(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Biodiversity Heritage Library | Children's Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Transactions of the American Society of Mechanical Engineers"

Google 



This is a digital copy of a book that was preserved for generations on Hbrary 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 maiginalia 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 liave 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 fivm 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 attributionTht GoogXt "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 liabili^ 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/| 



I 

Tf 

y.HI 






 u*v*i.v-. 



.> 



1 



"mE AMERICAN SOCIETY OP 
MECHANICAL ENGINEERS 



TRANSACTIONS 



VOLUME 41 

DETROIT MEETING 

NEW YORK MEETING 

1919 



NEW YORK 

PUBUSHED BY THE SOCIETY 

c» Wsji 39th Sthbet 

1920 



Copyright, 1920 by 
THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 



CONTENTS OF VOLUME 41' 

Detroit and New York Meetings 

No. Paqb 

1686 Biography of Mortimer Elwyn Cooley. Annual Report of the 

Council 1 

1687 Meetings, January- June 1919 19 

1688 Arthur M. Greene, Jr., The Present Condition of Research in 

the United States 31 

1689 A. D. Little and H. E. Howe, The Organization and Conduct of 

an Industrial Research Laboratory 69 

1690 C. E. K. Mees, Industrial Research Laboratory Organization 83 

1691 Enrique Touceda, Research Work on Malleable Iron 91 

1692 Arthur H. Young, Industrial Personnel Relations 145 

1693 L. P. Alpord, The Status of Industrial Relations 163 

1694 J. H. Walker, Central-Station Heating in Detroit 209 

1695 W. F. Verner, The Production of Liberty Motor Parts at the Ford 

Plant 239 

1696 Charles H. Fox, Fire Engines and the Essentials of Fire Fighting . . 263 

1697 Jacob M. Spitzglass, An Electrical Device for Measuring the Flow 

of Fluids in Pipes 277 

1698 J. W. Morton, Crude-Oil Motors vs. Steam Engines in Marine 

Practice 299 

1699 D. L. Arnou), A Suggested Formula for Rating Kerosene Engines. 323 

1700 O. C. Berrt, Standards of Carburetor Performance 333 

1701 N. C. Harrison, Pulverized Coal as Fuel 355 

1702 Frederick A. Scheffler and H. G. Barnhurst, Pulverized Coal 

for Stationary Boilers 377 

1703 C. R. Weymouth, Economy of Certain Arizona Steam-Electric 

Power Plants Using Oil Fuel 437 

1704 Wai^tbr C. Durfeb, Elements of a General Theory of Airplane- 

Wing Design 465 

1705 G. Francis Gray, John W. Reed and P. N. Elderkin, Air Fans 

for Driving Electric Generators on Airplanes 481 

1706 J. F. RoBBiNS, Mechanical Lifts, Past and Present, and a New 

Method for Their Balancing 499 

1707 Alfhonse a. Adler, The Design of Riveted Butt Joints 533 

1708 Cary T. Hutchinson, Economical Section of Water Conduit for 

Power Development 555 

1709 Meetings, Septembei^December 1919 565 

1710 Mortimer E. Cooley, Presidential Address 577 

^ The Society shall not be responsible for statements or opinions adyanced in papers or in dis- 
atitameetincB (C65). 

V 



362523 



No. Paqb 

1711 George H. Perkins, Steam Use in Textile Processes 593 

1712 T. S. Taylor, The Thennal Conductivity of Insulating and Other 

Materials 605 

1713 F. W. Dean and Henrt KreisingeA, Emergency Fleet Corporation 

Water-Tube Boiler for Wood Ships 623 

1714 WnxiAM L. De Baufre and Milton C. Stuart, Flow of Water 

through Condenser Tubes 655 

1715 H. A. S. Howarth', Slow-Speed and Other Tests of Kingsbury 

Thrust Bearings 685 

1716 Walter J. White, A Dredging Pump of Novel Construction 709 

1717 Allen H. Blaisdell, Turbo-Compressor Calculations 731 

1718 Laurence F. Seaton, Kerosene as a Fuel for High-Speed Engines . . 761 

1719 Chester B. Ix)RD, A Perfected High-Pressure Rotary Compressor. 773 

1720 Christopher H. Bierbaum, Common Errors in Designing and 

Machining Bearings 791 

1721a Frederick P. Fish, The Causes of Industrial Unrest and the Remedy 807 

1721& William L. Leiserson, Systems for Mutual Control of Industry.. 809 
1721c Ralph E. Heilman, What May We Expect of Profit Sharing in 

Industry?. . . ; 811 

17214 A. L. De Leeuw, Wage Payment 813 

1722 Col. E. A. Deeds, The Future of Aviation 827 

1723 Forrest Naoler, A New Type of Hydraulic-Turbine Runner 829 

1724 E. B. Biakeley, The Hvid Engine and Its Relation to the Fuel 

Problem 855 

1725 Leon Cammkn, Combustion of Heavier Fuels in Constant-Vohime 

Engines and in Engines of the Super-Inductive Type 875 

1726 Charles dk Freminville, Reliability of Materials and the Mecha- 

nism of Fmcturos 907 

1727 John Yoi'NCiER, Motor-TransiK)rt Vehicles for the United States 

Army 925 

1728 Alfkfj) Mrsrto, An Investigation of Strains in the Rolling of Metal. 961 

1729 W. E. M(K)ke, Motlem Electric Furnace Practice as Related to 

Foundritns in Particular 971 

1730 B. F. Waterman, Thread Forms for Wonns and Holxs 972 

1731 John E. Muhij^ij>, S<*ientific IX^velopment of the Steam 

liocoinotive 9tW 

1732 S. A. Sdlkntic, Oil riiKj Linos 1059 

173ii Manual on American Standard Viytc Threafls 10t>7 

1734 Reiwrt of the Joint ConfenMice Commit t<H» . . 1095 

1735 Necrouhiy 1 105 

173G Index 1 159 



VI 






THE AMERICAN SOCIETY OF 
MECHANICAL ENGINEERS 



To THB MCMBKRS: 

THE acoompanjring forty-first volume of Transactions 
records the activities of the Society darinc^ the past 
year and inclodes a number of papers on industrial problems 
which have been brought to the fore by the unsettled 
conditions following the great war. 

The volume comprises forty-six papers, addresses and 
discussions on engineering, industrial and economic subjects, 
three reports to the Society — one being the report of the 
Council, another the Manual on American Standard Pipe 
Threads and the third the report of the Joint Conference 
Committee of the four Founder Societies. 

One group of papers deals with the subject of research 
in private, industrial and educational institutions, while 
another is concerned with personnel relations. Two papers, 
with a voluminous discussion, consider the burning of pow- 
dered-fuel and give the results of tests of powdered-fuel 
plants. Other papers deal with power plants, aeronautics, 
measuring devices, the mechanism of fractures, intemal- 
oombustion engines, machine shop practice, drainage 
pumps and locomotives. 

The preparaticm of the volume has been carefully 
supervised by the Publication Committee and special pains 
have been taken in the selection, arrangement and index- 
ing of the material to make it of the greatest possible 
reference value. 

Calvin W. Ricb, Secretary 

99 West 89th Street 

New York 
1920. 



1711 GCMKB ff. 1 






my of hmdnunm^nd 







k 



OFFICERS 
THE AMERICAN SOCIETY OF MECHAN- 
ICAL ENGINEERS 

FORMING THE STATUTORY COUNCIL 

1919 



PRESIDENT 
AoBTDfEB E. CooLET Ann Arbor, Mich. 

VICE-PRESIDENTS 

I 

Tenng e:qnre December 1919 

^OHN Hunter New York, N. Y. 

-Spencer Miller New York, N. Y. 

||Max Toltz St. Paul, Minn. 

Terma expire December 1920 

Fred R. Low New York, N. Y. 

Henry B. Sargent New Haven, Conn. 

John A. Stevens Lowell, Mass. 

MANAGERS 

Terms expire December 1919 

Robert H. Fernald Philadelphia, Pa. 

William B. Gregory New Orleans, La. 

C. R. Weymouth San Francisco, Cal. 

Terma expire December 1920 

Fred N. Bushnell Boston, Mass. 

Fred A. Geier Cincinnati, Ohio 

D. Robert Yarn all Philadelphia, Pa. 

Terms expire December 1921 

Charles L. Newcomb Holyoke, Mass. 

Charles Russ Richards Urbana, 111. 

Frank O. Wells -. Greenfield, Mass. 

PAST-PRESIDENTS 

Members of the Council for 1919 

James Hartness Springfield, Vt. 

John A. Brashear Pittsburgh, Pa. 

D. S. Jacobus New York, N. Y. 

Ira N. Hollis , Worcester, Mass. 

Charles T. Main Boston, Mass. 

CHAIRMAN OF FINANCE COMMITTEE 

W. E. Symons New York, N. Y. 

TREASURER 

William H. Wiley New York, N. Y. 

vu 



SECRETARY 
Calvin W. Ricb New York, N. Y- 

EXECUTIVE COMMITTEE OF THE COUNCIL 

Mortimer E. Cooley, Chairman D. S. Jacobus 

Ira N. Hollis Charles T. Main 

John Hunter Henry B. Sargent 

STANDING COMMITTEES OF ADMINISTRATION 

FINANCE 

Supervises all financial affairs^ makes all contracts^ orders all expendittares, 

administers the general offices 

W. E. Symons (3), Chairman and Representative on Council 

Alfred E. Forstall (1) F. E. Law (4) 

George M. Forrest (2) Alex Dow (5) 

MEETINGS AND PROGRAM^ 

Conducts Annual and Semi-Annual Meetings j arranging professional and 

entertainment programs 

D. S. Kimball (2) , Chairman and Representative on Council 

Walter Rautenstrauch (1) W. G. Starkweather (4) 

A. L. De Lbeuw (3) R. V. Wright (5) 

Assisted by sub-committees which examine papers for special semans 

PUBLICATION AND PAPERS 

Supervises publication of The Journal (monthly) ^ Transactions (annually), 
Year Book (annually). Condensed Catalogues (annually), 

and other publications 

George A. Orrok (5), Chairman and Representative on Council 
Joseph W. Roe (1) George J. Foran (3) 

H. H. Esselstyn (2) Ralph E. Flanders (4) 

MEMBERSHIP 

Receives and scrutinizes all applications for admission or transfer, and 

reports names to Council 

S. D. CoiiLETT (1), Chairman and Representative on Council 

Frederick A. Waldron (2) W. S. Timmis (4) 

R. F. Jacobus (3) Hosea Webster (5) 

CONSTITUTION AND BY-LAWS 

Reports on all matters relating to Constitution, By-Laws and Rules referred 

to it by Council 

Jesse M. Smith (3), Chairman and Representatit^ on Council 

George M. Basford (1) E. S. Carman (4) 

Ira H. Woolson (2) James E. Saque (5) 

1 Sub-Committees are published in 1919 Year Book. 

Non: — Nomben in parentheaes indicate the number of years the member has yet to serve. 

viii 



LOCAL SECTIONS 

SypenfueB eondud of affairs of aU Local Sections throughout the country 

D. BoBSBT Yabnall (1), Chairman and Representative on Council 

W. H. Kknbbson (2) LbuiB C. Mabburq (4) 

C. RxTBB RicBABDB (3) S. B. Elt (5) 

STANDING COMMITTEES 

HOUSE 

Mainienanee of rooms of Society, furnishings, paintings, memorabilia, etc. 

Obbdb p. CmooNGS (1), Chairman J. D. Maouirs (4) 

Maxwkll p. Upson (2) F. A. Muschenheim (5) 

H. O. Pond (3) 

UBRARY 

Represents Society on Library Board of United Engineering Society 

Waijtkr M. McFaeland (1) Jesse M. Smith (4) 

A. M. Hinrr (2) The Secretabt 

W. N. Best (3) 

PUBLIC RELATIONS 

Considers matters referred by Council bearing on relations of engineer 

to community 
F. H. Newell (1) James £. Sague (3) 

Fred R. Low (2) J. Waldo Smith (4) 

RESEARCH » 

Appoinled 1909, for supervising aU research actUniies, collaborating with 
committees cf kindred societies, and obtaining result^ of researches 

conducted in other countries 



Arthur M. Greene, Jr. (1), Chairman 
Carl C. Thomas (2) 
Albert Kingsbury (3) 



Walter Rautenstrauch (4) 
Reginald J. S. Pigott (5) 



STANDARDIZATION COMMITTEE 

Standardizes method of making and arriving at standards ihemsdoes 
Endeavors to bring about a unification of the standards 

of the Society 
Hehrt Hess (3), Chairman Walter S. Twining (5) 

Lionel 8. Marks (1) W^illiam F. Kiesel (2) 

Clarence F. Hirshfeld (4) 

SOCIETY REPRESENTATIVES* 

Appointees by the Council on committees, etc., of other bodies 

> 8qb<kiminittew are published in 1919 Year Book. 

' Ftetbl Lilt of Society RepicaenUtives. Complete list published in 1019 Year Book; also 

eonmittees and Local SeetioiM* Committees. 
Non:— Nunben in pamtheses indirate the number of years the member has yet to aerve 

ix 



TRANSACTIONS 

OF 

THE AMERICAN SOCIETY OF 
MECHANICAL ENGINEERS 

Volume 41 — 1919 

nPHIS volume contains an account of the activities of the Society 
for 1919, and in it will be found papers and addresses given at 
the Spring Meeting, held in Detroit, and the Annual Meeting, held 
in New York, with the discussions thereon. 

In selecting the material published, the intention has been to 
include all papers having permanent reference value. For this reason 
condensed accounts are given of much discussion on papers covering 
the less technical subjects and reference is made to issues of 
Mechanical Engineering in which the subject matter may be 
found in a more complete form. 

MORTIMER ELWYN COOLEY 

Mortimer Cooley, born March 28, 1855, was the fifth member 
of a family of eight children, all reared on a farm in Canandaigua 
Township, Ontario County, N.Y. He is of the ninth generation of 
Coole)r8 in this coimtry. Benjamin, the first of the name, came from 
England and settled in Springfield, Mass., in 1642, where for many 
years he was a selectman. Benjamin was an ensign in the Hamp- 
shire Regiment commanded by Major John Pynchon in the King 
Philip War. He was a weaver by trade and hved not far from Mill 
River. The Barney and Berry skate factory, overlooking the Con- 
necticut River, is located on the rear end of the old Cooley home- 
stead. Later, a dozen or more Cooleys owned homes in Longmeadow, 
four or five miles south of Springfield on the other side of Mill River. 
As the tribe multipUed some of them moved to Granville, Mass., 
a beautiful and picturesque locality at the south end of the Green 

1 



of naming Her wttb appropriate naval ceremonies. tSbe was duly 
christened Alliance in the ship's honor and if the tales of rear ad- 
mirals of today who were cadets in those daj'S are to be believed the 
occasion was a notable one in the annals of the Navy. 

Cadet Engineer Cooley was forthwith detached and ordered 
home for a few weeks, then to the Bureau of Steam Engineering 
at the Navy Department, In June 1881 he was examined and 
promoted to Assistant Engineer and in August was ordered to the 
University of Michigan to teach steam engineering and iron ship- 
buOding. At the end of three years on request of the Regents his 
detul was continued a fourth year. Being then detached and 
ordered to the Pacific Station, the Regents conferred on Assistant 
Engineer Cooley the honorary degree of Mechanical Engineer and 
invited him to resign and accept the chair of Mechanical Ekigineer- 
ing. This he did, his resignation taking effect December 31, 1885. 
It was with a great deal of regret that he resigned as he was in love 
with the Service- There was at the time no prospect for any great 
increase in the naval force, and it seemed to him the opportunity 
for real work afforded him at the University ought not to be declined. 

Professor Cooley has given his entire life since he was twenty- 
six years of age to university work — thirty-eight years up to now. 
He has been Dean for fifteen years, having been appointed in Febru- 
ary 1904. The Michigan Agricultural College conferred on him the 
degree of LL.D- in 1907, and the University of Nebraska the degree 
of Eng.D. in 1911. When he came to the University there were 
but sixty or seventy engineering students out of a total of about 
1300 in the University, and the entire technical work in engineering 
was done in seven rooms at the south end of the main university 
building. The first engineering laboratory was built the winter 
after he came. It was a two-story brick veneer building 24 x 36 ft. 
costing $1500 and the equipment $1000. In it Professor Cooley 
himself taught forging, pattern making, and machine shop practice. 
It was styled by his colleagues "the scientific blacksmith shop." 
It was the beginning of an effort, now alt<^ether general, to give to 
ei^ineering students while in collie some practical knowledge of 
the materials and processes used in the execution of engineering 
projects.  

But Professor Cooley could not wean himself altogether from 
the naval service. He was from 1895 to 1911 the Chief Engineer 
officer of the Michigan State Naval Brigade and is now a retired 



4 SOCIETY AFFAIRS 

officer in the Brigade. In 1898 he returned to the Navy as Chief 
Engineer during the Spanish War. He was attached to the U.S.S. 
Yosemite and later to the League Island Navy Yard, his period of 
service being altogether about ten months. His honorable dis- 
charge was handed him by the commandant of the navy yard with 
words of commendation for his efficient work. 

While on blockade duty off San Juan, P.R., the Yosemiie engaged 
in a five-hour battle with tlie Spanish forts, gunboats and torpedo 
boats following the hiterception of the Antonio Lopez, a Spanish 
cruiser loaded with munitions, putting into the harbor. During 
the blockade a serious fire broke out in the coal bunkers of the Yose- 
mite which for a time threatened serious consequences. The fire 
was deep down and could not be reached. Chief Engineer Cooley, 
recalling the method of sinking piles on western rivers by means of 
a water pipe attached to the pile, had a hose and nozzle triced to a 
long slice bar with which, under fire pressure from the pumps, the 
fires were successfully quenched. The slice bar could be shoved 
down into the coal like a knife into soft butter. 

Following his return to the University in 1899 Professor Cooley 
was invited by the Citizens' Committee of Detroit, of which Gover- 
nor Pingree was Chairman, to appraise the power plants, rolling 
stock and stores and supplies of the Detroit street railways, which 
the city was contemplating purchasing. It was a hurry job and was 
done in a hurry. The ap|X)intment was made on Friday, the staff 
was organized Saturday and the report submitted the following 
Saturday covering $2,000,000 of proixuly. The following year, 
1900, at the request of Ciovenior Pingree, the Board of State Tax 
Commissioners, and the Board of State Auditors, Professor Cooley 
undertook to appraise the specific tax-paj-ing projxirties of the State 
of Michigan which included the steam railroads, the telegraphs, the 
telephones, the plank roads and the river improvements. ThLs was 
late in August. The field work wits coinplettMl in ninety days and 
the results submitted at the end of Decemlier in time for the incoming 
legislature. The work involved the insiHTtion of 1(),(XX} miles of 
track, thirty-odd thousands of tl»e freight cai*s, all the passenger 
and special equipments, all the locomotives, telegrai)h, and tele- 
phone lines, in short everything involved in the different kinds of 
properties. Some 150 men \\pre employed. The total of the ap- 
praisal was about $240,000,000 and the cost less than 875,000. 

As a result of this work the legislature enacted laws placing the 
railroads on an ad valorem tax basis which increased their taxes 



SOCIETY AFFAIRS 5 

threefold and more. When the assessment was made mider the new 
law in 1903 the railroads brought suit to enjoin their collection of 
taxes. This made necessary another appraisal as of the date of the 
assessment in which the value found for the railroads was $240,000^000 
an increase of $40,000,000 due largely to using 1903 prices for labor 
and materials instead of the average from 1890 to 1900. The case 
was carried to the U.S. Supreme Court and being finally decided in 
favor of the state, brought into the state treasury twelve or fifteen 
millions in back taxes. 

Michigan's pioneer work in valuation of large pubUc utiUty prop- 
erties was soon followed by other states. First among them was 
Wisconsin in a valuation of her steam railroads. Substantially the 
same methods were employed as in Michigan. Professor Cooley 
was consulting engineer. 

In the twenty years which have elapsed since that first appraisal 
in Detroit, Professor Cooley has had chai^ of many hundreds of 
appraisals in various states and municipaUties, in most of them em- 
ployed by the pubUc. In all of them he has stood consistently for 
correct results regardless of employer, "hewing to the line letting the 
chips fall where they may." In the aggregate the value of property 
appraised under his direction Ues somewhere between one and one- 
quarter and one and one-half billion dollars. 

Nor has Professor Cooley neglected opportunities to serve in 
other capacities. He waa^for a time chairman of the Board of Fire 
Commissioners, and President of the Common Council in Ann Arbor 
in 1890-1891. He served on the Board of Awards for the World's 
Fair in Chicago in 1893, and for the Pan-American Ekposition in 
Buffalo. He has for twenty-five years served as mechanical expert 
in patent causes, and testified many times on mechanical matters 
before juries and commissions. He was for five years (1907-12) 
chairman of the Block Signal and Train Control Board of the In- 
terstate Commerce Commission. 

Professor Cooley is a Fellow of the American Association for the 
Advancement of Science, member, since 1884, of The American 
Society of Mechanical Ekigineers, American Society of Civil Elngineers, 
American Institute of Consulting Engineers, Franklin Institute, 
Society for the Promotion of Engineering Education, Society of 
Naval Engineers, Michigan Engineering Society, Detroit Engineer- 
ing Society, Sigma Phi, Tau Beta Pi, Sigma Xi, the Army and Navy 
Clubs in Washington and in New York, the Detroit Club and the 
Yondotega Club in Detroit. 



6 SOCIETY AFFAIRS 

Professor and Mrs. Cooley have four children, three daughters 
and one son. All are married and seven grand children now keep 
them from growing too old. The son is a Commander in the United 
States Navy. 

Thus has Mortimer Cooley established his record of imusual 
accomplishment and honest devotion to his democratic ideals. And 
he is still "doing the chores." 



. y 



ANNUAL REPORT OF THE COUNCIL 

In his annual address^ President Cooley has reviewed the or- 
ganization of the Council and Committees and much of the work 
which has passed through these channels during the year 1919. 
This annual report of the Coimcil therefore supplements the Presi- 
dent's address by giving the more statistical record of actions taken 
by the Council as the body administrating the Society's affairs. 

COMMITTEES 

Under a revision of the Constitution C45, which was put into 
effect this year, changes in the grouping of the standing and special 
conmiittees of the Society have been made and further changes are 
pending, all leading towards an ultimate grouping of all conmiittees 
into administrative, professional and non-professional, with each 
group having two classes, standing and special. 

The work of adjusting the membership of the committees dur- 
ing the year, to have as equable as possible distribution of their 
membership among qualified men all over the country, has been a 
matter of serious consideration by the President and has finally led 
to the appointment of a special conunittee of the Council, the Com- 
mittee on Conmiittees, to report on all committee activities. Con- 
tinuous membership on certain types of committees will be dis- 
couraged in the future, but a suflBcient number of experienced and 
active members of such committees will be continued to secure con- 
tinuity of poUcy. It is beUeved that the adoption of these poUcies 
will greatly stimulate the committee work. 

COMMriTEES — NUMBER AND SUBJECTS 

A review of the pages at the beginning of the volume will show 
the large number of committees having in charge the many activi- 
ties of the Society. 

» Transactions, 1919, No. 1710. 



s 



sc^rnrrr 






Thc:5^^ i\>:r.:v.:::tv:> hAvc r^jki-^ irji'^ rrw>n5, as required by 
the I Constitution, whiv^h ir. .vnir^je:!:- fcrzi airte on file in the records 
of I ho SivioTv anvi Arv suiv-r.u^r.Kvi ir. the following: 

F*>j«j*uY. Tiio \\\sr 19 li* bj%5 "r«!:x:- a nanioulariv difficult vear 

« m. m m 

in tlio ttnanoini: v^f :ho S\v;o:y*< .vvl'iitie?, 'i;ie to increased costs 
of all work, rb.o o\;xr.d::ures o: :ht^ S:*:ie:y per member for the 
fiscal yoar vVtolvr 1. l\MS :o vV.v^lxr 1. lvI9 weite $32.43, as against 
$'jr>.tM tho pi\\\\ii:ig \TAr. Pnose exr»eni::ures were covered by 
ounvnt ii\i\Mno. 

hiVi^iUxi F:,',.is .:':.i T^.^i F:r\.:^. Tie pn?sent standing of the 
invostoil fvinvls of ti\o Svv:o:y i< showr. in tho following table: 

ivv; vN*. i: <v.yur 



Sooit»t\*i« one «)iinrtor »ntcrt>t in rtuwco-.-c >vv.»:>** l^,.. 



^ >lk- aA^^A. 



>486,792.70 



l.ibrnry l^ooks 

Furnituro HUii Fi\t»»M\-« 



5. -XXI 00 



18.000.00 



Stores, inchulin^; phitra ruul tini>lu>l puMio.Hti. :;.* 
KngiiurrinK Indi-x 



Trust Fund Inve.ittnont: 

New York City HJ ' . . lO.M ,p.ir Ji:..i>W) in) 

St. L., Pet)ri!i, iV N. W Kt ."» ., 1*.M> \k\t $;i».<»<>0 

Unitinl Now JtrM'y Canal Co., p.-ir $l.<>tK» 

Caith in Hanks rrprortrntum Priist Fund'. 



Liquid AiMeta: 

Liberty Boniii* 

United Kn^incerinK Sorirty . 
Account* Receivable: 

Members Dues 

Initiation Fees 

Sales of Publicatiun?, Adverti^inK. etc 

Advance Payments 

Cash: In banks for Reneral puriMMM't* . . 
Petty Cash Fund 



$3 '.>.•■.■>'• SI 
7.M^ 17 



lo.OOS 10 
11.712.50 
.'>S.112.'k3 



r>..->35.S7 
l..VMtOO 



29.307.11 

io.ooaoo 



oS,7 98.87 

o."^0(»0.00 
in.OOO.OO 



87.s;w.l3 
4.573.^8 



_ 7 AH5 87 
$7G7.341.75 



SOCIETY AFFAIRS 9 

LiabilUiea 
Trust Funds: 

life Membership Fund 46.102.81 

Library Development Fund 4,902.71 

Week's Legacy Fund 1.057.00 

Mdville Fund 1,127.36 

Hunt Memorial Fund 208.99 

Juniors* and Students' Prise Fund 2,000.00 

C. T. Main Award Fund 2.500.00 

58.798.87 

Dues paid in advance 2,241.06 

Initiation Fees uncollected 14.712.50 

Replacement Fund 1.163.18 

Accounts Payable 5»244.12 

Unappropriated Revenue 30.392.46 

Unexpended Appropriations-Excess 6.527.48 



Capital Investment $514,792.79 

Surplus and Reserve 146.524.25 



23,864.98 



$661,317.04 
$767,341.75 



Meetings and Program. Under the Spring and Annual Meetings 
of the Society is recorded the result of the work of the Meetings and 
Program Committee, but between the lines must be apparent the 
tremendous amoimt of preparation required to complete the pro- 
grams of these meetings. 

Notable sessions of the year were those on Industrial Relations, 
and Research. The committee's annual report points out that the 
committee has long been under pressure from progressive members 
of the Society to broaden the scope of the general meetings and 
papers presented. 

Standing sub-committees of the Meetings and Program Com- 
mittee are being gradually reorganized into Special Committees 
and now into Professional Sections reporting to the Council. 

Pvblicaiion and Papers, This committee has had a busy year 
in plans for the enlargement of the monthly pubUcation, Mechanical 
Engineering, and progress has been made in the face of many dis- 
couragements in adverse conditions in the printing trade, and in- 
creased costs of pubUcation and materials. 

Under this committee's jurisdiction, there have also been issued 
the annual Transactions, now in its fortieth volume, the Condensed 
Catalogues, ninth annual edition. The Engineering Index and 
the Year Book. 

Condensed Catalogues increased nearly fifty per cent in size 
over the preceding year, indicating progressive recognition of its 
usefulness. 



10 SOCIETY AFPAIKS 

The Engineering Index Annual, taken over from IndiLsirial 
Management in 1919, has been practically doubled in the number 
of pages, and also in the number of copies issued. 

The Year Book has been published in abbreviated form due to 
conditions in the printing trade, but this coming year it is hoped 
to return the book to its former scope. 

Membership, The Society's membership has increased during 
the past year by 2182, and is now 11,882. The report of the Mem- 
bership Committee includes reconmiendations for 11 reinstatements, 
and 63 promotions to higher grade. The committee has held twenty- 
nine meetings and had under consideration nearly three thousand 
applications. 

The full table of activities for the year shows the following: 

Recommended for membership 2182 

Deferred indefinitely 52 

Deferred 33 

Denied promotion 22 

In course of procedure G35 

Reinstatements 11 

Reconsiderations granted higher grades 0>3 

2998 

Honorary Memhership has been conferred b}- the Council upon 
Charles Alphonse Clement de la Poix de Freminvillc, of Paris, France, 
for services rendered his country and the engineering world in the 
practical development of transportation, and Auguste C. Rateau, 
also of Paris, France, pioneer investigator in the field of the steam 
turbine and turbo-compressor. 

In Memoriayn. The report of a special conunittee on War 
Service and Members* Memorial, Major Fred J. Miller, chainnan, 
was received at the Annual •Meeting]:, and the Council records the 
resolutions taken by rising vote at this meeting in memory of those 
meml)ers who made the supreme sacrifice in the Creat European War. 

Resolved: That the Society hereby expresses its jiioatcst api)re(ia- 
tion and pride in the s<ervice of its nienilK^rs who jrave their Hves that 
freedom might be preserved among the nations of the earth. ..." 

COMMITTEE OF ADMINISTRATION AND STANDING 

COMMITTEES 

Local Sections, The Ix)cal Sections of the Society now numl)er 
thirty-six. In the following cities where local section meetings 
are regularly held, cooperation with existing local engineering 



SOCIETY AFFAIBS 11 

organizations has been (ieveIoi>ed: Atlanta, Baltimore, Birmingham, 
Boston, Bridgeport, Buffalo, Chicago, Cincinnati, Cleveland, Denver, 
Detroit, Erie, Hartford, Houston, Indianapolis, Los Angeles, Meriden, 
Milwaukee, Minneapolis, New Haven, New Orleans, New York, 
Philadelphia, Portland (Oregon), Richmond, Rochester, Saint Louis, 
Saint Paul, San Francisco, Schenectady, Toronto, Troy, Tulsa 
(Okla.) Washington, Waterbury. 

The Secretary made three trips to the local sections during the 
year, and visited practically every section. These visits, together 
with the visits of the Committee on Local Sections, resulted in the 
formation of new sections at Cleveland, Colorado, Eastern New 
York, Houston, Tulsa (Mid-Continent Section) Rochester *and 
Washington, D. C. 

In the Cleveland Section there has been installed a plan of 
joint membership in our Society and in the Cleveland Engineering 
Society. 

The Local Sections Conmiittee also hopes to carry out into 
future meetings its cooperation in the professional programs of 
meetings under the jurisdiction of the Meetings and Program Com- 
mittee. This was done at Indianapolis this year. 

Increase of Membership Committee work has this year been 
placed in the hands of the Local Sections Conunittee. Great benefit 
is derived from the local mformation concerning applicants for 
membership. The inclusion of increase of membership work is ex- 
pected to be especially successful. 

Constitution and By-Laws. This committee acts as an advisory 
conunittee, reporting on all matters referred to it by the Council 
relating to the Constitution, and By-Laws and Rules, and harmon- 
izes with the Constitution any new policies of the Society. 

Suggested changes in the rules are being drafted by this 
committee as the result of the recommendations of the special com- 
mittee on Aims and Organization. 

During the past year a notable change in poUcy has been 
effected in providing that the "Nominating Conunittee shall be 
elected annually by the voting membership of the Society." By- 
laws are now in effect which detail the manner of election. 

Library. This Conunittee acts as the representative of the 
Society on the Joint Library Board through which the five Ubraries 
of the Founder Societies and United Engineering Society are ad- 
ministered as one. 

Increased use has been made of the Library Service Biu-eau. 



12 SOCIETY AFFAIRS 

Recataloguing the joint libraries has been started; this is much 
needed in order to make more quickly available the great source of 
information in these individual collections. 

House, The House Committee has given much time to the 
rearrangement and redecorating of the rooms of the Society. Part 
of the necessity for this was for suitable installation of the Memorial 
tablet to Frederick Remsen Hutton, prepared under the direction 
of a special committee, Past-Presidents Ambrose Swasey and Jesse 
M. Smith and Wm. H. Wiley, Treasurer. The foyer hall has a 
small committee room as a result of the remodeling and a more 
harmonious scheme of arrangement and decoration has resulted. 

Research. This year the Research Conunittee, led by Arthur 
M. Greene, Jr., as chairman, has been able to carry out some of the 
ambitions which the committee has had for several years. 

A reading of the complete report of this conmiittee brings out 
many interesting features of the work. 

Svb-Commiitee Reports have been presented in the following 
fields — Bearing Metals, Lubrication and Fluid Meters. A Heat 
Transfer Committee has been organized and started work which 
will take several years. The Research Committee had charge of 
one session of the Spring Meeting in Detroit. 

On one of the trips of the Secretary to I^cal Sections, the chair- 
man of the committee accompanied the Secretary and as a result 
sectional research committees have been established in several 
centers. Manufacturers having research laboratories have expressed 
their cordial interest and desire to cooperate with the committee; 
also educational institutions having engineering experiment stations. 

Through the Research Section of Mechanical Engineering the 
Research Committee has contributed data covering, research results, 
research progress and problems, research equipment, personnel and 
bibUographics. The Council, when asked to appoint representatives on 
the Engineering Division of the National Research Council, appointed 
men from the Research Committee of the Society in order that the 
work of this committee might tie in with the work of the national 
organization under the United States Government. Professor 
Greene, Chairman, was this year\s apix)intee. 

Siandardizalion. This Committee has IxKjn most active, in 
cooperation with the American Engineering Standards Committee, 
which latter has within its organization a memlx*rship representa- 
tive of the Civil, Mining, Electrical, Mechanical Engineers and 
A. S. T. M. The work of this joint organization is reported later. 



SOCIETY AFFAIBS 13 

PBOFESSIONAL AND SPECIAL COMMITTEES 

A scrutiny of the list of committees will show the great number 
of professional and special committees and the vast field they cover. 
All have contributed splendid work in their various lines. Since, 
however, the purpose of this report is to outline only final and com- 
pleted results, all the committees may not receive special mention. 
These committees included some war measure committees and with 
the coming of peace have been " mustered out, " or in some instances 
their research work continued through other channels. 

Professional Committees. Most important in the life of a pro- 
fessional society is the work of its technical committees. The 
^lechanical Engineers may be well proud of the splendid work being 
done by its professional committees at great expense of time and 
personal sacrifice on the part of the many members, each chosen to 
membership in a committee because of his special experience in the 
field covered. The completed work of these professional com- 
mittees is permanently recorded in Transactions and becomes in 
many cases Standard Practice. Notable progress has been made 
by the Boiler Code Committee and the Power Test Codes Committee, 
the latter with its nineteen indi\idual committees. 

Some further mention is made of these committees in the Joint 
Acti\'ities under Standardization. 

Aims and Organization Committee. This committee appointed 
in 1918 has completed its report which has been separately printed 
and widely discussed through all the committees, local centers and 
at two of the general meetings of the Society. Similar committees 
of the other Founder Societies have, through specially designated 
representatives of their main committees, combined in a Joint Con- 
ference Committee, to secure the benefit of "united action of the 
engineering and aUied technical professions in matters of common 
interest to them." 

Drawing their conclusions regarding the final action from the 
two general discussions by the members of the recommendations 
of the report and from the sj-mpathetic attitude of the Council 
regarding the recommendations, the Standing Committees of the 
Society have not been slow to sense the spirit behind this move- 
ment and have had the poUcies recommended under discussion for 
many months with the result that the final action turning the recom- 
mendations over to them, finds them ready to proceed. 

Awards and Relations with Colleges. In his address the Presi- 



14 SOCIETY AFFAIB8 

dent has covered quite completely a review of the plans of the Coun- 
cil to inaugurate closer and more helpful policies with the student- 
engineer and the young engineer as represented in the Jimior and 
Student Member; President Cooley also covered quite thoroughly 
the question of awards and prizes for contributions to mechanical 
engineering, either of a practical nature or in literature. The Charles 
T. Main Award was established this year, from a fund of $2500 
partly contributed by Past -President Main. 

SPRING AND ANNUAL MEETINGS 

In the conduct of the two general meetings this year the Meet- 
ings and Program Committee has been assisted by the Committee 
on Local Sections and the local committees and by the Research 
Committee. 

The Spring Meeting was held in Detroit, and a full account of 
the meeting has been published in the July issue of Mechanical 
Engineering. The keynote session on Industrial Relations was so 
successful that the committee was asked to continue this session at 
the Annual Meeting. Another strong session of the Spring Meeting 
was one on Research, under the auspices of the Research Committee. 

The Annual Meeting showed the best attendance of any meet- 
ing of the Society, the total number registered being 2116. There 
were sessions on Appraisal and Valuation, Gas Power, Industrial 
Relations, Machine Design, Power ^lachincry, Textiles, and Ma- 
chine Shop. The first of these was a joint session with the American 
Society of Refrigerating Engineers, and wiis followed by a request 
to have a session on this subject at the Spring Meeting in St. Louis. 

KEPKESEXTATIOX 

Honorary Vice-Pn^sidents were appointed to represent the 
Society at various functions thnnighout the year, such as conferring 
Honorary Degrees on M. Eugonc^, Schneider by the Stevens Insti- 
tute of Technology, the Annual Meeting of tlui KngincHTing Insti- 
tute of Canada, the National Rivers and Harbors ( 'ongress, the 
Society for the Promotion of Engineering E<luoatioii, the Promo- 
tion of Vocational Education, and the James Watt Centenarj- in 
Bingliam, England. 

PROFESSIONAL SECTIONS 

An amendment to the Constitution is More the Society to 
create a Standing Committee on Professional Sections. Pending 



SOCIETY AFFAIRS 15 

action on this amendment the Comicil has approved the policy of 
inaugurating professional sections and has authorized a special 
committee to report on various aspects of such an organization, 
— as local work, finances, meetings, etc. The new policy is in line 
with recommendations of the Committee on Aims and Organization 
and is intended to provide for those groups of men in the Society 
who are interested in a particular subject, and so sub-divide the 
membership professionally in the same way that the Local Sections 
groups provide a geopraphical sub-division. 

STUDENT BRANCHES 

There are 49 Student Branches. One of the features of the 
Annual Meeting on December 3, 1919 was the holding of a student 
branch conference and Dr. Hollis, Chairman of the Committee on 
Relations with Colleges presided. The plans for the Student 
Branches have been fully approved by the Coimcil as outlined in 
the report of the Committee on Awards and Relations with Col- 
leges submitted at the Spring Meeting in Detroit, and published 
in the October issue of Mechanical Engineering. 



JOINT ACTIVITIES 

Certain joint activities of the Founder Societies, — Civil, 
Mining, Mechanical, Electrical engineering societies — are com- 
bined under United Engineering Society, the holding corporation 
for the property of the Engineering Societies Building and the ad- 
ministrative organization for the funds of Engineering Foundation, 
Engineering Societies Library, and Engineering Council. 

Engineering Foundation. The Engineering Foundation, founded 
in November 1914 by a gift from Mr. Ambrose Swasey, Past-Presi- 
dent and Honorary Member of this Society, for the furtherance of 
research in sciences and engineering or for the advancement in 
other manner of the profession of engineering and the good of man- 
kind, has found its principal activity in cooperation with the Na- 
tional Research Council. In October, Engineering Foundation 
issued a progress report and copies of this pamphlet are on file 
in the Secretary's office. 

Research Council. During the past year the National Research 
Coimcil has been reorganized on a "peace basis." The American 
Society of Mechanical Engineers, three representatives on the Engi- 



18 SOCIETY APPAIRS 

The American Engineering Standards CommiUee was formed in 1918 
with a desire to bring into existence an organization similar to the 
British Engineering Standards Association and cooperate in the 
organization and control of engineering and industrial standards 
in the United States. 

Anglo-American Standardization has been brought into active 
thought of the day, the Secretary of the British Association visiting 
the United States, and in one of the statements covering his mis- 
sion said "It seems to me with so much talk of fierce trade compe- 
tition between America and England, everything ought to be done 
to turn that harmful idea into Anglo-American cooperation rather 
than trade rivalry. The world is surely large enough for the great 
Anglo-Saxon producing countries to find ample scope for their prod- 
ucts >vithout in any way endangering their bonds of friendship, 
drawn so closely by their great defense of Right." 

Permanent Franco-American Engineering Committee. It will 
be recalled that Charles T. Main as President in 1918 formed one of 
a delegation of representatives from the Founder Societies to the 
French Engineering Congress in Paris. Out of this has grown what 
is known as the Permanent Franco-American Engineering C!om- 
mittec, which is a joint committee with the other Founder Societies 
to assist the Government of France and all organizations of France 
desiring information. 



No. 1687 

MEETINGS JANUARY-JUNE 

MEETINGS OF SECTIONS 

]yriNETY-FOUR meetings were held by the twenty-four or- 
ganized Sections of the Society and the Providence Engineering 
Society, an affiliated body, dming the first six months of the year. 
Seventeen of these meetings were arranged jointly with one or more 
local technical organizations or branches of other engineering so- 
cieties. A nmnber of the papers of more general interents were pub- 
lished in Mechanical Engineering during 1919. 

ATLANTA 

February 1: A dinner in honor of Secretary Calvin W. Rice at 
the Druid Hills Golf Club. In the afternoon, Secretary Bice ad- 
dressed the students at George School of Technology. 

March 21: Joint meeting with Birmington and New Orleans 
Sections. In the morning an inspection trip to the Fulton Bag and 
Cotton Mills. In the evening an address by N. C. Harrison on 
Powdered Fuel. 

April 24: Address on Materials for High-Pressure Steam-Pipe 
Work, Wm. J. Neville. 

BALTIMORE 

January 27: Address by Secretary Calvin W. Rice. 
April 29: Papers on Oil Engines, by Leon Wygodsky, and on 
Boiler Explosions, by R. E. Munro. 

BIRMINGHAM 

January 23: Addresses by J. R. McWane and Oscar Wells on 
labor and finance. Abstract of Past^-President Main's President's Ad- 
dress, Broader Opportunities for the Engineer, read by J. J. Greggan. 

February 3: Addresses by R. W. McWane on Efficiency Work 
in the Emergency Fleet Corporation, by Col. T. 0. Smith on Finance, 
and by Secretary Calvin W. Rice. 

April 13: Address on Steel Specifications, O. U. Cook. 

May 23: Informal banquet. 

19 



20 SOCIETY AFFAIRS 



BOSTON 

January 31: Address by Past-President Main on experiences 
in France. Abstract published in Mechanical Engineering. 

March 13: Address by H. W. Rowley on Final Disposition of 
City Wastes. 

April 2: Tenth Annual Engineers' Dinner. Addresses by 
George H. Moses on A I^eague of Nations, by Prof. George Fillmore 
Swain on Reflections Suggested from a Recent Trip to France and 
by Richard H. Rice. 

May 21 : Joint meeting with Boston Society of Civil Engineers. 
Report on the Chicago Public Service Committee, by Mr. Metcalf. 

June 27: Outing at Villa Napoli, Nantasket, Mass., with 
A.I.E.E., Boston Society of Civil Engineers, Engineering Society 
of Western Mass., Providence Engineering Society and Worcester 
Section. 



buffalo 

January 29: Address by Nathan L. liieberman, on Horsepower 
Requirements of Aeroplanes and Power Consumption through 
Parasite Resistance. 

April 2: AVith the Engineering ."Society of Buffalo; illustrated 
address by E. S. CoUins on Industrial Applications of Electric Fur- 
naces. The meeting was preceded by a dinner. 



(:iiicA(;o 

Jariuary 13: Informal got-togothor meeting and dinner in 
honor of Secretary Calvin W. Rice. 

January 27: Address on Sugar Manufacturing, by M. J. Kermer. 

.4/>r?7 21: Witli Western Society of Engineers. Pajx^rs on the 
Triplex Process of Making Steel, by Robert J. Yf)ung, shown by 
motion pictures, and on Fatigue of Metals, by HerlKTt F. Moore, 
illustrated with motion pictures and huitem slides. 

May 20: Ladies' night. Pa|XT on Stress and Strain, by WilHam 
S. Sadler. 

September .30: Address by Mr. Whitten, Chairman, Zoning 
Conunittee, Cleveland, 0., on City Zoning as It Pertains to the 
Requirements for Residential and Manufacturing Districb?. 



SOCIETY AFFAIB8 21 

CINCINNATI 

February 1: Address by Mayor John Galvin on Financial 
Problems of Cincinnati and by Bert L. Baldwin on Conditions of 
Light-Railways Service in France. 

Feiyruary 17; Informal address by Secretary Calvin W. Rice. 

March 20: Joint meeting with the Engineers' Club. Address 
by Chas. H. Fox on The Evolution of the Fire Engine. 

March 25: Business meeting. 

May 15. Illustrated address l)y Ernest F. DuBrul on Trade with 
South America. 

CLEVELAND 

February 4: All-day convention. Addresses in the morning 
by H. E. Simmons on Rubber and Its Manufacture and by G. W. 
Shem on Electric Travelling Crane Development (illustrated). The 
luncheon was followed by an address by J. R. McQuigg on Expe- 
riences of Engineers in France. An inspection trip to the National 
Acme Manufacturing Company's plant occupied the afternoon, 
after which a dinner was held. C. A. Otis and Colonel G. M. Barnes 
gave an account of how the big guns were developed. 

June 10: Joint meeting with Cleveland Engineering Society. 
Morning address, Open-Hearth Charging Machine, by S. T. Well- 
man and I. D. Thomas. liUncheon aboard steamship City of Buffalo, 
foUowerl by a trip to the American Ship Building Plant, Lorain, and 
a talk by J. C. Workman. 

CONNECTICUT 

Bridgeport Branch 

June 26: With local members of A. S. C. E., A. I. E. E., S. A. E. 
and A. C. S. Paper on Mechanical Apparatus in the Treatment 
of the Wounded, by H. W. 0. Thompson; paper on Investigation 
of Case Carburizing and Case Hardening, by Mr. Boeghold. 

Hartford Branch 

May 12: With American Chemical Society. Inspection trip 
to Hartford Rubber Works and to Laboratories of Henry Souther 
Engineering Co. Illustrated lecture on Rubber, by Theodore 
Whittlesey. 

June 5: Address by Hiram Percy Maxim on Sound. 



22 BOCIETT AFFAIBS 

Meriden Branch 

February 28: With Meriden Manufacturers' Association. Ad- 
dress by John C. Spenee on The Training Department After the War. 
Addresses by Charles T. Clayton and H. C. Miles. 

April 28: Address by Joseph F. Keller on The Machine Making 
of Dies. 

May 20: Address by George H. Thacher on The Hand Stoker, 
What It Is and What It Does. 

June 6: Informal meeting and dinner. 

Nev) Haven Branch 

January 8: Address by Douglas K. Warner on The Friction 
of Ball Bearings. 

Jamiary 31: With Yale Mechanical Engineers' Club. Ad- 
dress by G. Douglas Wardrop on War A\dation in Retrospect; Com- 
mercial Aviation in Prospect. 

March 7: With Yale Mechanical Engineers' Club. Address 
by D. C. BucU on Long-Range Na\'y Guns with Railway Moimt. 

March 10: Address by President M. E. Cooley. 

May 27: Address by Hariy Gordon Hayes on Labor Problems. 

Watcrbury Branch 

January 6: Address l)y R. A. Cairns on Waterbury's Water 
Supply. 

April 8: Solution of Factoiy Waste Problems. Professor New- 
lands and others s|)okc. 

DETROIT 

January 11: Informal dinner to President M. E. Cooley and 
Secretary'' Calvin W. Rice. Address by President Cooley on An 
Unoccupied Rung in the Kn mincer's Ladder of Fame, and by Secre- 
tary Rice on l^roader Opportunities for the Engineer. 

April 4: Joint meeting with Detroit EngineiTing Society. Ad- 
dress by William B. Stout on Commercialization of Air Craft. 

EKIE 

February 19: Informal address by Secretary Calvin W. Rice. 



J 



SOCIETY AFFAIRS 23 

INDIANAPOLIS 

January 15: Informal reception to Secretary Calvin W. Rice. 

LOS ANGELES 

March 13: Addresses by Secretary Calvin W. Rice and George 
G. Anderson, Chairman of the Ix)cal Society of the A. S. C. E. Ad- 
dress by Professor Ford on Behavior of Steels imder Test. 

April 16: Address by F. Homberger on The Manufacture of 
Gas from Crude Petroleum. 

MID-CONTINENT 

February 5: Meeting of organization resulting in petition to 
Coimcil for the establishment of the Mid-Continent Section with 
headquarters at Tusla, Oklahoma. Secretary Calvin W. Rice ad- 
dressed the meeting. 

May 23: All-day meeting; business meeting in the morning. 
Inspection of airplanes in the afternoon. Address by Dean J. H. 
Felgar on What Should Be the Content of a Course of Instruction 
Designed to Fit a Man to Become a Petroleum Engineer; (a) From 
the Production Standpoint; (6) From the ReiBning Standpoint. 
Other addresses by George Tayman on Volumetric and Mechanical 
Efficiency of Gas Compressors with Varying Combinations of Pres- 
sure and Vacuiun, and by C. E. Pearce on Graphic Methods and 
Charts for Design of Steam Boilers and Other High-Pressure Vessels. 
Following an informal dinner addresses were given by 0. J. Berand 
on Appraisement and Valuation of Oil and Casing-Head Properties; 
by P. F. Walker on Industrial and Manufacturing PossibiUties in 
the Mid-Continent Section; by Paul Bateman on Tank-Car Main- 
tenance; and by W. S. Smith on Effects of Compressed Air or Gas 
on Petroleum Oil Production. 

MILWAUKEE 

January 15: Address by Henry Ij. Dale on Engineering Ex- 
periences at the Front. 

February 13 : Address by Secretary Calvin W. Rice. 

April 16: Joint meeting with the Engineers' Society of Mil- 
waukee, Milwaukee sections of all National Engineering Societies 
and the Aero Club of Wisconsin. Illustrated address by George R. 
Lawrence on Flying, Today and Tomorrow. 



24 SOCIETY AFFAIRS 

May 21: With Engineers' Society of Milwaukee. Address by 
Arnold Pfau on a Trip to Japan. 

MINNESOTA 

February 10: Dinner to Secretary Calvin W. Rice by the Minne- 
apolis Steel & Machinery Co. Addresses by Secretary Rice, John R. 
Allen, Max Tolz and James L. Record. 

March 4: Address by W. H. Adams on The Manufacture of 
Beet Sugar. 

April!: Regular monthly meeting. 

NEW ORLEANS 

April 14: Address by W. B. Gregory on Pumping Machinery. 
Used by the American Army in France. 

NEW YORK 

January 14: Address by L. C. Marburg on Aims and Organiza- 
tion; by Edwin J. Prindle on The Patent Situation in the United 
States; and by W. W. Macon, H. L. Aldrich and A. J. Baldwin on 
their experiences abroad on a trip of inspection of the battlefields 
of Europe. 

February 10: Joint meeting with Founder Societies. Delegates 
to Joint Engineering Congress spoke on the work of Congress and 
conditions in Franco. 

February 24: Address by P(^tor P. Dean on the Application of 
Electrical Control of Cfate Valves, buffet supjx^r served, followed by 
an address on the Application to Industry of the Personnel Work in 
the U. S. Army, by Lt.-Col. J. J. Swan. At the close of this 
address, motion pictures of animated t(H'hnical drawings for com- 
mercial and scientific use, showing electrical starting and lighting 
systems, the Burroughs adding machine, etc., were shown. 

March 2(): Engineers' Symposium under the general auspices 
of the Local Sections of the American Institute of Mining and Metal- 
lurgical Engineers, American Society of Mechanical Engineei-s, and 
the Society of Automotive Engineers, and in which the meml)ers 
of the American Institute of Electrical Engineers, American Society 
of Civil Engineers, American Chemical Society-, American Electro- 
chemical Society, Ameri(;an Institute of Chemical Engineers, Amer- 
can Society of Heatuig and Ventilating Engineers, American Society 



SOCIETY AFFAIRS 25 

of Refrigerating Engineers, Brooklyn Engineers' Club, Illuminating 
Engineering Society, Institute of Radio Engineers, Municipal En- 
gineers of the City of New York, Soci6t6 de Chimie Industrielle, 
Society of Chemical Industry, and the Society of Naval Architects 
and Marine Engineers were invited to participate. 

The general title of the meeting was The Engineer as a Citizen. 
Gano Dunn, President of the J. G. White Engineering Corporation, 
presided, and the following addresses were deUvered. The Civic 
Responsibility of the Engineer, by PhiUp N. Moore; The Relation 
of the Engineer to Legislation, by Calvert Townley of the Westing- 
house Elec. & Mfg. Company; The Relation of the Engineer to Ad- 
ministration, by Nelson P. I^wis of the PubUc Service Commission; 
The Relation of the Engineer to Public Opinion, by Spencer Miller 
of the Lidgewood Mfg. Company, and The Relation of the Engineer 
to Production and Distribution, by Comfort A. Adams, President of 
the A. I. E. E. 

April 9: Joint meeting with the MetropoUtan Section of the 
Society of Automotive Engineers in a Symposium on the Heavy 
Oil Engine. 

May 28: Addresses by John M. Goodwin on The Five-Color 
System of Camouflage and by Frederick Meron on The Layout and 
Ek^uipment of Factories (illustrated). 

Jvne 10: General discussion of report of Committee on Aims 
and Organization. 

ONTARIO 

May 16: Discussion on the Metric System: Pro: E. F. Burton 
and W. Percy Dobson; Con: Chester B. Hamilton and Ernest V. 
Pannell. 



PHILADELPHIA 

January 28: Address by William B. Dickson on Relations 
between Employer and Employee. 

February 19: Out-of-town meeting at Wilmington, Delaware. 
Address by F. A. Wardenburg on Power Development of the Old 
Hickory Plant. 

Ffbruary 25: Address by E. B. Morden on The Work of the 
Construction Division of the Army from Coast to Coast. 

March 25: Address by Joseph A. Steinmctz on the question 



26 SOCIETY AFFAIRS 

What Are We to Do With Our Returned Aviators and Their Battle 
Planes? 

May 27: Discussion of report of Committee on Aims iEUid Or- 
ganization. 

PROVIDENCE ENGINEERING SOCIETY 

January 7: Address by E. L. Wooley on Work Accomplished at 
Providence for the Emergency Destroyer Program. 

February 12: Annual banquet. Among the speakers were 
P. H. W. Ross, President of the National Marine League of the U.S.A., 
Leonard W. Cronkite of Boston, Special Agent for the U.S. Depart- 
ment of Labor, Captain Delport of the French Army, representing 
the French High Commission, Lieutenant J. A. H. Muirhead, En- 
gineer, officer of the British Army, and Alfred D. Flinn, Secretary 
of the Engineering Council. 

As an added feature moving-picture films of the assembling and 
operation of the 14-iiich naval guns at the front were shown by 
Ensign C. P. McCrae, U.S.N.R.F., of the Naval Bureau of Ordnance. 

M(iy 0: Address by Nicholas Stahl on Central Station Growth. 

May 13: Address by Mark AVhitehead on The Potter and 
Johnston Automatic Lathe and Its Tooling. 

June 7: Inspection trip to the concrete steamship being out- 
fitted })y the Ijord Construction Company. 

June 17: Annual Meeting. 

SAN FHANCISCO 

February 13: Address by F. A. Anderson on Electric Arc Weld- 
ing (illustratod). 

March 19: Insix^ction trip to concrete ship V)oing built by the 
Shipping; Board at Alameda. 

April \i): Illustrated addresses by (\jnnnan<ler Reed, U.S.N., 
describing a destroyer launched in fifteen d'.iys, and by A. P. Allen 
on The Modern Destroyer and the Part It Played in Winning the 
World War. 

ST. i.oris 

January 24: Address by J. M. Olvin on Tlie Manufacture of 
Army Cartridges. 

January 29: With the Associated JilngineerinK Societies of St. 
Louis. Address by L. C. Xonlinever on Refrigeration and ICggs in 
China. 



SOCIETY AFFAIBS 27 

FAruary 6: Dinner to Secretary Calvin W. Rice. 

Febmary 28: Informal dinner. Address by C. B. Lord on 
Women in War Industries. 

March 21: Address by A. S. Langsdorf on Industry, Research 
and the Engineer. 

April 25: Address by Dwight T. Farnum on what the industrial 
engineer is trying to do in industry. 

May 23: Address by Dr. Edward J. Swift on The Human 
Element in Industry. 

June 4: Illustrated address by Wallace C. Capen on Certain 
New Developments in Rear-Axle Construction. 

WASHINGTON 

April 30: Addresses by S. W. Stratton on Standardization of 
Screw Threads; Colonel E. C. Peck on Gage Work of the Ordnance 
Department for the U.S. Army; H. L. Van Keuren on the Certifi- 
cation of Gages at the Bureau of Standards and C. G. Peters on 
The Use of Interference Methods in Cahbrating Length Standards. 

June 6: General discussion on the relation of the mechanical 
engineer to his work, to the commimity and to other engineers. 

WORCESTER 

February 19: Address by M. Eskil Berg on Recent Development 
of Propelling Machinery for War and Merchant Vessels. 

March 6: Address by A. E. Kennelly on Field Ordnance and Field 
Ordnance Appliances. 

May 28: Discussion on Fuel Conservation by J. F. Tinsley. 

THE SPRING MEETING 

Detroit, Mich., June 16 to 19 

The Spring Meeting* of the Society was held in Detroit, Michi- 
gan, Jime 16 to 19, with headquarters at the Hotel Statler. The 
outstanding features of the meeting were the large attendance; 1180 
registered, of which 638 were members; extended discussion on aims 
and organization, sessions on industrial relations, research and 
pulverized fuel and many entertainments. 

The discussion^ of the Report of the Aims and Organization 

^ See MscHANiCAL Engineerinq, July 1919, for summary of discussion. 



28 SOCIETY AFFAIRS 

Committee began on Monday afternoon and was continued on 
Tuesday and Wednesday. The largely attended Research Session, 
lasting all day, showed the awakening interest of engineers in con- 
ditions resulting from the war and an appreciation by them of the 
need for research work in the development of American industries. 
A session on Industrial Relations, in which the human side of in- 
dustrial management was developed, evoked so much enthusiasm 
that a resolution was passed at its conclusion calling for a continua- 
tion of the discussion at the Annual Meeting. The discussion of 
the papers on pulverized fuel on Thursday morning also drew a large 
attendance. 

In accordance with the policy of the Meetings and Programs 
Committee to secure papei-s characteristic of the engineering work 
done in the part of the countrj'^ where the meeting is held, a Sections 
Session was arranged for Wednesday morning with papers contrib- 
uted by the Society's sections of the Mid-West. 

On Monday evening the Local Committee arranged an in- 
formal reception in the balhoom of the hotel, with a brief address 
of welcome by Mayor James Couzens, to which President Cooley 
replied. The reception was followed by dancing. On Tuesday 
evening a concert was given l)y the Burroughs Band at the Arena 
Gardens. On Wednesday, tl»c Committee arranged a sail on Detroit 
River and Lake St. Clair Flats. This trip, which lasted through- 
out the afternoon and evening, gave an opportunity of meeting 
members and friends from all sections of the country. Both the 
Council' and Sections Committee held meetings during the trip. 

Many ladies were in attendance at the meeting. Automobile 
trips were provided, there was a drive around Belle Isle, luncheon 
at the Detroit Boat Club, tea at Red Run Golf Club, and oppor- 
tunity was offered to insjx'ct many points of interest, including the 
United States General Hospital and Priscilla Inn. 

The success of the meeting was due to the faithfulness and 
untiring efforts of the various local committees. A list of the chair- 
men of these committees follows. GtMUMal Committee, H. H. Essel- 
styn; Reception, James Couzens; Printing and Publication, John 
C. McCabe; Transportation and Information, W. E. Cann; Enter- 
tainment, AL W. Talx^r; I^idies' Entertainment, Mrs. F. G. Ray. 
The Detroit Ix)cal Section committee, E. C. FisluT, chairman, con- 
tributed two excellent pajxTs to the meeting. 



SOCIETY AFFAIRS 29 

PROGRAM 

Monday Morning, June 16 

Registration of members and guests at headquarters. Meetings of Council 
and Society's Conmiittees. 

Monday Afternoon 

BUSINESS MEETING 

Report of tellers of amendments to constitution, reports of committees, 
discussion of aims and organization of the Society. 

Monday Evening 
Reception jmd dance. 

Tuesday Morning, June 17 
RESEARCH SESSION 

The Present Condition of Research in the United States, Arthur 
M. Greene, Jr. 

The Organization and Conduct of an Industrial Laboratory, A. D. 
Little and H. E. Howe. 

Address bt Acting Chairman of Engineering Division of National 
Research Council, G. H. Clevenger. 

Research Work on Malleable Iron, Enrique Touceda. 

Reports of SuB-CoMMrrrEEs on Flow Meters, Bearing Metals and 
Lubrication. 

Tuesday Afternoon 

INDUSTRIAL RELATIONS SESSION 

Industrial Personnel Reslations, Arthur H. Young. 
The Status of Industrl^l Relations, L. P. Alford. 

Tuesday Evening 
Musical entertainment and dancing. 

Wednesday Morning, June 18 
SIMULTANEOUS PROFESSIONAL SESSIONS 
SECTIONS SESSION 

Centr\l-Station Heating in Detroit, J. H. Walker. 
Production of Liberty Motor Parts at the Ford Plant, W. F. Verner. 
Fire Engines and the Essentiaia of Fire Fighting, C. H. Fox. 
An Electrical Device for Measuring the Flow of Fluids in Pipes, 
J. M. Spitzglass. 

GAS POWER SESSION 

Crude Oil Motors vs. Steam Engines in Marine Practice, J. W. 
Morton. 

A Suggested Formula for Rating Kerosene Engines, D. L. Arnold 
Standards for Carburetor Performance, O. C. Berry. 



30 BOCIBTT AFFAIRS 

Wednesday Afternoon and Evening 
Steaml)oat trip through St. Clair Flats. 

Thursday Morning, June 19 
SIMULTANEOUS PROFESSIONAL SESSIONS 

FUEL SESSION 

Pulverized CoAii as a Fuel, N. C. Harrison. 

Pulverized Coal for Stationary Boilers, Fred'k A. Scheffier and H. G. 
Barnhurst. 

Economy of Certain Arizona Steam-Electbic Power Plants Using 
Oil Fuel, C. R. Weymouth. 

general session 

Elements of a General Theory of Airplanb-Wing Design, Walter C. 
Durfee. 

Fans for Driving Electric Generators on Airplanes, Capt. G. Francis 
Gray, Liout. John W. Reed and P. N. Elderkin. 

Mechanical Lii-ts, Past and Present, and a New Method for Their 
Balancing, Lieut. J. F. Robbins. 

The Design of Riveted Butt Points, Alphonsc A. Adler. 

Economical Section of Water Conduit for Power Development, 
Cary T. Hutchinson. 

Thursday Afternoon 
Excursions to Ford Motor Company and Packard Motor Company. 



No. 1688 

THE PRESENT CONDITION OF RESEARCH 

IN THE UNITED STATES 

By Abthub M. Gbeene, Jr., Trot, N. Y. 
Member of the Society 

This paper, hy the Chairman of the Research CammiUee of the Society, deals 
wiih the conditions under which research is now being carried on in the United Slates. 
The author first discusses research in its relation to the technical school and gives 
a list of the universities having mechanical engineering laboratories. Engineering 
experiment stations are next considered, lists being given of the stations and of those 
which publish research bulletins. Codperative research and the research activities 
of the Government are next presented and finally the author considers commercial 
and industrial research work, giving in connection thereunth lists of the private 
research laboratories in the country and of manufacturing companies having their 
own research facilities. 

TTISTORY records the development of science from fortuitous 
observations and from systematic, careful and exhaustive re- 
search. The work of the Greeks as shown by Lucretius on the con- 
stitution of matter could only have been made after a careful and 
exhaustive study of the laws of nature. The work of Hippocrates 
certainly indicates a previous study of anatomy and the action of 
certain drugs. The studies of Galileo and Newton, of Kepler and 
Herschel, of Watt and Stephenson, tell of careful thought appUed to 
the interpretation of facts which led to the establishment of laws. 
The story 'of BerzeUus and his kitchen laboratory illustrates the 
spirit of adventure into the unknown which brought to us our early 
quantitative knowledge of the elements. The accidental observa- 
tions of Galvani, Newton, Bell and Crookes by future development 
and study led to results of incalculable value to science, while the 
theoiitical studies of Kepler, Maxwell and Hertz predicted results 
which in giving confirmation to theory gave also confirmation to 
the correctness of experimental observations. 

2 Although all past ages have had men devoted to research, 
they are not niunerous in any one period. If there is one thing, how- 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Societt or Mechanical Engineebs. 

31 



32 PRESENT CONDITIOX OF RESEABCH IX VSTTED STATES 

ovcT, that marks the present epoch, it is the prevalence of this idea. 
In the fjast, research was carried on bj- the few, but today as a result 
of r;ducation and as a result of the success of research in many fields 
ouc can scarcely read a periodical of any branch of science without 
finding some article on this subject. For the last fifty 3"ears the 
coinrnrrrcial value of research has become more and more evident 
to induHtrial plants. This appreciation has been coincident with 
the (frowih of these plants and has made their development possi- 
\fU:, whilf; n;ciprocally the growth of the plants has made extensive 
renoarclj [x^ssiblo. This growth has been intimately connected in 
tlie Harne manner with the growth of our universities and the more 
ext^;nHivc education of our people. In spite of the statement of 
many engineers that the young engineer or college man is of little 
valiu; when ho lx;gins his practical work, these men have proved 
tlif;ir worth in the development of the industries and of the present 
civilization. 

'4 i'or many years our colleges and universities have had their 
laUiratories where some research has been done, and with the de- 
v#:lrif)ment of our industries following the Civil War the necessity 
of commercial lalxjratories for examining or controlling products 
wan f;virifjnt. Forty years ago saw many small chemical laboratories 
arid a Utw Irj.stinK machines scattered throughout the country. The 
•'.tiiir('}t for knowloflgc by some of the investigators and the possibiUty 
of ?ii>i;li';»tif;n of the results of their researches undoubtedly reacted, 
ho th;it tJif: lahonitories of examination and control became labora- 
tori':.-; for invr-.-tijf.'itions in new and unkno\*Ti fields. 

i lU-^.('!ii'<:]i work, quite general >>efore the present war, became 
f/jor<; f.xii'iirivc in overcoming the dastardly appliances of the Hun 
in rjr-virin^ uf-w apj^anitus, y)roducts and manufacturing methods 
.'iri'l in improving riuality and production. The war has demon* 
:tr;it7-'J, if df-rnorihl ration was necessary', the value of researeh, and 
it i.-. uow tlifr fipiiiion of most of us that tliis stimulating viewpoint 
■liouM not \n: lost ;inr| that the war-time interest should be con- 
tin ij<'j. \(>i tills n-asoii it is well to consider the present conation 
of n-.-'-.ircli in the count rv. 

r> Jicf(»n' rjiscussin^ the matter of the physical condition of the 
vMrioij.^ l.'ihoraifjrios iJicrc arc a few general considerations which 
-houM l»e mentioned. The cost of ri'search in the past has been 
-ii^'li that in many cases it could only l)0 undertaken in an escten- 
-ivr- way ly lar^e corporations. The necessity and value have been 
f'.\-'i(li-ui Imj! the small plant has het-n unal.)le to inaugurate that 



ARTHX7B M. GREENE, JR. 33 

which it has known to be of value. In other instances there have 
been investigations which have been of such a nature that there 
would be no commercial gain from their results, although of great 
use to mankind. The necessity of such work has been clearly seen 
and appreciated by some, and in this country institutions and foun-. 
dations have been established by pubUcnspirited citizens to carry 
on investigations or to give grants to those who are working on 
problems for the general good. The reference is to such organiza- 
tions as the Rockefeller Institute for Medical Research, the Carne- 
gie Institution of Washington, and the Sage Foundation. 

6 A number of private conmiercial laboratories have been 
undertaking commissions for cUents, but recently the plan worked 
out years ago by the Associated Factory Mutual Fire Insurance 
CJompanies of New England in their cooperative laboratory and the 
Insurance Engineering Experiment Station has been appUed in the 
cooperative work of certain industries. These investigations have 
been undertaken by an industry as a whole or by a group of manu- 
facturers, and the results have been distributed among the contrib- 
utors to the expense funds, or, in certain cases, the results of the 
experiments have been freely given to the world. 

7 The various states in our Union and the National Govern- 
ment have beUeved that they should make investigations for the 
farmer, and for over thirty years agricultural experiment stations 
have been carrying on research in relation to soils, crops and Uve 
stock and within the last fifteen years the appreciation of their duty 
to the manufactiu^rs and the industries has been shown by the 
establishment of engineering experiment stations. Here the gen- 
eral problems of the manufacturers may be solved and the resources 
of the state developed. 

RESEARCH IN TECHNICAL-SCHOOL LABORATORIES 

8 To turn now to the present research activities, let us con- 
sider first our universities and technical schools. The general equip- 
ment of the university laboratories is planned to give training to 
the undergraduate in methods and to illustrate certain laws. The 
engineering laboratories are equipped so that research work is 
possible, but in many cases the schedule for instruction work is so 
heavy that Uttle or no research work can be done. Nevertheless, 
under these adverse conditions some work of great value has been 
produced during the last thirty years in the technical schools by 
faculty members, graduate students and even by undergraduates, 



34 PRESENT CONDITION OF RESEARCH IN UNITED STATES 

and the proceedings of our engineering societies indicate the extent 
of this work. The lack of time for experimentation has been cared 
for in the engineering experiment stations by the employment of 
full- or part-time investigators. In this way work can be carried 
on continuously to a conclusion. At present the disorganization of 
these laboratories by war activities has caused most of the work to 
cease. According to numerous letters, however, a return to normal 
conditions is looked for within a year. The laboratories of chemis-' 
try, physics, biology and the other sciences have been doing much 
graduate research work. This has been of a theoretical nature 
rather than of the applied form of research more evident in our 
engineering laboratories. The small number of graduate students 
of engineering has partially accounted for the limited amount of 
research from the engineering schools. 

9 The equipment of these technical schools is usually quite 
diversified and adapted for research work of a varied nature. The 
equipment has been planned in many cases for certain problems 
and in some instances special contributions have been made by 
some associated industries for equipment to make investigations of 
problems of that industry. Thus, at Johns Hopkins University the 
gas interests in and around Baltimore donated a fund for the equip- 
ment of a laboratory to study gas manufacture and its by-products. 
At the Carnopiie Institute of Tcchnolog}^ at Pittsburgh a laboratory 
for rollinp;-mill research and instruction is being established from 
funds which are contributed by a number of steel manufacturers. 

10 One of tlio jrrcat noods of tlie present time as voiced by 
directors of a largo (loviM-ninont laboratory and of a large com- 
mercial laboratory is the nocnl for more rosoarch men. Research 
demands a ni:iii of clear vision, gn.'at iniMjrination, tremendous 
resources, absolute lion<'st y, good training and devotion to work. The 
love of the work will have to be th(» incentive as in many cases the 
monetary returns arc small. If our eolleg(»s of engineering and 
science c<)uld by some means instill into more men the great desire 
for discovery through research, they would aid nnich in the con- 
tributions of this age to tin* futur(\ Training is also necessary, and 
that should be done by men engag(^(l in research. 

11 The field of research is broader than ever. As a great 
philosopher expresses 1 it, the relation between tin* Known and the 
Unknown is that of the surface of the sphere: the greater the sphere 
of knowledge l.>eoomes, the greater the surface* of contact with the 
unknown. 



ABTHUB H. GREENE, JB. 35 

ENGINEEBING EXPERIMENT STATIONS 

12 To aid the work of research for the industries and manu- 
facturers by supporting' men on whole or part time to carry out 
investigations, engineering experiment stations have been estab- 
lished in many state universities. These have been active and the 
development of such institutions is considered by some to be of 
such national importance that several bills have been introduced 
in Congress for Government aid in establishing them throughout 
the United States. 

13 The Engineering Experiment Station of the University of 
Illinois, organized in 1903, usually comes to mind when discussing 
this question, although there are fourteen such stations at other 
state universities. The Engineering Experiment Station of the 
University of Illinois up to January, 1919, has issued 110 bulletins 
and 10 circulars on its researches. Twenty-eight of these deal with 
structural problems, 28 with problems relating to fuel, its mining, 
storing, combustion and analysis, 10 with problems of mechanics, 
strength of materials and machine design, 14 with heat problems 
and 11 with problems of electricity and electrochemistry. These 
papers are sent free of cost to interested parties in some cases, and 
in other cases a nominal charge is made. 

COOPERATIVE RESEARCH 

14 While discussing the subject of the experiment station, the 
possibility of the cooperative research as shown by present condi- 
tions and the different methods of solving this problem should be 
mentioned. The problems of hot-air-furnace heating have been solved 
by empirical rules which have had Uttle if any scientific foundation. 
An association of builders of hot-air furnaces has granted the Engi- 
neering Experiment Station of the University of Illinois certain 
funds of money to finance an investigation of these problems, the 
results of the investigation to be made pubUc at its conclusion. 

15 The problems of metal rolling are complex and have been 
studied in the past with difficulty because investigations must not 
interfere with production. The cooperative plan of equipping a 
full-size rolling mill at the Carnegie Institute of Technology equipped 
with special apparatus for varying conditions and making quanti- 
tative determination of different data exemplifies what is being 
done by another industry. 



36 PBESENT CONDITION OF BESEABCH IN UNITED STATES 

16 The paper by Prof. Enrique Toueeda at this meeting illus- 
trates how the manufacturers of malleable iron have formed an 
association to employ a private laboratory to become their research 
laboratory for the purpose of improving and making uniform a 
product which was irregular in its properties. 

17 The Canners' Laboratory in Washington, D. C, gives 
another example of the cooperative method of an industry. In 
this laboratory the problems of the proper harvesting, handling, 
storing and canning of natural food products has been studied and 
the industry guided. 

18 The Mellon Institute of Research of the University of 
Pittsburgh is unique and illustrates a development of research by 
funds contributed to a laboratory by individuals, corporations or 
industries for the solution of problems confronting them. The 
Institute was organized about thirteen years ago by Dr. Robert 
Kennedy Duncan, and the contribution of funds for the support 
of the research was continued by each contributor for one or more 
years. The money so received served to pay the salary of the man 
or men on a special piece of research work and to pay for very special 
apparatus. Tlie Institute houses the research, furnishes ordinary 
supplies and apparatus, affords library and consultation facilities 
and directs the work. The investigations are made for the donor 
and the results belong to him. At present the work is under the 
charge of a director, acting through two assistant directors in charge 
of the fellows on individual and multiple fellowships. In the Insti- 
tute a method of developing complete unit experimental plants to 
study processes for certain donors has been used. In this way 
commercial processes have been developed from laboratory re- 
search in a way not done in many other research laboratories. 

