II 


-3  S-  S  a 


0»         Oft 


«       * 


5 

"3 


S5« 


I 


5 

•3 


III 

8S' 


9 
5 


I 


H    S 


g 


§ 


l-*l«* 
°'^i--i 


I 

I 

2 


I' 


i 

I 


o 


1 


Digitized  by 


Google 


APPENDIX  II 


785 


l■^|| 


j>%t 


;is 


■s-nr^^ 


o 


a 


J 


^«' 


a 
§ 


5 


£5 


25 


1^ 


a 
I 

I 


Digitized  by 


Google 


786 


APPENDIX  n 


N*s 


M       C4       eo 

w^  wt  ^t 

Oft  O)  Oft         O)  o 


.Mr 


d 

a 

o 

s 

fi 

s 

S 

«< 

1 

K 

H 

»; 

a 

.2 

I 

£ 

1 

J 

i 

P 

H 


■I 


O  :< 


|o 


I 


d 
o 
O 


a 

t 


9 


I 


:a 

1 

< 


■So 


I 


3    S 


Digitized  by  CjOOQ IC 


APPENDIX  n 


787 


V  '3 


a 
a 
o 
O 


.» 

S 


s 

3 


H 

a 

M 

OD 

H 

s 


Si 


a 

s 

OB 

OQ 

•s 

a 

a 

3 

I 


hthi 


H  S 


"flS 

2iS 


go 

S  s 


P3 


d 

U 

o 

If 

8 

Oi 

S 

•« 

« 

♦»3 

& 

^ 

s 

« 

a 

•T3 

n 

«> 

Ai 

o 

Digitized  by 


Google 


APPENDIX  III 


STANDARD  TESTING  CODE  FOR  HYDRAULIC  TURBINES 


The  following  Code  has  been  prepared  by  a  Committee  of  the  Hydraulic 
Turbine  Manufacturers  to  assist  in  avoiding  misunderstandings  in  regard 
to  stipulated  performances  of  hydraulic  turbines.  It  is  subject  to  such 
revision  from  time  to  time  as  will  be  required  by  any  new  developments 
in  turbine  testing  methods. 

iNTRODnCITON 

1.  Intended  Scope.  Hydraulic  turbine  tests  are  of  two  distinct  kinds: 
First,  acceptance  tests  on  completed  tiu*bines  after  installation  in  the  power 
plant;  second,  experimental  tests  either  on  full-sized  tiu-bines  or  models, 
carried  out  at  manufacturers'  laboratories  or  at  a  testing  flume.  Tests  of 
the  first  kind  are  for  the  purpose  of  determining  the  fulfillment  or  non- 
fulfillment of  contracts  between  the  turbine  builders  and  the  purchasers. 
Tests  of  the  second  kind  are  carried  out  for  the  purpose  of  obtaining  experi- 
mental data  on  which  the  design  of  an  installation  may  be  based;  for  sci- 
entific research  work;  or  for  the  investigation  of  special  problems.  This 
code  is  intended  to  apply  only  to  tests  of  the  first  kind.  When  tests  of  the 
second  kind  are  used  for  determining  the  performance  of  a  full-sized  in- 
stallation, this  application  should  be  made  only  in  accordance  with  principles 
which  will  be  stated  in  section  10,  below. 

2.  Principal  Factors,  Meaning  and  Intent  of  Terms  Used.  In  cod>- 
puting  the  efficiency  of  an  installation  a  distinction  must  be  made  between 
the  efficiency  of  the  plant  and  the  efficiency  of  the  turbine.  The  efficiency 
of  the  plant  may  include  all  losses  of  energy  up  to  any  stated  point  of 
delivery,  such  as  the  delivery  of  electric  power  from  the  transformers,  at 
the  switchboard  or  at  the  generator  terminals,  or  may  be  confined  to  the 
total  efficiency  of  the  hydraulic  installation,  for  which  purpose  the  power 
is  to  be  computed  as  that  delivered  by  the  turbine  to  the  generator  shaft. 

For  the  piurpose  of  computing  the  plant  efficiency  the  total  or  gross  head 
acting  on  the  plant  is  to  be  used,  and  is  to  be  taken  as  the  difference  in 
elevation  between  the  equivalent  still-water  surface  before  the  water  has 
passed  through  the  racks,  to  the  equivalent  still-water  surface  in  the  tail- 
race  after  discharge  from  the  draft  tube.  When  the  water  in  the  forebay 
in  advance  of  the  racks  flows  with  sufficient  velocity  to  make  its  vdocity 
head  an  appreciable  quantity,  the  actual  elevation  of  the  water  surface  shaO 
be  mcresj^  by  the  amount  of  this  velocity  head.    The  same  process  shall 

788 


Digitized  by 


Google 


APPENDIX  ni  789 

apply  to  the  point  of  measurement  in  the  taihrace;  that  is,  the  velocity 
head  at  the  point  of  measurement  in  the  tailraoe  shall  be  added  to  the  actual 
elevation  of  the  suif ace,  the  sum  being  considered  the  equivalent  still-water 
elevation. 

Except  where  specifically  stated  herein,  this  code  shall  be  understood 
to  apply  to  tests  of  the  turbine  proper,  and  the  terms  power,  efficiency, 
effective  head,  etc.,  are  to  be  taken  as  referring  to  the  turbine.  In  com- 
puting the  efficiency  of  the  turbine,  the  losses  through  the  racks,  in  the 
intake  to  the  penstocks,  and  in  the  penstocks  shall  not  be  charged  against 
the  turbine;  nor  shall  tiie  head  necessary  to  set  up  the  velocity  required  to 
discharge  the  water  from  the  end  of  the  draft  tube  be  charged  against  the 
turbine.  The  net  or  effective  head  acting  no  the  turbine  shall  be  measured 
from  a  point  near  the  intake  to  the  turbine  casing  in  turbines  equipped  with 
casings,  or  from  a  point  immediately  over  the  turbine  in  turbines  having  an 
open-flume  setting,  to  a  point  in  the  tailrace  in  the  manner  set  forth  below 
under  the  heading  ''Measurement  of  Head."  Since  the  tiu-bine  cannot 
develop  power  without  discharging  water,  a  correction  for  the  velocity  head 
required  to  discharge  the  water  into  the  tailrace  shall  be  added  to  the  taU- 
water  elevation;  and  a  similar  correction  applied  at  the  intake  to  encased 
turbines,  as  called  for  under  the  heading  ''Measurement  of  Head."  The 
power  developed  by  the  turbine  shall  be  taken  as  the  mechanical  power 
delivered  on  the  turbine  shaft  and  transmitted  by  the  turbine  shaft  to  the 
generator  or  other  driven  machine  or  system. 

In  drawing  up  a  general  code  it  is  recognized  that  under  particular  cir- 
cumstances sometimes  occurring,  methods  of  measuring  or  computing  certain 
factors  entering  into  the  test  different  from  those  specified,  may  appear  pos- 
sible and  reasonable;  it  is,  however,  the  intent  of  this  code  that  the  meaning 
of  the  terms  efficiency,  effective  head,  etc.,  shall  be  the  efficiency,  effective 
head,  etc.,  determined  as  herein  specified,  and  that  such  terms  shall  be  under- 
stood only  as  thus  defined.  _^ 

Genebal 

8.  Inspection.  Careful  inspection  should  be  made  before,  during,  and 
after  the  tests  to  insure  the  proper  operation  of  the  turbine  and  conditions 
of  measurement. 

The  turbine  runner,  guide  vanes,  and  casing  should  be  inspected  before 
and  after  test  to  guard  against  obstructions  clogging  the  vanes.  Any  change 
in  performance  during  a  test  should  be  investigated. 

4.  Operating  Conditions  During  Test.  Apparatus  installed  for  the  pur- 
pose of  the  test  shall  not  affect  the  performance  of  the  turbine  during 
the  test.  When  any  doubt  exists  regarding  this  point,  a  special  experiment 
shall  be  carried  out  to  detect  any  effect  of  removing  and  replacing  the 
apparatus  in  question,  other  conditions  being  maintained  constant. 

The  unit  shall  be  in  normal  operating  condition  throughout  the  test, 
and  shall  have  been  operated  under  load  for  an  aggregate  time  of  at  least 
three  days  prior  to  the  test. 

4.  (a)  Leakage.  Care  should  be  taken  that  all  air  inlets  into  the  draft 
tube  are  closed,  and  that  leakage  of  air  into  the  tube  or  drawing  of  air  into 


Digitized  by 


Google 


790  APPENDIX  HI 

the  penstock  intake  ub  not  taking  place,  as  indicated  by  exoessive  amounts 
of  air  in  the  discharge,  or  presence  of  vortices  in  the  intake.  Precautions 
against  leakage  of  water  from  penstock  or  turbine  casing  should  be  taken, 
particularly  through  drain  valves,  relief  valves  or  other  connections.  The 
rate  of  fall  of  the  standing  water  surface  in  the  turbine  casing  below  the 
point  of  intake  through  the  turbine  gates  should  be  observed  duhng  shut- 
down as  an  indication  of  possible  leakage. 

(&)  Unsteady  Conditions.  Tests  should  not  be  made  under  conditions 
of  changing  head,  load  or  speed.  Variations  of  load  during  an  individual 
run  shall  not  exceed  3  per  cent  above  or  3  per  cent  below  the  average  load, 
and  variations  of  head  shall  not  exceed  2  per  cent  above  or  2  per  cent  below 
the  average  head,  and  variations  of  speed  shall  not  exceed  1  per  cent  above 
or  1  per  cent  below  the  average  speed.  Instrument  calibrations  and  cor- 
rection curves  should  be  prepared  in  advance  of  the  test,  and  measures 
taken  to  enable  results  to  be  computed  as  quickly  as  possible  during  the 
coiu'se  of  the  test  or  before  the  work  of  testing  shall  be  considered  to  have 
been  completed. 

6.  Calibration  of  Instruments.  Important  instruments  shall  be  installed 
in  duplicate  and  all  instruments  shall  be  calibrated  both  before  and  after 
the  test.  Only  the  readings  of  those  instruments  in  which  the  two  cali- 
brations agree  shall  be  used  in  computing  the  results.  Where  results  are 
appreciably  altered  by  reason  of  instrument  calibrations  made  after  the 
test  disagreeing  with  those  made  before,  the  test  shall  be  repeated. 