19 At the University of Michigan the Detroit Edison Company 
has established a number of fellowships for research. This indi- 
cates another method of coopi^rative effort and the utilization of 
the equipment of our educational institutions. 

20 A list of co<")ix>ralive efforts in research must mention tiie 
work of the laboratories of the Factory Mutual Fire Insurance Com- 
panies. The work of this association has covered many years, some 
of its bulletins being issued over thirty years ago. Many papers 
and discussions were contributed to the early Tkansactions of this 
Society from its staff. 



ABTHUB If. GREENE, JB. . 37 

EDUCATIONAL INSTITUTIONS HAVING MECHANICAL ENGINEEBING 

LABORATORIES 

21 The educational institutions with which the Research Com- 
mittee corresponded and which are equipped with mechanical engi- 
neering laboratories are enumerated below. 

Alabama 

Alabama Polytechnic Institute, Auburn, Ala. Dean J. J. Wilmore. 

UniverBity of Alabama, University, Ala. Prof. George Jacob Davis. 
Arizona 

Umversity of Ariiona, Tucson, Ariz. Prof. W. W. Henry. 
Arkanaaa 

Univefsity of Arkansas, Fayetteville, Ark. Prof. B. N. Wilson. 
California 

University of California, Berkeley, Cal. Prof. B. F. Raber. 

Leland Stanford Junior University, Stanford Univ., Cal. Prof. W. F. Durand. 
Colorado 

University of Colorado, Boulder, Colo. Prof. J. A. Hunter. 
ConnediciU 

Yale University (S.S.S.), New Haven, Conn. Prof. L. P. Breckenridge. 
Ddaware 

Delaware State College, Newark, Del. Prof. M. van G. Smith. 
Georgia 

Georipa School of Technology, Atlanta, Ga. Prof. R. S. King. 
IlUnois 

Armour Institute of Technology, Chicago, lU. Prof. G. F. Gebhardt. 

Lewis Institute, Chicago, IlL Prof. A. W. Moseley. 

Northwestern University, Evanston, HI. Prof. H. S. Philbrick. 

Umversity of Illinois, Urbana, HI. Dean C. R. Richards. 
Indiana 

Purdue Univeraty, Lafayette, Ind. Dean C. H. Benjamin. 

Rose Polytechnic Institute, Terre Haute, Ind. Prof. F. C. Wagner. 
Iowa 

Iowa State College of Agriculture and Mechanic Arts, Ames, Iowa. Prof. 
M. P. Cleghom. 

State University of Iowa, Iowa City, Iowa. Prof. S. N. Woodward. 
Kansas 

University of Kansas, Lawrence, Kan. Dean P. F. Walker. 

Kansas State Agricultural College, Manhattan, Kan. Dean A. A. Potter. 
Kentucky 

State University of Kentucky, Lexington, Ky. Dean F. P. Anderson. 
Lomsiarui 

Tulane University of Louisiana, New Orleans, La. Prof. W. B. Gregory. 
Maine 

Universily of Maine, Orono, Me. Prof. W. J. Sweetser. 
Maryiand 

Johns Hopkins University, Baltimore, Md. Profs. C. C. Thomas and- A. G. 
Christie. 



38 PRESENT CONDITION OP RESEARCH IN UNITED STATES 

MassachiLseUs 

Massachusetts Institute of Technology, Cambridge, Mass. Prof. E. F. 

Miller. 
Harvard University, Cambridge, Mass. Prof. L. S. Marks. 
Tufts College, Tufts College, Mass. Dean G. C. Anthony. 
Worcester Polytechnic Institute, Worcester, Mass. Prof. W. W. Bird. 
Michigan 
University of Michigan, Ann Arbor, Mich. Dean M. E. Cooley. 
Michigan College of Mines, Houghton, Mich. Pres. E. F. McNair. 
Michigan Agricultural College, East Lansing, Mich. Dean G. W. Bissell. 
Minnesota 

University of Minnesota, Minneapolis, Minn. Prof. J. J. Flather. 
Missouri 
University of Missouri, Columbia, Mo. Prof. H. W. Hibbard. 
Washington University, St. Louis, Mo. Prof. E. L. Ohle. 
Nebraska 

University of Nebraska, Lincoln, Neb. Dean 0. V. P. Stout. 
New Jersey 
Stevens Institute of Technology, Hoboken, N. J. Prof. F. L. Pryor. 
Rutgers College, New Brunswick, N. J. Prof. R. C. H. Heck. 
New Mexico 

New Mexico College of Agriculture and Mechanic Arts, State College, N. Mei. 
Dean A. F. Barnes. 
New York 

Polytechnic Institute of Brooklyn, Brooklyn, N. Y. Prof. E. F. Church. 
Cornell University, Ithaca, N. Y. Prof. H. Diederichs. 
Columbia University, New York City. Dr. C. E. Lucke. 
New York University, New York City. Director of Testing Lab. C. P. BlisB. 
Clarkson College of Technology, Potsdam, N. Y. Prof. A. R. Powers. 
Rennselaer Polyteclmic Institute, Troy, N. Y. Prof. A. M. Greene, Jr. 
North Carolina 

North Carolina College of Agricultural and Mechanic Arts, W. Raldgh, N. C. 
North Dakota 

North Dakota Agricultural College, Agricultural College, N. D. 
Ohio 

University of Cincinnati, Cincinnati, Ohio. Prof. A. L. Jenkins. 
Case School of Applied Science, Cleveland, Ohio. 
Ohio Slate University, Columbus, Ohio. Prof. W. T. Magruder. 
Oklahoma 

University of Oklahoma, Normivn, Okla. 

Oklahoma Agricultural and Mechanical College, Stillwater, Okla. 
Oregon 

Oregon State Agricultural College, Corvallis, Ore. 
Pennsylvania 

Lafayette College, Enston, Pa. Prof. Donald B. Prentice. 
But^kncll University, Lewisburg, Pa. 

University of Pennsylvania, Philadelphia, Pa. Prof. R. H. Femald. 
Carnc^gie Institute of Technology', Pittsburgh, Pa. Prof. W. Trinks. 
University of Pittsburgh, Pittsburgh, Pa. 



ABTHUB M. GBEENE, JB. 39 

Lehigh UniverBity, South Bethlehem, Pa. Prof. Arthuf W. Klein. 

Pennsylvania State CoUege, State College, Pa. Prof. E. A. Fessenden. 

Swarthmore College, Swarthmore, Pa. Prof. G. F. Blessing. 
Rhode Idand 

Rhode Island State College, Kngston, R. I. 

Brown University, Providence, R. I. Prof. W. H. Kenerson. 
South Dakota 

South Dakota State College of Agricultural and Mechanic Arts, Brookings, 
S. D. 

University of South Dakota, Vermillion, S. D. Prof. M. W. Davidson. 
South Carolina 

Clemson Agricultural CoUege, Clemson College, S. C. Prof. W. B. Earle. 
Tennessee 

Univeraty of Tennessee, Knoxville, Tenn. 

Vanderbilt University, Nashville, Tenn. Prof. C. S. Brown. 
Texas 

University of Texas, Austin, Tex. 

Agricultural and Mechanical College, College Station, Tex. 
Utah 

University of Utah, Salt Lake City, Utah. 
Vermont 

Univeraty of Vermont, Burlington, Vt. Prof. E. Robinson. 
Virginia 

Virginia Polytechnic Listitute, Blacksburg, Va. 

University of Virginia, Charlottesville, Va. 
Washington 

State College of Washington, Pullman, Wash. 

University of Washington, Seattle, Wash. 
West Virginia 

West Virginia University, Morgantown, W. Va. 
Wisconsin 

University of Wisconsin, Madison, Wis. 
Wyoming 

University of Wyoming, Laramie, Wyoming. 



UNIVERSITIES HAVING ENGINEERING EXPERIMENT STATIONS 

University of Arizona, U. S. Bureau of Mines Station, Charles E. Van Bameveld, 

Supt., Tucson, Ariz. 
University of Blinois, Charles R. Richards, Director, Urbana, 111. 
Iowa State College of Agriculture and Mechanical Arts, Dr. S. W. Beyer, Acting 

Director, Ames, Iowa. 
Kansas State College of Agriculture, A. A. Potter, Director, Manhattan, Kan. 
University of Kansas, Perley F. Walker, Director, Lawrence, Kan. 
University of Minnesota, School of Mines, W. R. Appleby, Director, Minne- 

iH;x>lis, Minn. 
University of Missouri, E. J. McCaustland, Director, Columbia, Mo. 
Missouri School of Mines, Mining Experiment Station, A. L. McRae, Rolla, Mo. 
Pennsylvania State College, R. L. Sackett, Director, State College, Pa. 



40 FBESEXT C03CI>inOX OF BKPIfAWH IS VSIOD ffCkXEB 

Poriiie UnzrenssT, C. H. B^vrH. Dbocsor. LA£&Telte^ lad. 

Agricahural &cd Merhtni-a? Cc&rf d Tsh» J. C. Xi^ Director, Ccrftege 



it J of r--4iL J. F. MerrZl Dsrertor. Sah Lake CStr, UtdL 
Unirershy of Waehinctoa. C. F. Ntagigggcc Aeds^ Director, Seattle^ WmIl 
Umrersny of Wisoonsn, Aiiress Dirwtcr. Msdisoo. Wii. 



rxi^TRsmns issuing exginiiilring beseabch bulletins 

Rf:ZLSi^hjer FrAy.echnic ti5titu:r, P C. Riiesis. Direcux. Troy, X. Y. 
Universitv of CalL'onJa. Charles Derleth. Jr.. Editor. Berkley, Gd. 
Uoiversity of Minnesc t;^ Addr«:s Din?ctor. Expenmental Engmeeriiig Dept., 
Minnesota, Minn. 

GOVERXMEXT ACTIVITIES IX RESEARCH 

22 Tlie Bureau of Standards at W:ishington and Pittsbui]^, 
the lalx>rator\' of the Unite*! States Bun^iu of Mines at Pittsbiu^, 
the Foo^J Laboraton- and Forest Pnxlucts Lalx)ratory of the Depart- 
rrjf-nt of Aericulturf- and the Naval Experiment Station at Annapolis, 
are a few of the Government aetiNities interested in research. 

23 At the Bureau of Standards research work is being done in 
pliv-ir---, c-hemistr}-. metallurgy', manufaoturine and engineering. 
'Hjere i-- hardlv a branch of human endeavor which is not touched 
hv thi.s enormous research labomtorv. In 1917-1918 there were 
over 14W employees connected with the Bureau, and accounts 
acjrrejratin^ more than $3,400,000 were handled. During this year 
the Bureau issued fifty-three publications and these may be obtained 
tliroujrh correspondence. 

21 I'he primary work of the Bureau is the definition and fixing 
of Ht.'indjirds of iiif-asurements, >tandard constants, standards of 
qiifility, j-tMnd.'irds of fK.*rformance and standards of practice and 
to do this thr*y have flivided the scientific and technical staff into a 
di virion of weights and measures, a division of heat and thermom- 
c-try, '»n electrical division, an optical division, a chemical di\'ision, 
a m;it*Ti;ds division, an enjrineerin^ research division, a metallurgi- 
('ii\ divi.«-ir)n, and a ceramic division. Each division is imder a Chief 
of Division and undcT him there are numerous cxjx^rts and assist- 
antH. The Bureau f(?els that its f miction is one of service to the 
nation and it enric»avors to aid all who apply for infonnation or 
guidance. 

25 T\u) work of the (lage S(»ction of the Bureau of Standards 
during Mi*- recent war activities must 1)C remembered as of the 
l^reut^Ht iiniK)rtance. This dei)artment un<lertook to regulate the 



ABTHUB M. GREENE, JR. 41 

gages used in the various manufacturing plants through its head- 
quarters in Washington and its branches in the East and the Middle 
West. The Section has developed instruments for testing screw- 
thread gages for profile and pitch, instruments for end measurements, 
and in fact it is prepared to test any commercial gage or templet 
for accuracy. The Section has studied the salvaging of gages and is 
vitally interested in the problems of duplicate production. 

26 The activities of the laboratories of the Bureau of Mines 
and the Department of Agriculture are devoted to their special 
fields of endeavor, and in each case scientists of training and experi- 
ence are in charge of the research. 

27 The U. S. Naval Experiment Station at Annapolis, Md., 
is used to study the apparatus and materials used by the U. S. Navy 
or certain Government bureaus. The work consists in making tests 
on these, and in addition researches regarding the general laws 
imderlying the apparatus have been undertaken. The Station is well 
equipped with apparatus and an excellent staff. The work of the 
Station is for Government information, but frequent papers by 
members of the staff appear at times before various technical societies. 

^28 The Forest Products Laboratory of the United States De- 
partment of Agriculture at Madison, Wis., is devoted to problems 
relating to the applications of forest products. 

29 The Watertown Arsenal is equipped for research in materials 
of engineering. The reports from this laboratory have been for 
years the soiu'ce of many data on the strength of materials. 

30 The Philadelphia Navy Yard is equipped for research in 
fuel oils, while the Washington Navy Yard is equipped for testing 
ship models, propellers, airplanes and air propellers. The wind 
tunnels and testing basin are of special merit. 

31 The research activities of the American Society of Heating 
and Ventilating Engineers in connection with the Pittsburgh Labora- 
tory of the United States Bureau of Mines is important and illus- 
trates the activities of certain groups of scientists and engineers. 
This society plans to make researches regarding problems arising 
in its field of endeavor for the benefit of the profession and the 
public. In this project the expenditure of $20,000 per year for a 
number of years is proposed. 

32 Thp dentists of the United States have a number of labo- 
ratories devoted to the solution of their problems. The Research 
Institute of the National Dental Association in Cleveland is one 
which illustrates the cooperative endeavor of allied scientists. At 



42 PRESENT CONDITION OF RESEARCH IN UNITED STATES 

one of their recent conventions the dentists have declared them- 
selves in favor of giving the results of all research to the profession 
without compensation. 

33 According to the pubUc press an Institute for Drug Research 
is about to be established. It is to be supported by the profits of 
the Chemical Foundation which has been formed to take over 
Government-held German patents on chemicals, dyestuffs and 
drugs. In the plan as outlined this work would be done by chemists, 
pharmacologists and physiologists working in the "vast undis- 
covered field of drugs and chemicals for the welfare of mankind.'^ 

COMMERCIAL RESEARCH 

34 The private research laboratories of the country are pri- 
marily devoted to investigations of materials for commercial pur- 
poses, to check products or raw materials or to improve the product. 
These are quite numerous and the list of laboratories given below 
shows in a partial way the private research resources of our country. 
Much of the work done by these laboratories is of the nature of 
inspection, but in many of them the commissions imdertaken for 
clients have been of a true research nature, in finding the cause of 
defects, the methods of improving product and in some cases plan- 
ning actual production methods or processes. 

35 Many of these laboratories have been in existence for al- 
most a half a century; others have been developed in the last decade 
from a local need for such institutions. The work of such a labora- 
tory is to be described in a paper at this meeting of the Society and 
the varied nature of its activities will be seen. 

36 In the correspondence of the Research Committee with the 
various private lal)oratories, the willinjjness to cooperate in the 
work of the Committee was coupled with the statement that most of 
their research work was for clients and, therefore, could not be made 
public. The Committee hopes that in some cases the persons for 
whom the work is done will contribute information after this has 
been properly protected. 

37 The work of these private laboratories covers all fields of in- 
vestigations and new equipment is obtained in many cases for special 
investigations. In some cases a laboratory has l>een specializing 
in problems of a definite character and its eciuipment for this work 
is expensive and complete. The list given Ijelow represents the 
private laboratories known to the Research Committee at the present 
time. 




ABTHX7B M. GBEENE, JB. 43 



PRIVATE LABORATORIES 

Booth, Gairefct & Blair, Philadelphia, Pa. 

Cement plants, cement, building materials, chemistry. 
Dayton Enc^eering Laboratories, Dayton, Ohio. 

Spark plugs, auto and airplane engines, wireless apparatus. 
Detroit Testing Laboratories, Detroit, Mich. 

Dairy products, lubricants, soaps, road and building materials. 
Electrical Testing Laboratories, New York. 

Instruments, lan^ie, insulation, fuels, lubricants, paper. 
Fitxgerakl Laboratory, Lie, Niagara Falls, N. Y. 

Electrodiemistry. 
James H. Henxm, Cleveland, Ohio. 

Metallurgy, chemistry, ceramics, inspection. 
Robert W. Hunt & Co., Chicago, HI. 

Metallurgy, chemicals, materials, apparatus, inspection. 
Institute of Industrial Research, Washington, D. C. 
Institute of Fermentology, Chicago, ID. 
B. B. Lathbury, Philadelphia, Pa. 

Cements, materials, inspection. 
Leeds & Northup, I^iiladelphia, Pa. 

Electric instruments, electrochemistry, heat treatment. 
Lehigh Valley Testing Laboratory, Allentown, Pa. 

Cements. 
Lincoln Hanson and Abbott, Portland, Me. 

Electrical apparatus. 
Arthur D. little Co., Inc., Cambridge, Mass. 

Qiemical analysis, processes, paper, foods, textiles, metallurgy. 
New York Testing Laboratory, New York, N. Y. 
Pittsburgh Testing Laboratory, Pittsbiu^ Pa. 

General testing, metallurgy, chemical, materials, inspection. 
Rome Testing Laboratory, Rome, Ga. 
Rubber Trade Laboratory, Newark, N. J. 
S. P. Sadler, Philadelphia, Pa. 

Chemical analysis, processes. 
Henry S. Spackman Eng^eering Co., Philadelphia, Pa. 

Cements, materials, chemicals. 
Textile Trade Laboratory, Newark, N. J. 
Enrique Touoeda, Albany, N. Y. . 

Metallurgy, chemical analysis. 
Underwriters Laboratories, Chicago, 111. 

Hre-protective apparatus, electrical apparatus, insulation, chemical analysis. 
United States Conditioning and Testing Co., New York, N. Y. 

Textiles. 
John H. Yocum, Newark, N. Y. 

Leather and oil trade. 



46 PRESENT CONDITION OF BESEABGH IN UNITED STATES 

high as 200,000 or currents of 12,000 amperes may be obtained. 
The building is piped with city water, river water, illuminating gas, 
high-pressure hydrogen, low-presbiire hydrogen, oxygen, high-pres- 
sure steam, compressed air and vacuum suction and vacuum clean- 
ing. Distilled ^ater is supplied to any room by gravity, and liquid 
air may also be obtained. Various kinds of gas and electric vacuum 
and arc furnaces are installed in one part of the building. 
A furnace is installed for argon purification. Various crushers, 
grinders, rolls, punches and a 60-ton hydraulic press are installed. 

49 The illuminating laboratory, which is distinct from the re- 
search laboratory, is devoted to special problems in stud3ring the 
best lighting units or methods of utilizing these units. The con- 
sulting engineering department laboratory, devoted to high-tension 
phenomena, the testing laboratory for materials, the standardisa- 
tion laboratory for instruments and the development of new instru- 
ments represent activities at Schenectady which are devoted to 
research in their commercial routine duties. Many engineering 
departments are constantly making investigations which are of a 
research nature. 

50 The work at the laboratories at Ljiin and Pittsfield is largely 
applied to the production problems of these plants and at Harrison 
and Cleveland the problems of lamp production are studied. 

61 The problems of the Westinghouse Company are of a simi- 
lar nature to those of the General Electric Company. 

62 The research la bo rat or}' of the Eastman Kodak Company 
represents the research activity of another manufacturing corpora- 
tion. The staff of this laboratory' consists of about fifty men, some 
fifteen of whom are specialists. The budget amounts to more than 
$100,000. The work of the laboratory is devoted to physics, chem- 
istry, to plant problems and new development. In this laboratory 
full-size apparatus is used at times in making research. As shown by 
the equipment of many laboratories the study of complete processes 
in the laboratory' on a commercial scale is one of the features of 
the times. The problems of this laboratory are organic, inorganic 
and colloidal chemistrj-, optics, color photography, film products 
and applications of general photography, chemical products and 
emulsions. 

63 llie laboratory of the National Limp A.ssociation, now the 
Nela Park Laboratory at the National Lamp Works of the General 
Electric Company at Cleveland, Ohio, represents one of the best- 
known research institutions of this country. The research work of 



ARTHUR M. GREENE, JR. 47 

this laboratory is devoted to the physics, phj'siology and psy- 
chology of light, the production, utilization and efficiency of lumi- 
nous energy. In this laboratory ^^ith a staff of eight investigators 
of the highest ability directing the work, all kinds of illiuninat- 
ing problems are studied from every angle. The laboratory has a 
policy of sending out its experts to study local conditions, and in 
many cases investigators from other institutions come to this labo- 
ratory to carry on research. During the first eight years of its exist- 
ence from 1908 the laboratory has produced 125 high-grade papers. 
These are abstracted by the authors and the abstracts published 
at intervals. In having abstracts made by the authors the impor- 
tant points of the researches are sure to be covered. 

54 The laboratory of the Packard Motor Car Company is de- 
voted to the study of materials used in their plant, the inspection 
and test of suppUes and finished work as well as the development of 
new devices or processes. 

55 The great extent of research faciUties is disclosed by the 
following list of laboratories used by the Research Conmiittee. 

MANUFACTURING COMPANIES HAVING RESEARCH LABORATORIES 

Air Reduction Company, New York City. 

Aluminum Castings Company, Cleveland, Ohio. 

Aluminum Company of America, New Kensington, Pa. 

American Agricultural Chemical Company, New York City. 

American Beet Sugar Company, New York City. 

American Brass Company, Waterbury, Conn. 

American Locomotive Company, Schenectady, N. Y. 

American Optical Company, Southbridge, Mass. 

American Rolling Mill Company, Middletown, Ohio. 

American Sheet & Tin Plate Company, Pittsburgh, Pa. 

American Smelting & Refining Company, New York City. 

American Telephone & Telegraph Company, New York City. 

Amoskeag Mills, Manchester, N. H. 

Arlington Mills, Lawrence, Mass. 

Armour & Company, Chicago, HI. 

Armstrong Cork and Lisulation Co., Pittsburgh, Pa. 

Atlantic Refining Company, Philadelphia, Pa. 

Babcock & Wilcox Company, Bayonne, N. J. 

Baldwin Locomotive Works, Philadelphia, Pa. 

The Barrett Company, New York City. 

Bausch A Lomb, Rochester, N. Y. 

Berlin Mills Company, Berlin, N. H. 

Bethlehem Steel Company, So. Bethlehem, Pa. 

Browne & Sharpe Manufacturing Company, Providence, R. I. 

Buick Automobile Company, Flint, Mich. 



Xm 



yt.ffx»^!: zx^szmzjs 29 



TfH ^V \.>IIEClUT. r T VT nr rr, ^ 






PKLJ. i^ni .1 



Q-jT'^jr GJME • 333. ".I — >»r y. Y. 






Ford A:r-c=f:iii* CccirazLT. De^c^:. Mjzi. 
FtiI-x-:: B&g izi Ccr«:i. Mils. Aili^-Jw Ga. 
Gtasnl Bairl;-^ Cc z:p«iz.j Nr«- Ycck Chv. 
G€::*ral Chr=i:al Cczip5*cT. X*w Yc« Chj. 

Oiio EiTTJ.:::. N. J.. Lviji. MiSk 

m 

'jrSia«:~i rr— ..i- V_' j,a-_j. ^.'r»-_i— -. v. — .0. 

Ho! :«:•=.''. S:cr! C:z-rizy. S:.Ti.:.:f':, X. Y. 

Ingerr-'/.l-Rmi C-:=ipiny. N-:w V:rk C::y. 

Intern i*. :. L il H Ar-.e^: e r C-. ::. p ah y , C :.: : i£ :■ . 1 1:. 

IriV:rrii*i'.:ii! A.lvh=-:r: GripLi:^ CoiLi-iny, Xi:i^iri F^L?. X. Y. 

H. M. .J::.ii5-Mir.'.illc Co.. Xtw- York. X. Y. 

L.iokawi:.Li .St* -rl Conipiny. BuiTil.^. X. Y. 

Lifi'i'i Air Pr-yi'jcta Co:r.paLy. Xew York ».'ry. 

Lu'llurii Sv;'l Co mp Liny, Coloni'?. X. Y. 

Mi'lv-'ilr; S' 1^.-1 Cor;;p.*r.y, Philjivlphi;*. P:!. 

XaMofiil A.'jiiirii; .v Chvrr.iojl W. -rk.-. BufT.xlo, X. Y. 

Xntiofi'il ^'■'irr^oii Cor^ff^ny, Cirvo]:tr.'i, Ohio. 

X.-ition.'il ^'.ihii H»Ln^ter Conjp:»ny, l);iy*.*.'n, 'X'.i'. 

N-'iMon.-il Ijjriib«T M:inufa':turi;r5' A«.H.".iaii'.»n. Chi';i*:".>. 111. 

.\':l;i K«-'firoh L;il»oriitorv, X*-la Purk Lat'Orutorv. CK-vi.l:ir.ii, Ohio. 

N':w J'T^'-y Zinr <"orijp.'iny, Xow York City. 

N«;w York Sliii^hiiiMirifr <"oriKj ration, Camdrn, X. J. 

.\''\vjMirf. St\sn ."^hiphiiiMirig <^'ornpany, XewiK.irt Xtws, Va. 

Pji«k;iri| AutoMioliil*: T'ompany, Detroit, Mich. 

Pennsylvania It.iilrrirul Company, Altoona, Pa. 

Pfnii^ylvania Salt Mfjr. C'onjpany, Philadelphia, Pa. 

Pierce-Arrow Aiitomofiilo T'oinpany, BuiTalo, X. Y. 

Pit.tHliiirKh IMat^r (jl.mH ('oiupany, Pittsburgh, Pa. 

Pratt Af Whitney Company, Ilartfonl, Conn. 

pHMiiHion IiiHiriimcnt Company, Detroit, Mich. 

PymJiTtrir Inhlninient Company, Trenton, X. J. 

Kfiny Klrririr Company, Detroit, Mich. 



ABTHUB M. GREENE; JR. 49 

Reo Motor Company, Laudng, Mich. 

Sangamo Meter Company, Springfield, 111. 

Sears, Roebuck Company, Chicago, 111. 

Solvay Prooefis Company, Syracuse, N. Y. 

Standard Oil Company, New York City. 

Studd>aker Corporation, South Bend, Ind. 

B. F. Sturtevant Company, Hyde Park, Mass. 

Tayk>r Instrument Company, Rochester, N. Y. 

Titanium ADoys Manufacturing Company, Niagara Falls, N. Y. 

Tidewater Oil Company, Bayonne, N. J. 

Union Switd^ A Signal ComxMmy, Swissvale, Pa. 

United Gas Improvement Company, Philadelphia, Pa. 

United States ladustrial Alcohol Company, South Baltimore, Md. 

United States Smelting Company, New York City. 

United States Steel Corporation, New York City. 

^^ctor Phonograph Company, Camden, N. J. 

Welsbach Company, Gloucester, N. J. 

Western EUectric Company, New York City. 

Westin^ouse Airbrake Company, Wilmerding, Pa. 

Westinghouse Electric & Manufacturing Company, East Pittsburgh, Pa. 

S. S. White Dental Manufacturing Company, Philadelphia, Pa. 

56 I cannot close this paper without referring to one agency 
which not only has been of utmost importance during the present 
war but also in its continuance in the times of peace will have a 
still greater influence on the developments in science and industry. 
I refer to the National Research Council, the purpose and work of 
which is best explained by the following executive order issued by 
the President of the United States, May 11, 1918: 

The National Research Council was organized in 1916 at the request of the 
President by the National Academy of Sciences, under its Congressional charter, 
as a measure of national preparedness. The work accomplished by the Council 
in organizing research and in securing codperation of military and civilian agencies 
in the solution of military problems demonstrates its capacity for larger service. 
The National Academy of Sciences is therefore requested to perpetuate the 
National Research Council, the duties of which shall be as follows: 

1 In general, to stimulate research in the mathematical, physical and 
biological sciences, and in the application of these sciences to engineering, agricul- 
ture, medidne and other useful arts, with the object of increasing knowledge, 
of strengthening the national defense, and of contributing in other ways to the 
public welfare. 

2 To survey the larger possibilities of science, to formulate comprehensive 
projects of research, and to develop effective means of utilizing the scientific 
and technical resources of the country for dealing with these projects. 

3 To promote codperation in research, at home and abroad, in order to 
secure concentration of effort, minimize duplication, and stimulate progress; 



^. j^Liiy^^ ; j.A^i'?M7iy zw ffrnffymnr is laimp states 

4 T: ssrr? w i '"»'-™« if Trmrnir 









f 1: rv.'*t*r uii scilii^ si^asinjii; kii£ •sgfnnnafc Jtomatasaam al faooie and 

cc 'Jiii iiriisiiifi: izti vrbr::iraii rcuisiss :c Sit GriTenaififf, enc^ niEuiy and 
^rrZ- 7 : iiJ: •£=*! r«.Psai*ff- A Tr r5g :i zhn Grr^'SSBoatrL ipsa ike maaiDaikm of 
^b^ ?r^>t=.* lif 'Jm ^iZiiicAl AskSsciT -x Sssehmbl v^ be doHnated bj the 
Pr«siie=.; jiS =RCLt«rs :c *.^ Ccc-fil Ai iisrscccir* aai «^ keftdi of the de|iert- 
=ir:z.i5 - — r^zLi.'.'zi'j xC'MmfC tlI xc'iiz.ifci ic Ax^xcsse a CTOT wiy that may 

57 TL-r '.v::k ini j :::.:: v :: :jir X-i::::-iI Research Councfl 
*.\ill ':.-=: iisjifi^-: m ir.-: ;: :ir r.i7»{r? :i.: ihis meeting, and I wiD 
iio: i'.vr:.:: :■: -i:. ir^ivrj.:.: ::.-: v.iriii : ::ll-::r^ which it has under- 
tiikvL. To ::.-ik-. i:.ii i.i:>r :::::!-> 'i:^ever, I will give an 
or.!l:.e «,■:* ::.♦.- vri:-ir.:z:a:::ir. Ji5 .: -xl^v : iir^r.z :he war period. 

o> T:-*j '.'"ir. ■:: ;^\i.^ ur. irr :':.-. ."::..:r::::i::sh:p o: Dr. George E. 
Hule '.v::h ::.:►-:- vijo-jhuirn.-.n. :*:: ox*. ..:::vo s<ore:itr\*. a treasurer 
and -.wo a.-'-i-*:;:.: -';:M-:.ir:v-. T'.-? oxv;:::vv bviir,: under the chair- 
ruan-fiip of Dr. J-.-hn J. T'liry .or^:-:*.-: .::* ::.-; :■:!:. er< of the Council, 
til': ciiairman ari'i virT.-'/lMiirrr-iri of ■!:■.■:-:..:.> nr.-.i the chairman of 
j-er-tion- of T.h»: ^M.-n^Tn! R».'l.ri..ii> Divi-iv:.> :..-j»j:her with six elected 
iii^-rfjh<-r.-:. 'J'ho DivLsion.s of tlie <Juuii«.li worv ;is f*.»iluws: 

1 Thf:r Divi.-ion of Genoml Relutiun.-? 

2 Military DivLsion with its R*/>i*arch Inft>niiation Service 
.'i I)ivi.siori of Eii(riiif,*eriiig 

4 l>ivi.-sion of Physjicsj, Matheinatic?, .\<trunomy and Geo- 

[ihy.-ics 
i) Hi vision </f Chorni.-tr>' an<l Chemical Technology- 
Division of TJoology ami Tleography 

7 Divihiiin of Medirino and Related Sciences 

8 Division of Agriculture, Botany, Forestry, Zoolog}- and 

Fiabcriee. 



DISCUSSION 51 

EsLcti Division was divided into committees and sections, covering 
special features of the work. There were over one himdred scientists 
representing various technical and scientific societies, educational 
institutions, commercial laboratories and manufacturers. 

59 Tlie work accomplished by the Ck>uncil during the war has 
been so imp(»i;ant and many of the investigations and researches 
which were not completed gave so much promise for the future that 
the Coimcil has been reorganized to continue its work for the further- 
ance of science and its appUcations. 



DISCUSSION ON ENGINEERING RESEARCH 

npHE following general discussion of the subject of en^eering 
research has to do with papers Nos. 1688, 1689 and 1690. 

P. F. Walker (written). Professor Greene has truly said that 
in most of our educational institutions the members of the faculty 
are so burdened with routine teaching that large amounts of re- 
search work cannot be expected. There is good reason, however, 
why the schools need to give active attention to scientific investi- 
gation. It is the schools that must produce the men to go into 
active professional work, including research. In the schools these 
men have implanted in them certain ideals which will remain, per- 
haps imconsciously, as determining factors in their lives. To pro- 
mote research therefore is of vital importance in order that the 
student should be brought face to face with research problems, and 
thus imbibe something of the spirit of the investigator. 

A matter in which the writer is at present personally interested 
and to which he is giving a great deal of attention is the state in- 
dustrial problem. It is a research side of that new branch of the 
profession sometimes designated as industrial or commercial en- 
gineering. Every state and every community has problems peculiar 
to itself and it is a function of state educational institutions as well 
as of the engineering profession to interest themselves in such 
problems with the aim of rendering service. A combined survey 
and study of industrial possibilities is being made. This is men- 
tioned because it is a kind of research not always thought of as 
coming within the scope of the engineer. To the writer, however. 



52 PRESENT CONDITION OF BS8SABCH IN UNITED STATES 

it seems that it is an activity which comes distinctly within the 
terms of that definition which states that engineering is the applica- 
tion of all of the sources of power to the use and convenience of 
man. In any community, and particularly in one where the natural 
resources are not yet developed by industries to the full extent that 
economic conditions will warrant, it is the business of the en- 
gineer to encourage and pave the way for new indushial develop- 
ment as truly as it is to develop the natural agencies for its 
succossful prosecution. 

F. J. SciiLiNK {imritten). The importance of real research 
work and its possibilities in returning a quite extraordinary profit 
for a very moderate expenditure cannot be over-emphasized. If 
this end is to be attained, however, the venture must be undertaken 
with a broad vision and the investigator must be permitted within 
reasonable limits to carry on studies which may not promise imme- 
diate pecuniary benefits. 

It is nothing less than astonishing to find, as we occasionally 
do, thjit a lonj^-ostablislied product, supposedly fully standardised, 
is being manufactured without even a superficial knowledge of the 
simple engineering and scientific facts that imderlie its performance. 
Some of the portable or hand-operated fire extinguishers are splendid 
examples of this class, and in these simple devices, which are not 
nearly so compl(»x, from the designing and manufacturing stand- 
|)()int, as an alkmi clock or a lawn mower, the varied and manifest 
types of mechanical and oj)erating failures presented are well-nigh 
unbelieval)le. One can say with a high degree of certainty that 
the ex|)en(liture of $500 in a real engineering study of the problems 
of material and function in any one of several such extinguishers 
would have made unnecessary the waste of many thousands of 
dollars in ineffective designs, some of which arc an absolute menace 
in that they give the owner a false and unfounded sense of security 
against fire hazards. 

The t'ondition of American business in which large profits 
could he made without the necessity of careful and rigorous 
attention to engineering research and standardization is rapidly 
passing away in the face of the present perplexing industrial 
situation. The advancing costs of labor and material, with the 
concurrent unwillingness of the public to pay increased prices for 
the manufactured product it consumes, are bringing to every manu- 
facturer a problem of almost crucial character, of holding down or 



DISCUSSION 53 

reducing his manufacturing costs without depreciating his product. 
This difficulty must be met on the one front by research and stan- 
dardixatioUy and on the other by the efficient management of labor; 
and the former, though not so well advertized or perhaps even so 
highly regarded, is likely to prove at least as powerful an ally as 
the latter. 

Arthxtb J. Wood (written). Professor Greene states that one 
of the great needs of the present time is for research men and 
follows with a statement which in itself explains why there are 
no more men available. He says, "The love of the work will prove 
to be the incentive as in many cases the monetary returns are 
small." I believe that the author is conservative in saying "many 
cases," — it should read "most cases." 

Monetary returns will never make a research engineer out of 
one whom nature never fashioned for such work, but it will produce 
results of unmeasured value from the one who has the love of and 
ability for research work if it frees him from financial cares which 
he otherwise must carry. 

Surely the need is for more research men and the industrial 
laboratories look primarily to the colleges for their material, but 
what are the colleges, as a whole, doing to develop men along these 
lines? 

In the period of awakening to the value of research there will 
be plunges into subjects without proper preparation or preliminary 
study and analysis, and some half-completed results of tests will 
doubtless be put forth as research, partly because the colleges can- 
not or do not pay salaries to teachers and investigators which would 
train the right men in the proper lines. The making of a research 
worker is a long-time process, and it calls for a normal, straight- 
thinking mind, well informed and evenly balanced, which must be 
left free to get results. One of the apparent needs for men in col- 
legiate research is for more leisure in which to think out methods, 
to plan the work in fields not already well occupied and to keep the 
mind thoroughly saturated with the subject at hand. How much 
leisure should be granted? The man himself should be the one best 
qualified to judge. 

Regarding industrial vs. pure research, there should be no 
conffict; no line of demarkation can be drawn between the two 
although they are essentially different. One is helpless without the 
other. No research is so "pure" but that it may some day lead to 



54 PRESENT CONDITION OF BE8EARCH IN UNITED STATES 

results of commercial value. It is unfortunate that some en^eers 
do not have a true conception or an adequate appreciation of the 
value of scientific research work. 

The Society through its Committee on Research and the various 
sub-committees may be an important factor in guiding some of 
this work. It may help to raise the standards of pure research 
so that industrial and educational institutions alike will accept the 
definition of research as given by the late Dr. R. H. Thurston as 
the "art of revelation and prophecy." The Society must be a leader 
in the great work and place research in its accepted and well- 
deserved place among engineering enterprises. 

H. S. Coleman. The system of research at the Mellon In- 
stitute was formulated by the late Robert Kennedy Duncan and 
placed in experimental operation at the University of Eomsas in 
1906. In 1911 the system was inaugurated at the University of 
Pittsburgh, and in 1913 established on a permanent basis there 
through a gift, by Messrs. Andrew William and Richard Beatty 
Mellon, bankers of Pittsburgh, of a modern research building and 
an endowment to cover the general overhead expenses and salaries 
of the administrative staff. 

Any company or association of manufacturers having a prob- 
lem or group of problems requiring investigation may become the 
donor of an industrial fellowship by contributing to the Mellon 
Institute a definite amount of money, for a period of not less than 
one year. The foundation sum must be adequate for the purchase 
of all necessary special apparatus or other equipment as well as to 
furnish the annual i-tipcnd of tlic research man or men selected to 
work on the particular problem. Tlic Institute houses the investi- 
gatory work, furnishes it with the use of its permanent equipment, 
affords library and consultative facilities, gives careful direction to 
the progress of the research, and provides an atmosphere which is 
conducive to pnxhictivti iiKiuiry. All results obtained during the 
course of the in<lustrial fellowshif) belonjr exclusively to the donor. 

The Institute is not, in any sense of the word, a commercial 
institution, beinp entirely independent and deriving no financial 
profit from any investigation conducted under its auspices. In fact, 
during the last fiscal year, it was necessary to draw upon the en- 
dowment fund for almost $70,000. 

Up to the present time the engineering research carried on at 



DISCUSSION 55 

the Mellon Institute has been rather limited in scope as the present 
building and equipment were designed mainly for chemical research. 
There are, however, three fellowships at present in operation in- 
volving engineeiing research, and plans are now under way for the 
construction of an engineering research building which will provide 
adequate facilities for extensive engineering investigation. 

Investigations are usually worked out in three stages. First, 
there is the laboratory stage; then comes the unit plant or semi- 
commercial stage and finally the process is placed on a commercial 
basis in the plant of the donor. The fellow who has developed the 
process through Uie laboratory and unit-plant stages is usually at 
this time taken over by the company and placed in direct charge of 
the new process. 

As a result of the investigations involving the development of 
new processes, a large and valuable collection of special equipment 
has been acquired. This equipment, in most cases becomes the 
property of the Institute al the expiration of the fellowship for 
which it was purchased, and is available for further use in con- 
nection with new problems. 

John R. Allen. The American Society of Heating and Ven- 
tilating Engineers decided about a year ago to establish a research 
laboratory. After a thorough investigation of the available loca- 
tions the committee in charge decided to locate this laboratory at 
the Bureau of Mines in Pittsburgh. The Bureau of Mines kindly 
agreed to supply the necessary laboratory space in their new ex- 
perimental building. 

The fimds for carrying on the work of the laboratory have been 
largely supplied by the members of the society and by manufac- 
turers interested in research of this character. 

The work in general will not be of a commercial nature but 
the problems taken up will be of a fundamental character. The 
committee in charge of the work has laid down three principal 
activities for the Bureau: 

1 The collecting of all the references covering research work 

along the Ihaes of heating and ventilation. This will be 
a card index and will be particularly for the use of the 
Bureau but will be available for all members of the 
society. 

2 The work of research, which is divided into two portions 



56 PRESENT CONDITION OF RESEARCH IN UNITED STATES 

— research at the laboratories in Pittsburgh and research 
in other institutions. 
3 The standardizing of all instruments and methods of 
testing. The purpose of this is to establish uniform 
conditions of arriving at conclusions so that all results 
can be properly compared and to ascertain what d^ree 
of accuracy instruments should give in order that results 
may be reliable. 

It is not the intention of the Bureau of Research to repeat 
work that has already been carefully and accurately done or to 
develop apparatus at the Bureau which is already well developed 
in other institutions. It is the intention of the Bureau to make 
use of all the available equipment in the country as far as is pos- 
sible. Part of the work will be the collecting and editing of this 
material and the conducting of experiments along lines not hitherto 
conducted in other institutions. 

The intention of the Research Laboratory in general is to give 
heating and ventilating engineers more definite data in regard to 
the main questions arising in their business so that the work of 
the profession can be done with greater accuracy and certainty 
than is possible at the present time. 

Zay Jeffries.^ The Aluminum Castings Co. is engaged in a 
line of endeavor which for a long time has had for its object the 
refining of methods of production. 

The Aluminum Casting Company's laboratory employs be- 
tween 80 and 90 individuals. It is a two-stor>'^ building 50 by 230 
ft.; and its buildings and equipment are valued at approximately 
$150,000. The annual expenditure for this work alone is about 
$300,000. We are finding out a great many things in connection 
with non-ferrous alloys, but we are only advanced far enough to 
apply these fundamentals to our industry. Our work is new, and 
yet its application is even now producing results, and no doubt will 
have more extended use in the future. 

Charles Russ Rk^hards. Some 25 years ago the first attempt 
was made to provide for federal aid in the establishment of Engi- 
neering Experiment Stations, but for one reason or another the 
effort to secure congressional action was futile. A few years later, 

* Director of Research Laboratory'i Aluminuni Castings Companj-, Cleveland, 
Ohio. 



J 



DISCUSSION 57 

the officers of the College of Engineering of the University of 
Illinois succeeded in interesting the university authorities in engi- 
neering research, and on December 8, 1903, the Engineering Experi- 
ment Station was established by an act of the Board of Trustees. 
It has since undertaken industrial research in a great variety of 
lines. 

The Engineering Experiment Station is an organization within 
Uie College of Engineering. The control of the station is vested in 
the Executive Staff which is composed of the Director and his 
assistant, the heads of the several departments of the College of 
Ekigineering, and the Professor of Applied Chemistry. The mem- 
bers of the faculty are encom-aged to devote to research work as 
much time as they may have at their disposal. If any one desires 
to imdertake the solution of an important problem, we attempt to 
relieve him of some of his exacting teaching or administrative 
duties so that he can give a portion of his time to the research. 
Most of the research work in the station, however, is carried on by 
the Research Corps composed of various full-time and part-time 
research assistants and special investigators who are employed for 
specific purposes. 

In addition to the research work conducted from university 
fimds, the Engineering fixperiment Station has imdertaken from 
time to time cooperative investigations of problems of importance 
to an industry or a group of industries, imder an arrangement which 
provides that the cooperating agency shall pay the principal ex- 
penses connected with the investigation. At the present time there 
are four or five such investigations in progress. 

As an institution supported by the state, it is necessary for 
us to safeguard the interests of the public in all research work 
which we undertake, whether it is at our own initiative or in 
cooperation with outside agencies, by publishing the results of the 
investigations and by reserving the ownership thereof. 

The Engineering Experiment Station at the present time has 
an annual budget of approximately $60,000, but since it uses all of 
the facilities afforded by the College of Engineering, the annual 
expenditures for research work are considerably in excess of the 
regular budget. However, in view of the fact that our funds are 
apportioned to ten departments, the annual amount available for 
any one department is necessarily small. It is my hope that we 
may increase the number of cooperative investigations under the 
general direction of the Engineering Experiment Station, so that 



k 



58 PRESENT CONDITION OF RESEARCH IN UNITED STATES 

each member of our Faculty, who is a specialist in his particular 
line, may have an opportunity to direct a corps of special investi- 
gators on some important investigation, which will insure results 
of real value to the profession. It is only through cooperation of 
this sort that we can hope to imdertake the solution of some of the 
larger problems, for no educational institution can supply sufficient 
funds to conduct research work which involves large expense. If 
we can secure the cooperation of the industries of Illinois and of the 
adjoining states in such a program, I am sure that the work of the 
station can be greatly extended and improved. 

Charles H. Benjamin. The engineering experiment station 
at Purdue is but two years old. We now have a paid staflf of 
workers, and a good field in which to work and our future 
is very bright. Our object in establishing the station was primarily 
to coordinate the research work which has been carried on at 
Purdue for so many years, to enable us to do it more satisfactorily 
and to publish the results more widely. Another purpose in form- 
ing this organization in cooperation with the industries of the state 
is that primarily we are interested in Indiana, its products, its manu- 
factures and its future, and it is our first aim to collaborate with the 
men of Indiana to increase this productiveness. There is a certain 
advantage in a university laboratory. A university or engi- 
neering experiment station is commercially unprejudiced, and 
this removes from it any suspicion of commercial interest. Some- 
times we are required to investigate problems that perhaps could 
be better investigated by a private concern, but that establish- 
ment feels that its own efforts would be misinterpreted and that the 
results from the engineering experiment station will receive more 
universal credence and support. The organization and our methods 
of work are very similar to thojse which Dean Richards has so 
well outlined. Among the investigations under way may be men- 
tioned the testing of road materials and cof'jperation with the 
highway commissioners in the building of good roads, the efficiency 
of the carburetor, and the testing of farm tractors, in which 
we are cooperating with the agricultural school of the university. 
We expect to see a testing plant at the university which will bear 
somewhat the same relation to farm tractors as the locomotive 
laboratory does to the railroad, and our only difficulty is to take 
care of the great amount of work that it brings to us without 
solicitation. 



DISCUSSION 59 

In conclusion, I vnsh to express my appreciation of the work 
of the Research Committee. I do, however, want to urge caution, 
not on the part of the Committee, but on the part of some governing 
agencies in trying to shape research work. The- research man is 
a genius. He is bom, not made. He is peculiar and he must work 
in his own way and on his own initiative. He must not be inter- 
fered with and he must be allowed to follow a thing according to 
his bent. 

J. R. BiBBiNs commended the spirit of the Research Session 
and the character of the contributions produced by Professor 
Greene's Committee. There were, however, certain broad aspects 
of the subject, involving state and even national policy, which 
seemed to have been forgotten in the special interest centering on 
research problems. These he brought forcibly to mind by the slogan 
"Millions for Agriculture, not one cent for Industry — Why?" 

He then presented the following resolutions: 

Whereas, it is a matter of generally accepted concern that any 
advanced policy of The American Society op Mechanical Engi- 
neers, in common with other complementary learned societies and 
associations, should incorporate prompt and extended recognition of 
scientific and industrial research as a specific means of advancing 
technology and proper industrial development of the nation; and. 

Whereas, the Great War has fully demonstrated the principle 
of research, applied broadly as well as in detail, to be of inestimable 
benefit in advancing national welfare; and. 

Whereas, the Committee on Aims and Organizations has 
specifically recommended that The American Society op Me- 
chanical Engineers take a more active interest in this field ; there- 
fore, be it 

Resolved, That it be declared the sense of this Research Session 
of the A.S.M.E., that Engineering Coimcil shall be encouraged to 
undertake active support of a plan and organization of scientific 
and industrial research to the end that the following objects may be 
accomplished with all reasonable dispatch: 

1 To secure the passage in the present Congress of a special act 
furthering nation wide research in State units through congressional 
appropriations for this purpose under the general coordination of 
Engineering Council, or other national agency thoroughly represent- 
ative of the engineering profession. 



60 PBESEMT CONDITION OF BE6EABCH IN UNITfiD STATES 

2 To encourage trade, industrial, and utility associations to 
interest themselves in the advance of the Arts and the constructive 
benefits to be derived from research work in their respective fields 
and to cooperate* with them in their efforts in these directions. 

3 To encourage the various research institutions or instrumen- 
talities now or to be established by the Federal and State govern- 
ments in their close cooperation in this general research policy. 

4 To encourage and assist in the establishment of organised 
departments of engineering research at the various imiversities, 
adequately equipped with material and personnel and to bring 
such department as closely as possible in touch with the vital prob- 
lems of industrial development confronting the Nation. 

5 To institute organized publicity with the industries of the 
country and ascertain broadly by a thorough canvass their vital 
needs, with a view to directing the research work of tiie country 
and the cooperative development of the industries through the 
agency of the technical laboratories both public and private. 

6 To organize and support a separate department of the 
societies' activities, in close cooperation with similar departments 
of other technical societies, to act, through Engineering Council, or 
other representative national agency, as a permanent clearing house 
for all research work. 

In support of these resolutions, Mr. Bibbins pointed out that 
tiie Chicago local section's desire to ascertain what work could best 
be done by the local sections within the limited time and facili- 
ties at their command, emphasizing the necessity of publicity and 
that tlie Chicago Committee believed in the necessity of each state 
developing itself as a unit (too much could not be done in this 
direction), encouraging by every jwssible means every form of in- 
telligent and efficient research, both scientific and also industrial 
research of a more practical nature. There had been some dis- 
cussion of the propriety of the state undertaking activities of this 
character, especially with roferenre to allying itself thus with in- 
dustry. But in view of the high moral standing of the engineering 
profession, it seemed rather lar-fitrhed precaution to deny to the 
people of the state opportunity of thus dt'veloping their industries 
to the inaxiniiun extent. 

Following Mr. Bihhins' presentation of the resolution offered 
by the Chicago Comniittt'c, there was u general discussion which 
centered around the quest iiHi of j^oviTumental supervision of 
research. Wm. T. Magruder cited the need of state research work 



DISCUSSION 61 

in using Ohio coals in gas' production to supplement and replace 
the fast-disappearing natural gas supply; considerable work along 
this line had been done already in Illinois. The resolution was 
also discussed by Dean A. A. Potter, who stated that while he be- 
lieved the Society should encourage research, and while it is un- 
questionably the purpose of the Society to urge governmental 
recognition, it would nevertheless be a mistake to approve a reso- 
lution asking Congress to appropriate a certain amount of money 
to support research in certain types of institutions. He believed 
that The American Society of Mechanical Engineers should 
express its approval of all kinds of scientific and industrial research, 
but should not approve any specific bill. 

J. R. Bibbins emphasized the point that the Chicago Commit- 
tee had no desire to complicate proceedings by attempting to classify 
institutions; it only favored the broad application of the principle, 
through publicity and national recognition of research by the 
passage of supporting legislation. The detail could be worked out 
by administrative authorities, possibly with the help of Engineering 
Coimcil. 

Doctor Mees, Director of the Eastman Kodak Research Labo- 
ratory, stated that he was in accord with Dean Potter's remarks 
and did not believe in urging Federal research. It was a question 
whether Federal research was desirable. The leaders of industrv 
were against it, and the ad\4ce that he would give to the Society — 
he was not a member of it — was to pay for its own research. 

Several members of the Society who favored financial assist- 
ance on the part of the Government next presented their views 
and urged that the Society endorse any measure which would 
obtain an appropriation from the National Government for the 
benefit of such institutions as might be selected. 

C. H. Bierbaum also discussed the resolution. A bill of the 
kind proposed was proper enough from a theoretical point of view, 
but from a practical point of view he did not believe anything 
worse could be done than to have Congress pass a bill for Federal 
research, for there would then be a most inefficient distribution of 
funds. The securing of funds, however, for doing this work was, 
he believed, the smallest problem and the securing of men to do the 
work, the greatest. He was decidedly in favor of private or semi- 
private research work, but not for Federal control. 

Albert Kingsbury urged a careful consideration of the resolu- 
tion and suggested that the report of the Chicago Committee be 



62 PRESENT CONDITION OF RESEARCH IN UNITSD STATES 

brought to the attention of all memberB of the Society through 
suitable means of publication. 

A substitute motion to the effect that the Society should not 
favor federal research was not seconded. The resolutions were 
then voted upon by individual items and carried. 



DISCUSSION 

Address on Organization op the Division of Engineering 

OF THE National Research Council 

A T the Research Session of the Spring Meeting, Mr. Galen H. 
Clevenger, Vice-Chairman of the Division of Engineering of 
the National Research Council, was invited to give an address on 
the organization work of the Council and of the Division of En- 
gineering with which he is directly connected at present, in charge 
of the headquarters of the Division of Engineering in the Engineer- 
ing Societies Building, New York. Mr. Clevenger said in part: 

Before discussing the work of the Engineering Division, it will 
be well to briefly review the history of the National Academy of 
Sciences and the National Research Council, pointing out the 
close relationship which exists between the two bodies. Early in 
1863 after the Civil War had been in progress many months and 
serious and unexpected reverses had occurred, a number of leaders 
in science and engineering, recognizing clearly the need for a 
national organization embodying the whole range of science to 
advise the Government on scientific questions in connection with 
the conduct of the war, planned the National Academy of Sciences 
Eind through their efforts a bill incorporating the Academy was 
introduced in the Senate on February 21, 1863, and after passing 
both the Senate and the House was signed by President Lincoln 
on March 3 of the same year. Later the original bill was amended 
to remove the Umitation on membership and to enable the Academy 
to receive bequests. 

At this time, with but few exceptions, the now well-organized and 
powerful scientific and technical bureaus of the Government were 
aot in existence, so that the new organization at once became and 
continued to be of great assistance to the War and Navy Depart- 
ments throughout the rest of the Civil War. 

During the years intervening between the close of the Civil War 
md the beginning of the war with Germany the Academy has on 



Preaented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Societt of Mechanical Engineers. 

63 



64 ORGANIZATION OF THE DIVISION OF ENGINEBBING 

many occasions advised the Government in scientific matters, the 
most notable case being perhaps in connection with certain problems 
arising in the construction of the Panama Canal. 

WAR ORGANIZATION OF THE NATIONAL RESEARCH COUNCIL 

In April 1916, following the attack upon the Sussex^ President 
Wilson called upon the Academy to aid in organizing the scientific 
and engineering forces of the United States in the defense of the 
country. The Academy accordingly again turned to the work for 
which it was originally organized and which it performed so well 
during the Civil War. 

In order to perform most effectively the greatly enlarged service 
possible under these new conditions, the organizing committee of 
the Academy after careful consideration recommended that a new 
body be formed, which would be so organized as to take every ad- 
vantage of the very Uberal charter of the Academy. Such an 
organization would therefore share fully in all the privil^es of the 
Academy both at home and abroad and would have the further 
advantage of permitting of the widest range of freedom in the selec- 
tion of its membership. 

In accordance with this recommendation the National Research 
Council, comprising the chiefs of the technical bureaus of the Army 
and Navy, the heads of the ci\dlian bureaus of the Government 
engaged in scientific research and engineering, investigators repre- 
senting the educational institutions, research foundations, and repre- 
sentatives of industrial engineering research, was formed by the 
Academy with the active cooperation of the leading scientific and 
technical societies of the country. 

Under date of July 24, 1916, President Wilson addressed a letter 
to the President of the National Academy of Sciences expressiiig 
his approval of the preliminary report covering the newly formed 
National Research Council and promising his support to the move- 
ment. 

On Februar}' 28, 1917, the Council of National Defense passed 
a resolution asking the National Research (^ouncil to codperate with 
it in niattors of research for national defense, and soon after 
the Research Council was requested to act as the department of 
science and research of the Coimcil of National Defense, its par- 
ticular function bcin^ the organization of investigations on military 
and technical problems. 



Q. H. CLEYENGEB 65 

In July 1917 the National Research Council was requested by 
the Chief Signal OflBcer to organize a Division of Science and Re- 
search of the Signal Corps. 

The war organization of the National Research Council con- 
sisted of eight divisions in addition to the Research Information 
Service; namely, 

1 Division of General Relations 

2 Military Division 

3 Division of Engineering 

4 Division of Physics, Mathematics, Astronomy, and Geo- 

ph3rsics 

5 Division of Chemistry and Chemical Technology 

6 Division of Geology and Geography 

7 Division of Medicine and Related Sciences 

8 Division of Agriculture, Botany, Forestry, Zoology, and 

Fisheries. 

The officers of the Council consisted of a chairman, three vice- 
chairmen, a treasurer, an executive secretary and two assistant 
secretaries. The organization of the Divisions differed somewhat, 
but in general each ha<l a chairman, a vice-chairman and an execu- 
tive committee. 

WAR ORGANIZATION AND WORK OF THE DIVISION OF 

ENGINEERING 

The Division of Engineering at this time comprised four sections: 
a Section on Metallurgy, a Section on Mechanical Engineering, a 
Section on Electrical Engineering, and a Section on Prime Movers. 
The work of each section was imder a chairman, who was directly 
responsible to the chairman of the Division. 

The Section on Metallurgy had for its principal work the solving 
of metallurgical problems arising in connection with the conduct of 
the war, more particularly those brought to it by the military. This 
work was accomplished through the medium of committees, whose 
personnel included leading authorities upon metallurgy. 

The Section on Mechanical Engineering established a drafting 
room in charge of a chief draftsman at Research Council head- 
quarters, and through the generosity of the Carnegie Institute of 
Technology, a machine shop at Pittsburgh under the direction of a 
foreman. These were used for the development of inventions 
referred to the Section by the Divisions of Engineering and Physics. 



66 ORGANIZATION OF THE DIVISION OF ENGINEEBINQ 

The Section on Electrical Engineering concentrated its eflforts 
upon the problem of electric welding, more especially electric weld- 
ing as applied to shipbuilding. This Section worked in very close 
cooperation with the Emergency Fleet Corporation, who financed 
its investigative work. 

The Section on Prime Movers devoted its attention chiefly to 
the design and development of power plants for aircraft. 

The eflforts of each section were so directed as to be of the great- 
est service in the solving of the problems of greatest immediate need 
to winning the war; each has to its credit important achievements 
during the war period. (See Report of the Academy of Sciences for 
the Year 1918.) 



PRESENT ORGANIZATIONS AND AFFILIATIONS 

Permanent Organization of the National Research Council. Under 
date of May 11, 1918, President Wilson issued an executive order 
asking that the National Research Council be perpetuated, and 
in accordance with this request the permanent organization of the 
Council has been rapidly accomplished. 

The membership of the Council consists of: 

1 Representatives of national scientific and technical societies 

2 Representatives of the Government, as provided in the 

Executive Order 

3 Representatives of other research organizations and other 

persons whose aid may advance the objects of the 
Council. 

The officers of the National Research Council are a chairman, 
one or more vice-chairmen, a secretary and a treasurer. 
The Council is organized in Divisions of two classes: 

A Divisions dealing with the more general relations and 

activities of the Council 
B Divisions dealing with ri'lated branches of science and 

technology. 

A Divisions of Cencral Relations are: 

1 (lovcriinuMit Divi>ion 
11 Division o{ FonMiin Relations 
HI Division of States lU'lations 



Q. H. CLBVENGEB 67 

IV Division of Educational Relations 
V Division of Industrial Relations 
VI Research Information Service. 

B Divisions of Science and Technology are: 

VII Division of Physical Sciences 
VIII Division of Engineering 
IX Division of Chemistry and Chemical Technology 
X Division of Geology and Geography 
XI Division of Medical Sciences 
XII Division of Biology and Agriculture 
XIII Division of Anthropology and Psychology. 

The aflfairs of each Division are administered by a chairman, a 
vice-chairman, and an executive conmiittee, who are elected annually 
by the Division and confirmed by the Executive Board of the Council. 

The purpose of the National Research Council is to promote re- 
search in the mathematical, phjrsical, and biological sciences, and 
in the appUcation of these sciences to engineering, agriculture, 
medicine, and other useful arts, with the object of increasing knowl- 
edge, of strengthening the national defense, and of contributing in 
other ways to the pubUc welfare. 

AflSliation with similar organizations abroad is rapidly bringing 
about an International Research Council. 

The Division of Engineering consists of three representatives of 
each of the four founder engineering societies, the societies so repre- 
sented being The American Society of Mechanical Engineers, the 
American Institute of Electrical Engineers, the American Institute 
of Mining Engineers, and the American Society of Civil Engineers; 
further, there is one representative each from the foiu* more impor- 
tant non-founder societies, the societies so represented being the 
American Society for Testing Materials, the American Society of 
Illuminating Engineers, the Western Society of Engineers, and the 
Society of Automotive Engineers. In addition to the representatives 
of the engineering societies there are twelve members at large, making 
a total membership in the Division of twenty-eight. Eight members 
of the Division are also members of The Engineering Foundation. 

The work of the Engineering Division has gone steadily forward 
during the reorganization period and to such an extent that the 
newly organized Division is now in full operation. 

The Engineering Foundation and the Division of Engineering. 
The Engineering Foundation has from the beginning taken a very 



68 ORGANIZATION OF THE DIVISION OF ENGINEERING 

active and important part in furthering the work of the whole Coun- 
cil; indeed, in the earlier stages of the war organization, when the 
funds available for carrying on its work were very limited, The Engi- 
neering Foundation gave the services of its secretary and substan- 
tially its whole income to the support of the Council, this arrangement 
continuing until support was secured from other sources. 

Recently a plan of close affiliation of The Engineering Founda- 
tion and the Division of Engineering has been approved by the mem- 
bers of these bodies and also by the Executive Board of the Council. 
In compliance with the terms of this agreement The Engineering 
Foundation has provided the Division of Engineering with an office 
in the Engineering Societies Building at New York, together with 
necessary clerical a.ssistance. They further have agreed to make 
appropriations of their funds to aid specific undertakings of the 
Division from time to time as may be later determined, and in fact 
at the present time an arrangement has been effected whereby 
The Engineering Foundation undertakes the financial support 
of the work of the Committee on Fatigue Phenomena. 

The Division of Engineering is not to be regarded as an instru- 
ment of research, but rather as a stimulator and coordinator of re- 
search. Its principal object is to get more and better research done 
in engineering, carefully avoiding the position of being a dictator, 
or of assuming credit for work which it has encouraged others to do. 
The Division of Engineering now has nineteen committees work- 
ing upon a variety of subjects. These are in various stages of 
organization. Every effort is being made to take up researches of 
broad general interest. At present twentj^-one states extending 
from the Atlantic to the Pacific are represented on these com- 
mittees and the number is rapidly increasing. 

The National Research Council, like the National Academy of 
Sciences, was brought into being during a ix'riod of war and although 
growing out of a war-time need its utility in time of peace is now 
fully dcMiionstrated. Although it received aid from the Government 
during the war it is not a Government bureau, and in the future it 
will b(^ supported by private endowment. 

The National Research Council as a federation of research in- 
ter(\sts of the United States covering the whole field of pure and 
applied scic^nce, and with the effective coojxTation iK)ssible with 
foreign investigators through the International R(»si»arch Council, 
is in a ix)sition through its Division of Engin(»(»ring to jx'rform a 
most valuable service in furthering Engineering Research. 



No. 1689 

THE ORGANIZATION AND CONDUCT OF AN 

INDUSTRIAL LABORATORY 

Bt a. D. Little, Cambridge, Mass. 
Member of the Society 

and 

H. E. HowE^ Cambrioqe, Mass. 
Non-Member 

During the war indvMrial research in the United States was naturally stimu- 
UUedf and as a result there now exists a deeper interest than heretofore in the appli- 
cations of science to manufacturing processes. New laboratories will undoubtedly 
he built and many old ones reorganized in order to render more efficient service. It 
is the purpose of this paper to point out the organization and conduct of such a re- 
search laboratory. 

The authors first outline the aims of a research organization^ following which the 
divisions of the laboratory are enumerated and discussed, the laboratories of Arthur 
D. LittlCf Inc.f being taken as a type. The methods of management, writing of re- 
ports and the commercial organization of the laboratory are also discussed at some 
length, and the paper concludes with a description of the building and equipment 
best suited to carry on this type of work. 

T>REVIOUS to the war there were about 375 industrial research 
laboratories in the United States, including those maintained 
by manufacturers for their own benefit, and commercial laboratories 
prepared to render similar service, continuously or intermittently 
to the establishments without such faciUties. At the present time 
there are no figures available regarding the number of new labora- 
tories established as a result of the war, but there is no doubt but 
that the war created a deeper interest in industrial research, and 
the application of science to manufacturing processes. It is also 
evident that those laboratories which existed before the war are 
displa3dng a greater interest in fundamental research, and in rehabi- 
litating their organizations are paying far more attention to the 
research phases of their problems than they have been willing to do 

^ With Arthur D. Little, Inc. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Societt of Mechanical Engineers. 
For discussion see p. 51. 

69 



70 INDUSTRIAL UkBOBATORY OBGANUATION 

heretofore. This, then, is an opportune time to discuss the organiza- 
tion and conduct of an industrial laboratory and it is hoped that 
those interested in the establishment of new laboratories, or the 
expansion of old ones, may find something in this paper to assist 
them. 

2 In discussing the industrial laboratory we may choose be- 
tween the one organized for the purpose of exploring some small 
comer of the broad field s\Tnbolic of our ignorance and an establish- 
ment concerned with the greatest variety of problems. 

3 A laboratory of the latter type should consist of a collection 
of special laboratories carefully articulated to produce results most 
efficiently, and the work common to all of them should be organized 
separately in a large general laboratory. 

4 Fortunately for our countrj- there are several such labora- 
tories doing splendid work, and notwithstanding the care exercised 
to avoid undue specialization, nearly all of them contain depart- 
ments which dominate, due either to stronger men or the greater 
appeal which these departments make to the company; or perhaps 
to a seemingly greater importance of their class of problems at 
the moment. The great majority of these laboratories are main- 
tained in the plants of industry at an annual expense running up to 
two millions in at least one case, and with many spending hundreds 
of thousands each year. 

5 Another plan which should be mentioned involves the train- 
ing of men as a primary consideration, and the Mellon Institute, 
at Pittsburgh, affords a conspicuously successful example of what 
may be done in educational institutions in solving the problems 
of industr}', while at the same time men are trained in research. 

TUB AIMS OF A RESEARCH ORGANIZATION 

6 Broadly stated, tlic aims of a research organization should be: 
a To find, develop and train men 

b To create such a background in the public mind as shall 
insure support for research and the industrial utilization 
of research results 

c To secure coojx? ration l)etween different branches of 
science, as, for example, l>etwcen chemists and mathe- 
maticians. (The fortuitous combination of the mathe- 
matical mind with the viewpoint of the chemist in Wil- 
lard Gibbs laid the basis for physical chemistry. But 
such a combination in a single individual is ver>' rare.) 



A. D. LITTLE AND H. E. HOWE 71 

d To avoid repetition and duplication of effort, first by ren- 
dering present knowledge readily available to research 
workers, second by appljdng clearing-house methods to 
research projects 

e To stimulate research by emphasizing the importance of 
specific problems, making special grants, rendering ma- 
terial and faciUties as generally available as possible 

/ To fiunish a general staff for research which shall work 
out the plan of attack for major problems, assign the 
several lines to competent workers and coordinate and 
focus the whole 

g To bring home to manufacturers the advantages of re- 
search with the view of promoting the establishment of 
private, corporation, and group laboratories 

h To make and publish a census of available research fa- 
cilities in men and equipment 

i To survey the natural resoiu-ces of the nation and direct 
research toward their development 
To appraise oiu* great industrial wastes and develop plans 
and methods for turning them to profitable use 

7 As regards any research laboratory, it goes without saying 
that it is the personal factor which determines performance and this 
is preeminently true of the laboratory director. Sir Humphrey Davy 
truly said that his greatest discovery was Michael Faraday, and no 
greater problem is likely to confront a research laboratory than that 
involved in the discovery of a director. Successful laboratory di- 
rectors may be of several types, but a miUtant optimism, contagious 
enthusiasm, controlled imagination and quick human sympathy are 
conunon to them all. Such a man will naturally in selecting his 
subordinates look for these personal quaUties almost as carefully as he 
will weigh specialized scientific trainmg, and having been thus guided 
in his selections will find it relatively easy to inspire throughout his or- 
ganization those relations of good fellowship and that esprit de corps 
which multiply enormously the effectiveness of any working force. 

8 Exceptional men are hard to find simply because they are 
exceptional, and the director in laying out the work of the labora- 
tory and extending its personnel will endeavor to augment the out- 
put of the exceptional man through the coordinated effort of properly 
directed men of secondary capacity. Fairness in apportioning credit, 
frequent conferences, and opportunity for self-development are also 
essential to the attainment of high efficiency. 



72 INDUSTRIAL LABORATORY ORGANIZATION 

9 The so-called commercial laboratory devoting its efforts to 
industrial research and operated on a strictly business basis will 
best serve our present purpose, and that of Arthur D. Little, Inc., 
in Cambridge, Mass., will be taken as a type in the belief that much 
of interest will be found in this establishment, which is "dedicated to 
industrial progress." During the past thirty-three years this labo- 
ratory has grown from a partnership of two chemists to an organiza- 
tion of sixty people, and scheme after scheme has been devised for 
the management of the enterprise, only to find new conditions and 
rapid growth calling for constant revision. Being a corporation, it 
is managed by the usual officers with a board of directors, all of whom 
do not devote their entire time to the business. 



THE DIVISIONS OF THE LABORATORY 

10 Within such a laboratory there are two distinct sets of 
duties which may be designated as scientific or technical, and com- 
mercial or financial. These two divisions have at least two points 
in contact, one being through a service manager, and the other the 
departniont charged with obtaining new business for the organization. 

11 C onsidering first the purely scientific side of the commercial 
laboratory, its fundamental duty is to interpret the results of pure 
science in the terms of industrv. While the work of the commercial 
laboratory is of the same order as that done in any laboratory even 
where the dollar is nev(?r discussed, it nnist be conducted with full 
recognition of the fact that many industrial i)roblems are as inti- 
mately concerned with ec<»noniic (luestions as with scientific. In 
other words, while for instance a laboratory process in glass may 
be intensely interesting and of fun<laniental inii)ortance, the client 
can hardly be expected to be satisfied with a report unless a commer- 
cial method for operating it can be devised. The technical work 
should be in charge of the* presid(»nt, under whom various depart- 
ments should ])e organized, so that each pha.^e c)f a given problem 
may have the att(*ntion of a specialist, provided with adequate 
equipment to facilitate the work. 

12 In this connection it may be emphasized that it pays to 
provide congenial, inspiring surroundings for the la!)oratory worker. 
The laboratoiT can l)e made attractive without being ornate, or 
involving unreasonabh* (»xpense, and every elTort should Ic nunle 
to have the workers rea.M)nably happy. I'nder no other condition 
can the best work be expected, and it must be n'memlMTed tlial 



A. D. LITTLE AND H. E. HOWE 73 

the heaviest investment is in the time of these workers, the salary 
cost being much greater than that for equipment or material main- 
tenance. Rewards other than monetary for faithful service also 
play an important part. 

13 The departments into which the technical division are divided 
will naturally differ in each laboratory, but a fairly definite line 
can be drawn between research, engineering and standardized or 
routine work. It is advantageous to have all of the standardized 
work, including that incident to research and engineering, carried 
on under one department head, for in this way it can be done to 
better advantage both as regards efficiency and economy. There 
will be many occasions when individuals from each of the depart- 
ments will confer on special problems. 

14 The research department should be organized for both 
laboratory and small-factory-scale work. There will be a multi- 
pUcity of subjects, and since special faciUties cannot be provided in 
advance of close acquaintance with the problem, the organization 
of departments for research along special lines will concern personnel 
more than a division of floor space or equipment. 

15 Engineering will embody plant inspection, design, construc- 
tion and operation, as much of its work will be in the field, al- 
though many phases of its problems will be worked upon concur- 
rently in the laboratory by those departments best suited to handle 
the work. 

16 The analytical department will be subdivided under such 
headings as textiles, fuels, food, metallurgy, and metallography, 
chemical microscopy, water, lubricants, construction materials, 
pulp and paper, fermentology, etc. Some of these subjects will 
require special acconunodations, while others can share a large 
laboratory which provides space for certain apparatus kept in place 
for a large number of similar determinations. 

17 Nothing is more expensive or demoralizing than experi- 
mentation in the plant. An industrial research laboratory should, 
therefore, be adequately provided with equipment of semi-conmiercial 
size. Infant mortality among processes is high in any case and the 
most critical period in their young Uves is that covering the transi- 
tion from the laboratory to the plant. They require and the research 
laboratory should provide a nursery to protect and foster them 
during this period of their development. Some large manufacturers 
have even found it desirable to operate in connection with and \mder 
the sole direction of their research laboratory a small plant in which 



74 INDUSTRIAL LABOKATOBT OBGANIZATION 

actual commercial manufacture is regularly conducted. Such ex- 
tension of the laboratory's fimction permits the complete reduction 
to practice of new methods and the commercial demonstration of 
the suflSciency of the product before the innovations are introduced 
into the main plant. 

18 Even when no such provision appears feasible, it is, never- 
theless, highly desirable to have the industrial research laboratory 
actually engaged in some small scale, highly specialized, commer- 
cial manufacture, preferably of some product which it has itself 
originated. The least advantage of this procedure is that such 
manufacture of a properly selected product may frequently defray a 
substantial proportion of the expenses of the laboratory. The major 
benefits are the acquirement of a certain commercial sense by the 
laboratory staff, an appreciation of the conditions and difficulties 
of actual production, and finally the strengthening of the position 
of the laboratory through the increase in its turnover and equipment. 
Such procedure, while perhaps not general, has been followed to 
great advantage by the research laboratories of the General Mectric 
Company, the Eastman Kodak Company and Arthur D. Little, 
Inc. 

19 It is easy to visualize the organization chart for such a 
laboratory, and a brief description of how a new piece of work will 
be handled may convey a still better idea of the method of man- 
agement. 