6.  Conduct  of  Test.  Both  parties  to  the  contract  shall  be  represented  and 
shall  have  equal  rights  in  determining  the  methods  and  conduct  of  the  test. 

All  points  of  disagreement  shall  be  settled  to  the  satisfaction  of  both 
parties,  and  the  results  of  the  test  be  agreed  on  as  acceptable,  before  the 
test  shall  be  considered  terminated  or  the  test  equipment  removed. 

The  measiu^ment  of  the  various  quantities  entering  into  the  computa- 
tion of  tiu*bine  power  and  efficiency  shall  be  in  accordance  with  the  follow- 
ing regulations: 

Measurement  of  Power  Output 

7.  (a)  By  Electrical  Measurement  of  Generator  Oatpot  and  Generator 
Losses.  In  turbines  direct-connected  to  electrical  generators  the  power 
output  of  the  tiu*bine  may  be  measured  as  provided  below. 

The  intent  of  the  provisions  contained  herein  is  that  the  power  output 
of  the  turbine  shall  be  taken  as  the  power  output  of  the  generator  plus  all 
losses  supplied  by  the  turbine  up  to  the  point  of  measurement. 

The  generator  may  be  tested  for  efficiency  either  in  the  shops  of  the  builder 
or  after  installation,  the  losses  being  determined  either  by  direct  measurem^it 
of  input  and  output  or  by  the  separate-loss  method;  the  electrical  measure- 
ments being  carried  out  in  accordance  with  the  Standardization  Rules  of 
the  American  Institute  of  Electrical  Engineers  of  Septermber,  1916,  but 
subject  to  the  provisions  contained  herein. 

The  generator  losses  and  efficiency  as  herein  defined  are  for  the  generator 
considered  as  a  dynamometer,  and  are  independent  of  the  performance 
guarantees  of  the  generator  which  are  not  within  the  scope  of  this  code. 


Digitized  by 


Google 


APPENDIX  III  791 

The  generator  efficiency  shall  be  determined  for  the  values  of  load,  power- 
factor,  temperature  or  other  conditions  existing  during  the  turbine  test. 
When  the  generator  is  run  during  the  turbine  test  at  speeds  different  from 
that  used  in  the  generator  test,  the  generator  efficiency  shall  be  corrected 
for  the  changes  in  speed. 

When  practicable,  the  generator  is  to  be  separately  excited  during  both 
generator  and  turbine  tests,  and  the  excitation  loss  is  not  to  be  included 
in  computing  generator  efficiency,  and  is  therefore  also  to  be  omitted  in 
computing  turbine  output  during  the  turbine  test. 

When  determined  by  the  separate-loss  method,  the  generator  efficiency 
in  the  case  of  polyphase  alternators  when  separately  excited  is  to  be  taken  as 

(Kilowatt  Output  at  Generator  Terminals) 

7Kaowatt\^r/'fiar-u|«P«'»''^-K  /stray  Load-uJ  ^7*°'  1 

{ Output  }+{^tu«}+l-;^-}+{  ws  h[-^J 

all  losses  being  expressed  in  kilowatts. 

The  stray  load-losses  are  to  be  determined,  in  accordance  with  Paragraph 
458  of  the  above  Standardization  Rules  of  the  A.I.E.E.,  by  operating  the 
generator  on  short-circuit  and  at  the  current  corresponding  to  the  load  to 
be  used  in  turbine  test.  This,  after  deducting  the  windage  and  friction  and 
I^R  loss,  gives  the  stray  load-loss,  the  total  amount  of  the  loss  so  determined 
being  included  in  the  above  formula,  in  place  of  }  or  i  of  this  value  as  some- 
times used  in  former  practice.  It  is,  however,  understood  that  whenever 
under  the  special  conditions  of  an  installation  other  losses  exist,  these  are 
to  be  added,  in  accordance  with  the  second  paragraph  of  this  subdivision, 
to  the  stray  load-losses  determined  as  here  given. 

The  value  of  generator  windage  and  friction  should  be  directly  measured 
in  the  shop,  or  after  installation.  In  units  containing  direct-connected 
exciters,  the  windage  and  friction  may  be  measured  by  driving  the  generator 
by  the  exciter  run  as  a  motor.  When  the  windage  and  friction  cannot  be 
directly  measured,  it  is  to  be  taken  either  from  shop  tests  of  generators  of 
similar  design  or  from  a  retardation  test  made  after  installation.  When 
possible  more  than  one  method  should  be  used  in  order  to  obtain  a  check. 

In  making  such  a  retardation  test,  the  turbine  shaft  and  runner,  or  the 
turbine  runner,  are  to  be  disconnected  when  practicable  from  the  generator 
shaft,  in  order  to  enable  the  windage  and  friction  of  the  generator  alone  to 
be  computed.  When  the  turbine  shaft  or  runner  cannot  be  disconnected, 
the  generator  windage  and  friction  are  to  be  computed  by  deducting  from 
the  total  windage  and  friction  that  of  the  turbine,  which  for  this  purpose 
may  be  found  with  sufficient  accuracy  from  the  formula: 

Turbine  windage  and  friction  in  Kw.  ^^KBD^N*  in  which 

B = height  of  distributor  in  feet ;  ^ 

D= entrance  diameter  of  runner  in  feet; 
iV^= revolutions  per  second; 

K^&n  empirical  coefficient  which  may  be  taken  as  0.000115  as  deter- 
mined from  available  test  data. 


Digitized  by 


Google 


792  APPENDIX  ni 

In  computing  the  turbine  output  in  the  turbine  test,  this  is  to  be  taken 
as  the  kilowatt  output  of  generator  divided  by  the  generator  eflSdency  as 
computed  above,  the  result  being  converted  from  kilowatts  to  horse-power. 

If  an  exciter  generator  is  also  mounted  on  the  unit  shaft  and  is  used  to 
excite  the  unit  imder  test,  then  to  the  output  of  the  main  generator  com- 
puted as  above  without  reference  to  excitation  there  is  to  be  added  the 
kilowatt  output  of  exciter  divided  by  the  exciter  efficiency,  this  converted 
to  horse-power.  It  is  recommended,  however,  for  simplicity  that  when  pos- 
sible th^exciter  shall  be  run  without  load  and  the  unit  separately  excited. 

It  is  recommended  to  avoid  retests  and  to  provide  a  reliable  check,  that 
the  electrical  instruments  used  in  all  tests  be  installed  in  duplicate.  These 
instruments,  together  with  the  instrument  transformers,  shall  be  calibrated 
both  before  and  after  the  tests  in  the  same  condition  as  used  in  the  tests. 
When  tests  are  made  under  slightly  fluctuating  loads,  the  output  shall  be 
determined  both  by  indicating  wattmeters,  read  at  short  intervals,  and  by 
recording  watt-hour  meters.  During  the  turbine  test  the  speed  of  the  unit 
shall  be  observed  by  accurately  calibrated  tachometer  or  by  revolution 
counter. 

(6)  By  Absorption  Dynamometer.  When  a  dynamometer,  either  of 
the  Prony  brake,  friction  disc,  or  other  type,  is  used,  the  dynamometer  is 
to  be  so  arranged  as  to  avoid  imposing  either  end  thrust  or  side  thrust  on 
the  turbine  shaft  and  bearings,  or  to  avoid  adding  any  friction  load  which 
is  not  measured. 

The  brake  must  be  capable  of  operating  with  the  weighing  beam  floating 
free  of  the  stops  during  the  entire  duration  of  a  run.  A  dash  pot  or  equiv- 
alent device  may  be  used  to  assist  this  action  if  so  arranged  that  the  accuracy 
of  measiuing  the  actual  torque  acting  on  the  tiu'bme  shaft  is  not  impaired. 

The  dynamometer  must  be  so  constructed  that  the  lengths  of  all  lever 
arms  used  for  transmitting  and  reducing  the  loads  can  be  aocuratdy  measured. 
The  zero  load  of  the  dynamometer  must  be  capable  of  accurate  measurement 
and  should  not  be  large  in  comparison  with  the  net  load  to  be  measured. 

When  power  is  determined  by  dynamometer,  particular  care  is  to  be 
used  in  obtaining  accurate  measurement  of  the  speed  of  the  shaft.  If  ta^ 
chometers  are  used  these  are  to  be  frequently  calibrated  by  counting  the  revo- 
lutions over  an  ample  length  of  time.  Under  usual  conditions  it  is  recom- 
mended that  the  speed  be  directly  measured  by  revolution  counter,  a 
tachometer  being  also  used  as  a  check  and  to  indicate  variations  in  speed 
during  a  run. 

Mbasubement  of  Power  Input  ob  Wateb  Hobse*foweb 

8.  Measurement  of  Head.  The  intent  of  the  provisions  contained  herein 
for  the  measurement  of  head  is  the  true  determination  of  the  difference 
between  the  total  energy  contained  in  the  water  immediately  before  its 
entrance  into  tl^e  turbine,  and  its  total  energy  immediately  after  its  dis- 
charge from  the  draft  tube. 

The  turbine  shall  be  tested  if  possible  under  the  effective  head  stated  in 
the  contract,  and  at  the  speed  specified  in  the  contract.  If  during  the  test, 
however,  the  effective  head  shall  differ  from  the  specified  head  by  an  amount 


Digitized  by 


Google 


APPENDIX  in  793 

not  exceeding  10  per  cent  of  the  latter,  the  speed  of  operation  of  the  tur- 
bine shall  be  adjusted  to  correspond  to  the  head  under  which  the  test  is 
nmde.  The  principle  is  recognized  and  accepted  that  if  the  speed  is  changed 
in  proportion  to  the  square  root  of  the  head,  the  horse-power  output  will 
change  in  proportion  to  the  three-halves  power  of  the  head,  and  the  turbine 
efficiency  will  remain  the  same;  that  is,  when  the  head  differs  from  the 
value  specified  in  the  contract,  the  contract  guarantees  shall  be  considered 
to  apply  if  the  hydraulic  equivalents  of  the  power  and  speed  of  the  turbines 
are  substituted  for  the  power  and  speed  enumerated  in  the  contract.  The 
hydraulic  equivalent  of  the  speed  is  eqiial  to  the  specified  speed  multiplied 
by  the  square  root  of  the  ratio  of  the  effective  head  existing  during  the  test 
to  the  specified  effective  head.  The  hydraulic  equivalent  of  the  horse- 
power is  equal  to  the  specified  horse-power,  multiplied  by  the  three-halves 
power  of  the  ratio  of  the  effective  head  existing  during  the  test  to  the  specified 
effective  head. 

The  test  shall  not  be  carried  out  if  the  head  differs  from  the  contract 
value  by  more  than  10  per  cent  either  above  or  below,  or  if,  due  to  an  excess 
of  the  head  above  the  contract  value,  or  to  a  reduction  in  tailwater  eleva- 
tion, the  total  draft  head  approaches  within  5  feet  of  the  limiting  value 
corresponding  to  the  barometric  height.  By  total  draft  head  is  meant  the 
height  of  the  centerline  of  the  distributor  of  vertical  tiu'bines,  or  of  the 
highest  point  of  the  discharge  space  of  the  runner  of  horisontal  turbines 
above  tailwater,  added  to  the  velocity  head  at  the  point  of  minimum  internal 
diameter  of  the  runner  band. 

If  during  the  test  it  is  not  practicable  to  adjust  the  speed,  or  if  the 
final  calculation  should  show  the  speed  to  have  been  incorrectly  adjusted 
to  suit  the  head,  provided  that  the  discrepancy  in  speed  does  not  exceed 
2  per  cent  either  way  from  the  correct  value,  the  values  of  power  and  efficiency 
fihown  by  the  test  shall  be  corrected  on  the  basis  of  the  test  curves,  of  the 
eame  or  a  homologous  turbine,  made  at  a  testing  fliune  or  on  a  wheel  tested 
in  place  according  to  the  methods  of  this  code,  when  such  curves  are  available. 

(a)  Encased  Turbines.  In  turbines  having  closed  casings  the  head  is 
to  be  measured  by  at  least  two,  and  when  possible  not  less  than  four  pie- 
zometers located  in  a  straight  portion  of  the  penstock  near  the  turbine  casing 
intake,  and  by  two  or  more  rod  or  float  gauges  in  the  tailrace,  placed  at  points 
reasonably  free  from  local  disturbances. 

Such  board,  rod  or  float  gauges  are  to  be  free  of  velocity  effects,  and  if 
this  is  not  obtainable  when  the  gauges  are  set  in  the  open  channel,  they  shall 
be  placed  in  properly  arranged  stilling  boxes. 

All  piezometers  shall  be  connected  to  separate  gauges.  The  conditions 
of  measurement,  including  velocity  distribution,  length  of  straight  run  of 
penstock,  and  conditions  of  piezometer  orifices  shall  be  such  that  no  piezom- 
eter shall  vary  in  its  readings  by  more  than  20  per  cent  of  the  velocity  head 
from  the  average  of  all  the  piezometers  in  the  section  of  measurement.  The 
piezometer  orifices  shall  be  flush  with  the  surface  of  the  penstock  wall,  the 
passages  shall  be  normal  to  the  wall,  and  the  wall  shall  be  smooth  and  parallel 
with  the  flow  in  the  vicinity  of  the  orifices.  The  piezometer  orifices  shall 
be  approximately  i  inch  in  diameter,    If  any  piezometer  shall  be  obviously 


Digitized  by 


Google 


794  APPENDIX  in 

in  error  due  to  some  local  cause  or  other  condition,  as  indicated  by  its  reading, 
after  the  addition  of  the  velocity  head,  giving  a  head  in  excess  of  the  initial 
available  head  corresponding  to  the  elevation  of  the  surface  of  headwater, 
the  source  of  the  discrepancy  shall  be  found  and  removed,  or  the  piezometer 
eliminated. 

When  stilling  boxes  are  used  in  the  tailrace  the  conununication  between 
the  box  and  channel  shall  consist  of  one  or  more  piezometer  openings  in  a 
plane  surface  parallel  to  the  flow,  in  order  to  avoid  velocity  effects.  When 
board  gauges  are  used  at  the  side  of  the  channel,  they  shall  be  flush  with 
the  wall  surface. 

The  effective  head  on  the  turbine  is  to  be  taken  as  the  difference  between 
the  elevation  corresponding  to  the  pressiu-e  in  the  penstock  near  the  entrance 
to  the  turbine  casing,  and  the  elevation  of  the  tailwater  at  the  highest  point 
attained  by  the  discharge  from  the  unit  under  test,  the  above  difference  being 
corrected  by  adding  the  velocity  head  in  the  penstock  at  the  point  of  measure- 
ment and  subtracting  the  residual  velocity  head  at  the  end  of  the  draft  tube. 
The  velocity  head  in  the  penstock  shall  be  taken  as  the  square  of  the  mean 
velocity  at  the  point  of  meajBurement,  divided  by  2g;  the  mean  velocity 
being  equal  to  the  quantity  of  water  flowing  in  cubic  feet  per  second,  divided 
by  the  cross-sectional  area  of  the  penstock  at  the  point  of  measurement 
in  square  feet.  The  residual  velocity  head  at  the  end  of  the  draft  tube  shall 
be  taken  as  the  square  oi  the  mean  velocity  at  the  end  of  the  draft  tube, 
divided  by  2(7,  the  mean  velocity  being  equal  to  the  quantity  flowing  in 
cubic  feet  per  second,  divided  by  the  final  croes-eectionai  discharge  area 
of  the  closed  or  submerged  portion  of  the  draft  tube  in  square  feet. 

(&)  Open  Flume  Setting.  In  the  case  of  turbines  set  iU'Open  flumes,  the 
head  is  to  be  measured  by  board,  rod  or  float  gauges  located  immediately 
above  the  center  of  the  turbine,  and  by  board,  rod  or  float  gauges  in  the 
tailrace,  all  gauges  being  placed  at  points  reasonably  free  from  local  dis- 
turbances, and  not  less  than  two  gauges  lieing  installed  in  the  flume  and  not 
less  than  two  in  the  tailrace. 

Such  gauges  are  to  be  free  of  velocity  effects,a  nd  if  this  is  not  obtainable 
when  the  gauges  are  set  in  the  open  channel,  they  shall  be  placed  in  properly 
arranged  stilling  boxes.  When  stilling  boxes  are  used,  the  communication 
between  the  box  and  channel  shall  consist  of  one  or  more  piezometer  openings 
in  a  plane  surface  parallel  to  the  flow,  in  order  to  avoid  velocity  effects. 
When  board  gauges  are  used  at  the  side  of  the  channel,  they  shall  be  flush 
with  the  wall  surface. 

The  effective  head  on  the  turbine  is  to  be  taken  as  the  difference  be- 
tween the  elevation  of  the  free  water  surface  immediately  above  the  center 
of  the  turbine,  and  the  elevation  of  the  tailwater  at  the  highest  point  attained 
by  the  discharge  from  the  unit  under  test,  the  above  difference  being  cor- 
rected by  subtracting  the  residual  velocity  head  at  the  end  of  the  draft  tube. 
The  residual  velocity  head  at  the  end  of  the  draft  tube  shall  be  taken  as  the 
square  of  the  mean  velocity  at  the  end  of  the  draft  tube,  divided  by  2g;  the 
mean  velocity  being  equal  to  the  quantity  flowing  in  cubic  feet  per  second, 
divided  by  the  final  cross-sectional  discharge  area  of  the  closed  or  sub- 
merged portion  of  the  draft  tube,  in  square  feet. 


Digitized  by 


Google 


APPENDIX  ra 


795 


Measurement  op  Quantitt  op  Wateb 

9.  The  quantity  of  water  discharged  from  the  turbine  is  to  be  meas- 
ured by  one  of  the  following  methods.  It  is  recommended  that  whenever 
possible  -more  than  one  of  these  methods  be  used,  the  quantity  being  taken 
as  the  average  of  the  results  of  two  or  more  simultaneous  measurements. 

(a)  By  Weir.  When  the  quantity  of  water  is  measured  by  weir,  weirs 
with  suppressed  end  contractions  shall  be  used. 

The  weir  or  weirs  shall  if  possible  be  located  on  the  tailrace  side  of  the 
turbine,  and  care  shall  be  taken  that  smooth  flow,  free  from  eddies,  surface 
disturbances  or  the  presence  of  considerable  quantities  of  air  in  suspension 
exists  in  the  channel  of  approach.  To  insure  this  condition  the  weir  should 
not  be  k>cated  too  close  to  the  end  of  the  draft  tube,  and  stilling  racks  and 
booms-  should  be  used  when  required.  The  channel  of  approach  should  be 
straight,  of  uniform  cross-section  and  should  be  unobstructed,  by  racks  and 
booms,  for  a  length  of  at  least  25  feet  from  the  crest.  The  racks  should  be 
arranged  to  give  approximately  uniform  velocity  across  the  channel  of 
appiroach.  The  uniformity  of  velocity  should  be  verified  by  current  meter 
or  otherwise. . 

The  head  <m  the  weir  should  be  observed  by  hook  gauges  placed  in  stilling 
boxes  communicating  through  orifices  approximately  1  inch  in  diameter 
in  the  sides  of  the  channel  of  approach,  approximately  1  foot  below  the  level 
of  the  crest  and  a  distance  of  not  less  than  5  or  more  than  10  times  the  head 
upstream  therefrom,  the  head  being  observed  independently  at  both  sides 
of  the  channel.  In  measuring  quantities  of  water  corresponding  to  the 
loads  on  which  the  turbine  guarantees  are  based,  the  head  on  .the  crest  shall 

TABLE  OF  VALUES  OF  C  FOR  VARIOUS  HEADS  AND  HEIGHTS 

OF  CREST  P 


Head 

h 
in  Feet. 