THE METHOD OF MANAGEMENT 

20 The authorization for the work will go to the service manager, 
who sees all incoming mail, and to the authorization will be attached 
any correspondence or data bearing on the case, all of which will be 
given a case number for identification, and this number will be en- 
tered in a case register, which will indicate the name of the cUent, 
the subject of the problem, the date the authorization is received and 
the date when the work shall have been completed. The service 
manager, who must be familiar with the ability of each member of 
the staff, as well as with the work in hand, will assign the case to the 
division which can render the best service. Conferences will then 
be called, into which any member of the organization who can con- 
tribute anything to the solution of the problem in hand will be 
drawn, and outside associates or independent consultants may be 
included. The problem will then go into work by means of instruc- 
tion sheets, setting forth what is to be accomplished, suggesting 



A. D. LITTliE AND H. E. HOWE 75 

methods of attack, relating any special circumstances, references 
to literature, and standard methods which may be appUcable, and 
as much light as possible given to the individual who is to do the 
work. Accompan3ang the case there will be a tag bearing the case 
number, upon which a date at which it is expected the work can be 
completed, or a progress report made, must be indicated. The tag 
is then returned to the service manager. Through the means of data 
sheets, time sUps and verbal reports the progress of the problem 
will be readily followed. This procedure will be followed in all the 
divisions, the individual reporting to his superior, and the service 
manager will be alert to insure prompt and eflScient service to all 
cUents. 

21 At the completion of the work the report to the cUent vary- 
ing in extent from a single printed form, upon which the results of 
analysis may be set down, to a bound volume of several hundred 
pages, will pass through the hands of all concerned, and will thus 
be distinctly the report of the organization and not of an individual 
in the organization. 

REPORTS 

22 Report writing requires skill, for it must be comprehensive 
and complete without padding. It should begin with a clear state- 
ment of the problem, followed by the conclusion reached as a result 
of the work, which may then be described in detail. Patents, cost 
data, tables, graphs, photographs and samples should be dealt with 
in an appendix, and in some instances descriptions of apparatus 
should be included. The whole must be carefully indexed, and a 
copy sent to the Ubrary to be bound and kept as confidential infor- 
mation in locked cases, but as part of the Ubrary it should be carded 
for the Ubrary card index. Obviously no fast rule can be laid down 
for writing reports, but it should be borne in mind that many of 
those who read technical reports are not interested in minute details, 
and that the subject-matter must be presented in a form that will 
be interesting and understood by the layman. It must also have 
its important points so emphasized that they can be readily picked 
out by those not caring to read the entire report, but at the same time 
it should include suflScient data to serve the purpose of a fuUy quaU- 
fied technical man to whom the report may be referred at some 
later time. 

23 This brings up the question of the Ubrary, which may easily 
be considered the backbone of the industrial laboratory. Its extent 



76 INDUSTRIAL LABORATOBY ORGANIZATION 

will depend upon other library material available in the (XMnmunity, 
but there are few things which obstruct research more seriously than 
the absence of easily accessible proper library facilities. A few dollars 
spent in books and literature frequently saves as many hundreds 
otherwise spent in work of dupUcation. The useful periodicals must 
be pro\'ided, elaborate indexes will be found a good investment, 
also abstracts and patents; in short, every means for quickly locating 
literature references should be at hand. The current literature, 
with articles of interest indicated on an attached slip, should be 
circulated among the members of the staff whose names are checked 
on this slip, and some one, preferably a chemist, should have assigned 
to him the task of constantly reading the literature n order that no 
scrap of informat'on shall escape. Such a chemist-librarian wiU 
conduct searches in other libraries, prepare abstracts, and in fact 
direct the information service for the laboratory, and through the 
laboratory- to its clients. This will frequently mean recasting a 
scientific article in order to make it of practical value to the works 
manager or superintendent. 

THE COMMERCIAL ORGANIZATION 

24 ( oiicerning the conmiercial di\'ision of the organization, the 
cas(; we have described will have originated through the recom- 
riiondation of a satisfied client, through magazine advertising, the 
literature of the laboratory, lectures and informal talks, direct 
appeals, traveling representatives, or through some similar channel. 
The publicity of the laboratory' will be in the hands of the commer- 
cial department responsible for securing new business, and answerable 
to the president so far as technical matters are concerned, and to 
i\w, treasurer on the financial pluises of a prosjx^ctive problem. The 
(^onirnercial department will answer inquiries, will present attrac- 
tive problems for industrial research, and endeavor to obtain au- 
thorization from manufacturers for work contemplated along lines 
which are frequently made the subject of a conference. When the 
authorization is secured through |x^rsonal interviews, by correspond- 
en(!(», and frcMpiently by telegraph, the commercial division records 
b(»come a part of the case, and nothing more is to l)e done by the 
division until after the final report. Then it is quite reasonable to 
make work satisfactorily completed the Ix^st recommendation for 
attack upon some new problem. 

25 The commercial department will maintain a- carefully classi- 



A. D. LITTLE AND H. E. HOWE 77 

fied mailing Ust, a follow-up system on prospects, and its own letter 
nies. The treasurer, who has at hand data as to the cost of each 
man's work, will be consulted before any proposals involving money 
are made, and the financial division, under the treasurer, will attend 
to aU matters of time cost on work, determine the overhead, the 
items of maintenance, depreciation, etc., all of which go into the 
total cost of doing business, something of which the average indus- 
trial laboratory remains in almost complete ignorance. By means 
of records constantly kept up to date, and revised in the Ught of 
experience, the treasurer will know accurately the percentage which 
must be added to the salary of each individual in order to arrive at 
the cost of his work, and will soon be able to rate the men according 
to their capacity and estimate something of what a particular under- 
taking should cost. This will be of great assistance in making sug- 
gestions regarding appropriations for prospective work, and also 
which members of the stafiF are of the greatest service to the organ- 
ization. Costs will be determined with the help of time sUps, which 
each individual will submit daily, indicating the cUent's name, the 
case number, the amount of time spent and how it was spent, that 
is, whether in the laboratory, in the Ubrary, traveling, etc. Another 
function of the treasurer's division will be the management of indus- 
trial enterprises where such ser\'ice is desired, together wdth tech- 
nical supervision or assistance. 



BUILDINGS AND EQUIPMENT 

26 With this general plan of management before us, we may 
now consider the equipment and space required for effective work. 
EiXperience has shown that a satisfactory building is one approxi- 
mately 50 ft. by 150 ft. (planned so that another building may be 
easily connected in H-formation) and consisting of a basement, 
with three floors, the basement itself being so designed that it is as 
light and airy as the upper floors. Such a building with a wing hous- 
ing the power plant and small grinding rooms (this being the con- 
nected wing of the H) has a total floor space of approximately 
30,000 sq. ft. The building occupied by Arthur D. Little, Inc., is 
of this type and the various departments are allotted the following 
space: Anal3rtical division, 5328 sq. ft.; research di\ision, 3458 
sq.ft.; engineering, 1 155 sq. ft.; commercial department, 768 sq. ft.; 
management (meaning the management of outside enterprises), 538 



78 INDUSTRIAL LABORATORY ORGANIZATION 

sq. ft.; special department devoted to pulp and paper, 4118 sq. ft. 
The portions of the building which are non-producing, such as 
miscellaneous offices, which includes stairs, corridors, halls, lava- 
tories, etc., comprise 8000 sq. ft., and the space unassigned amounts 
to 3815 sq. ft. This includes laboratory space provided for emer- 
gencies and expansion but not in constant use. 

27 These area measurements form the basis of apportioning 
overhead due to interest on investment, depreciation, repairs, insur- 
ance, upkeep, etc., and among the departments is distributed the 
charge for carrying 768 sq. ft. devoted to an industrial museum and 
1536 sq. ft. devoted to the library. 

28 The basement will provide room for the power plant, the 
current sample room, the general stock room, two very large rooms 
for small-factory-scale equipment, two small rooms for coal and 
other crushing and grinding operations, and a machine shop in which 
the physical testing machinery can be installed. A laboratory for 
testing construction materials, such as cement, may also be placed 
in the basement to advantage, and here, too, a room and vault can 
be set aside for inactive letter files and records. 

29 The first floor will provide a series of offices each of about 
250 sq. ft. area, two larger ones which may be used by officers of the 
company, consulting engineers and others, a reception room, and an 
information booth with switchl)oard. The museum can also be on 
the first floor, together with the rooms devoted to the commercial 
department's work. Quarters for the financial division, with ample 
vault space, and room for current correspondence and the general 
stenographers* office complete the floor. The engineering division 
might also occupy rooms on the first floor. 

30 The second floor may be properly devoted to research and 
the library. It is frequently advantageous to be able to segr^ate 
research problems, and it will be well to provide a series of small 
rooms, say of approximately 250 sq. ft. in area, which can be fitted 
up in ac(;ordance with the requirements of the problem, and easily 
dismantled at the conclusion of the work, to l)e refitted according 
to the next undertaking. A branch stock room should l)e located 
on the second as well as the third floor, and these should be served 
by elevator from the general storeroom in the basement. There 
are always a numlx^r of scattering problems in research that can be 
handled in one lalK)ratory, and so a special-problems laboratory 
should also be provided on the second floor with an office for con- 
sultation purposes. Finally a large room, say of 1500 sq. ft., should 



D. lilTTLa AHD H. E. HOWE 




BAfatHKNT AND FiBBT-Fioos Pl&ns or THs Laboratobt or 
Abthcr D. Ijttlk, Inc. 



80 INDUSTRIAL LABORATORY ORGANIZATION 

also be available for large undertakings, and as a space for emer- 
gency overflow. 

31 The third floor may comprise the general analytical labora- 
tory, a room of about 1500 sq. ft., adjacent to which should be a 
room for titration and the balance room (each of approximately 
125 sq. ft.), a fuel-testing room and a special room for extraction. 
A branch stock room and the offices of the head chemist and assist- 
ants should also be on this floor. The optical room should be 
placed where north light can be obtained, and a small dark room 
must be provided, as well as a specially equipped room for the 
physical testing of paper and textiles. The kitchen can adjoin the 
assembly room. A locker room for the men and a rest room for 
the women must also be provided. Space under the roof can be 
used for fans and a ventilating appliance, water tanks, etc. 

32 Such a building as has been thus briefly described cost about 
$200,000, 1917, and $50,000 will provide genera^ equipment. There 
is always something new to buy for a laboratory. Such an estab- 
lishment will provide working space for approximately 150 people, 
and more could be accommodated if necessary, depending largely 
on the type of work being conducted. The cost of operation and 
maintenance, based on a staff of sixt}-, w^U be about $20,000 per 
month. 

33 While the tyix* of building described may not appeal in all 
cases, it is certainly true that better work is done in a separate build- 
ing provided for the purpose than if an attempt is made to remodel 
an old building or provide space in some existing building. Labora- 
tories started in this way are c(Mtain to develop, if at all successful, 
and moving is very expensive, princi[)ally because it interferes 
with the efficienc}' and production of the establishment, which is a 
direct loss due to the high salary cost. The permanent investment 
in a laboratory building and equipment is smaller than the salary 
investment, and any enterprise of this sort should be planned witb 
reference to the fact that it is the men's time and accumulated knowl- 
edge that form the stock in trade. 

34 Not only does the* success of an industrial research labo- 
ratory dei)end upon its (Mjuipment and environment, but far more 
defininitely upon the capacity and ability of the director, and 
his nnmediate associates. The constant problem is to obtain 
hearty cooperation without over-organization and without in any 
way danqxining the enthusiasm of the individual. After all, the 
greatest progress in science has l)een made through individual 



A. D. UTTLE AND I 



cEfort, and it is the function of the industrial research laboratory 
and its organization to provide and maintain conditions under 



SECOND FLOOR PLAN 



THIRD FLOOR Pit 



8econi>- and Third-Floor Plans of the Laboratort op 
Artbok D. Little, Inc. 

which such individual effort will achieve the greatest results, 
utilizing cooperation as far as circumstances in each instance 
may warrant. 



No. 1690 

INDUSTRIAL RESEARCH LABORATORY 

ORGANIZATION 

By C. E. K. Mees/ Rochester, N. Y. 

Non-Member 

The great value of scientific research, both to the industries and the nation at 
large, is now generally recognized. The industrial research laboratory is an im- 
portant JacUjT in maintaining the supremacy of an industry, and its siuxess depends 
to a considerabls degree upon its relation to the other departments of the company 
with which it is associated. In this paper these statements are discussed, arid the 
author presents his views regarding the establishment and function of industrial 
laboratories, giving in connection with the latter three annular diagrams. The 
form and operating costs of industrial laboratories are also discussed. 

THE triumphs which have ah-eady been won by research labora- 
tories are common knowledge. The incandescent-lamp in- 
dustry, for instance, originated in the United States with the 
carbon-filament lamp, but was nearly lost to this country when the 
tungsten-filament was developed, only to be rescued from that 
danger by the research laboratory of the General Electric Company, 
who fought for the prize in sight and developed first the drawn- 
wire filament and then the nitrogen lamp; and we may be sure 
that if the theoretical and practical work of the research laboratory 
of the General Electric Company were not kept up, the American 
manufacturers could by no means rest secure in their industry, as, 
undoubtedly, later developments in electric lighting will come and 
the industry might be transferred, in part if not completely, to the 
originators of any improvement. Manufactiu'ing concerns and 
especially the powerful, well-organized companies who are the 
leaders of industry can, of course, retain their leadership for a num- 
ber of years against smaller and less completely organized com- 
petitors, but eventually they can insure their position only by having 
in their employ men who are competent to keep in touch with and 

1 Director of the Research Laboratory of Eastman Kodak Company, 
Rochester, N. Y. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 

AlfEBICAN SoCnSTT OF MECHANICAL ENGINEERS. 

For discusBion see p. 51. 

83 



84 INDUSTRIAL RESEARCH LABORATORY ORGANIZATION 

to advance the subject, and the maintenance of a laboratory staffed 
by such men is a final insurance against eventual loss of the control 
of its industry by any concern. 

2 The success of an industrial research laboratory depends to 
a considerable extent upon its position in the organization and upon 
its relation to the other departments of the company with which 
it is associated. 

3 If industrial enterprises had been organized afresh with the 
research laboratory as a definite part of the organization, it is prob- 
able that by this time some general opinion would have been formed 
as to the position which the laboratory should occupy, but in fact 
nearly all industrial research laboratories have been added to or- 
ganizations already formed, and their relations to the other depart- 
ments of the organization are usually closely associated with their 
origin. 

METHODS OF ESTABLISHING LABORATORIES 

4 Laboratories are established in many different ways. If 
there is a technical scientific expert in the executive of the manu- 
facturing company he may feel the need of a laboratory and become 
its director, and in this case the laboratory will necessarily be very 
closely associated with the work of the executive who initiated it. 

5 A laboratory may also be established under a separate 
director, not himself associated with the executive officers of the 
company, but as a reference department for the executives, and in 
this case also it will be very closely associated with the oflScers 
of the company and will tend to be concerned more with questions 
of policy and the introduction of new products than with any other 
of the problems of the company. 

6 In a large compan}'^ a research laboratory may be estab- 
lished as a separate department, having its own organization and 
bemg available as a reference department for all sections of the com- 
pany, in which vasv its activities will cover a very wide field, but 
at the same time it will not have as direct an influence upon the 
policy of the company as will happen if it is closely associated with 
one or more of the* executive officers. 

7 The earli(»st research laboratories grew out of the works 
testing and control laboratories and were therefore responsible 
directly to the works manager. More recently, laboratories have 
generally been established as independent departments of the com- 
pany and responsible only to the general manager. 



C. E. K. MEES 85 

8 The executive official to whom the laboratory should re- 
port will depend upon the nature of the work to be done. There 
may be industries in which research work is required for only a single 
department, and in this case the research workers should be re- 
sponsible to the head of that department; there are others in which 
the interest in the research is confined to the works, and in such 
cases the laboratory should be responsible to the works manager; 
but in most technical industries research work will have a great 
bearing not only on the methods of production but even on the 
general poUcy of the industry, and in such cases it is necessary that 
those who direct research should be in touch with, and responsible 
to, the executives who control poUcy. 

9 The position of the research laboratory in an industrial 
organization is perhaps best determined by the criterion that the 
research department should be responsible to the officer of the 
company who is in charge of the development of new products. 
If the introduction of new products is in the hands of the works 
organization, then the research department should be responsible 
to the works manager; if there is a definite development depart- 
ment, or, if new products are introduced through the agency of 
some definite executive, then it is to that executive that the re- 
search department should be responsible. The research laboratory, 
in fact, should primarily be associated with development. 

FUNCTIONS OP THE LABORATORY 

10 The chief functions of the laboratory are as follows: 

1 The provision of information regarding the technical and 

scientific matter in which the industry is interested, and 
the supply of this information in a form suitable for the 
education of the employees, of the customers, and of 
the general pubUc. 

2 Service in the form of the provision of specifications and 

standards for materials, the making of analyses and 
tests, assistance to the works in regard to difficulties 
and to customers in the relation of problems arising 
from the use of the product. 

3 The development of new processes or products, utiUza- 

tion of by-products, the development of new depart- 
ments of the industry. 

11 These functions may be expressed by means of annular 
diagrams such as those shown in Figs. 1, 2 and 3. Fig. 1 shows 



86 



INDUSTRIAL RESEARCH LABORATORY ORGANIZATION 



the information diagram. The information originating from the 
library and from the research staff is issued in the form of ab- 
stractS; reports, scientific publications and monographs, and first 
goes to the executive, manufacturing, purchasing and sales depart- 
ments for their information; then to the advertising and 
educational departments to be placed in the form required for 
publication in the scientific and general press. 

12 The organization of the development work is shown in 
Fig. 2, where the work is shown to be foimded upon pure research 
done in the scientific department, which imdertakes the necessary 



INFORMAT/ON DIAGRAM 






P\JltCHA 9iMO AMD 
9M4X9 OCPXWTMrMT, 






tOUCATiO^ 
OCP/«ltrM 










Cw»To»*en» 



9CtHTtnC P9t »* 



Fig. 1 Information Diagram, Functions of an Industbial Resbabcb 

Laboratory 



practical research on new products or processes as long as they 
are on the laboratory scale, and then transfers the work to special 
development departments to form an intermediate stage between 
the laboratory and the manufacturing departments. These 
development departments are really small scale manufacturing 
departments which may be operated either by the works depart- 
ment or by the laboratory; but which are controlled, as regards 
the work done in them and the method used, by the laboratory 
itself, being run as experimental departments in order to develop 
a new process or product to the stage where it is ready for large 
scale manufacture. 



C* B* K* MKKS 



87 



13 In Fig. 3 the service diagram shows the scientific divisions 
as the operating centers, each of which supports and controls the 
necessary service departments, which prepare specifications and 
standards, midertake testing and anal3rsis, the investigation of 
works troubles, complaints of customers, and suggestions from 
the sales department, the results of which are communicated to 
the departments interested. 

14 The laboratory organisation will therefore consist of a 
section of administration which will be responsible for the direction 



D£\/£LOPMEJST DIAGRAM 




Fig. 2 Deyelopiient Diagram, Functions of an Industrial Rxseabch 

Labobatort 

of the work, the control of accounts and the issuing of reports; a 
section of information, which will operate the Ubrary, prepare ab- 
stracts of the Uteratiu^, keep in relation with the patent depart- 
ment and constitute its technical wing, and prepare reports and 
pubUcations of all kinds; and the scientific section, which wiU 
carry on the operation of the laboratory work. 



FORMS OF ORGANIZATION 



15 There are two forms of organization possible for the scien- 
tific work of the laboratory. For brevity these may be spoken of 
as the " divisional " system and the '' ceU " system. In the divisional 



K8 



INDUSTRIAL RESEARCH LABORATORY ORGANIZATION 



Hystoin the organization is that familiar to most businesses. The 
work of the laboratory is classified into several divisions; physics, 
chemistry, engineering, and so on, according to the number neces- 
Miry to cover the field, and each of these divisions has a man of 
suiUble scientific attainments in charge of it. In a large division 
each of these men will in turn have assistants responsible for 
sections of the division, all the heads of divisions finally being 
n^sponsible to the director of the laboratory. 

10 rnder the alternative or cell sj'stem the laboratory consists 
of a number of investigators of approximately equal standing in the 





SCR VICE DIAGRAM 




^JL^^'^'^^^C^^ 




^O^PU *>•«'* ^^ 


< 


^^ ^^^^ 3u6<^»"»'*0^ 


y 




^ 

i 


/ i*// \ "^ \ 


»> 


Lr- f^CMT#.c \ ^\A 


^ 


I taCMIVW^N I'M ^f #1 


\ 
o 


\>^i^y / 


V 


Nw n:»T.>^ y/^ > 


^ 


^ "'^ C^ 




. <* 




^^ « o*^**"* 




■^>,.«-Av""' 


i 

1 

1 




i- 





V 






'.;i:\':,-.',»v\ r;uV. .^t" tV.otv. :x^siv^r.s*.l^'.o or.'.y :o :ho director, and each 
*^: :':..'••. v^^.'.Cvn: v.jV^v. s.^'v.r siw :v..- :Y><\srvh. EaoH such invest!- 
K . . ^ .,',.■ >^x* • V. :. \  \' :* : \ ' \ . . : ,\: \\ ; : \\ .■•, ns >: ;i v. : s .■%> nui y be necesBary. 

•.'■... .*v- ^.^■'•. ^>^:,"-. .x:^;^^" t:;;-^^ two srstems of 

*- -^^ • >* . s . K .. .-. \ -^ . • . v\x:.' :.*..: v>n the otlier hand 

^"^^ • ^ ..X .■..■.•. .K f^tx^nvc. 

^ ^^ ^ x . \ • --^^ :. -y is hc^ directed 

^ " >. . - .  . - ' .^-s v>: all the men 

■^ K V '» • <. X > %^ . ; h-sw the further 



.^if ihe defalk of 



C. B. K. HEES 89 

the direction of the scientific work, it will keep him constantly 
informed as to all the work going on in the laboratory. 

COST OF RESEARCH LABORATORY 

19 From various sources, but chiefly from the convenient list of 
American laboratories given by Fleming, there can be found the 
cost of a research laboratory per scientific worker employed, at the 
time when that list was prepared. It might seem that there would 
be very great variations in this, but, provided that the laboratories 
are all of the physical and chemical type, there is a surprising agree- 
ment between the figures, which show that the cost of building and 



Fic. 4 The Labohatory op the Eastman Kodak Company 

equipment for a laboratory was then between $3000 and $4000 per 
man. From the same sources the annual cost of maintenance of 
such a research laboratory appears to be slightly lower than the first 
cost. Probably S3500 per man is a fa ir estimate of the cost of main- 
tenance, and of this we may take 60 per cent as representing 
salaries and wages and the remainder all other expenses. These 
figures must, of course, now be increased in proportion to the rise in 
costs. 

20 As laboratories are organized and experience gained in the 
types of laboratory suitable for different industries it will doubt- 
less be possible to lay out definite schemes of organization for a 
laboratory suitable for the requirements of any industrial under- 



90 INDUSTRIAL RESEARCH LABORATORY ORGANIZATION 

taking. At the present time, however, it is possible to do this 
only in the most general way, and it is necessary to consider each 
case independently, taking into consideration the requirements of 
the particular industry involved. 

21 Fortunately data on this subject are being accumulated 
at the present time and the whole question of laboratory organiza- 
tion is being studied carefully both here and abroad. In this country 
the National Research Council has appointed a special committee 
to promote industrial scientific research and this body is now en- 
gaged in drawing up a scheme for the estabhshment of an alloys 
research laboratory and is considering other lines of industrial re- 
search all of which will surely meet with the approval and support of 
The American Society of Mechanical Engineers. 



No. 1601 

RESEARCH WORK ON MALLEABLE IRON 

Bt Enbiqttb Touceda*, Albany, N.Y. 
Non-Member 

The author presents an account of foiw yeare of research work undertaken for the 
American MaUeable Castings Association as a plea for industrial research among 
manufacturers and as a striking example of what such research can accomplish. He 
sketches the organization and purpose of the Association and shows how the quality of 
the product of its members has steadily increased since the beginning of the research 
work, MaUedble^iron castings, due to lack of uniformity and dependability, were 
rapidly being replaced by other materials. There were many fallacious ideas and 
theories regarding the physical properties of such castings and the methods of annealing 
them. Records of tests of l-in. bars from seven different concerns made by the author 
in 1911 showed that the average ultimate strength loas 39,882 W. and the elongation 
under 5 per cent, A report dated March, 1919, to the members of the AssocieUionf 
each of whom regularly submits test bars from some one heat of each day's runs, showed 
that 44 per cent of the test bars submitted during that month had an ultimate strength 
over 52,000 W. and an elongation of 14.67 per cent, indicating the progress made since 
research work was undertaken. 

It is further stated that the average of test bars of the Association from January 1, 
1917, to March 31, 1919, has shown an ultimate strength of 51,000 lb. and an elongation 
12.5 per cent. The records of tests show, contrary to generally accepted theory, that 
the etongoHon increases with the ultimate strength. The purpose of the Association, 
however, is not to increase ultimate strength and elongation but to increase the uni- 
formity of a product upon which the engineer can rely, and this is being accomplished 
through exhaustive research and advice to members Oirough the considting engineer of 
the Association. 

A description is given of the process of manufacturing malleable iron, of the air 
furnace, and of the annealing ovens and the annealing process. The structures of 
iron containing free carbon and iron containing combined carbon are shoum by micro- 
graphs and the metallurgy of cast iron is carefuUy explained with abundant micro- 
graphs of typical structures. 

The effects of the time element in cooling through the critical temperature, of stac- 
cessive anneals, of varying percentages of carbon, sulphur, silicon, phosphorus and 
manganese and of subsequent heating to high temperatures are clearly described and 
Ultistrated. Picture-frame fractures are also disciused. 

^ Consulting Engineer, Albany, N. Y. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Socibtt of Mechanical ESnqineers, 29 West 39th Street, New 
York. 

91 



92 RESEARCH WORK ON MALLEABLE IRON 

The author closea by exploding three popular theories with regard to maUeabU 
iron. He shows that the strength of malleable iron is not confined to its skin hut that 
this may be machined off without destroying strength, if the quality of the iron iscuit 
should be. Secondly ^ fie shows that it is possible to eliminate the carbon throughout 
the entire specimen and not merely near the surface* LasUy, he shows that thick as 
well as thin castings can be annealed successfully. 

TT is the writer's belief that three factors mainly operate to darken 
the vision of many of the manufacturers regarding the value of 
research work. The first is fundamental. Stated brutally, it is 
ignorance as to its possibilities. Following this regrettable fact is 
their obsession to the thought that the prime and essential requisite 
for success lies in the advertisement, with oftentimes the accent 
on the last four words. The third factor has reference to the extent 
to which research has been discredited through the emplojTnent for 
that purpose of men not qualified through temperament, proper 
training or resourcefulness, to undertake such work — the square 
peg in the round hole, the neophyte, and at times the quack. While 
the revelations born of the world's war have in large measure done 
much to open the drowsy eyes of the manufacturer, it is nevertheless 
certain that hard and diligent work still rtmiains to be done before 
the majority will Ik? aroused from their lethargy. 

CONDITION OF THE INDUSTRY PRIOR TO RESEARCH 

2 Prior to the time four years ago that research work was under- 
taken in the interests of the American Malleable Castings Associa- 
tion, the industry was in a more or less chaotic condition. There 
had Ikhmi at letkst three years of serious business depression, ruinous 
competition as a consequence had been running its insensate course, 
but back of and l)eyond all this, was the damning accusation of thei 
engiiuHT, that the material, except in the case of a limited number 
of concerns, w:is not only lacking in deixuidability, but of low strength 
when de|HMulable. In niilway-car fabrication particularly, the 
nuinl>or of nialloable-in.)n castings useil had dwindled from a very 
lar^e quantity |K'r car, to an ahuost insignificant number consisting 
mainly of uniiniH)rtant details. Malleable c:ist iron was rapidly 
Unnir ivplaccd by the stool casting, and in other directions as well, 
tho latter was on oroa oiling on the legitimate field of the former, and 
inoidontaliy plaonm it in an exaggeratedly false position, for the 
n\ison that when substitutions woiv nuule the {bitterns were re- 
designed auvi niailo nuioli heavier to aooonumKlato the less-fluid 
easting projHTties o\ that metal. When a steel c:usting failed, the 



99 



I'.'.J'.'j'.'i' " ' v*^ t_r 

« 



'JilZ-.: TlXIii-r z- 



■•I* 



pAr::-:j.: >--:■. 
meiallu-*z>" \'\ • 

part of *ii- •_.-, 
at the tzzr: -:■..• 
prf.'IiS; Trr-r- ■: 

niil'ie^' :r... z t-; 



' ..r 



I • 



•"w -r- — 



.--i'^- 



--- :...:r- 






-: . .f 



4 A::^: 



* • ..i 



>• •> tt 



■■■••■ 



- - -....■— ",.r T-^- .. 






  ^- - •-• I -,■-..,- , 

. ..  •^' • ■-•' J-—. ■• . 

^ .,- »•- _ 

-*. ." ■* - i- •-». - .. . 



■i -■:■:■: v 



t" " 



"L ^ . •  —1 - . L •'• 

•-■•■...•■ ■^- 

'<^ • .. . . - — ..-a..t-:i i- - 






.-J'.- - . 



ity 

to 
are 
:ial 
the 
are 
:he 

ai 

to 
on 
?r>' 
nd 
ite 
ige 
lb. 
est 
lb. 
is 
ad 
ns 
dd 
as 
•o- 
at 
5e. 
he 
•St 
er 
ic 

ID 

d. 
)e 



ts 
nd 
uld 



• . 



94 RESEABCH WORK ON MALLEABLE IRON 

this product conclusions drawn from experiments made on white- 
heart malleable iron, such as used on the other side. 

FORMATION OF THE ASSOCIATION 

6 Among the various manufacturers of malleable-iron castings, 
at the time referred to, were some twenty-five who were progressive 
enough to understand the benefits that accompany cooperation, 
and that one might better have an intimate friend with whom to 
compete than an enemy. These had formed an association that had 
been in existence some ten years or more. The association had as 
its objectives the exchange of ideas in the direction of business 
economy, improved works practice, the study of proposed foundry 
and factory enactments, the securing of more favorable insurance 
rates, the study of problems relating to cost, labor, housing and 
sanitation, and finally the forming of friendships that are the natural 
outcome of frequent and close personal contact. It is to these 
men largely that credit should be given for the renaissance of the 
industry. They decided to enlarge their field of action and, irrespec- 
tive of cost, determined to go the full road along the lines of metal- 
lurgical research. They determined as well that every statement 
made as to progress would be conservative and accompanied by 
data that would be incontrovertible, which course they have followed 
to the letter. 

7 The preliminary part of this paper will be descriptive of the 
steps that were taken to bring system out of chaos, the second will 
deal with some metallurgical details of the malleable-iron process, 
and the third with the fallacies that have been handed down in the 
manner already indicated. The first step taken towards systematiza- 
tion was to make a hasty survey of the different plants. At the 
conclusion of the survey, papers were written by the consulting 
engineer in connection with matters deemed of most importance at 
the time to the membcrshii>, and in these recommendations were 
made as to suggested improvements in works practice generally. 
These papers were sent to the secretary", who had them printed in 
uniform bulletin fonn for distribution. In these bulletins to date 
a very wide field ha.s been covered, as they contain quite complete 
details in regard to physical proix?rtics and tests, the metallurgy 
of the process, air-furnace and anncaling-ovcn construction and 
practice, combustion, special investigations of difficulties encountered 
from time to time by the various members, and other mattera too 
numerous to mention. 



ENRIQUE TOUCEDA 95 

METHOD OF PHYSICAL TESTING 

8 When the research work was started, it was found that by 
far the majority of the members had no system of testing the qnaUty 
of their product, aside from the twisting and bending of a casting, 
in order to ascertain its ductiUty, or the bending over of test lugs 
attached to castings. Consequently there was no way in which 
could be compared the quaUty of product of one member with that 
of another. To be candid, there was available no information of 
value that could be given to an engineer who might be seeking 
information of this character. Mr. Benjamin Walker, who at the 
time was vice-president and general manager of the Erie Malleable 
Iron Company, of Erie, Pa., a number of years previous to this, had 
devised for the purpose of testing the quaUty and uniformity of his 
product, what he called a test wedge. The wedge was 6 in. long by 
1 in. wide, tapering from ^ in. at its base to iV in. at its other end. 
His practice was to distribute a series of these wedges throughout 
the annealing oven, and at the conclusion of the anneal subject 
them to test — accomplished by holding the butt end of a wedge 
in a narrow-jawed tongs, placing it upright on an anvil, and striking 
its top end with a 6- or 8-lb. sledge. In this manner the thin end of 
the wedge would gradually curl up imder these repeated blows, and 
it is apparent that the more blows the wedge would stand before 
fracture, the shorter would be the butt left in the jaws. The short- 
ness of this butt he considered was a measure of the metaFs ductiUty. 

9 It was decided that an attempt would be made to standardize 
this test for adoption by the* Association. With this end in view a 
machine (Fig. 1) was so designed that a weight of 21 lb. when raised 
to a height of 3.33 ft. above the top of the wedge, when placed in 
position for test on the anvil of the machine, would be automatically 
tripped to deUver a fairly constant blow on the thin end of the wedge. 
The blows were counted, and for convenience the number dehvered 
before rupture took place, was recorded as the blow efidency (as- 
sumed to be a measure of toughness), while the length of the butt 
(assumed to be a measure of ductiUty) was measured and expressed 
in terms of hvU efficiency. These were arbitrary terms, understood 
by the members, and intended for their use only. It wiU be noted 
that the test is one of great severity, for the reason that the first 
blow must be borne by a section but sUghtly larger than 1 in. by tV 
in., and equals, expressed in pounds of static pressure, over 1350 lb. 
As the wedge curls and consequently shortens, subsequent blows 



98 RESEARCH WORK ON MATiLEABLE IBON 

11 In this manner it is possible to learn just how each member 
is progressing, and if it is considered by the Research Committee 
that his progress has not been as rapid as it should have been, the 
consulting engineer is requested to pay him a visit with the object 
of aiding him more quickly to better his condition. 

AVERAGE PHYSICAL PROPERTIES PRIOR TO RESEARCH 

12 It is pertinent at this point to present in as fair and impartial 
a manner as possible a comparison of the physical properties of 
malleable iron manufactured within the past few years with that 
made in the period prior to 1913. Before entering into this matter 
the writer wishes to go on record as stating that any data in con- 
nection with the ultimate strength of malleable iron is valueless 
as far as serving to show its real worth, unless accompanied by infor- 
mation regarding the ductility of the metal as measured by the 
elongation. If one knows how, there is no diflBculty whatever in 
uniformly making a metal of 85,000 lb. ultimate strength, provided 
ductility be sacrificed down to what would be represented by a 
5 per cent elongation. Further on the structure of such material 
is shown. In the past there have been occasional records of tests 
that have run somewhat similar to these figures, and such tests have 
been quoted by others in different articles, but in the light of present 
practice this is not considered good malleable iron, unless it is to be 
used for special purposes where ductility is not of consequence and 
a high ultimate strength is imperative. When such high ultimates 
were obtained it must obviously have been by accident rather than 
design, because if 5 per cent elongation was a rather high average 
at the time, it follows that it would have been accompanied by a 
high ultimate strength had it been known how to obtain it. There 
are still stronger reasons for making this statement. For many 
years the writer has had an unusual opportunity to' learn either at 
first hand or on good authority what character of product most of 
the concerns made, and aside from one concern who had always 
enjoyed an enviable reputation for the uniformity and exoellenoe 
of its product and another verj' large company whose plants were 
\ery painstaking in their methods of manufacture, the ordinary nin 
of malleable iron was undoubtedly inferior. The record of tests 
made in this laboratorj' during the period mentioned do not exceed 
three hundred. Thoy do, however, represent the product of many 
different concerns. As some of the bars were of square section, 



SNBIQUS TOUGEDA 99 

some rectangular, and others round, it is plain that no uniformity 
existed in their dimensions. The latter vary anywhere from ^ to 
1 in. in diameter, while the former for the most part are 1 in. square 
and 1 in. by § in. The great majority of these tests show that 
the ultimate strength was under 39,000 lb. per sq. in., while the 
elongations were for. the most part under 3.5 per cent. There are 
instances of fairly high strength, slightly over 48,000 lb., while the 
highest elongations, ran 7 per cent in 4 in. 

13 It happens that the writer has a record of tests made in 1911 
on bars made by seven concerns that were deemed at the time to 
be imquestionably among the very best producers of malleable-iron 
castings. These foimders were each asked to make 20 of the very 
best bars they could produce for test, 10 to be 1 in. in diameter, and 
10 to be § in. in diameter. In these tests the average ultimate 
strength of the 70 bars of 1 in. diameter is 39,882 lb., and the average 
elongation exactly 5 per cent. The lowest ultimate is 31,990 lb. 
and the highest 45,560 lb., the lowest elongation 1.7 and the highest 
9.8 per cent. In the |-in. bars, the average ultimate is 41,693 lb. 
and the average elongation is 5.5 per cent. The lowest ultimate is 
33,600 lb. and the highest 47,430 lb., lowest elongation 1.2 and 
highest 6.3 per cent. Inasmuch as each of these seven concerns 
were informed that whaf was wanted was 20 test bars that would 
represent the very best product they could make, and inasmuch as 
these manufactiu^rs were considered among the best of the pro- 
ducers, it would appear that the writer is warranted in assuming that 
the foregoing tests would represent a high rather than a low average. 

14 Table 1 is reproduced from Dr. H. M. Howe's book. The 
Metallography of Steel and Cast Iron, pages 96 and 97. In the first 
series of this table there is a very good bar, though the writer 
is inclined to doubt the accuracy of the figures under the elastic 
limit column. In the second there is a bar, sUghtly better than 
the other. The one with 8.2 per cent elongation is also good. 
The second series furnishes a guide as to what was deemed to be 
unusually good malleable iron several years ago. 

THEORIES OF MALLEABLE IRON 

15 In Hatfield's Cast Iron in the Light of Recent Research, 
1912, page 213, is a table containing the results of 14 tensile tests 
of bars 1 in. by | in. in section. The bars run very uniformly and 
the material, while of low strength, is ductile. The latter should 



100 



HESEAHCH WORK ON MALLEABLE IBON 



certainly be expected in bars that are but f in. thick. The lowest 
ultimate strength is 42,750 lb. and the highest 51,200 lb., the lowest 
elongation 10 per cent and the highest 15.3 per cent. Under this 
table appears the following: "If attempts are made to increase the 
maximum stress obtained from such iron, the elongation would 
appear to have to be sacrificed, and to a considerable degree. In 
illustration of this the following records are given." 



Maximum stress, 


Elongation, 


Reduction of 




tons per sq. in. 


per cent in 2 in. 


area, per cent 




18.9 


12 


9 




21.1 


13 


8 




24.0 


9 


8 




26.5 


7 


5 




27.8 


5.5 


3.5 




29.0 


5.0 


4.2 




34.3 


3.0 


2.5 





16 Under this table the following words appear: "The cause 
of the increase in tonnage is the retention of increasing proportions 
of combined carbon. This combined carbon stiffens the material 
and incidentally reduces its ductility." The writer believes that 
he was the first to prove and furnish indisputable evidence as to the 
falsity of the statement that the ductility of malleable iron decreased 
as its ultimate strength increased. Wliile Hatfield is correct in 
assuming that the ultimate strength is increased and the ductility 
decreased with an increase in combined carbon, the statements 
indicate clearly that he was, at the time at least, not familiar with 
the manner in which malleable iron can be obtained having the 
characteristics of high strength accompanied by high ductility. 
It can b(^ stated, that in normal malleable iron as made today the 
higher the strength the higher will l>e the ductility, and in this 
particular this metal Ls unitjuc. In a paper read by Dr. Richard 
Moldcnki*, bc^fore the American Foundrymen's Association in 1903 
can be found these words: ''The tensile strength of malleable castings 
should run hot wren 42,0(K) and 47,000 lb. i)er sq. in.; castings showing 
only 3r),(K)0 lb. arc servi('eai)le for ordinary work. It is not adnsabU 
to run bcynfid r)-l,(K)0 11). ]M}v s(i. in. f(fr the n.s/7/c^/a? is reduced, and one 
of the valuable propertic^s of the malleable* casting impaired. The 
elongation of a piece of iiood malleable will lie between 2J and 5| 
\)cr cent.'' In the MecliMiiical Kngine(*rs' Handbook, edited by 




ENRIQUE TOUCEDA 



101 



Prof. Lionel S. Marks and published as late as 1916 appears the 
following by the same author : "These castings should not be machined, 
as the interior is not as strong as the metal at and near the surface. 
Tensile strength 35,000 to 48,000 lb. per sq. in. European malleable 
cast iron, made by a somewhat different process, is not as sensitive 
to machining, the castings which are thin only, are practically 



TABLE 1 PROPERTIES OP MALLEABLE CAST IRON (H. M. HOWE) 



Tensile 

strength, 

lb. per 

sq. in. 



Yield 
point, 
lb. per 
HI. in. 



Elongation 



per cent 



inches 



Malleable eastings. Standard Properties 



E. Schoemann. Open-hearth furnace. 



W. P. Putman, 



I 



Kent, Master 1.52 X 0.25 in 

Car Builders' I 2.0 X 0.78 in 

Association. 1891 | 1.54 X 0.88 in. . . . 

Sixe of 8i>ecimen [ 1.52 X 1.54 in 

Kent 

Touceda. Avg. strength of commercial 



44,230 
48,640 
39,638 
50.849 

34,700 
25.100 
33,600 
28.200 
32,000 
41,000 



28,326 
41.792 

21.100 
15,400 
19.300 



3.9 
4.5 
7.03 
10.15 

2.0 
1.5 
1.5 
1.5 
2.0 



Malleable castings. Unusually good properties 



Contraction 
of area, 
per cent 




10.51 
20.92 



C. H. Gale 


 I 


60.000 
70,000 












3 tests 13/16 in. diam. 


55,100 1 


5.2 




5.5 




3 tests 13/16 in. diam. 


64.500 




2.8 




1.3 


H. R. SUnford. 


2 tests, 13/16 in. diam. 


69.100 




4.0 




2.6 


average of 


2 tests, 13/16 in. diam. 


56.700 




8.2 




8.4 




3 tests. 13/16 in. diam. 


51.600 




7.0 




7.7 




42 tests, 13/16 in. diam. 


49.810 




6.61 




6 23 


S. B. Chadsey. . 


{ 


45.810 
55,230 




6.25 
10.55 


4 





decarbonized in the annealing process, whereas in the America 
black-heart malleable iron only the skin is decarbonized, the metal 
adjacent for about J in. partially so, and the central portions contain 
the full carbon percentage of the original hard white casting." The 
writer has italicized certain parts of the matter quoted in order to de- 
vote if possible a little attention to these parts later on. It is beUeved 



102 



RESEARCH WORK ON MALLEABLE IRON 



that the foregoing fau*ly sets forth the state of affairs as to physical 
properties during the period mentioned. 



IMPROVED PHYSICAL PROPERTIES DUE TO RESEARCH 

17 Let us see what improvements have been made within 
recent years, that is, since the research work was started for the 
Association. We will not select the best month's record, but will 
reproduce in part the very last report sent to the members, that is, 
the record for March 1919. 

18 Table 2 covers the physical tests for ultimate strength and 
elongation on bars received during the month of March 1919. 
It would be impossible, almost, to consider the figures in this record 
and fail to note that as the tensile strength increases, so does the 
elongation. These monthly records have been kept in this manner 

TABLE 2 ANALYTICAL EXAMINATION OF PHYSICAL TESTS FOR m^TIMATE 
STRENGTH AND ELONGATION ON TEST BARS SUBMITTED DURING MARCH 1919 



Limits of ultimate strength 



Per cent of bars 


Per cent eloncation 


0.40 


4.50 


88 


6.86 


2.79 


7.67 


6.47 


8.13 


10.29 


9.50 


17.16 


10.09 


17.95 


11.55 


44.06 


14.67 



Under 40.000 lb 

Between 40,000 and 42.000 lb. 
Between 42,000 and 44.000 lb. 
Between 44.000 and 46.000 lb. 
Between 46,000 and 48,000 lb. 
Between 48.000 and 50,000 lb. 
Between 50,000 and 52,000 lb. 
Over 52,000 lb 



for four years or over and there is no exception to this rule. It will 
be noted that only 0.40 per cent of the total bars received tested 
under 40,000 lb. per sq. in. As a matter of fact, only 10.54 per cent 
were under 46,000 lb., while 44.06 |X}r cent stood over 52,000 lb. 
with an average elongation of 14.67 per cent. The best individual 
record showed an average of 59,681 lb. ultimate and 21.47 per cent 
elongation. The worst was 39,942 lb. ultimate and 4.20 per cent 
elongation. The latter record belongs to a member who but very 
recently joined the Association, and bears out quite well the thought 
that the writc^r has been endeavoring to convey. 

19 The inenihers who submitted bars classified under Railway 
Work were twenty-two in number. The average ultimate strength 
and elongation of test bars submitted by the eleven members having 
the highest averages are found to be 53,559 lb. and 15.56 per oent| 



ENRIQUE TOUCEDA 103 

respectively. Carrying through the same operation with the 
twenty-six members who are not thus classified, it is foimd that the 
average ultimate strength and elongation of these thirteen are 
respectively 52,327 lb. and 12.42 per cent. Taking the average of 
these twenty-four members, we find that the ultimate strength is 
52,943 lb. and the elongation is 13.99 or practically 14 per cent. 
Lest our intention be misunderstood, it should be explained that 
our effort is not directed toward securing an increased ultimate 
strength and elongation so much as uniformity of product. The 
aim is to secure a product that the engineer ^^411 readily acknowledge 
possesses excellent ph^'sical properties, which vary but little from 
heat to heat. Within what can be considered quite narrow limits, 
this is what these particular twenty-four members are doing, while 
most of the others are not far behind. It may also be of interest 
to state that from January 1, 1917, to March 31, 1919, the average 
ultimate strength of the test bars of the Association as a whole has 
been over 51,000 lb. ultimate and the elongation 12.50 per cent. 



INFLUENCE OF WAR CONDITIONS 

20 In considering the last statement the following facts should 
be taken into accoimt. War conditions during 1917 and 1918 
made it quit« impossible to secure appropriate pig iron and fuel. 
It is only fair to state that most of the companies were greatly 
handicapped during this period. It was solely and only through 
an intimate knowledge of the metallurgy of the process derived from 
the research work that made such a showing possible. Aside from 
this the membership has been and is constantly growing and the 
total average is and for some time to come will be necessarily affected 
as a consequence, as it takes some few months to get a new member 
in line. It is not imfair to assume that a still better showing could 
have been made had the times been normal, and had the membership 
been confined to those only who were members when the research 
work was first started. At the beginning of our investigations an 
elongation of 10 per cent was considered to be an indication of a 
superior product. As our knowledge of the metallurgy of the 
process increased, accompanied by better air-furnace practice and 
annealing-oven conditions, the elongation particularly began to climb. 
An elongation of 20 per cent is not now looked upon as unusual; 
elongations of 25 per cent occiir with considerable frequency, while 
we have had numerous bars that have run as high as 30 per cent and 



104 BESEABCH WORK ON BIALLEABLE IRON 

several of 31 per cent, which for an untreated cast-iron product we 
believe to be quite extraordinary. 



METALLURGY OF CAST IRON 

21 The raw material from which malleable-iron castings are 
made is pig iron, but this must be of suitable composition for the 
process. The usual elements or "impurities" (as they are fre- 
quently called) in pig iron are carbon, phosphorus, sulphur, man- 
ganese and silicon. If any of these elements combine with the iron, 
or combine with each other in definite proportion, compounds will 
be formed and it must always be kept in mind that compoimds have 
very different physical properties from those possessed by the ele- 
ments that form them. For instance, pure iron is so soft that it is 
difficult to machine it in such a manner as to leave a nice clean sur- 
face, as the chips have a tendency to shear or tear ofiF before the edge 
of the tool has cut through, while graphite, the form in which free 
carbon exists in pig iron, is certainly vcrj- soft; and still, when these 
two unite to form carbide of iron, in the proportion always of al)out 
6.67 per cent carbon to 93.33 per cent iron, they yield the hardest 
substance that can be produced from iron or steel b}"- any known 
method. Now it happens that there exists a preferential and reci- 
procal attraction between some of these elements. For instance, if a 
piece of pure siUcon is dropped into a bath of pure iron, or even iron 
contaminated with phosphorus, sulphur and manganese, when soli- 
dification takes place the silicon will he found to exist in the iron not 
as such, l)ut united with it to form a definite compound called sili- 
cide of iron, because any tendency it has to remain by itself is over- 
come by the iron, for which it has a greater attraction than for any 
of the other elements present. In the event that any element in pig 
iron remains uncoml)ined, that is, fails to unite with the iron, or 
with any of the other elements present, then the element is said to 
exist "free." 

22 Co7)ipou7His of Iron. In the absence of much silicon^ carbon 
will always unite with the iron to form the compound called carbide 
of iron, a structural constituent known as cementite. We have 
already stated that silicon unites with iron to form the compound 
known as silieide of iron. Phosphorus in commercial pig iron suit- 
al>l(* for use in tin* manufacture of malleable iron «ilwa\'s combines 
with the iron to torni the eoin pound phosphide of iron. Sulphur 
and manganese have a nM'iproeal attra<-tion for each other, greater 



ENRIQUE TOUCEDA 105 

than either has for the iron or the other elements, and in consequence 
we can consider that the conditions are always such commercially 
that the sulphur and manganese will unite together to form man- 
ganese sulphide. Any manganese in excess of this requirement will 
unite with the carbon to form what we will at present call manganese- 
iron carbide (manganiferous cementite). It can now be stated that 
of the five impurities or elements referred to, none, except the carbon, 
can occm* in the iron in the free state, and the latter can never exist 
free, as graphite, in whole or in part, unless it happens that there is 
present in the iron an amount of silicide of iron sufiBcient to prevent 
the formation of iron carbide. As indicated, silicide of iron has a 
tendency to break up, that is, render imstable, the carbide of iron 
that forms during, or shortly after solidification, causing it to disso- 
ciate into its two original soft constituents, iron and carbon; but as 
we will see later on, this action will not start unless a certain amount 
of silicide of iron be present. 

23 Influence of Time Element. It must be borne in mind that 
all reactions are governed more or less by a time factor. If the 
pig or the casting, although containing high silicon, be rapidly 
cooled the iron may become rigid so quickly that the carbon will 
be denied the time to separate out as graphite in spite of the presence 
of an amount of silicon that would have precipitated the carbon 
under conditions of normal cooling. Sulphur, on the contrary, acts 
to encourage the union between carbon and iron; that is, it acts to 
stabilize the carbide of iron. Inasmuch as sulphur imites with 
manganese to form sulphide of manganese, a compoimd of the 
nature of slag and supposed to be inert, the writer fails to imderstand 
just how it functions to stabilize the carbide. He acknowledges 
that this appears to be the case even when there is present at times 
more manganese than suflScient to satisfy the sulphur. He acknowl- 
edges also his ignorance as to the cause, but beUeves that the sulphur 
does not unite with the manganese until after it first has exerted its 
influence in some manner unknown on the carbon. 

24 Free and Combined Carbon, When the carbon exists free, 
it occurs more or less imiformly distributed throughout the mass of 
metal, in plates of varying shape, size and thickness as shown in the 
micrograph of Fig. 2. When in this condition, the fracture of the 
pig iron usually will be coarsely crystalline and very black, the iron 
will be very soft and easy to machine, and it will lack high strength. 
If, on the other hand, the carbon is present entirely in the combined 
form, the fracture of the pig will be white and vitreous, it will be 



106 RESEARCH WORK ON MALLEABLE IRON 

impossible to machine it due to its extreme hardness, and while it 
will have a much higher ultimate strength than when the carbon is 
in the free state, it will be brittle. The great difference in structure 
when the carbon is combined and when it is free can be seen by 
comparing the micrograph of Fig. 2 with that of Kg. 3. Between 
these two limits it is possible to have pig iron in which the carbon 
will exist in both the combined and graphitic form in any proportion 
of the total carbon content, while the appearance of the fracture, 
machinability and strength, obviously will depend upon what part 
of the total carbon content remains combined and what part remains 
graphitic. If then we see a sand-cast pig (one that has cooled nor- 
mally) whose fracture is white we know immediately much about its 
physical characteristics and a few things about its composition; 
that is, we know that it is extremely hard and brittle, that its silicon 
content must be low, and that all the carbon must be in the combined 
form. If, on the other hand, the fracture is very black and coarsely 
crystalline, we will know that the iron is very easy to machine, 
that is will not have a very high ultimate strength, that the silicon 
is not very low but on the contrary is probably quite high, and that 
most if not the entire amount of carbon exists in the free state. 

25 Effect of Silicon, The foregoing practically signifies that 
whether the carbon will be in the combined or free state in a pig 
iron or a casting depends primarily upon the percentage of silicon 
present, though the part played by the rate of cooling must also be 
considered, as we may have thick or thin castings or castings with 
thick and thin sections. Consequently, through the control of the 
silicon, the condition in which the carbon will remain in the iron can 
be determined. As the control of the siUcon is easy, so is the control 
of the condition in which the carbon will exist in the casting. 

26 Ignoring for the moment the influence of the rate of cooling, 
it has been stated that in order to obtain an iron that will be white 
in fracture, the silicon must be low enough to lack a tendency to 
break up the carbide of iron into its original two components. White 
iron is made up structurally of two elemental constituenta, carbide 
of iron (cenientite), and carbonless iron (ferrite), the former existing 
in part free, and the remainder forming a mechanical mixture in 
definite proportion with all of the carbonless iron, to make a con- 
stituent to which the name pearlite has been given. In Kg. 3 the 
white constituent is the cementitc existing free, while the dark areas 
arc the pearlite. Many years ago it was discovered that if particles 
of carbide of iron, in intimate contact with carbonless iron of the 



Fig. 2 MicBOGRAPH Showinq Free Cabbon in Cast Iron 



MicBOQRAPH SHOwnta Combined Carbon in Cast Ibon 



108 RESEARCH WORK ON liALLEABLE IRON 

character referred to, were maintained at a temperature of bright 
redness for some 40 to 60 hours and then allowed to cool with extreme 
slowness, that this very hard constituent could be, through this 
treatment, broken up into the two very soft constituents through 
whose union it was formed. 

PROCESS OF MAKING MALLEABLE CASTINGS 

27 It can now be stated that the process for making black-heart, 
malleable castings involves two steps. The first step consists in 
making a casting in which the totalit}'' of the carbon will exist as 
carbide of iron, when the iron will have a structure shown in Fig. 4, 
which structurally is like Fig. 3 but which contains less free cemenf ite, 
because air-furnace white iron has an average carbon content of 
but 2.40 per cent as against an average of 3.50 per cent in whit« pig 
iron. In this step, then, is produced a casting white in fracture, 
hard and as brittle as glass. The second step consists in subjecting 
this white-iron casting to a heat treatment such as will Qerve to 
break up this hard carbide into its two original components, both 
of which are very soft, and hence from a whit€-iron casting we can 
obtain through beat treiitnient one that possesses the properties of 
strength, toughness and ductility. Fig. 5 shows the structure of a 
normal well-annealed piece of malleable iron. The white ground 
mass is the carbonless iron (ferrite), while the dark continent is 
the carbon that precipitated out during the anneal. If Fig. 5 be 
compared with Fig. 4, an idea will be gained of the profound change 
that has taken place in the structure during the annealing process. 

28 As pre\'iously pointed out, the raw material for the manu- 
facture of these castings is pig iron, though pig iron does not con- 
stitute the entire charge, as not only must the sprue from the previous 
heats be melted, but it is more than likely that when these two alone 
are used the carbon in the mixture will l>e too high to yield a white 
iron of suitable composition to produce the strongest product. Con- 
sequent ly, in order to lower the carbon to the necessary limit, steel 
or other voiy low-carbon scrap must Ixi used to bring about that 
end. In by far the greater majority of cases this mixture is melted 
in a rtnerberatory furnace, commonly known as an air furnace, but 
the cupola, the open-hearth and the electric furnace can be and are 
used for that puri)ose. Inasnnieh as the iron for more than 95 per 
cent of all of the nialleal'le eastings produced is melted in the air 
furnace, we will confine^ ourselves to that particular apparatus. 
A photof^raph of the furnace is shown in Fig. G. 



BNIUQITE TOUCEDA 109 

DBSCBIPnON OF TBS AIR FURNACE 

29 The furnace consists essentially of a fire pot, hearth and 
stack. Some furnaces have a sohd roof, the charge being "peeled 



Fia. 4 MiCBOOBAPH or Hakd, Whits Cabt Iboh 



Fto. 5 MiCBoORAPH OF NoBMAL, WbutAnnsaud Maixbable Ibon 

in" throi^h the chai^g door, but in the larger number of cases 
the roof is made up of bungs that can be removed durii^ repairs 



110 BESELABCH WOBS ON lULLSABLB DtON 

and a sufficient number of them lifted off when necessary for the 
purpose of admitting the charge. While a few of the furnaces 
operate with natural draft, most of them use a forced draft of from 
3 to 4 oz. pressure. Although oil is advantageously used for fuel 
in a few instances, where its cost is not prohibitive, bituminous coal 
having a volatile combustible of from 25 to 35 per cent is the heating 
^ent used by the majority. In order to bum the volatile products 
of the coal, as well as the CO generated from it, when, as should be 
the case, a deep bed of coal is used, secondary air is admitted through 
a series of tuyeres or a continuous tuyere, which is located far enough 
in front of the grate bridge wall and so inclined that the air will 



Flo. 6 Typical Air Furnace 

enter and be deflected about 15 in. from the base of the bridge. 
In this miinner a maximum temperature of about 2800 to 2950 d^. 
fahr. can be eventually obtained. Considerable time and thought 
have been expended in an endeavor to see if it would not be possible 
to regulafe the amount of sccondarj' air to just insure perfect com- 
bustion by the u.>^ of a C()i recorder. After considerable experi- 
mentation it was found that tlic plan was <lefeatcd by the inability 
of the recorder to act with .sufTicicnt prom|)tness. The furnaces vary 
in capacity from about 7 tons for the small furnaces to 35 tona 
for the larger ones. The best are verj' inefficient and in practice 
the fuel ratio b about three of iron to one of coal. The average 
practice does not exceed 2.5 to 1, In numerous cases waete-heftt 



ENRIQUE TOUCEOA 111 

boilers are used and while this conserves heat, it obviously does not 
add to the furnace efficiency. The average furnace has a capacity 
of about 15 tons. 

THE CHARACTER OF THE AIR-FURNACE CHARGE 

30 It has been explained that the object of the first step in the 
process is to obtain white-iron castings, or those in which the totality 
of the carbon is in the combined form, and we have seen that this 
object can be attained through a control of the silicon; that if the 
silicon is high, the carbon will be precipitated in the iron as graphite, 
which will yield a gray-iron fracture, while if it is low, the carbon 
will combine with the iron and the fracture of the casting will be white. 
As pig iron in which all the carbon is combined can be easily made 
and purchased, it would appear logical to assume that inasmuch as 
the sprue which, of necessity, forms part of the charge must be white, 
why not use this character of pig, when all that will have to be done 
will be to melt the charge, superheat it to a temperature that will 
successfully run the castings and let it go at that. Actually this is 
what is done when the inferior cupola product is made. To under- 
stand why it is impractical to do this in the air furnace, it is necessary 
to have a previous knowledge of what takes place in this apparatus 
during the melting of the charge. 

OXIDATION IN THE FURNACE 

31. The electric furnace is the only apparatus in which a high 
temperature can be attained with a reducing atmosphere. A high 
temperature in any other melting furnace consequently impUes an 
oxidizing atmosphere. This in turn means that during the melting 
and superheating of the metal in the air furnace, those elements in 
the iron that will oxidize and remain oxidized will be eliminated to 
an extent that will depend upon the time the charge is exposed to 
the oxidizing influence of the furnace atmosphere. Practically, all 
of the constituents are oxidized in part, but only four remain in that 
condition. These four are the iron, silicon, manganese and carbon. 
Oxidation actually starts as soon as the charge on the furnace hearth 
begins to get fairly well heated up. The bulk of the oxidation 
takes place, however, just as the iron begins to melt, for at this 
stage the molten iron as it runs off the melting sides of the pig is 
freely exposed to the oxidizing furnace gases, and presents a large 
surface in proportion to its weight, with the resultant elimination of 



112 RESEARCH WORK ON MALLEABLE IRON 

some silicon, manganese, iron and carbon. The iron is oxidized to 
ferrous oxide, the manganese to manganous oxide, the silicon to 
siUca, and the carbon to CO which escapes as a gas. As the silica 
has acidic properties, while iron and manganese oxides are basic in 
their action, they will unite together to form a slag consisting of a 
double silicate of iron and manganese, although the composition of 
the slag as tapped is much more complex, being modified by the 
sand on the pig and sprue and the unpreventable erosion of the 
hearth side walls and bottom. During an average normal heat 
there will be a loss in siUcon of 0.35 per cent actual, a loss in carbon 
of about 0.35 per cent actual, and an average loss of about three- 
fifths of the manganese. 

REASONS FOR NOT USING WHITE PIG-IRON CHARGE 

32 The reason why it is impractical to use white pig iron can 
now be made clear. In the making of coke pig iron in the blast 
furnace, the amount of silicon and sulphur that will be in the product 
is largely a function of the furnace temperature. When this is low, 
there will be produced a pig in which the silicon will be low and the 
sulphur high. With a hot furnace the conditions are reversed and 
there will be obtained a pig high in silicon and low in sulphur. As a 
rule, a low-silicon coke pig means one that is prohibitively high in 
sulphur, but even if this were not the case, pig of this character could 
not be successfully used in the air furnace to make white-iron castings 
for the manufacture of malleable iron, because such castings must 
have a siUcon content between certain limits. If the sihcon is too 
low, then on annealing we will fail to obtain a casting whose fracture 
will be normal. If then we ased a pig as low in siUcon as obtains 
in the case of white pig iron, and with this were forced to use 35 
to 40 per cent of sprue that must of necessity be still lower in silicon 
than the charge from which it was made, then we would obtain 
metal too low in this element to yield good castings. The situation, 
however, is worse than this, l)ccause to lower the carbon to the neces- 
sary limit steel scrap must be used, which contains practically no 
siUcon, while by the time the molten metal is hot enough to be 
tapped from the furnace 0.35 per cent of the silicon in the charge 
has been eliminated. Fortunately there is no necessity of using 
whit€ iron, because it is a |x»rfcctly simple o|X}ration to convert pig 
iron that is gray in fracture, and in which the carlx)n exists mostly 
.•ee, into castings in which all of the carbon is combined. 



ENRIQUE TOUCEDA 113 

CONTROL OF SILICON CONTENT 

33 In Fig. 7 can be seen the fractures of six air-fumace test 
sprues. No. 1 of gray fracture was cast very shortly after the 
charge was completely melted, while No. 6, white in fracture, was 
cast when the metal was just hot enough to run the castings. No. 
2 was cast about 25 min. after No. 1 and the others at succeeding 
equal intervals. In comparing the fractures in the order in which 
the sprues were cast, it can easily be seen that as the silicon and 
carbon content in the bath were being gradually lowered, the graphitic 
carbon lessened by degrees, until finally the siUcon became so low 
that it no longer possessed the power during soUdification to drive 
the carbon out of combination with the iron. It is simply a matter 
of holding the molten iron in the air furnace a sufficiently long time 
to eliminate the siUcon down to a point where the amount that 
remains in the bath will be insufficient to precipitate any carbon 
when solidification takes place. Irrespective of how much siUcon, 
within reason, were in the charge, it would be possible to finally 
eliminate enough to accomplish this object, but as this would involve 
a great waste of fuel, it would not be commercially practical to do it. 
The logical thing to do, and the thing that is done, is to charge a 
mixture with a siUcon content just high enough to admit of the 
carbon combining, coincident with the arrival of the bath to pouring 
temperature. While too high a siUcon content wiU result in a waste 
of fuel and valuable time, too low a siUcon will not only defeat the 
obtaining of a superior product but the castings ordinarily wiU be 
unsound. 

PHOSPHORUS CONTENT 

34 The usual practice is to have the phosphorus content under 
0.20 per cent, but if aU the other elements are correctly proportioned 
and the white-iron castings correctly annealed, a much higher 
phosphorus content can be used and a very good product obtained. 
The writer is convinced that in the absence of combined carbon, 
considerable Uberty can be taken with this element in sections that 
do not exceed | in. It may also be stated that there seems to be 
no advantage to the quaUty of the product to run the phosphorus 
as low as 0.10 per cent, while there is a disadvantage in so doing that 
results from a lessening fluidity of the iron. Inasmuch as a pig of 
0.20 per cent phosphorus can be purchased as cheaply as one con- 
taining 0.30 per cent, there is no commercial advantage gained in 
using the latter. 



RESEARCH WORK ON MALLEABLE IRON 



ENBIQUE TOUCEDA 115 

SULPHTJK AND MANGANESE CONTENT 

35 In considering the proper sulphur content, we must at the 
same time take into accoimt the manganese, since due in a measure 
to their reciprocal attraction these elements imite to form man- 
ganese sulphide, as already explained. Provided the sulphur is 
properly balanced by the manganese, one need not fear injury from 
this element up to, say, 0.10 per cent, but approaching this point 
and beyond it to 0.12 per cent as the highest limit, great care must 
be taken not only to see that it is properly balanced by the manganese, 
but that the annealing temperatm^ be kept low. Just how low 
can be better imderstood when the rationale of the annealing process 
is briefly entered into. As far as the writer has been able to ascer- 
tain, there is absolutely no advantage to be gained and, as a matter 
of fact, considerable disadvantage in striving for a low sulphur 
content in the castings. It is getting increasingly difficult to 
piu-chase malleable pig iron that will average much below 0.65 per 
cent in manganese and it is not easy to obtain it this low. When 
this percentage in the mixture or charge is exceeded, it is quite diflScult 
if not impossible to obtain best results, unless some sulphur be 
present to control subsequently its action through the formation of 
sulphide. Under present average conditions the writer would 
prefer to see 0.05 sulphur in the product than one-half that amount. 

36 The manganese content in the product should not be lower 
than 0.18 per cent nor higher than 0.36 per cent, the former when 
the sulphur is at its low, and the latter when it is at its high, limit. 
This does not mean that a good product cannot be made with a 
slightly lower manganese provided the silicon is not abnormally 
low, ot with one sUghtly higher, but as a departure is made from 
the figures given, greater care must be exercised all along the Une. 
An excess manganese makes the material very sensitive to heat and 
when the manganese in the hard-iron casting is high, the annealing 
temperature must be kept at the lowest point at which the carbon 
can be precipitated. 

CARBON CONTENT 

37 Concerning the desirable carbon content for the hard-iron cast- 
ings, the writer believes that the higher the carbon the more easily 
can the carbon be precipitated in the anneal, but the weaker the 
product. It is impossible to obtain such a product as is now being 
turned out if the carbon be high. In looking over the Uterature of 



116 RESEARCH WORK ON MALLEABLE IRON 

even recent date one can find statements to the effect that a superior 
product can be made with a carbon content of over 4 per cent, which 
is further evidence in regard to what has been considered a superior 
product by some authorities. The Uterature of the subject is fraught 
with just such statements but invariably unaccompanied by data 
to back them up. To cover the ground briefly, it can be said that 
the lower the carbon, up to a point at which the carbon can be 
successfully precipitated, the more ductile and trustworthy will be 
the castings. If, however, the carbon is run too low, trouble will 
l>c experienced from lack of fluidity and in complicated castings 

TABI.E 3 CARBON ANALYSIS OF TEST BARS OF ELONGATION OVER 
20 PER CENT AND ULTIMATE STRENGTH OVER 52.000 LB. 



Total carbon, 


Elongation, 


Total carbon. 


Eloncation, 


l>cr cent 


per cent 


per cent 


pa- eent 


1.72 


21.00 


1.50 


20.50 


0.72 


20.31 


1.43 


21.09 


1.18 


22.00 


1.52 


22.66 


l.M 


23.00 


1.46 


21.88 


1.4rt 


21.00 


1.39 


21.09 


1.30 


20.00 


1.35 


21.50 


1.85 


22.00 


1.81 


25.78 


2.03 


20.31 


1.50 


20.31 


0.83 


25.00 


1.51 


21.00 


1.31 


25.50 


1.64 


90.50 


1.73 


21.00 


1.78 


20.60 


1.83 


22.00 


1.76 


20.00 


1.07 


20.50 


1.19 


21.00 


1.77 


20.00 


1.49 


23.60 


1.41 


2tV00 


1.62 


22.60 


1.55 


20 IX) 


1.70 


21.00 


1 01 


2l> Oi> 


1.42 


20.50 


1.75 


25 iX* 


1.52 


22.50 



contraction cracks will Ix? in evidence, while if the carbon is run 
still lower, a stivly fracture will exist in the castings. A carbon 
oinitont of 'J.;^.") {HT ivnt in the hani iron will yield sufficient fluidity 
for nu>st work :uul after such riMuovnl of the carbon as will take 
pluv during the anneal a touch, stn^nc: pnxluct will result. A few 
lip:invs in this ooniuvtion will prove illuminating. 

ivS 'V:\h\c 1^ oontaii^s the carlxMi analysis of many ban in all of 
\\\\w\\ il\o douirntion was over 20 ix^r ciMit and the ultimate strength 
over ,VJ.lHH^ 11 ^ 

i^i^ In oiilor that thoix^ will Iv no iloubi :k> to the reliaUlity <tf 
tlic carlHMi licttMininations. it can K"» statt\l that the drillings 



ENRIQUE- TOUCEDA 1 17 

secured by milling off the metal at the reduced part, the milling 
cutter passing over the entire cross-section until sufficient drillings 
were obtained for the carbon determinations in dupUcate. The 
drillings were then very thoroughly mixed. The carbons were run 
through by combustion in a platinum tube with all precautions 
deemed essential for trustworthy work. It can be further stated 
that all of the bars were regular ones sent in for test to this office, 
and it was unknown to any one that these analyses were to be made. 

40 Without practically writing, at too great a length on the 
subject, it would not be possible to cover completely certain facts 
of interest in connection with air-furnace work, proper construction 
of furnace, furnace operation and suggestions for further improve- 
ment in this direction. 

ANNEALING OF HARD IRON 

41 When a piece of air-furnace hard iron is gradually heated, 
a temperature is finally reached where many of its properties are 
very different from what obtains at a very sUghtly lower temperature. 
It will cease to be magnetic. Its structural composition will be 
different. The size of the crystals will be much finer than was the 
case under this particular temperature. It can be carburized be- 
yond its original carbon content if packed in a carbonaceous material 
and held at this temperature for a sufficient length of time, while 
if packed in material that yields oxygen it can be decarbonized 
almost completely if the piece is thin. Also the carbide of iron 
can be broken up into its two soft constituents at the temperature 
referred to. This temperature is called the critical temperature, 
or critical range, and for air-furnace hard-iron castings it is in the 
vicinity of 1440 deg. fahr. It is the lowest temperature at which 
hard-iron castings may be successfully annealed. This statement 
must be modified by the further statement that in an oven under 
perfect control this temperature is the one that would be selected. 
In practice it would not be safe to adhere too closely to it, for the 
reason that should the castings while being held "at temperature" 
fall under the critical range, it would undo in large measure what had 
been accomplished above it. In the annealing of the castings one 
of the things to be avoided is oscillating temperatures, or tempera- 
tures alternating above and below the critical range. For this 
reason it is necessary to select a temperature some 100 to 150 deg. 
fahr. above the critical, say, 1550 deg. fahr. in which event, even if 
due to carelessness the temperature does drop a Uttle, it will not be 



118 RESEARCH WORK OK MALLEABLE IRON 

liable to fall to a dangerous point. There is another reason why 
this latitude is deemed essential, though this does not obtain today 
to the extent it did formerly, due to the improvement in oven con- 
struction and operation. This has to do with the fact that in lai^ge 
ovens it requires considerable ingenuity to arrange flue openings, 
drafts, etc., in such a manner that the temperatures in all parts of 
the oven will be uniform, for which reason it is necessary to make 
sure that the temperature at the coldest comer is somewhat above 
the critical range, which will serve to safeguard oscillations in these 
locations. 

PREPARATION OF CASTINGS FOR ANNEALING 

42 In order to anneal the hard-iron castings that have previously 
been barreled or sand-blasted, chipped, gates ground oflF and in- 
spected, they arc packed in cast-iron pots where they are surrounded 
by an oxidizing packing. The packing has a dual function: to 
furnish oxj-^gen through whose agency the castings will be decar- 
bonized to the extent that is possible, and to avoid kiln warp, that is, 
prevent the castings from distorting. The pots, or stands, are 
sectional and each comprises a casting which forms its bottom, 
upon wliich four or five sections are superimposed. Elach section 
consists of a rectangular or circular "ring" by which name they are 
known, whether they are of the former or latter shape. These rings 
are about li in. thick and vary in size at different plants, depending 
upon the dimensions of the castings to be annealed, but they would 
average, if rectangular, about 14 in. by 24 in. by 14 in. high. In build- 
ing up a stand, a ring is placed on the stand bottom and then care- 
fully filled with coastings that arc surrounded with packing. When the 
ring is completely filled, it is hanmicred on the sides with a light 
sledge in order that the packing will run down and fill in all voids. 
The second ring is then placed upon the first one, and this is filled 
in the same manner, which procedure is followed until the stand of 
four rings is completed. In Fig. 8 can \yo seen an annealing oven, 
partly filled with stands that are each four rings high. The top ring 
is filled with (*astings only to about two-thirds of its height, for if 
tluiy w(M-o brought to the top thoy would be exposed to the oven 
gasos. Instead, th(^ top third of the ring is filled with packing and 
this in turn is covonMl with an iron plate. The top, and all joints 
in tho stand, are then mudded or luted in order to prevent the en- 
tran('(^ of the oven ^ases, after which the stand is lifted up by the 
charjring truck and placed in position in the oven. 



BNKIQDE TOUCEDA 119 

DEBCBIPTtON OP ANNBAUNQ OTGNS 

43 The ovens are usually of rectangular shape, and vary in 
capacity from 15 tons for a very small oven to 50 for the largest 
ones. Their average capacity is about 25 tons. The usual fuel is 
bituminous coal, but hard coal, powdered coal or oil are used. The 
matter of construction must be dismissed with the statement that 
these ovens are being standardized and designed with a determina- 
tion of securing uniformity of temperature throughout, and a great 
deal of study and experimentation is being devoted to this proposi- 
tion. The flues are not only being properly proportioned for the 



FiQ. 8 Annealing Oven pAari.T Filled with Annealing Pots 

draft used, but the flue openings so dimensioned that the beat can 
be drawn to any part of the oven in amounts sufficient to equalize 
temperatures, while provision is made whereby they can be easily 
kept clean. In the plants of the Association practically all of the 
ovens are under pyrometer control, and equipped at a central station 
where it is possible at any moment to ascertain the temperature at 
the hottest and coldest part of any oven, while a master pyrometer is 
used as a check on those that are permanently located. In this 
manner it is rather difficult for things to go wrong without detection. 
Air-^umace and annealing-oven operations have also developed from 



120 RESEARCH WORK ON MALLEABLE IRON 

extremely crude methods to intelligent control, and this in large 
measure accounts for the improvements in the uniformity of the 
product. 