Hbiqht  of  Cbbbt  P 

4 

5 

6 

7 

S 

9 

.  10 

12 

14 

16 

20 

I.O 

3.376 

3.356 

3.344 

3.335 

3.329 

3.325 

3.322 

3.317 

3.314 

3.311 

3.308 

1.2 

3.391 

3.366 

3.350 

3.339 

3.332 

3.326 

3.322 

3.316 

3.311 

3.308 

3.305 

1.4 

... 

3.379 

3.359 

3.346 

3.336 

3.330 

3.324 

3.316 

3.311 

3.307 

3.303 

1.6 

. . . 

3.370 

3.354 

3.343 

3.334 

3.328 

3.319 

3.312 

3.308 

3.302 

1.8 

... 

... 

3.363 

3.350 

3.340 

3.333 

3.322 

3.315 

3.309 

3.303 

2.0 

... 

... 

3.358 

3.347 

3.338 

3.325 

3.317 

3.311 

3.304 

not  be  more  than  two  (2)  feet  or  less  than  one  (1)  foot,  and  the  velocity  of 
approach  shall  not  be  greater  than  1  foot  per  second. 

The  discharge  shall  be  computed  by  the  Francis  formula  in  the  form  given 
below,  using  the  accompan3ring  table  of  coefficients.  These  coefficients  are 
believed  to  represent  the  best  available  information.  The  values  of  turbine 
efficiency  resulting  from  weir  tests  made  in  accordance  with  this  code  are 


Digitized  by 


Google 


796  APPENDIX  III 

understood  to  be  efficiencies  computed  by  the  use  of  the  formula  and  coef- 
ficients here  given. 

where  Q  ^quantity  in  cubic  feet  per  second; 
L  s length  of  weir  in  feet; 
^  » observed  head  above  crest  in  feet. 

P  is  the  height  of  the  crest  above  the  bottom  of  the  channel  of  approach' 
in  feet. 

To  facilitate  computations,  all  corrections  for  velocity  of  approach  have 
been  included  within  the  coefficients  as  given;  these  are  therefore  to  be 
used  in  the  formula  stated  above,  the  observed  head  being  used  without 
modification. 

Note:  The  above  coefficients  are  the  averages  of  values  computed  by 
the  following  three  formulas: 

(1)  Bazin, 

(2)  Behbock, 

(3)  Fteley-Steams, 

in  which  h,  «head  due  to  velocity  of  approach. 

The  weir  shall  be  sharp  crested,  with  smooth,  vertical  crest  wall,  complete 
crest  contraction,  and  free  overfall.  Complete  aeration  of  the  nappe  shall 
be  secured  and  observation  of  the  crest  conditions  and  form  of  nappe  shall 
be  made  during  the  test  to  avoid  defective  conditions  such  as  adhering 
nappe,  disturbed  or  turbulent  flow,  or  surging.  The  sidewalls  of  the  channd 
shall  be  smooth  and  parallel  and  shall  extend  downstream  beyond  the  over 
fail  above  the  level  of  the  crest. 

Weirs  of  a  length  exceeding  approximately  twenty  times  the  head  (ex- 
cepting in  cases  where  the  velocity  of  approach  is  extremely  low);  or  weirs 
of  moderate  crest  length  having  high  velocities  of  approach;  or  those  in 
which  the  velocity  of  approach  is  irregularly  distributed,  or  in  which  the 
leading  channel  is  subject  to  action  of  the  wind,  should  either  be  subdivided 
into  a  nimiber  of  sections  or  the  head  should  be  observed  not  only  at  both 
sides  but  also  at  intermediate  points  across  the  chamiel  of  approach.  The 
elevation  of  the  crest  should  be  measured  at  short  intervals  of  its  length 
in  determining  the  zero  readings  of  the  hook  gauges. 

(6)  By  Current  Meter.  When  the  discharge  is  measured  by  current 
meter,  observations  shall  be  taken  by  two  different  t3rpes  of  meter,  one  type 
having  preferably  such  characteristics  that  it  will  slightly  over-regi»ter 
under  conditions  of  turbulent  or  oblique  flow,  and  the  other  type  having 


Digitized  by 


Google 


APPENDIX  III  ;  797 

characteristics  such  that  it  will  under-register  under  similar  conditions.  The 
true  velocity  obtained  by  reducing  the  meter  readings  on  the  basis  of  their 
still-water  ratings  may  then  be  taken  as  a  weighted  mean  between  the  two 
series  of  observations. 

As  a  basis  for  arriving  at  the  proper  weighting  of  diverging  meter  results, 
the  instruments  in  question  should,  in  addition  to  their  regular  still-water 
ratings,  be  given  simultaneous  oscillation  or  angularity  tests  at  several 
velocities  near  those  which  will  probably  be  experienced  during  tests.  By 
means  of  the  resulting  data,  curves  showing  the  over-  and  under-registering 
characteristics  of  each  meter  may  be  plotted  for  varying  degrees  of  obliquity 
or  velocities  of  osciUation.  The  total  deviation  of  the  two  meters  may  then 
be  noted  for  any  obliquity  or  lateral  velocity.  When  the  relative  deviation 
of  the  two  meters  is  observed  in  the  field,  the  curves  will  then  indicate  the 
proportions  in  which  the  total  deviation  should  be  divided  to  give  the  proper 
correction  for  each  meter. 

The  point  method  of  observation  shall  be  used  and  sufficient  points  shall 
be  obtained  to  enable  both  vertical  and  horizontal  velocity  curves  to  be 
plotted  for  all  portions  of  the  section  of  measurement.  The  average  velocity 
shall  be  determined  from  these  curves  by  planimeter.  . 

The  section  of  measurement  shall  be  rectangular  and  smooth  flow  con- 
ditions shall  be  obtained.  It  b  reconmiended  that  in  order  to  avoid  abnor- 
mally long  durations  of  run  a  number  of  meters  of  each  type  be  used  simul- 
taneously. The  elevation  of  water  shall  be  continuously  observed  during 
the  current  meter  measurement  by  stilling  boxes,  piezometers,  or  other 
reliable  means.  If  the  supporting  rods  for  the  meters  are  in  the  same  plane 
as  the  meters,  the  area  of  these  rods  shall  be  subtracted  from  the  wetted 
area  of  the  flume  in  calculating  the  quantity.  The  meter  should  preferably 
be  supported  by  rods  placed  a  sufficient  distance  behind  them  to  avoid  any 
obstructive  effect.  When  a  heavy  mast  or  supporting  frame  is  used,  it 
should  be  designed  to  offer  a  minimum  disturbance,  and  should  be  located 
several  feet  downstream  from  the  meters. 

(c)  By  Pitot  Tube.  When  the  Pitot  tube  method  is  used,  the  Pitot 
tube  shall  be  located  in  a  straight  run  of  penstock  or  conduit,  at  a  distance 
equal  to  at  least  ten  pipe  diameters  from  any  upstream  bend  and  at  least 
five  diameters  from  a  downstream  bend.  When  the  observation  is  made 
in  a  circular  pipe  or  penstock,  at  least  two  Pitot  tubes  shall  be  arranged 
to  traverse  two  relatively  perpendicular  diameters,  but  in  the  case  of  very 
large  penstocks  or  those  having  imsymmetrical  flow,  Pitot  tubes  shall  be 
arranged  to  traverse  completely  or  partially  the  intermediate  diameters, 
giving  traverses  at  forty-five  degree  intervals. 

In  determining  the  velocity  in  the  penstock  by  the  Pitot  tubes  the  static 
pressure  over  the  cross-section  shall  be  measured  by  from  four  to  eight 
carefully  constructed  piezometers  equally  spaced  around  the  wall  of  the  pen- 
stock at  a  section  1  foot  in  advance  of  the  Pitot  tube  section  to  avoid  the 
effect  of  the  Pitot  tube  supporting  structure,  the  penstock  being  of  uniform 
cross-section  between  the  piezometers  and  the  points  of  the  Pitot  tubes. 
All  piezometers  shall  be  connected  to  separate  gauges.  The  conditions  of 
measurement,  including  velocity  distribution,  length  of  straight  run  of  pen- 


Digitized  by 


Google 


798  APPENDIX  in 

stock,  and  condition  of  piezometer  orifices  shall  be  such  that  no  piezometer 
shall  vary  in  its  readings  by  more  than  10  per  cent  of  the  velocity  head 
from  the  average  of  all  the  piezometers.  The  piezometer  orifices  shall  be 
flush  with  the  inside  surface  of  the  penstock  wall,  the  passages  shall  be  normal 
to  the  wall,  and  the  wall  shall  be  smooth  and  parallel  with  the  flow  in  the 
vicinity  of  the  onfices.    The  orifices  shall  be  i  inch  in  diameter. 

The  velocity  at  each  point  in  the  penstock  shall  be  computed  by  the 
formula  V  =  y/2gh,  in  which  h  represents  the  difference  in  feet  between  the 
total  dynamic  pressure  recorded  by  the  Pitot  tube  at  that  point  and  the 
average  static  pressure  recorded  by  the  piezometers.  The  velocities  so 
determined  shall  be  plotted  as  ordinates  against  values  of  the  areas  of  the 
sections  of  the  penstock  corresponding  to  the  points  of  measurement  aa 
abscissas,  a  smooth  curve  being  drawn  through  the  points  obtained.  The 
mean  velocity  in  the  penstock  will  then  be  taken  as  the  mean  ordinate  of 
the  above  curve  multiplied  by  0.976.  This  coefficient  is  based  on  the  average 
of  various  comparative  tests,  and  is  required  to  correct  for  oblique  or  sinuous 
flow  under  the  usual  conditions  in  straight  penstocks. 

When  the  length  of  straight  run  of  penstock  is  insufficient  or  when  the 
flow  is  disturbed  by  a  severe  bend  or  obstruction  upstream  from  the  tube 
or  when  the  average  velocity  is  less  than  6  feet  per  second,  the  above  coef- 
ficient will  not  apply  correctly,  the  correct  value  being  considerably-  lower 
in  such  cases,  which  do  not,  therefore,  come  within  the  scope  of  this  code. 
The  coefficient  corresponds  to  a  tube,  the  point  of  which  is  |  inch  in  diam- 
eter with  a  i  inch  hole,  the  face  being  normal  to  the  axis,  and  at  least  3  inches 
from  the  nearest  surface  of  the  supporting  pipe. 

(d)  By  the  Screen  or  Diaphragm  Method.  When  the  screen  method  is 
used  a  sufficient  length  of  straight  flume  of  uniform  cross-section  shall  be 
constructed  with  a  close-fitting  screen  filling  the  cros»-section.  Provision 
shall  be  made  for  accurately  observing  the  velocity  of  the  screen,  preferably 
by  electric  contacts  and  chronograph.  The  length  of  run  of  the  screen  shall 
be  sufficiently  in  excess  of  the  portion  used  for  measurement  to  orovide 
ample  space  for  starting  and  stopping  the  screen,  so  as  to  insure  uniform 
conditions  over  the  measured  portion  of  the  run.  In  determining  the  dis- 
charge the  velocity  of  the  screen  shall  be  multiplied  by  an  area  intermediate 
between  the  net  immersed  area  of  the  moving  screen  and  the  average  area 
of  stream  cross-section  of  the  portion  of  the  channel  traversed.  The  varia- 
tion of  the  level  in  the  flume  shall  be  observed  during  the  course  of  the  nm 
and  the  average  elevation  shall  be  used  in  determining  the  area. 

(e)  By  Titration  or  Chemical  Method.  When  ^e  chemical  method  is 
used  in  measuring  discharge,  care  shall  be  taken  to  insure  that  at  the  point 
of  introducing  the  dosing  solution  no  portion  of  the  solution  shall  be  carried 
off  by  back  currents  and  shall  therefore  fail  to  pass  to  the  sampling  station, 
and  that  the  sampling  station  shall  be  so  placed  that  no  pollution  shall  be 
caused  by  reverse  currents,  causing  fresh  water  to  pass  the  station  from  down- 
stream. Wlien  necessary,  owing  to  a  short  length  of  mixing  passage  or  lack 
of  sufficient  disturbance  to  cause  thorough  mixing,  the  dosing  pipes  shall 
be  so  placed  that  an  equal  degree  of  concentration  over  the  entire  section 
of  the  sampling  station  shall  be  obtained.    Samples  shall  be  taken  from 


Digitized  by 


Google 


APPENDIX  III  799 

points  distributed  over  the  entire  sampling  section.  All  necessary  precau- 
tions shall  be  observed  in  taking  samples,  and  in  observing  the  end-point 
of  the  reaction  during  titration. 

In  short  tests,  care  shall  be  taken  to  preserve  a  uniform  rate  of  intro- 
duction of  the  dosing  solution.  Preliminary  observations  shall  be  made  to 
determine  the  time  required  aft^r  the  dosing  is  started  for  uniform  conditions 
to  become  established  at  the  sampling  station;  and  in  the  actual  tests  the 
dosing  shall  be  continued  for  double  this  time  before  sampling  is  begun. 
Uniformity  of  dilution  of  samples  both  with  respect  to  location  in  the  section 
and  the  time  of  taking  shall  be  considered  essential  for  an  acceptable  test. 

Power  Tests  op  Turbine  Supplemented  by  Efficiency  Tests  of  a  Model 

10.  When  the  conditions  of  an  installation  are  such  as  to  involve  serious 
difficulty  or  expense  in  the  application  of  any  of  the  above  methods  of  water 
measurement,  the  tests  of  the  installed  turbine  may  be  made  when  accept- 
able to  both  parties  without  measuring  the  quantity  of  water,  a  homologous 
model  of  the  turbine  being  constructed  and  tested  at  the  expense  of  the 
purchaser,  and  the  power  delivered  by  the  installed  turbine  compared  with 
that  computed  from  the  model  tests. 

This  method  must  not  be  confused. with  the  practice,  which  has  some- 
times been  followed,  of  comparing  a  turbine  with  a  model  having  a  homol- 
ogous runner,  but  dissimilar  with  respect  to  setting,  draft  tube  or  other  parts. 
The  runner,  guide  vanes,  draft  tube,  casing,  or  other  adjacent  water  passages 
should  be  geometrically  similar  in  the  turbine  and  model;  and  when  so 
constructed,  the  power  stepped  up  from  the  model  tests  for  the  hydraulic 
equivalent  of  the  speed  gives  a  reliable  basis  of  comparison  with  the  power 
actually  obtained  from  the  installed  unit. 

The  power  of  the  model  when  operating  at  the  hydraulic  equivalent 
of  the  speed  of  the  large  unit  in  the  tests  of  the  latter,  at  the  same  propor- 
tional gate  opening,  is  to  be  multiplied  by  the  ratio  of  the  area  of  the  discharge 
orifices  of  the  large  turbine  runner  to  that  of  the  model,  and  by  the  three- 
halves  power  of  the  ratio  of  the  head  existing  in  the  tests  of  the  large  unit 
to  the  head  in  the  model  tests.  When  the  power  so  computed  agrees  exactly 
with  that  obtained  from  the  installed  unit,  the  efficiency  of  the  large  unit 
shall  be  considered  to  be  identical  with  that  of  the  model;  and  when  the 
power  of  the  large  unit  exceeds  that  thus  computed  from  the  model,  the 
efficiency  of  the  large  unit  shall  be  considered  to  be  in  excess  of  that  of  the 
model.  In  measuring  the  gate  opening  the  actual  opening  of  the  gates  shall 
be  determined,  and  care  shall  be  taken  to  avoid  errors  due  to  the  effect  of 
the  pressure  on  the  vanes. 

Appendix 

11.  Special  Methods  of  Water  Measurement.  The  following  methods 
of  water  measurements  may  sometimes  be  applied;  these  are,  however, 
subject  to  limitations,  and  are  available  only  under  special  conditions. 
They  have  not  as  a  rule  been  in  sufficiently  general  use  in  turbine  testing 
to  permit  full  reliance  to  be  placed  on  them  until  opportunities  are  afforded 
for  checking  them  against  the  methods  already  given. 

(a)  By  the  Bulk  or  Volumetric  Method.    Water  measurement  by  weight 


Digitized  by 


Google 


800  APPENDIX  m 

or  volume  is  not  usually  available;  the  fonner  is  limited  to  laboratory  use, 
which  is  outside  the  scope  of  this  code.  The  bulk  method  is  applicable 
only  when  there  is  available  a  reservoir  of  regular  form,  the  volume  of  which 
up  to  various  water  levels  may  be  accurately  measured,  and  when  the  fol- 
lowing conditions  may  be  observed: 

The  draw-down  or  filling  of  the  reservoir  must  not  cause  a  variation 
in  head  on  the  turbine  during  a  run  exceeding  the  limits  specified  under 
section  4  (&),  namely,  a  total  of  4  per  cent  of  the  head.  It  must  be  possible 
to  shut  off  completely  aU  inflow  into  or  outflow  from  the  reservoir.  The 
tightness  of  the  gates  and  reservoir  walls  must  be  tested  by  closing  aU  gates, 
and  observing  over  a  time  of  several  hours  the  rate  of  rise  or  fall  of  water 
level  in  the  reservoir  throughout  the  full  range  of  variation  of  level  which 
will  be  used  in  the  turbine  test.  At  the  same  time  any  leakage  through 
the  turbine  head  gates  is  to  be  measured.  The  surface  elevation  in  the 
reservoir  is  not  to  be  so  affected  by  velocity  or  wind  effects  as  to  cause  local 
variations  in  level  of  more  than  5  per  cent  of  the  total  draw-down  used  in 
the  turbine  tests.  This  variation  is  to  be  observed  by  gauges  distributed  over 
the  whole  reservoir,  which  are  to  be  read  simultaneously  at  short  intervals 
throughout  the  test.  The  effect  of  surface  evaporation  shall  be  investigated 
and  corrections  applied  to  cover  it. when  local  conditions  are  such  that  it 
becomes  appreciable. 

(6)  By  Venturi  Meter.  When  it  is  possible  to  install  a  Venturi  meter 
not  exceeding  in  dimensions  or  differing  in  conditions  from  meters  whose 
coefficients  have  previously  been  determined  in  accurate  tests,  the  Venturi 
meter  may  be  used.  The  meter  shall  be  similar  in  proportions  to  meter 
previously  tested. 

(c)  By  Color  Velocity  Method.  When  the  water  used  by  the  turbine 
passes  through  a  conduit  suited  to  the  purpose,  the  color  method  of  quantity 
determination  may  be  used,  depending  upon  the  time  of  passage  between 
two  points  of  a  mass  of  color  injected  into  the  stream.  The  distance  between 
the  two  points  where  the  passage  of  the  color  is  observed  must  be  sufficiently 
great  to  render  the  interval  between  the  times  of  passage  of  the  color  at 
the  two  stations  large  compared  to  the  time  required  for  all  the  color  to  pass 
either  station.  The  conduit  must  be  of  sufficiently  regular  form  to  per- 
mit its  cro6&-eectional  areas  to  be  accurately  measured  at  all  points  between 
the  stations. 

{d)  By  Brine  Velocity  Method.  A  method  similar  to  11  (c)  adapted  to 
closed  conduits  has  been  used,  consisting  in  the  injection  of  a  mass  of  brine, 
the  time  of  passage  of  which  is  detected  by  the  variation  in  electrical  resist- 
ance between  two  contacts  placed  in  the  stream.  .  A  pair  of  such  contacts 
is  placed  at  each  station,  and  the  time  of  passage  of  the  brine  between  the 
stations  is  chronographically  recorded  by  a  specially  arranged  wattmeter. 
The  stations  should  be  arranged  as  imder  11  (c). 

(e)  By  Color  Density  Method.  The  coloration  or  color  density  may  also 
be  employed  for  approximate  tests,  this  method  depending  on  the  use  of 
a  colored  dosing  solution  in  place  of  a  salt  solution  in  a  manner  similar  to 
the  chemical  method  of  9  (e),  observation  of  the  color  density  replacing 
the  titration. 