LENGTH OF ANNEALING TIME AND TEMPERATURE LIMITS 

44 In annealing, the castings are brought "to temperature," 
that is, to 1550 deg. fahr. or as high as 1600 deg. fahr. if thought 
best, as rapidly as it is deemed they can absorb the heat. Too great 
a forcing of the heat during this period is avoided, for if it is done 
the rings expand much more rapidly than the material within them, 
which leaving a space between ring and contents will allow the 
packing to bleed down from the top towards the bottom of the 
stand, lessening the compactness in the upper rings. In average 
practice it takes about 48 hr. for the oven to arrive "at tempera- 
ture." The temperature of anneal is then maintained for a minimum 
of 48 hr., the time recommended being 60 to 72 hr. Firing is 
then stopped and the oven sealed tight in order that the castings 
wiU not cool faster than from 8 to 10 deg. per hr. while passing 
through the critical range. To safeguard this very important 
detail, this rate of cooling is maintained until the pyrometers indicate, 
that the oven temperature is less than 1100 deg., for on cooling, the 
castings are liable to be some 200 deg. higher than indicated by the 
pyrometer in the oven. After the temperature has been lowered to 
that point an opening is made in the front of the oven in order to 
allow it to cool more rapidly, for the reason that once the castings 
are at a temperature under the critical range no change can take 
place in their structural composition, so the only remaining pre- 
caution is to see that the castings do not cool so fast that inter- 
nal strains Ciin develop in them. From the foregoing, it can be 
readily seen that the average length of anneal occupies about seven 
(lays. From theoretical, as well as [)ractical considerations, the 
writiM- (Ux\s not believe that there is a possibilitj' of safely lessening 
the time for this oiK?ration, by much more than one day without 
taking chances. He has designed an oven in which the temperature 
of I (UK) (le^. can W easily attained in 25 hr. and he is aware that 
when the composition of the hard-iron castings is such that the 
lianl carbide is in its most unstable condition that even less than 48 
hr. will sullice for the precipitation of the carbon, so that these 
two iHM-ioils can be reduced somewhat, but danger is ever present if 
lilHMti(»s :uv taken with the cooling through the critical range* In 
order tiien to safejxuard the consumer as well as his own reputatioD, 



ENRIQUE TOUCEDA 121 

the manufacturer should make no serious attempt to shorten the 
anneal imduly. The annealing capacity should be such as to make 
the attempt unnecessary. If it is made, however, then the pyrom- 
eter element should be inserted directly into the center of the pot, 
placed for that purpose in contact with the side wall of the oven, 
in order better to determine when the castings have actually arrived 
"at temperature" and the moment when he can commence to 
record the time the temperature can be started on its downward 
course, which procedure will enable him to operate more closely and 
accurately. 

A CONVERSION RATHER THAN AN ANNEALING PROCESS 

45 From what has preceded it should be evident that the 
second step in the manufacture of these castings should not be known 
as an annealing process, but more appropriately as a conversion 
process. The dominant function of an anneal is to obliterate coarse 
crystallization or an unsuitable one, and replace it by the most 
suitable that it is possible to produce in the object treated, and 
iDcidentally remove internal stresses. Annealing does not imply 
structural changes in the piece when cold, aside from grain size and 
grain refining. In the anneaUng of maUeable-iron castmgs the 
dominant object is to convert white, hard iron, in which all of the 
carbon is combined, into a soft, tough, ductile iron in which no 
part of the carbon is in that state. In order to achieve this, it is 
necessary to maintain for a suflBcient length of time a temperature 
just in excess of the critical range, which, as has been pointed out, 
coincides closely with that at which grain refining occurs, so it 
happens that during the conversion both objects can be practically 
attained. 

STRUCTURAL CONSTrTUENTS IN IRON AND STEEL 

46 Before briefly entering into the rationale of the annealing 
operation and subsequently taking up certain facts in detail con- 
cerning the finished product and the conditions that influence the 
appearance of the fracture and physical properties, some particulars 
regarding the structural constituents in iron and steel should be 
presented in order that the exposition may be intelligently followed. 
Steel and iron crystalUze upon soUdification, and if a piece is properly 
polished and etched, the crystaUine boundaries can be developed 
and seen under proper magnification. If a sample of pure iron is 



122 BBSEABCH WORK ON MALLEABLE IRON 

thus treatetl, its struttural compositions is considered to coasist of 
100 per cent forrite, becaiL'^e the latter name has been given to iron 
which contains no carbon. (See Fig. 9.) If the iron were contami- 
natwl witli tlin other four impurities to which we have alluded, it 



if.sv Fki 

:. 11 Mll'lKXM 



wdiilil .•'till be I'riri^iilcrcd to consist of 100 per cent ferrite. For 
('x;iniplc, ftTiili' h:is till' imiiHTtyof disscilvinn silicon and its physical 
pnjpiiiics iMn Ijc cliiinjri'cl tlicnliy, under which conditiui however 
it woiikl .•'lill remain fi-rritc. 



ENBIQUE TOUCEDA 123 

47 Effed of Varying Proportions of Carbon. We have already 
seen that when carbon is added to pure iron it unites with it to form 
carbide of iron in definite proportion, to which compound has been 
given the name cementite. It happens, however, that cementite 
on forming invariably insists upon mechanicaUy mixing with ferrite, 
in the proportion of about 12 per cent of the former to 88 per cent 
of the latter, and will never exist free imless there is a deficiency' of 
ferrite to satisfy it in accordance with the above ratio. As the 
carbon ocmtent in iron is gradually increased from zero up to 
about 0.90 per cent, structures are obtained in which obviously the 
percentage of ferrite decreases from 100 per cent to zero, while the 
pearlite increases from zero to 100 per cent. Any carbon steel 
therefore that contains less than about 0.90 per cent of carbon 
consists of free or excess ferrite and that amount of pearUte that the 
carbon content was able to make. Fig. 10 shows a steel which 
contains about 0.20 per cent of carbon, the white constituent being 
the ferrite and the dark the pearUte. Under a carbon content of 
0.90 per cent, however, while not all of the ferrite will be used up by 
the cementite to form pearUte, all of the latter will be, so that none 
of it can under these conditions remain free, for it will all be used 
up in forming the mechanical mixture with ferrite, which in a slowly 
cooled steel will consist of alternate layers of these two constituents 
in the definite proportion of 12 per cent cementite and 88 per cent 
of ferrite. When a steel contains about 0.90 per cent carbon we find 
through calculation that it will form 12 per cent of cementite, and 
as the remainder of the material must be ferrite and figures to 88 
per cent, such a steel must contain 100 per cent of this mechanical 
mixture, to which the name pearUte has been given, the appearance 
of which in malleable iron can be seen in Fig. 11. All carbon steels, 
therefore, in which the carbon is less than 0.90 per cent consist of 
pearUte and some excess ferrite existing as such. If the steel con- 
tains more than 0.90 per cent of carbon, then its structural composi- 
tion consists of pearUte and the amoimt of cementite that was 
formed in excess of what was required to make pearUte, the pearUte 
growing less and the excess cementite greater with each increase in 
carbon content. Fig. 12 represents the structure of a steel con- 
taining about 1.10 per cent of carbon, the white rivers being the 
free cementite and the dark ground-mass the pearUte. The structural 
composition of both air-furnace white iron and white pig iron there- 
fore consists of pearUte and excess cementite, the former shown in 
Fig. 4 and the latter in Fig. 3, the white constituent being the excess 



124 RESEARCH WORK ON MALLEABLE IRON 

cementite and the dark the pearlite. While these constituents are 
stable under the critical range, on passing through and over it they 
undergo a change, and a new one is formed called austenite. 

48 Austenite, This constituent differs greatly from the others 
that we have considered, particularly in the fact that it is of indefinite 
composition; that is, no matter how poor in carbon the product 
may be, or how rich, we will obtain austenite on passing over the 
critical range, for in the former case we obtain an austenite lean in 
carbon and in the latter one rich in carbon. In order that carbon 
may be precipitated during the anneal, the iron must be in the 
austenitic condition, and if it were not for the fact that austenite is 
of indefinite composition the precipitation of the carbon would not 
be possible. The mechanism of the operation can be made clear by 
stating that during the anneal, as soon as the hard iron attains its 
austenitic condition, graphitization starts with considerable slowness, 
the austenite crystals rejecting some of their carbon in a minute 
nucleus and in so doing becoming leaner in carbon. This operation 
continues "at temperature" until the austenite is practically carbon- 
less and the carbon nuclei have balled up and grown through a 
segregation of the carbon into nodules. Coincident with this opera- 
tion some of the carbon diffuses out of the iron due to the oxidizing 
action of the pot atmosphere, which fact in turn would not be possible 
were it not that the austenite was of indefinite composition. It is 
a well-known fact that it is not possible to heat any iron product, 
for such operations as rolling, forging, annealing or hardening, in 
an oxidizing atmosphere without surface decarbonization. In these 
operations, however, the material is as a rule exposed to a temperature 
exceeding that of the critical range for a comparatively short interval 
of time, while in the case of the annealing of malleable iron we have 
seen that the shortest time to which the product is subjected "to 
temperature" is 48 hr. so that diffusion has ample opportunity 
to exert a very pronounced influence on the carbon content. 

49 That diffusion takes i)lacc right to the center of such sections, 
as are generally found in quite the heaviest malleable castingSy the 
writer has demonstrated in many instances, and when the niechanism 
of the operation is considered, it can be seen that this could hardly 
i)e otherwise as the principle involves the simple fact that if we 
liavo a layer of iron with a lower carbon content than its contiguous 
layer, carbon from the latter will diffuse into the former until equi- 
librium is established between the two. As the surface of the 
casting during the anneal is being continually impoverished in 



SNBIQUE TOUCEDA 125 

carbon, there must be a travel of the carbon from the center toward 
the surface, for it is correct to assume that once the process is started 
that part of the section from the surface to the center consists of an 
infinite number of layers, each sUghtly lower in carbon than its 
contiguous one. The underlying principles in the manufacture of 
cement steel, and in case-hardening are just the opposite of those in- 
volved in the anneal, and there is abundant proof that the carbon 
can diJGfuse into the iron right to the center of the section. As the 
rate at which the pot atmosphere acts to remove the carbon from 
the inunediate surface and the rate of diiGfusion of carbon from one 
layer to another is not the same, we can expect to see an almost 
carbon-free surface rim or frame in the section. It can be stated 
that were it not for the fact that the precipitation of the carbon was 
taking place coincident with its elimination, all of the carbon could 
be removed from the iron. 



MATERIALS FOR PACKING ANNEALING POTS 

50 In connection with the material used as a packing, con- 
siderable could be written if time permitted. The object to attain 
is to secure one that will evolve oxygen just as fast as the carbon 
can be eliminated, and no faster. If one is used that is too strong 
in this particular, then the surface of the casting will be attacked 
and we will have scaled castings, which are very troublesome and 
costly to clean. In addition to this we wiU run into another trouble 
that will be referred to later. If, on the other hand, it is too inert, 
we will fail to eliminate suflScient carbon and will not obtain the 
highest degree of ductiUty. The usual packing consists of a pre- 
dominating proportion of inert material, such as ground air- or 
blast-furnace slag, pulverized firebrick, etc., to which has been added 
iron oxide in some form, such as rolling-mill scale, hammer scale, 
etc. With a correctly sized inert packing it is possible to secure 
excellent results, provided the voids are of such capacity as will 
contain the right proportion of air to furnish oxygen for the reaction. 
This scheme is in a measure rather difficult to operate uniformly 
unless after each anneal the packing is again sized. 

51 It is the practice in some plants to anneal without packing 
in what are known as muffle ovens. Provided precautions are 
taken to have a tight muffle and the latter is well filled with castings 
to allow of its holding but Uttle air before the muffle is sealed, very 
good results can follow, provided further that the carbon in the 



126 RESEARCH WORK ON MALLEABLE IRON 

hard-iron casting is low. As a matter of fact very creditable results 
can be obtained from castings that are very low in carbon (not so 
low as to prevent carbon precipitating) even if no carbon is eliminated, 
which means that the lower the carbon content in the hard-iron 
castings, the less dependent one need be in connection with the 
extent of carbon elimination. This matter is rather important in 
some instances where a reUable grade of malleable that can be 
machined at very high speed is desired. A rim of decarbonized 
iron can be so soft that in some machining operations the metal 
will crumble and tear in front of the tool edge and generate so much 
heat that it will be softened. Many samples have been sent to this 
office with the statement that the material was hard, which when 
examined were found to possess the character of skin referred to. 
The skin was dead-soft, and the castings proved hard to machine, 
due to the reasons stated. The general worth of a casting, its 
machinability and peculiar physical characteristics depend so much 
upon the structural make-up of the finished product, that some 
time must be devoted to a consideration of the factors that act to 
prevent either the structure or fracture or both from being normal. 

APPEARANCE OF FRACTURE OF TEST PIECES 

52 It takes a lengthy experience before one can tell from the 
fracture of the iron what may have caused its abnormal condition, 
which is one directionin which great progress has been made. When 
a normal, well-annealed piece of malleable iron is subjected to a 
steady, direct pull in a testing machine, a point is reached at which 
the crystalline grains of which it consists are elongated permanently, 
the stretch continuing imtil fracture takes place. If the fracture is 
examined it will be seen that the grains have elongated into finely 
pointed spines which gives what is called a "tooth" to the fracture. 
When light falls obUquely on such a fracture, there is produced 
a play of colors that yields a sheen caused by a reflection from the 
points and sides of the spines and the shadows that fall between 
them. As the grains in the decarbonized rim are more ductile than 
the rest of the metal in the section, they will elongate to a greater 
extent, and if the fracture is held in certain directions to the lig^t, 
the width of the decarbonized rim can be seen by contrast, its color 
appearing under those conditions a little lighter than the rest of the 
section. (Sec Fig. 13.) This explanation is made and entered into 
because such a rim or border must not be confused with and mmt^lrftti 



ENRIQUE TOnCEDA 127 

for the character of fracture that has a well-defined frame, that is, a 
border having not only a sharp line of demarcation between it and 
the core of metal which it surrounds, but an appearance wholly 
distinct from it. If the writer were asked to pass judgment as to 



Fio. J3 Fractures Showing Width or Decarbonizisd Rdi 

the quality of a piece of malleable iron, based upon either the appear- 
ance of its fracture, or what would be shown by a polished and 
etched section under magnification, he is positive that he could 
render a more reliable opinion in the case of the former than would 



Fio. 14 Fbacttibes Dub to Low Silicon Accoufanied bi Low Carbon 

AND Manganese 

obtain in the case of the latter. The reason lies in the fact that 
even in a non-ductile product it is possible to have an absolutely 
normal structure, one which consists of a matrix of ferrite, throughout 
which are uniformly distributed nodules of free carbon, such as shown 



12S RESEARCH WORK ON MALLEABLE IBON 

in Fig. 4, while if the fracture is as has been described, a normal 
structure at least can be predicted. If the ferrite is not ductile, 
then the crystalline grains will not elongate, with the result that we 
obtain a structural appearance that would be interpreted by those 
not familiar with the facts as belonging to a piece that had been 
insufficiently annealed. It has already been stated that when the 
silicon ie too low a steely fracture will result, and also that the metal 
is liable to be unsound. In making this statement the writer is 
assuming that the silicon b low, not because too little was used in 
the charge, but low due to exce.isive elimination in the air furnace. 



Fig. 15 Micdograph of Specimen with 0.33 per cent Silicon and 
0.0S9 PER CENT Manqanese 

In such cases tlic low silicon will be accompanied by low carbon and 
manganp,sc. Such fractures are shown in Fig. 14. In Fig. 15 can 
lie scon the structure of a piece that had a silicon as low as 0.33 and a 
manganese of 0.089 per cent. The fracture of this piece was 
uniformly bright and coarsely crystalline. It had no frame. Tlie 
structure eonsi.it« of a matrix of pearlite and throughout it are 
(list riliu ted particles of undecomposed hard carbide (cenwDtife) 
and small, well-rounded nodules of gmphite, or temper carbOD, Bfl 
llijit rarbon is called which separates out during an anneoL TTie 
laslinK friPKi wliirh (his piece was taken was in an oveo in wbidi 
caslings of cmirt coni|K>sitinn annealed perfectly, and the | 



ENBIQUE TOUCEDA 



129 



of the particles of undecomposed carbide simply means that the 
amieal was not carried on for a sufficiently long time for this char- 
acter of product. The casting, however, would have been very 
inferior even if all of the carbide had been broken up. 

PICTURE-FRAME FRACTURES 

53 Very frequently low silicon-carbon-manganese of certain com- 
positions will 3deld what are known as picture-frame fractures, such 
as are shown in Fig. 16, which are typical and have the following 
composition: Silicon, 0.54; phosphorus, 0.162; sulphur, 0.053; 
manganese, 0.108; total carbon, 2.01. This piece when polished 
and etched showed the following characteristics: a decarbonized 
surface border, an inner ring of coarsely laminated pearUte, and 
within this a core corresponding in structure to that of normal 




Fig. 16 Picturb-Fhame Fracittie in Low-Silicon-Cahbon-Manganese 

Specimen 

malleable iron. Fig. 17 shows the decarbonized border surrounding 
the pearUtic ring. Fig. 18 the structure of the pearUtic ring, and 
Fig. 19 the core within the pearUtic ring. It is the presence of this 
ring of pearlite whose ductiUty is so much less than that of the 
metal in either the decarbonized border or core that produces on 
fracture the sharp line of demarcation between frame and core. 
While in this particular fracture the frame is fier>' bright and finely 
crj'stalline and the core black, there are picture-frame fractures 
that show various color characteristics of frame and core, but it 
will be found that invariably the frame has its pearUtic ring of 
greater or less breadth. The pearUte is not always coarsely 
laminated, but as a rule has the appearance and consists of an amount 
of pearUte that would correspond to a 0.35 per cent or 0.45 per cent 
normalized carbon steel. 




RES8ARCH WOBK ON 



ENBiqUE TOUCEDA 131 

54 If the sulphur in the hard iron is unduly high and particularly 
if not well balanced by the manganese, the casttogs will almost 
invariably show a picture frame on fracture, and especially is this 
true if the temperature of anneal is too high for such a composition. 
If the manganese is too high and not well balanced by the sulphur, 
the same result will follow. In each case there is an appearance 
to the frame and core indicative of which is which. The frames 
shown in Fig. 20 are rather typical of the latter. In this case the 
frame is dove-colored, while the core is black but sparkling. 

55 While time is not available to enter into a full discussion of 
what has been discovered in regard to picture-frame fractures, there 
are some points that can be recorded. There are some compositions 



that unquestionably have frame-producing tendencies. These com- 
positions will not produce a frame when annealed in an atmosphere 
that is not oxidizing. The surface structure of the hard iron has 
nothing to do with the problem as the writer has had } in. ground 
off of one side of hard-iron samples and upon annealing the frame 
was in evidence equally on all sides. It is beheved that the following 
facts are pertinent to the situation: Not only do certain composi- 
tions affect the ductility of ferrite, but the same is the case with 
a pearlitic structure. We can have ferrite that will elongate into 
very long spines and ferrite that will fail to elongate at all. We 
can have a pearlitic grain that can be ductile and those that are not. 
The forgoing is stated because the writer believes that whether or 



IKSEARCll WORK ON MALLEABLE IRON 



ENRIQUE TODCEDA 133 

not a frame will be produced in the fracture depends upon the 
breadth and ductility of the pearlitic ring, because a sUght pearlitic 
ring can be present within a decarbonized border without a picture- 
frame fracture being produced. It is his belief that whether there 
will be a pearlitic ring or not depends upon the rate of surface 
decarboniaation, as compared with the rate at which a dissociation 
of the cementite takes place. When conditions are such that there 
will exist a region between the decarbonized surface border and the 
core that will have a carbon content of about 0.90 per cent, equi- 
librium seems to be established in this region, and if any carbon 



Fig. 22 MicaoaBAPH of Malleabij: Ison of High Ultimate Strength 

passes from this region to the decarbonized border it is replenished 
by carbon from the core. 

THE PEARLITIC RING 

56 The writer believes that such is the case, and is of the opinion 
that perhaps the samples shown in Fig. 21 may have a bearmg on 
the case. A well-annealed bar was cut into eight pieces. A section 
from the first piece was poUshed and etched, and the other pieces 
all packed together and given another anneal. The second specimen 
was then prepared like the first. The remaining six were then 
given a third anneal, and the third piece poUshed, As this procedure 



ENRIQUE TOUCEDA 135 

was continued, it follows that the eighth bar had eight separate and 
complete anneals. It will be noted that this very faint pearlitic 
ring which shows whitish and very faint in the once-annealed bar, 
is very distinct and well defined in the second and that it is wider 
and further in toward the center. It will be seen that with each 
anneal the pearlitic ring has widened and has a smaller periphery. 
As it was found that wdth each anneal the total carbon content 
decreased, it is plain that once the pearUtic ring has formed, it does 
not act as a seal for the passage of carbon from core to surface and 
that the width of the pearUtic ring is buUt up and added to more 
quickly from the core than it is robbed of carbon by the decarbonized 
border. The matter can be sunmied up as follows: If conditions 
are such that a region containing about 0.90 per cent or less carbon 
is formed, this region while permitting carbon to migrate or diffuse 
through it will at the same time be incapable of having its carbon 
precipitated. Under proper conditions the region can alter its 
position and increase in extent. For such a region to have a start 
it is essential that there be a very substantial difference in carbon 
content in two parts of the section. 

STBUCTURE OF HIGH-STRENGTH MALLEABLE IRON 

57 Malleable iron of very high strength accompanied by a 
ductility that can be considered good enough for certain purposes, 
has already been alluded to. While no effort has been made to 
exploit this product as yet, it would appear to the writer that there 
is a very large field in which it could be used to advantage. In 
Fig. 22 can be seen the structure of a sample that stood an ultimate 
strength of 84,000 lb. and had an elongation of 5.20 per cent. This 
material can be made with uniformity and it is beUeved from experi- 
ments that have been under way for some time that a 90,000 lb. 
ultimate and a 10 per cent elongation might be uniformly maintained. 
It will be noted that the structure consists of a ground mass of 
pearlite, in which are more or less uniformly distributed nodules of 
temper carbon. The structure readily explains why the product is 
of high strength. 

EFFECTS OF HEATING MALLEABLE IRON 

58 As it is frequently necessary to heat the finished product for 
the purpose of straightening it, for galvanizing and other purposes, 
it may prove instructive to see what happens when an annealed 



nKSKARCil WORK ON MALIJIABLE IROS 



ENRIQUE TOUCEDA 137 

piece of malleable iron is heated up to and beyond the critical range. 
Ten pieces were cut from a normal, well-annealed malleable-iron 
bar. Fig. 23 shows the structure of this bar at X200. The ten 
pieces were then placed in an annealing oven in the writer's laboratory 
that can be controlled with great accuracy. When the pyrometer 
registered 1250 d^. fahr., one piece was withdrawn and allowed to 
cool in the air. When the temperature reached 1300 deg. fahr. 
another piece was withdrawn. The other pieces were withdrawn at 
temperatures of 1350, 1400, 1450, 1475, 1500, 1550, 1600 and 1675d(«., 
respectively, Fig, 24 showing structure at 1400 deg., F^. 25 at 1450 



Fid. 32 Example of Large MALLBABLE-IitoN Castlvo 

deg.. Fig. 26 at 1475 deg.. Fig. 27 at 1500 and Fig. 28 at 1675 deg. 
All of these micrographs were taken at a magnification of 200 diam- 
eters. It is apparent that no change has taken place in the struc- 
ture up to 1400 deg., but that somewhere between 1400 and 1450 
deg. the 8tructiu« starts to alter in appearance. An examination of 
Figs. 25 to 28 shows that increased amounts of pcarlite result as 
the temperature is increased, and that in Fig. 28 nearly all of the 
temper carbon has been dissolved. It follows that for straightening, 
brazing and other operations that necessitate the heating of a 
malleable-iron casting, the temperature used should be well under 
1400 deg. fahr. 



138 RESEARCH WORK ON MALLEABLE IRON 

59 In order to obtain some idea of the mechanism of what 
takes place, micrographs at a magnification of 750 diameters were 
taken of the crystalline bomidaries of pieces corresponding to Figs. 
25, 26 and 27, which are shown in Figs. 29, 30 and 31. It would 
appear that as the carbon is dissolved the solution takes place through 
the amorphous iron in the crystalline boundaries, a fact that the 
writer has not heretofore seen mentioned. This is possibly the 
manner in which the carbon starts to diffuse into iron during cementa- 
tion. What might be written in connection with the numerous 
observed facts pertaining to this product would fill many pages. 
The writer has lacked time to even take up the matter on which 
he has written in the way in which he would have liked to have 
presented the subject, but he believes that enough has been covered 




Fig. 33 Example of Abuse Malleable Castznos will Withstand 

to illustrate that through research and tlirough that means only 
can the manufacturer make real and permanent progress. 

60 For the benefit of those who labor under the impression 
that malleable iron is a product unsuited for any but small castings, 
Fig. 32 is shown. One casting of which the writer has a photograph 
is 5 ft. long, 23 in. high and some of its sections are 3 in. thick. 
That the metal, when well made, can stand great abuse is illustrated 
in Fig. 33; while the problem in regard to disproportionate sections 
that confronts the manufacturer, coupled with intricacy of design, 
and too often the intricacy of ill design, is well illustrated in Figs. 
34 and 35. Irrespective of how good the metal may be, as shown 
by plu'sical tests on bars and wedj^es, patterns are furnished so 
outrageously out of proportion that the good qualities of the metal 
can easily be destroyed and i{)0 frequently the metal is blamed 
when the design is at fault. These troubles are overcome as far as 



ENBIQUE TODCEDA 139 

poeaible by a thorough study of the best method of gating the casting, 
in order that no evidence of shrink will be present in any part. 
Research has made considerable advance in this direction and well- 
placed, n-ell-proportioned shrink heads are now generally used. Not 



Fig. 34 Example of Castiko of Disfropohtionate Sectioxs 

infrequently, in the case of such castings as the ones referred to, 
the sprue, runners and heads weigh about as much as the casting. 



COllUOJI FALLACIES 

61 Before taking up the matter of fallacies, which during the past 
few years have been pretty well exploded although some still linger 




Fig. 3c Example of Cabtino of Disproportionate SEcnoHB 

in the minds of a few of the engineers and consumers, the writer would 
like to make some further remarks by way of explanation concerning 
what was stated in the first part of this paper in connection with the 
literature of the subject. Enough has been shown to make clear 
the fact that in order to make a good quahty of malleable iron, 



140 



RESEARCH WORK ON MALLEABLE IRON 



it is absolutely necessary that the white-iron-casting composition 
be correct. Many very interesting and laborious researches have 
been made in connection with the precipitation of the carbon during 
the anneal and in numerous other directions on hard-iron samples 
whose physical properties would be so low as to be worthless when 
annealed, due to their impossible composition. Had the same work 
been done by the eminent men who have carried through the experi- 
ments rcfcrn^d to on a hard-iron composition that was normal in all 
particulars except in connection with the particular element they 
were inv(\s(igating, the metallurgist in this particular field would 
have been greatly l>enefitod and his path made easier. 

02 The fallacies that have l)ecn handed down and accepted by 
many of the engineers and consumers as true, are numerous, but 



TABLE 4 TESTS OF MALLEABLE-IRON BARS WITH DECARBONIZED 

SKIN REMOVED 





Mark 


Ultimate Btrength, 
lb. per 8q. in. 




Per cent elonc*tion 
in 2 in. 




12-2-1 


52.084 




17.60 




12-3-1 


47.182 


1 


10.00 




12-3-2 


51,107 


( 
1 


17.50 




12-1-1 


56,732 




14.00 




12 r>-i 


46.482 




7.00 




12-.V2 


52,246 




23.00 




12-6-1 


47.889 




19.00 




12-7-1 


48.080 


F 


18.00 




12-7-2 


49.610 




18.00 



the following only will be toucheii ujwn as they are the most im- 
portant : 

a The strength of malleable iron lies in the skin. When it 
has been reinoved the remainder of the metal is found 
to he very inferior and not dependable. 
// Durintc the anneal, the elimination of the carbon is con- 
fined to the surface, aiul the amount removed from the 
rest of tin* s(M'ti4»ii is inennsecjuential. 
c ^\'llen the section of a castiiii: excee<ls I in. in thickness^ 
it cannot 1»(* ann(\ile<l thnni^hout. 
t)3 ( '(Micernintr itiMn (7. the «iata in TaM(» 4 will prove of value. 
Nine reiiular test i>ars were niacliiiMMl until the clecarlx)niEed surface 
was remove. 1. 'I1ie<t^ bars were all I'nMn jJitYerent heats and marked 
as indicated in tht^ table. 



ENRIQUE TOUCEDA 141 

64 As the writer did not have duplicates of these bars, he was un- 
able to make a comparison between the machiaed and the bars as cast 
and lacked time to nm through a set for illustration, but the experi- 
ment should be unnecessary in any event in view of the above. It 



is obvious and must be acknowledged that the metal in the decar- 
bonized skin is more ductile than the coi'e, so when a bar fails it must 
be conceded that it is the core that has parted first, for the reason 
that the metal in the skin has not at the instant of fracture reached 



142 RESEARCH WORK ON MALLEABLE IRON 

its maximiiin elongation. Aside from the foregoing we have the 
practical evidence that presents itself in the case of the automobile 
industry in which thousands of tons of machined malleable-iron 
castings are used annually on parts that receive in service great 
abuse, such as wheel spindles, etc. On the other hand, the writer 
not only admits that when the skin is machined off some malleable- 
iron castings the remaining part is worthless, but admits as well that 
the castings would be such with the skin on. This, imfortunately, 
will continue to be the case imtil the purchasing agents cease to 
shop around and a contract is made on price as the basis rather 
than quality. 

65 Taking up item 6, the writer can, without encumbering this 
paper with the large amount of data he has on the subject, prove 
the falsity of this contention. In the figures quoted in Table 3 
for bars of over 52,000 lb. ultimate strength and over 20 per cent, 
elongation, there will be noted two bars, one of which has a carbon 
content of 0.72 per cent and the other of 0.82 per cent. Aside from 
this there are fourteen with a carbon content of 1.50 and under. In 
Par. 39 an explanation is given of the manner in which the drillings 
were taken for analysis. It has already been pointed out that the 
carbon in the hard iron must be kept up to a certain figure, failing 
which the castings will not only misrun, but contraction cracks will 
spoil them. If we assume, in the case of the first two bars referred 
to, that the carbon was reduced by one-half, then in the bar that 
had but 0.72 per cent carbon, the carbon in the hard iron from which 
it was cast must have been 1.44 per cent, and we all know that it 
would be almost impossible to run such work, say nothing about 
subsequently annealing it. In a |-in.-diamet«r annealed bar such a 
low carbon content is unusual, but it proves the point that is being 
made, nevertheless. The writer has polished the section of two f-in.- 
bars and has photographed them at about seven diameters. These 
are shown in Figs. 36 and 37. They will furnish a fairly good idea as 
to how the carbon is distributed throughout the section, and indicate 
that the carlx)n does not vary by uniform gradation from surface to 
center, but in one region can vary slightly from what it may be in 
another. This does not signify that in the regions of highest carbon 
content the carbon lias not been lowered through diffusion into its 
contiguous region, for many investigations have shown that this is 
just what do(*s happens. 

66 Item c can be disproved in a few words. We know that in 
order to break up the hard carbide in whit^ iron it is simply 



JSi^^l-f 



ENBIQUE TOUCEDA 143 

that the casting be not only heated until the iron is in an austenitic 
eonditioni but maintained at that temperature for a certain interval 
ol time. To state that thick castings of white iron cannot be an- 
nealed is to state that they cannot be brought to a uniform tem- 
perature throughout and maintained at that temperature. Such a 
daim would be an absurdity. 



No. 1682 

INDUSTRIAL PERSONNEL RELATIONS 

By Arthur H. Young,* Chicago, III. 
Non-Member 

npHE present focusing of attention on personnel relations will 
result in a new era in industry — as much an epochal change 
as we have had through various fundamental causes heretofore; 
such as the change from the original craftmanship to larger shop 
organization, made possible by power development, power trans- 
mission, changes wrought by methods of communication, and re- 
finement of the methods of transportation, the era of consoUdation 
of interests, and so on, and one has to stop and wonder why at last 
we have come to the consideration of the human factor. 

We have probably had as great a refinement as would produce 
any revolutionary changes in machinery and methods and prac- 
tices in everything except that relating to the human factor. All 
of us have witnessed the birth and the growth and the final develop- 
ment of the safety movement, and more recently, too, of employ- 
ment management. 

I refer to these two developments, particularly, as factors in 
personal relations of today, and illustrating how rapidly changes 
are coming about. I think it was only back in 1906 that the first 
organized effort in safety was made. It is generally credited to the 
South Chicago plant of the Illinois Steel Company, under the leader- 
ship of Mr. R. J. Young. He was a sort of genius who suddenly 
discovered that accidents were preventable — that it was no longer 
necessary to kill and maim men as a part of the making of steel. 

DEVELOPMENT OF THE SAFETY MOVEMENT 

At first not much attention was paid to his proposition, but it 
rapidly gained support, due to its humanitarian appeal, its evangeli- 

* Manager, Industrial Relations Department, International Harvester 
Company. 

Address, slightly condensed, delivered at the Industrial Relations Session 

of the Spring Meeting, Detroit, Mich., June 17, 1919, of The American Society 

OF Mechanical Engineers. 

For discussion see p. 186. 

145 



146 INDUSTRIAL PERSONNEL RELATIONS 

cal aspect. The first attack was naturally very largely along me- 
chanical lines — the removal of projecting set screws and other 
parts of revolving machinery, the railing of platforms, the guarding 
of gears, changes in construction, marking of aisle spaces, and all 
such things. 

A very disappointing result was had, because at the end of two 
or three years, after several hundred thousand dollars had been spent, 
at that particular plant accidents had not been reduced over 20 
per cent, and generally speaking, today safety engineers rate the 
mechanical factor or the correction of engineering practices as not 
contributing more than 20 per cent to the possible efficiency of the 
safety movement. 

It was immediately discovered that the real problem was 
psychologically to teach a naturally careless man to become habitu- 
ally careful, and that more than 75 per cent of our accidents were 
preventable by a correction of habits of thought — of thinking 
safety all the time. To illustrate: It is perfectly natural to cross 
a street just exactly as a chicken does — to look neither to the right 
nor the left, but step right out from the curb; and it is only by a 
process of education that we learn to pause at the curb and look up 
and down the street and then walk into the middle of the street 
and then glance again, and finally to cross the street. That is a 
process of education which will prevent accidents, and it was what 
was immediately arrived at in industrial safety engineering. 

The shaping up of that problem brought its immediate solu- 
tion, and brought into play a most interesting development from 
the laboratory of safety engineers. There were movies, bulletin 
boards, mass meetings, and finally the formation of public safety 
councils, the introduction of the study of safety in the schools. 
Its serious consideration by civic poUce departments and other 
parts of society is really the product of the industrial safety engi- 
neer, because he reaUzed that in order to make a man think safety 
all the time it was necessary to work on him not only while he was 
in the shop, but at home as weU. He carried his message to the 
man in the shop through the children whom he encountered at 
school, and during the hours that he was walking to and from the 
plant or riding to and from the plant by pubUc safety bulletins and 
so had a continuous propaganda working 24 hours a day in order to 
make him think safety. 

Immediate results were achieved. Taking again, for example, 
the South Chicago plant of the Illinois Steel Company, with 



ABTHT7B H. YOUNG 147 

I happen to be familiar, back in 1906 the frequency of fatal acci- 
dents in that plant was 47 during the year, or about one a week. 
By 1913 that rate had dropped to 7, and it has since stayed at about 
that figure. 

As in fatal accidents, so in the less serious accidents, the fre- 
quency rate dropped in the same proportion; and in the Illinois 
Steel Company, there has been a reduction in the frequency of 
accidents to employees on duty of over 85 per cent since the 1906 
rate. That was duplicated in nearly every plant of the United 
States Steel Corporation, and in thousands of other industrial con- 
cems^as well. 

There have been a nimiber of by-product results which have 
directly contributed to the growth of personnel movements and 
industrial relations, as we term them today, coming out of this 
safety movement. In the first place, it was found that not only 
did this human interest conserve lives, but that it was a paying 
proposition; originally it was an interest in the cause of humani- 
tarian principles, and safety was regarded as somewhat of a fad by 
a great many employers. 

After a suflScient period of experience on which to establish 
definite figures, it was found that safety was really "good business." 
The Steel Corporation has published the most interesting figures 
along this line, showing a net saving almost exactly equal to the 
cost of the safety work, or 100 per cent on the investment. 

There is a striking unanimity in the form of organization for 
efficient safety work. The original movement in 1906 resulted in 
the formation of shop committees; as soon as it was realized that 
the problem was one of education of the workman and that his 
interest must be aroused, means were sought to turn the problem 
over directly to him, and I believe the first safety conmiittees were 
organized at the South Chicago plant of the Illinois Steel Company. 

Shop committees — the workmen themselves — were given 
charge of the safety work. It was their duty to investigate the 
cause of each accident, fix its responsibility, and make recommenda- 
tions for a prevention of its recurrence. In addition, they were to 
make regular inspections of the plants, and by their foresight and 
recommendations were to prevent, as far as possible, the recurrence 
of accidents. 

Some form of a shop committee is almost unanimously used by 
safety workers today in all industrial establishments of any size. 
It may be of the workmen themselves, it may be workmen in one 



148 INDUSTRIAL PERSONNEL RELATIONS 

committee reporting also to a committee of foremen, or it may be 
a joint committee, or it may be a joint committee of workmen, fore- 
men and officers of the company, but the successful safety program 
of today includes shop committees. 

In addition to the definite return on a cash basis of the safety 
work, there have been several interesting by-produipts. The first 
is the reduction of labor turnover. The Steel Corporation's reports, 
recently published, show that since 1906 they prevented, roughly, 
25,000 fatal or serious accidents. Thoy rate as serious any accident 
which disables a man more than 35 days, or results in a permanent 
disability, such as the loss of a thumb, or an eye, or other member 
of the body. 

It is reasonable to assume that s:uch an injury would require 
the replacement of the trained worker, and Captain Fisher and 
other writers on the subject have given some interesting figures as 
to the cost of replacement of trained workers. They vary anywhere 
from $10 to $100, depending somewhat on the way it is figured, 
and somewhat on the worker. Of course it is fair to say that steel- 
mill workers are rather well trained, and probably $100 as the cost 
of replacing a trained man with an inexperienced man is not an 
over-estimate. If we multiply 25,000 prevented replacements of 
trained workers, because of the efficiency of a safety movement, by 
$100 we hiwe a substantial by-product, due to reduced lalx)r turnover 
through efficient safety work. 

There was also another manifestation. As is generally known, 
the Steel Corporation lias been facing a suit for dismemberment. 
While the hearings were on several years ago there appeared volun- 
tarily a number of old employees who petitioned that the corpora- 
tion l)e not dissolved, l>ecause that would mean a return to the days 
of ruthless competition when they did not have standardized safety 
programs. Of course, no one knows what effect that action had on 
the decision, but it presented concrete evidence of the boosting of 
the morale of flu* workers through safety work. 

THK QIJKSTION OF EMPLf)YMKNT MANAGEMENT 

Turning for a moment to employment management, I wiU 
cite i\u) circumstances of my first employment. At the age of thir- 
t(»on years I w(*nt to work in a steel mill during vacation time and 
reported to the eiiginecM* who turned me ov(t to his assistant. The 
assistant said, "Do you know anything about oiling?" Being 
thirteen years old, naturally I knew everything about anything. 



ARTHUR H. YOUNG 149 

He said, "AD right, here is the cylinder oil, and there is the beeswax 
and some sand. Here is a ladder. Go to it!" And that was the 
extent of my instruction in oiling. 

I well remember the first day. An emery-wheel explosion 
occurred. It was frequent in those days, just as frequent as crane 
runway accidents were, and those things that are not heard of any 
more. 

The third day we had a fire caused by an overheated bearing. 
I discovered the fire when it first started, threw a bucket of sand 
on it, and put it out. 

I still have distinctly a feeling of horror as I think of the days 
that I climbed on a rickety ladder, unsafe as imsafe could be, up to 
a high-speed shaft. In those days we oiled while the machinery 
was in motion, and the high-speed pulley on one side of me without 
any protection whatsoever, and couplings and collars on the shaft, 
had projecting set screws and bolts galore. There was absolutely 
no thought given to safety. 

I also think of the possibility of a great financial loss through 
the neglect properly to educate a new man in a rather important 
and dangerous job. If that fire had not been discovered within the 
first five minutes it would have swept through the shop. 

PROCEDURE OF A PRESENT-DAY EMPLOYMENT MANAGER 

Contrast such methods with the employment program in effect 
in many shops today. In every well-managed institution there is 
an employment manager, not necessarily known by that title, but 
functioning as such, to whom a request to fill the vacancy of oiler 
would be referred. In those days if there had been a vacancy the 
assistant engineer would have gone to the gate and picked a man 
from the waiting crowd, and a crook of a finger would have put 
that fellow on the pay roll. The paymaster would have been in- 
formed that this man was oiling. 

Today there would be a process of selection. Instead of the 
group hanging around at the gate, there would be a well-furnished 
employment office with a waiting room, and when the request for 
an oiler was received by the employment manager, he would go 
out and seek a man who had some experience or who was adaptable 
for that position, and he would also know in full the duties and just 
what specifications he would have to meet in getting the right man 
for that job. 

If the man whom he chose were a foreigner that man would be 



150 INDUSTRIAL PERSONNEL RELATIONS 

interviewed across a desk by a man who could talk in his own lan- 
guage, and be invited to tell in full his past experience. K he lied 
about it, that would show up in the replies to the letters of investiga- 
tion that are sent out as a rule. 

I am not one of those who believe that the employment man- 
ager is the final judge of the fitness of a man for the job; that rests 
wholly in his performance on the job, and the foreman is probably 
the final judge there, but certainly the employment manager can 
by intelligent first sifting, by coarse sifting, get rid of many palpably 
undersirable applicants. The appUcant who seems after that sift- 
ing to be the man for the position would then be given a complete 
physical examination, not necessarily to reject him if physically 
impaired, but to make sure that he would not be assigned to a job 
that would further injure him if he had any physical disabiUty. 

If he were acceptable after that physical examination he would 
probably be told something of the policies of the company; be given 
an introduction to his employer — the corporation — and be told 
of its safety program and handed a book of rules of conduct. He 
would probably be told that this rule book was made up by the 
Safety Committee, employees in the shop, and that if he did not 
fully understand it and wanted to quiz anybody on it he had only 
to seek the nearest old-timer who had been concerned in the revision 
of the book, and he would be given full and complete information. 
Undoubtedly, he would have greater respect for the rules when he 
learned that they were made by the men themselves. 

Then he would be told of the promotional opportunities, — 
that it was the policy of the company to promote from the ranks, 
not to hire anybody at a given wage unless they were satisfied that 
nobody then in their employ at a lower wage could be promoted to 
fill that vacancy. He would be quizzed particularly as to his ambi- 
tion. What did he want to be? He would not necessarily be put 
upon the job as an oiler permanently, but in an efficient filing system 
for recording applications his ambition would be registered. If he 
aspired to be a machinist and had stated that he had had certain 
experience as a bench hand and was attending night school, that 
would be recorded; and later, when a machinist was desired, prefer- 
ence would bo given those applications from employees already in 
the scTvice. Probably at tliat time the employment manager would 
send for tliitj chap and say to him, "When you first went to work 
here six months ago you were put on as oiler. I understand you are 
still doing it, and you said you wore going to night school to learn 



ABTHUR H. YOTJNG 151 

to read blueprints and take up manual training, and so on. Just 
how proffieient are you as a machinist? Now I have a vacancy 
and you might be of use." And possibly he would be fitted into 
that job. 

When he was hired there would be entered on his employee's 
record card a pretty complete history of him, his social condition, 
whether married or single, how many children he had, just how 
many dependents he had, where he was bom, his age, his schooling, 
his military service, if any, whom to notify in case of sickness, result 
of physical examination and so on, for statistical purposes and then 
he would be sent to the job on sort of a personally conducted tour. 
He would be taken by an agent of the employment oflBce, watchman, 
or special guide, and first shown the gate nearest his work so that 
any unnecessary hazard of traveling through a long plant might be 
avoided in coming to work. 

In the meantime, if he were a single man he would be assisted 
in getting a proper boarding house. If he were a married man, 
temporarily located in a boarding house, the employment agency 
would probably show him a map of the vicinity of the plant and the 
desirable residential districts for his type, and give him a list of 
reputable real estate men as an assistance to him in locating per- 
manently, and show him the street-car lines whereby he might best 
get to work. Finally, when delivered to the foreman for whom he 
was to work, he would be introduced by his own name and learn the 
name of his foreman. Nowadays, the worker, whether he be foreigner 
or American bom, is introduced to his foreman by name and the 
guide makes sure that they understand each other. 

The foreman probably repeats "John Sobrinski — did I get it 
right? Well, my name is John Smith. I am your foreman." And 
the last word of the personal guide is, "The labor supervisor par- 
ticularly requests you, Mr. Foreman, to give this man any special 
instructions necessary in the particular hazards of his job." And 
then and there the foreman, either himself or through a fellow- 
workman, tells this man something about his job, and introduces 
him by name also to one or two of his neighboring workmen, so they 
will know each other, and the man can go to his fellow-workmen 
and seek and get information. 

Some employment agents go still further, and during the proc- 
ess of hiring a worker give him a movie show picturing the com- 
plete operation of the plant — taking it from the receipt of the 
raw material, through the various manufacturing processes, on 



152 INDUSTRIAL PERSONNEL RELATIONS 

through shipphig, just as general as it may be or just as complete 
as it may 1x5 in the space of time available, and increasingly that 
time available is lengthening. 

We used to brag in employment management about hiring a 
man in a minute, or half a minute, or in a number of seconds, or 
about the number we could put to work in a year. But that has 
clianged; now we are beginning to boast of taking an hour and a 
half, and even a week and a half to put a man properly on his job. 

Now, it needs no more tlian an appreciation of that contrast 
to bring to mind how well sold a man may be on his job, how much 
he may be made to feel tliat he is a constructive part of that estab- 
lishment, if he is put to work by the modern method, as contrasted 
with the fonner. If when he is put on the job the foreman calls his 
attention to the movie pictures of the particular operation that he 
is engaged in, and if it only be wheeUng cinders away from the central 
boiler plant — if he is told at the time tliat that is the central boiler 
plant that furnishes the power to run the plant that does the things 
that he saw in the movies and tliat these cinders are the refuse from 
the boiler tliat generates all that power — then and there he links 
himself to the plant as a constructive force, as an important part 
of that machine. This is very necessaiy if we are going to settle 
this unrest that we now feel. 

The unrest in industry today is evidence of a social, a poUtical 
unrest that must be met freely, frankly and squarely, and corrected. 
In fact, it is manifested as Bolshevism in its ultraradical form. It 
is no less than tliat. Probably when the liistory of the Bolshevik 
movement in Russia is finally writt^^n, and we know just what its 
causes wore, it may be found that a large contributing cause was 
the fact tluit the citizc»ns of that country had no part whatever in 
their government. Things were done at them and far them, but 
never by them. Th(»y were never taken "in on the know." 

Whil(» it is ix)ssible tliiit their condition as citizens of that coun- 
try under despotism was as good, their living conditions and all 
otlier conditions just as fine as they themsi*lves might have done 
throuj^h demoeracy, th(*y never had that filling because they had 
not been consulted. It is simply a staU'inent of basic psychology 
to siiy thiit we are only mildly inten^sted in the tilings that other 
j)('o|)le are doinj»: for us, but we an* intensely interested in the thinge 
that we are doing for oui'sclves. 

Cloing just a little further, if we luive seen, as in the safety 
movemrnt, the efiieiency that may Ix^ gauied through consulting 



ABTHUR H. YOUNG 153 

the workmen themselves on such a matter of common interest, is 
it not perfectly logical to go to the men themselves for consulta- 
tion on other problems of mutual interest — recreation, sanita- 
tion, health, and then the controversial matters of hours and 
wages? 

It has been^stated by Professor Hoxie that naturally the aims 
and purposes of employee and employer are antagonistic; that 
the employer seeks long hoiu^ and low wages as a means to a low 
cost of output, and that conversely the employee is constantly 
seeking shorter hours and higher wages, which must necessarily 
mean a decreased output. 

There is abimdant experience in the industrial world today to 
show that the reduction from twelve to ten hours in the length of 
time employees are working has not decreased production — that 
the reduction from ten to eight hours has been accomplished in fac- 
tories without any decrease in the output, and that today reductions 
are being made from nine to eight hours without any decrease in 
the output. Shorter hours do not necessarily mean reduced output. 
Neither do higher wages necessarily mean increased cost, if they 
mean a higher standard of li\ang, a better mental and physical 
development. I firmly believe that there is a common ground upon 
which employer and employee can meet for the consideration of 
their problem, and it need not be a controversial affair either in 
the consideration of wages and hours any more than any considera- 
tion of safety or health or recreation or plant canteens or anything 
of the kind.. 

THE INTERNATIONAL HARVESTER COMPANY ^S PLAN OF 

EMPLOYEE REPRESENTATION 

Our plan is known as the Harvester Industrial Council plan of 
employee representation, and wjis offered to the employees of the 
company on March 12 of this year. It was a frank invitation to 
them to participate in the determination of the policy of the com- 
pany on all matters of mutual interest, including wages and hours, 
on an equal basis. They were invited to elect, by secret ballot, 
one-half of the membership of a works council whose function was 
to determine the policy of the company on the various items I have 
enumerated. The management appointed the other half of the 
membership. A guarantee of equal participation was had by the 
adoption of a unit ballot system. There are only two ballots cast. 
A majority of the employees' section determines their attitude and 



154 INDUSTRIAL PERSONNEL RELATIONS 

casts a unit ballot, as also does the majority of the management 
representatives. 

These unit ballots have the same value, regardless of the num- 
ber present on each side. The employees are guaranteed the right 
to a free performance and a free action in all of their activities as 
employee representatives. If there is any question of discrimina- 
tion on their part, they may appeal directly to the president, and 
if not satisfied with his adjudication of the matter, it may then be 
arbitrated upon the selection of an arbitrator, mutually agreeable, 
whose decision would be binding upon both parties. 

The plan was not put into effect at any plant which did not 
vote by considerable majority for it. It was first offered at the 
seventeen American and three Canadian plants. It was adopted 
at seventeen and failed of adoption at three. The day after the 
failure to adopt at these three plants petitions were circulated ask- 
ing for another ballot and an opportunity of coming under the plan. 
The statement was made that the employees did not fully under- 
stand the plan on the previous day because it was written only in 
English. We have since published it in foreign languages because 
the employee representatives find it exceedingly difficult to convince 
foreign-bom employees of the exact meaning of a certain clause if 
they themselves have to translate it. 

An error was also made in the form of ballot. The ballot stated 
"For adoption. Against adoption." Some very good friends on 
the outside of the plant who were working against the plan told 
many foreigners to vote on the bottom line. When they saw the 
phrase "Against adoption" it did not mean anything to them and 
they voted according .to instructions on the bottom line. 

There was no positive effort on the part of the company to 
"sell" the plan. It was simply a dignified announcement. Each 
employee was furnished with a copy of the plan with a short r^umd 
of a facsimile of the ballot, and asked to indicate, after three days 
of consideration, his wish for or against it. At the present time 
the plan is in operation in nineteen of the twenty plants erf the 
Harvester C\)mpany. 

Aiu^thor fumianiontAl included in the plan is the guarantee of the 
pn^tootion of tho employee against any discrimination because of 
r«iv or st^\ or inomN^rship in any religious body or kbor organiia- 
lion. 

VVNOriON OF THE WORKS COUNCIL 

I1v txuu^iion of tho works council is limited to the determina- 
v,^- ,n; ^^o |x^Ux\\ ol tho cinnpany with reference to wages, houTB, 



ABTHUR H. YOUNG 155 

recieation, health, sanitation, restaurants, and other matters of 
mutual interest. A policy having been determined, its execution 
Ues wholly with the management. But the manner of execution 
being open to question at any time, it may be brought up through 
the works coimcil. In other words, we have given to the employees 
equality in participation in the legislative and judicial functions, 
but not with reference to the executive. That, we believe, still Ues 
wholly with the superintendent and foreman, and it is further ac- 
cented in the procedure in bringing matters before the works council. 

The plan states that any employee desiring to bring a matter 
before the works coimcil shall present it first to the secretary, who 
shall ascertain whether it has been presented to the superintendent 
through the regular channels. If this has not been done, he shall 
see that it is done promptly. If the adjudication of the matter by 
the superintendent is not satisfactory to the employee or employee 
representatives, it then and then only comes before the works coun- 
cil. It must be presented in writing by the secretary to all members 
of the council at least three days before a regular meeting. The 
decision of the works council is final and binding. 

When it agrees upon a matter — and it can only fimction 
through an agreement because the two ballots cast in opposition to 
each other completely deadlock the decision of the works council 
— it is forwarded to the superintendent for execution. In case 
the council deadlocks, it is then in order to reopen a discussion or 
propose an alternative or compromise resolution. If the deadlock 
still continues the matter is then referred to the president of the 
company, the highest executive oflScer, who is given ten days in 
which to propose a settlement acceptable to the majority of the 
employee re^presentatives. 



THE GENERAL COUNCIL 

If he fails to do that within the following five days he may elect 
to put it into arbitration direct, and arbitration is by mutual con- 
sent before a disinterested and non-partisan arbitrator, if one can 
be chosen. If not, each side selects one. If they agree, the matter 
is settled and their decision is final and retroactive. If they 
cannot agree and choose a third arbitrator, a majority of the three 
is binding on both parties; or the president may elect to throw it 
before a general council — provisions being made that by such a 
reference, or in the event that a matter is introduced into a works 



156 INDUSTRIAL PERSONNEL RELATIONS 

council which is common to more than one other plant, the presi- 
dent may indicate the other plants which are interested, and call a 
general council of those plants, whereupon the works council origi- 
nating the proposition ceases its consideration of the matter. 

In a general council the employee representatives of each of 
the plants designated by the president send at least two of their 
representatives. For a general council they select one for each 
thousand employees or a major fraction thereof, but in no case less 
than two. The president names a number of management repre- 
sentatives, not greater than the total of employee representatives, 
who function exactly as the works council in regard to the method 
of ballot, etc. 

Provision is made that either council may be recessed *at any 
time and the employee representatives and management employees 
be privileged to withdraw and canvass their factions. In a general 
council a recess may be taken to enable the plant representatives 
to consult with the other members of their works council. Pro- 
vision is also made, in order to do this, that the superintendent shall 
convene the*, works council in whole or any part of the employee 
represeiita,tives for conference with the representatives who have 
been elected to the general council. The traveling, hotel and other 
expenses of the representatives are paid while in the performance 
of this work. 

After consultation with their plants upon a matter pending, the 
general council may be reconvened, and decision is then binding 
upon all plants affected. Pro\dsion is made that all employees 
serving on the works councils shall be paid for the time they are so 
serving. They are privileged to call l>efore them any employee of 
the plant to give testimony in a case under consideration, and the 
time of the employee so summoned is paid. 

Provision is made, however, that in case it is not acceptable to 
the employ(»e representatives to receive their money from the com- 
pany, th(*3^ are at liberty to arrange for a pro rata assessment of 
tlu» enipl()3'(H\s directly. That was Ix^cause of the objection which 
had been made by organized labor and other students of the subject 
to the gcncTal plan of paying employee represtMitatives while serv- 
ing as such. 

Copies of tlu^ plan arc available, and no less than 658 specific 
r(»qu(»sts for tluMu havi* conu* from other industrial establishments, 
from colleges and universiti(\s, from individuiils, and from various 
sources. 




ARTHUR H. YOUNG 157 



RESULTS ACCOMPLISHED BY WORKS COUNCILS 

We have been most agreeably surprised at the splendid results 
accomplished through these works councils since March 12, but 
siuprised only at the rapidity with which they were accomplished, 
because there was no question in our minds as to the ultimate result. 

There were elected by the employees 148 representatives in the 
19 different plants. The average age of those employees is 38 years 
and 10 months. They are mature. One hundred and twenty-seven 
of them are married, and only 21 are single. Their average length 
of service is 7 years and 7 months, so they are relatively old-timers 
and mostly sedate married men. One hundred and two of them are 
native-bom Americans, and 46 are foreign-bom naturaUzed citizens. 

Ninety-seven per cent of all of the employees who were present 
and eligible actually voted for or against the plan when it was pro- 
posed for adoption. The results at the various plants ranged from 
an almost imanimous vote for it to a vote of sixty to forty against 
it in two of the plants. One or two of the plants showed only a 
scant majority for the plan, considering the number of men involved, 
and we were somewhat puzzled at the time as to just what the result 
would be. For instance, at one plant the majority on the whole 
plan was only 200, and in one particular department — we will say 
the malleable foundry — the vote was 265 against the plan and 
125 for it. And we thought that those 125 should be the only ones 
who would participate in the subsequent nominations and election, 
and we might then have a condition where only 65 of them would 
elect a representative for practically 400 men, which we could not 
feel was true representation. We were very much relieved, however, 
to learn that 98 per cent of all of the employees present and eligible 
actually voted at the nominations. 

In the department just mentioned every man working on that 
day — nearly 400 of them — participated in the nominations, and 
again in the elections. On the final ballot for elections a nominat- 
ing ballot was provided first in order that the elective representa- 
tive might have at least a majority vote of his constituents. It can 
easily be seen that if we had a department of 300 men and only 
one elective ballot and there were, say, 30 candidates running 
pretty evenly, possibly a man with only 25 or 30 votes would win 
the election. The nominating ballot makes it sure that at least a 
majority will be had by the successful candidate. 

The average vote cast for the successful candidates at all plan ts 



158 INDUSTRIAL PERSONNEL RELATIONS 

was 68 per cent of the total, so that the winning candidates repre- 
sented, or were elected, by more than two-thirds of their constitu- 
ents. With a single exception, every local plant superintendent 
has written in response to an inquiry that if he had been permitted 
to select the employee representatives he would not have been able 
to choose a more satisfactory and more truly representative group 
than did the employees by their own secret ballot. 

The Harvester Company has always operated under the open- 
shop principle, but a number of union men have been elected as 
employee representatives and are serving on works councils. Our 
experience shows that these men appreciate as readily as non-union 
employees the constructive possibiUties of the plan and there is no 
indication that their participation in the cooperative activities of 
the council is not fully as satisfactory as that of the non-union 
representatives. 

One of the first results under the plan was, naturally enough, 
a demand at several of the plants for shorter hours and increased 
wages. As one old-timer said, it looked very much as if the com- 
pany was giving a sort of a Christmas party when it passed around 
those booklets saying that the works council would determine wages 
and hours. 

With a single exception these requests were withdrawn volun- 
tarily by the employee representatives upon presentation of the man- 
agement's side of the case, which was to the effect that this was 
not an opportune time for such action — that the agricultural- 
implement business was a competitive industry. The management 
was able to show that wages and rates were as high or higher than 
in similar industries in the vicinity, and that only through construc- 
tive work in this council, through a greater efficiency in the reduc- 
tion of the costs, would it be enabled to pay higher wages and still 
remain in a competitive market. If the employees were willing to 
do their part the management would do its part, exchange figures 
with them and show exactly what conditions were at any time. 
When it would be felt that the time had come to consider it again, 
this would be done. 

Thus far under the ** Harvester Industrial Council" plan the 
works councils have been able to reach mutually satisfactory con- 
clusions on all matters proposed with a single exception. That 
exception was in reference to a demand for a wages and hours re- 
vision, afToctiiig about 25 per cent of the employees of one of the 
plants. The proposition as put up by the employee representatives 



ARTHUR H. YOUNG * 159 

did not meet the approval of the local management. After an 
extended discussion that was wholly friendly and frank, the ballot 
of the works coimcil resulted in a tie. This was probably due to 
"an agreement to disagree," because both sides felt that the matter 
was one which could well be referred to the president for settlement 
and were entirely willing that this course should be followed. 

Automatically the matter came before the president who was 
able to make a compromise oflFer to the employee representatives 
which met with their entu-e approval and at a special meeting, held 
four days after the original action had been taken in the works 
coimcil, the proposal of the president had received the unanimous 
approval of the employee representatives and the matter was settled 
to the satisfaction of all concerned. In fact, it has resulted in a 
marked advance in the morale of the plant and the friendly attitude 
of the employees toward the council is doubtless more firmly estab- 
lished by the manner in which this matter was handled. 

Probably 75 per cent of the actual business of the works council 
is transacted outside of the regular meetings. Many suggestions 
or complaints have been brought to the attention of employee 
representatives and referred by them direct to the foreman or super- 
intendent, who have promptly cared for the matter to the com- 
plete satisfaction of all concerned. This has been particularly true 
with reference to correction of ventilating equipment, improvement 
of shop practice, more convenient time of the weekly payday, occa- 
sional reviews of piece-work rates and similar matters. This fimc- 
tioning of the council has been of especial value in acquainting the 
employees with the principles of time and motion studies, the rea- 
sons for adoption of certain shop rules, etc. 

The employees have universally seemed appreciative of the 
opf)ortimity to familiarize themselves with the facts as to any situa- 
tion and have been exceptionally fair in passing judgment after 
complete discussion. They have displayed particular interest in 
production problems and seem to realize that the basis of larger 
returns to them for their labor lies in an increased production and 
more eflBicient operation of the shops. 

Some of our plants are in the middle west side of Chicago, 
right in the heart of the radical district. Another in particular is 
out on the far south side, where probably you have read of race 
riots in Gary and Indiana Harbor and that vicinity. 

I believe firmly that there is a well-organized propaganda on 
foot to start Soviet organizations in this country, and to play up 



160 INDUSTRIAL PERSONNEL RELATIONS 

the rebellious spirit in the foreign element in the vicinity of some 
of our shops. 

The foreign members of one of our works councils desiring to 
meet this movement, asked for information which enabled them 
to get up a report showing the participation of the foreign-lx>rn 
employees in military' service of this country. They drew parallel 
columns of the number of foreign-born employees per thousand of 
the men at the plant; and the number of each nationality per thou- 
sand who engaged in the military program. The average for the 
whole plant, of all the employees engaged who went into military' 
service was 22 per cent. It was much higher at the other plants. 
The average of all the foreigners was 22.2 per cent, practically the 
same, but particularly the average of the Polish people against 
whom sentiment had been directed, and against whom a definite 
campaign was being waged in playing up their racial spirit, was 
30 per cent. The average participation of the Poles was greater 
than that of the Americans. Then they drew some conclusions: 
that the foreigner, while he was not a citizen, and therefore not 
hound b}' the same ties as the American, had contributed almost 
the Siime to the military' program, and individual nationalities to a 
greater extent than had the Americans; that his contribution had 
really been much greater from the fact that he was not a citizen, 
not bound by those ties; and they drew the conclusion that the 
anti-foreign sentiment which was being engendered was all wrong 
and not justified by facts. 

Now, that was done in the works council committee on publicity 
without any instigation by the management whatsoever. It was 
the thought of the m(»n themselves as a metho<l of competing with 
this insidious and destructive propaganda. 

RKSPONSIBILITIES VNDERTAKEX BY WORKS COUN'CILS 

^^^* had a restaunint which was not very satisfactory. It was 
losing nion(\v for us and not getting the n^sults which we wanted. 
We turned it over to the works council and the patronage has trebled 
since that time*. It no longer shows a deficit. It was frankly and 
friM'ly turntMl over to the men to manage. They hire their own 
manager, they mike th »ir own rules, they s?t th *ir own prices, and 
they have leanictj something about the n»staurant business that we 
wantt'tl them to know, and sonu^thing of its trials and tribulations, 
and we have "snld" the proposition entiivly. 

There have biM.*n one or two n»quests for ivinstatement of dis- 



AKTHTJB H. YOUNG 161 

charged employees. There are functionalized employment bureaus 
and all recommendations for men on the part of foremen are reviewed 
at the employment oflSce. There are no restrictions in oifr plan as 
to what may come before the works coimcil. In two cases — the 
only cases thus far that have been presented — the works council 
by unanimous vote upheld the decision of the employment manager 
and refused to reinstate the employee who thought he had a just 
grievance. 

The safety program has been rejuvenated. We thought we 
were getting up efficient bulletins and we thought we were carrying 
our message all the way down the line. We thought that our safety 
program was satisfactory, and we couldn't go much farther with it. 
Since the inauguration of the works council, however, our accident- 
frequency rate has steadily fallen month by month. The character 
of the bulletins has changed somewhat. They are written in shop 
parlance, and by the men on the safety committees. 

The sub-committee organization activities of the works council 
have been wonderfully well arranged by the employees themselves 
without any action of the management. They have unanimously 
said that they do not want a grievance conmiittee. They do not 
want any committee on wages and hours. Those are subjects that 
they want to come out for a full discussion in the works coimcil, 
and their reasoning was somewhat along these lines: They said, 
"If we had a grievance committee which only went to the boss on 
controversial matters, he couldn't help but get down on us a little 
bit, and believe that we never think of anything except how to make 
trouble. And we don't want to have that put up to just one or two 
men, we want the whole council to work on that; and thien, too, we 
have this feeling that if a man comes to us with a grievance or a 
request for piece-work revision, or revision of prices or hours, we 
don't ourselves want to say to him, *Well, you go over in the black- 
smith department and see John Doe, because he is on that committee. 
We don't want to have anything to do with that.' Those are things 
they want us to get into, and if we do, we want to do it to our com- 
plete satisfaction." 

The sub-committee activities, so far as grievances and hours 
and wages have been concerned, are always taken up by the council 
as a whole, and it has meant that no one or two individuals sat 
themselves up as business agents in the plant. 

The question has been asked how much of this himian relations 
problems Ues within the field of the mechanical engineer. Engi- 



162 INDUSTRIAL PERSONNEL RELATIONS 

neering, to my mind, has been the consideration of an exact science, 
and these matters of personnel in the industrial sense are not re- 
ducible to stable factors at all. In the whole scale of industrial 
relations we are always dcahng with the human being — a body 
with a mind, a creative force, and a soul, and therefore not reducible 
to any sort of stabiUty as regards its valuation. You can't get up 
any sort of a formula which will solve a given condition, and repeat 
on it. And I don't believe the personnel problem ever should be 
treated as a science and coldly analyzed or formulated. 

I beUeve that its direction and its constant progress should be 
made by men, not necessarily technical in their education, although 
we do need the assistance of every bit of science that can be brought 
to bear, but more particularly men who are given over to the idea 
of real service. 

There are many employment managers functioning today in 
splendidly furnished, mahogany-Uned offices, and who receive the 
requisitions from foremen for help and who analyze the jobs' speci- 
fications quite technically, and select by means of phrenology, and 
goodness knows how many other "ologies," applicants from the 
waiting line, if there is such a thing, or by scientific advertising, if 
the waiting line is not there, and possibly fill those jobs. 

But I believe the work is best done by men who forsake the 
oflSce and don't hire through an employment window but get out 
and interview a man across a desk, and look upon him as another 
human being, one who is coining into a strange land, into a strange 
company, and make him thoroughly at ease; who will sell him his 
job as a constituent part of a great assembly, who will look after 
him, not in an apparently disinterested way, but in an atmosphere 
of good-fellowship after he goes to work, and who will look upon 
his problems with a light of experience and know exactly, or nearly 
exactly, what his thoughts are, what his suspicions are — and those 
I do not thuik can be had by a science as well as by experience. 

I think the employment engineer of today — the personnel 
manager of today — should first of all be chosen on personality. 
That is certainly more than fifty per cent of the necessary make-up. 
He should be of an engaging i)crsonality, a man who can converse 
with other men and get them in turn to converse with him; who 
has a synipat luetic oar; who has a certain poise; and withal, who is 
a good, keen judge of human nature; and then on top of that, as 
nmch technical training as he may have. 



No. 1683 

THE STATUS OF INDUSTRIAL RELATIONS' 

By L. p. Alfosd, New York, N. Y. 
Member of the Society 

At the Anmud Meeting of the Society of 1912, a re-part vxu presented by the 
SuXhCcmmittee on AdnwnietraHon, of which the late James M. Dodge was Chair- 
man, and L. P. Alford, Secretary, on The Present Stale of the Art of Industrial 
Management. This report was replete with information upon the broad aspect 
of the management problem as it then existed in the industries of the country. 

During the seven years which have intervened since the preparation of this 
report, the question has been studied from many different angles and has come to 
be viewed in quite a different light from that in which it was regarded when the original 
report was prepared. In consequence, the Committee on Meetings and Program 
appointed Mr. Alford a committee of one to prepare the following new report upon 
the subject for presentation at the Session on Industrial Relations al the Detroit 
meeting. 

This report not only comprises a review of the new aspects of the problem which 
have recently developed, but also a historical summary of the progressive stages in 
the development of industrial relations since the period immediately following the 
Civil War. It has proved to be inevitable that after any great economic disturbance 
like that produced by the CivU War, or the present period of unrest following the 
world conflict in Europe, there should be unrest and uncertainty in the field of labor 
and employment; and it thus seems appropriate at this time to outline briefly the 
most important transitions which have occurred in this field, beginning with the 
time of the Civil War and including the situation which now exists, so similar in 
character, but greatly amplified. 

TN the presidential address made to this Society in 1882 is the 
following remarkable statement of the responsibility of engi- 
neers in solving the problems involved in the relations of employers 
and employees: 

In singular and discreditable contrast with all the gains in recent and current 
practice in engineering, stands one feature of our work which has more importance 
to us and to the world, and which has a more direct and controlling influence 
upon the material prosperity and the happiness of the nation than any modem 
invention or than any discovery of science. I refer to the relations of the em- 
ployers to the working classes and to the mutual interest of labor and capital. 

^ Report prepared at request of Committee on Meetings and Program. 



Plresented at the Spring Meeting, Detroit, Mich., June 1919, of The 
Amxbican Socibtt of Mechanical Engineebs. 

163 




164 THE STATUS OF INDUSTRIAL RELATIONS 

It is from us, if from any body of men, that the world should expect a complete 
and satisfactory practical solution of the so-called "Liabor problem." More is 
expected of us than even of our legislators. And how little has been accomplished. 

2 Dr. Thurston was speaking of conditions as they were a full 
generation ago, yet his words are as true today as they were then. 
The world is looking to the engineers for leadership in these matters 
"and how little has been accomplished." 

3 The topic of industrial relations is so complex and far-reaching 
that any treatment within the space of a professional paper must 
of necessity be restricted to certain aspects of the problems. So 
in attempting to outline the present position of the body of fact that 
is comprehended in the term "industrial relations," the scope will 
be limited to a survey of the more important developments of the 
last thirty or thirty-five years, and to a statement of a few of the 
more outstanding tendencies revealed by events of the immediate 
present. 

4 To shorten the paper, certain of the supporting facts have 
been largely grouped in two appendices or are referred to in foot 
notes. Free use has been made of the writings and work of 
others and an earnest attempt has been made to give due credit, 
but in the examination of many sources of information it may be 
unwittingly that violations of courtesy have crept in. If such are 
discovered, the author hopes that the situation will be viewed 
leniently in the face of the difficulty of compressing into a limited 
number of pages a discussion of a subject of such complexity, and 
one on which such a tremendous mass of literature has been produced. 
Many have supplied information of value in personal letters and 
such assistance is gratefully acknowledged. 

5 Before we can attempt to outUne the status of industrial rela- 
tions it is necessary to define what we mean by the term. To that 
end this statement is offered: 

Industrial relations comprises thai body of principle, practice 
and law growing out of the interacting human rights, needs and 
aspirations of all who are engaged in or dependent upon producHve 
industry, 

6 It will be observed that this definition does not include di- 
rectly the feelings of uneasiness and unrest among industrial workers, 
but \Tiews them as the expression of real, or fancied, rights or needs, 
and considers strikes and lockouts as but the assertion of the same 
or similar claims. On the other hand, the definition does include not 



L. P. ALFOBB 165 

only the interests of employers and employees who are actually en- 
gaged m industry, but Ukewise of others who are dependent upon 
productive industry for the satisfaction of some of their needs, or 
for the safeguarding of some of their rights. 

7 From the viewpoint that industrial relations comprises a 
body of principle, practice and law operating in productive indus- 
try it is a major subject for examination by this Society, for the 
members of The American Society of Mechanical Engineers are 
drawn largely from the great group of responsible executives in in- 
dustry. It is clearly a function of this Society to enUghten its mem- 
bership in regard to matters that so vitally concern manufacturing 
and production as the one under consideration. 

8 Furthermore, now is a particularly appropriate time for 
such a study as we are in a period that promises to yield develop- 
ments of the highest importance. In this respect the present is 
similar to the decade 1880 to 1889, during which time several of the 
important methods and practices that have been worked out in 
industrial relations had their beginnings. 

9 To extend this comparison of times, the Civil War period was 
one of prosperity in America and Europe and of growth of labor 
organization. The years from 1865 to 1869, the half decade imme- 
diately following the close of the Civil War, are referred to in our 
history as the period of reconstruction. After 1870 there set in a 
long period of business depression which culminated in the financial 
panic of 1873 and continued until about 1879. During this time 
there was a decline in trade unionism. Beginning with about 1880 
there was a recovery which brought business prosperity, high wages, 
a decrease in unemployment and an enlargement of trade unionism. 
Dming the 80's there were many hard-fought strikes, widespread 
labor unrest, and frequent and insistent demands upon the part of 
workers in industry. 

10 At the present time we are witnessing the payment of higher 
wages than were ever before known in this country, there is a general 
feeling of imeasiness and unrest throughout our entire industry, 
labor is making many demands, and strikes are so frequent and 
widespread that it is doubtful if a single member of this Society has 
not at least been inconvenienced during the present year by the 
temporary cessation of some function of industry upon which he is 
in some degree dependent. 

11 But this comparison carries further. During the 80's there 
were developed and put into use two practices expected to mitigate 



166 THE STATUS OF INDUSTRIAL RELATIONS 

or help solve the labor difficulties of that time, namely, profit sharing 
and methods of wage payment intended to give the worker a direct 
share in the benefits of increased production. At the present time 
we are seeing another heightening of interest in profit sharing and 
the widespread installation of the so-called shop-committee system. 
Thirty years ago and today are similar periods of experimentation 
in industrial relations. 

12 So we cannot avoid the conclusion that times of labor unrest 
are productive of new plans and methods that attempt to satisfy 
the needs and harmonize the rights of those who are engaged in 
industry. And it is reasonable to expect that the period we are now 
in will mark the rise of new methods and practices that will properly 
belong in the classification of industrial relations. 

13 Although the beginnings of the development of industrial 
relations took place during the 80's of the last century the situation 
in industry did not become acute in the United States imtil about 
1905. That date fixes approximately the time when the evils of 
absentee directorate of large corporations came into prominence. 
Many of our great industrial consolidations had taken place before 
1905 and the owners had put into effect a system of control and 
management arbitrarily determined in a few of our large cities, prin- 
cipally New York City, while the plants of these corporations were 
spread throughout the country. 