Digitized  by 


Google 


APPENDIX  m  801 

if)  By  Resistance  of  Salt  Soloti<m.  A  method  which  has  been  used  ex- 
perimentally is  similar  to  the  chemical  method  of  9  (e),  except  that  the 
amount  of  chemical  (salt)  in  solution  is  detennined  by  measurement  of  the 
electrical  resistance  of  the  solution  instead  of  by  titration.  Care  is  required 
to  guard  against  changes  in  resistance  due  to  small  temperature  variations. 

12.  Measurement  of  Water  Horse-power  in  Plants  Containing  a  Fall 
Increaser.  In  case  of  an  installation  including  a  fall  increaser  or  other 
device  utilizing  an  auxiliary  flow  for  increasing  the  effective  head,  the  fol- 
lowing provisions  shall  be  observed:  In  determining  the  efficiency  of  the 
turbine  proper,  considered  separately  from  the  fall  increaser,  the  faU  increaser 
shall  be  closed,  and  precautions  shall  be  taken  that  no  water  except  that 
passing  through  the  turbine  shall  enter  the  system  between  the  points  at 
which  the  head  is  measured. 

In  order  to  determine  the  performance  of  the  combined  hydraulic  instal- 
lation, including  both  turbine  and  fall  increaser,  the  total  water  horse-power 
shall  be  computed  from  the  sum  of  the  turbine  discharge  multiplied  by  the 
head  on  the  turbine,  and  the  auxiliary  discharge  multiplied  by  the  head  on 
the  fall  increaser.  The  head  on  the  turbine  shall  be  measured  from  a  point 
immediately  in  advance  of  the  point  of  intake  to  the  turbine  proper,  as  above 
provided,  and  the  head  on  the  fall  increaser  shall  be  measured  from  a  point 
inmiediately  in  advance  of  the  intake  gates  of  the  increaser,  the  head  in 
each  case  being  measured  to  a  point  below  the  junction  of  the  two  streams 
at  the  outflow  from  the  plant.  For  the  computation  of  water  horse-power 
it  will  be  necessary  to  determine  the  division  of  the  total  discharge  between 
the  turbine  and  fall  increaser.  This  may  be  done  when  practicable  by 
separately  measuring  the  water  admitted  to  the  tiu'bine  during  the  operation 
of  the  fall  increaser. 

If,  owing  to  the  arrangement  of  the  fall  increaser,  it  is  impracticable 
to  separate  the  water  horse-power  of  the  turbine  from  that  of  the  fall  increaser, 
the  gross  efficiency  of  the  combined  installation  may  be  determined  by  meas- 
uring the  combined  total  flow,  and  the  total  head  from  a  point  common 
to  the  two  flows  before  entering  the  plant  to  a  point  after  they  are  reunited 
below  the  final  point  of  discharge. 