14 It is frequently stated that the necessity for establishing 
industrial relations today is the growth of the factory system wherein 
all personal contact is lost between owner or manager and the worker. 
While this is true, it does not sum up the entire loss. In the system 
of absentee directorate there are other evils as well, and these taken 
together have set up situations where there have been clashes over 
the rights, needs and aspirations of those who belong to the class of 
employers and those who form the great group of employees. These 
losses in fitness for control may be stated in this wise: 

a The loss of personal contact and relationship that formerly 

existed between the master and his skilled workmen and 

apprentices 
b The loss due to the lack of personal knowledge of the 

work being done on the part of present-clay directors and 

mana<2:crs 
c The loss due to the lack of personal knowledge of the 

tools and equipment used in production on the part of 

present-day managers 



L. P. ALFOBD 167 

d The loss of the direct oversight of saving and conserving 

materials and human eflfort on the part of present-day 

managers 
e The withdrawal from productive work of the families of 

the directors and managers 
/ The loss of equality of li\dng conditions between the 

families of the directors and managers and the workers. 

15 The eflfect of these losses in creating a situation where there 
may be a clash of interests, and failure on each side to understand 
and appreciate the other, is brought home when we contrast the 
human relationships in the days of craftsmanship with those of the 
factory system. In former times the employer or master knew how 
to do all parts of the work himself, had in fact done so with his 
own hands, was in personal contact with all of the tools, equipment 
and materials used in his shop, had complete personal oversight of 
everything that was done, and held a relationship of almost father 
to son with his apprentices who, in many cases, lodged and boarded 
in the master's home and were ser\'ed by the master's wife and 
daughters. Not only did the master instruct his apprentices in the 
requirements and skill of the trade, but he likewise set them the 
example of right Uving. All of the activities of the master opened 
to the apprentice as he became able to exercise them. There was 
a complete community of interest between the master, his family, 
his workmen, their famiUes and his apprentices. 

16 By contrast, in American industry today far too often the 
owners and directors live hundreds and thousands of miles away 
from the workers in their plants, under entirely different conditions, 
as var}dng as New York City and a Massachusetts mill town or a 
Middle-West factory community. All exact knowledge one of the 
other is lacking, there is no community of interest or purf)ose, and 
no assurance that a f)olicy determined uf)on in the directors' room 
will meet the needs and rights of the workers in some far distant 
locaUty. 

17 It is the creation of this situation of absentee directorate 
that has done much to focus attention upon the necessity of develop- 
ing a body of principles, practice and law to satisfy the needs and 
safeguard the rights of all who are engaged in or dependent upon 
American industry'. 

18 Examination reveals six major lines of development amid 
the various methods, plans and systems that have been tried in 
seeking to work out better industrial relations. These are: 



168 THE STATUS OF INDUSTRIAL RELATIONS 

a Profit-sharing plans 

b Methods of wage payment 

c Methods and laws to reduce the hazards in industry and 
mitigate the effects of injuries and occupational diseases 

d Employment management 

e Declaration and enforcement during the period of war of 
three rights of workers, namely, collective bargaining, 
restricted hours of labor and the living wage. Declaration 
of these same rights and others in the Treaty of Peace. 

/ Systems for mutual or joint control by employers and 
employees. 

19 A controlling reason for considering these six lines of develop- 
ment is their actual or promised permanence and widespread accept- 
ance and application. For this same reason it does not seem perti- 
nent to the purpose of this paper to devote space to welfare methods, 
industrial betterment, suggestion boxes or systems, shop gardens, 
factory bands, dances, ministrels, glee and athletic clubs, employees' 
loans, benefit associations, pensions and many other activities that 
properly classify under our definition of industrial relations. With- 
out doubt successful appUcations of every one of these activities 
can be mentioned, and likewise, without doubt, experiences can be 
pointed out where they have had a beneficial effect in promoting 
satisfactory conditions and assisting to develop a "spirit of the 
organization." It is more than likely that many of these activities 
will always find a place in industry, but none of them seem to be 
a major line of development, and in fact all classify imder welfare 
work or industrial betterment, which have fallen into disrepute be- 
cause of the motive of charity or paternalism that has inspired them 
in many places. 

20 The element of failure in all these agencies is the lack of 
removal of the fear of unemployment. For the fundamental cause 
of industrial unrest is the dread of losing the opportunity to work 
and thereby secure the necessities of life, or of I)eing cut off from 
deserved promotion. 

PROFIT SHARING 

21 Profit sharing has Ixjcn mentioned as one of the lines of 
development that become i)rominent during the period ending in 
1889. Approxiiiiiitc^ly 32 firms that put into effect some form of 
profit-sharing plan during this i)eriod.^ A few of these were mer- 
cantile establishments, but all were large employers of labor. 

» See Profit Sharing Between Employer and Emplo>-ec, N. D. Gilman, 1889. 



L. P. ALFOBD 169 

22 Not only is there reason to believe that these attempts were 
inspired by conditions of industrial unrest, but there is direct testi- 
mony to this efifect in two cases. The Globe Tobacco Company, 
of Detroit, Mich., entered into an agreement in regard to its profit- 
sharing plan in 1886 with the district board of the Knights of Labor; 
and the plan of the H. O. Nelson Company, of St. Louis, Mo., which 
was annoimced on March 20 of that year, was an outgrowth of the 
great raikoad strike, strikes m the building trades and the move- 
ment for the eight-hoiu" working day. 

23 Within the past' six months a number of large industrial 
plants in this coimtry have announced or put into efifect profit-shar- 
ing plans, so we seem to be having a repetition of what took place 
during the 80's of the last century. 

24 It is worth while to point out that although profit sharing 
was put into efifect in a number of American industries a generation 
ago, it has never had widespread adoption. This fact may cause us 
to question its efifectiveness as a promoter of good industrial rela- 
tions. Several reasons are recognized for this situation: Payment of 
gains under profit-sharing plans are deferred and so lack an imme- 
diate appeal; the gains do not come dbectly nor entirely from the 
efiforts of the workers, but are dependent upon the hazards of the 
enterprise; the amoimt distributed to any individual worker is 
equivalent to an increase of only a few cents an hour for the year; 
the amount shared by each person among the owners and managers 
is many times greater than that received by each workman. 

25 Plans for stock participation have the same purpose and 
value as those for profit sharing. 

METHODS OF WAGE PAYMENT 

26 The pioneer work that has since yielded the science and 
practice of industrial management was performed during this same 
period of industrial unrest, that is, from 1880 to 1889. During this 
decade Mr. Henry R. Towne presented two papers before this 
Society that have been frequently referred to as the beginnings of 
the literature of industrial management. The first was read in 
1886. In it Mr. Towne called the attention of his fellow-engineers 
to the need of a study of the financial and profit-making aspects of 
shop management. He urged his associates to become "economists" 
because the engineer is one who essentially efifects economy. His 
second paper was presented in 1889 and in it was described a gain- 
sharing plan that Mr. Towne had appUed in his own shop. The 



170 THE STATUS OP INDUSTRIAL RELATIONS 

object was to enable his employees to share the profits of the 
business, depending upon gains in efficiency as shown by careful 
accounting. 

27 Influenced by his study of the epidemics of strikes during the 
same decade, Mr. F. A. Halsey originated his premium plan of wage 
payment that was disclosed to this Society in a professional paper 
read at the Providence meeting in 1891. In a recent letter written 
by Mr. Halsey to the author the activities of the Knights of Labor 
and the great street-car strike in New York City are mentioned as 
two of the events that influenced him to study the possibility of 
some form of wage payment that would reward the worker for 
increased effort and production. 

28 This same decade was the period of the work of Dr. Fred- 
erick W. Taylor, which later on gave to this Society several papers 
outlining his system of shop management. It will be recalled that 
a part of his methods was a system of wage payment known as the 
* ' differential system. ' ' 

29 Since that time several other methods of wage payment 
have been originated and developed, so that today we recognize 
some half a dozen that have more or less extensive application. 
Each one has been developed with the purpose of improving the 
relationships between employer and employee in regard to the 
division of the earnings and profits of industry, or to provide an 
extra reward for extra productive effort. 

THE "safety-first" MOVEMENT AND WORKMEN'S COMPENSATION 

30 About 1910 American juries began to award large sums 
in suits for personal damages, where the plaintiff had been injured 
by machinery or otherwise in industry. Employers turned to lia- 
bility-insurance companies to defend these suits and settle with 
their employees, thus bringing an outside party between them and 
their workers. Appreciation of the hazards in industry and the 
hardships endured by incapacitated workmen and their families, 
not only during the period of recovery from the injury itself but 
|)crhaps throughout the life of the wage earner due to decreased 
earning power, gave rise to the "safety-first" movement and the 
enactment of workmen's compensation laws. 

31 It is interesting to note that the safety-first movement was 
fostered largely by men residing west of Pittsburgh, many of whom 
were in the employ of large corporations where directorates were in 



L. P. ALFORD 171 

New York City. Furthermore it was from the outset a young man's 
movement. Here and there a few progressive firms paid generous 
compensation voluntarily before laws were passed making this 
practice compulsory. Mr. John L. Henning instituted such a plan 
in 1904 in a mining operation in Louisiana of which he was chief 
engineer. This is the earliest attempt of this kind in the United 
States that has reached the author's attention. 

32 In the enactment of compensation laws the rights and needs 
of the community were recognized. An entirely different view of the 
legal relationship of employer and employee was taken than that 
which had existed before, and which was smnmed up in the expres- 
sion "master and servant." The responsibility for the injuries was 
placed squarely upon the industry through the employer, and in 
boards and conmiissions, machinery was set up to make sure that 
employees would receive the compensation to which they were 
entitled. There are now 39 states which have compensation laws. 
The first was New Jersey and the effective date was July 4, 1911. 

33 But the safety-first movement has yielded much more than 
this group of legal enactments. There has grown up a large body 
of practice in regard to safeguarding machinery and working spaces, 
establishing and operating first-aid rooms and factory hospitals, 
toward improving the heating, ventilation and sanitary conditions 
in factories, removing or mitigating conditions tending toward occu- 
pational disease, providing medical, dental and nursing service for 
the families of employees, establishing medical examinations before 
employment and at stated intervals thereafter, teaching personal 
hygiene to employees and their families, and, in fact, a complete 
new outlook in regard to the health and physical welfare of all who are 
engaged in industry. Through it all there has been a moral motive. 

34 The movement has also brought into being two strong 
organizations, the National Safety Council and the Workmen's 
Compensation Bureau. In addition there is an association of phy- 
sicians who are engaged in industrial practice. 

35 The American Society of Mechanical Engineers has contrib- 
uted to this development through taking the initiative in preparing 
several safety codes or standards. It is estimated that the safety- 
first movement of the past ten years has reduced the number of 
annual fatal accidents in the United States from 35,000 to 22,000 
with a corresponding lessening of maiming and disabling accidents. 
This line of development is the most significant of all from the view- 
point of the interests of the conununity. 



172 THE STATUS OP INDUSTRIAL RELATIONS 

EMPLOYMENT MANAGEMENT 

36 Since 1916 there has come into prominence what is now 
recognized as a . new profession in industry, that of employment 
management. It comprehends in its broadest interpretation the 
establishment of all policies and direction of all of the functions hav- 
ing to do with personnel. It has taken over, expanded and developed 
the former work of hiring and discharging, has sought to reduce 
labor turnover and has fostered and directed those activities which 
are usually comprehended under the term welfare. 

37 The oldest organized employment department of which 
the author has knowledge has been in operation about nineteen 
years in the plant of the B. F. Goodrich Company, Akron, Ohio. 
In 1907 Mr. H. F. J. Porter in a discussion of a paper presented 
before this Society called attention to the evils of labor turnover 
and outlined some methods that he had taken for its reduction. 
In 1914 Mr. Magnus W. Alexander presented a striking address 
before a convention of the National Machine Tool Builders' Asso- 
ciation in which he gave statistics gathered from some twelve metal- 
working plants revealing the tremendous amount of the turnover 
of labor and its excessive cost. A small group of men in and around 
Boston, Mass., worked on this problem of employment for a number 
of years and as early as 1910 organized the Boston Employment 
Managers' Association. 

38 But the movement did not gain headway until about 1916 
when it was ready for the truly marvelous expansion that has taken 
place during the period of the war. The April 1919 issue of 
Personnel, the official Bulletin of the National Association of 
Employment Managers lists 27 employment managers' associations. 
In addition, a National Employment Managers Association was 
formed in May 1918. No other line of development in industrial 
relations has had the rapidity of growth of employment manage- 
ment. But the impelling motive has not been entirely that of 
fostering good industrial relations, although that result has como in 
many cjises where the work has been well done. The major reason 
in the minds of most industrial executives in establishing employ- 
ment departments has been to secure employees during the period 
of labor scarcity and to find out why men leave. Another impetus 
to the movement came through the action of the United States 
Government in insisting that such departments should be installed 
in plants manufacturing munitions, war supplies and ships. To 



L. P. ALFOBD 173 

meet the demand for trained managers a nimiber of colleges and 
miiversities established six-week courses in employment management 
imder the direction of the War Department. Since the signing of 
the armistice several of these have been modified and put on what 
will probably prove to be a permanent basis. 

DECLARED RIGHTS IN INDUSTRIAL RELATIONS 

39 As a war measure President Wilson by proclamation created 
a National War Labor Board to establish principles and poUcies in 
regard to the employment and utilization of labor during the period 
of war, more particularly in war industries, and to set up machinery 
for considering and adjusting grievances. At the outset this Board 
declared three rights of labor: The right to organize and bargain 
collectively; the right to a limited number of hours of labor; the 
right to a living wage. 

40 The Peace Treaty written at Versailles recognizes these 
three rights and several other principles that are of "special and 
urgent importance." Appendix No. 1 gives the text. 

41 The recognition of these rights is a great step in the develop- 
mentr of industrial relations, and they can never be abrogated in 
American industry. The same situation is reflected in British con- 
ditions as shown by the summary of the report of the Employer's 
Industrial Commission of the United States Department of Labor. 

MUTUAL OR JOIXT CONTROL 

42 The sixth major line of development has to do with control 
in industry. Students of the present condition of unrest have pointed 
out that the fundamental is a "struggle for control," the opposing 
forces being the owners and the workers. The expression "industrial 
democracy" is frequently used as describing a state that is about to 
come, or is now being ushered in. The parallel between our poUtical 
democracy and the expected industrial form is sometimes put in this 
wise: The slogan that gave birth to this nation and brought our 
poUtical democracy was, "No taxation without representation." 
The parallel slogan expressive of the movement to bring industrial 
democracy is "No control without representation." 

43 The method that is being followed to put this ideal into 
practice is the shop committee. It is a new development. Approxi- 
mately 106 firms in this country have some form of representative 



174 THE STATUS OP INDUSTRIAL RELATIONS 

shop committee. The oldest was put into effect in 1903 and most 
of them were started in 1918. Bridgeport alone has 44 plants 
where this system is in force. The plan installed by Mr. H. F. J. 
Porter in 1903 in the plant of the Nemst Lamp Company in Pitts- 
burgh was described by him in an article published in The Engine- 
ering Magazine in August 1905. The shop committee was made up 
of secretly-elected employees under the chairmanship of the shop 
superintendent. Conditions in the plant that might be bettered for 
the employees were discussed, and the committee evidently func- 
tioned as a safety-first committee as well as a representative shop 
committee. 

44 Without doubt the success of the safety committees with 
which there has been some six or seven years of experience has 
paved the way in many plants for the representative shop committee. 
In fact, a prediction of this development is found in a paper presented 
before this Society in 1915 written by Mr. W. H. Cameron.^ 

45 A general classification of these shop committee plans 
yields three types: 

46 The first sets up an organization roughly paralleling the 
Cabinet, Senate and House of Representatives of the United States 
Government. The Cabinet may consist of some or all of the direc- 
tors or higher executives of the plant. The Senate may consist of 
all or a portion of the foremen, while the House of Representatives 
is a body secretly elected by the employees. 

47 A second type divides the workers in the plants into divisions, 
each having a definitely determined and equal number of employees, 
not necessarily defined according to craft or occupation. Any one 
division may include employees from several trades and doing 
varying kinds of work. E^ach division secretly elects its own repre- 
sentative and these representatives coming together form the shop 
committee. This plan is simple, but is not adapted to industries 
where there is a high degree of organization or where the workers 
naturally divide into a series of recognized trades. 

48 The third plan is the one adopted by the National War 
Labor Board and installed under its direction in a number of plants 
making war materials. Its essential features are outlined in the 
following official statement of procedure for election: 

Shop committees shall be selected to meet with an equal or lesser number of 
representatives to be selected by the employer. Each department or section of 

* The Attitude of the Employer Towards Accident Prevention and Woric- 
men's Compensation, W. H. Cameron, Trans. Am. 9oc. M. E., voL 37, p. 90S. 



L. P. ALFOBD 175 

ibe diop shaQ be entitled to one oommitteeman for each one hundred employees 
employed in the department or section. If in any department or section there 
riiafl be eoq^lojrees in excess of any even hundred, then an additional oommittee- 
man may be dected, providing the additional employees beyond the even hundred 
ahaU be fifty or more; if less than fifty no additional representation shall be 
allowed. As an example: In a department or section emplojring 330 men, three 
eommitteemen will be dected; in a department employing 375 men, four oom- 
mitleemea will be dected. 

49 In plants where shop committees have been in operation 
for some time a wide variety of topics has come up for discussion 
and determination. One grouping lists some fort^^-two different 
kinds of matters, of which only one was wages. The plans are too 
new, however, to estimate the extent of the effect they may have 
in developing good industrial relations. But a sufficient amount 
of experience has already been accumulated to indicate that the 
appUcation of the ideas of joint consideration and control on the 
part of employers and employees will produce favorable results. 
To reach a workable basis the employer must voluntarily limit his 
own authority and agree to conduct his business by the rule of rea- 
son and even-handed justice as interpreted by the representative 
shop committee that he may set up. The plan seems to restore, 
so far as possible in a large-scale business, the simple and effective 
relationship that used to exist between the master and skilled work- 
man, and which largely exists even today between the small employer 
and his half-dozen employees. 

50 A rather more formal mechanism for establishing representa- 
tion in industry is the protocol sj-^stem worked out in the coat and 
suit trade in New York City, and which has been appUed in a few 
other instances in similar trades. 

51 From one \'iewpoint the rapidly developing movement to 
put shop committees into effect in the United States is a confession 
on the part of employers and those who have the responsibihty for 
industrial enterprises that they have already lost some of their 
control. Under such condition there is a readiness to experiment. 
From another \4ewpoint it is an earnest attempt to find a basis of 
democratic cooperation in the control of industry. 

52 A parallel movement has taken place in Great Britain lead- 
ing to the proposal to establish Industrial Councils, through which 
the British Government may promote effective cooperation between 
the organized employers and workers, believing that representative 
government in industry will foster good relations. 



176 THE STATUS OF INDUSTRIAL RELATIONS 

DEVELOPMENT OF MOTIVES 

53 After having sketched these six Unes of development in 
improving industrial relations and in building up a great body of 
principle, practice and law, it is wise to examine some of the pre- 
dominating motives. 

54 One that came into play eariy and brought to the front 
those practices and activities summed up under the headings wel- 
fare work and industrial betterment was the motive of altruism. 
The successful manufacturer, taking the part of the autocratic 
benefactor, enjoyed the swelling of the heart and feeling of personal 
gratification that came to him when he arranged for the spending 
of money to provide conveniences and benefits for his employees, 
as bathrooms, restaurants, flower gardens, reading rooms, rest 
rooms, and the like. But workers are quick to resent favors if they 
are substituted for justice, and the welfare movement as such has 
very properly been discredited and has practicaUy disappeared, 
although many of its activities have been retained but inspired by 
a different and proper motive. 

55 The safety-first movement from its inception was influenced 
by engineers who saw the essential economy in preserving the life 
and limb of the workers. The employers translated this engineer- 
ing economic viewpoint into the commercial motive "it pays." 
Selfish though this was and is, nevertheless it was neither charitable 
nor paternalistic, and because the movement had a firm basis in 
soimd industrial economics and in morals it has met with deserved 
and widespread success. 

56 But even this engineering-commercial motive is inadequate 
to provide the impetus for the developing of the industrial relations 
that we all hope for. It does not comprehend the interests of all 
who are dependent upon and must be served by productive industry; 
so many of the leaders of thought on industrial matters have de- 
clared and emphasized another motive, saying that it must pre- 
dominate and prevail else we will never have the development that 
we need. This is the one of service, Mr. H. L. Gantt, in an article 
in the May, 1919, issue of Industrial Management, presents the motive 
of service as applied in bis own consulting work. Once it becomes 
active we can hope for the working out of a new body of principle 
and practice in regard to industrial relations that will bring far happier 
conditions than any we have yet experienced. 




L. P. ALFOBD 177 



INTERESTS IN INDUSTRY 



57 'V^th this development of motive has been a broadening 
of the recognition of those who are interested in the proper carrying 
on of industry itself. In the early days the expression "master and 
man" or the legal "master and servant" siunmed up these interests 
and relationships, but as industry developed and the clash of rights, 
needs and aspirations became apparent the famiUar term became 
"capital and labor," used more particularly to denominate antago- 
nistic forces. In fact, this term is used very generally at the present 
time, although it has been pointed out that there can be no con- 
flict between capital as such and labor, for we cannot conceive of a 
man fighting with accumulated wealth represented in the physical 
means of production. So, to clarify thinking some authors have 
insisted that we should use the term "capitalists and laborers" in- 
stead of capital and labor. Point is given to this contention by a 
quotation from a speech delivered by President Lincoln before the 
Wisconsin Agricultural Society in 1859: "Labor is prior to, and in- 
dependent of, capital — in fact, capital is the fruit of labor, and could 
never have existed if labor had not first existed. Labor can exist 
without capital, but capital could never have existed without labor." 

58 The next development of interests brought consideration of 
the needs and rights of the public, which depends upon the proper 
carrying on of industry for much of its well-being. This thought 
brought a new grouping — "capitalists, laborers and the public." 
Quite recently in studies of industrial relations some writers have 
included five parties, namely: The capitalists who supply the 
materials and means for production; the laborers who supply pro- 
ductive capacity; the managers who provide direction and control; 
the community in which the industry is located and upon whose 
operation its welfare to a certain extent depends; and the public 
that purchases the articles and goods produced. 

59 There is more than ample reason for including the interests 
of the community and public in any such classification. The laws 
regulating the hours of labor for women and children, for the elimi- 
nation of hazards in industry, and for compensation for industrial 
accidents are but one expression of the rights of the community 
as well as the worker in the operation of industry. 

60 Before justice can be framed in the form of law there must 
have been developed a body of general principles. It is evident 
that the principles underlying industrial relations are now in a 
process of rapid formulation. It is probable that before long our 



178 THE STATUS OP INDUSTRIAL RELATIONS 

courts will have to pass upon an increasing number of industrial- 
relations controversies. Such matters are justiciable today; they 
were not twenty years ago. We may look forward to a time when 
controversies in regard to such rights will become just as justiciable 
as any controversy in regard to property. It is possible that when 
this time comes it will be properly referred to as the era of industrial 
democracy. 

61 Three tendencies in this development of industrial relations 
seem to be new though they are not novel. 

62 The first is the acceptance of the motive of service, which on 
moral grounds declares for recognition of the rights, needs and aspi- 
rations of every one engaged in or dependent upon industry. It is 
the engineering viewpoint rendered unselfish. 

63 The second is the wilUngness to consider workers in groups. 
By training and experience the engineer has only been willing to 
look upon workers as individuals. Two of the governing principles 
that have brought modem industry to its present heights are the 
division of labor which minutely subdivides the job, and the selec- 
tion and adaptation of the worker which individualizes him and 
attempts to fit him to some particular task, tool or machine. This 
is the viewpoint of specialization which deals only with units. 

64 But in industrial relations the workers must be considered 
in groups or in the mass. This is the viewpoint of the industrial 
psychologist as contrasted with that of the technical engineer, and 
the latter has been slow to understand that his methods of sub- 
division and specialization cannot be used successfully in dealing with 
the problems of industrial relations. Much has been said about 
fostering cooperation, but little progress has been made. A reason 
for this situation is found in the lack of understanding on the part 
of industrial executives that to build morale or the spirU of the organi- 
zation their working people must be appealed to in the mass and 
not as individuals. 

65 The development of the safety committees began to open 
the eyes of industrial executives as to what might be accomplished 
once employees as a body had a chance to express their desires 
and opinions. The present movement to establish shop committees 
will carry- this experience further and into new aspects of the prob- 
lems of industrial relations. The experience so gained will show 
tlie possibilities and advantages of discarding the individualistic 
viewpoint of the engineer. It will also bring the passing of arbi- 
trary and autocratic decisions. 



L. P. ALFOBD 179 

66 The third tendency is toward mutual or joint control, toward 
mutuality and the working out of representation. It is an expression 
of democratic ideals. 

67 From the experience of the past and in the face of the ten- 
dencies and forces now operating we may confidently expect a greater 
development in industrial relations during the inunediately forth- 
coming years than in any preceding equal period of time. May 
engineers accept their entire responsibiUty and perform fully their 
duty in working out proper solutions of the problems presented! 



APPENDIX NO. 1 

The general principles in regard to labor incorporated in the 
German Peace Treaty signed at Versailles and found in Part XIII, 
Section 11, Article 427, are as follows: 

The High Contracting Parties, recognizing that the well-being, physical, 
moral, and intellectual, of industrial wage earners is of supreme international 
importance, have framed, in order to further this great end, the permanent 
machinery provided for in Section 1 and associated with that of the League of 
Nations. 

They recognize that the differences of climate, habits, and customs of 
economic opportunity and industrial tradition, make strict uniformity in the 
conditions of labor difficult of immediate attainment. But, holding as they do, 
that labor should not be regarded merely as an article of commerce, they think 
that there are methods and principles for regulating labor conditions which all 
industrial communities should endeavor to apply, so far as their special circimi- 
stances will permit. 

Among these methods and principles, the following seem to the High Con- 
tracting Parties to be of special and urgent importance. 

First: The guiding principle above enunciated that labor should not be 
regarded merely as a commodity or article of commerce. 

Second: The right of association for all lawful purposes by the employed 
as well as by the employers. 

Third: The payment to the employed of a wage adequate to maintain 
a reasonable standard of life as this is understood in their time and country. 

Fourth: The adoption of an eight-hour day or a forty-eight hour week 
as the standard to be aimed at where it has not already been attained. 

Fifth: The adoption of a weekly rest of at least twenty-foiu* hours, which 
should include Sunday wherever practicable. 

Sixth: The abolition of child labor and the imposition of such limitations 
on the labor of young persons as shall permit the continuation of their education 
and assure their proper physical development. 

Seventh: The principle that men and women should receive equal re- 
muneration for work of equal value. 



178 THE STATUS OP INDUSTRIAL RELATIONS 

courts will have to pass upon an increasing number of industrial- 
relations controversies. Such matters are justiciable today; they 
were not twenty years ago. We may look forward to a tune when 
controversies in regard to such rights will become just as justiciable 
as any controversy in regard to property. It is possible that when 
this time comes it will be properly referred to as the era of industrial 
democracy. 

61 Three tendencies in this development of industrial relations 
seem to be new though they are not novel. 

62 The first is the acceptance of the motive of service, which on 
moral grounds declares for recognition of the rights, needs and aspi- 
rations of every one engaged in or dependent upon industry. It is 
the engineering viewpoint rendered unselfish. 

63 The second is the willingness to consider workers in groups. 
By training and experience the engineer has only been willing to 
look upon workers as individuals. Two of the governing principles 
that have brought modem industry to its present heights are the 
division of labor which minutely subdivides the job, and the selec- 
tion and adaptation of the worker which individualizes him and 
attempts to fit him to some particular task, tool or machine. This 
is the viewpoint of speciaUzation which deals only with units. 

64 But in industrial relations the workers must be considered 
in groups or in the mass. This is the viewpoint of the industrial 
psychologist as contrasted with that of the technical engineer, and 
the latter has been slow to understand that his methods of sub- 
division and specialization cannot be used successfully in dealing with 
the problems of industrial relations. Much has been said about 
fostering cooperation, but little progress has been made. A reason 
for this situation is found in the lack of understanding on the part 
of industrial executives that to build morale or the spirit of the organic 
zation their working people must be appealed to in the mass and 
not as individuals. 

65 The development of the safety committees began to open 
the eyes of industrial executives as to what might be accomidiihed 
once employees as a body had a chance to express their deKVfSi 
and opinions. Tiic present movement to establish shop (nminittnfp 
will carry this experience further and into new aspects of the pfilh'' 
lems of industrial relations. The experience so gained will diM 
the possibiUties and advantages of discarding the individualil|j|L 
viewpoint of the engineer. It will also bring the passing of 9Sf^^ 
trary and autocratic decisions. . .'»- 



L. P. AIJX3SD 1S3 

nuoBE off 1^ lAbor Proiikaik. li^l^ ;'Iji Inm A^e. t. 
102, p. 22S»-eL 1222-5. Xcft. n-2s : 

IWilM iii Coadarenvnt an C^^apenaJcoL, 1S*1S^ In Amer. Macxl.. t. 
^ p. 749^32. OcsL. 24 
r, J. P. 
TVinp'iiig^ riKphjtl Had l«bar Togedier. 1S*1S. .Is Ind. Vwiag^ 
r. 5a, }!. 40-1. Jail. 

flf Ij^xb- TTuresi Puizis<ed Cnn rnr X&Dana] Moduxkm BoKrd. 
WIS. 3ii Efac Bj. Jul. T. 51. p SSa. Fflt. 16 ; 
OKoaavc, J. P. 

Mn PuiPBr. li*l^. li: Amer. Ihbl. lExzi. Ea^z^. Bull. 137. p 
963-41, liij. Inm Trwx: Bft^ t. «2. p. 1€11-S. June 27. JxJ. Anif?. 
6oc llecL. liDp^. X. 40. p. 4&4-5. Junt. 

CoopEBaaoL ctf EmpkiTea und lizxi^ikiTfd in IndosciT'; Ssfetj Dericef 
and MoiBiirs: GMiperh.'dai; iih: XflxianiJ IdeiJ: S«i» of Posien Isined 
bj the Nibduxiib! Inds^ozifel GuitBerr^^iaL Mcrrement. IfilS. In Azoer. 
Ind-, T. IS. p. 36-7, Mfcx. 

FmABOKl. C. 

Plm cd DemcMTkry in BumMsaE. 1S*1S. In Iron Af^e. t. 101. p. 
140^-4, VLhT 30 
Good, £. T. 

WjigBE and Pra&u ii. BrhiBL Indisczx. 1S*1S. In Tatgr.. t. 125. 

p. 47-S. Jfcl:. IS. 

Gcrrenmian Wm-Lhiior Puiiry. 1£»1S. In Askt. M&ei... t. 4S. 
p. 4!JSh90. Mju-. 21 

^wwwJl InduBiirT ill & rtamcKzunr: Abscnictc. 191S. Tl Jztl. 
Ajnex. Sol. MetL. Encrt.. x. 40. pL. 1 . p. 533-S. Jahr. Iran Tt. 'Rev., x. 
€2, p. 1607-10. June 27. Extxsjr^ tan A^t. x. 10^ p. 77. Juhr 11. 

HCATOK. J. 

Ccos^niisarT Arimz&^an i£ C^>pc«(sd ry Inx«s«%uarE of Bmisb 
IndiBChfc] TnrsBi- ItlS. li Itol Trfcde Rex., x. <S, p. 255-7. Aug. 1 . 
IxnpoTLUu I>Ex«:i'jpaxtSL''.^ £«iikitixi£ "uc Lbbur uid Capsul 191S. In 
M€!L. 4 Ciiexi-. Ijnoic.. x. li j.. -y:t.>-5. Apr. 15 

lusparinz HjjmaEx Berv*)*^ ChpTfib! hud Lbbor. Y. M- C. A. con- 
itraux: tR L^kt G^3»exk. Wk. I^IS. In Amer. Lambermiin. 2252. 
p. 41-2. J'Jx IS . 
JzsrsrzsEQS. £. J. 

Addroe of Pr»sideLi *«'^ Lc»'jkl A. L M. E. SwucmE. 15»1S. In 
Amer. Inst, Mii. Tc^zn. Bull- t. 137. Sup. 11-li, Miit. 



JiMOi^irsil Pttfc'jt — FjJTt Ii»K-*uijiJ ii. Wiiiiing Wu. 191S- In 
Auvjrmyih-e lui.. t ^ ;,. i»cr-?. Nc'x. 22 . 
LucG, C. G. 

Sctciih! Rwccifc-Lruruiic B_»jud '-^ fi>il-yw Vicuinr. 161S. Tn 
FrifTig N*'vt. X, i^j - .y>.'-7. Mtr. 21 
M£££E3F. E. J. 

Win.: of Til*: litJ'jyT Sn/uifciii'jL* 1«*1*. li KngTig Xevs, t. 80, 

p. 341-2, T*L. 7. P'.^*?, X. 47, p. 2'i*^.. F*jb. 2G- 
T. 5(0, p. 472-S, M*r. 7., 



182 THE STATUS OF INDUSTRIAL RELATIONS 

Stonby, G. 

President's Address to the Engineering Section of the British Asso- 
ciation. 1916. (In Engr., v. 122, p. 205-7, Sept. 8. Excerpts: Elec. 
Wld., V. 68, p. 870-1, Oct. 28.) 
Weiss, G. 

Bridging the Chasm Between Capital and Labor. 1916. (In 
Amer. Ind., v. 17, p. 22-3, Nov.) 
Weston, G. 

Securing Industrial Democracy. 1916. (In Elec. Ry. Jnl., v. 48, 
p. 1205-6, Dec. 9.) 
Weston, J. 

Industrial Democracy with Particular Reference to the Relations 
between Capital and Labor. 1916. (In W. Soc. Engrs. Jnl., v. 22, 
p. 125-30, Mar., 1917. Abstract, Elec. Rev. & Western Electrician, v. 
69, p. 1011, Dec. 9, 1916. Discussion, W. Soc. Engrs. Jnl., v. 22, p. 131- 
53, Mar., 1917.) 

1917 — Barnes, C. 

Industrial Peace. 1917. (In Engr., v. 123, p. 35-6, Jan. 12.) 
Bebrt, G. L. 

Sharing the Responsibility of Industry. ,1917. (In Inland Ptr., 
V. 58, p. 670-1 Feb.) 

Changing Existing Standards; Statement of the National Industrial 
Conference Board Respecting National Labor Situation. 1917. (In 
Amer. Ind., v. 18, p. 33-5, Nov.) 
Hodge, J. 

How Great Britain is Meeting the Labor Problem. 1917. (In 
Factory, v. 19, p. 21-4, July. 111. Plans.) 

Industrial Councils; Report of Sub-oommittee of Government Re- 
construction Conmiittee. 1917. (In Engr., v. 124, p. 10, 13, July 6.) 

Industrial Reconstruction. 1917. (In Ulum. Engr., v. 10, p. 243, 
Sept.) 

Industry and the Whitley report. 1917. (In Engr., v. 124, p. 169, 
Aug. 24.) 
Kent, S. 

War Problems. 1917. (In Engng. News, v. 79, p. 957-8, Nov. 22.) 

LoNGRnOGE, M. 

President's Address, Institution of Mechanical ESngineers. 1917. 
(In Engr., v. 123, p. 373-4, Apr. 27.) 
NoxoN, F. W. 

Paradox of Unrest. 1917. (In Amer. Ind., v. 18, p. 15-8, Aug.) 

Secret of Development. 1917. (In Engr., v. 123, p. 85, Jan. 26.) 
Walker, P. F. 

Ethical Tendencies in Modern Industrialism. 1917. (In Ind. 
Management, v. 53, p. 877-85, Sept.) 
Wheeler, H. A. 

Organized Labor and Business Must Accept Equal Obligi^tionB. 
1917. (In Eng. Rec, v. 75, p. 219-21, Feb. 10.) 

1918 — Alexander, M. W. 

Elements of the Labor Problem; with Disouflsion. 1918. (In 
Textile Wld., v. 53, p. 5143-4a, May 4.) 



L. P. ALFOBD 183 

InDqpcnrtant Phases of the Labor Problem. 1918. (In Iron Age, v. 
102, p. 1259-61, 1322-5, Nov. 21-28.) 

Babeon Conference on Cooperation. 1918. (In Amer. Mach., v. 
49, p. 749-^ Oct. 24.) 
Bbofht, J. P. 

Bringing Capital and Labor Together. 1918. (In Ind. Manage- 
ment, Y. 55, p. 40-1, Jan.) 

Canses of Labor Unrest Pointed Out by National Mediation Board. 
1918. (In Elec. Ry. Jnl., v. 51, p. 333, Feb. 16.) 
Channing, J. P. 

Man Power. 1918. (In Amer. Inst. Min. Engrs. BuU. 137, p. 
963-9, May. Iron Trade Rev., v. 62, p. 1611-3, June 27. Jnl. Amer. 
Soc Medi. Engrs., v. 40, p. 464-5, June.) 

Cooperation of Employers and Eknployed in Industry; Safety Devices 
and Measures; Codperation the National Ideal; Series of Posters Issued 
by the National Industrial Conservation Movement. 1918. (In Amer. 
Ind., V. 18, p. 26-7, May.) 
Fraboni, C. 

Han of Democracy in Business. 1918. (In Iron Age, v. 101, p. 
1402-4, May 30.) 
Good, K T. 

Wages and Profits in British Industry. 1918. (In Engr., v. 125, 
p. 47-«, Jan. 18.) 

Government War-Labor Policy. 1918. (In Amer. Mach., v. 48, 
p. 489-90, Mar. 21.) 
Hatnbs, G. H. 

Small Industry in a Democracy; Abstracts. 1918. (In Jnl. 
Amer. Soc. Mech. Engrs., v. 40, pt. 1, p. 533-8, July. Iron Tr. Rev., v. 
62, p. 1607-10, June 27. Excerpts. Iron Age, v. 102, p. 77, July 11.) 

HOBTON, J. 

Compulsory Arbitration b Opposed by Investigators of British 
Industrial Unrest. 1918. (In Iron Trade Rev., v. 63, p. 255-7, Aug. 1.) 
Important Developments Relating to Labor and Capital. 1918. (In 
Met. &, Chem. Ekigng., v. 18, p. 393-5, Apr. 15.) 
Inspiring Harmony Between Capital and Labor. Y. M. C. A. con- 
ference at Lake Geneva, Wis. 1918. (In Amer. Lumberman, 2252, 
p. 41-2, July 13.) 
Jennings, S. J. 

Address of President to Local A. I. M. E. Sections. 1918. (In 
Amer. Inst. Min. Engrs. Bull., v. 137, Sup. 11-13, May.) 
Kent, S. 

Industrial Peace — Vmt Essential in Winning War. 1918. (In 
Automotive Ind., v. 37, p. 907-9, Nov. 22.) 
Lang, C. G. 

Social Reconstruction Bound to Follow Victory. 1918. (In 
Engng. News, v. 80, p. 566-7, Mar. 21.) 

MSHBBN, E. J. 

What of The Labor Situation? 1918. (In Engng. News, v. 80, 
p. 241-2, Fd). 7. Power, v. 47, p. 298, Feb. 26. Discussion, Engng. 
News, Y. 80, p. 472-^, Mar. 7.) 



184 THE STATUS OF INDUSTRIAL BELATIONS 

New Industrial Dispensation Must Be Made Labor Insuranoe. 1918. 

(In Engng. & Min. Jnl., v. 106, p. 318-9, Aug. 17.) 

Of What Use is the Millionaire? 1918. (In Engng. & Min. Jnl., v. 105, 

p. 470, Mar. 9.) 

Philadelphia Rapid Transit System Cooperative Plan Extended. 

1918. (In Elec. Ry. Jnl., v. 52, p. 459-62, Sept. 14.) 

Relations Between Capital and Labor in India. 1918. (In Amer. Ind., 

V. 18, p. 44, Mar.) 

Rockefeller, J. D. Jr. 

Relation of Capital to Labor. 1918. (In Amer. Arch., v. 114, p. 
262, Aug. 28.) 

Shop Stewards and the Coventry Strike. 1918. (In Engr., v. 125, p. 8, 
Jan. 4.) 

Significant Changes in Business Management. 1918. (In Amer. 
Mach., V. 49, p. 191-3, Aug. 1.) 
Tipper, H. 

Agreement vs. Bargaining. 1918. (In Automotive Ind., v. 39, p. 
784-5, Nov. 7.) 

Labor Representatives Lack Sufficient Power. 1918. (In Auto- 
motive Ind., V. 39, p. 835,' Nov. 14.) 

Lack of Confidence Between Employer and Employee; the White 
Plague of Labor Efficiency. Diags. 1918. (In Automotive Ind., v. 
39, p. 625-6 +, Oct. 10.) 

Which Side Are You On? 1918. (In Automotive Ind., v. 39, p. 
669-71, Oct. 17.) 
Warren, L. L. 

Propaganda in Industrial Relations. 1918. (In Ind. Management, 
V. 55, p. 197-8, Mar.) 

Whom Are We Fighting? 1918. (In Amer. Mach., v. 49, p. 211, Aug. 1.) 
1919 — Allen, R. 

Industrial Unrest. 1919. (In Power, v. 49, p. 296, Feb. 25.) 
Alwtn-Schmii>t, L. W. 

Organizing the Nation for Peace. 1919. (In Indus. Management, 
V. 57, p. 45-8, Jan.) 
Gillespie, J. J. 

Indifierentism: The Present Danger in Manufacturing. 1919. (In 
Indus. Management, v. 57, p. 38r-9, June.) 
Tipper, H. 

Social Surroundings Have Important Bearing on All Labor Ques* 
lions. 1919. (In Automotive Industry, v. 40, p. 366-7, Feb. 13.) 

BOOKS ^ 

1912 — Talbot, W. 

Select Bibliography of Recent Publications on the Helpful Relations 
of Employers and Employed. W. Talbot, Cleveland, Ohio. 1912. 

1913 — Commons, J. R. 

LalK)r and Administration. New York. MacMillan. 1913. Re- 
lation of labor to scientific management. 
Nearinq, Scott. 

> These books are not av&iUble in the U. E. 8, Library* 



L. P. ALFOBD 185 

Financing the Wage Earner's Family. 1913. New York. 
Hudbsch. Discussion of wages and living conditions from radical 
standpoint. 
Spargo, John. 

Sindicalism, Industrial Unionism and Socialism. New York. 
Huebsch. 1913. An exposition of the radical attitude towards indus- 
trial relations, by a Socialist. 
Tridon, a. 

New Unionism. New York. Huebsch. 1913. The attitude of 
the extreme industrial unionists (I. W. W.) toward industrial relations. 
Wilson, Woodrow. 

The New Freedom. Garden City, N. Y. 1913. Deals with the 
relations between capital and labor. 
New York (State). Factory Investigating Conmiission. 

Preliminary report. New York. State Factory Investigating Com- 
mission. 1912. 3 V. 

Reports 2-4, published 1913-1915. 8 vols. 

1914 — U. S. — Industrial Relations Conmiission. 

First annual report for year ending Oct. 23, 1914. U. S. Govt. 
Printing Office. 1914. 
U. S. — Mediation and Conciliation Board. 

First annual report of the Commissioner of Mediation and Concilia- 
tion. Wash. U. S. Supt. of Documents, 1913-1914. 

1915 — Grant, Luke. 

National Erectors Association and the International Association of 
Bridge and Structural Iron Workers. U. S. Com. on Industrial Rela- 
tions. F. P. Walsh, Chairman, Kansas City, Mo. 1915. 
Rockefeller Foundation, New York City. 

Information furnished by the Rockefeller Foundation in response 
to questionnaires submitted by United States Commission on Industrial 
Relations. New York. Rockefeller Foundation, 1915. 
U. S. — Industrial Relations Conmiission. 

Report on the Colorado strike. U. S. Commission on Industrial 
Relations, F. P. Walsh, Chairman, Kansas City, Mo. 
U. S. — Industrial Relations Commission. 

Report of the investigation of the wages and conditions of telephone 
operating. U. S. Commission on Industrial Relations. F. P. Walsh, 
Chairman, Kansas City, Mo. 1915. 

1916 — American Labor Year Book. (Annual.) v. 1-. 1916-. New York. 

Rand School of Social Science, 1916 +. A review of labor condi- 
tions in various countries, giving statistics, industrial relations, etc. 
Co^?5N, J. H. 

Law and Order in Industry. New York. MacMillan. 1916. 
Collective bargaining or the working out of trade agreements between 
employers and employees. 
Nearinq, Scott. 

Poverty and Riches. Phila. Winston. 1916. A criticism of the 
wage system by a radical, recommending sweeping changes in the rela- 
tion of labor and capital. 
U. S. — Industrial Relations Commission. 



186 THE STATUS OF INDUSTRIAL RELATIONS 

Final report, including the report of Basil M. Manly, Director of 
Research and Investigation, and the individual reports and statements 
of the several commissioners. (Repr. from Senate Doc. No. 415, 64th 
CJongress. Washington, Govt. Printing Office, 1916.) 
U. S. — Industrial Relations Commission. 

Industrial relations; final report and testimony. (64th Congress, 
Ist Sess., Senate Doc. 415, 1916.) 11 volumes. Full report of hearings 
of the Commission on wages, hours of labor, management, strikes and 
industrial relations. 

1917 — American Academy of Political and Social Science. 

Present labor situation. Phila. Publ. by Academy. 1917. On 
industrial arbitration of labor disputes. 

1918 — Henderson, Arthur. 

Aims of Labor. New York. Huebsch. 1918. Sets forth the 
policies of the British Labor Party and the British labor imions on 
industrial relations, reconstruction, etc. 

1919 — American Academy of Political and Social Science. 

Reconstruction Labor Policy. Phila. Pub. by Academy, 1919. 
Reconstruction and industrial relations. 



DISCUSSION 

The following excerpts from the general discussion on Industrial 
Relations apply to the preceding papers, Nos. 1692 and 1693. 
An extended account of the discussion of these papers was given in 
Mechanical Engineering for July 1919. 

Fred J. Miller was of the opinion that the industrial problem 
resolved itself mainly into the establishment of a business relation 
between employer and employee of such a character as usually 
exists between any other classes of people who do business together 
— relations which must be mutually and permanently satisfactory 
to each side involved. 

The employee must feel that he is doing at least as well as he 
could do with another employer in the same line, and the employer 
must feel that he is doing about as well as he could with any other 
body of employees. 

A business relation implies, also, the right of either party to 
present a proposition at any time for a change in the tenns of that 
relation. It should be agreed that in any negotiations regarding the 
terms of employment, either side may have the right to be repre- 
sented by a representative of its own choosing. In these days of 
large corporations, the owners of a business have chosen, under 
the name of officers, superintendents, foremen, and so on, men who 



DISCUSSION 187 

represent them in all negotiations concerning the business relation 
that should exist between employer and employee. Nobody dis- 
putes the rights of the owners to select such representatives of their 
own choosing, and if a business relation between employer and 
employee is to exist, the employees should have exactly the same 
right to choose their representatives. These may constitute a com- 
mittee of their own number or they may be union representatives; 
but so long as they are chosen by the employees they should be 
treated as their representatives and allowed to state their case. 

When an employer has sat down for a discussion across the 
table with representatives of that sort, whether they be employees 
of his own or union representatives, or what, and views have been 
expressed candidly and freely as man to man, it will often be found 
that many of the difficulties which may have been anticipated will 
disappear; in fact, that thereafter there is no difficulty whatsoever. 

Forrest E. Cardui;i40. We have at the present time two con- 
flicting ideas of what constitutes industry. One of these is based 
on the old idea of competition between all men for the good things 
of life. It falls back on the proposition that the primary purpose 
of industry is to make money for the capitalist or owner. Incidently 
it furnishes a means of livelihood to the worker and in doing so 
affords the capitalist or owner an opportunity to secure a commod- 
ity which he desires, namely, service in exchange for wages. The 
owner has the primary rights and privileges and the entire control, 
while the worker is put in the position of competing with his fellow 
workers in order to dispose of a perishable commodity, service, sus- 
ceptible only to the operation of the laws of supply and demand. 

Now we are coming to a point of view which regards the pri- 
mary purpose of industry as service to the community and con- 
tributory to the general welfare of the nation; while the general 
welfare of the capitalists, owners, managers and employees who 
are engaged in the industry becomes secondary. 

Our present-day attitude of mind on questions of capital and 
labor is such that we regard industry as passing from the competi- 
tive to the cooperative basis; from the idea of the entire and exclusive 
control by the owner to the idea of joint control, primarily in the 
interests of the community. In this we base our ideas on the theory 
of a democratic government which demands equal rights and privi- 
leges for all men. 

The defects of the competitive system we have all seen and will 



188 THE STATUS OP INDUSTRIAL RELATIONS 

continue to see as long as it remains such, which it is liable to do for 
some time to come. They are (1) that labor is under no moral obliga- 
tion either to the community or to the employer to give its best, or 
to cooperate for a joint benefit; and (2) labor believes that it is 
compelled to adopt forms of organization and methods of procedure 
which emphasize the fighting power of the union and which develop 
inefficiency rather than efficiency and a spirit of antagonism rather 
than a spirit of cooperation. The men in charge of the activities of 
the union are dependent for their support upon fostering, at least, 
a reasonable amount of friction, so that they may have something 
to do and may point with pride to their achievements. Before we 
can do away with this spirit we must substitute something which 
will assure to the workmen a larger measure of the good things of 
life and larger measure of control in the things in which he is 
interested. 

This can only come about gradually and as the result of eco- 
nomic education of the workingman who must understand what 
he is to strive for, what kind of men he must elect to take care of 
his interests, and that efficiency nuist l)e the l)asis for any increase 
in his material welfare. 

Otto P. Geier.* There is great need today for trained indus- 
trial physicians who will not be satisfied simply to be called in as 
specialists, but who will actually make a contribution to the solu- 
tion of some of the very difficult problems that are facing industry 
today. When we take into account that there is, perhaps, a billion 
dollars' worth of loss in this country per year, at least half of which 
is preventable by proper medical supervision in industry, then the 
problem that looms up for solution by the industrial physician, an<l 
by engineers and managers, is worth talking about. 

Industrial Relations means to me that feeling of responsibility 
which the farseeing management has; that in having large units of 
society, such as industry, in its charge it has all the responsibilities 
that a comnmnity has toward large groups; that it must exercise all 
the care in regard to the matters of health and sanitation and safety, 
and all the things that naturally come up, and perhaps the thing 
that has not been stressed today sufficiently Ls the matter of housing 
— the matter of living conditions. 

Many of you know a man out in the field today, who is a per- 

^ Director Employees' Service Department, Cincinnati Milling Machine 
Co., Cincinnati, Ohio. 




DISCUSSION 189 

sonal service man, who is in overalls, and going from plant to plant 
in this country trying to find out what workmen are thinking about. 
I recently spent about five or six hours with him to find out, if possi- 
ble, what his reaction had been as a result of about four months in 
overalls, working in coal mines, in steel mills, and doing the things 
that other men do who earn their money by the sweat of their brow. 
Perhaps the thing that he stressed more than anything else 
was the fact that in spite of all the talk that we have had about 
sanitation, safety and medical examinations, we haven't even 
scraped the surface. We are not getting this message across to the 
workmen, and we haven't begun to realize how much we can do 
for them by actually caring for their health. Living conditions are 
the things that are making men start this unrest, and they are com- 
ing to the factory in the morning tired and miserable and ill at ease 
with everybody and the world at large. 

Boyd Fisher^ said that he was interested in presenting the 
social and economic justification for the type of plan which Mr. Alford 
had described and Mr. Young enlarged upon, whereby the struggle 
for control in industry might to some extent be compromised. 

The problem of the maladjustment of capital and labor is of 
interest because such maladjustment reduces production. The 
public is interested because lessened production increases the cost 
of commodities; and the manager, engineer and capitalist alike are 
interested because profits depend on rate of output. 

As to labor, the speaker said that it had occurred to him that 
one of the reasons why labor willfully restricts and limits output, 
although it is one of the partners of production, is because it is the 
one holding the least advantageous position in the apportionment 
of the production, and in formulating the rules by which that appor- 
tionment is made. 

This is due to the fact that those who undertake industry pay 
off all of the other factors of industry. They pay rent and interest, 
they pay the makers of the tools and suppliers of material, they 
pay the managers and they pay labor, and then they keep what is 
left. 

These factors, other than labor, are mostly beyond the control 
of those who imdertake industry. The manufacturer cannot con- 
trol rent, for example, and even his paid managers have a bargain- 
ing power fixed in part by market conditions and in part by their 
^ Consulting Engineer in Management, 362 Burlingame Ave., Detroit, Mich. 




190 THE STATUS OF INDUSTRIAL RELATIONS 

ability to stay out of the market if the price offered does not suit 
them. Labor, on the other hand, has no such advantage and has 
to come into the market without suflScient resources and reserves 
and take what the market conditions provide. So it has been, at 
least, in the past. 

Inasmuch, therefore, as capital has to pay ofif all of the other 
factors at prices over which it has no full control, at figures that 
must be satisfactory to the other factors before the bargain is made, 
and does not have to pay off labor in accordance with a bargain 
that is satisfactory to labor, capital is often able to derive an advan- 
tage at the expense of the workers in industry. 

Mr. Young has proposed a method whereby labor can advantage 
itself equally with the other factors, so that, in a sense, it can say 
whether it will or will not stay out of the market. It can protect 
itself and get a fundamentally right bargain, which is the first step 
toward the basic need of the public in relation to industry, namely, 
to increase production. For with such protection labor has no 
just grounds for attempting to restrict output. 

L. W. Wallace ^ said that labor problems have always existed 
and are likely to continue to arise as long as humanity is constituted 
as it is. He, therefore, wanted to sound this warning: 

"I do not anticipate that at this time or at any future period 
there will be evolved a panacea that will forever solve any and all 
problems that may arise between employer and employee. This 
is no more possible than that a plan can be evolved whereby there 
will be no more wars between nations. Some form of industrial 
democracy on the one hand and a league of nations on the other 
will unquestionably be an agency of great value and influence, but 
these agencies within themselves will not eliminate labor troubles 
nor make impossible future wars. In fact, no instrumentality or 
agency will accomplish much unless there be behind them and dis- 
seminated throughout every fiber and thread the spirit of fairness, 
honesty and justice. 

**In all sincerity the principle of the * Golden Rule' must obtain: 
do unto your employee as you would have him do unto you. I 
believe it is the duty of the employer first to demonstrate that he 
is operating on tliis principle. It is his responsibility to engender 
into the minds of the employees perfect confidence, absolute warm- 
heartedness and cordial respect for him. 

^ Director Red Cross Institute for the Blind, Baltimore, Md. 



DISCUSSION 191 

"It is also my conviction that no man will ever succeed as a 
leader of men and solve the industrial problems, miless he has a 
large store of hmnan sympathy in his heart. Unless we are sym- 
pathetic, we are cold and indifferent to those matters that are near- 
est and dearest to men. We are apt to be impatient with hmnan 
weaknesses, we are apt to make demands that are unfair and un- 
reasonable, and we are apt to be cruel in our decisions and rash in 
our actions. Unless the principles enunciated are carried out, no 
satisfactory results will be obtained, whether the plan is a committee 
system, an industrial democracy system, or a House and Senate 
plan. The plan or the machinery^ whereby you are to operate is not 
nearly so important as the motive that prompts, the ideals that pre- 
vail and the sincerity that obtains." 

Sam H. Libby stated that they had had a cooperative com- 
mittee in operation at the Sprague Electric Works since last De- 
cember. Ninety per cent of the 2000 employees voted to try the 
plan, which is much like that described by Mr. Young. In the 
Sprague committee, however, the elected members number about 
two to the hundred, with chairman, vice-chairman and secretary. 
Three members were elected to each of six sub-committees and on 
the sub-conmiittees the management appointed three other mem- 
bers, making six members each. These sub-committees each selected 
a chairman, which in every case proved to be an appointed repre- 
sentative, showing the confidence of the men in the proposition. 
Mr. Libby described at length some of the practical workings of the 
committee. 

Richard A. Feiss, as the result of a discussion introduced by 
Mr. Young as to the extent to which the question of management 
is an engineering problem, contended, strongly that it is strictly 
such and that one's views should be big enough to make engineering 
stand for the things with which the engineer has to deal today 
rather than for the mere word "engineering'' as defined in the old 
dictionaries. He contended that primarily mechanical engineering 
is production. The purpose of a machine is to produce and the 
machine itself must be produced in the process of manufacture. 
Many times an engineer has seen the best theoretically designed 
machine go to pieces in the hands of the average workman, because 
it had been forgotten that a man had to control the machine and 
its production, and that its successful operation could not be de- 



192 THE STATUS OP INDUSTRIAL RELATIONS 

termined solely by the working drawings on which it is based. How, 
then, can one say that a problem affecting the man whose labor is 
assisting in production, or perhaps being replaced by a machine 
which is designed, does not present what is distinctly and directly 
an engineering problem? Again, we note that the running of two 
or three machines in a group is much more complicated than the 
running of one. It must be evident that the questions of the man 
and the machine cannot be separated and that their relationship 
is so essential in securing production that the question of labor and 
its management is distinctly an engineering problem. 

All works councils and other plans to develop the relationship 
between the management and men are merely part of the general 
scheme needed to develop and improve the conditions of the worker 
in order to make him a more contented and better workman with a 
view to enhancing production. None of these plans in themselves 
will do. 

The solution to th problem, which is variable from time to 
time, depends upon taking into consideration every factor with a 
clear vision as to one's own objects and the ultimate relationship 
to be established. It has been well said that there are no panaceas. 
It should be said however, from one point of view that there is one 
panacea, viz. truth. It is necessary to study and realize the truth 
on the part of the management and to give full publicity in every 
possible way to the truth in the plant and outside of the plant in 
order to educate workmen and public at large and bring them closer 
to the realization of the mutual problem evolved in industry. And 
the ultimate solution of the industrial problem depends upon our 
ability to see and to make others see the truth. 

H. F. J. Porter (written). In his admirable paper Mr. Alford 
makes mention of my article in the Engineering Magazine for 
August, 1905, which presents a fairly clear idea of my installation 
of a factor}' committee in the plant of the Nernst Lamp Company, 
in Pittsburgh, in the winter of 1903-4, which I understand was the 
pioneer installation of its kind in this country. In 1905 I presented 
a paixir before The American Society of Mechanical EIngineers 
entitled The Realization of Ideals in Industrial Engineering/ in 
wliich I referred to the merits inlierent in enlisting the interest of 
the human element regarding matters afifecting it in the manage* 
ment of an industrial plant. The basis for this paper was the 

^ Trans. Am. Soc. M. E., Vol. 27, page 343. 




DISCUSSION 193 

experience obtained with my shop committee in the Nernst Lamp 
Company. 

When I took charge at the Nerast plant I decided to install the 
suggestion system used successfully by the National Cash Register 
Company, but disliking the method of handling all such features 
for the management, I determined to have the suggestions passed 
on first by a committee of the employees themselves and, therefore 
requested the latter to elect by secret ballot a representative to it 
from each department. The committee so formed elected its own 
chairman and secretary and all suggestions were collected from 
the boxes by the secretary and prepared by him for the committee. 

Among the questions considered by the committee were the 
following: 

Wage Payment, the premium system being finally adopted. 

Permanency of Employment. The committee considered the 
taking on and laying ofiF of all employees and succeeded in stabiliz- 
ing the working organization, lowering the labor turnover developing 
functions now accorded to an employment manager. 

Sickness and Accident Prevention, resulting in suggestions for 
safety and the appointment of a nurse in charge of an emergency 
hospital and a visiting physician. 

Fire Drill. There had been a panic due to a false alarm of fire 
in the factory before I took charge, which led to the development 
of a factory fire drill. I found to my surprise after a canvass of 
representative factories of the country that further than developing 
a fire-fighting brigade for handling the hose, etc., no such thing as 
a fire drill existed anywhere. The fire drill at the Nernst plant was 
the first established anywhere, which at a given signal would take 
the employees out of a factory building in quick time. 

Mr. Alford refers to the fact that my shop committee was com- 
posed of workmen and foremen and that the superintendent pre- 
sided. The first committee was formed wholly of working men and 
women but as they were advanced to positions in the management 
and the employees reelected some of them to office, some foremen 
were on the committee and the superintendent was made chairman. 
There was, however, a second committee representing the man- 
agement composed of heads of departments and of which I was the 
chairman. 

The second shop conmiittee which I installed was in the Nelson 
Valve Company's plant in Philadelphia in 1907. Here we had a 
shop conmiittee of workers only and an advisory board composed 



194 THE STATUS OP INDUSTRIAL RELATIONS 

of foremen and the superintendent. These committees met weekly, 
first separately and then together. I think this arrangement will 
be found to give the greatest satisfaction. 

Mr. Alford mentioned in his paper that it does not seem per- 
tinent to devote space to certain activities that classify under the 
definition of industrial relations. I feel that he minimizes the im- 
portance of some of these features, particularly the suggestion 
system, benefit associations and pensions, which seem to me to be 
of as vital importance as profit sharing, methods of wage payment, 
the safety first movement, employment management, mutual or 
joint control, etc. 

Scientific Management, always diflScult to install in a factory 
under autocratic management, goes in as a matter of course under 
committee management. 

Cyrus McCormick, Jr. ^ (written) I disagree profoundly with 
Mr. Alford^s statement that "students of the present condition of 
unrest have pointed out that the fundamental is a struggle for con- 
trol, the opposite forces being the owners and the workers." The 
idea underlying employee representation is not a question of class 
struggle, but rather of removing class distinction. It is a construc- 
tive endeavor to secure added benefits which are not granted by our 
present system, rather than a restricted effort to prevent the spread 
of ills incident to industrial warfare. It may, it is true, prevent the 
spread of these ills, but if it cannot at the same time ameliorate 
conditions, it must philosophically be regarded as a failure. The 
first two principles of employee representation hereinafter described 
prove these points. 

Employee representation, if it is to succeed, must include a 
practical application of the following fundamental principles: 

a There must be full representation for employees concern- 
ing working conditions, protection of health, safety, 
wages, hours of labor, recreation, education, and the 
like. This statement, of course, involves a recognition 
of the right of collective bargaining. It assumes that 
the workers, individually and as a group, are intellectu- 
ally capable of maintaining their share in the joint dis- 
cussion with chosen representatives of the management. 

b Joint conference between men and management. It is not 

* Works Manager, International Harvester Co., Chicago, HI. 



DISCUSSION 195 

sufficient to grant the employees merely the right to 
discuss their own affairs. This discussion must take 
place with the management or its representatives. If 
any subject can be brought out into the open where it 
can be tested by the clash of men's minds and where 
honest opinions are exchanged in free and frank dis- 
cussion, a happy solution of any debated point cannot 
be long delayed. The fundamental point is to get to- 
gether; to talk things over; to debate them; and so to 
imderstand each other's opinions and points of view. 

c Shop committees differ only in the basis of representation. 
The great point is to secure a sufficient representation to 
build the plan upon an essentially democratic foundation. 

d Employee representation must include an easy channel of 
approach from the lowest workmen to the highest official 
in the company whereby the former can appeal to the 
justice of big minds. In this way it is to be hoped that 
personality can be reintroduced into industry without 
depriving industry of the efficiency of broad organization. 

e There must be no discrimination on account of race, sex, 
political or religious affiliation, or membership in any 
labor or other organization. This is an essential tenet 
of democracy, and if it is not included in the groundwork 
of any plan of committee representation, that committee 
cannot succeed. 

/ Finally, there must be executive supervision for the plan. 
In this way the experience gained from day to day can 
\ye collected and coded into a working principle for the 
future. The handling of labor problems is becoming the 
work of specialists, and as the years progress the special- 
ist in labor matters will be more and more important 
even than he is at present. Certainly this is not an 
engineering problem in the narrow use of the word 
"engineering," nor is it a problem for psychologists or 
sociologists. It is rather an attempt to translate hu- 
manity into the language of modern industry. The 
whole problem must be looked at, not merely on ethical, 
but also on economic grounds, and the future develop- 
ment of Industrial Relations will, it is believed, depend 
upon a satisfactory solution of this dual view of the 
problem. 



196 THE STATUS OF INDUSTRIAL RELATIONS 

Mack Gordon ^ (written). Although the shop committee and 
collective bargaining outside of trade-union control are compara- 
tively new and untried methods in management, some conclusions 
can be drawn from the limited experience at hand. 

If the worker does not understand clearly what are the func- 
tions of the shop committee, and what his part is to l)e, it will not 
have his confidence, which Ls essential. 

How Ls this confidence to be built up? The management must 
make up its mind absolutely to turn over to the workers the right 
of bargaining with it on any matter whatsoever that pertains to 
the workers* interests, such as: 

a Wages. The quantity to be produced for those wages. 
The method of determining what the quantity to be 
produced shall be, such as time study. Individual 
requests for increase in pay 

b Hours of work 

c Any rules or regulations affecting the conduct of the worker 

d HLs right to a hearing in case of discharge. 

More than five years' experience has proved that the workers 
are reasonable. As long as there is confidence in the committee and 
in the honesty of the management, they will remain so. They do 
not ask for impossible things and when they do it is only because 
of lack of knowledge as to conditions. A frank and honest discus- 
sion will soon dissipate the misunderstanding. 

However, the mere existence of a shop committee that can 
present grievances, ask for increases in pay, changes in hours, review 
of discharges, etc., is not enough. When the committee finds some- 
thing wrong and asks that it be remedied, in the most important 
cases, the conditions can only be remedied by a change of manu- 
facturing policy, or of methods of management. The management 
must be willing and able to make these improvements. If it cannot, 
the mutual confidence that is so necessary will be dissipated, due to 
the lack of ability to satisfy the workers. It is in this phase of the 
situation that scientific management can help to solve the problem. 

WiLLARD G. Aborn ^ (written). In the first place the employer 
must be entirely convinced that the employee committees can and 
shall work out to the mutual l:)enefit of all concerned. Fair-minded- 
ness on the part of the employer and a willingness at the Ix^ginning 

^ Consultant, 226 Marion Bldg., Cleveland, Ohio. 
« 619 West 113th St., New York City. 



DISCUSSION 197 

to go more than half way to convince his employees that he is ear- 
nestly endeavoring to bring about a mutually beneficial condition 
is a necessary preliminary to the successful inauguration of such a 
scheme. 

There have been and always will be dissatisfied radical workers 
in every plant, whose influence increases or decreases in proportion 
as the fairness of the employer decreases or increases. Therefore, 
by all means, in instituting these committees, reduce the influence 
of these radicals by giving the rationals, who are about 80 per cent, 
entire freedom in the selection of their representatives; in other 
words, remove all possible suspicion of undue influence at the elec- 
tions by arranging that they shall be directly under the supervision 
of the workers themselves or their representatives. Do not permit 
any official or foreman to in any way dominate, interfere, or even 
suggest, except as they may be called upon by the workers for 
advice. 

Ejections without nominations seem desirable and have worked 
out extremely successfully where used, in that conservative em- 
ployees of long service in the plant have almost invariably been 
elected. Also the time required is much less and the result is really 
more democratic. 

Department committees of three are also recommended for the 
reasons that three are more constructive than one and when repre- 
senting their own department only have intimate personal knowl- 
edge on any subject affecting their constituents. 

Welcome joint discussion on any matter pertaining to the plant 
— in fact, if necessary or desired, originate matters for discussion 
so that the committees may function and the workers' interest in 
the plant welfare be stimulated. Those who have had dealings 
with committees know that there is nothing much more dead than 
a non-working committee. Joint meetings of employer and em- 
ployees' committee should be held frequently enough to keep up 
a Uve interest. Allowing the committeemen to honestly feel that 
they are originating and inaugurating matters for common good 
will tend to keep their enthusiasm aUve and working. 

Guy p. Miller^ (written). Although the plan of Industrial 
Relations adopted by the Bridgeport Brass Co. has been in opera- 
tion only nine months the results have been far greater than antici- 
pated. About 100 meetings of committees have been held during 

^ General Manager, Bridgeport Brass Co., Bridgeport, Conn. 



198 THE STATUS OP INDUSTRIAL RELATIONS 

this time, at 25 per cent of which hours and wages have been dis- 
cussed, and in no case has the conclusion reached been other than 
unanimous and entirely satisfactory to the employees and the 
management. Every employee has an opportunity to be heard 
whenever he has a grievance, which enables the company to adjust 
little things which cause annoyance and to explain other things to 
the satisfaction of the men. In order to get the greatest benefit out 
of these committees, the safety and sanitation work, as well as the 
work of the sick benefit association with the insurance features, 
have been turned over to them. Joint committees are also handling 
recreational and athletic activities in all plants of the company. 

Every three or four months an evening meeting of all the com- 
mittees is held with a dinner in our cafeteria, and I explain the gen- 
eral condition of the business and ask for expressions from the men, 
which has helped cement the spirit of confidence which has been 
established. No plan will be successful which fails to instill confi- 
dence, but as soon as the men are confident that the management 
will give them a square deal they will go half way. 

Many feel that agitators may gain control of the committees 
and organize them as union representatives, but I am convincetl 
that this danger is not to be feared as long as the management has 
the confidence of the employees, and when this is gone no plan can 
be successful. 

P. J. Reillt ^ (written). The work of industrial engineers 
and employment managers must be supplemented by a new type 
of foreman — one who has been trained for his foremanship. In- 
formation that will assist foremen to administer their departments 
more effectively should be organized and presented to them in a 
form that will make the material readily available. The philosophy 
behind any rules of organization which foremen are expected to 
enforce should be explained so that they can enforce such rules 
without appearing arbitrary. Regular meetings for foremen should 
be held for the discussion of problems affecting the planning of 
work, the quahty, quantity, and economy of production, the han- 
dling of new workers and the promotion of deserving older workers. 
Such meetings are effective in the enUghtenment of foremen so that 
they are in entire accord with the management's policies in each of 
these fields. Much can be done in the development of foremen by 

^ Head of Personnel Division of the Retail Research Association, 225 Fifth 
Ave., New York. 



DISCUSSION 199 

relieving them from duty in their department for short periods so 
that they can work under the direction of the employment manager, 
the industrial engineer, or the master mechanic. 

Opportunity should be given to foremen to attend special courses 
on industrial management, or to visit other factories that they 
may get the broadening which so many of our foremen badly need. 

If the industrial leaders will give them the chance, the foremen 
will learn to treat help with a degree of human sympathy that will 
result in better team work. They will develop more patience, judg- 
ment and tact, and will eventually realize that men under them will 
produce best when their heads and hearts are in their work as well 
as their hands. 

D. G. Stanbrough (written). With reference to the develop- 
ment of the personal relations, this is a matter of organization. The 
foreman, in the minds of the workmen, represents the management, 
and consequently, if we are to be successful in maintaining personal 
relations, it becomes necessary that we develop a high class of fore- 
manship, and this instruction work must be done by executives 
familiar with the broad policies of the business and who have had 
the necessary practical experience to enable them to talk to the 
foreman from the foreman's viewpoint. 

The foreman must be a man of the proper amount of personal 
kindness and should be able to understand the psychology of man- 
agement. Such foremen will be found among every class of workers, 
or at least potential foremen of this caliber, and to develop such 
men it goes without saying that the factory executive must have 
the proper viewpoint. 

I want to take particular issue with the closing paragraphs in 
the paper. I do not think that we can ever hope for any real meas- 
ure of success if we are to consider workers in groups or masses. 
You cannot build morale by appeal to masses. I believe that the 
appeal must be made to the individual, through the organization 
line. The spirit of organization can be strongly built up if each indi- 
vidual worker feels that the management is conscious of his efforts 
and that the appeal is to him personally. Even in a large organiza- 
tion much can be done by the executive in knowing his men. At least 
he can know a few men in each department, and through them the 
spirit can be communicated to the organization. 

John L. Henning (written). As Mr. Alford states, the loss of 
personal contact is one of the contributing causes of present unrest; 



200 THE STATUS OF INDUSTRIAL RELATIONS 

but if employers will use the engineer to the fullest extent and per- 
mit him to take some responsibility in handling labor questions he 
will, I believe, demonstrate that his so-called fault of dealing with 
labor from an individualistic viewpoint may be turned to a very 
good account. He will use a little more human sympathy in dealing 
with labor, which after all is simply a collection of human beings 
who inherently have the same "rights, needs and aspirations" as 
the employer and those dependent upon their product of industry. 
Referring to Mr. Alford's conclusions as to the reasons for the 
apparent failure of remedies heretofore applied, I have proved to 
my own satisfaction, in a small way, that reasonable assurance of 
a permanent job is the greatest single weapon against unrest and 
Bolshevism. The moment uncertainty as to tenure of job and rat^s 
of pay is introduced into a workman's mind he is ready to believe 
almost anything, and you cannot expect a worker in any class to 
save for a home when he has no or only small assurance of the 
regularity and permanency of his work. 

William M. Leiserson ^ wrote saying that the tendency of the 
engineers is to consider workers only as individuals, while the ''in- 
dustrial psychologist" holds that they must be dealt with in the 
mass. On the other hand, the economist, who has been studying 
labor problems and industrial relations for more than a hundred 
years, takes a broader view based on a study of the labor problem, 
its history, and its causes and manifestations under different sys- 
tems of industry. 

The economist, he continued, analyzes industrial relations and 
finds not one problem requiring individualistic or collective handling, 
but two sets of problems, one requiring dealing with each worker 
as an individual, the other necessitating dealing with wage-earners 
iis a chiss. The first of these may \)c called the personal relations in 
industry, tlie second, the economic or governmental relations. The 
personal relations include such subjects as recruiting, selecting, hir- 
ing, training, placing and promoting workers, looking after their 
health, safety, comfort and welfare. These problems are indi- 
vidual problems; they are technical questions that must be de- 
cided by technical experts like the engineer or the physician. The 
second set of relations, however, has to do with bargaining, wages, 
hours, shop government and discipline. These (luestions are essen- 
tially controversial. They cannot lx» decided l)y technical experts. 
* Working Conditions Service, U. S. Dept. of Labor, Washington, D. C. 



DISCUSSION 201 

They are matters of opinion requiring decision by a democratic 
majority. 

Failing to catch the economist's fundamental analysis of the 
problem, engineers, employment experts and industrial managers 
are wont to group all industrial relations together and to include 
the fixing of just rates of pay in the managers' function. Accurate 
records of individual production and fairness and square dealing 
with employees, they think, will assure justice to the workers. This 
may be true as far as injustices between individuals within a plant 
are concerned under a general scale of remuneration that is already 
fixed, but it overlooks entirely what President Wilson has called 
the progressive improvement in the condition of the wage earner, 
the raising of the scale of all so that labor may receive a very much 
larger share of the product of industry than it gets today. 

H. L. Gardner ^ wrote outlining in detail the personal quali- 
fications of the employment manager, his duties and the functions 
of the employment department. He said that the manager should 
be a broad-gage man, perferably a high executive or an oflBcer of 
the concern, for which he preferred the title ''service manager." 
He should be of sufficiently large caliber to ''sit in" with the officers 
or a committee responsible for all relations between employer and 
employee; and being thus constantly apprised of general problems 
and accepted poUcies he can then organize and direct his depart- 
ment to coordinate with all other service departments and activities. 