Digitized  by 


Google 


Digitized  by 


Google 


INDEX 


A 

PAGS 

Absorption 49 

Agricultural  work 28 

A.I.E.E.  standardization  rules 308 

Air,  reluctance  of 289 

Air  tanks  for  pressure  regulation 142 

Air  valves 128 

Altitude,  effect  on  temperatiure 312 

Ammeters 540,  541 

Ammeter  transfer  plugs  and  receptacles 560 

Apparatus,  arrangement  of 175 

exciters 176 

general  consideration 175 

generators 176 

governors 176 

lightning  arresters 192 

reactors 178 

switching  equipment 179,  566 

transformers 176 

turbines 175 

transportation  and  erection 193 

Arcing  ground  suppressor 618 

Area,  land  and  water  of  United  States 14 

Armature  reactance 286 

reaction. . ; 285 

Atmospheric  pressure 42 

Auxiliary  stations.    See  Steam  aux.  stations 692 

B 

Banding  of  wooden-stave  pipe 131 

Bazin's  formula 106,  796 

Bearings,  generator 331 

thrust 333 

suspension 333 

turbine 241 

Bearing  value  of  soils 168 

Brakes 348 

Breathers,  transformers 454 

803 


Digitized  by 


Google 


804  INDEX 

PAOB 

Buft-baw 489,  665 

expansion 571 

heating 566 

mechanical  short-circuit  stresBes 471 

mimic 563 

permissible  current  density 566 

reactance 569 

sectionalizing 474 

skin  effect 569 

structure 563 

supports 569 

Bushings,  entrance 571 

oil  circuit  breaker 505 

transformer 442 

C 

Cables 625 

current-carrying  capacity 633 

ducts  and  conduits 628 

heating 472,  633 

insulation 625 

mechanical  short-circuit  stresses 471 

reactance  and  resistance 642 

single  vs.  multiple  conductors 629, 632 

sixe 633 

troubles 628 

voltage  tests 634 

Calibrating  terminals 561 

Canals 106 

concrete  lining 108 

cross-section 107 

evaporation 110 

seepage 110 

side  slopes 109 

Central  stations  in  United  States 25 

Chezy  formula 106,  116 

Choke  coils 617 

Circuit  breakers.    See  Oil  circuit  breakers 496 

Coal  production  in  United  States 13 

Commercial  opportunities 28 

agricultural  work 28 

electro-chemical  industries 34 

irrigation 30 

mining 33 

railroad  electrification 38 

Concrete  pipe 137 

Conductor  spacing 627 


Digitized  by 


Google 


INDEX  805 

PAGX 

Ck)iiduit 628 

Ck)imections,  system '. 486 

exciter 355 

generator  armature 282,  295 

instrmnents 551 

transformers 391 

Conservation  of  natural  fuel  resources 13 

Control  switches 561 

Cooling  water  for  transformers 381,  457 

Corona 639 

Corrosion  of  turbine  runners 233 

Cost  of  hydro-electric  plant 702 

development  expenses 703 

estimated  and  actual  costs 706 

overhead  charges 705 

physical  costs 704 

Cost  of  hydro-electric  power 742 

Cost  of  steam  power  stations 745 

Cost  of  steam  power 745 

Cranes 171,  195,  197 

Current-limiting  reactors.    See  Reactors 458 

Current  meters 65, 796 

Current  transformers 546,  563,  638 

Curve-drawing  instruments 541,  546 

D 

Dams 74 

arched 85 

buttressed 85 

choice  of  type 74 

classification 74 

earth-fill 76 

gravity 79 

location 74 

masonry 79 

multiple-arched 88 

pressure 79 

rock-fill 78 

rolling 99 

rules  governing  design 88 

timber  crib 75 

Depreciation » 743 

Developments,  history 1 

electrical 1,  10 

hydraulic ; 1, 9 

Disconnecting  switches 577 

Distribution  voltage 273 


Digitized  by 


Google 


806  INDEX 

/  FAQI 

Diversity  factor 678 

Drainage  area 649 

Drying,  exciters  and  generators 199 

transformers 445 

transformer  oil 449 

Ducts ess 


E 

Economical  aspects ^ 644 

auxiliary  stations 692 

available  energy 675 

cost  of  plants 702 

cost  of  power 742 

interoonne(;ted  systems 698 

investigating  an  enterprise 699 

load  and  diversity  factor 678 

power  demand 677 

primary  and  secondary  power 683 

water  power  reports 645 

water  storage 685 

Efficiency,  generators 312,  314,  791 

installation 788 

transformers 384 

turbines 211,  217,  791 

Electrical  developments 1, 10 

Electro-chemical  industries 34 

Energy,  available 675 

flowing  water 66 

kinetic 66 

potential 66 

Entrance  bushings 571 

Equivalents 68 

Erection  of  apparatus 196 

Erosion  of  turbine  runners 233 

Evaporation 47,  110,  164 

Excitation,  synchronous  generator 288,  290 

Exciters 350 

arrangement  in  power  house 176 

batteries 361 

capacity  and  rating 351 

characteristics i 352 

connections 355 

control 552 

drying 199 

insulation  resistance 200 

mechanical  design 353 


Digitized  by 


Google 


INDEX  807 

FAOB 

Exciters,  method  of  drive 353 

separate  excitation 350 

shunt  V8.  compound  wound 352 

speed 353 

voltage 351 

Expansion  joints  for  pipe  lines 128 


F 

Field  control 552 

Field  discharge  switches 554 

Field  rheostats 556 

Financial  aspects 644 

Fishways 99 

Flashboards 91 

Float  method  of  stream  flow  measuring 65 

Flood  gates 91 

Floods,  prevention 22 

Flow  of  water,  canals 106 

flumes, 110 

pipe  lines 114,  121 

tunnels 113 

Flow-summation  curve 685 

Flumes 110 

concrete Ill 

wood Ill 

steel 112 

Flux,  leakage 286,  377 

Flywheel  effect 247,  321,  327 

Foundations,  power  house 168 

Francis  formula 795 

Freezing  of  water  in  pipe  lines 129 

Frequency 273 

eflfect  on  generators 275 

illumination 280 

induction  motors 277 

railroad  electrification 279 

synchronous  converters 278 

transformers 275 

transmission  lines 276 

Frequency  changers 274 

Frequency,  high 596 

absorbers 602 

indicators 543 

Friction,  losses  in  pipe  lines 116 

coefficient 106.  117 

Fuel  resources  in  United  States 13 


Digitized  by 


Google 


808  INDEX 

G 

PAOl 

Gas  power,  relation  to  steam  and  water  power 27 

Gauges,  hook 61 

Gauging  stations 62 

Gates.    See  also  Valves 144 

gate  valves 148 

operation  and  control 149 

rolling 97 

tainter 97,148 

tilting 96 

sliding 93 

sluice 146 

wicket 220,236 

Generators,  induction 348 

comparison  with  synchronous  generators 348 

operation 349 

output  and  excitation 348 

utilisation 351 

Generators,  synchronous. 280 

A.I.E.E.  standardization  rules 308 

armatiu^  connections 282, 295 

armature  reactance 286 

armature  reaction 285 

armature  self-induction 286 

arrangement  in  power  house 176 

bearings 330 

brakes 348 

characteristics 290 

determination  of  efficiency .  .  .' 314 

division  of  load 321 

drying 199 

effect  of  altitude  on  temperature 312 

effect  of  power  factor 284 

efficiency 312,  791 

excitation  range 290 

excitation  required 288,  290 

erection 199 

flywheel  effect 321,  327 

frequency 275 

grounding  of  neutral 304 

horizontal  o».  vertical 324 

induced  E.M.F 280 

insulation  resistance 200 

leakage  flux 286 

losses 313 

lubrication 338 

mechanical  design 324 


Digitized  by 


Google 


INDEX  809 

PAOB 

Generators,  synchronous,  parallel  operation 318 

permissible  temperatures 309 

rating 307 

reactance 469 

regulation,  voltage 291 

repair ;. .  172 

saturation  curves 289 

short-circuit  current 292 

speed 315 

83mchronous  impedance 292 

synchronous  reactance 294 

temperature  measurements 310 

ventilation 173,  345 

voltage 270,  316 

voltage  regulation 291 

.  wave  form 282,  301 

windings 325,  330 

Governors 246 

action *. 249 

arrangement  in  power-house 176 

arrangement  and  operation 251 

energy  output 251 

methods  of  control ^  254 

power  cylinders 253 

pumping  outfit 252 

pressure  supply 252 

speed  regulation 220,  246 

Grade,  hydraulic 106,  117,  119 

Grounding,  generator  neutral 304 

lightning  arresters 612 

transformer  neutrals 392,  393,  396,  398,  399,  401,  404,  408, 

413,420 

transformer  secondaries 393,  550,  638 

Ground  detector,  electrostatic 545 

Guide  vanes 220,  236 

H 

Head 67,  114,  210 

effective  or  net 114,  789,  794 

elevation 114 

gross 67,  114,  788 

limitations 675 

loss 1 16 

measurement 792 

pressure 114 

variation 214 

Headworks 74 


Digitized  by 


Google 


810  INDEX 

FAOB 

Heating  of  power-house 175 

High  frequency 596 

High-frequency  absorbers 602 

History  of  hydraulic  and  electrical  developments 1 

Hook  gauge 61 

Hydraulic  gradient        119 

Hydtaulic  radius 106,  117 

Hydro-electric  systems,  data 783 

references  to  descriptions 757 

Hydrograph  records 66 

Hydrology 39 

I 

Ice 103 

Ice  guards 71 

lUumination  of  power-house 174,  538 

Impedance,  cables 642 

effect  of  parallel  operation  of  transformers 425 

natural 597 

synchronous 292 

transformer 426 

Indicators,  frequency 543 

power  factor 542 

synchronous 544 

temperature 545 

transformer  cooling  water 435 

Inductance 467 

Insulation,  generator 325,  330 

transformer 439 

wires  and  cables 625 

Insulators,  bus-bar 569 

Instruments 538 

ammeters 540, 541 

connections 551 

current  and  potential  transformers 546 

curve-drawing 641,  546 

electrostatic  ground  detectors 545 

frequency  indicators 543 

indicating  wattmeters 542 

power-factor  indicators 542 

reactive  volt-ampere  indicators 543 

synchronism  indicators 544 

temperature  indicators 545 

voltmeters 640,  641* 

watthour  meters 549 

Intakes,  water 100,  163 

Interconnected  systems 698 


Digitized  by 


Google 


INDEX  811 

FAGS 

Investigation  of  an  enterprise 699 

Irrigation 30 

K 

Kutter's  formula 106 

L 

Layouts  of  power  stations 180 

Leakage  flux,  generators 286 

transformer 377 

Lightning  arresters,  aluminum  cell 604 

charging 614 

charging  current  indicator 614 

charging  resistance 614 

choke  coils 617 

discharge  recorder 616 

grounding 612 

location 192,  609 

Line  drop  compensation 368 

Load,  division 321 

regulation  on  system 323 

curves 680 

factor 678,  682 

Location  of  development 644, 652 

Lubrication.    Sec  Oil .^338 

M 

Magnetizing  current,  transformers 376 

Management 746 

Manufacturing,  power  requirements 27 

Market  for  power 677 

Mass  curves 685 

Mechanical  stresses  on  short  circuits 471 

Meters,  electric.    See  Instruments 638 

price  current 65 

venturi 262 

water  flow 262 

Mimic  buses 563 

Mining  industry 33 

Multi  recorder 589 

N 

Natural  impedance 597 

Nozzles,  auxiliary  relief 224 

deflecting 222 

jet  deflecting 222 

needle 221 


Digitized  by 


Google 


812  INDEX 

O 

PAOB 

Qfl,  lubricating 108,  338,  341 

transformcc 46,  178,  443,  449,  451 

Oil  circuit  breakers 496 

bushings ' 505 

rating 497 

rupturing  capacity '. .  / 470,  497 

selection  of  type 497 

structures 563 

time  of  opening 470, 498 

types  and  design 498 

Oil  circuit-breaker  batteries 589 

Oil  production  in  United  States 15 

Organization  and  Operation 746 

management 746 

operating  force 746 

operating  and  maintenance  instructions 754 

operating  records 748 

Oscillations 596 

Outdoor  stations 192,  577 

Overspeed  of  turbines 226 

Over-voltage  protection 593 

arcing  ground  suppressor 613 

classification  of  over-voltages 693 

^   lightning  arresters 604 

protection  of  telephone  lines 620 

short-circuit  suppressor 620 

P 

Parallel  operation,  generators 318 

transformers 421 

Penstocks.    See  Pipe  lines 114 

Perimeter,  wetted 107 

^ezometers 793 

Piping,  lubricating  oil 339,343 

transformer  oil 466 

transformer  cooling  water 457 

Pipe  lines 114 

anchors 128 

concrete  pipe 137 

economic  diameter 121 

expansion  joints 128 

friction  loss 116 

gradient 119 

loss  of  head 116 

number 120 

pressure 128,  246,  250 


Digitized  by 


Google 


INDEX  813 

PAOa 

Pipe  lines,  sue 120 

steel  pipe 126 

thickness  of  plates 125, 126 

wooden-stave  pipe 131 

Pitot  tube 797 

Polarity  of  transformers 421 

Pondage 159 

Population  of  United  States 14 

Potential  transformers 646,  666,  638 

Power,  cost 742 

development 654,  675,  684 

market 666,  677 

primary  and  secondary 683 

Power  factor,  effect  on  generator  operation 284 

effect  of  reactance 462 

Power  factor  indicator 542 

Power-house 166 

arrangemet  of  apparatus T 176 