Interviewing is perhaps the keynote of his work, and with the 
right viewpoint and with practical knowledge of plant operations 
and conditions he can be of tremendous influence in building the 
right kind of working force. Continuing, Mr. Gardner wrote: 

"One of the more abstract functions of Employment Manage- 
ment deserves detailed attention. If we are to offer a satisfactory 
substitute for the decreased personal contact in industry, the fore- 
men are the channels through which we must work. To the work- 
men, the foremen are the personification of the manager and the 
company; it seems vital that such representation should harmonize 
with the general policies of the concern, yet too little has been done 
to develop this contact and to insure the desired results. To my 
mind, one of the most important services which employment man- 
agement can render, and one which has a most important influence 

* Manager Personal Relations Sec, E. I. DuPont de Nemours Co., Wil- 
mington, Del. 



202 THE STATUS OF INDUSTRIAL RELATIONS 

on industrial relations; lies in securing real cooperation of foremen 
through understanding of, and sympathy with, employment methods 
of attack on the common problems of personnel and production. 
This may soimd too theoretical, but it is, in fact, a tangible problem, 
and quite possible of surprisingly satisfactory solution. 

*'At the risk of repeating certain thoughts which I emphasized 
at the National Association of Employment Managers Convention 
in Cleveland, I would criticize the average employment department 
as too selfish. We too frequently devise theoretical solutions for 
the existing problems and impose them on the plant without suffi- 
cient sympathy for the other fellow's opinions and troubles. Com- 
plete success demands that the employment department thoroughly 
acquire the general plant viewpoint and merge its individual activi- 
ties into the broader service program." 

Mark M. Jones ^ (written). Functions in industry can be 
arranged according to many very interesting classifications. The 
five M's — money, methods, materials, machines and men — have 
appealed to me as a simple classification. Engineers will recognize 
that in our progress on the fifth M — men, we have from an ad- 
ministrative standpoint reached a point of about 30 in case we 
consider progress on the other four M's — money, methods, ma- 
terials, and machines — as being at 70, with 100 the ideal. 

Employment management as a part of Industrial Relations 
has as its object the administration of recruiting, selecting, placing, 
transferring, promoting, and releasing workers on an engineering 
basis. It seeks to apply the labor policy of the enterprise so far as 
it may affect those functions named. Its effectiveness is only limited 
by the strength of the belief of the management in the value and 
possibilities of such a service. 

Employment management is distinctly a "service" function. 
It is an aid to production and a department for the purpose must 
always be operated with the proper understanding of the import- 
ant part it plays in turning out the finished product. 

A well-organized employment department surrounds the initial 
contact of the worker with those activities which influence him favor- 
ably and contribute definitely toward the final object of production. 
It pro\ades a proper reception place and courteous treatment 
while the applicant is being studied. It studies him from mental, 
moral, physical, financial and social standpoints with the object of 
^ Director of Personnel, Thomas A. Ediaon Industries, Orange, N. J. 




DISCUSSION 203 

placing him where he can work to his own best interests. The employ- 
ment manager knows definitely that an individual cannot work to the 
best interests of the enterprise unless he is working to his own best 
interests. The interests of worker and business are identical. From 
the standpoint of proper placement there is no divergence. From the 
standpoint of effectiveness, however, men must be weighed more 
carefully and a more exact method of so doing remains to be 
developed. 

If during the coming months engineers will have in mind just 
one need of this field, namely, the same recognition and study of 
the fifth M — men — as of the other factors in production, they 
can do much in assisting American industry toward the great ideal 
of industrial nations, which is that of ''increasing individual 
production." 

Mr. Alford's Status of Industrial Relations is a timely sum- 
mary of past and present, full of valuable data and epitomized in 
a warning to be heeded. 

Harrington Emerson (written). I have been requested to 
contribute to the discussion by notes on Wage-Payment Plans as 
one of the problems of modem industry. 

Wages are but a phase of a much deeper problem. On the one 
side are the necessities of life, to live morally, to have health, time 
for study and industrial improvement, and to have opportunity; 
on the other, what the earnings will afford, and again also the market 
rate for similar services. An empirical equiUbrium is struck. Not 
as high wages as an ideal life would require, but often more than 
the earnings justify. 

But what I recognize is that three conditions enter into work 
and wages: (1) Thebasichourly wage, however determined; (2) the 
equivalent in output for the hourly wage; and (3) the special excel- 
lence of the worker. Of these the third, the individual excellence of 
the worker, is the most important. 

The great truth, as yet only partially recognized, is that the 
superior worker is worth so much more than the average or inferior 
worker that any amount of care and supervision and all the extra 
pay required to secure him is a good and pa3dng investment. 

From this conviction it is evident that I am wholly out of 
sympathy with so-called profit-sharing, which repudiates individu- 
ality and makes of business a kind of providence that rains on the 
just and the unjust alike. 



204 THE STATUS OF INDUSTRIAL RELATIONS 

There is no connection between profit and skill and effort. The 
product of the highest skill may be sold at a loss, the product of 
malingerers may be sold at high profit. 

The extra wage paid for individual competence is not a dole, 
a gratuity, a present. It should be a measured and full equitable 
compensation for a measured delivery. 

There are current wages so low as to be dishonest. There are 
also current wages so high as to be dishonest. 

Starting with a basic hourly rate, increased yield should com- 
mand higher pay as long as unit costs drop. An increase in wages 
that increases unit costs (unless money is falling in value) Ls robbing 
the three other divisions of the community. 

The problem always is so to lower costs as to increase output, 
to lower unit costs yet to pay more per hour, to give greater security 
and volume of investment yet to lower interest and dividend rates, 
so to compensate executives as to stimulate them to secure the best 
combinations of men, materials and machines, thus again lowering 
unit costs and adding to the common fund. 

Fundamentals, not expedients, underlie all real solutions of 
the wage problem. 

R. G. A. Phillips* (written). To my mind industrial rela- 
tions problems came into being along with the **Big Business" idea 
and I think it well and timely that we give these various problems 
serious thought and study. 

In our industrial relations work at the works of the American 
Multigraph Company the biggest topic of the day is our industrial 
democracy. Therefore, although it comes last in point of things 
done, it ranks first in order of importance. 

In our case Industrial Democracy was no ''spur-of-the-moment 
idea." It is a subject we have been studying almost as long as we 
have been in business. Our active interest dates back about five 
years — we have been all that time progressing toward our goal. 
It took about a year and a half of constant study before the final 
plan with its constitution became a fact. It took definite shape 
toward the end of February 1919, and finally on March 1 the plan 
was put up to the employees. 

Since then wo have had many incidents that have made us 
wish we had started sooner. For instance, I will refer to some of 
the minutes of the congress meetings and pick out suggestions that 
wore approved by the senate as being wise and necessary. 
^ Vice-President, The American Multigraph Co., Qcveland, Ohio. 



DISCUSSION 205 

Educational Committee: Suggestion that certain employees 
be taught what the Multigraph does so they might do 
their work with greater understanding. (Assemblers and 
final inspectors). 

Produdion Control Committee: Suggested rearrangements of 
departments that will produce greater results. 

Suggestion Committee: Presented about fifty new ideas, the 
principal one of which will result in considerable saving 
in handling of tools, etc. 

Sales Cooperation Committee: Working all the time with Sales 
Department. Pushing manufacturing and keeping up 
standards. Latest job the reduction of time taken to 
fill and ship foreign orders about 80 per cent. 

Sanitation and Safety Committee: Always at work. Producing 
great results. Total suggestions adopted to date, about 
65. 

Recreation Committee: Managed several dances, an indoor 
baseball league, bowling league and all gymnasium classes. 

Spoiled-Work Committee: Very active. Reported last week 
on correcting a condition that reduced scrap and increased 
production on a certain part about 150 per cent. 

Shop Training: Managed all class work during season just 

finished. Promoted mathematics classes (two each week 

after hours) and a big general shop eflBciency class that 

met every Friday night having an average attendance 

of 150. 

Industry has got to begin to get all the effort it is buying. It 

should no longer be satisfied with the services of the hands of its 

workers — the brains, too, must be induced to participate. This 

matter of brain and hand cannot be commanded either — it must 

be more of a pull-together effort. We think we are headed that 

way through our Industrial Democracy. 

C. B. AuEL (written). It seems almost certain that the six 
lines of development listed by Mr. Alford will go a long way, if 
fairly universally adopted, toward lessening industrial unrest; but, 
they will hardly eliminate it since the fundamental cause for this 
unrest as admitted by the author is fear of unemployment, and 
methods of overcoming it have not been included. Some persons 
may point out in opposition to this statement that during the recent 
period of tremendous industrial activity, labor imrest was perhaps 



206 THE STATUS OP INDUSTRIAL RELATIONS 

at a maximum; but, in answer to this it may be said this unrest was 
quite abnormal, due very largely to war conditions, and was more- 
over aggravated by employers practically bidding against one 
another in the labor market, with the very natural consequence 
that labor tended to oscillate to and fro, wherever wages were high- 
est, or in the contrary event to insist on wages being brought to the 
high level. 

One very important item the author has omitted from his ILst 
is ^^Americanization,*' which has been carried on by many corpora- 
tions for a number of years past, principally through the medium 
of schools for their employees; but, fine as is the work already done 
by them, the task is so huge that it needs to be supplemented by 
greater efforts on the part of the various states if real headway is 
to be made. In Pennsylvania, for example, one authority has an- 
nounced that with a population of 8,000,000, there are 1,500,000 
foreigners over the age of 10 years and of these half a million cannot 
read or write English, while a third of a million cannot read or write 
any language, but what is worse than these statements is the further 
fact that the half-miUion that cannot read or write English increased 
to this figure from a quarter-million in the short space of 10 years. 
Illiteracy is a fertile field for the propagation of every kind of **Lsm" 
and it seems astonishing that a condition like this should exist in 
any state in our Union, and doubtless a similar situation exists in 
certain other states. 

D WIGHT T. Farnham (written). At the close of the paper 
Mr. Alford opens a question whose just solution will, I believe, 
within the next ten years tax to the utmost the ability and resource- 
fulness of our financial, economic, ethical and engineering minds. 
The fact that the International Harvester Company had hardly 
installed their "industrial democracy" plan before they were del- 
uged with recommendations for rate raises illustrates this trend. 
When their committeemen, however, were taken to the company's 
l)ooks and shown that the business could not continue if their de- 
mands were granted, unreasonable demands for the most part 
ceased. From the standpoint of human relations, then, the first 
general principle necessary in order to establish a sincere industrial 
democracy would seem to be the possession of profits which will 
survive the light of public opinion. 

The fair division of the rewards is the question which we face, 
and it will call for the development of the general principles under 



DISCUSSION 2b7 

which controversies may be adjudicated. The first step is perhaps 
the establishment of a fair retmn upon capital which is so invested 
as to be reasonably secure. The next step in logical order to insure 
stability would then be the determination of adequate sinking funds 
to make reasonably certain steady dividends from each sort of busi- 
ness. Setting up such reserves would take investment in industrials 
out of the gambling class and would make them attractive at a 6 or 
7 per cent return instead of having to offer chances of 15 to 20 per 
cent in order to sell stock. 

One of the rocks upon which profit sharing has in the past gone 
to pieces has been its call upon the workers to share in the sacrifice 
during dividendless periods. The other has been the determination 
of the share to which labor is entitled. The solution of such ques- 
tions requires first the scientific analysis which the engineering 
mind is best qualified to make. 

What Mr. Alford describes as the second great tendency in 
the development of industrial relations — the willingness to con- 
sider the workers in groups — I interpret as not so much a plea for 
an imderstanding of mass psychology as a plea for that knowledge 
which brought the ancient Greeks to the belief that he who imder- 
stood l^e was all-powerful. Every age has had its interpreter of 
motives which actuate humanity. Today the human psychologist 
is an industrial psychologist. We applied the mind of the engineer 
first to materials, then to machines. Now we have reached labor. 

C. E. Knoeppel (written). Through organization and spe- 
cialization in industry, relationships have each year become more 
and more complex, so that today this great problem of human 
contact is by far the most important confronting us, and my pre- 
diction is that from now on Industrial Relations will be considered 
the keystone of the industrial structure. 

I am, therefore, pleased indeed to see that The American Society 
of Mechanical Engineers is giving this matter of Industrial Rela- 
tions the prominent place it occupies. If the engineering world does 
not give the subject attention, where will the initiative come from? 
From the workers? No. From the employers? No. I say "no'' 
in both cases advisedly, because in the last analysis each side is 
generally suspicious of the other's motives when suggesting im- 
provements having to do with human relationships. 

The world-wide tendency toward socialism and revolution in 
ideas, if not in acts, is due not so much to desire for political changes 



208 THE STATUS OP INDUSTRIAL RELATIONS 

as to a demand for economic changes. What the great masses of 
people want are homes, food, farms, clothes, jobs, wages which 
balance cost of living, participation, representation in affairs, oppor- 
tunity for self-expression and development. 

The gigantic convulsion the world is now going through seems 
to me to be a protest against the way modern society is organized, 
against much in the present plan of man-to-man contact. The 
purpose of the coming era, as I see it, is to find the right basis, and 
this the people the world over will do, regardless of the strife and 
bloodshed necessary to its accomplishment. 

John Younger (written). I would heartily endorse the 
thought expressed in Par. 63 and 64 — that the workers should be 
considered in groups; and while it is true that the tendency is for 
their operations to be sub-divided to a point of intense specializa- 
tion, yet a process similar to that of establishing workshop limits 
can be used and limits placed on' such specialization so that the 
men inside these particular limits fall into groups. 

The trade unions of England feel that the problem of employ- 
ment is not solely for the employer to solve, but that they also 
have a share in the proper solution. With this object in view, Ruskin 
Hall was founded many years ago at Oxford, being practically a 
branch of Oxford University but reserved exclusively for students 
of all ages sent by the different trade unions. 

This college has specialized on economic and sociological prob- 
lems and questions of employment, capital and labor. These are 
studied, of course, as disinterestedly as possible with a slight natural 
bias towards the viewpoint of labor. 

It is my belief that the training of men in this work is distinctly 
beneficial and cannot help but give more intelligent cooperation in 
the human problems of manufacturing interests, and I would sug- 
gest that the workings of Ruskin College and its results be studied 
very carefully by capitalists and laborers. 



No. 1694 

CENTRAL-STATION HEATING IN DETROIT 

By J. H. Walker,* Detroit, Mich. 
Non-Member 

This paper discusses the general problem of the utUtzaiion of the heat ordinarily 
discharged to the condensing water in a central electric generating station. The 
impossibiliiy of its complete utUizalion for the purpose of heating buildings and the 
difficulties in the way of even its partial utilization are pointed out, with particular 
reference to conditions existing in Detroit. 

The development of the central heating system of The Detroit Edison Company 
is traced, showing how the use of exhaust steam for heating was abandoned in favor 
of live steam. The reasons why it is more commercially expedient under the existing 
local conditions to supply live steam to the healing system and to generate all electric 
current in the condensing stations are also fuUy brought out. 

The latter part of the paper describes some interesting features of the central 
heating system in Detroit, such as the boiler plants, distributing system, under- 
ground pipe and tunnel construction, consumers^ installations and meters. Special 
mention is also made of distribution losses, condensation return lines, and the 
method of transmitting steam through feeders at high velocities and with large 
pressure drops. 

The paper concludes with a discussion of the advantages of central heating 
service and of the obstacles to its under use and points out the possibility of 
operating individual plants in combination with the central plant. 

f\NE of the natural results of the groupmg together of human 
^^ beings m civiUzed communities is the existence, in our cities, 
of central plants for the generation and distribution of heat to sur- 
rounding buildings. The advantages of central heating service to 
the user, over the alternative of operating a heating plant in his own 
building, are comparable to those accompanying any other public 
service. To the community also a properly operated central heating 
plant is a distinct asset, commercially and economically. 

2 Started in a limited way about forty years ago, the central 
heating industry has grown steadily, though not with the rapidity 
of some other utiUties, until at the present time there are between 
two and three hundred enterprises operating as public utiUties in 
cities of all sizes in most of the northern states and doing an annual 

* The Detroit Edison Company. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Societt of Mechanical Engineers. 

209 



210 CENTRAL-STATION HEATING IN DETROIT 

gross business estimated at from ten to fifteen million dollars. The 
capital invested is perhaps thirty to forty million dollars. As media 
for distributing the heat, both hot water and steam are used, but 
because of the impracticability of metering hot water service and 
because of the better adaptability of steam to the requirements 
of the average building the trend of progress is toward the latter. 
Further development of the business now hinges mainly upon the 
possibility of the establishment of the proper relationship between 
the selling price and the cost of the service so as to insure adequate 
return upon invested capital. In many instances this relationship 
is not satisfactory at present, chiefly because the actual cost and 
the value of the service are not fully appreciated. The engineering 
practicabiUty has been amply demonstrated; and that there is a 
pronoimced economic demand for the service is beyond question. 

8 The central heating business is closely allied with the elec- 
tricity supply business and in most cities is carried on directly or 
indirectly by the electricity supply companies. The task of supply- 
ing the demand for heating service falls quite naturally to the 
electric company because of the partial similarity in the methods of 
production and distribution of the two commodities; and furthermore 
the ability to offer to its prospective customers both electric and 
heating service, with the consequent entire elimination of any sort 
of power plant from the customer's building, is of great advantage 
to an electric company doing business in a large city. 

4 Other reasons for the combining of the two utilities are the 
economies in the consumption of fuel, and to some extent in the 
investment costs, which are sometimes made possible by the physical 
combination of the electric and heating plants. 

THE UTILIZATION OP EXHAUST HEAT 

5 The heat carried away by the condensing water in the central 
electric stations of the United States, equivalent to about 60 per 
cent of the total fuel burned by them, is one of the more obvious 
(although not by any means the greatest) sources of waste of the 
country's fuel resources. Like many other similar losses this one 
exists, not because its reduction is theoretically impossible but be- 
cause it is seldom conunercially practicable. But while commercial 
considerations have always dictated certain practices in the utiliza- 
tion of fuel in the past and will continue to do so in the future, the 
increasing cost of coal and the present impulse towards its conserva- 
tion now direct attention to some of the fundamental problems. 



J. H. WAUEBB 211 

6 This great quantity of heat, rejected at low temperature from 
the generating units, may be considered as a by-product of electricity 
supply and as such it£ rate of production will depend upon the 
rate of production of electricity, the primary product. Since neither 
electrical energy nor heat can be stored to any great extent, it is 
necessary for the complete recovery of the by-product heat that the 
demand for it be equivalent, hour by hour, and day by day, to the 
rate of electricity supply. The warming of the interior of buildings 
is a natural means of making use of this heat, but the great 



'O 5 K)IS20 25303S«I«S0 
meVsofRieYMf 

Fta. 1 Load Cubvxb or Electric and SrEAu-HEATiNa Plants 

diversity in the rates of use of the two commodities renders impossible 
even an approximately complete utilization of the by-product heat. 
The lack of agreement in the rates of use of electricity and heat from 
week to week throughout the year is illustrated in Fig. 1, in which the 
1918 load curves for electrical and heating service in Detroit are 
plotted to a percentage scale with the maximum point on each curve 
taken as 100 per cent. 

7 The possibility of establishing a better relation between the 
rates of use is rather slight. The rate of heat supply is largely 
fixed by unalterable climatic conditions. The use of electricity for 
lighting is also governed by natural elements. The demand for 
electricity for industrial use, which now constitutes a major fraction 



212 CENTRAL-STATION HEATING IN DETROIT 

of the output of most central stations, is not governed by these 
factors and is the only kind of load whose characteristics could 
conceivably be adjusted to suit the requirements for exhaust heat, 
but even this could probably not be done to any practicable extent. 

8 Another important obstacle to the full use of this by-product 
heat through the warming of buildings is due to the great develop- 
ment of the central electric stations which in many industrial centers 
in the United States have so increased in size that the amount of 
exhaust heat which would be available as a by-product is greatly in 
excess of that which it would be commercially feasible to distri!)ut^ 
for the heating of buildings. For example, in 1918 the central 
electric generating stations in Detroit pro<luced approximately 
774,000,000 kw-hr. of electricity. The exhaust steam which would 
have been available if discharged at pressures above atmosphere 
and which could have been utilized for heating, considering the 
winter months only, would have amounted to over 9,000,000,000 lb. 
This is five times the quantity actually distributed in the existing 
central heating system of the city, which entirely covers the only 
portion of the city in which the heating load is sufficiently dense to 
render the laying of distribution mains commercially justifiable. 
For because of the great investment cost^s the distribution of heat 
by the medium of steam is feasible only in the districts of relatively 
great density of load, which, in most of our cities, comprise only the 
business district and the very l)est residence districts. Nor are the 
economies to be gained by using by-product heat sufficient to enlarge 
this area appreciably. The central heating business in the average 
American city could keep pace with the electricity supply business 
only if the density of population were far greater than is compatible 
with present standards of living. 

9 Thus it is apparent that the exhaust heat from central 
electric stations can at best l)e utilized for heating only during the 
times when heat is required and then only in so far as it can be 
commerciallv (listril)uted. But there are certain obstacles which 
stand in the way of its utilization even to this extent — obstacles 
which arise primarily from the difficulty of transmitting steam over 
long distances. 

SYSTEMS OF CENTRAL HEATING 

10 Assuming that a central heating load exists and is to be 
supplied by the electric company, there are in general three methods 
by which this can be done: 




J. H. WALKER 213 

a The heating load can be served from a condensing generat- 
ing station so designed that steam is available for heating 
at pressures above atmospheric after partial expansion 
in the electric generating units, the remainder of the 
steam used for current generation being fully expanded 
and condensed at high vacuimi 

b Separate heating plants may be built in locations near 
the heating load and equipped with non-condensing 
generating units which will generate current only to the 
extent of the requirements for exhaust steam for heating, 
the remainder of the electricity being produced in a 
condensing station 

c The heating system may be supplied entirely with live 
steam from boiler plants located near the center of the 
heating load. 

11 It may so happen that the natural location for the main 
generating station serving a city is near the heating load, and if this 
is the case, the first method is preferable. In such a plant the use 
of bleeder turbines, designed so that steam can be extracted from the 
intermediate stages after partial expansion, offers great advantages. 
This arrangement has been successfully carried out in some instances 
and is probably the nearest possible approach to ideal conditions, 
since the duplication of equipment is reduced to a mininmm. 

12 Often, however, a consideration of land values or of railroad 
connections requires that the main condensing station be located at 
such a distance from the heating load as to preclude the possibility 
of transmitting steam from it. This is the case in Detroit as will be 
seen from Fig. 2. The Delray plant and the Connors Creek plant, 
the two main generating stations operated by The Detroit Edison 
Company, are respectively 3^ and 4^ miles from the heating district. 
A condensing plant, located on high-priced land near the heating 
district and with inconvenient railroad connections, or none, would 
be necessary if this first method were to be used. To such a plant, 
built for electricity supply and consequently, for reasons previously 
stated, burning more coal than a plant built solely for heating, this 
matter of proper railroad faciUties is particularly important. Here 
again enters the matter of the transmission of steam, for seldom 
could the bleeder-turbine plant be located as favorably with reference 
to the heating load as could Uve steam plants, and the additional 
investment in transmission lines must therefore be charged against 
this plan. 



214 



CENTRAL-STATION HBATINQ IN DETROIT 



13 The size of the pipes required to transmit a given quantity 
of steam over a given distance decreases as the density of the steam 
and the amount of pressure drop along the pipe increase. The 
most economical method from the standpoint of investment costs 
would be to extract steam from the high-pressure stages of the 
turbine; but the amount of electricity which could be generated 
per pound of steam would then be reduced. The relative values of 
these factors for an assumed river-front bleeder-turbine plant in 
Detroit are illustrated by the curves m Fig. 3. Curve A shows the 
total credit for the saving in coal and boiler capacity which could be 
allowed such a bleeder-turbine plant located on the river front and 




Fig. 2 Map SHOwma Location of Detroit's Heating District and 

Electric Generating Stations 



serving the existing Detroit heating system. Curve B shows the 
total of the additional investment charges due to the higher-priced 
land on which the plant would be built, plus the investment charges 
and line losses involved in transmitting the steam from this river- 
front plant to the centers of the heating load. Even at higher 
extraction pressures the bleeder plant would not be justified under 
Detroit conditions Ixjcause of its unfavorable location with respect 
to the heating load. 

14 The second plan above mentioned, namely, the building of 
plants located near the heating load and generating current only to 
the extent of the exliaust-st<*am requirements, involves somewhat 



J. H. WALKEK 215 

difTereat factoni. In this case the capacity of the generating units 
will be detennined by the heating requirements and the electricity 
generated by them will be the by-product. An essential requirement 
for the success of this plan is that the relation of these generating 
unita to the electrical system as a whole be such as to allow their 
loads to be adjusted without restriction according to the momentary 
requirements for exhaust steam. 

15 The value of such current as the heating plant will produce 
will be detennined tai^ly by the cost of producing an equal amount 



Pressure of Exfracfed&team-Lb.&j^ 



in the condensing stations. If the condensing stations do not 
produce current at an extremely low cost and, if other conditions 
are not unfavorable, this plan may prove very attractive and is in 
fact the most widely used method of combining the electrical and 
steam-heating plant. But unfavorable conditions surrounding the 
production of this relatively small amount of current may render 
its cost unattractive. The size of such generating units as might be 
installed in the heating plant may be so insignificaDt compared to 
the large size of the units in the condensing plant that the investment 



216 CENTRAL-STATION HEATING IN DETROIT 

in the main station is not measurably reduced and the smaller units 
are therefore additional investment. The investment in steam- 
distribution mains is also much greater than when steam at higher 
pressures is used. 

16 Thus it may in some cases prove in the end more commer- 
cially expedient to omit entirely the generation of current in the 
heating plant. 

17 In Detroit many of these conditions conspire to make the 
use of exhaust steam for heating unattractive. Generating units 
exhausting into the heating mains were operated in one of the heating 
plants for several years but this practice has since been definitely 
abandoned and the heating system is now being supplied with live 
steam. Current generation is limited to the output of small house- 
service turbo-generators whose exhaust is utilized to heat the feed- 
water. The history of the central heating industry in Detroit is 
one of steady progress toward this method of live-steam operation. 

DEVELOPMENT OF CENTRAL HEATING IN DETROIT 

18 The immediate motive which, in 1903, led to the establish- 
ment of the central heating industry in Detroit was the possibility 
of obtaining a high thermal efficiency in the generation of electricity 
through the utilization of the exhaust steam. The generating plant 
from which steam was first suppUed for heating had previously 
dfscharged its exhaust to the atmosphere. Owing to the building, 
at this time, by The Detroit Edison Company, of a large condensing 
plant, the smaller plant, with several others, would have been shut 
down, but the possibility of selling the exhaust steam made it appear 
desirable to continue the generation of current there to the extent 
of the exhaust steam requirements. Also, since the plant was located 
in a district served with direct current, the loss involved in converting 
an equivalent amount of alternating current received from the main 
plant to direct current would be saved for such current as might be 
generated there. The plant was in a high-class residence district 
and tlie heating service proved very popular, but the actual overall 
economy of the plant was not as great as had been anticipated. 

19 A year later the construction of a central heating plant was 
begun in the business district of the city. This second project was 
undertaken, not a^j a means of disposing of exhaust steam, but for 
the express purt)o8e of supplying the demand for central heating 
service among the owners of downtown buildings, who were con- 



J. H. WALKER 217 

sideling shutting down their plants and purchasing electric servdce. 
The downtown plant was built primarily as a heating plant, and 
thou^ provision was made for electric generating units they were 
never installed and the heating mains were supphed with steam from 
the boilers through reducing valves. 

20 With the development of the large condensing generating 
stations the generation of current in the heating plants grew less 
and less attractive and when a third heating plant became necessar>% 
because of increasing demand for steam heat, no provision was made 
for electric generators. The same practice was followed when, in 
1916, the original exhaust-st.eam plant was rebuilt; and when the 
steam-distribution sj'stem of another compwmy which had been 
engaged in the generation of electricity and the distribution of 
exhaust steam in the business district was purchased by The Detroit 
Edison Company, the new plant which was built to supply this 
district was also designed as a heating plant only. The entire 
combined distribution system is now being supplied with live steam 
from the boilers through reducing valves. 

21 Since there are no generating units exhausting into the 
heating mains the pressure carried in them is not limited by considera- 
tions of back pressiue. On those sections of the system formerly 
supphed with exhaust steam at from 2 to 5 lb. pressure, the pressure 
now maintained ranges from 5 to 15 lb. and is being increased from 
year to year, toward the upper limit permitted by the strength of 
the pipes. The pressure on the section originally designed as a Uve- 
steam system is about 30 lb. Because of the increased capacity of 
the distribution system at higher pressures, due to the greater density 
of the steam and the greater allowable pressure drop, these higher 
distribution pressures are desirable. 

22 The steam is dehvered to the network of mains through 
connections made at the plants and also through feeders radiating 
from the plants and delivering steam to certain '' feeding points " 
in the distribution network, the method being similar to the feeder 
and main method employed in electricity distribution systems. 



REASONS FOR USING LIVE STEAM IN DETROIT 

23 The supplying of hve steam to the heating system and the 
abandonment of the generation of current in the heating plants 
is commercially justifiable, even though such current would be 
generated at a high thermal efficiency. The underlying reason 



il^lB- ' ' 



218 CENTRAL-STATION HEATING IN DETROIT 

for this is that there exist certain unfavorable conditions which 
outweigh the thermal advantage and make the total cost of such 
current higher than the generating cost at the large and eflScient 
main generating stations. 

24 In considering the cost of generating current, when the heat 
in the exhaust is recovered, it should be borne in mind at the outset 
that the amount of coal consumed, though small, is not by any 
means negligible as compared to that consumed in a condensing sta- 
tion. For each kilowatt-hour so generated there would be extracted 
from the steam, if the conversion were 100 per cent eflScient, its 
heat equivalent, 3415 B.t.u. Taking into account mechanical and 
electrical losses in the generating unit and the losses involved in 
generating steam from the coal, the actual number of heat units 
devoted to the generation of electricity is not less than 6700 B.t.u. per 
kw-hr. This is 27 per cent of the corresponding figure for the 
Connors Creek plant (a condensing station) which in 1918 generated 
its total output at an average of 20,900 B.t.u. in the coal per kw-hr. 
of output. So that at the outset, the additional fuel which would 
have to te burned in the heating plants if current were generated, 
would be over one-quarter of the amount required to produce the 
equivalent amount of current at the Connors Creek plant. 

25 One of the principal elements of cost which miUtates against 
current generation in the heating plants is the attendance labor, 
which, because of the low load factor and small size of any generating 
units which might be installed there, is much higher per kw-hr. gen- 
erated than in the large stations where the size of the units and 
the load factor are much greater. 

26 But the really deciding factors are the investment charges. 
The change of policy involving the final abandonment of exhaust-steam 
heating was made in 1916, with the rebuilding of the Willis Avenue 
heating plant, the original exhaust-steam plant. At this time the 
electrical load in Detroit was increasing rapidly and several new 
turbo-generator units were being purchased. At the Willis Avenue 
heating plant the existing and future exhaust-steam requirements 
would have called for a generating unit of about 2000 kw. capacity. 
The unit purchased at that time for the Connors Creek plant was 
the 45,000-kw. machine which has since been put into service. 
Would the existence of small generating units in the heating plants, 
aggregating altogether possibly 4000 kw., actually reduce the number 
of machines in the Connors Creek plant at that time or at any future 
time, if the latter were to increase in steps of this magnitude? 



J. H. VALKEB 219 

Regardless of theoretical cousideratioos, as a matter of fact it actually 
would not have done so and it became clearly evident, therefore, 
that any investment in generating units in the heating plants must 
be reckoned as additumal investment and the cost of_]any electricity 
generated by them must include the fixed charges on that investment 
and could not be credited with having saved any investment else- 
where. The relatively insignificant proportion of the total system 



Fia. 4 Load Cubves Showinq Possible £k.ECnuc Output of Heating 
Pi-iNT AS Compared with Output of Main Station 

load which could be borne by a unit in the Willis Avenue plant is 
strikingly illustrated in Fig. 4. 

27 The small size of the unit and its low load factor would make 
these investment charges relatively high per kilowatt-hour generated. 
Furthermore, while the chief advantage of the heating-plant generat- 
ing units would lie in their ability to avoid conversion losses by 
generating direct current required for the downtown district, this 
advant^e coidd not be completely made use of because of the lack 
of coordination between the heating and electrical loads of that 
district, which would make it necessary to convert some of the 



220 CENTRAL-STATION HEATING IN DETROIT 

direct current back to alternating current for transmission to some 
other district or else would reduce the load factor of the unit. 

28 For a quantitative economic study of the subject the Willis 
Avenue heating plant offers a favorable example, for it had been 
operated as a generating plant and the distribution mains had been 
designed for exhaust steam pressures. Based on the performance 
of the old engine-driven units, there would have been generated 
by the 2000-kw. turbo-generator which it was proposed to install 
in the new plant about 6,000,000 kw-hr. per year. 

29 In accordance with the considerations which have been men- 
tioned, the proposed turbo-generator installation should be charged 
with its operating and maintenance costs including labor, supplies, 
and the fuel equivalent of the energy produced. It should also be 
charged with the fixed charges on the unit itself and on the building 
space occupied. Based on the fairly stable pre-war prices existing 
in 1916, these items might be conservatively estimated as follows, 
neglecting the disadvantage due to the lack of coordination between 
steam and electrical loads. 

Charges Against Generating Unit — Operation and Maintenance: 

Labor — 3 oilers $ 3,360 

Supplies, etc 200 

Fuel cost — 1221 tons at $2.66 3,247 

Maintenance 200 

$7,007 
Fixed Charges (Additional Items Only) : 

2000-kw. turbo-generator at $15 per kw $30,000 

Installation cost, wiring and piping 4,000 

Building and foundations 10,000 

$44,000 

Depreciation at 4 per cent ] ,760 

lleturn on investment at 6 J per cent 2,860 

Total charge against turbine $11,627 

30 The turbo-generator nmst of course be credited with the cost 
of generating an equivalent amount of energy at the main plant, 
of transmitting it to the heating-plant district, and of converting 
it to direct current, since these costs would be incurred if the turbo- 
g(»nerator were not installed. 

31 The production cost of electric current generated by a 
central station consists of two parts, the demand or ''readiness to 
serve'' component and the ^'energy'* component. The former may 
be defined as that portion of the total production cost which would 
be incurred if no electricity were actually delivered, but if the plant 



J. H. WALKER 221 

were merely held in readiness to deliver the loads actually sustained 
— held, in other words, with steam pressure up, turbines and auxiliaries 
in motion, a sufficient number of boilers banked and the operating 
crew on duty. The energy component may be considered as that 
part of the total cost which is directly proportional to the amount 
of energy dehvered. Although the exact separation of these com- 
ponents is impossible, the distinction between them is none the less 
real and is generally recognized. 

32 In the present case the heating-plant generating unit can be 
credited only with energy component, since it is not to be considered 
as having reduced the size or number of units in the main generating 
plant. The energy component may fairly be considered as including 
75 per cent of the fuel cost, 50 per cent of the cost of maintenance of 
plant equipment and that part of the labor, such as coal handling, 
which may be considered as depending upon the amount of electricity 
generated, and amounting in this case to 9 per cent. 

33 The actual combined efficiency of transmission to the heating- 
plant district and of conversion to direct current is approximately 
80 per cent, so that to deliver 6,000,000 kw-hr. of direct current to 
the heating-plant district there would be generated at Connors 
Creek 7,500,000 kw-hr. 

34 The credit which can be allowed to the heating plant for the 
current which it would generate would then be as follows: 

Fuel — 4112 tons at $2.08 $9,786 

Wages 285 

Maintenance 617 

Total credit allowable $10,688 

35 This credit of $10,688 compared with the larger figure of 
$11,627 which is to be charged against the heating plant unit, thus 
indicating that the generating of current in the heating plant would 
not be justified under pre-war price conditions. 

36 The foregoing study is based only upon conditions in 
the plant itself and no mention has been made of the effect upon 
the distribution system when exhaust steam is distributed. It is 
common practice in exhaust-steam systems to carry a back pressure 
of from 2 to 15 lb. At these pressures the specific volume of the 
steam is high and the allowable pressure gradient throughout the 
system is very limited, making it necessary to install much larger 
pipes than is the case when live steam at high pressures is used, and 
consequently increasing the investment. While it is true that 



222 CENTRAL-STATION HEATING IN DETROIT 

turbines have been built for exhaust pressures up to 30 lb., the 
electrical energy which can be extracted per pound of steam under 
such conditions is reduced and the cost per kilowatt of turbine 
capacity is increased; furthermore, in the method of feeder operation 
which is actually employed and which has proved of inestimable 
value, the pressure of delivery to the feeders is much higher than 
this. 

37 The method of live-steam operation was adopted in Detroit 
before the recent great advance in the price of coal. The present 
high price of coal makes exhaust-steam operation appear somewhat 
more favorable and it is of course conceivable that at some future 
date a very high coal price may compel a change of poUcy. But 
with the cost of underground lines also increasing, the saving in 
distribution investment will probably continue to be sufficient to 
justify a continuation of the present methods. 

38 Most of the foregoing facts apply only in cases where the 
generating capacity in the heating plant is negligible in comparison 
with the main generating stations. If this is not the case, if the 
discrepancy in size is not great so that investment in the main plants 
is actually saved, or if the additional current required from the 
condensing station is actually not produced at a relatively low cost, 
then the situation may be entirely changed, and there are many 
instances of this in the United States. The consideration of the 
steam-transmission investment, when low-pressure steam is used is 
often controlling, however, and is being increasingly well recognized. 

DISTRIBUTION SYSTEM 

39 The popularity of the heating service in Detroit has led to 
its development on an extensive scale. The present distribution 
system covers an area about 2 miles long and half a mile wide 
which includes the entire central shopping, business, and financial 
districts and a small portion of the residence district. About 
2,700,000 sq. ft. of radiation, Ix^sides numerous water heaters and 
cooking fixtures, are served. The distribution sj'stem contains 
about 20 miles of underground mains and 2 miles of tunnels. The 
four boiler plants which supply steam to the system contain 17,470 
rated boiler horsepower, and they delivered in 1918 nearly two billion 
pounds of steam to the system. Over 1700 consumers are served. 

40 The distribution mains and the buildings served are shown 
in Fig. 5. Though originally built in three distinct sections, the 



Fio. 5 Map of Centkal HEAnNo Srarmi in Detroit 



224 CENTRAL-STATION HEATING IN DETROIT 

system is now a practically continuous network, and the plants are 
so much interconnected that the load can readily be shifted from 
one to another. In the spring and early autumn two of the four 
plants serve the entire area. 

BOILER PLANTS 

41 The four boiler plants which supply steam to the heating 
system are shown in Fig. 6. They are equipped with water- 
tube boilers and underfeed stokers and are of modem design 
throughout. Their location in the central district of the city 
necessarily restricts the amount of land which they may occupy 
and demands a suitable type of architecture, absolutely smokeless 
operation, and extreme cleanliness in the handling of coal and ashes'. 

42 A cross-section of the Congress Street plant, the newest of 
the four, is shown in Fig 7. It is dpsigned to contain eventually 
four l300-hp. boilers and two 2600-hp. boilers of the "W" type. In 
the effort to reduce the amount of attendance lalx)r at this plant the 
auxiliaries are located, for the most part, on the boiler-room floor 
so as to 1x5 within convenient reach of the few men constituting 
the operating crew. Coal is hauled from bunkers at the railroad 
sidings to the plant in drop-bottom buckets of 5 tons capacity, 
which are lifted by a crane and emptied into overhead hoppers 
from which the coal is distributed by belt conveyors to the boiler 
bunkers. Ash-handling equipment is practically nil, the boilers being 
set at a sufficient elevation to allow wagons to be driven beneath the 
hoppers. 

43 Because of the fact that only a relatively small amount of 
condensation is returned to the plants, careful treatment of the 
raw water is necessary. The feedwater flows through live-steam 
purifiers, ofx^rating at boiler pressure, in which the scale-forming 
materials are precipitated. In addition, sodium carbonate is fed in 
automatically graduated amounts to reduce the slight amount of 
hard scale-forming material which finds its way into the boiler. 
Although Detroit water is not a bad boiler water, these precautions 
are necessary because of the large percentage of make-up water 
and the rather high rates of steaming at which the boilers are 
s()inetiin(\s drivc^n. 

44 Tlu^ auxiliaries are almost entirely motor driven. The 
current is supplicMl by a ToO-kw. turbo-generator unit exhausting 
into an op(Mi fi^dwater heater. The load carried on this generator b 



CoDpcn Street Plant Puk PUm PUnt 

Fia 6 HcATiNa Plants of Thb Detroit Edison Coupant 



226 CENTRAL-STATION HEATING IN DETROIT 

adjusted according to the requirements for exhaust steam for heating 
the feedwater and any excess current generated is delivered to the 
outside electric-distribution system. Conversely, if the electricity 
requirements of the plant exceed the output of the generator, current 
is drawn from the outside supply. Exhaust steam is thus made 
available for heating the feedwater and the advantages of motor- 
driven auxiliaries are also secured. Moreover the turbo-generator 
constitutes a source of electricity supply for the plant in case of 
local interruption of the outside service. 

45 Steam is generated at a pressure of 130 lb. gage. This 
pressure is chosen in order to provide for considerable pressure 
drops in the outgoing feeders as calculated for present conditions. 
It may be raised to 160 lb. at some future date. The outgoing steam 
lines leave the plant through a tunnel shaft. 

FEEDERS 

46 Because of the great increase in connected load the transmis- 
sion capacity of the distribution network, most of which was installed 
several years ago, is now quite inadequate. To have raised the 
pressure throughout the sjrstem would have increased its capacity; 
but the S3rstem pressure was permanently limited by the fact that 
most of the underground fittings are of a low-pressure pattern, and 
temporarily by the fact that in those sections of the system formerly 
suppUed with exhaust steam the customers' installations are not 
provided with reducing valves. Instead of attempting to change 
these conditions, which would have involved the reconstruction of 
much of the distribution network, the less expensive plan was adopted 
of running feeders from the plants to various centers of distribution. 
In selecting the pipe sizes for feeders, advantage is taken of the 
difference between boiler pressure and distribution pressure, and 
the size of the pipe is so chosen that at times of maximum load much 
or even all of this pressure drop will take place in the pipe itself. 
The diameter and the cost of the pipe line are thus materially reduced. 
In operation, the pressure of the steam delivered to the feeder is 
raised or lowered as required by the adjustment of a reducing valve 
at the plant, in order to maintain a constant pressure at the center 
of distribution which the feeder supplies. The pressure existing 
at this center of distribution l<? recorde<l at the plant by a long-distance 
gage, electrically operated. 

47 The velocity of the steam flow in the feeders at times of heavy 



J. H. WALEEB 227 

load is very high. Velocities as great as 75,000 ft. per min. have 
been measured. This high velocity does not appear to be at all 
objectionable, however, there being no apparent erosion of the 
pipe and no hammer, or objectionable vibration. The fact that 
the steam is in a superheated state because of the pressure drop is an 



Fio. 7 CROSS-SscnoN of the Conoress Strxbt Plant 

advantage in these respects. Feeders are constructed with long radius 
bends and where the connection is made to the distribution mains 
the diameter of the pipe is gradually increased by special taper fittings 
BO as to reconvert a portion of the velocity head to static pressure. 
48 A considerable pressure gradient is also allowed to take 
place in the distribution mains as well as in the feeders; but this is 



228 CENTRAL-STATION HEATING IN DETROIT 

relatively small, since the upper limit is fixed by the allowable work- 
ing pressure of the older mains. The more recently laid mains 
are capable of withstanding a working pressure of 125 lb. and the 
gradient in the mains can therefore be increased at some future 
date. From a standpoint of safety, however, the desirability of 
carrj-ing pressures in excess of alxiut 50 lb. on tlie service connections 
to buildings is questionable until further development in pres-^^ure- 
reducing apparatus is made. 

49 This method of steam distribution is obviously an adaptation 
to pre\'iously existing conditions and would doubtless U* mollified 
if an entirely new system were being laid out. 

UNDERGROUND CONSTRUCTION 

50 The distribution mains range in size from 20 in. near the 
stations to 4 in. at the outskirts of the system. The original under- 
ground pipes were laid in a segmental wood casing bound with 
wire. This construction has been fairly successful under favorable 
soil conditions, but the concrete conduit shown in Fig. 8 has been 
found to Ije superior in many resi)ect<? and has been used exclusively 
for several vears in all new construction. In this construction the 
pipe is insulated with a standard tliickness of pipe covering and 
surrounded with an envelope of concrete poured over a wooden form, 
leading an air space around the pipe. Proper underdrainage is of 
course essential. 

51 The longitudinal expansion and contraction of the pipe, due 
to changes in its temperature, are absorbed, in the earlier construc- 
tion, by means of expansion fittings of the copper-diaphragm type. 
In recent construction a slip joint, consisting of a brass sleeve, 
sliding in a packed gland, has been used. The use of slip joints 
decreases somewhat the cost of the pipe line as they will absorb 
more travel ami can l>e placed at wider inter\'als than the diaphragm 
fitiinir. 

52 The underground mains have not l)een laid sufficiently long 
to permit of an accurate estimate of their life. The oldest lines 
laid in worn! casing have now been in servii*e 15 years and, while 
the casing in many places has deterioratevl considerably, in other 
places it is in fairly pootl condition. The concrete construction, 
dating back 10 years, seems to have dotoriorateil verj- little. The 
only repairs or replacements which have bi^en made to date have 
Jjeen made necessary- by external corrosion of the pipe arising from 



J. H. WALKBB 229 

some outside and local cause such as a water-pipe leak. An average 
life of 20 years for the wood log construction and 30 years for the 
concrete construction would seem to be a very conservative estimate. 
Soil conditions in that part of the city where the heating mains 
are laid are particularly favorable, the soil being very well drained. 
Manholes are located at intervals of about one city block (300 to 
400 ft.), to house the shp joints, valves and bleeder traps, 

53 In the heart of the business district the pipes are in tunnels, 
of which there are about 2 miles, lying from 25 to 40 ft. below the 
street level. In cross-section they are similar to a horseshoe and 
are built of brick, with concrete floors. (Fig. 9.) For the most 



FiQ. 8 UNDERQitouND Stbau-Line Constbdction 

|)art they are about 6 ft. in height and 6 ft. wide. The tunnels are 
ventilated by suction fans which draw a small amount of air through 
them continuously and much larger amounts when it becomes neces- 
sary for men to work in them. Their temperature ranges 
between 90 and 130 deg. fahr. When two or more pipes are to be 
laid under a street in a congested district it has proved desirable 
to build a tunnel to avoid the inconvenience to the pubUc attendant 
upon tearing up the street, either for the original construction work 
or for subsequent changes. The tunnel permits of ready access to 
the pipes at any time, and since it is far below any other structures, 
no obstructions are met with. Tunnel construction is exceptionally 
simple in Detroit where the subsoil is a blue clay nearly impervious 



230 CENTRAL-STATION HEATINQ IK DETBOIT 

to water and well adapted to tunneling operations. The co3t oF 
tunnel construction is high but the advantages gained are com- 
pensatorj'- 



Fid. 9 Tunnel Conhtkuctios yok Steam Links 



DISTRIBL'TION LOSSES 

54 Distribution losses from various sources are consideiftble. 
Heat losses from the mains are the principal item. This Iobb can be 



J. H. WALKER 231 

controlled within limits in designing the system by the application 
of the proper amoimt of insulation, so as to produce the most economi- 
cal relation between heat loss and investment costs. Besides the 
condensation losses in the mains and service connections, there are 
losses due to the leakage of steam from consmners' piping and air 
values, and loss due to the escape of unmetered condensation, and 
the slip of meters. About 80 per cent of the steam delivered by 
the plants is metered as condensation in the consumers' buildings. 
This figure compares favorably, considering the nature of the busi- 
ness, with the efficiency of electrical distribution. In 1918, for 
example, the ratio between the electrical plant output and the 
consmners' meter readings for the Detroit district was 83 per cent, 
the large items of loss being transmission, transformation and 
conversion. 

CONDENSATION RETURN LINES 

55 The condensation from the buildings heated is returned 
to the plants only to a very limited extent. In the tunnels it is 
necessary to install a return line to receive the discharge from the 
traps on the steam lines, and wherever possible the condensation is 
drained from the adjacent buildings to this line. It is difficult and 
costly, however, in many cases, to arrange a gravity discharge from 
the building basements to the tunnel, and the cost of installing and 
operating pmnps to handle the condensation would more than offset 
the value of the heat and the feedwater which would be salvaged. 

56 In the districts not served from tunnels, all of the condensa- 
tion is wasted to the sewers. Even with the present high cost of 
coal, return lines would scarcely be a profitable investment. Further- 
more, they are short-lived, and any leakage from them is disastrous 
to adjacent steam lines. The proper method of salvaging the heat 
in the condensation is by means of an economizing coil in the con- 
sumer's building. 

consumers' installations 

57 The consumer's equipment includes, besides the usual radia- 
tors and piping, a pressure-reducing valve, which reduces the service 
pressure to the lower pressure required for heating, and a trap, whose 
function is to discharge the water of condensation from the system. 

58 Any existing steam-heating system can be adapted for 



232 CENTRAL-STATION HEATING IN DETROIT 

service by making a few minor changes. A hot-water system can be 
served with steam by substituting for the fuel-burning water heater 
a surface heater in which the water in the system is heated by the 
steam. A hot-air system can be adapted to the use of steam by 
substituting steam coils for the furnace and using the same duct 
system, though this is rarely done. 

59 Economizing coils, utilizing the heat in the condensation, are 
recommended by the Company but not required. They are seldom 
installed in any but the largest buildings as few consumers car(» 
to make the necessary investment even though a demonstrated saving 
can be made. In some buildings the condensation is passed through 
a surface heater which preheats the water used for lavatory purposes, 
and this seems to be the most satisfactory form of economizer for 
large buildings. 

60 The consumer's installation is the least reliable factor in the 
maintenance of uninterrupted and satisfactory service. Boiler 
plants can be and are designed and operated so as to be extremely 
reliable. A distribution system, if properly laid out, with ample 
capacity and duplicate feeding routes, can be operated so that 
the steam supply to a building is practically never interrupted. 
But the consumer's piping system, and the special appliances 
attached to it, operated by persons unfamiliar with mechanical 
apparatus are a frequent source of trouble. Tagging of the valves 
to indicate their proper operation, distribution of printed instruc- 
tions and other educational measures are employed with varying 
success, and the service is on the whole much more reliable than 
that obtained from the ordinary individual plant. A "trouble 
service'' is maintained day and night to insure immediate attention 
to minor repairs and adjustments of consumers' equipment. 

61 The regular heating season covers 8 months of the year, 
though summer service for cooking and water heating is supplied 
to a few consumers conveniently located. Cool weather in September 
usually makes necessary the commencement of service before the 
contract date of October 1. It has, in fact, been started in August. 
Steam was originally sold to some extent for power purposes but 
this service is now practically discontinued. 

METERS 

62 With the exception of a numl)er of calibrated steam jets used 
in cooking fixtures, the steam is sold entirely on a condensation 



J. H. WALKBB 233 

basis, the condeDsation flowing from the system through a meter 
and thence to the Company's return line or to the sewer. 

63 The art of metering condensation was comparatively new 
when the Company commenced operations and althoi^h many 
advances have been made, it has not yet reached, and probably 
never will reach, the standards of reliability and accuracy of electrical 
metering. The troubles experienced are lai^ly mechanical; but 
there are certain fundamental obstacles in the way of their entire 




COTtportmintRai Filling. Water Eilinding CompartmeirtNal Riled and OwrfloKinq. 
toBightof Center Tgrns Drum in into Compar+mentHoS. 

Dirtction of Arrow 




Nol Ready teEmphj.Noe Full 

and Overflowing. 

into Mo. 3. 



Ho-1 Nearly Empty. No.EOverflowinq 
andNo.3 Nearly Full. 



Fia. 10 Condensation Mbtbb and its Peinciplb or Opbeation 



elimination. The operating conditions, because of excessive tempera- 
ture, moisture, and dirt are severe, and the allowable size and 
cost of the meter are limited. 

64 The meter first used in Detroit was of the tilting type, con- 
sisting of a rectangular pan of two compartments hung on knife- 
edge or ball bearings, the compartments filling alternately and, 
when full, tipping the pan so as to bring the opposite compartment 
to tiie filling position. In 1907 a revolving meter was devised. 



234 CENTBAL-BTATION HEATING IN DSTBOIT 

consisting of a drum of four compartments which filled and dumped 
Buccessively, the drum being revolved by the weight of the water. 
This meter was fairly successful and a few of that design are still in 
use, but a greatly improved design of revolving meter, operating on 
the same general principle (Fig. 10), was originated in 1909 by Hans 
Resert and is.being used at present, with shght modifications. 

LOAD CHARACTERISTICS 

65 The daily boiler-plant load curves vary considerably both 
in magnitude and in shape with different outdoor temperatures. In 






Fto. 11 Typical Loai> Curves of Heatino Ststui 

moderate weather steam is used in many buildings only for a few 
hours in the mommg, resulting in a decided peak at that time with 
' a steady decrease in load throughout the remainder of the day. 
On the very coldest days, however, the heat is used continuously 
throughout the 24 hrs., in most buildings, giving a daily load factor 
on such days of between 85 and 95 per cent. This is illustrated in 
Fig. 11. The monthly sales of steam are very closely proportional 
to the difference between 70 dcg. fahr. and the average outside 
temperature. The annual load factor is about 35 per cent. 

06 The load curves of the individual buildings show these same 
general characteristics although different types of buildings have 
markedly different daily load curves. In Fig. 12 are shown the 



J. a. WALKER 235 

load curves erf three buildings of different types. The high peak, 
in the case of the theater, was caused by the throwing on of the 
fan system. The steam in the office building was shut off at night, 
while in the hotel it was used continuously. 

67 Detroit rates at the present time take no account of varying 
load factor though this is done in some other cities in an approximate 
way. Possibly future progress in the art of metering will lead to 
the general establishment of rate schedules having a demand com- 
ponent as well as a consumption charge. 

ADVANTAGES OP CENTRAL HEATING SERVICE 

68 The existence of a central heating system in a city holds 
certain distinct commercial and economic benefits to the consumer, 
the electric company, and the community in general. 



Fio. 12 Ttpical Load Curves op Lwitiduai. Conbombbs 

69 In a residence district the simplification of domestic ar- 
rangements, the eliniinatioD of smoke and of the dirt and annoyance 
resulting from the handling of coal and ashes, and the availability 
at all times of any desired amount of heat, when coal is high priced 
and scarce, are factors of such real merit that they are reflected in 
higher land and rental values. The scarcity of domestic labor in 
American homes and the steady trend toward more luxurious stand- 
ards of living are yearly increasing the value of these advantages to 
the consumer. The high cost of underground mains compared to 
the amount of steam which can he sold in an area of detached resi- 
dences, however, makes this the least desirable class of service. 

70 The owner of the medium-sized commercial building is par- 
ticularly appreciative of the service since by so arranging the piping 



236 CENTKAL-STATION HEATING IN DETROIT 

that the heat used by his various tenants is metered separately 
and paid for directly by the tenants, he is entirely relieved of 
this burden. In the large downtown store or oflBce building, 
the elimination of coal and ash handling is a very tangible ad- 
vantage and the space saved by the exclusion of boilers has a 
measurable rental value. 

71 To the community at large the elimination of smoke and 
soot, the better handling of coal and ashes, and the reduction in 
manual labor, are distinct economic benefits. 

72 The ability of the service to displace the individual plant 
in residences and business buildings rests upon the value of these 
factors and upon the savings which are possible through the more 
economical purchase, handling and utilization of coal in the large 
central stations as compared with the wastefulness of the small 
individual plant. 

73 The obstacles to the wider use of the service are the burden 
of the investment in the distribution system and the losses which 
necessarily accompany the distribution of heat. Both of thase 
factors are of greater moment in an area of detached residences, 
and consequently, the most favorable field is in the more densely 
loaded business districts. 

74 That central heating service in general can compete at 
profitable rates with the individual plant is sometimes questioned. 
Its ability to do so must rest, in the last analysis, upon a proper 
appreciation, by the consumer, of the intangible advantages which 
have been mentioned. The improvement in standards of hving 
and the general progress toward cleaner cities are undoubtedly 
serving to increase the value of these advantages. 

COMBINATION WITH INDIVIDUAL PLANTS 

75 In some of the larger cities operating companies have been 
formed which take over and operate existing boiler plants in down- 
town buildings and supply steam for heating the buildings in which 
they arc located and often adjoining buildings. This plan owes its 
existence to the fact that the gain in economy of the central station 
over the individual plant is less in the case of large buildings where 
the boiler plants are often of considerable size and can be operated 
with fair efficiency, and also to the fact that the cost of installing 
of distribution mains in such districts would be very high. By re- 
lieving the building owner of the burden of operating the plant, the 
operating company performs a service of measurable value. 



\ 



DISCUSSION 237 

76 Some of the chief advantages of the central heating service 
are defeated by this plan, however, since the dirt, smoke, and labor 
nuisances still exist. By combining such existing plants with an 
efficient central station and operating them only in the coldest 
weather, advantage can be taken of their capacity to reduce the 
size of the main boiler plant, and of the feeders and distribution 
mains. Possibly this plan will be the ultimate solutioq in some of 
the larger cities. It may be described as a relatively cheap invest- 
ment, with relatively high operating costs, to take care of occasional 
peak loads, while the more costly and efficient central plant and 
street pipes take care of the body of the annual load curves with 
obvious economic possibilities. 



DISCUSSION 

August H. Kruesi emphasized the importance of the author's 
statement of the reasons for using live steam in Detroit which 
showed that the additional fuel necessary, if current were generated 
as a by-product of heating, would be one-quarter of that which 
would be used in a condensing steam plant. The reason was the 
great improvement in the performance of large turbo-generators. 
In the plants with which he was associated, he said, they charged 
13.5 per cent of the steam passing through the engines and turbines 
to power and the remainder to heating, and he thought that 10 to 
13 per cent a more representative figure of ordinary industrial prac- 
tice than the 27 per cent which prevailed under the conditions at 
Detroit. In such a case the by-product plant should receive a greater 
credit, that is, the author's item of $10,688 should be doubled, and 
the return on the investment of an exhaust-steam heating plant 
would be in the neighborhood of 25 per cent. 

The Author, in answer to a question, stated that the use of 
exhaust steam for district heating had been definitely abandoned in 
Detroit four years ago and repeated that the underlying reason for 
giving it up was that current could not be generated in the heating 
plants as cheaply as in the main generating station, this being due 
to the fact that the small generating units installed in the heating 
plants must necessarily be reckoned as additional investment since 
they certainly would not reduce the number of generating units in 
the larger plants. 

In his closure, the author emphasized the fact that the paper 



238 CENTRAL-STATION HEATING IN DETROIT 

was intended to describe the conditions existing in Detroit and did 
not imply that the use of exhaust steam for central heating was not 
generally feasible. It was true, however, that these conditions might 
ultimately be reached in many other cities, for with the increasing 
size and efficiency of condensing generating plants it wiU become 
more difficult for central heating stations to produce electricity which 
can compete in cost with that produced by the condensing station. 



No. 1685 

THE PRODUCTION OF LIBERTY MOTOR 
PARTS AT THE FORD PLANT 

By W. F. Vbrnbb, Detroit, Mich. 
Member of the Society 

This paper deals with the prodvctwn of Liberty motor cylinders and conneding- 
rod crankehaft hearings as carried on at the Ford Motor Company's plant at Detroit. 
The contract made with the United States Government called for 5000 motors and 
these were to be produced at the rate qf 50 per day of eight hours. To do thiSj 
important developments in the methods of manufacture were brought about by the 
Production Department of the Ford Motor Company. 

One of these was the method of producing cylinders from tubing. Six opera- 
tions were necessary and the author describes them in detail. The methods employed 
to produce oonnecting^od crankshaft bearings likewise resulted in a great saving of 
time. Twenty-one operations were found necessary for this work and a complete 
description of each is given. The paper concludes with an explanation of the 
method of installing bearings in the upper and lower haloes of the Liberty motor 
crankcase. 

/^N November 22, 1917, the Ford Motor Company entered 
into a contract with the United States Government to build 
5000 Liberty motors. The contract was accepted at a time when 
Ford cars were being manufactured at the rate of 3500 per day 
and to change over from their production to that of Liberty motors 
was by no means a minor undertaking. The manufacture of Lib- 
erty motors differs in many essentials from the manufacture of the 
ordinary type of motor used for automobiles, and of the 14,000 
tools used on Ford car production only 987 were adaptable to the pro- 
duction of Liberty motors. 

2 It was estimated that 350,000 sq. ft. of floor space would be 
required to produce 50 Liberty motors per day of eight hours. The 
ultimate space required was 550,000 sq. ft. As no floor space was 
available, this necessitated the dismantling and removing of several 
thousand machines and the rearrangement of over 50 per cent 
of the regular plant equipment. The new arrangement was so made 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of Thb 
Amsrican Socnmr of Mechanical Enginsebs. 

239 



240 



PRODUCTION OP LIBERTY MOTOR PARTS 



as to allow an entire building for the production of Liberty motors. 
This permitted the concentration of production and therefore maxi- 
mum efficiency and production in the shortest time, and since pro- 
duction in the shortest time possible was the great issue, no expense 
was spared in this shop rearrangement. 

3 The shifting of labor from standard-car production to Liberty 
motor production was gradual. No labor trouble was experienced 
as the curtailment in regular products released enough labor for 
the production of Liberty motors. Table 1 shows the growth in 
the number of men employed in the Liberty Motor Department. 



TABLE 1 



SHOWING GROWTH OF FORCE BUILDING LIBERTY MOTORS 
AT FORD MOTOR CO. WORKS 



Date 



Nov. 22, 1917. 
Feb. 2, 1918. . 
March 1. 1918, 
April 1, 1918 . 
May 1. 1918.. 
June 1, 1918.. 
July 1. 1918. . 



No. of Men Employed 




675 
779 
1.550 
2,450 
3.412 
5.141 



Date 



Au£. 1, 1918. 

Sept. 5. 1918. 

Oct. 1. 1918. 
I Nov. 1. 1918. 
\\ Deo. 1, 1918. 
i Jan. 2, 1919. 



No. <^ Men Employed 



7.976 

9,390 

10.653 

11.288 

826 

543 



4 In accordance with the terms of the contract deliveries were 
planned as follows: 

- April 1918 200 

May 1918 800 

June 1918 1000 

July 1918 1000 

August 1918 1000 

September 1918 1000 

Total 5000 

5 This schedule was subsequently revised with instruction from 
the Government to attain a maximum production of 100 motors 
per day in anticipation of an order for 7000 additional motors. 
The schedule was revised to obtain a production of 2500 motors 
per month by December 1918. 

G Production was fast approaching the goal when the armistice 
was signed, as evidenced by Fig. 1, which shows the pnxluction of 
Liberty motors at the works of the Ford Motor Company for the 
period from November 1917, to December, 1918. As is invariably 
the case in any new undertaking, the production of these motors 



W. F. VEBNEB 



241 



was retarded by many factors which it was the eflfort of the engmeers 
to eliminate. Among the predominating ones were the following: 
a Orders for raw material could not be placed immediately 

on account of incomplete detailed specifications 
b There were many changes. For a period of 14 months, 
1013 changes were authorized: March 1918 showing 167, 
April 109 and May 115; then tapering off to December 
1918 which showed only eight changes. These changes 



2800 




Hov. Oec. Jan. Feb. Mar. Apr. May. June. July Aug. Sept. Oct Nov. Dec. 
- Months, 1918 J 



Months, ldlT>|<- 



Fig. 1 Production op Liberty Motors at the Ford Motor 

Company's Plant 



represent Government changes and do not include those 
made by Ford Motor Company in dies, jigs and fixtures 
which averaged from three to five for each of the above 
c Material specifications were constantly revised 
d Shortage of fuel, especially with sub-contractors 
6 Railroad embargoes delayed shipment on shop machines 

and raw materials 
/ A-4 priority on shop machinery instead of A-1 



242 PRODUCTION OF LIBERTY MOTOR PARTS 

g Lack of acceptable thread gages 

h Lack of actual shop experience by Government inspectors. 
The faUing off in production in November was of course due to the 
signing of the armistice. On the day the armistice was signed, the 
highest mark was reached in the number of motors assembled in one 
day, namely, seventy-five. 

METHOD OF PRODUCING MOTOR CYLINDERS 

7 Several major and important developments were brought 
about by the Production Department of the Ford Motor Company. 
First among these was the cylinder forging made from tubing. 
This method resulted in an enormous saving when considering the 
cost of machining a cylinder from a soUd forging and also the cost 
of making a solid forging. The Liberty motor cylinder was forged 
from a high-carbon steel tube 5f in. outside diameter, 4f in. inside 
diameter by 39f in. mill length, and completed in six operations as 
follows: 

Operation A — Cut-off 

Operation B — Close head 

Operation C — Form the head 

Operation D — Rough-drill bosses for inlet- and exhaust-valve 

ports 
Operation E — Upset and form flange 
Operation F — Heat-treat 

8 Operation A — Cvi-Off. The tube was heated, at the point at 
which it was to be cut to about 1200 deg. fahr., in a specially de- 
signed rectangular gas furnace having a series of circular openings 
along two sides, and through which the tubes were inserted. The 
capacity of the furnace (Fig. 2) was such that the successive tubes 
were heated sufficiently for cutting within the time required for the 
cutting operation. Once started, the operation was continuous 
with a production rate of 150 tubes per hour per machine. 

9 Upon removing the tube from the furnace it was placed in 
the shearing? die of a press equipped with a special punch and die. 
The tube was then fitted with an arbor so constructed that, as the 
punch of the press sheared the outer wall of the tube, the arbor 
transmitted tlic shearing power to the lower wall, thus shearing 
the whole witliout distorting the tulje. The punch and die were 
set on the press so that the end of the tube was cut at an angle of 
19 deg. with the center line of the tube, to the required length of 



W. F. VBRNER 243 

20} in. at one side of the angle and 19} in. at the other side. This 
angular cut was essential to Operation B. 

10 Operation B — Close Head. It was considered for a long 
time next to impossible to forge a Liberty motor cylinder from a 
tube on account of the manufacturing difficulties encountered in 
closing the head. When the end of the tube to be closed was cut at 
right angles to the center line, it was found unsatisfactory due to cold 



Fia. 2 Gab Fumjace fob Eeatino Tobes — Operation A 

shuts or unfused sections in the metal occurring in the center of the 
dome. 

11 However, by cutting the tube at an angle of 19 deg. with the 
center line of the tube it was found that the forming dies could be 
so constructed to cup or draw inward the tube wall, the high or 
extended portion of the wall causing the converging or closing of 
the metal to one side of the center line of the tube until, in the 
final forming of the head, the metal was joined at right angles to the 
19 deg. cut (see Fig. 3). After this operation the appearance of 
the closed end resembled the common type of explosive shell with 
the nose jMirtion at an angle of 19 deg. The central portion of the 



244 PBODUcnoN of libbbtt motor pabts 

dome U thus formed without a weld and retains to the fullest 
extent its fibrous strength. 

12 The angular head of the tube was then heated in a furnace, 
similar to the one previously described, to about 1900 deg. fahr. 
preparatory to forming the head so that the point could be used as 
a part of the boss which later was drilled for an intake or exhaust 
post. 

13 The die used in this operation was of the double-action type 



Fia. 3 Pkeliminabt Forhinu of the Cylinder Head — Operation B 

u,ii(i comprised two scmi-circulur steel Jan's, tapered on the upper 
outside diameter and pivoted in the rear to swing horizontally. A 
cast-iron locking plate was attached to the blanking foot of the 
press and taix^n-d to correspond to the taper of the jaws, so that 
when the blanking foot wjis in down position the tapered surfaces 
of the jaws scrvcwi to lock the tube in position. The interior upper 
parts of the jan-s were bored and fitted with split bushings or bronse 



245 

bearings to fit the punch. Resting Sush, normally with the upper 
surface of the jaw, was a semicircular steel supporting band equipped 
with three guide pins supported on springs. 

14 When the jaws were swung in position around the tube, 
they formed a steel ring which gripped the tube around the top or 
heated portion and as the punch descended, the ring shpped down 
the tube, and the supporting springs depressing under the pressure 
of the punch thus prevented the tube from bulging when the punch 
closed in the head. The punch was designed so that the dome was 



Fia. 4 pBESs Used in Final Formino or Cylinder Head — Operation C 

drawn to a point 19 deg. to one side of the center line of the tube, 
as previously described. 

15 Operatwn C — Form the Head. This operation was per- 
formed on a press (Fig. 4) provided with a Specially designed punch 
and die. To serve as a die, a bolster or base plate was mounted on 
the bed of the press. The base plate was bored in the center to 
receive the shank end of an upright cylindrical locating arbor and 
counterbored to receive a thrust plate. Two sections made up the 
arbor, the lower of which was made of soft steel bored to receive the 
hardened-steel tip or top section. The top section tapered slightly 



246 PRODUCTION OP LIBERTY MOTOR PARTS 

inward at the extreme end and the top surface was curved to properly 
form the dome or head of the cylinder. 

16 Horizontal sliding jaws around the locating arbor were held 
open by springs and operated by cams attached to the ram of the 
press. The upper interior portion of the jaws was shaped to form 
the expanded area for the combustion chamber. When the jaws 
were closed by the action of the cams, they fitted snugly around 
the punch. 

17 As the ram descended, the cams attached thereto forced the 
jaws together so that as the punch pressed down the head of the tube 
on the arbor it formed the valve-port and spark-plug bosses and 
the jaws formed the expanded area for the combustion chamber. 
On the back stroke, the jaws were forced apart by the springs as 
soon as the pressure of the cams relaxed. In case of adhesion to 
the cylinder, a wedge attached to the ram and operating between 
the two ends of the jaws breaks the adhesion on the back stroke. 
A knock-out sleeve located about the base of the arbor and operated 
by an arm beneath the bed of the press on the back stroke was carried 
upward and loosened the cylinder on the arbor. Production on one 
machine totaled 150 per hour. 

18 Operation D — Rough-Drill Bosses for Inlet'' and Exhaust- 
Valve Ports, The cylinder was held in a trunnion fixture attached 
to a drill press so that the center line of the valve port swung in line 
with the spindle. The cylinder was located by using the valve-port 
bosses. 

19 Operation E — Upset and Form Flange, These operations 
were performed on a 5-in. forging machine in one heat-treat on two 
separate dies. The die used for the first portion of the operation 
(upset) comprised two horizontal sliding steel jaws operated by 
cams and bored at both ends to fit tighly about the body of the 
cylinder. At the middle the jaws were undercut or recessed so that 
in the upsetting the metal would so flow as to form a heavy ring of 
metal at the section at which the flange was to be located. The 
punch was in the form of a mandrel with a shoulder which fitted the 
closed jaws and which on striking the bottom of the skirt or open end 
of the cylinder forced the heated metal to the proper upset dimen- 
sions al)0ut the mandrel and into the recess of the jaws. 

20 The die employed in the second portion of the operation 
(form flange) was of the same type above described, with the excep- 
tion that the jaws were counterbored at the entrance end to the 
forged flange dimensions, instead of being reoeesed in the middle. 



W. p. VERNER 247 

The punch consists of a mandrel with a shoulder surrounded by a 
sleeve which extended beyond the shoulder. The sleeve fitted the 
closed jaw of the die and the inside extended portion upset the 
metal by pressing it against the die, thus forming the flange. The 
sleeve was provided with two vent holes permitting gases that 
might be formed during the forging, to escape. 

21 In operation the skirt of the cylinder was first heated to 
about 1900 deg. fahr. in a furnace, then dipped in water to a depth of 
about 1^ in. This cooling was done to form a ring of hard metal 
at the bottom of the skirt for the punch to act upon. The cylinder 
was then placed in the forging machine and the flange made. Pro- 
duction totaled 85 per hour on one machine. 

22 Operation F — Heat-Treat. The completed forged cylinder 
was placed in a large rectangular heat-treating furnace and heated 
to a temperature of 1525 deg. fahr. and then quenched in a brine 
solution. After quenching it was heated in an annealing furnace 
to a temperature of 1125 deg. fahr. and cooled in air. Brinell test 
was 217-255. This heat treatment left the cylinder ready for the 
machine operations. 

METHOD OP MANUFACrURING CONNECTING-ROD CRANKSHAFT 

BEARINGS 

23 Next in importance to the method just described of pro- 
ducing Liberty motor cylinders was the development of a special 
process of making bronze babbitt-lined bearings for the crankshaft 
end of the connecting rods which would stand up under the Gov- 
ernment's 50-hour test. The method comprised 21 operations, as 
follows: 

1 Rough-drill closed end 

2 Rough- and finish-bore bronze and face one end 

3 Turn outside diameter to fit babbitting fixture and face 

one end 

4 Babbitt 

5 Cut-off gate, rough- and finish-bore babbitt 

6 Finish-turn outside diameter to 3.095 in. and face ends 

to length 

7 Press in broaching ring 

8 Broach hole to 2.4275 in. in diameter 

9 Press out of broaching ring 

10 Grind outside diameter to 3.075 in. 



248 PRODUCTION OP LIBERTY MOTOR PARTS 

11 Cut in halves, using -^in. saw (2 on) 

12 Close in 

13 Swage 

14 Finish-mill the parting line 

15 Face ends to length 

16 Fillet both ends 

17 Cut two grooves for forked-end rod 

18 Drill and ream dowel holes in lower half-bearing only 

19 Cut two J-in. semicircular oil grooves on the parting line 

20 Cut twelve oil pockets in both halves 

21 Burr. 

24 Operaiion 1 — Rough-Drill Closed End. A 21-in. drilling 
machine was used in this operation, which was necessary only on 
castings where a thin web of metal entirely closed one end. 

25 Operation 2 — Rough- and Finish-Bore Bronze and Face One 
End. This operation was performed in a 12-spindle 14*in. multiple 
machine. The bushings were gripped in a chuck and bored at the 
rate of about 80 per hour. 

26 Operation 3 — Turn Ouiside Diameter to Fit BabbMing 
Fixture and Face One End. A 12-spindle 14-in. multiple machine 
was also used for this operation. The bushings after boring were 
clamped on special arbors and the outside diameter turned to fit 
babbitting fixture. 

27 Operation 4 — BabbiU. The equipment required for this 
operation consisted of acid vats, tinning furnaces, die-casting ma- 
chines with water-circulating systems, a furnace large enough to 
supply a unit of four die-casting machines with molten babbitt and 
a compressed-air outfit to use with the gas furnaces. 

28 The bushing was first dipped into a flux made proportionately 
of 11 lb. sal ammoniac, 9 lb. zinc chloride, 6 qt. muriatic acid and 
13 qt. water. The hydrometer reading was 23 to 25 deg. B. 