basement 166 

cranes : 171 

doors 171 

floors 170 

foundations 168 

general  design 166 

heating 176 

illumination 174 

roof 170 

typical  layouts 180 

ventilation 173 

windows 171 

Power  systems,  load  factor '. 682 

peak  loads 682 

yearly  output 682 

Power  transmission,  development 4,  10 

Precipitation.    See  Rainfall 44 

Pressing,  atmospheric 42 

pipe  line 122,  126,  138 

regulation 246 

regulators.    See  Relief  valves 268 

Primary  power 683 

Primary  power  in  United  States 24 

Q 

Quantity  of  flowing  water 67, 795 

R 

Racks 100 

Rack  cleaners 101,  147 


Digitized  by 


Google 


814  INDEX 

Radius,  hydrauHc 106, 117 

Railroad  electrification 38 

Rainf  aU 44,  650 

disposal 47 

records 46 

variations 44 

Rating,  current-limiting  reactors 459 

exciters 351 

generators,  synchronous 307 

oil  circuit  breakers 497 

transformers 383 

Ratio,  transformers 375, 424 

Reactance,  armature 286 

bus-bars 569 

cables 642 

effect  on  power  factor 462 

effect  on  regulation 463 

generator,  synchronous 292,  294,  469 

transformers 377, 469 

transmission  lines 469 

Reaction,  armature 285 

Reactive  volt-ampere  indicator 543 

Reactors,  current-limiting 458 

arrangement  in  power-house 178 

bus-bar 466,  474 

calculation  of  three-phase  short-circuit  currents 460,  468 

calculation  of  single-phase  short-circuit  currents 480 

effect  of  reactance  on  regulation 463 

effect  of  reactance  on  power  factor 462 

feeder 466,  476 

generator 465 

inductance '. 467 

location 465 

losses 467 

mechanical  design 485 

number 468 

purpose 458 

rating 459 

rating  as  affected  by  current 461 

rating  as  affected  by  frequency 461 

rating  as  affected  by  voltage 461 

reactive  drop 459 

size 468 

Stott-system 467 

temperature  rise 459 

Receptacles,  ammeter  transfer 560 

voltmeter  and  synchronizing 559 


Digitized  by 


Google 


INDEX  815 


Register,  water  stage 265 

Regulation,  speed  of  turbines 122,  140 

Regulation  of  stream  flow 689 

Regulation,  voltage 363 

effect  of  reactance 463 

generator,  83mchronous 291 

hand  regulation 363 

K.  R.  system 369 

line-drop  compensation 368 

synchronous  condenser 371 

T.A.  regulator 336 

transformer «* 378 

Regulators,  T.  A 363 

Relays 507 

balanced 521 

control 526 

differential - 522 

high-tension,  series 525 

high-voltage,  high-current 371 

interconnected  reverse  power 518 

low  voltage r . . .  626 

overload 509 

over-voltage 525 

pilot  wire 523 

reverse-power 514 

selection 492 

signal 526 

split-conductor 521 

time  settings 494 

trip-free*. 526 

underload 526 

Relief  valves 141,  221,  251 

Reluctance,  air 289 

iron  and  steel 289 

Reports,  preparation 645 

water  power 645 

Reservoirs.    See  Storage  reservoirs 159 

Resonance 595 

Rheostats.    See  Field  rheostats 556 

Run-off 51 

mean  annual 54 

records 649 

Rupturing  capacity  of  oil  circuit  breakers 470 

S 

Saturation  curves 289 

Secondary  power 683 


Digitized  by 


Google 


816  INDEX 

PAOp 

Seepage 164 

canals 110 

Shipping  limitations 193 

Short-circuit  currents 458 

calculation  of  three-phase 460,  468 

calculation  of  single-phase 480 

mechanical  stresses 471 

synchronous  generators 292 

Shori-drcuit  suppressors 620 

Signal  systems 583 

Skin  effect 569 

Slope,  hydraulic «. 106,  117,  119 

Spacing  of  conductors 627 

Specific  speed  of  turbines 206,  209 

Speed,  exciters 353 

generators , 315 

turbines 214 

Speed  regulation 122, 246 

turbines 220 

Spillways 83 

Stand  pipes.    See  Surge  tanks 141 

Starting  up  of  station 198 

Station  wiring.    See  Wiring 625 

Steam  auxiliary  stations 692 

base-load  stations 697 

cost 745 

low-water  stations 695 

peak-load  stations 697 

prime  movers 693 

stand-by  stations 694 

Steam  power,  cost 745 

relation  to  water  and  gas  power 27 

Steel  pipe 125 

Storage  batteries,  excitation 361 

oil  circuit  breaker 589 

Storage  reservoirs 59,  159 

intakes 163 

limitations 160 

location 160 

prevention  of  floods 22 

regulating  effect 59 

seepage  and  evaporation 164 

storage  and  pondage 159 

Storage  of  water 650,  685 

Stream  flow 53 

definition  of  terms 53 

duration  curves 676 


Digitized  by 


Google 


INDEX  817 

PAOB 

Stream  flow,  economical  development 645,  684 

energy  available 676 

factors  affecting  stream  flow 56 

mass  curves , 685 

measurements 58,  65,  795 

records 66 

regulation 689 

smxmiation  curves 685 

variations 53 

Supports,  bus-bar 569 

Surge  tanks 141,  251 

Switchboards 529 

panel  type 531 

bench  boards 535 

Switches,  control 561 

disconnecting 577 

field  discharge 554 

throw-over 561 

Switching  equipment 485 

ammeter  transfer  plugs  and  receptacles 560 

arrangement  in  power-house 179 

bus-bars 565 

bus  and  switch  structures 563 

calibrating  terminals 561 

control  switches 561 

current  and  potential  transformers 546, 565 

disconnecting  switches 577 

entrance  bushings 571 

exciter  and  field  control 552 

field  discharge  switches 554 

field  rheostats 556 

instrument  equipment 538 

mimic  bus-bars 563 

multi  recorder 589 

oil  circuit  breakers 470, 496 

oil  circuit-breaker  batteries 589 

outdoor  arrangement 577 

relays 492,  494,  507 

relay  protection 486 

signal  systems 583 

switchboards 529 

system  of  connections 486 

voltmeter  and  s3mchronizing  plugs  and  receptacles 559 

Switching  high-tension  circuits 509,  748 

S3mchronous  condenser  regulation 371 

S3mchronous  generators.    See  Generators 280 

Synchronous  impedance 292 


Digitized  by 


Google 


818  INDEX 

PAOB 

Synchronous  reactance 294 

S3mchronifian  indicator 544 

Synchronizing  plugs  and  receptacles 559 

T 

Taps,  transformers 387, 441 

Telephone  protection '. 620 

Temperature,  indicators 545 

measurements 310 

Temperature  rise,  permissible 309 

bus-bars 566 

current  limiting  reactors 459 

exciters 351 

generators 308 

transformers .' •  383,  453,  455 

Testing  code  for  turbines 788 

Thermometers  for  transformers 445 

Three-wire  system,  Edison 392,  401,  402,  408,  418,  419 

Throw-over  switches 561 

Thunderstorm  records 617 

Transformation,  phase 409 

two-  or  three-phase  to  single-phase :,: 409 

two-phase  to  six-phase 410 

three-phase  to  two-phase :•. : 412 

three-phase  to  three-phase,  two-phase 415 

'  three-phase  to  six-phase 417 

Transformation,  voltage 391 

single-phase 391 

two-phase 393 

three-phase,  delta-delta 397 

three-phase,  delta-Y  and  vice  ver^. 398 

three-phase,  Y-Y 403 

three-phase,  open  delta 404 

three-phase,  T 407 

Transformers 373 

arrangement  in  power-house 176 

breathers 432 

bushings 442 

connections 391 

cooling  coils ; . . .• 433 

coo^g  water 381,  457 

cooling  water-indicators 435 

cores 437 

core  type 379,  439 

corrosion  of  cooling  coils 434 

current  and  potential 548 

drying 445 


Digitized  by 


Google 


INDEX  819 

TraDsformen,  efficiency . : >....'..' 384 

frequency 275 

fundamental  principles 373 

grounding  of  neutrai .  .392,  303,  396,  398,  399,  401,  404,  408,  413,  420 

grounding  of  secondaries 393 

grouping 489,  492 

impedance 426 

induced  E.M.F 375 

magnetizing  current 376 

mechanical  design 429 

method  oi  cooling 379 

number  and  sixe 390 

oil 443,  456 

oil  drying ; . .  449 

oil  testing 451 

operatKHi 462 

parallel  operation 421 

rating.. ;: 883 

ratio 875 

reactance 377,  469 

regulation,  voltage 378 

shell-type 379, 437 

shipping 445 

single  and  polyphase 381 

tanks .- 429 

taps 887,441 

temperature  measurements 383 

temperature  rise 383,  453,  455 

voltage 385 

voltage  regulation 378 

windings 438 

Transmission,  developments 4, 10 

principal  data  of  systems 783 

reactance 469 

voltage 8,  271 

Transportation  of  apparatus 193 

Traveling  waves 596 

Tunnels 113 

Turbines 202 

arrangement  in  power-house 175 

bearings 241,  330 

brakes 348 

buckets 242 

casings 237 

characteristic  curves 214 

corrosion 233 

draft  tubes 239 


Digitized  by 


Google 


820  INDEX 

PAOB 

Turbinea,  efficiency 211,  789 

erosion 233 

flywheel  effect 327 

gate  mechanism 235 

history  of  developments 1,9 

horizontal 227,  242 

housing 246 

impulse 204 

lubrication 338 

mechanical  design 226, 242 

noizles 241, 246 

number  of  units  and  capacity 204 

over-speed 226 

reaction 202 

regulation 140,  220 

runners 231,  242,  244 

selection  of  type 204 

speed,  actual 214 

speed  regulation 140, 220 

speed  rings ^ 237 

speed,  specific 206 

speed  variations , 202, 204 

test 217,  788 

vertical 229,  242 

wicket  gates 236 

windage  and  friction '. 791 


U 
Unloading  of  apparatus 194 


V 

Valves.    See  also  Gates. . 144 

air 128,157 

gate 148 

Johnson  hydraulic 156 

operation  and  control 149 

pivot 155 

reUef 141,  221,  251,  258 

Velocity  of  water 106,  107,  116,  798 

canals 106 

flumes 110 

pipe  lines 114,  246 

tunnels 113 

Ventilation,  generators 173,  345 

power-house 173 


Digitized  by 


Google 


INDEX  821 

Venturi  meten 262 

registers 264 

manometers 265 

Voltage 270 

distribution 273 

exciter 351 

generator 270, 316 

induced 280, 375 

transformers 386 

transmission 4, 271 

Voltage  drop  in  conductors 639 

Voltage  regulation 363 

generators 291 

transformers 378 

Voltage  rise.    See  Over-voltage 525,  593 

Voltage  test,  generators 326,  331 

oil  circuit  breakers 497 

transformers 386 

Voltmeters 540,  541 

Voltmeter  plugs  and  receptacles 559 

W 

Water 39 

critical  temperature 40 

effect  of  atmospheric  pressure 42 

energy  of  flowing  water 67 

latent  heat 41 

measurements 43 

properties .  .  .*. 39 

quantity  of  flowing  water 67,  795 

safe  velocities 107 

specific  gravity 39 

specific  heat 42 

velocity 66,  106,  110,  113,  114,  116,  121,  798 

weight 39 

Water  conductors 104 

canals 106 

classification 104 

flumes 110 

pipe  lines 114 

tunnels 113 

Water  hammer 138 

Water  flow  indicators,  transformers 435 

Water  power,  history  of  developments 1 

Water  power  in  United  States 16,  17 

Water  power  in  the  world 13 

Water  power  from  inland  waterways 22 


Digitized  by 


Google 


822  INDEX 

Water  power,  relation  to  steam  and  gas 27 

Water  power  reports 645 

Water  stage  registers 265 

Water  storage.    See  Storage  of  water 686 

Water  supply,  source 44 

Water  supply  systems,  power  from 23 

Waterways,  power  from 22 

Water  wheels.    See  Turbines 202 

Watthour  meters 549 

Wattmeters,  indicating 542 

Waves,  form 282,  301 

traveling 696 

Weir 58 

Wetted  perimeter 107,  795 

William  and  Hazen  formula 116 

Windings,  generator 282,  285,  295,  325 

transformer 391,  438 

Wiring,  station 626 

cables  in  duct  or  conduit 628 

control  and  instrument  wiring 629,  633 

corona  limit  of  voltage 639 

•    economical  considerations 639 

exciter  and  field  wiring 631 

general  practice 631 

generator  and  transformer 631 

high-tension 633 

insulation 625 

open  wiring 626 

resistance  and  reactance  of  cables : 642 

single  V8.  multiple  conductors 629,  ^ 

spacing  of  conductors 627 

voltage  drop 639 

Wooden-stave  pipe . . , 131 


Digitized  by 


Google 


Digitized  by 


Google 


Digitized  by 


Google 


Digitized  by 


Google 


THIS  BOOK  IS  DUE  ON  THE  IiABT  DATE 
STAJCPED  BBI«OW 

AN  INITIAL  FINE  OP  26  CENTS 

WILL   BE  A«9E5S£D    FOR    FAILURE  TO    RETURN 
THIS   eOOK  ON   TH€   DATE   DU£.    THE  WEHAVTT 
WILL  INCREASe  TO  SO  CENTS  ON  THE  FOURTH 
DAY     AND    TO    «t.OO     ON     THE    SEVENTH     DAY 
OVERDUE. 

f^AR    10  19; 

!fl 

,^.  Nr 

rrfic     ic;sjJ!U 

rC©  * 

lap  f  l^fifM-T/87 

Yu   19799 


yy^f'^U 


"V. 


VT"    f  M  ■■»        J      9tUM      BWI'JIIHI    ■  L     P"    Jifi    LIJ  J 


THE  UNIVERSITY  OP  CALIFORNIA  LIBRARY       i 


Digitized  by 


Google 


4