29 Before placing the bushing in the specially designed die- 
casting machine to be babbitted it was immersed in the molten tin. 
The die-casting machine, Fig. 5, consisted of a rectangular plate 
mounted on suitable legs or base, having a cored hole for receiving 
a crucible containing the metal; a crucible with suitable rim having 
two bosses on which were fitted bearings for a pump-lever shaft; 
a pump fastened to two bosses of the rectangular plate for forcing 
the metal through a nozzle into the die proper; and a fixture mounted 
on the plate straddling the pot for casting the bearing. 



W. F. VEHNBR 249 

30 The fixture for casting the bearing was made up of a circulat- 
ing-water-cooled mandrel or arbor sliding vertically in a housing. 
Thie bousing was secured to two side-support brackets which were 
bolted to the plate. The lower die holder carrying the die spans 
the crucible and shdes vertically on guide pins. Springs received 
the weight of the holder in suspensio i. Just inside of the guide- 
pin bearings and screwed into the holder were two suitable rods 



Fio. 6 Babbittinq Machinb — Operation 4 

which extended through the housing. The bousing was drilled 
and counterbored to receive rods and springs. The lower die 
holder in its normal or loading position was suspended on the springs 
so that the opening or gate of the die was slightly above the nozzle 
of the pump. 

31 For babbitting, a tinned bushing was inserted in the lower 
die and the water-cooled arbor was moved downward until stopped 



250 PRODUCTION OP LIBERTY MOTOR PARTS 

by the arbor stripper ring, clamping on the upper end of the bushing 
to be babbitted and the gate end of the lower die on the pump nozzle. 
The movement of the arbor was controlled by an upward movement 
of the hand lever. While holding the arbor firmly in position with 
one hand, the operator with the other hand pulls upward on another 
hand lever attached to the pump lever, thereby forcing the molten 
babbitt from the pump cyUnder through the nozzles and into the 
bushing. 

32 The metal was allowed to set for about 30 sec, when the 
pump-control lever was pushed downward. The piston of the pump 
upon its return uncovers ports permitting molten metal to flow 
into the cylinder preparatory to another casting. Simultaneously 
the arbor-control lever was thrown downward, assisted by the 
weighted end, causing the arbor stripper ring to strike violently 
against the lower face of the arbor housing and thereby stripping 
the babbitted bushing from the arbor. This upper movement of 
the arbor allowed the lower die holder to regain its normal position 
and severed the connection between the lower die and pump nozzle, 
which assured the bushing sticking to the water-cooled arbor and 
not becoming gate-anchored to the lower die. The thickness of 
the babbitt wall at the top for best results in babbitting was ^\ in. 
and for the bottom ^^ in. 

33 Operation 5 — Cut-Off Gate, Rough- and Fintsh-Bore Babbiii. 
Geared-head screw machines were used for this operation. The 
bushing was held by a three-jaw clutch; an ordinary cut-off tool 
held in the tool block of the cross-slide was used for cutting off 
the gate, and a boring bar with (wo cutters (one for roughing and 
one for finishing) was mounted in the turret for boring hole to proper 
diameter, with an allowance for broaching. 

34 Operation 6 — Finish; Turn Outside Diameter to 3.085-in. and 
Face Ends to Length, In this operation 14-in. x 4-ft. lathes with 
back-arm attachments wore used. The bushings were held on an 
expanded arbor, which was held in the spindle of the machine. 
A cutter mounted in the tool block of the lathe cross-elide turned 
the outside diameter to fit the broaching ring. The back arm 
carried a tool Mock with two cuttc^rs spaced to face the bushing to 
its proper length. 

35 Opiration 7 — Prci>s in Brinichiug Ring. For this opcTation 
(see Fig. 0) a bolster plate with two holes lx)red large enough to 
allow the l)ushing to drop through and counterbored to suit outside 
diameter of the broaching ring was strapped to the bed of a press. 



W. f. TBRMEB 251 

In the ram of the machine were carried two cylindrical punches 
slightly smaller in diameter than the outside diameter of the bushing 
but differing in length. 

36 In operation a ring into which had been pressed a broached 
bushing was placed in the counterbored hole under the long punch, 
and a bushing that was not broached was slightly entered into an 



Fio, 6 FRCsaiNQ in Broachinq Rinq — Operation 7 

empty ring and placed under the short punch. When the press 
was tripped, the long punch forced the broached bushing out of 
the ring and the short punch forced the bushing, which had not 
been broached into the ring, 

37 Operation 8 — Broach Hole to 2.4275-tn. Diameter. A special 
broaching machine, Fig. 7, was designed for this operation. The 



252 PBODUCTION OP LIBEBTT MOTOR PABTS 

legs were removed from a No. 1 Knowles keyseater and the body 
complete with gears, rack, sHde, etc., was bolted in a vertical position 
to a special base casting. An upper guide of cast iron, with a hard- 
ened and ground steel bushing for the broach-holder quill was 



FiQ. 7 iJi'KciAL UuoAaiiNo Machine — Operation 8 

dovetailed to the ways of the machine. Below this upper guide 
was the extended work-holding bracket (a part of the Ix^ of the 
Knowles key.scater) which carried the work holder. This work 
holder was a hardened steel bushing ground on ita outside diameter 



W. p. VERNER 253 

to fit the hole in the bracket and counterbored to fit the broaching 
ring. On the special base casting below the work-holding support 
was bolted the lower guide, a gray iron casting with a hardened 
bushing groimd to fit the broach holder. The broach holder was a 
long toolnsteel bar hardened and ground on one end to fit the guide 
bushings and on the other end to fit the holes in broach and broach- 
holder quill. 

38 The foot-pedal bracket was bolted to the base of the machine 
and was bored to fit the broach holder, which it served to guide 
and keep in alignment. Between the lower guide casting and the 
foot-pedal bracket was disposed a collar firmly fastened to the 
broach holder, and attached to this collar was a yoked lever. This 
lever was so fulcrumed that the weight of the broach was just slightly 
more than counterbalanced by a cast-iron weight which insured 
the broach end of the holder being piloted in the quill when the 
broaching was being done. A hand lever with lift rods was attached 
to the yoked lever to control the loading. 

39 The broach had eight cutting edges varying in size from 
2.422 in. diam. to 2.4273 in. diam. In addition to the cutting edges, 
the upper end of the broach had three burnishing surfaces, the 
diameter of which were 2.4274 in. The broach was 5 in. long and 
had a hole groimd to 1.625 in. diam. to fit the end of the broach holder. 

40 In operation, the keyseater functions normally, with the 
exception that the down (or what would be the return) stroke of the 
ram is utilized for pushing the broach through the work. The 
hand lever controlling the broach holder is pulled down imtil the 
latch on the foot lever engages the top of the collar on the broach 
holder. This operation holds the pilot end of the holder and the 
broach holder quill apart, allowing the broaching ring containing 
the work to be mounted in place. The broach holder was then 
allowed to ascend until the end of the holder protruded enough to 
allow slipping on the broach, and then further until its end was 
piloted in the hole of the quill. The machine was then tripped 
and the broaching was done on the downward stroke of the sUde 
carrying the quill, the broach being forced through the work by the 
pressure exerted on descending quill. This downward stroke was 
carried far enough to allow the foot-lever latch to engage the collar 
on the broach holder automatically. The machine was reversed, 
disengaging the quill and the broach holder. The broaching ring 
containing the work was then lifted out and the broach was removed 
from the holder; leaving the machine ready for the next operation. 



254 PROOtlCTION OP LIBERTY HOTOB PAitTS 

41 Operation 9 — Press Out of Broaching Ring. This operation 
has Ijeen previously described under Operation 7, and is shown in 
Fig. 6. 

42 Operalion 10 — GHnd Outside Diameter to 3.075 in. The 
bushing was ground between centers on a 6 x 18-in. plain grinder. 
A split hardened-steel ring, ground on its outside to a diameter of 
2.4275 in., was inserted in the bushing and a hardened arbor ground 
with a slight taper was in turn inserted into the split ring, the hole 
of the latter being ground tapering to conform to the arbor. A light 



Fio. 8 Machine for Besdcnq BBARiNaa — Operation 12 

pressure applied to the end of the arbor expanded the ring suffi- 
ciently to firmly hold the bushing for grinding. 

43 Operalion 11 — Cut i?i Halves using j\-tfi. Saw (2 on). This 
operation was (wrformed on plain milling machine with an indexing 
type of fixture. The work-liolding arbor of the fixture was made 
long enough to accommodate two bushings. The bushings were 
clamped on tlic arbor with a "C" washer and stud bolt. A com- 
mon Mj-in.-widt! milling cutter was used. 

44 Operation 12 — 67o.sc In. After being cut in half, the Ixiar- 
ings were Iwnt to decrease their diameter so as to allow their fitting 
easily into the swaging fixture used in Operation 13. Tlie bending 
fixture, Fig. 8, was made with a semicircular arbor bolted to a base 



W. F. TBBHSB 255 

plate. A housing with three cam lever slides was fastened to the 
base plates. OpeiuDgs in both sides of the housing allowed for the 
insertion of the bearing over the arbor. The top cam lever clamped 
the work in position and the side cam levers closed in the bearing 
the necessary amount. 

45 Operation 13 — Swage. This operation, which was performed 
on a press, was the keystone operation in the successful production 
of accurate bearings, as the set given them insured their holding the 
shape of the master forms. 

46 The fixture, Fig. 9, was comprised of a hardened and ground 



Fia. 9 Upseiting of Beakings to Finished Dimensions — Operation 13 

steel base plate fastened to a hardened-steel form. Two eyebolts were 
held by pins in each side of the form. These parts assembled formed 
the female section and were fastened to the bolster plate. The 
male section was made up of a balf-round arbor of hardened steel 
on the clamping plate, the joining surfaces of which were also ground. 
Two slots in each end of the clamping plate allow the eyebolts to 
hold the two sections together. A tongue-and-groove construction 
on the joint surfaces of the sections kept them in alignment. A 
filler piece made of hardened steel and ground to the finished bearing 
dimensions was used to take the flow from the ram of the press. 



256 PRODUCTION OF LIBERTY MOTOR PARTS 

A hardened-steel cylindrical punch with a flat-ground bottom surface 
was fitted into the ram of the machine. 

47 In operation, a half-bearing with a filler piece on top 
was clamped between the male and female section of the fixture; 
the press was then tripped and the cylindrical punch on descending 
struck the filler piece, which projected slightly above the upper 
surface of the fixture. This upset the metal to the finished bearing 
dimensions with an allowance on the parting-line surfaces of about 
0.003 in. for finish-milling. 

48 Operation 14 — Finish-Mill the Parting Line, A plain mill- 
ing machine on this operation was very satisfactory considering the 
close limits of plus or minus 0.00025 in. A fixture with a half-round 
seat or nest bored to fit the outside diameter of the half-bearing 
was bolted to the plate of the machine. The parting-line surfaces 
of the half-bearing were leveled by the hinge gage attached to the 
fixture. The work was clamped with a half-round hardened and 
ground steel clamp, the curved surface of which was made to fit the 
inside diameter of the half-bearing. Two plain milling cutters were 
mounted on the arbor and were so spaced that they straddled the 
clamping bolt and nut. 

49 Operation 15 — Face Ends to Length, A 14-in. x 4-ft. lathe 
without a tailstock was used for this operation. A hinged clamping 
ring was used to clamp the work tight on the steel arbor, which was 
mounted in the spindle of the lathe. Two halves, or one complete 
bearing, were machined at one setting. A special tool block with 
two cutters straddling the clamping ring faced the bearing to the 
proper length. 

50 Operation KS — Fillet Both Ends, The fillet cuts in both 
ends of the bearing, to clear the radius of the crankshaft pin, were 
made on 21-in. drill presses. The fixture comprised a circular-shaped 
base bored in the center to receive a flange<l hardened-steel pilot 
busiiing. The outside diameter of this pilot bushing above the 
flange was made to fit the inside diameter of the bearings and the 
hole of the pilot bushing was a fit to the pilot of the filleting cutting 
holder. The half-bearings were clamped in pairs about the piloting 
bushing by means of two hinged clami)s rotating on a pin located in 
the rear of the fixture. A single-fonned filleting center was fastened 
in a slott(ui holder. The holder was held in the spindle of the drill 
press with a tajx^red shank. 

51 Operation 17 — Cut Two Grooves for Forked-End Rod, A 
14-in. X 5-ft. lathe. Fig. 10, with a special cross-slide arranged with 



W. T. TESNEB 257 

front and back tool blocks, was used for this operation. A stub 
arbor mounted in the spindle of the machine was made with a flange 
to serve as a stop for locating the work while clamping with a hinged 
ring, similar to the one used for Operation 15. The half-bearings 
were clamped on the arbor at one setting. The arbor was provided 
with a center bo ihai the lathe tailstock could be utilized to stiffen 
the support of the work under the pressure of the cut. The grooving 
cutters were of the circular forming type, with six cutting edges. 
The adjustment of the cutting edges was controlled by the movement 



FiQ. 10 Method or Cuttinu Grooves fob Forked-End Rod — Operation 17 

of a toothed lever. The teeth of this lever iiieHhed with the teeth 
in the boss about the center of the grooving cutter. The cutter in 
the rear tool block was used for roughing, and the front cutter for 
finishing. Suitable stops were arranged on the cross-slide to control 
the depth of the cut. 

52 Operation 18 — Drill and Ream Dowel Holes in Lower Half- 
Bearings Only. A 2-spindle 14-in. drill was used on this opera- 
tion. The drill jig comprised a base with side supports, to which 
was fastened the drilling plate. On the under side of this plate 
was fastened a locating block formed to fit the contour of the work. 



258 FBODncnoN of xjbebty uotob pabts 

A slide disposed between the side supports and actuated by a cam 
lever served to clamp the work in position for drilling and reaming. 
Slip bushings were used in the drilling plate to insure accuracy, 
one for the drill and one for the reamer. 

53 Operation 19 — Cui Two j-in. Semicircular (Ml Grooves on the 
Partiiig lAne. This operation was done on a hand miller. The 
cast-iron fixture supported the work in a semicircular nest and was 
clamped in place with a strap and thumbscrew. 

54 Operation 20 — Cu/ Twelve Oil Pockets in Both Halvea. This 



Fig. 11 Metbod of Currwa Oil Grooves — Ofe&atiom 20 

operation was also done on a hand miller. The fixture, Fig. 11, wu 
made to so clamp the work that the parting surfaces were in a verticat 
position. The oil pockets were cut on a radius, the cut being 
<;'} in. deep at the parting line and running out to the inner surface 
of the bearing } in. below the parting line. An arbor with six i^^. 
wide plain cutters properly spaced was mounted in the spindle ot 
the machine and supported by tlie over arm. 

55 Operatiim 21 — Burr. A No. 1 keyeeater was used for 
removing the burrs on the inside or babbitted surface of the bearings- 



W. F. VKBNEB 259 

llie fixture was made in halves, the lower half having a cylindrical 
section \t4uch served to hold the fixture in the machine. Ways for 
the slide carrying the broach were machined in the lower half, and 
also recesses cut for the hardened and ground blocks on which the 
parting surfaces of the work rested. The upper half was made in 
the form of a clamp, this part being bored to fit the half-section 
of circular hardened and ground steel liner, the inner surface of 
which fits about the outside of the work. Bosses bored for guide . 
pins extended from the fixture and served to keep the halves in 



Fio. 12 Uppeb and Loweb Halybs of Libkbtt Motob Ckankcase 

alignment. The weight of the upper half was disposed on springs 
coiled about the guide pins. The broach was of a semiciraular 
section and bolted to the slide. This sUde was connected to the 
ram of the machine. A copious flow of oil was kept on the work, 
keeping the locating block and broach free from chips and dirt. 
The cutting was done on the pull or regular stroke of the machine. 

HEn^OD OF FirnNQ BBAKINQS TO CRANKCASE 

56 Fig. 12 shows the upper and lower halves of the crankcase 
as it comes from the assembly department, and after being bolted 



260 PRODDCTION OF LIBERTT MOTOR PABTS 

togrthor the bearings liiic ia reained to 3.000 in. or slightly less, 
['lulvr a to1omiu-o of 0.0015 in. below 3.000 in. 50 crankcoses can be 
i\>!un<Hl Ix'foiv it lH'«inu's nocessarj- to rogrind the reamer bars. 

.'•7 'rho nuist satisf!ii'tor>' n'«nu'r is of a soliiUbar pattern having 
iiistTttnl ciittiT blados iM't'iiwI ill plaiv witli two taper collais on 
n-amor Iwr, ('jm- imist W lakoii in st>fiiring the main liearing bolts 
to thf lowtT lutlf of thi' fitiiiki-as*' U'fore n-aming. The nuts are 
ilr:iwn up until tin* shoiildor (dowvl {X)rtiun) on the main liearing 
Imli omu's motal to iiu-tal with tho frank«»s<>. Tlie nut is then 



■-•.^V .VSSEVKl^ 



^-w half 
V'mt like 

<^ ^jog bar 



W. F. VERNEB 261 

show the amount of scraping to be done. This first scraping opera- 
tion removes all slight distortions in the bearings and brings the 
holes in alignment. 

59 Next the lining bar is again blued or blacked, placed as before 
the halves of the crankcase, bolted as previously described, and 
then the bar is rotated so as to again leave an impression. If the 
first scraping is properly done about 25 lb. pressure at the end of the 
bar handle will rotate the bar. The case is now ready for the final 
scraping, leaving the bearing holes absolutely to size and in perfect 
alignment. By this method, of lining and sizing the bearing holes 
in the crankcase, strict interchangeabiUty of all bearings is obtained 
and a perfect backing is made for the babbitt-lined bronze-backed 
bushings. 

60 The crankcase is now ready for the bearings. The backs 
of the bearings, both upper and lower, are blued or blacked, also 
the edges of the parting Une of the upper halves, and a 2.6275-in. 
lining bar is then placed in position. The crankcase is again bolted 
together, using care as before in screwing down the nuts. The 
crankcase actually clamps the bar, and a wrench with heavy handle 
is used for rotating the bar. If conditions are ideal, the wrench 
when in a horizontal position will, when given a light tap, fall 6 in. 
to 8 in. to a position as shown. 

61 The crankcase is again pulled apart. The lower half of the 
crankcase takes an impression around the main bolting from the 
upper half which has been blued or blacked. This indicates that 
the two halves of the crankcase came together and that the 
bearings did not hold them apart. Impressions are next looked for 
made by the parting line of the upper half of the bushing on the 
lower half of the bushing. An impression shows that the two 
halves of the bushings came together and that the bar did not hold 
them apart. 

62 The bearings are numbered so that they can be identified 
and put back in the same place from which they were removed. 
The backs of the bearings having been previously blued or blacked 
leave impressions on both the upper and lower halves of the crank- 
case, and if the impressions show high spots, the cases are again 
scraped. This is the final scraping, after which the crankcase is 
ready for the crankshaft assembly. 

63 The above described method may appear to be long and 
tedious. Two men, however, can scrape four complete crankcases 
in 8 hr. The bearings were furnished complete to a tolerance of 



262 PBODUCTION OP LIBERTY MOTOB PABTB 

plus or minus 0.00025 in. in both inside and outside diameter. All 
halves of the bearings are interchangeable. No allowance was made 
for reaming the bearings in place. It was found that the babbitted 
smf aces, when left in their original broach-finished state gave better 
results than when reamed and scraped in place. 

DISCUSSION 

H. M. Crane * (written). Mr. Vemer's description of the 
method of producing Liberty Engine cylinder forgings from steel 
tubing is extremely interesting, and I was closely enough connected 
with the Liberty Engine program to be able to appreciate the great 
aid that this system of production was to the rapid and economical 
manufacture of the Liberty Engine. 

The method of manufacturing the bearing boxes also, as de- 
scribed by Mr. Vemer, is extremely ingenious and advantageous, 
in view of the fact that while providing bearings held to extremely 
close limits of accuracy, it is still possible to do most of the work 
on a complete round bushing, which after spUtting is brought to 
the correct shape for use without shims by what are practically 
punch-press operations. 

To me, however, the most interesting thing in this paper is 
the description of seating the main bearing bushes in the crankcase. 
The method, of course, is not a new one, for it has long been known 
to be essential that the bearing bush must be properly seated to get 
a satisfactory job. Unfortunately, in the last few years, due to 
the pressure for quantity production and low cost the very essen- 
tial operations described have been largely omitted in automobile 
engine practice. I am thoroughly convinced that this has not 
been economical from the automobile owners point of view, and 
that the sUght saving in first cost has always resulted in consider- 
able increased maintenance expense or in the use of imnecessarily 
large bearings to obtain a given result. 

* Vice-President, Wright -Martin Aircraft Corporation, New Brunswick, N. J. 



No. 1696 

FIRE ENGINES AND THE ESSENTIALS OF 

FIRE FIGHTING 

By Charles H. Fox, Cincinnati, Ohio 
Member of the Society 

The importance of the fire engine and Uie methods employed 4n fire fighting are 
frequently underestimated by both layman and engineer, and this is due in no smaU 
part to the fact that the principles involved are not sufficiently understood. In this 
paper the author presents the essentials qf effective fire fighting and shows the impor^ 
tant rekUion thereto of the fire engine, 

A brief historical sketch cf the development of the fire engine is first given, and 
performances and methods of rating fire engines, as outlined by the National Board 
of Fire Underwriters, are next presented. The paper concludes with a discission 
qf the losses in fire hose and the determination qf nozde areas, 

OTEAM power was not successfully applied to fire engines until 
the beginning of the year 1853. Up to that time the so-called 
*'hand engines" were used exclusively and it should also be under- 
stood that at that time the present-day system of water works, 
was still in its infancy and, therefore, the chief dependence for a 
supply of water for fire-extinguishing purposes was upon methods of 
storage in vogue before water mains came into general use. 

2 The conventional hand fire engine of that day comprised a 
rectangular wooden box suitably mounted on four low wheels. 
Pumps, of the piston type, were housed within and firmly fixed to 
the floor of the box; working levers were provided and motion was 
imparted to the pistons by a host of firemen lined up on opposite 
sides of the apparatus. At this early period fire hose was not plenti- 
ful, the best was crudely made up of leather, and the pumps were, 
therefore, placed close to the scene of the fire. Water, largely con- 
veyed by a hand-to-hand passing of fire pails, was poured into the 
engine trough, where it was picked up by the pumps and forced 
through the leading hose and onward to be thrown on the fire. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of Thx 
American Society of Mechanical Engineers. 

263 



264 FIRE ENGINES AND ESSENTIALS OF FIRE FIGHTING 

Somewhat later it became customary to equip these hand engines 
with a non-collapsible suction hose, so that water could be drawn 
directly from cisterns or wells, but the wooden tub or reservoir 
always remained a cliaracteristic feature of these old-time machines. 

EARLY STEAM FIRE ENGINES 

3 Ivirly in .hmuary 1853, Mr. A. B. Latta successfully tested 
his new stoam-drivcm fire engine. Mr. Latta was a citizen of Cin- 
cinnati and although his pioneer effort resulted in the production 
of an extremely hea\y machine, the engine was purchased by the 
city and known as the Joe Ross, This first steamer marked the 
beginning of a Jiew epoch in fire fighting. 

4 The real basic element of Latta's invention was embraced 
chiefly in a quick-steaming water-tube boiler, dependent entirely 
upon a forced circulation of water through the steam-generating 
coils. The fire pumps were of an ordinary' piston type. They were 
laid horizontally, and occupied positions forward on opposite sides 
of the machine. Two steam cylinders, in alignment with the pumps, 
were placed so that the pump rods extended to the steam pistons; 
wliile the steam piston rods, passing through the rear cylinder 
heads, were each coupled to a crosshead and by the use of connect- 
ing rods motion was commimicated to the two rear wheels of the 
engine. This arrangement obviously provided means for propell- 
ing the apparatus under its own power. As a matter of fact, how- 
ever, the Joe lioss and later engines of similar model were alwaj's 
drawn by horses, as the rear driving wheels could be disconnected 
from the engine and the pumping mechanism by means of clutches. 

5 In the light of what is now known concerning the actual 
power requirements of a fire engine, one must believe either that 
the voluntiM*i-s who manned the primitive hand pumps were mus- 
cular giants, or that the machine, iis compared with power-driven 
engines, must have been woefully ineffective. The latter view is of 
course the correct one, and since the volunteer work was not satis- 
factory th(*re was a gn^at demand for more advanced methods. 

Mr. Miles (ireen wood, another of Cincinnati's public-spirited 
citizens an<l i)roniinent as a manufacturer, was at that ime greatly 
interested in fire-depart men t affairs. lie was a staunch advocate 
of anytliinjr wliicli would iinprov(? eoiiditions, and to further his 
unselfish aims li(» took personal charge and became Chief of the 
Department with the result that, within three months following 



CHARLES H. FOX 265 

the installation of the Joe Ross engine, the fire-fighting forces were 
completely reorganized under his leadership and the first paid fire- 
department was inaugurated in the city of Cincinnati. The year 
1853 therefore also marks the beginning of the present paid system 
and which has since been extended and generally applied to all 
large fire departments. 

7 Latta's second steam fire engine was built and installed in 
the year 1853. The purchase of this machine was made possible 
by popular subscription and the engine was named and long known 
in Cincinnati as the Citizens' Gift. From the beginning, Latta's 
new industry did not result in a monopoly, for rival steam fire-engine 
builders were soon in the field. Before tracing the further develop- 
ment which followed, it seems proper to refer briefly to the salient 
factors associated with fire streams and the fire engine. 

ESSENTIALS OF EFFECTIVE FIRE FIGHTING 

8 It may first be stated that many of the methods employed by 
fire departments are not properly appreciated, either by the lay- 
man or the engineer. The importance of the fire engine is fre- 
quently underestimated and the possibilities of averting disaster 
are also undervalued, this being due in no small part to the fact 
that the principles involved are not suflSciently understood. There 
is, therefore, need for greater knowledge of fire prevention and fire 
fighting. 

9 When fire must be fought, fire streams can be effective only 
when the water is expelled from the nozzle at an appropriate speed. 
In other words, imless enough of the initial pressure available for 
starting the flow through the hose survives at a point immediately 
back of the nozzle orifice, the resulting jet will not measure up to 
its mission. The characteristics of a fire stream — good, bad or 
indifferent — are directly dependent upon the velocity of the jet 
and obviously the velocity is proportionate to the surviving pres- 
sure just mentioned. For the best results the flow may be too slow, 
while on the other hand disappointment will follow when the velocity 
of discharge goes beyond what might be termed the maximum 
economical limits of nozzle pressure. 

10 It can be shown that Uttle is to be gained by forcing nozzle 
discharges at much beyond 100 lb. pressure and the proper remedy 
when only such high pressures are available is to substitute a nozzle 
of larger bore. On the other hand if the jet seems to lag, it is an 



266 FIBE ENGINES AND ESSENTIALS OF FIRE FIGHTING 

indicatioD that the surviving pressure is not high enough to afford 
the required velocity of discharge, and, under these circumstances, 
if the initial hydrant or pump pressure is already at its highest 
possible point, or if it is also impossible to augment the volume of 
water passing, then the proper remedy is to substitute a nozzle hav- 
ing a smaller bore. 

11 A well-trained fireman does not necessarily have to be 
familiar with the hydrauUc formulae by which these varjring effects 
can be accurately analyzed, but it would be most desirable if he 
could be educated to the proper use of fire nozzles under the widely 
varying situations encoimtered in the fighting of fires. 

12 It may be of interest in this connection to note that the 
range of nozzle discharge pressures is not large between a stream of 
inferior reach and the swiftest jet within the economical power 
limit. While it is true that larger nozzles will "carry" farther at 
the same velocity than nozzles of smaller bore, yet for practical 
purposes it will be found that the most eflScient fire streams are 
developed under the following conditions: 

13 A jet issuing at a velocity under 60 ft. per sec. (about 25 
lb. pressure) would not be considered a good stream. Above 60 ft. 
per sec. the jet stiffens and as the velocity of flow increases the 
stream becomes the more effective. However, when the flow attains 
a speed of approximately 120 ft. per sec, with a corresponding dis- 
charge pressure of nearly 100 lb. per sq. in., a point is reached where 
additional forcing fails to contribute much in the way of added 
distance. A rough but nevertheless rational assumption and one 
which works out well in practice is that stream velocities from 60 
to 90 ft. per sec. can be easily managed, but above 120 ft. per sec. the 
streams are difficult to control and the pressures are needlessly 
wasteful. 

FIRE-ENGINE PERFORMANCE 

14 The function of a fire engme is either to draw water from 
any basin or other conveniently located source or, when fire hy- 
drants are available to make up the pressure which is seldom high 
enough in ordinary water mains to serve for effective fire service. In 
fighting fire, it is not uncommon to elevate the nozzle far above the 
source of the water supply. This procedure of course involves loss 
of forcing pressure, which is in proportion to the static head of the 
column. The greatest power-absorbing medium between the source 
of supply and the point of discharge is the fire hose. It may also be 



CHABLBS H. FOX 267 

said that here is involved the point which is least understood in 
the subject of hydraulics as applied to fire-fighting practice. 

15 Fire-engine pump pressures must necessarily be kept within 
limits compatible with the strength of the fire hose, and when work- 
ing pressures are maintained at upward of 250 lb. there is always 
considerable risk, because the average fire department is not always 
supplied with hose which will safely withstand the high pressures 
which modem fire engines are capable of developing. The initial 
pressure as indicated at the pumps drops off steadily toward the 
point of discharge and it should be well understood that all pressure 
which is not thus absorbed by the friction of the flow through the 
hose is finally manifested as velocity at the nozzle orifice. Hence, 
imless the several factors that enter are considered together it can- 
not possibly follow that the best results will be attained. 

16 The preceding paragraph expresses the gist of fire-engine 
performance and the variable features always associated with the 
work may be definitely stated as follows: 

Pumping Capacity: The number of U. S. gallons discharged per 
minute. 



a At 120 lb. 
6 At 200 lb. 
c At 250 lb. 



pressilre registered at pump. 



Water Supply: The source and adequacy of the supply — 

d If at draft — height of lift represented by the vertical 
distance of water surface below center of pump intake 

e If at hydrant — capacity as indicated by pressure regis- 
tered when water is flowing. 
[Note: High static pressure, as indicated when hydrant 
is not flowing, is no index to its capacity to discharge 
any required volume.] 

The Layovi: Conditions affecting the discharge are — 

/ Size and character of the fire hose 

g Number of hose lines in use 

h Lengths of hose stretched between engine and nozzle 

i Niunber of streams played 

j Bore and style of nozzles used 

k Elevation at the point of discharge. 



268 FIRE ENGINES AND ESSENTIALS OF FIRE FIGHTING 

METHODS OF RATING FIRE ENGINES 

17 During a long period, manufacturers of fire engines were 
unrestricted as to the ratings which they assigned to their machines 
and when "gallons per minute" were given, no definite discharge 
pressures were associated with the expressed pumping capacity. 
In pumping tests, the actual discharge was seldom checked or veri- 
fied, with the result that comparisons of fire-engine performance 
were more a matter of guess than of accurate determination. This 
condition, however, no longer obtains, and the change is largely due 
to the exacting supervision initiated and persistently pursued under 
the auspices of the National Board of Fire Underwriters. A corps 
of expert engineers is constantly in the field for the purpose of in- 
vestigating and keeping in touch with the fire-fighting situation 
in all cities, and pursuant to this policy of the insurance interests, 
it is now generally known what may be expected of fire engines 
when new and in prime condition, and periodical inspections dis- 
close weaknesses subsequently developed in service. 

18 The competency of fire-engine operators is a matter which 
the manufacturers cannot control, and as the conditions under 
which fires must be fought are so variable, it is also quite impossible 
to make fire engines so completely automatic that they can be 
operated and successfully maintained without fairly skilled atten- 
tion. The latent possibilities in the best fire engines, hose and 
other modem appliances can only be realized when such apparatus is 
well manned. 

19 According to the standards formulated by the National 
Board of Fire Underwriters in addition to the normal rating of a fire 
engine, expressed in gallons per minute, there must also be sufficient 
power to expel the rated volume of discharge at a pressure of not less 
than 120 lb. per sq. in. Inasmuch, however, as certain situations 
may demand higher initial pressures, the pumps must also be so 
related to the attending power plant that higher pressures can be 
realized. 'i1i(Tefore further qualifications arc demanded, as follows: 

20 One-half the rated capacity should be discharged by the 
pumps at 200 11). pressure and one-third the capacity at 250 lb. 
pressure. Furthermore, fire engines should not be limited to 250 lb. 
pressure, for the fire departments of our large cities are confronted 
with the so-called ''sky-scrapers," and when water must be forced 
to the upper floors of these tall structures, the overcoming of the 
static head alone greatly lowers the eflfective pressure. However, when 



CHABLES H. FOX 269 

the power of an engine is fully utilized in forcing water to extinguish 
fire in a lofty building the work and consequent strains are perfectly 
legitimate. On the other hand, when pressures much in excess of 
250 lb. are employed simply to overcome resistance, which could 
be reduced by a more intelligent use of hose and nozzles, then the 
work evidently represents not only a waste of power but also very 
bad practice. 

LOSSES IN FIRE HOSE 

21 A good engineering axiom to observe is that all work should 
be accompanied with the least expenditure of force, and the aim 
here is to express the fact, that many firemen have yet to learn how 
to smooth their own way. The point can best be illustrated by an 
example. 

22 The inside diameter of fire hose ordinarily used is 2^ in. 
Cotton-jacketed, rubber-lined hose is the most common kind and 
lengths of 50 ft. (each such length constituting a section) are the 
recognized standard. The conventional way of expressing friction 
loss, or the loss of pressure when water is flowing through the hose, 
is in terms of pounds per square inch for a unit length of 100 ft. for 
any given rate of flow stated in gallons per minute. This drop in 
pressure or so-called friction loss varies somewhat with quaUty or 
make-up of the hose; varying diameters and the fact that some 
makes of hose present a smoother waterway than others are also 
causes which prevent formulating a coefficient which can be applied 
to all kinds of 2^-in. fire hose and by which friction losses can be 
predetermined with precision. 

23 It, therefore, follows from the foregoing statements that all 
tables setting forth friction loss in fire hose should be accepted only 
as close approximations, and in actual practice the results may, 
therefore, not agree. However, within the range of velocities and 
the lengths of hose ordinarily encountered in actual service, it has 
been found that for water flowing through 100 ft. of average good 
quaUty 2§-in. rubber-lined and cotton-jacketed fire hose, the fric- 
tion loss will be about 14 lb. when the rate of the flow represents 
250 gal. per min. Accepting this approximation as a basis, it must 
also be kept in mind that the friction loss increases in direct propor- 
tion to the length of the hose; hence, for the same rate of flow 250 
gal. per min. through 200 ft. the total loss would be twice as much 
as indicated for 100 ft. 

24 The effect of varying the rate of flow follows a different and 
more intricate law, the friction loss increasing more nearly in pro- 



270 FIRE ENGINES AND ESSENTIALS OF FIRE FIGHTING 

portion to the square of the flow. Therefore, if the loss per 100 ft. 
is 14 lb. when 250 gal. are flowing, it will be quite correct to assume 
that it will require nearly 2 X 2 or four times as much pressure, i.e., 
56 lb. to discharge 500 gal. through 100 ft., 112 lb. through 200 ft., 
etc. Stated as formula 

L = 0.00023 (? 

where L is the friction loss per 100 ft. of 2§-in. rubber-lined, cotton- 
jacketed fire hose of good quaUty, and G is the flow in gallons per 
minute. The results obtained by using this approximate formula 
agree closely with those in the tables compiled by Mr. John R. 
Freeman, which appear in Fire Stream Tables, issued by the In- 
spection Department of the Associated Factory Mutual Fire Insur- 
ance Companies, Boston, Mass. 

25 The red book. Fire Engine Tests and Fire Stream Tables, 
published by the National Board of Fire Underwriters, New York, 
gives the formula: 

L = 2<? + Q 

in which L represents the friction loss, as before, and Q the volume 
flowing in gallons per minute divided by 100. Mathematical precision 
is always laudable, but in action, firemen could hardly be expected 
to work out problems other than such as might readily be solved 
mentally. However, extreme accuracy is entirely unnecessary in 
fire fighting, but, if guns can be aimed accurately while a battle is 
raging in war, it would seem altogether feasible that fire engines 
could be set to work with hose layouts arranged with nozzles com- 
patible with the general conditions of a situation, so that the effi- 
ciency latent in modem apparatus and appliances would be more 
frequently realized. 

26 A close approximation, reasonably well gaged according to 
a correct and definite line of reasoning is preferable to merely a 
guess or no system at all. Given sufiicient study, the difficulties of 
apparently intricate hydrauHc formulae will vanish most surprisingly, 
and the real need today is for a greater cultivation of mental capac- 
ity to the end that commanding officers in the fire service become 
especially proficient in obtaining the best possible results with the 
apparatus at their disposal. 

NOZZLE AREAS 

27 It will be evident, from what has been presented, that the 
characteristic of a fire stream develops according to the following 
interlocking factors: 



CHABLES H. FOX 271 

a The pressure which survives at the nozzle 
b The volume of water reaching the nozzle 
c The relation of the nozzle bore to a and b. 

The difficulties which attend in the existing system of nozzle bores 
is largely due to the fact that the diameters vary after the order of 
ordinary machine-shop reamers, i.e., | in., 1 in., 1| in., 1^ in., and it, 
therefore, follows that the areas of the orifice have increments which 
increase in arithmetical progression. 

28 A more logical system has long been proposed, wherein the 
nozzle orifices would be multiples of a fixed standard. In accord- 

TABLE 1 CALIBER SYSTEM OF STANDARD FIRE STREAMS 

DXVISSD AXID PbOPOSID BT CH4BT.BB H. FoX 



CaUber 


Ar«a 


Diam. of 


Nominal 




Value 


8q. In. 


Bore, In. 


Diam., In. 




25 


0.3068 


0.625 


f Ezaet 




50 


0.6136 


0.883 


i plus 0.008 




75 


0.0204 


1.082 


1^4 plus 0.019 




100 


1.2272 


1.250 


1} Exact 




125 


1.5340 


1.397 


11 plus 0.022 




150 


1.8408 


1.530 


li plus 0.030 




175 


2.1476 


1.652 


If plus 0.027 




200 


2.4544 


1.767 


1| plus 0.017 




225 


2.7612 


1.875 


li Exact 




250 


3.0680 


1.976 


US plus 0.039 




275 


3.3748 


2.072 


2,^ plus 0.010 




300 


3.6816 


2.165 


2| plus 0.040 




325 


3.9883 


2.252 


2} phis 0.002 




350 


4.2951 


2.337 


2x\ plus 0.025 




875 


4.6019 


2.420 


2i plus 0.045 




400 


4.9087 


2.500 


2iExaot 





Nora: — The true basic caliber unit is the area of a circle Hn. in diam., vis.: 0.012271875. 

ance with this method it was suggested that the l^-in. nozzle, a 
size now most common, be designated as ''100 caUber." Other 
nozzles in the same system would be made with orifices of such 
areas that the water-discharging capacity at the same pressures 
would be respectively as the caUber number by which each is to be 
distinguished. Therefore, instead of nozzles measuring f , f , |, 1 
Hf H) If » li> lf> IJi 1J> ^^^ 2 in. in diameter, the caUber system 
would leave the |- and 1^-in. sizes as before, calling these 25 and 100 
caUbers, with others, to cover practically the same limits, but with 
fractional bores arranged in multiple, such as 25, 50, 75, 100, 125, 



272 FIRE ENGINES AND ESSENTIALS OF FIRE FIGHTING 

150, 175, 200, 225, 250, 275 calibers. Table 1 jqves these proposed 
caliber values in more detail and for a complete discussion the 
reader should consult, Caliber of Fire Streams, Report of Transactions 
of the International Association of Fire Engineers, 1911. 

29 Referring briefly to the gasoline-powered fire engines of today, 
the development has passed beyond the stage which first marked 
early endeavors to supplant the horse-drawn steam fire engines. 
It is now conceded that in all points of eflSciency and economy, 
the well-built gasoline tj'pe is superior to the steamer. This state- 
ment docs not exclude the vital essential of reliability and it is also 
true that a comparison made upon the basis of their nominal ratings, 
will show that the better gasoUne-driven engines will deliver fully 
one-third more water when working steadily for any considerable 
length of time simply because no stoking is required. It is not 
an easy task to keep the boiler of a steam fire engine up to its 
maximum working limits, and the elimination of the labor involved 
explains why this estimate in behalf of the gasoline engine is 
reasonable. 

30 Experience has also shown that the earlier objections with 
respect to the possibilities of tractive failures by reason of adverse 
weather conditions were more imaginary than n^l and as the result 
of all the advantages attending the introduction of gasoline fire 
engines, the steam-driven types are pa.ssing out of use by force of 
the dictum which spi^lls the survival of the fittest. 

DISCUSSION 

Clarence Goldsmith (written). The ixiix»r as presented 
covers the general subject under di<cusj?ion in a most complete 
manner and no particular phase of the many items involved has 
been given undue prominence to the exclusion of others. However, 
in order to refresh the minds of those who liave lirtilt with some of 
the probli'nis in timers piist and to put on guiird those who may be 
called uiH^n to apply s<inie af the prinripl*^ without iuiving time to 
pve tht'iii pruptT cunsidenition. it may l»e dt^i ruble to treat certain 
jKirts of the subject more in det;ul. 

Althuuph the principles of design of the oentrifujKil pump and 
of the pisoline enpine are well known by the dt^signen? of each claflB 
of ecjuipnuiit. y«t, the obs*. nat i« >ns of the writer durine the past 
ten years show that much trouble k encountered when the centrif- 
ugal-pump manufacturer purchases a gasoline engine and attempts 



DISCUSSION 273 

to drive his pump with it, or when the manufacturer of a gasoline 
engine attempts to drive a centrifugal pump with his engine. 

The pump manufacturer has tested his pump and determined 
its most economic speed, capacity and pressure, while the gasoline- 
engine manufacturer has determined the horsepower deUvered at 
different speeds. In many cases sufficient study is not given to the 
performance curves of the two pieces of equipment before the capac- 
ity and size of each which are to be selected to form the combined 
motor-driven pump are determined. 

In order to have the equipment successful it is fundamentally 
necessary that the power developed by the gasoline engine at any 
speed shall exceed that required to operate the centrifugal pump 
at the same speed. To the casual reader this statement may appear 
entirely superfluous, yet the failure to meet this axiomatic require- 
ment has cost many manufacturers much time, money and trouble. 

The writer is in absolute accord with the statement that the 
efficacy of fire streams up to 2i in. in diameter cannot be materially 
improved by increasing the nozzle pressure beyond 100 lb. per sq. 
in. Observations made on a l^-in. stream by experienced servers 
showed that the increase in the effective vertical reach when the 
nozzle pressure was increased beyond 100 lb., was not over 15 ft. 
up to 150 lb. pressure and that when the nozzle pressure was in- 
creased beyond 150 lb. the effective vertical reach decreased. This 
decrease was evidently due to the increased resistance which the 
air offered to the surface of the water jet. It is thus evident that 
a pressure of from 80 to 100 lb. at the nozzle is required to develop 
efficient streams from deluge sets, turret nozzles, deck guns, ladder 
pipes and water towers; for hand lines, even when operated by 
trained men, 60 to 80 lb. is sufficient. There are a few cases how- 
ever when pressures in excess of 100 lb. may be of some advantage 
on 2 and 2j-in. tips. For instance, in overhauling lumber when 
stacked, overtiiming walls or breaking through obstructions when 
the nozzle can be brought up close and horizontal or vertical reach 
are not desired, pressures as high as 200 lb. at the nozzle enable the 
operator to take advantage of the added momentum of the stream 
and accomplish more efficient work. 

In p)assing, it may not be out of place to remark that one of 
the best methods of placing a fire stream in favorable position for 
observation is to make the observations at night with the stream 
directed between a bank of electric lights and the observer. 

The capacity of a pmnper or steamer should always be stated 



274 FIBE ENGINES AND ESSENTIALS OF FIBE FIOHTINO 

in gallons per minute against the observed net pressure, the latter 
to be the actual difference in the pressures observed on the suction 
and discharge sides of the pump. As an example two cases will be 
cited: (1) An engine is taking suction from a reservoir 5 ft. below 
the suction inlet, the friction loss through the suction hose when 
delivering 750 gal. per min. is 6^ ft., making the total lift 11.5 ft. 
or 5 lb.; the observed discharge pressure is 115 lb., therefore the net 
head pumped against by the engine is 120 lb. (2) An engine is 
taking suction from a hydrant fed by a street main carrying a pres- 
sure of 55 lb., when the engine is delivering 750 gal.; the friction loss 
through the hydrant and branch is 3 lb. and through the suction 
hose 2 lb., making the total friction loss 5 lb., thus delivering a pres- 
sure of 50 lb. on the suction side of the pump; the observed discharge 
pressure is 170 lb., therefore the net pressure is 120 lb. 

Many engineers of steamers have maintained that their engines 
worked better at draft than when suction was furnished imder pres- 
sure from a hydrant. It is the opinion of the writer that this is an 
erroneous notion and he has never seen it substantiated in practice. 
At any rate it is an established fact that the pressure furnished on 
the suction side of the pump increases the capacity of a pumper when 
delivering at pressures above that pressure at which it is rated. 

If a pumper is rated at 750 gal. at 120 lb. net pressure and water 
is delivered to the suction side of the pump at lb. pressure, then 
it will discharge somewhat more than one-half this quantity at 200 
lb. pressiu-e and a Uttle more than one-third the quantity at 260 lb. 
pressure. Now if water is delivered to the suction side of the pump 
under 80 lb. pressure the discharge pressure will be 200 lb. and the 
capacity will remain at 750 gal., and if the water is deUvered to the 
suction under 130 lb. pressure the pump will continue to discharge 
750 gal. per min. with a discharge pressure of 250 lb. 

It is thus evident that the pressure in the water mains should 
be taken advantage of in all cases. This is particularly important 
in large fires where high pressures are required to develop the more 
powerful streams and to overcome the friction losses in long lines 
of fire hose which must be laid from the engines located farthest 
from the fire. Where good hydrant pressures are available, say, 
from 40 to 60 lb., the more nearly can the rated capacities of the 
V engines be maintained when required to work against discharge 
pressures greater than 120 lb. 

The one and main object of the pumper or steamer is to deliver 
water at the nozzle at a pressure suitable to develop an efficient fire 



DIBCUB8ION 275 

stream. Now it has been shown that any increase in discharge 
pressure reduced the delivery of the apparatus ; therefore every means 
should be adopted which wUl relieve the necessity of raising the dis- 
charge pressiu-e. Increasing the carryiag capacity of the hose lines 
from the engine to the nozzle will cut down the friction loss which 
in turn will cut down the pressure at the engine which is required 
to deliver the water at the nozzle imder suitable pressure. 

The use of 3-in. hose will accomplish this result in the most 
economical manner and practically the same results can be accom- 
plished by siameaing 2^-in. hose lines. For a given quantity of 
water flowing, the friction loss through a 3-in. hose line is about 
one-third that through a 2^-in. line of equal length. Two 2|-in. 
lines siamesed offer about one-fourth the frictional resistance of a 
single 2^-in. line. It, therefore, is advisable to lay 3-in. hose lines 
from pumpers to the entrance of buildings and to ladder pipes, tur- 
ret nozzles, deluge sets and water towers in order to enable the 
pmnpers to deliver larger quantities of water. 

The successful fighting of a fire depends -much upon the de- 
velopment of powerful hose streams and these in turn depend upon 
the efficient operation of the pumpers or steamers. An untrained 
and imexperienced engineer may so operate an engine that is in 
first-class condition that it will not give as good service as a much 
poorer engine in well-traiaed hands. So much depends upon the 
operation of pumpers and steamers that it is of prime importance 
that the crews be trained in the operation of their machines at 
capacity at frequent intervals, say, once a month. 

Abthur M. Greene, Jr., spoke of the prize offered about 1838 
by the Mechanics Institute of Boston for a successful fire engine 
and described the apparatus brought out by the firm of Braithwaite 
and Ericsson. Beneath the driver's seat was a bellows, attached 
by a connecting rod to the rear whee's of the device so that draft 
was provided while the vehicle was being drawn to the scene of the 
fire. At the front of the engine was a large air chamber around 
which was coiled the smoke pipe leading from the boiler. 

The Author. Mr. Goldsmith expresses some further truths 
associated with the ratings of fire engines which shoidd be better > 
understood and, referring especially to the work done on the intake^ 
side of the pump, i.e. when the water supply is drafted, or, the gain 
realized when the flow to the pump is imder pressure, I am pleased 



276 FIRE ENGINES AND ESSENTIALS OF FIRE FIGHTrNG 

to see these points incorporated in his discussion of the subject, be- 
cause it frequently occurs in practice, that by neglecting this, either 
not enough or else too much credit is accorded the engine which is 
tested. 

It also is true that effective work may frequently be done by 
using streams of exceptionally high velocity, all of which is justified 
when circumstances warrant such use of excessive power. The 
writer's argument however should be taken to center on the desir- 
ability of attaining the maximum effect at the nozzle with the mini- 
mum waste of power and, as Mr. Goldsmith indicates, this can be 
accomplished by using either 3-in. hose instead of 2i-in. or the 
friction losses between the pump and the nozzle may be materially 
reduced by using double leads of 2J in. hose, joined into one at a 
point near to the nozzle. 

Next to an intelligent management of the engine and pumps 
and yet of equal importance, comes the judicious disposal of the 
avenues through which water is forced to the nozzle orifice, and it 
is to this end that the writer's paper was written. Much more could 
be mentioned, but the saUent features only could be given adequate 
expression. 



\ 



No. 1697 

AN ELECTRICAL DEVICE FOR MEASURING 
THE FLOW OF FLUIDS IN PIPES 

Bt Jacob M. SprrzoLASs, Chicaqo, III. 
Member of the Society 

This paper deals with the theory and experimental develapment of an electrical 
dance in which an ammeter and watt-hour meter are utUized in measuring the flow of 
fluids in pipes. The dectric current which is used is regtUaied by the differential 
pressure of the fluid passing through the pipe. 

The principle of the device invoUfes a combination of the physical laws governing 
the flow of fluids in pipes and the flow of an electric current. The units of flow meas- 
urement care represented by general equations covering the relation hettoeen the velocity 
of the fluid in the pipe and the differential column obtained by the device. For the 
units of the electrical measurement^ in the standard adopted the maximum capacity 
of flow is represented by a current of 1 ampere at a constant pressure of 40 volts. 

The diagrammatic relation of the units involved in the electric measurement of 
the flow is shown by two curves; one in the form of a parabola representing the relar 
lion of differential head to the current — the other in the form of a hyperbola represent- 
ing the relation between the current and the corresponding resistance in the circuit, 

A summary is given of the experimental work in preparing a resistance which 
would vary according to the relation established, and which would operate under the 
usual conditions of flow, Parliculars are given of the early experimental devices and 
of the improvements made until a satisfactory model was obtained, and the method of 
testing is discussed and a typical run described in detaS, 

In condiision, a number of instances are dted where the instrument described 
has made possible the adoption of central measuring stations in large manufacturing 
plants, resulting in improved economy of operation. 

DESPITE the fact that the science of mechanical engineering 
is much older than that of electrical engineering, its methods 
of measurement are nevertheless in many respects much behind 
those afforded by the latter. A striking example of this is found 
in a comparison of the methods of measuring fluid motion in pipes 
and the flow of an electric current. The instrument used for the 
electric current is simple and direct-reading, and while there have 
been many excellent devices adopted for measuring the flow of 
fluids in pipes, it has been quite generally agreed that an instru- 
ment similar to the ammeter or w^attmeter would be of great value. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of Ths 

AlfSBICAN SOCIETT OF MECHANICAL ENQINEEBS. 

277 



278 BLBCTBICAL UEASUBINO OF FLmD FLOW 

2 Recently the writer had the privilege of experimenting with 
a flow-measuring device in which these instruments ara applied. 
Measurement is accomplished by means of an electric current which 
is BO regulated by the differential pressure of the flow that it repre- 
sents the amount of fluid passing through the pipe. 

3 The main features of the device are shown diagrammatially 
in Fig. 1. The U-tube, partly filled with mercury, is made to bal- 
ance the impact pressure of the flow in the pipe by the rise of 
mercury in the low-pressure side of the tube. The mercury column 
also forms a part of the electric circuit, as shown in the figure. This 
electric circuit contains a flxed external resistance Ri in series with b 



Pio. 1 DuoKAMMATic Sketch or Actbok'b Elbctbical Dbvici worn Hsis- 
UBiNQ Flow or PLCtDs in Pifbs 

variable internal resistance Ri, a constant electromoti^'e force B, 
an ammeter A and a watt-hour meter W. In the contact chamber 
C, which forms the low-pressure side of the U-tube, there are a 
number of conductors of varj-ing length placed above the mercuiy 
column, and as the mercury rises it makes contact with one «»- 
ductor after another. The variable resistance ft is subdinded by 
these conductors into resistance steps corresponding to the vaiying 
length of the conductors, so that the rise and fall of the mercuiT 
column varies the amount of resistance and thereby regulates the 
amount of current passing through the circuit, 

4 The biisic principle of the derice accordingly involves the 
laws governing the Row of fluids through pipes along with those 



JACOB M. 8PITZGLA88 



279 



governing the flow of an electric current. The problem of estab- 
lishing the theoretical relation between these fundamental laws 
oS^ered little difficulty because of the similarity between the units 
of flow measiu-ement, such as pressiure and velocity, and the units 
of electric measurement, such as voltage and current. On the other 
hand, the attempts to apply the theory to a working model were 
beset with numerous difficulties, and the obstacles that were over- 



DincfionafFhw v 



SfoflK 



•pzzz 

1^ 



*vynamic 



(Piiof Tube) 









I (Ho^kTube) 




I (Orifice flak) 



Fig. 2 Methods of Detebmixinq Velocity Pbessure 

come during the long period of experimental work presented many 
problems which are briefly dealt with in later paragraphs. 



UNITS OP FLOW MEASUREMENT 



5 The relation between the pressure and velocity of fluids in 
its simplest form is represented by the well-known equation 



2(7 



y = Pi - Pi - AuJ 



or 



.-v/ 



2 ff(P, - P.) 



y^2gw>j\ 



[1] 



280 ELECTRICAL MEASURDfC OF FLUID FLOW 

where v and y represent the velocity and density of the fluid; 
(Pi - Pi) the equivalent differential pressure; h the height and w 
the density of the liquid column balancing the differential pressure 
of the flow. 

6 This differential pressure hw may be obtained, as shown in 
Fig. 2, either directly by balancing the difference betw*een the 
dynamic and static sides of a pitot tube inserted in the line, or in- 
directly by balancing the difference between the high- and low- 
pressure sides of a venturi tube, nozzle tube or orifice plate. In 
the case of the pitot tube, the differential column in the U-tube 
represents the flow or motion existing at the given section of the 
line, but in the venturi tube, nozzle or orifice, the column obtained 
represents the change of motion produced by the artificial obstruc- 
tion of the passage at the given section of the pipe. 

7 In anj' case, however, the relation between the differential 
column thus obtained and the velocity of the fluid in the pipe may 
be representeil by Equation [1], pro^^ded there is introduced the 
exix^rimental coefficient derived for the given tube or orifice. Thus 
in RtwnU, 

V = C^ 2^\ - [2] 

The Vi^hune of the fluid Q (Vissing per unit of time throu^ an area 
A is givon by the ei|uation 

<?--4r».4r\2snr\ - p] 

tho a>rn\^ixnuling weight C |xt unit of time is 

:uuJ ilio total woiiiht lor a givon jx^riiHl of liino t is 

ift KCt\ h^ [6] 

S ILn'.:*.^; aiiv^pttsl tlio foJv»i\Mu»i j^xnionil ixiuations for the flow 
v^t t'v.iviN *lu* vx^: :vNjHn\viu\>i o!ivt:u* nu\'W<urt*:uo;ns nis^v be oovered 

v*.":v*;; .:*. ;r,v.iv:v> !'v>\\ii^»; lhi\nii;l; ;ho oUvirio circuit of 
•,:-.x* :'.^\iN •.::;•.'.»; »K*\ uv Tl'.o i!>;:viir,o:;: w^k? dosiglied 80 
;;> to V.aw v^v.y* .r.v.iv'v ;vivvsvr.t tlu* nu'iximum e^Mcity 









JACOB M. SPITZGLASS 281 

E = electromotive force of the circuit. A uniform pressure of 
40 volts was selected to represent the average density 
of the fluid measured 
Wt = amount of electric energy expended in the circuit of the 
device in a period of time t 

R » total resistance of the circuit in ohms 

F = rate of flow in the pipe corresponding to the electric current 
in the circuit, or the ratio of <? to /. F is the "indicat- 
ing factor" of the flow meter. 

T = total amount of flow or weight of fluid corresponding to 
the electric energy passed through the circuit. T is 
the ratio of G to W, and is designated as the "totalizing 
factor" of the flow meter. 

9 Since by definition, FI ^ G and from Equation [4] G = KC 
Vyy/h, therefore FI = KCy/yVh, or 

I--fvWh [6] 

The value of /C is constant for any given set of conditions as deter- 
mined from Equation [4]. The vahie of C, depending upon the 
particular design of the tube or orifice, is also constant for any given 
case. 

10 To find the value of F, let Imox- be the current in amperes 
corresponding to the maximum capacity of the meter, Gmax-, which 
in turn corresponds to the maximum differential colunm /u<». From 
Equation [6]: 

KC — 

Imax' = —pTX^y^/Tlmax [7] 

whence .- — 

F = KCy/y - L^J 

Inuix* 

and since Imax- is equal to unity, 

F = KCVWh^x [9] 

11 The quantity Vhmax- is called the characteristic or the scale 
of the given meter and it determines the capacity of the meter, 
depending upon the amount of differential column A^o*. which the 
meter is able to develop and record. 

12 Combining Equations [6] and [9], 



s/-. 



[10] 



Inuix* 



282 ELECTBICAL MEASURING OF FLUID FLOW 

It is interesting to note that hlKma*- represents the relative value 
of the differential column for a given rate of flow, and 100 A/Am*. 
is the percentage variation of the head in any given meter. From 
Equation [10] it follows that in order to represent the amount of 
flow, the current 7 should be numerically equal to the square root 
of the relative height of the mercury column in the U-tube of the 
meter. From the same equation, 

That is, the height of the column for a given flow is numericaUy 
equal to the constant Amos, times the square of the current flowing 
through the circuit. 

13 From Ohm's law {E = IK) we obtain by substitution 



R^E 



y/^ [12] 



That is, the resistance R in the circuit should be numerically equal 
to the voltage divided by the square root of the relative height of 
the differential column. 

14 It remains to determine the value of 7, the ''totalizing 
factor'' of the instrument, or the ratio of 6 to IF. Since, Wi - Eli 
and by definition TWi ^ Gt ^ Fit, therefore 

r = | CM] 

That is, the totaUzing factor of the meter is equal to the indicating 
factor divided by the voltage in the circuit. 

15 Fig. 3 shows diagrammatically the relation of the units 
involved in the electric measurement of the flow. The parabolic 
curve to the right shows the variation of the current in the electric 
circuit representing the capacity of the flow and corresponding to 
the percentage variation in the differential column balancing the 
velocity pressure of the flow. This curve represents the solution 
of Equation [11]. The hyperbolic curve to the left of the diagram 
shows the relation between the current and the corresponding re- 
sistance at the given voltage of the electric circuit. The solutioii 
of Equation [12], or the relation of iZ to A, is obtained indirectly 
by following from a given value of R on the resistance curve to 
the corresponding value of h on the current-and-capacity curve. 

16 It will be observed that the diagram does not include the 
first 10 per cent of the flow capacity inasmuch as this represents 



k 



JACOB U. BFITEOLASB 283 

only 1 per.ceot of the differential column, which is ss low as a prac- 
tical device is able to measure with any degree of accuracy. This 
disadvantage, however, is offset by the fact that the scale of the 
flow meter can be so chosen that the desired mesaurements will 
f^ within the active part of the scale. 

PRACTICAL APFUCATION 

17 After the relations between the various factors had been 
detennined, the problem reduced itself to the construction of a 



Fio. 3 DiAOBAM SHowiNa Relation betwben Units or Fluid Plow and 
Electric Units 

resistance which would be regulated by the differential colimm of 
the flow according to the solution of Equation [iS]. The first 
attempt in this direction was made by inserting a continuous resist- 
ance coil in a water manometer where the height of the water column 
would reduce the amount of resistance in the coil by short-circuiting 
the part immersed in the water. The first trials were made with 
direct current, and while it was anticipated that electrolytic action 
woiUd be set up between the metal conductors and the water, it 



284 ELECTRICAL MEASURING OP FLUID FLOW 

was nevertheless expected that this action would not take place 
when alternating current was used. 

18 There was little information available on the subject and 
it was therefore necessary to determine experimentally the amount 
of resistance needed. After obtaining some idea of the amount re- 
quired and being hindered by the accumulation of deposit in the 
container, which was at first attributed to the electrolytic action 
of the direct current, provisions were made to continue the experi- 
ments with alternating current. It was discouraging, however, to 
note that practically the same action took place between the metal 
conductors and the water when alternating current was used. Re- 
peated analyses of samples of the deposit disclosed that it was a 
formation of oxide, due to the corrosion of the metal conductors in 
contact with the water. 

19 Besides the formation of deposits, there were other disad- 
vantages in short-circuiting the metallic resistance by a wat^r 
column. On one hand the conductivity of the water varied with 
its hardness, thus introducing a variable resistance in the part of 
the column which was covered by water, and on the other the vapors 
formed over the surface of the water had a tendency to short-circuit 
the resistance coils, again introducing a similar variable in the por- 
tion of the column above the level of the water. 

20 When repeated attempts to eliminate these defects had 
failed it was decided to adapt mercury instead of w^ater for the 
regulating column of the instrument. The use of mercury, how- 
ever, necessitated radical changes in the form of the device. The 
effective column of mercury for the average velocity pressure would 
be too small to cover the continuous resistance coil and produce 
the desired regulation of the resistance. It was therefore found 
necessary to regulate the resistance by steps through conductors 
coming in contact with the top of the mercurj*^ column. Fig. 4 shows 
the elementary form of experimental device adapted for this purpose. 

21 In this elementary device the successive conductors were 
divided into steps of equal height from the zero level of the mercury 
column. The electromotive force of the circuit was maintained 
constant at 40 volts. The resistance of the circuit, which amounted 
in all to 400 ohms, was subdivided by the contact ro<.ls into 40 con- 
secutive steps, and the amount of resistance provided for each step 
was determined from Ecjuation [12]. Using these values the maxi- 
mum current of 1 am|x»re corres|x)nded to the minimum resistance 
of 40 ohms, while the mininmm current of 0.1 ampere corresponded 



\ 



lACOB M. BFITZ0I.A8S 285 

to the TnftTimiim reaistajice of 400 ohms. Between these limits the 
rise and fall of the mercury column produced by the variation of 
the head on the hi^-pressure side of the U-tube would vary the 
amount of resistance in the circuit in accordance with the hyperboUc 
curve shown on the left of Fig. 3, thereby regulating the amount of 



Fia. 4 Device i: 

current passing through the circuit in accordance with the para- 
bolic curve on the right of the figure. 

22 The operation of the elementary device was successful from 
the very start, but only as long as the contact chamber was kept 
free from water did the regulation of the current correspond with 
the variation of the head or the height of the mercury column. On 
the other hand, the instrument when equipped with an oil seal and 



286 ELECTRICAL MBASXJBINa OF FLUID FLOW 

connected to a pitot tube in the steam line could not be made to 
operate properly. When left imder pressure the oil would leak 
through the fiber plug and allow water to get into the contact cham- 
ber, which would immediately put the instrument out of order. 

23 To overcome this action the body of the instrument was 
extended to include also the resistance coils attached to the contact 
rods and the circuit completed through a plug connecting the in- 
ternal and external resistances together. This change eliminated the 
leakage of oil from the instrument, but still the water could not be 
kept out of the chamber for any length of time. In some cases the 
water would blow through the mercury as soon as the pressure was 
admitted to the instrument. It was thought at that time that a 
bypass valve connecting the static and dynamic tubes when the 
pressure is admitted to the instrument would eliminate this trouble. 
After many changes in the original design an instrument was made 
which for a short period of time was used as a steam-flow meter. In 
this instrument, besides the equalizing valve, an additional overflow 
chamber was provided to keep the water from reaching the con- 
ductors over the surface of the mercury. A terminal post replaced 
the connecting spark plug, thus allowing an adjustment of the posi- 
tion of the conductors with respect to the level of the mercury column. 
These additions, however, did not entirely eliminate the possibility 
of water coming in contact with the resistance elements of the device. 
With a uniform flow in the pipe the operation of the instrument 
would continue for some time, but when a slight disturbance of 
pressure occurred in the line it would cause the water to blow 
through the mercury into the contact chamber and this would imme- 
diately (lis(;ontinue its operation. 

24 Notwithstanding this objectionable feature, the conven- 
ience of the electric measurement of the flow and the fact that 
the instrument would function as long as there was no water in 
the contact chamber have encouraged further experimentirig for the 
elimination of defects. The problem was finally solved in the fall 
of 1917 by the addition of a mercury seal connected in parallel with 
the working !)ase of the instrument. The function of the mercuiy 
seal in this case is quite similar to that of a fuse plug in an electric 
circuit, with the added advantage that it is self-replacing. 

25 The principle of its operation is illustrated in Fig. 5, which 
shows a section of the meter body and seal chambers. It can be 
seen that the U-tube joining the two compartments of the seal will 
contain a colunm of mercury about one-half the height of the colimin 



JACOB u. sPiTzaiABS 287 

in the meter body. Under normal operation the mercury column 
in the seal acts in unison with the mercury column in the meter 
body and does not interfere with the proper transmission of the 
differential pressure in the meter. When, however, some disturb- 
ance of pressure occurs in the line sufficient to break the seal, the 
mercury spreads over the lai^er area of the compartment, equalises 



Fio. S Latbbi Ttfe or Author's Flow Metbk with Mkbcubt Seal 

the preasure in the two compartments in the same manner as would 
an autom&tic openmg of a bypass valve, and thus prevents the 
breaking of the higher column la the U-tube. As soon as the ab- 
normal differential pressure is released, the mercury drops back into 
its place and reestablishes the necessary seal between the two com- 
partments. In this model a large quantity of oil is trapped in the 



288 



ELECTRICAL MEASURING OP FLUID FLOW 



two compartments of the seal and in the meter body, eliminating 
the possibility of water blowing through the mercury and coming 
in contact with the resistance elements of the meter. 

26 In the latest type of the flow-measuring device the contact 
rods were changed from their former equal spacing above the zero 
level to spaces varying in height so as to give at each step equal in- 
crements of current representing equal amounts of flow. This was 
accomplished by gaging the length of the contact rods to follow the 
parabolic curve shown at the right in Fig. 3, which represents the 
solution of the equation h = Phma*. In the actual gaging of the 



TABLE 1 



h 

in. water 



0.45 
1.06 
2.06 
2.81 
3.56 
4.00 
4.56 
5.125 
6.875 
7.00 
7.56 
9.125 
9.875 
10.812 
11.25 
12.50 
13.56 
14.87 



DATA OBTAINED FROM TEST OF ELECTRIC 
FLOW-MEASURING DEVICE 
(^1 = 112 volte) 



Et 

volts 



I . 



40.15 
40.15 
40.15 
40.05 
40.05 
39.95 
39.95 
39.85 
39.85 
39.75 
39.65 
39.75 
39.75 
39.65 
39.55 
39.65 
39.55 
39.55 



/ 

amperes 



0.110 
0.170 
0.235 
0.260 
0.280 
0.305 
0.326 
0.347 
0.370 
0.392 
0.418 
0.434 
0.458 
0.480 
0.505 
0.523 
0.542 
0.566 



h 

in. water 



16.12 

17.43 

19.44 

21.00 

22.75 

24.00 

25.25 

27.00 

29.125 

30.00 

31.625 

34.06 

35.37 

37.00 

38.75 

40.75 

42.875 

44.75 



Et 

volts 



39.45 

39.45 

39.35 

39.25 

39.25 

39.25 

39.25 

39.25 

39.15 

39.05 

39.05 

39.05 

39.05. 

39.00 

38.00 

38.80 

38.70 

38.45 



/ 

amperes 



0.592 
0.613 
0.642 
0.666 
0.689 
0.713 
0.740 
0.765 
0.790 
0.814 
0.838 
0.860 
0.883 
0.905 
0.930 
0.053 
0.978 
1.000 



contact rods hmax. is tukcn as the distance between the zero level 
of the mercury and the end of the contact rod showing the maxi- 
mum flow. The succesvsivc heights of the rods for the given equal 
incrcrnonts of current arc detennined more conveniently by dif- 
ferentiating Equation [11], h = Phmax. The first differential, or 
dh = hmaz. (2 Idl 4- d/-), represents the respective increments of h 
corresponding to the given increments of /. The second differential, 
or iPh = 2hynax. dPj represents the resixjctive difference in the suc- 
cessive increments of h, from which it is noted that the distance 
between the successive contact rods is increased unifonnly by the 
constant quantity 2 /uux///*. 



i 



JACOB M. SPirzOLASS 



TESTING POH ACCUHACT 



27 The accuracy of a flow-measuring device is necessarily made 
up of two factors. One is the accuracy with which the device regis- 
ters the differential pressure equivalent to the flow in the pipe, and 
the other is the accuracy with which it will indicate or record this 
differential pressure or the equivalent units of flow. In the usual 
application of the flow meters, where the pitot tube, the venturi 



Fio. 6 CouPABiaoN op Cqbve Obtained tbom EktuATioN [llj wna Data 
PiMTtEn PBOU Table 1 

tube or the orifice plate is used for obtaining the differential pres- 
sure of the flow, the efficiencies of these devices have been determined 
by numerous tests and are at present well known. Their nature is 
such that a given flow will always produce the same effect under the 
same cooditions, since they do not possess any working parts to vary 
the relative effectiveness of their operation. On the other hand, the 
indicating or the recording elements of such devices may vary from 
time to time depending upon the condition of the moving parts in 



290 ELECTRICAL MEASURING OF FLUID FLOW 

these elements. It is necessary, therefore, to have convenient 
means for testing them in order to ascertain their accuracy at fre- 
quent inters'als. 

28 Table 1 gives the data of a typical test on the resistance ele- 
ment of the flow-measuring device, showing the relation between 
the differential column and the corresponding readings of the electric 
current. 

29 In Fig. 6 the points obtained from the test are indicated by 
the small circles, and for comparison a curve is shown gi\dng the 
theoretical relations according to the equation h = Phmas, In this 
test the differential pressure was obtained by varying the height 
of a water column connected to the dj^namic side of the meter. The 
electric current was supplied to the indicating instrument through 
a transformer, and the primary or line voltage was kept constant 
by varjing the field of the generator supplying the power, thus ap- 
proximating the actual condition of an average installation. The 
secondarj' voltage varied due to the transformer regulation, but the 
resistance element of the instrument is designed to compensate for 
such regulation, so that the indicated variations of current gave a 
fairly accurate measurement of the differential pressure. 

CONCLUSION 

30 The fact that the flow of fluids can be measured electrically 
has made passible many important installations where no other 
method could be employed. In one instance a large manufacturing 
concern had been contemplating for a long time the adoption of a 
system for measuring the amount of steam, air and water used by 
its various departments, but was hindered by the fact that the vari- 
ous lines were distributed over a wiile area and in some places were 
carried through sub-basements, where measuring de\'ices would be 
inaccessible; also much time and a large force of employees would 
be required to read the various instruments about the plant and to 
integrate the recording charts. As soon as the concern discovered 
that flow could be measured electrically, that the indicating instru- 
ments di<l not have to be located where the flow was to be measured, 
and that the integrating device was merely a watt-hour meter which 
integrated the How indep(Mnlently of the oilier instruments, a measur- 
ing system was instituted for all its products and many wasteful 
uses of power wen; thereby eliminated and an accurate distribution 
of costs estaMislMMJ throughout the factory. 



DISCUSSION ' 291 

31 The adaptability of the integrating feature to the electrical 
measurement of flow is of great importance since the readings from 
the watt-hour meter are more accurate than those taken from the 
recording ammeter and just as accurate as the instantaneous read- 
ings of the indicator. This feature therefore eliminates the necessity 
of planimetering the charts and insures accurate results for any 
variation of flow. 

32 When measuring the flow of steam generated by a battery 
of boilers the flow indicators are placed in front of each boiler, show- 
ing the momentary performance for the guidance of the fireman. 
At the same time, supplementary recorders connected electrically 
with the indicators are placed conveniently for the supervision of 
the chief operator. 

33 Recently the manufacturers of water gas adopted the use 
of low-pressure exhaust steam for gas generation, which created an 
urgent demand for a measuring device to operate intermittently, 
varying every few minutes from zero to maximum. After many 
imsatisfactory trials of mechanical devices the electrical method of 
flow measurement was adopted, as this made it possible to measiu'e 
successfully the steam required for the manufacture of water gas 
and resulted in a great economy. 

34 The main advantage, however, of the electrical method of 
flow measiirement is the accuracy with which the differential pres- 
sure is transmitted through a mercury column, which column is not 
hindered in its movements by any mechanism and is therefore free 
to attain the true level under all conditions of flow. Furthermore, 
the electrical instruments used to register the flow can be checked 
at any time without interfering with the operation or installation 
of the measuring device. 



DISCUSSION 

Herbert B. Reynolds (written). The flow meter described 
in this paper certainly has a great many advantages and should 
have a very wide field of application, due to its remote indicating 
features. 

As stated in the paper, the resistances and contact points are so 
arranged that the current flowing through the ammeter varies di- 
rectly as the fluid flow. If direct current is used, the deflection of 
the indicating needle will be in direct proportion to the fluid flow. 



292 ELECTRICAL MEASURINQ OF FLUID FLOW 

In other words, it would be possible to provide a meter with a uni- 
form scale, which would be more desirable than the scales found on 
most flow meters. I am under the impression that alternating cur- 
rent is used exclusively for the operation of these meters, which 
means that, although the current strength varies as the Uquid flow, 
the indications vary as the square of the Uquid flow, due to the 
well-known scale characteristic of all alternating-current meters. 
However, it seems to me that a uniform deflection could be obtained 
by simply adjusting the resistances and contacts, as is done in ob- 
taining a straight-line law between the Hquid flow and current flow. 
I would be interested to know if the author has made any attempt 
to obtain a uniform scale in this manner. 

In a great many cases it is desirable to have an indication of 
the total flow, through two or more pipe lines on one dial. It seems 
to me that this instrument could be used for this purpose, by plac- 
ing a mercury column, with its resistances and contact points, on 
each pipe line. The terminals of these mercury columns could then 
be connected in parallel or in series, with the indicating meter in the 
main circuit. This probably has been considered and found im- 
practicable; however, if a scheme like this could be worked out, 
it would broaden the field of the instrument. 

Geo. F. Gebhardt (written). During the past 20 years I 
have had exceptional opportunity for testing practically all types of 
steam flow meters exploited in this country, both in the laboratory 
and under service conditions. Most of these devices gave con- 
sistent results in the laboratory where the various factors entering 
into the calculation of the weight of flow were known, but under 
service conditions the indicators of the dial and chart frequently 
departed considerably from condenser weights. 

A study of the discrepancies showed that the error laid chiefly 
in the "factor" for converting velocity-pressure variations to weights, 
rather than in any inherent defect of the meter. These "factors," 
as established by the manufacturers, are based upon what are as- 
sumed to be service conditions for the proposed installation, and 
naturally failure to predict the true conditions will result in error. 
In several instances where the meter has been discarded as unrdia* 
ble it was ascertained that the factors were merely calculated and 
that no calibration tests of any kind had been made. 

The steam flow meter has been wonderfully developed during 
the last decade, and where the rate of flow does not fluctuate widely 



and the steam conditions are fairly constant it is a dependable and 
accurate means of meaauring the weight of the flow. However, 
where the rate of flow fluctuates rapidly and there is considerable 
variation in the pressure and quality of the steam, the indicated 
readings are generally not in accordance with the actual weights 
flowing. Proper calibration under service conditions greatly reduces 
the error but few plants are equipped for tiiis purpose. 

The many advantages of the electrical method of control over 
the mechanically operated mechanism are enumerated in the au- 



Fio. 7. Differences betmebh Sdccessive Square Roots or Heads 
AND Ampere Reading fob 35 Inckeuents of Pbessorb 

thor's paper. My own experience has shown the electrical control 
to be accurate and dependable. Meters operating on this principle, 
however, are limited to plants supplied with current and are subject 
to error on account of voltage variation. 

I would like to ask the author if he has had any success in apply- 
ing his electrically-controlled meter to pulsating flow, as in connec- 
tion with high-^9peed reciprocating engines or slow-speed reciprocat- 
ing pumps. 

Geo. H. Babrus (written). The paper contains a table giving 
the amounts of current which were shown by this instrument under 
various increasing heads, starting with 0.45 in. water pressure and 



294 



ELECTRICAL BfEASURINa OF FLUID FLOW 



rising by 35 increments of yar3ring amounts to a total of 44.75 in. 
water pressure. It is interesting to analyze this table and see vriih 
what degree of sensitiveness or accuracy the indicaHons of current 
respond to the gradiial increase of head. The analysis can best be made 
by taking the successive increments of head and comparing them 
with the corresponding increments of current. To be comparative 
the square-roots of the heads are first obtained and the differences of 
the successive square-roots taken. Then the differences of succes- 
sive currents are found, using for convenience the column of amperes. 
These two sets of differences are as follows: 





Differences 


EHfferences 




Differenees 


Differenees 




between 


between 




between 


IwCween 


Number 


Successive 

Square Roots 

of Heads, 

In. Water 


Successive 
Ampere 

Readings, 
Amperes 


Number 


Square Roots 
of Heads, 
In. Water 


Suoeesshre 
Rfdingi, 


1 


0.36 


0.06 


19 


0.15 


0.21 


2 


0.40 


0.065 


20 


0.24 


0.29 


3 


0.2.> 


0.025 


21 


0.17 


0.24 


4 


0.21 


02 


22 


0.19 


0.23 


5 


11 


025 


23 


0.13 


0.24 


ti 


0.14 


21 


24 


0.12 


0.27 


7 


12 


0.21 


25 


0.18 


0.25 


8 


o.in 


0.23 


26 


0.20 


0.25 


9 


0.22 


22 


27 


0.06 


0.24 


10 


11 


0.26 


28 


0.14 


0.24 


11 


0.27 


0.16 


29 


0.22 


0.22 


12 


0.12 


0.24 


30 


0.11 


0.22 


13 


0.15 


0.22 


31 


0.14 


0.28 


14 


0.06 


0.25 


32 


0.14 


0.25 


15 


0.18 


0.18 


33 


0.15 


0.23 


16 


0.15 


0.19 


34 


0.17 


0.25 


17 


0.18 


24 


35 


0.14 


0.22 


18 


0.16 


0.26 

1 









These differences are plotted on the chart of Fig. 7 in which the 
upper record is that of the ampere differences and the lower <xiet 
the differences of head computed in the manner stated. It will be 
seen at a glance that when viewed in this way there is great die- 
parity betwei^n the two records. In only a few instances do the 
variations in current respond immediately and proportionately to 
those of the head. No doubt the gaps between the contact pttate 
of the instrument are responsible to some extent for these conditione 
because no change of current would be hkely to occur while the 
head is clianging fn)m one point to the next, but it is not clear that 
other caiLses are absent. It would be interesting to have further 



DISCUSSION 295 

information regarding this matter and it is suggested that such in- 
formation might be obtained by taking a series of electrical readings 
corresponding to much smaller increments in head, say one-quarter 
or one-half an inch, throughout the whole scale. The comparison 
might also be extended by getting a series of readings going down 
the scale, and demonstrate whether contact is made or broken at 
the same reading of head whether rising or falling. 

The records shown on the chart bring out the fact that the 
instrument is not scientifically accurate. Nevertheless the state- 
ments offered in Par. 30 of the paper indicate that its utiUty as a 
practical device for the electrical measurement of the flow of fluids 
has been well proved in actual use. 

A. H. Anderson (written). I have had a wide experience with 
fluid meters since the old Sargent steam meter which accomplished 
something which is never attempted in present day meters. It 
compensated automatically for the change of steam pressure within 
a wide range. The Sargent meter was applicable only in sizes of 
pipe under 6 in. diameter and I do not know that it is used today 
in any size. Then came the Gebhardt steam meter, patented in 
1905 by Prof. Geo. F. Gebhardt, of the Armour Institute of Tech- 
nology, which was the fore-runner of the Republic meter. 

No mention is made by the author of the error introduced into 
his calculations by change of steam pressure or quality of steam. It 
is also essential that the pipes connecting the meter to the pitot tube 
be kept full with water. 

It seems to me that the value of the meter depends upon the 
selection of the correct coeflScient for the pitot tube or orifice. 

I have seen many fluid meters in operation, but in only one 
instance has any provision been made to check the accuracy of the 
installation by actual weight. Where such a check has been pro- 
vided, meter inaccuracies have been detected in time for early cor- 
rection. Fluid meters today are examples of marvelous ingenuity, 
but their presence must not lull us into false security. Every plant 
using fluid meters should also have a weighing device in the feed 
line for occasional use. 

One example of the utility of the Republic meter is on a boiler 
in the Union Stock Yards, Chicago. Normally the boiler is de- 
veloping 750 hp. but let the inspection door be opened for about one 
minute and the fluid meter will begin to drop until the horsepower 
is about 500 and after the door is closed, several minutes elapse before 



296 ELECTRICAL MEASUBINQ OF FLUID FLOW 

the meter returns to normal. The variation, of course, is due to 
the chilling effect of the excess air. 

William B. Fulton asked if the author had had any experi- 
ence with his meter in connection with thickened fluids, such as 
paper-mill stock containing 3 or 4 per cent stock, the remainder 
being water. The author suggested the use of a pitot tube with 
large openings. The change of resistance, due to the thickened 
fluid, would be recorded by the instrument, if the connection did 
not clog, which would not occur if the openings were large enough. 
The flow could be measured positively. The instrument would give 
the velocity from which could be computed the volume. 

Austin R. Dodge said that the electric type of flow meter, as 
the author had stated, could be easily checked at any time but 
this was true of other types of meters and did not depend upon the 
electrical method of operation. 

Also it seemed to him that with an automatic integrating at- 
tachment it was quite as simple to obtain the total flow as with 
the watt-hour meter. 

E. G. Bailey said that in metering steam there was a minimum 
of four distinct fluids to deal with: steam, the condensed water in 
the connecting pipes, air, which invariably separated and gave a 
false head, and the mercury in the measuring device. In most 
practical installations, to keep each of these fluids separate and in 
its proper place so as to prevent a false head was a big part of the 
battle, and anyone who could add two other fluids, oil and electric- 
ity, and keep them in their proper places certainly deserved a great 
deal of credit. Looking at it in another way, the fluid meter problem 
involved two factors, a means of obtaining a pressure difference 
and a means of measuring this pressure difference. The author 
has made 50 per cent more of a problem of it in using this pressure 
difference to control an electric meter electricity, adding a third and 
completely distinct step. 

There was a fifth mobile substance which had been mentioned 
in the discussion of paper stock, namely scale, sediment and other 
matter which invariably flowed through steam lines. This should 
be considered carefully in view of its possible effect on any type of 
flow meter. 

Mr. Bailey asked the author how near was the specific gravity 
of the oil used in the oil seal to that of the water and how was thia 



y 



DISCUSSION 297 

taken care of in the calibration of the instrument to prevent a false 
head in the oil seal. 

Wm. B. Gregory said that while he had had considerable ex- 
perience with pitot tubes and other means of measuring water he 
had never used an instrument of the type mentioned in the paper. 
He expected to use such a meter during the summer in tests at a 
large sugar refinery and wished to know in how far the results could 
be depended upon and what the probable error in operation would be. 

The Author. With reference to Mr. Reynolds' discussion, 
the purpose of the straight-line law between the liquid flow and 
current flow was necessary mainly for enabling the watt-hour meter 
to integrate the total current and thereby represent the total flow. 
It results in a uniform deflection on direct-current instruments, 
which are used sometimes, while on alternating current instruments, 
the deflection is not uniform. 

The measurement of the total flow through two or more pipe 
lines on one dial has been found possible in actual practice in many 
cases. In several large boiler installations, there are two outlets 
to each boiler and one watt-hour meter totalizes the flow of both 
outlets. 

With reference to the other written discussion, Mr. Barrus 
probably appreciates the fact that the accuracy of the meter lies 
in the cumulative indications and not in the difference between 
one contact point and the other. Mr. Barrus, in his last paragraph, 
stated that the records shown on the chart bring out the fact that the 
instrument is not scientifically accurate. This is true. When 
measuring steam with a pitot tube or an orifice plate, which are only 
accurate within two per cent at the best working conditions, it would 
be absolutely out of the question to look for a scientifically accurate 
instrument. All the errors found by Mr. Barrus are within one per 
cent, which is absolutely insignificant for commercial purposes. 

The author, in answer to a question by Alan E. Flowers re- 
garding the relation between the average and maximum flow as 
measured by the pitot tube replied that the question opened an 
entirely different field and one which he would not, at that time, 
go into. In reply to a second question by Mr. Flowers about tests 
which would show the relation between the total flow and the indi- 
cation of the meter the author said that he believed that Professor 
Gebhardt had answered it by stating that all meters were accurate 
enough in the laboratory under proper conditions of operations and 



298 ELECTRICAL MEASURING OF FLUID FLOW 

upon tests made under these conditions only could conclusions be 
based. For accuracy, the customer must test his meter in his own 
plant and under the actual conditions which prevail, thus estab- 
lishing a "factor" for its use. Tests of the meter, showing its ac- 
curacy, could be supplied in great quantity. 

With regard to the question of variation in voltage proposed 
by Professor Gebhardt, the author stated that the meters were 
generally installed in plants where voltage regulation was normally 
excellent and that they were connected with the main switchboard 
and not to a line where variations would be likely to occur. Further, 
in transforming the current, two transformers were used. 

With reference to the error introduced by pulsating flow, which 
in some instruments might be as high as 60 per cent, the author 
stated that the only way to overcome the difficulty was to calibrate 
the meter for the actual conditions. 

With reference to the device on the Sargent meter which com- 
pensated for pressure, mentioned by Mr. Anderson, the author said 
that he was attempting the design of such a device which would be 
perfected upon the completion of some research work. He thought 
that to buy apparatus for testing a flow meter was to spend more 
money than would be warranted. 

The author did not agree with Mr. Dodge that a mechanical 
device could be as accurate as a watt-hour meter. 



No. 1698 

CRUDE-OIL MOTORS VS. STEAM ENGINES 

IN MARINE PRACTICE 

By J, W. MoBTON, Philadelphia, Pa. 
Junior Member of the Society 

This paper is a discussion of the various Jadors to be considered in choosing the 
form of motive power for war vessels and cargo ships; also a presentation of the 
advantages and disadvantages as compared to steam engines of high- and low-powered 
oil motors of both the constant-pressure and constantrwdume type. The relative 
values of four-stroke cyde, two-stroke cycles double-acting ^ and horizontalf oU motors 
are also compared and some details of their construction such as lubrication sys- 
temSt piston cooling and scavenging pumps are analyzed. 

^HE number of articles which have appeared of late describing 
the perfonnances of the so-called motorships would seem to 
indicate that for marine purposes the crude-oil motor is rapidly 
replacing the steam engine. The chief reason for this is not to be 
found, as might be expected, in the lower operating cost of the 
crude-oU motor, but is rather due to other factors which, when con- 
sidered, lead to the conclusion that the crude-oil motor is probably 
the most economical prime mover of today. 

2 Perhaps the correctness of this statement can best be shown 
by comparing a steamship with a motorship. In the case of a war 
vessel, for instance, there are eight factors which ought to be con- 
sidered in choosing the form of motive power, namely weight, space 
occupied, radius of operation, preparedness, crew necessary, fuel, 
auxiliary equipment, and cost. 

3 The weight of the power plant is of great importance, for a 
saving therein can be utilized to increase the armor of the ship. If, 
for example, a warship of 3500 tons displacement be equipped with 
crude-oil motors there will result a saving as indicated by Table 1, 
which is based on a single steam engine developing 4400 hp. and 3 
units, of 1600 hp. each, of a standard make of crude-oil motor. 



Presented at the Spring Meeting, Detroit, June 1919, of The American 
SociBTT OF Mechanical Engineers. 

299 



300 



MABINE CRUDE-OIL MOTOBS 



4 In the case cited about the same space would be required for 
both motor plant and steam plant, but the total horsepower could 
just as well be developed by two units of slightly larger size, and if 
this were done approximately one-third the space would be saved. 
Moreover, if oil motors were used the space occupied by the coal 
bunkers could be partially released, as crude oil can be stored in 
tanks under the engine floor and below the protective line. 

5 It is obvious that a warship must possess a great cruising 
radius and if weight for weight of engines be considered then the 

TABLE 1 COMPARATIVE WEIGHTS OF POWER PLANTS IN STEAMSHIPS 

AND MOTORSmPS 



Steamship 



Engine, tons 

Per cent of displacement. . . 
Fuel, tons 

Per cent of displacement. . . 
Crew, provisions, water, tons. 

Per cent of displacement. . . 



335 



246 



01 



0.6 



7.0 



2.6 



ToUls. 



672 





ffship 


SftTinfliin 


Motorship 


Mote 


weight. 








tons 


Pweent 


264 




71 


• » • 




7.6 


• • • 


S.l 


86 




160 


• • • 




2.6 


• • • 


4.6 


82 




9 


• • • 




2.3 


• • • 


0.3 


432 


1 


240 


6.0 



R 



ratio R of the cruising radii of a steamship and a motorship may be 
expressed by the formula 

VcXHoXWc 

VoXHcX Wo 
where 

Ve = cu ft. occupied per ton of coal = 45 
Vo = cu. ft. occupied per ton of oil ■= 35 
He = B.t.u. per lb. of coal = 14,000 

Ho = B.t.u. per lb. of oil = 18,500 

We = lb. coal consumed per c.hp-hr. 
Wo = lb. fuel oil consumed per e.hp-hr. 

In the case of the warship under consideration if We l>c taken as 
1.75 and Wo as 0.4, then the ratio will be 

45 X 18500 X 1.75 



R 



-5.7 



35 X 14000 X 0.4 

The cniii5ing radius will, of course, vary with the type of engine and 
ship, but nevertheless the average ratio, as expressed above, may 
be taken at least as 1 to 5 in favor of the crude-oil motors. 



J. W. MOBTON 301 

6 A motOTship in sharp contrast to the steamship is always 
ready for action, as no time is lost in getting up steam. Moreover, 
fuel is consumed only when the motors are running, and the motor- 
ship is capable of maintaining full speed as long as there is a supply 
of fuel. 

7 As there is no need for stokers on a motorship, the crew can 
be decreased about 10 per cent, and this of course permits of a corre- 
sponding saving in provision, water storage and quarters. 

8 In the case of a steam-driven warship the smoke from the 
stack frequently betrays the location of the ship, for it is well known 
that a vessel can be easily located by its smoke long before the masts 
or hulls are in sight. Another drawback to the steam-driven warship 
Ues in the fact that the smoke covers the ship with a filnn of soot 
which, getting into vital parts of auxiliary machinery, necessitates 
frequent cleaning and imnecessary wear and tear. On the other 
hand, when oil is used there is no smoke, the handling of the oil is 
both cleaner and easier, and the life of machinery and equipment is 
greatly increased. 

9 When a change is made from steam to oil the auxiliary ma- 
chinery is usually electrically driven, as this has been found to be 
most satisfactory. The dynamos are usually driven by separate 
oU motors. Hot water for heating and sanitary purposes is obtained 
from a small oil-fired boiler and while the first costs of such installa- 
tions are high, they are in time offset by the lower operating costs. 

10 Prior to the war the first cost of a crude-oil motor plant was 
approximately twice that of a steam plant. This was partly offset, 
however, by the savings m operating expenses, under which item 
comes cost of fuel, labor and supplies. The cost of fuel naturally 
depends upon the prevailing market for coal and oil. In regard to 
maintenance there is practically no information available as the 
crude-oil motor is still in its infancy so far as this country is con- 
cerned. 

CARGO SHIPS 

11 Many of the preceding statements concerning the installa- 
tion of crude-oil motors on battleships can also be properly applied 
to cargo ships. During the last four years a nmnber of such vessels 
have been equipped with crude-oil motors of both the two-stroke 
and four-stroke cycle type, an(J while engineers are not agreed as to 
the best type, the four-stroke cycle is to be preferred. 

12 One of the most successful installations of this kind is that 



304 



MABINE CRUDE-OIL MOTORS 



M 



2 
o 

o 

< 
u 

O 






o 

CI 






ii 



c 

i 







|1 









^O "* ».0 Q O -^ 
» Q » O C O 
—• « — > W C>i C^J 



^ O O Q 
O ?i "^ ^ 
C^ C^ 04 W 



O C9 









_ lO lO 






•xa 'cI '\i 






•inui 'asjoijs 



OQQQQOOOQQQQQ 



•lUUI 'OJOQ 



8QOOOOOOO 



CO 5 ?5 
O to CD 



o 



j(>)<>ni qotTd UT 

KJOpUl^Xj JOOJS,* 



OOOOOOOCaOOXOOOOO«OCO«0«0«0«0(0 



I '^^ ^> <^< -^ ^ <^> <^< <^< ^ CO ro 04 c) cc CO CO CO C9 CO 



04 Ol oi ro ?0 iN Oi 



K^otn^ til 



^ ''"'V 95P95i5i5?228S5<28S5?8t2 8 

*M»<'->a I '::i g ^, 2 g ^ :4 ^:i 9 a 9 ^^ ;^ ^ 9 ,^ g ^ ^ 



'im;»»y 



^^O^:OQ^0O!XOOOec^5OOOOOO 

„ -^^ — O i't^ — -r — -"^ '" ''i '- -- ' - ' • '^ '5 T *^ 
1-^ I.-^ I- 1^ o i": '-t ••': I': ••■: o ••': ••■: ••': o o »c •-'t ».': 






ocoooccoococooooooo 

"^^Ti .... iilTllll 



r, d c'l i C cj .j: rj .^ 
5t '^ jR -r -r re — ."^ r^ 






^ ^ *NJ ^1 ^.k t^ 1^ 1^ (^ 1^ 

•M »-• f'^ ''^ ^^ ^m ^1 ^1 ^ ^ 



.^ .^ ^ ^ ^ 



X. ui >^. ;!: u: X. ;- z', ;^ u; u; V', z', ;i; u; L^; ;^ ;- W 






«j .^j -^ -s« ^ -^ ;.^ -* -♦. -t> -* -f 1^ O »" '^ •'J l*^ «0 






X 



: -'A \j 

 ^' •* u 

. y. 7. — 7. 



< 



r. 



C 

5 



»:, "• >«» »" "~ 






c C 






*-.■:: s~ si Sa 



< 



o o _ o 

^^ v,^ V,, V- \^ N^ • ^ N-« V^ V,.^ V^ V^ Vj, N^ ^ ^ O •-• 

^|;^ « , —■, -i', ps\ — rt ""  -"\ — N « Ws 'J-, "-fs .... 




J. W. MOBTON 303 

the motorshipy and transmitting power through a single slow-run- 
ning screw as in the case of the steamship. The propeller on the 
steamship has a diameter of 17 ft., while those of the motorship are 
only 10 ft. The ratio of indicated to eflfective horsepower is there- 
fore the same for both ships and the total efficiency of engines and 
propellers is also about the same. 

13 In order to correspond to the motorship in horsepower the 
steamship would be obliged to carry an extra boiler. This would 
increase its weight 570 tons, whereas the weight of the machinery on 
the motorship is only 470 tons. The length of the engine room of 
the motorship is 41 ft. while the length of engine room and necessary 
boilers in the steamship is 66 ft., and this despite the fact that the 
steamship has less horsepower. 

14 The fuel consumption during the trial trip of the Suecia was 
0.294S lb. per i. hp., or 0.3685 lb. taking the mechanical efficiency at 
80 per cent, a very satisfactory value. Table 2 a£fords a compari- 
son of this vessel with other motorships. 

MOTOR SIZES 

15 The horsepower of a crude-oil motor, whether of the con- 
stant-volume or constant-pressure type, is somewhat limited by 
practical considerations of construction. At the present time the 
maximimi size is 2500 hp. per shaft and from six to eight cylinders 
for the two- and four-stroke cycle constant-pressure type, although 
some experimental engines have been built as large as 6000 hp. 
This is in sharp contrast, however, to the power developed by steam 
which runs as high as 15,000 hp. for the steam engine and 35,000 hp. 
for the steam turbine. 

16 Crude-oil motors of small horsepower are usually of the 
high-pressure type, medium, so-called semi-Diesel or hot-bulb 
type and two-cycle constant-volume type. Prominent among such 
semi-Diesel engines are the Bolinder, Scandia, Avance, Alliance, 
Tuxham, Gidion and Holm. 

17 It has been proved both by experiment and practice that 
the constant-volume, high-power, single-cylinder type of motor is 
unsatisfactory, and chiefly because of the fact that the enormous 
heat generated at the pistons during the explosion must travel a 
great distance before it is transferred to the water jacket, and as 
a result the central part of the piston is heated to such an extent 
that preignition frequently occurs. Furthermore, the sudden rise 



304 



MABmE CBUDE-OIL MOTOBS 



S 

IS 

o 



o 

o 

< 
u 



OB 









CI 5 

< 



.9 






umi jocI Ria^.-nn 
'poods ua)u!cl 



lO ^ «0 O O "f •}« 

» p 00 o o 6 p 

i-< W iM CM c* ci w 



Fi •<*• 



li 



01 cs 



}0 OJ^CH 



Ift IQ 






•ra 'd 'XI 



Q Q O "O 10 O O O Q »0 lO Q Q •<> «0 »Q »0 P >0 



■nrai '03{ox;g 



i» o ac 3C t'* i^ t« ^ o o b- r* 



•urai 'ojog 



t.i iTS >0 O O i.'b k-^ t^ « CO lO »Q «D <o 



O 

3 



j())f>Tn Tfano ut | ooaoooooxooooootototo<o<D<DO<DO 



8SSS8S 

cs CO n cc ^ 70 



, 22222S22299 
?5M?ir:?7?JW?i'reow 



]),»0<ly 



»o o 



'f ^ -» " r» v-* ' •♦ ^ "* f '^«' ^:f v-f V* v^ ^ V "^ ^ O •"• 



M>fi.iMp :..i> : :-: :; :^ rX r: 12 Vi :-: i~ ?i ?i ^ 1^ ?i i g g 



i 



)0 



''i»^i"a I u"- 2 V.1 9 9 I?, '::i ;a s 9 9 15 ^ 9 ^ ^ ^ 






•Ul-'lj 



r?cc^00r^or5000con000000 
1.^ •-'i »": '': '"t •': '"t «■: •. •■: o ».■; •" •.•; •■. •-'5 ».. »c »o 



MI1-"JI 



occccccoccococooo 

ri 



• i -^ ri i c ^'i 'J: ri i'; c o m ri i i o •- 

•^ I ^ ^ ""^ ^" ^ ^r ^ >*- ^^ ^^ >•» «ap ^^ ^" • f • I 

^^^t^^-?^'r^:^^*^t^:■-r-r^:^Tf'r*r 



55 



v*' ^if O O s ^ 




>. J: X. ui J! >'. il^ ■/: :5 ;Ii :S -'i >< ii; X :!; ;ii ;j W 



> 



• 



.i.t \j.-»' 
I 1 ]• '.Mul J»:-»_\ 



/- 



X 



■> rj :? r- 






I" t*; ii I"; f0 

=^ cr. ?. a 31 



! : ^ ^ 



.- -« -^ 



,.,_,.<::: ^ u: 



? d 



> £ "s "a : 5 
S ^ r -a ft 



c 






I 

I 
1 



J. W. MOBTON 305 

in pressure sets up detrimental vibrating stresses in the material, 
and it is also exceedingly difficult to obtain a proper mixture of air 
and oil, rapid ignition and complete combustion and exhaust. 

18 On the other hand, the constant-pressure oil motor of high 
power must have niunerous cylinders all of the same size and the 
high pressure, which lasts approximately 10 per cent of the stroke, 
necessitates a careful design of the crank mechanism. This applies 
especially to the crankshaft, which is both difficult and costly to pro- 
duce. Moreover, in motors with only a few cylinders the flywheel 
must be heavy, large and expensive if a great variation in speed is 
to be cared for. Difficulties have also been encountered in keeping 
the piston leak-proof — if it fits too tightly the result is a large 
frictional loss, and if the clearance is too great there will be a loss 
of pressure. The greater the piston area the more effective must 
be the cooling system as the surface exposed per poimd of working 
medium decreases as the cylinder dimensions increase. This 
accounts for the lower thermal efficiency and the horsepower available 
per imit of piston surface with large cylinders. 

LOW-POWER MOTORS 

19 Advantages, The efficiency of an oil motor is greater than 
that of a steam engine, especially in small imits, and furthermore 
it decreases less with increasing load. The fuel consumption of an 
oil motor is independent of the attendant's skill, as it shows the 
same value in daily work as on a test stand. With a steam engine, 
however, the fuel consimiption is dependent upon the attendant 
and especially the stokers, who must be experienced. In steam 
plants the boilers and pipes must be able to withstand high pres- 
sures and temperatures, but with oil motors a high pressure exists 
only in the cylinder and a few small pipes. Moreover, the oil motor 
is simpler in construction, and the weight and space occupied con- 
siderably less. It is of interest to note in this connection that the 
steam turbines in torpedo boats weigh about 45 lb. per b.hp. and 
aeroplane motors but a little over 1 lb. per b.hp. 

20 The small oil motor has other advantages over the steam 
engine and these can be briefly enumerated as follows: Starting 
takes only a few seconds; fuel is consumed only during operation; 
combustion in the cylinders can take place iminterruptedly for a 
long period of time; the engines require little attention and the oil 
tanks can be replenished more easily than coal bimkers can be 
filled and without the dirt accompanying the latter procedure. 



306 MARINE CRUDE-OIL MOTORS 

21 Disadvantages. It must not be thought that oil motors have 
no disadvantages, for there are some, chief among which may be 
mentioned the following: Fuel oil is more expensive than coal and 
is not available in all ports; furthermore, lubricating-oil consump- 
tion is greater than in the case of steam engines as one must deal 
with higher temperatures and, as a rule, higher pressures. The oper- 
ation is also more irregular than that of a steam engine, as oil motors 
cannot be run as slowly as is desirable when reversing from ahead 
to astern. The causes of interruptions (especially with electrical 
ignition) are often diflScult to locate; the motor frequently works 
noisily with smoking and malodorous exhaust; the dangers due to 
fire and explosion are greater, and the life of the machine is less. 

HIGH-POWER MOTORS 

22 Advantages, The statements made in regard to the low- 
power motor are also true to a certain extent in the case of the 
high-power motor, but in addition the following items should be 
noted: With larger output the economy of a power plant becomes 
of greater importance. The low efficiency of most steam plants is 
due to the fact that the heat generated must travel through heavy 
boiler plates, and that there is a great loss of heat through the ex- 
haust, and through condensation of steam. Furthermore, the steam 
plant operates with a small temperature difference, and to increase 
this difference is exceedingly difficult because it is very desirable 
to work with higher temperature limits. With saturated steam the 
pressure soon increases beyond a practical value and with super- 
heated steam the walls of the superheater soon reach their limit, 
as they should never be allowed to become red hot. On the other 
hand, in the oil motor the heat formation and utiUzation take place 
in the working cylinders, and the higher temperature limit is there- 
fore the same as the temperature of combustion. The value of the 
lower temperature limit is greater than in a case of a steam plant, 
but the temperature difference of the process is higher. 

23 Another advantage of the oil motor is due to the ease with 
which the air supply necessary for complete combustion is regulated 
and since the point of ignition is practically constant, there results 
complete combustion. On account of the great heat which exists in 
a l)oiler room it is very difficult for the stokers to maintain normal 
pressure in a steam boiler; the coal consumption is accordingly 
increased by reason of careless firing but the speed of the vessel 
decreases. On the other hand, in the case of the motorship the 



J. W. HOBTON 307 

reverse holds true. If the fuel oil becomes warmer it is more easily 
vaporizedi with the result that fuel consiunption decreases while 
the speed of the vessel increases. 

24 Since fuel oil can be stored in tanks a considerable saving 
can be made in space, more oil can be carried and a greater cruis- 
ing radius thus afforded. The weight of a slow-running four-cycle 
stroke oil motor for cargo ships is about the same as that of a steam 
plant but far less space is required. The horsepower and speed are 
about the same for a twin-screw motorship as for a single-screw 
steamship, but as the former's screw diameters are smaller their 
eflSciency is greater and the vessel can thus be more easily maneuv- 
ered. The cylinders of a motorship are all similar, and since they 
work independent of each other, one or more of them can be put 
out of service and the motor still continue to operate. 

25 Steam boilers and their fittings require as a rule a consider- 
able outlay for repairs and maintenance; also the sides and bottom 
plates of the hull which are nearest to the coal bankers are likely to 
corrode because of the sulphur content of the coal, and in addition 
sea water in the ballast tanks may also cause corrosion of the plates. 
On the other hand, a motorship has no boilers and the fuel is stored 
in bottom tanks, and this very fact is a great advantage since the 
tanks are protected by the oil. 

26 Disddvantdges, In the case of a single-cylinder oil motor the 
torque acting on a single shaft is limited, but everything else being 
equal, such as shape and line of hulls, propellers, r.p.m. and b.hp., 
the i.hp. of the motorship will be greater than that of the steamship. 
Since the reciprocating parts in a crude-oil motor weigh more than 
those of a steam engine the balancing of the masses is more diflScult 
and the shifting forces greater. At light loads the pistons in a large 
motor are not tight enough to maintain high compression, and since 
the combustion space is not sufficiently warm for positive self -igni- 
tion, large motors can only be run when under light loads at approxi- 
mately one-third to one-fourth normal speed. Furthermore, the heat 
transmitted to the cooling water per unit area of the cyUnder is 
about two and one-half times as great in the combustion chamber 
as in the cylinder walls, and accordingly the heat drop in the cylinder 
wall is very considerable and tends to introduce stresses in the 
material. If the limits of elasticity are exceeded these stresses will 
be transmitted to distant parts. The material is also subjected to 
high stresses as a result of rapidly repeated power impulses, high 
temperatures and great pressure in the cylinder. 



308 MARINE CRUDE-OIL MOTORS 

CONSTANT-PRESSURE VS. CONSTANT-VOLUME MOTORS 

27 Advantages. The constant-pressure oil motor consumes less 
fuel than the constant-volume type, and this is due to the fact that 
it has a greater compression and a higher expansion ratio. Further- 
more, the fuel oil is introduced into the combustion chamber in its 
natural condition and is, therefore, more completely vaporized, with 
the result that combustion is more complete. A cheaper grade of 
oil can, therefore, be used. The mean eflfective pressure is also 
greater and therefore greater output can be obtained and preigni- 
tion practically excluded. 

28 Disadvantages. The constant-pressure oil motor also has 
its disadvantages. In the first place this type of motor costs more 
to produce than the constant-volume type, and in addition an air 
compressor is necessary. The motor itself operates at high pres- 
sures and temperatures and the air compressor necessitates tanks 
for storing the air for starting and running purposes. 

CONSTANT-VOLUME MOTORS 

29 Low-pressure Type. The low-pressure oil motor is perhaps 
the simplest prime mover of its kind and is especially adopted for 
use in small boats where the operators are usually unskiUed. The 
chief advantages of this type of motor are its low first cost and 
small operating expenses. Furthermore, the low temperature in the 
cylinder and the low explosive pressure, together with its simplicity 
of operation make it a motor of great durabiUty. This type, how- 
ever, has some disadvantages and chief among them are its small 
power output, large cylinder dimensions, excessive weight and the 
long time required for starting. 

30 MediuMy or so-called Hoi-Bvlb Type. The so-called hot- 
bulb motor operates at a medium pressure with or without water 
injection. ^Vllile the tendency today is to avoid water injectioRi 
if tlic hot-bulb motor be nevertheless operated with water injection 
its fiKj] consumption will be comparatively low; the temperature of 
inlet air low; overheating of cylinder and piston will seldom occur; 
the cylinder will remain clean for a longer period, and the piston 
rin^s will not gum or stick. The water tank, however, adds 
weight, and considerably reduces cargo space. Furthermore despite 
careful attention, corrosion and rust will rapidly destroy the motor, 
especially if sea water is used. The most serious objection, how- 



k 



J. W. MORTON 309 

ever, lies in the fact that the amount of water injected must be 
regulated according to load, and this necessitates constant attention. 

31 On the other hand, if the medium-pressure motor is 
operated without water injection, the cost of maintenance is low, 
the mechanism construction simple and Uttle attention is necessary. 
Its chief operating difficulty lies in the inabiUty to properly time the 
supply of fuel oil as its injection against the highly compressed air 
is exceedingly difficult in the short time allowed before ignition 
takes place. 

32 High-pressure Type. The advantages of high-pressure 
motors with self-ignition and injection of fuel oil during the high- 
compression cycle are practically the same as those of the 
medium-pressure motor without water injection. High-pressure 
motors, however, are usually operated with an oil chamber and such 
a type is very economical, has a high mean effective pressiu^, its 
cylinders are smaU, its weight low, and it can be started without 
any preliminary operations. Furthermore, since the fuel oil is not 
injected against a high pressure the oil pumps can be of simple 
design and are consequently easily maintained. Like all other 
motors, however, this type has its disadvantages — chiefly the 
fault of design and construction rather than operation. If the size 
of the oil chambers or atomizer holes be improperly proportioned, 
fuel-oil consumption will greatly increase. Since this type of motor 
cannot be started by hand, pressure air is necessary and therefore, 
the motor must be manufactured with great care. This, of course, 
means a higher initial cost. In operation, the ignition and com- 
bustion cannot always be controlled and if inefficiently operated, 
preignition will occur. 

FOUR-STROKE-CYCLE MOTORS 

33 Advantages. This type of motor is particularly well adopted 
to high-speed work since it has a separate suction and exhaust stroke 
and thus the cylinder is filled each time with a full and new charge. 
Compared to a corresponding two-stroke motor it uses less fuel per 
b.hp. Furthermore, as it does not employ a scavenging pump, re- 
ceivers, etc., the motor construction is simplified and better adapted 
for continuous hard work than the two-stroke-cycle motor. 

34 Disadvantages. The chief objections to this type of motor 
are its increased weight and the greater space required. On account 
of the larger number of valves the valve gear is more compUcated 



310 MABINE CRUDE-OIL MOTORS 

and noisy. Except with very large outputs the piston can act as a 
crosshead, and unless the cylinder diameter be greater than 20 in., 
water cooling is unnecessary. The crank motion of this type of 
motor necessitates a heavy crankshaft and heavy flywheel if smooth 
nmning is to be obtained. Furthermore, the exhaust valves are 
subjected to high temperatures and occasionally give rise to con- 
siderable trouble. This type of motor is exclusively used by the 
Danish and Dutch and now the American merchant marine and 
also finds a large field in the automobile and aeroplane industries, 
as gasoline motors work, as a rule, on the four-stroke cycle. 

TWO-STROKE-CYCLE MOTORS 

35 Advantages, Compared with a four-stroke motor the two- 
stroke-cycle type has a greater output per cylinder, and everything 
else being equal, this excess will usually be about 75 per cent. On 
account of its greater pressiu^ at the beginning of the compression 
stroke, the weight of the charge can be increased if the stroke volimies 
are equal. On the other hand, power is exerted only during three- 
quarters of the stroke, as during the remaining portion exhaust 
takes place. Thus the total stroke volume is not completely filled 
at the beginning of compression with a fresh charge, and the result 
is that for the same cylinder dimensions the output is only 75 per 
cent greater than for a four-stroke-cycle motor. This difference, 
however, becomes even less for high-speed two-stroke motors since 
the mean effective pressure is lower. By forced scavenging through 
the valves the fresh-air charge can, however, be made cleaner. In 
large imits only the starting and fuel valves are subjected to high 
temperatures. In the open-type four-stroke motor, vapor from 
lubricating oil and exhaust gases escape into the engine room, 
thereby causing impure air conditions. This, however, does not 
occur in a two-stroke motor which has a closed crankcase. Finally, 
the two-stroke engine is very smooth-running because of its more 
uniform torque. 

36 Disadvantages, The chief disadvantage of the two-stroke 
motor is its high operating cost. This is due to its higher loss of fresh- 
air charge through exhaust ports, smaller utilization of the working 
fluid, and higher heat loss due to cooling, friction and increased 
pump work. The mean effective pressure is lower because during 
the working stroke exhaust also takes place. The mean temperature 
during the cycle is higher than in the four-stroke motor and conse- 



J. W. MORTON 311 

quently cooling of the piston is necessary. Furthermore, the piston 
can act as a crosshead only in small motors; lubricating-oil con- 
sumption is greater, and the exhaust ports become overheated if 
not water-cooled. Finally, the stresses exerted on the moving 
parts are very large and, therefore, there must be ample sliding and 
bearing surfaces. The two-stroke type of motor is chiefly used in 
vessels where weight, space and first cost are the deciding factors. 
Its reliability and economy, however, have not been so marked as 
in the case of the four-stroke motor, although in the Scandinavian 
countries, Grermany and Italy it has been used to a considerable 
extent. 

DOUBLE-ACTING MOTORS 

37 Stationary double-acting motors have been used with 
fairly good results but double-acting marine motors are as yet in 
an early stage of development. The chief advantages of the double- 
acting motor are its light construction, better balancing, its econom- 
ical operation and the comparatively small space which it occupies. 
However, the first cost is higher, and the lubricating-oil consumption 
is greater than in the case of a single-acting motor. This type is also 
more compUcated and requires cooling of pistons and piston rods as 
the latter usually pass through the hot combustion chamber. 

HORIZONTAL MOTORS 

38 The horizontal motor is chiefly used for stationary plants 
because this type is more accessible, can be greatly overloaded, is 
cheap to construct and can be placed in a space with low headroom. 
Like all other types of internal-combustion engines, the horizontal 
motor has many disadvantages, chief among which are the follow- 
ing: The piston rings have a tendency to stick due to lubricating oil 
collecting on the lower half of the cylinder; imperfect lubrication 
thus results and this causes leakage and increases fuel consumption. 
The motor also has a tendency to rock and this necessitates a large 
foundation. Furthermore, the location of the crankshaft makes it 
exceedingly diflBcult ^o secure a direct drive of electric generators. 
These disadvantages in stationary plants usually outweigh all 
other considerations, and as a result the horizontal, double-acting, 
constant-pressure motor in large units is the type usually installed. 
On the other hand, in spite of the higher initial cost, the vertical 
motor is alwa3rs used on ships as its operation, durabiUty and 
great overload capacity make it an ideal installation. 



312 MARINE CRUDE-OIL MOTORS 

MOTOR DETAILS 

39 Mechanisms vs. Direct-acting Trunk. Trunk mechanism is 
applied in small-size motors and the crosshead guide is used only 
in large motors. This is due to the fact that motors with trunk 
pistons are cheaper to produce; the overall height is reduced and 
it meets the requirements for a non-leakable piston. The trunk 
mechanism, however, has certain disadvantages and these may be 
enumerated as follows: Side pressure is taken up by the hot part of 
the cylinder and since one side of the piston may become hotter 
than the other, the piston will naturally bend toward that side. 
The cylinder will, therefore, have a tendency to wear oval in shape, 
causing the piston to blow. Furthermore, the piston will, notwith- 
standing guards, draw lubricating oil up into the combustion cham- 
ber, and this of course increases operatmg expense. On the other 
hand, when the crosshead is used the lubricating oil can circulate 
in great quantities and inspection is also simplified. 

40 Forced-Feed vs. Gravity System Lubrication. The maximum 
pressures exerted between the wearing surfaces in an oil motor are 
greater than those in a steam engine, and consequently lubricat- 
ing systems must be different for each type. In large oil motors 
high pressures are used to force the oil to all vital parts, and the 
wearing surfaces are thus separated b}"- a film of oil. The loss due 
to friction is, therefore, reduced and the method also assists in keep- 
ing the surfaces cool and at the same tune increasing their life. 
Forced-feed lubrication also requires less attention and is independ- 
ent of the engineer's skill as long as the run is not hindered. The oil 
consumption is economical as the oil can be filtered and used again. 
There is no splashing of oil in or about the engine room and for this 
reason the system is well adapted to the high-speed motors. The 
forced-feed lubricating system with its double oil pumps, etc., is, 
however, quite expensive and since the mechanism is not visible, 
a breakdown is not easily detected; consequently, if one occurs, 
a serious delay will be occasioned. The forced-feed lubrication 
system with a close crankcase is used to great extent in high-powered 
motors whereas the centrifugal or gravity system is used in small 
niotoi*s. 

41 Piston Cooling. Piston cooHng may be accomplished by 
the use of either oil or water. If oil is used it may be combined with 
the forced-feed lubrication system, and in such case should the 
cooling oil leak into the bottom of the crankcase, .no harm will be 



J. W. MORTON 313 

done. On the other hand, if water is used and it gets into the crank- 
case the oil will be expelled from the sliding surface and serious 
damage may result. If sea water is used as a cooling medium, stor- 
age tanks are imnecessary, but if oil or fresh water is used, tanks 
and coolers must be provided and for large motors, these require 
considerable space. As a result large units use sea water for cool- 
ing purposes and medium-size motors employ oil. 

42 Scavenging Pumps. Piston Type. The simplest and cheap- 
set type of scavenging pump is a piston working in a closed crank 
pit. As a rule, however, this type of pump suppUes insufficient air 
since the quantity of air is limited to the stroke volmne. The cylinder 
is also insufficiently cooled and the volumetric efficiency decreases. 

43 Stepped-Piston Type. Scavenging air obtained from a pimip 
with a stepped piston is usually sufficient to meet all purposes, but 
if the piston is not tight, exhaust gases may mix with the fresh air, 
in which case there is great danger from explosion of the lubricating- 
oil vapors in the receiver. The stepped piston increases neither 
the length nor width of the motor, and if breakage occurs on a 
scavenging pump, only that individual cylinder is affected. The 
moving parts of such a pump are heavy and consequently large 
masses have to be accelerated and retarded. The motor, however, 
is increased in height, even where the wristpin is secured to the 
working cylinder, because the ratio of L (length of connecting rod) 
to R (crank radius) must be greater than normal in order to insure 
clearance between piston and shaft or bearing cap. This ratio, how- 
ever, becomes less if the wristpin is secured in the pump piston, but 
the overall height of the motor will be greater. On the other hand, 
if the piston pump takes the place of a crosshead, the length of the 
wristpin can be increased and the bearing pressure thus reduced. 

44 Special Pumps. The first cost of special scavenging pumps is 
high; furthermore, they require more room and, should they fail 
to function the several working cylinders will be greatly affected. 
The chief advantage of special pumps is their lower oil consumption 
and the ease with which sufficient cold air can be obtained. 

45 Each of the types of pumps mentioned has its various uses. 
The piston type is usually found on small motors, the stepped- 
piston type on submarine engines and motors of medium size, and 
separate or special pumps on heavy-duty motors installed on cargo 
and passenger ships. 

46 Ignition Devices of Constant-Volume Engines. Although it 
has been stated that the constant-volume crude-oil motor is an un- 



314 &IARINE CRUDE-OIL MOTORS 

satisfactory type, its ignition devices are of interest. If the " electric 
type of ignition " is employed the motor can be started inmiediately. 
The time of ignition can be controlled and varied at will. Electric 
ignition, however, is expensive and it is difficult for unskilled me- 
chanics to keep it in order. 

47 Another type of ignition device is the so-called "ignition 
tube." With this tube a greater compression ratio can be used and 
fuel consumption is also decreased. With this device the time of igni- 
tion can also be controlled and varied at will, but a steady or con- 
tinuously burning torch is necessary and this is a great objection, 
especially at sea. 

48 A third type of ignition system is the well-known " head " or 
" hot bulb,'' which is simple in construction, action and operation. 
At normal speeds this type of ignition is the most satisfactory and, 
furthermore, it is not aflfected by damp air or water spray. The chief 
objections, however, lie in the fact that it requires a long time to 
start the motor; the heads are easily cracked; the compression 
ratio is small and the mean effective pressiu^ is low. Furthermore, 
if the load changes during operation the hot bulbs will often run 
cold in the four-stroke cycle and hot in the two-stroke cycle. Broadly 
speaking, however, this type of ignition has been found to be most 
satisfactory and is very largely used for all types of small motors 
using either kerosene or heavy oil as fuel. 

DISCUSSION 

Louis Illmer (written). In the paper under discussion, Mr. 
Morton apparently favors the four-stroke oil engine and appears 
to give insufficient credit to the recent advances made in heavy- 
duty two-stroke engine design as applied to the merchant-marine 
service. 

Undue complication of mechanism is probably chiefly respon- 
sible for the slowness on the part of the American shipbuilder in adopt- 
ing the oil engine for the propulsion of large cargo ships, since each 
additional vital moving part adds to the first cost, the liability to 
accident and the overhauling required for maintenance of operating 
efficiency. 

Other things being equal, the oil engine with the least number 
of vital working parts will most adequately meet the demand for 
reliable low-cost power, particularly so under American conditions 
of high labor and relatively low fuel costs. 



DISCUSSION 315 

It is conceded that a perfected oil engine of the two-stroke type 
admits of the simplest possible construction for marine propulsion. 
The four-stroke marine engine suffers from inherent difficulty in re- 
versing the mechanically operated inlet and exhaust valves as driven 
from the half-speed camshaft drive. Furthermore, the long-stroke 
slow-speed engine of the four-stroke type does not compare favor- 
ably in weight, floor space or in first cost with the cargo-ship type 
of two-stroke engine. 

With but a single power impulse per cylinder for every two 
revolutions, the heavy power parts of a four-stroke engine do not 
work to advantage during the idle inlet- and exhaust-stroke periods. 
Hence the two-stroke engine is enabled to show a decided weight 
reduction and consequent lowering in first cost. For example, a 
six-cylinder slow-speed two-stroke engine suitable for cargo-ship 
propulsion can readily be made to deliver approximately If times 
as much shaft power for the same weight and floor space as will a 
four-stroke oil engine of the same speed and bore. This advantage • 
is of decided importance when larger merchant ships are to be en- 
gined, since the usefulness of such a prime mover is determined in a 
large measure by the ultimate power capacity to which a compact 
six- or eight-cylinder unit can be built. 

Another advantage of the two-stroke engine is that it requires 
no mechanically operated inlet and exhaust valves. The cylinders 
of such oil engines are best charged from the under side of the power 
piston, through inlet ports overrun by the piston. If desired, the 
scavenging air supply may be supplemented by additional low- 
pressure pumps but this is not essential when liberally proportioned 
injection-air compressors are provided. 

With this mode of charging a two-stroke cylinder, the timing 
events of the inlet and exhaust ports must be correctly proportioned 
and the shape of the piston deflector lug must be properly designed 
to produce effective scavenging of the power cylinder, otherwise the 
incoming fresh-air charge will blow out of the exhaust ports without 
displacing the burnt products of combustion. Deficiency in oxygen 
for combustion of the fuel oil has been responsible for the failure of 
a number of two-stroke engine designs. 

The described mode of charging two-stroke power cylinders 
reduces the valve gear parts to a minimum and requires only one 
air starting valve and one fuel valve opening into each cylinder 
head. Since these valves run in unison with the piston movements, 
they may be eccentric-driven and reversed by a Stephenson Unk 



316 IfABINE-CRnDE OIL MOTOB8 

in the manner of a steam-engine gear. The elimination of the half- 
speed camshaft drive is a feature of the two-stroke engine that makes 
for a simple and compact reversing valve gear. 

While the thermal eflBciency of a well-designed two-stroke en- 
gine is not quite as high as that of a four-stroke engine, the dif- 
ference is not large. Most of the loss of economy is chargeable to 
the increased pump work required to charge the power cylinder, 
but the relatively small gain of about 10 yyor cent in brake eflSciency 
in favor of the four-stroke engine is not in itself sufficient to oflFset 
the other advantages which a perfected two-stroke engine affords. 

A good two-stroke engine is somewhat more difficult to design 
than a four-stroke engine, partly because of the greater heat flow 
through the cylinder walls. For equal bore dimensions, the rate 
of heat flow in a two-stroke engine is approximately 1.6 times that 
in a four-stroke engine running at the same speed. 

The limiting rotative speed at which an oil or gas engine may 
be safely run is largely determined by the temperature assumed 
by the cylinder-bore wall, and if this is not kept within prescribed 
limits, troubles from piston lubrication and cracking of cylinder 
parts will result. 

The temperature head required to drive heat at a given rate of 
flow through the cylinder wall increases with the wall thickness. 
For equal shaft power the bore dimensions of a four-stroke oil en- 
gine are about 3 those required for a two-stroke engine. Allowing 
for the thicker four-stroke cylinder wall and assuming equal rotative 
speeds, the rate of heat flow in the case of the two-stroke engine 
should not exceed | that obtained in a four-stroke cylinder of 
equal power capacity. 

In high-speed marine oil engines, which are usually worked up 
to their limiting rate of heat flow, the four-stroke cycle offers some 
advantage. On the other hand, the relatively slow speed demanded 
for cargo-ship engines is exceptionally favorable to the two-stroke 
type, since tliese engines oi)erate under heat-flow conditions so 
moderate as to allow ample cooling of all vital parts even in the 
largest-sized cylinders. It is therefore readily possible to keep the 
l^ore-wall temperatures of such two-stroke engines well below the 
critical limits required for safe and reliable running. To further 
safeguard against fatigue and breakdown, it is advisable to provide 
for liner cylinders and such other constructive features common in 
high-powered oil-engine practice, so as to give the cylinder parts the 
requisite long life and complete immunity against cracking. 



V 



DISCUSSION 317 

The greater frequency of impulse and the consequent more 
even turning effort of the two-stroke engine largely improve the 
speed control and reduce to a minimum the flywheel effect needed 
for a smooth-running marine engine. 

StiD another feature of the two-stroke cycle that gives promise 
for rapid future development is its special adaptabiUty to the hot- 
bulb type of engine. 

Judging from the recent trend of marine oil-engine develop- 
ments, it now appears that a combination of the constant-pres- 
sure Diesel cycle with the hot-bulb constant-volmne cycle is likely 
to result in a cycle which is especiaUy suited to American oil-engine 
requirements. The lowering of the maximimi Diesel working pres- 
siure to approximately that used in the automobile engine should in- 
crease the reliability of operation. 

The constant-volume or explosive engine, when working with 
a compression pressure of about 250 lb. per sq. in., shows a thermal 
shaft efficiency but little inferior to that of the Diesel engine. By 
further reducing the compression to about 150 or 175 lb. per sq. in. 
the fuel economy of the explosive engine is but slightly sacrificed, 
while for a given size of crankshaft the weight and cost relations as 
taken upon the power-output basis show a considerable improve- 
ment as compared with the high-compression engine. 

An engine operating with the combined type of cycle should 
be provided with an efficient timed spray valve for the fuel-oil injec- 
tion, and in large engines this should be preferably of the air-injec- 
tion type used for Diesel engines. 

When working with such limited compression, self-ignition 
may best be obtained by holding a portion of the products of com- 
bustion from stroke to stroke within a water-jacketed vaporizer 
chamber, without, however, requiring the use of a hot plate. The 
trapped hot burnt gases may then be used to preheat the air that 
is pressed into the bulb chamber to a point where it is capable of 
promoting self-ignition of the injected fuel oil. This sets up a light 
explosion at constant volume, after which the remaining oil may 
be gradually fed into the power cylinder in the manner of a Diesel 
engine. 

While the water-cooled bulb will not of itself start the engine 
from the cold, it does not require an external flame; instead it is 
only necessary to preheat the small amount of air enclosed within 
the vaporizer chamber, which may readily be done by means of an 
electric coil heater or spark plug. After the first few explosive 



318 MARINE CRUDE-OIL MOTORS 

charges are thus ignited, the required heat transfer takes place 
from stroke to stroke to make the engine self-igniting. 

The jacketed vaporizer is especially applicable to oil engines of 
the single-acting type, the bulb chamber being preferably formed 
centrally in the cylinder head about the cylinder axis. The timed 
fuel valve can then be made to inject straight through the bulb 
chamber and directly against the hot piston head. This arrange- 
ment gives the nozzle considerable distance for proper spray forma- 
tion before striking the piston top, after which the oil charge spreads 
out over the piston-head surface and intimately mixes with all the 
air throughout the combustion chamber. 

Finally it is pointed out that this late development in true 
semi-Diesel engines is especially suited to the two-stroke-cycle engine. 
In the four-stroke engine considerable constructive difficulties are 
involved in placing the mechanically operated inlet and exhaust 
valves about the water-cooled vaporizer chamber. Furthermore, 
in a two-stroke engine the vaporizer chamber may be kept at a 
smaller size with respect to the piston displacement, due to the fact 
that the confined hot gases have less chance to cool off between power 
impulses. 

These and other advantages previously cited would indicate 
that a two-stroke semi-Diesel engine along the lines discussed should 
in the near future find favor for the propulsion of merchant ships 
and be capable of fully establishing the inherent possibilities of a 
slow-speed oil engine for marine service. 

W. D. Ennis (written). The value of this paper would be in- 
creased if the author would add particulars regarding the motors 
used. The mean effective pressures range from 87 to 91 lb. per sq. 
in., values often surpassed by stationary Diesel engines. Appar- 
ently the cylinders are the ordinary four-cycle single-acting form, but 
there seems to be no statement of the fact. 

A chief argument in favor of the crude-oil motor appears from 
Table 2, where "repeat orders" in 1916 are quoted for lines which 
began with this construction in 1912. What has happened since 
1916? 

It is to be assumed that the eight factors listed in Par. 2 are not 
presented as a complete list. In fact, the author refers to othen, 
admitting a shorter life for the crude-oil motor and referring to the 
absence of sufficient data on maintenance costs. He also HiMiiaaM 
what is almost an overwhelming disadvantage for any but slow- 



DISCUSSION 319 

speed cargo ships, the multiplicity of cylinders. The largest cyl- 
inder in Table 2 is below 30 in. bore and develops only 333 hp. 
The largest plant listed is 4000 hp. 

It is doubtful whether steam can ever be eliminated in naval 
practice. Bequirements for heating, humidifying and evaporators 
do not tend to decrease. It is even proposed now to use steam 
heat for submarines. It is difficult for internal-combustion engines 
to displace steam where they cannot displace it completely. 

Table 1 is obviously based on coal fuel for steam. If oil fuel 
is used, the ratio 22 directly following the table becomes about 2.0 
instead of 5.7. 

Forrest E. Cardullo characterized the adoption of the Diesel 
and analogous cycles for a standard as an advance too far in the 
direction of theory. He did not believe that there had been success- 
fully developed an engine capable of burning continuously all types 
of crude oil in a satisfactory manner. The engines become dirty 
and efficiency is lost. He said he knew of no reason why a satis- 
factory producer for gasifying oil could not be designed which would 
operate continuously at 90 per cent efficiency or even higher. With 
this producer it would be possible to develop an internal combus- 
tion engine of the automobile type which, with the same power, would 
have less internal friction than the Diesel engine. There would be 
certain practical disadvantages to such a combination, the principal 
ones being the extra space required and the additional labor for 
operation. He referred to the number of times a plant operating 
Diesel or semi-Diesel engines must be laid up in order to clean the 
engines. The practical weight of experience and of actual reliability 
and cheapness, he believed, was in favor of the producer-gas engine 
operating with gas generated from fuel oil. 

Lewis H. Nash concmTed in the idea of adopting the principle 
of gasifying oil in some type of gas producer. He knew of gas en- 
gines which had been in operation for 25 years with negligible cost 
for repairs. He considered that the mechanical difficulties with 
the Diesel engine far outweighed its theoretical advantages. 

Henry B. Oatley, referring to Table 1 in which Mr. Morton 
presented the comparative weights of power plants in steamships 
and motorships, said that in the comparative installations aboard 
the U. S. S. Maumee and Tiogay steam and motor-driven ships, 
respectively, the Diesel-engine plant weighed decidedly more than 



320 MARINE CRUDE-OIL MOTORS 

the steam plant. The difficulties of getting high pressure and the 
limitations of superheaters of which the author spoke while discus- 
sing the heat elements of temperature range for the various types 
of engines existed, according to Mr. Oatley, only in the present 
motor designs, but he saw no difficulty in the further development 
of the types which have been successfully operated at pressures as 
high as 500 or 700 lb. per sq. in. Thus, with higher pressures and 
with higher degrees of steam temperature the boiler plant, in point 
of weight and size, would become markedly smaller, and the con- 
trast in point of space occupied aboard the ship for the steam plant 
as compared to the motor plant would decrease, possibly to the 
point of being on a par. Mr. Oatley further criticized the argument 
advanced for preferring the motorship over the steam-driven ship 
by reason of the depreciation of the hull because of the sulphur con- 
tent of coal and the amount of coal space required in the steam- 
driven ship, on the ground that rapid progress had been made in the 
utiUzation of oil as fuel for steam-driven ships, and that this practice 
would in all likelihood continue to increase. 

(). C. Berry,* commenting on Mr. CarduUo's suggestion of gasi- 
fying oil in some type of gas producer, mentioned that in experi- 
ments conducted at Purdue University for the purpose of finding 
the best means of introducing into the mixture the heat required 
to vaporize the liquid fuel, it had been concluded that two elements 
are required to get the mixture in a burnable condition. In the 
first place, the temperature of the mixture must be high enough to 
maintain in the liquid fuel sufficient vaporizing pressure so that the 
fuel, once reduced to a vapor, will stay in a vaporized state and get 
into the cylinder as a vapor and not a liquid. The other was the 
time element. Because of the smallness of this time element, thou^ 
the temperatures are high enough to maintain the required tempera- 
tures, still the mixtures retain their wet condition when they arrive 
in the cylinder. For this reason, observed Professor Berry, the hot 
spot had come into automobile practice as one of the solutions for 
the vaporization of kerosene, and he expressed his belief that it 
could be applied to good advantage to the use of fuels heavier than 
kerosene. A properly designed hot spot, he said, would eliminate 
the difficulty found in gas producers, where the temperature is often 
so high that it is not possible to keep a high enough compression 
without getting preignition. 

* Professor of Mechanical Engineering, Purdue University. 



V 



DISCUSSION 321 

Mr. Cardullo said in reply that in mentioning the gas producer 
he had intended to convey the idea of the actual combustion of oil 
in a producer so as to form a clean, burnable mixture in the engine. 

L. H. Johnson spoke of his experience with the two-cycle semi- 
Deisel hot-bulb, hot-surface engine and stated that after investi- 
gating the various theories advanced to explain this type of engine 
he had finally formulated the following conclusion as to its opera- 
tion: That the oil must be injected into the engine in the form of a 
spray or a mist, and if low compression is to be used, it must be 
sprayed against a hot surface and vaporized; then this oil gas mixes 
with the air compressed in the chamber and explodes by the heat of 
compression or the heat stored in the combustion chamber. Tests, 
he said, had convinced him that it is necessary for successful opera- 
tion of the engine to remove the deposit of carbon or residue of mate- 
rial that is left in the cylinder. In the experiments to which he 
referred no trouble had been found in maintaining the temperature 
necessary for successful operation and the main problem had been 
the elimination of the deposits. 

The Author wrote in reply to Mr. Illmer's discussion that he 
was, at present, in favor of the four-stroke cycle oil engine for mar- 
ine purposes. Space restrictions in the original paper had prevented 
his comparing the various types in detail. He thought that the two- 
stroke cycle engine had not been perfected to a point which would 
warrant its use at the present time in marine installations. He 
doubted the advisability of charging the two-stroke engine from 
below the piston on account of the danger of adulterating the work- 
ing charge. Scavenging with valves located in the head was su- 
perior, but the design introduced complications. 

He believed that with solid-fuel injection, the four-stroke en- 
gine could be made with fewer working parts, and that larger out- 
put could be obtained with less weight. He felt that the trend of 
manufacturers was toward the Diesel engine and away from the 
hot-bulb engine, in the larger sizes. 

Answering Mr. Ennis* request for more complete information 
regarding the motors mentioned in the paper, the author stated that 
they were of the four-stroke, single-acting type and were built by 
Burmeister and Wain, Copenhagen, Denmark. He realized that 
mean effective pressures greater than 91 lb. per sq. in. were obtain- 
able in stationary practice, but this did not extend to marine prac- 



322 MARINE CRUDE-OIL MOTORS 

tice. There were excellent reasons for this, such as the difference in 
cooling media, ventilation, speed, and foundation. It was due to 
lack of raw material throughout the war that there had been little 
activity in the building of marine motors by the Copenhagen con- 
cern. The building numbers of present orders ran from 276 to 344 
with deliveries stipulated as far ahead as 1922. The East Asiatic 
Company of Denmark had 30 on order, the largest of which was to 
be 6500 i.hp. in 16 cylinders. It was quite true, as Mr. Ennis 
pointed out, that, compared with oil rather than coal, the value of 
R derived in Par. 5 of the paper would be between 2 and 2.5. 

With regard to the reliability of the motorship, the author 
cited the record of the George Washington which made a non-stop 
run of 68 days without mishap or cleaning of any sort. Every 
valve taken out for examination was found to be without fault and 
replaced. 



No. 1099 

A SUGGESTED FORMULA FOR RATING 

KEROSENE ENGINES 

Bt D. L. Arnold, Chicaqo, III. 
Member of the Society 

At the present Ume there is no standard method of rating kerosene oil engines, 
and consequently engines of the same displacement are given different ratings by 
each manufacturer. A standard formula would therefore he of value and in this 
paper the writer discusses the various formuloB now in use cmd suggests that a pis- 
ton displacement of 13,000 cu. in. per minute be taken as one brake horsepower. 
The subject of a standard name-plate is also discussed and a suggested form is given. 

A MONG the various companies manufacturing kerosene en- 
^^ gines there seems to be no uniform standard of rating in use, 
and as a result engines of the same bore, stroke and speed — or in 
other words, of the same piston displacement per minute — are given 
different ratings by aUnost as many manufacturers. In the past 
little attention has been paid to this minor detail, as it seems to have 
been considered, but the purchaser of engines and tractors has had 
much experience with this inconsistency. For instance, one buys 
an engine of 4 hp. that will actually develop 6.5 hp.; another buys 
one also rated 4 hp. but it will only develop 5 hp., and it will in all 
probability have a different piston displacement per minute. Ap- 
parently this means nothing as far as the customer is concerned, 
for in each case he receives more horsepower than Jie believed he 
was purchasing. A very great difficulty does, however, arise when 
these engines are judged by the amount of maximum work which 
they will perform in a given length of time, not on brake tests but 
on users' equipment, and the result is that the purchaser of the 
smaller engine is greatly dissatisfied, despite the fact that his engine 
will develop a 25 per cent overload. 

2 The foregoing is only a sample case of the inconsistency which 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of Thk 
American Societt or Mechanical EInqineebs. 

323 



324 



FORMULA FOR RATING KEROSENE ENGINES 



has resulted in many bills now being introduced, or about to be 
introduced, in state legislatures, specifying how an engine shall be 
rated. If these bills should all pass it might be that each state would 
require a different rating, and thus the manufacturers would be 
obliged to furnish a different nameplate in each state. It would 
therefore seem advisable for the manufacturers to get together 
and adopt a standard rating for kerosene engines; and this work 
would be greatly facilitated if the engineers of this Society and 
of the Society of Automotive Engineers would cooperate and adopt 
a standard rating for this type of power plant. 



TABLE 1 FORMULA NOW IN USE FOR FOUR-STROKE-CYCLE ENGINES 



n 



diameter of piston, in. 
number of cylinders 



L » length of stroke, in. 

N » number of revolutioos per minate 



No. 



3 
4 

5 

6 

7 

8 
9 

10 



Authority 



N.A.C.C. * 

or 
A.L.A.M. 

E. P. Roberta 

French Automobile Club 
Royal Automobile Club, British 

Royal Automobile Club, Swedish 

Prof. H. L. Collanders 
T. Thornycrofts 

M. Farouz 
M. Amon 

E. W. Roberts * 



Fouixrx^ 



2.5 



(at 1000 ft. per min.) 



Dl^LNn 

18.000 

0.45 I^n (at 985 ft. per min.) 
0.405 Ll^n 
I^LNn 



15,240 
0.565 D{D 



-l)n 



2700 
0.121 Z)**L®*« 
0.1 D'^n 
I^LNn 



^ Vol. 1, p. 30, 8.A.E. SUndards DaU Sheets. 

* Gas Engine Handbook, p. 131. X is a factor varying with the fuel and cycle. For I eycie 
engines X equals 16,000 fur natural gas and 14,000 for gasoline. 



3 If stanihircl rating is to be of si^rvice, it must give equal 
protection to the manufacturer and the customer, and any formula 
which is obtained must therefore be based upon practice covering 
a sufficient period of time. Before taking up the proposed fonnulEi 
however, the writer wishes to review some of the formulae that are 
in use tcxhiv and which are Usted on Table 1. 



D. L. ABNOLD 325 

4 The A.L.A.M. formula^ is based on the assumption that there 
is developed within the cylinder a mean effective pressure of 90 lb. 
per sq. in., that the engine operates with 75 per cent mechanical 
efficiency, and that the piston speed is 1000 ft. per min. With im- 
govemed engines this formula is perhaps as good as any, but for 
engines operating with governors it is obvious that if a piston speed 
of other than 1000 ft. per min. is maintained, it will be necessary 
to make a correction to cover this point, in which event the formula 
is unwieldly and inconvenient. 

5 An empirical formula * which is considered as very conserva- 
tive practice when compared with the various ratings now used by 
engine manufacturers is as follo\%'s: B.hp. = 0.00006042 D^LNn = 
V/13,000. Before deriving this formula the writer compared the 
ratings of 115 engines manufactured by 61 companies and found 
them to vary from 8,256 to 19,985 cu. in. per b.hp. Forty-six 
engines were rated using 11,000 cu. in. or less as 1 b.hp.; 29 used 

* Derivation of A.L.A.M. Formula: 
Let, HPb - brake horsepower 
HPi = indicated horsepower 
HPr = rated horsepower 

A «- area of piston, sq. in. 

D « diameter of piston, in. ^ 

E - mechanical eflBciency (assimied at 0.75) ** 

L « length of stroke, in. 

n - number of cylinders 

P = mean effective pressure, lb. per sq. in. (assumed at 90 lb.) 

V - piston displacement per minute, cu. in. 

N - revolutions per minute 

8 - piston speea, ft. per min. (assimied at 1000) 

MPk HP F ^ PASNn „ 90 X 0.7854 D* = 1 000 x n x 0.75 
^ ° * ^ 4 33,000 ^ " 4 X 33,000 x 12 



, or for practical purposes, 



Also, since s 



2.489* *^ '^ *^ ' 2.5 

2L.V 
12 



„p^ 90 X 0.7854 D» x 2L.V x n x 0.75 D'LNn nnnnnA«oQ n.r v 
^^' 4^33,000 X 12 - 14:939 = 000006693 D^LATn 

and substituting V for 0.7854 D^LNn 

„j. 90 X F X 2 X 0.75 F_^ 

° 12 X 33,000 X 4 " 11,733 

• Derivation of Empirical Formula: 

0.7854 D^LNn V 



HPr (recommended) = 



13,000 13,000 



HPr (transformed) — f^=in - 0.00006042 D'LXn 

1d,OOU 



326 



FOBMULA FOR RATING KEROSENE ENGINES 



11,000 to 12,000 cu. in.; 12 used 12,000 to 13,000 cu. in.; 10 used 
13,000 to 14,000 cu. in. and 18 used above 14,000 cu. in. 

6 It will thus be seen that 87 engines or approximately 75 per 
cent were rated as using less than 13,000 cu. in. per b.hp. and while 
this value may be more conservative than some manufacturers 
might wish to use, it gives a standard basis of comparison and 
allows the engine to develop a fair percentage of overload above 
its rating, since any engine should be able to develop as a maximum 




Fig. 1 



3 5 

Formula Numbers 

Comparison of ExoiNE-RATiNa Formula in Terms 
OF Piston Displacement per Mindtb 



1 b.hp. por 11,000 cu. in. pLston displacomont per min. when the 
engine i.s in ^ood condition. 

7 At this time it seciiLS desirable to compjire this new formula 
with those iilready in use, and for that reason Table 2 has 
compiled, from which have l^een plotted the values shown 
Fig. 1 which shows a wide variation in some of the formulje. 
It should also \jc noted that thLs chart does not show the maxi- 
mum horscinnver that can be expected from an engine but has 



D. L. ABNOLD 



327 



reference only to the rated brake horsepower which should be rea- 
sonably expected from a kerosene engine. If one engine can develop 
a higher maximum than another, or, in other words, can make good 
on a lower piston displacement per minute than its competitor, the 
credit is certainly due to the manufacturer who can consistently 
produce these conditions. However, practically all engines of cor- 
rect design will develop a maximum horsepower according to the 
formula heretofore given, namely 1 b.hp. per 11,000 cu. in. piston 
displacement per min. 

8 With reference to the rating of four-stroke-cycle engines, 
it would seem best for all concerned that this rating should be 



TABLE 2 COMPARISON OF ENGINE-RATING FORMULiB IN TERMS OF 

PISTON DISPLACEMENT (V) PER MINUTE 



FOBlfULA No. 


1 


j 2 

1 


3 


5 


10 


11 


Authority 


N.C.C.C. 

OB 

A.L.A.M. 


1 

E. P. Roberts 


1 

1 

Royal 
French Auto- 
Auto- 1 mobile 
mobile | Club 
; Swedish 

1 


E. W. Roberts 


Rated 
Horsepower 




16000 


14000 


11500 


ReoommeiH 
dation 


1 

Expressed in 
piston dis-' V 


V 
14,137 


V 


• 

V 


V 


V 


V 
9.032 


V 


placement per 11,733 
minute 


13,180 11,969 

i 


12,566 


10.995 


13.000 


! 
Expressed in 
t e r m s f D^LNn 


18,000 


1 
D'^LNn I^LNn D^LNn 


D*LiVn 


11.499 


ti^NLn 


Roberts form- 
ula 


14,939 


16,781 


15,240 


15,999 


13.998 


16,550 



made the nearest whole horsepower to that determined by the 
standard formula, it being a very easy matter to change the 
rated speed of the engine in order to accomplish the exact 
rating. In any event, however, the rated horsepower would not, 
if the foregoing condition is adhered to, vary more than plus or 
minus 0.5 hp. 

9 Having established a standard rating for this class of engine, 
there should also be a standardized nameplate; to state the hp. 
alone with no reference to speed is insufficient and misleading. This 



328 



FORMULA FOR RATING KEROSENE ENGINES 



nameplate should be clear and concise, leaving no cause for doubt 
as to what is meant. Many companies use the form of rating as 
follows : 




Others simply use: 




and still others reverse the figures with relation to the lettering. It 
would seem, however, that the most exact way of placing the rating 
on the nameplate would be according to the following form: 



16 HP at 500 RPM 



It has been the custom of many manufacturers to leave out the little 
word "at,'* which, when included in the fonnula, leaves no room 
for doubt as to what is meant. 

10 In considering the intemal-combustion-engine rating we 
must also consider the tractor rating as the intemal-combimtion 
engine forms its power unit and the tractor rating is therefore de- 
pendent upon the engine rating. The owner of a tractor is not only 
interested in the amount of power that can ho delivered by the 
engine but also in the amount that can be dolivored to the drawbar 
at tlie different speeds. Here again, however, manufacturers have 
been inconsistent in the ratings which they have made. The ma- 
jority of manufacturers have followed the nile that drawbar horse- 
power should l)e considered as 50 per cent of the rating of the power 
unit and in actual practice covering many years of experience this 
seeuLs to l)e a verj' conservative figure to use. It is true that the 
tractor will often develop greater drawbar horsejwwer, but when 
taking into consideration the wide range of conditions through 
which the tractor must work, such as changes of soil and class of 
work, a 50 |^t c(»iit rating for th(» drawbar pull seems on the whole 
to l>e the Ix'st value to use; then*fon\ on the tractor engines it would 
seem that the best nameplate would l>e the following: 



V 



D. L. ARNOLD 329 



16 Brake HP at 500 RPM 

8 Drawbar HP at 500 RPM of the engine 

Drawbar pull: 

lb., reverse speed at miles per hour 

lb., first speed at miles per hour 

lb., second speed at miles per hour 

lb., third speed at miles per hour 

Drawbar pull and HP are on the average good footing. 



11 The following formulae are those which it would seem are 
best adopted to the rating of internal-combustion engines and 
tractors: 

^ ^ , . , 0.7854 D'LNn ^.^ 

Rated engine horsepower = ^^ ryw> {X\ 

^ , , Rated engine horsepower ^^-i 

Drawbar horsepower = s L2J 

^ , „ „ , 375 X rated drawbar horsepower ^--i 

Drawbar pull, lb.*= r t-- ~^ u [3] 

^ ' travel in miles per hour *- -• 

^, . , o . „ 63,025.21 X brake horsepower ^^-i 

Motor torque,^ m-lb. = ^r? . . . L*J 

12 Practical work in experimental testing laboratories and on 
regular test floors has proved that four-stroke-cycle engines operat- 
ing on kerosene will develop for periods of two hours or more 1 b.hp. 

* Derivation of Drawbar Pull Formula: 

Let Fd = drawbar pull, lb. 
HPd = drawbar horsepower 
S » speed in miles per hour 

^, „ ZfPd X 33,000 X 60 375 HPd 
Then Fd^ ^280 x~5^ " ^S" 

* Derivation of Motor Torque Formula: 

Let HPb = brake horsepower 

I » length of brake arm, in. 
W = lb. pull at end of arm 
N = revolutions per minute 
T = motor torque, in-lb. 

^^"° ^^' ' 33;000 

_ H Pb X 3 3, 000 x 1 2 
2tN 

_ 63,025.21 X /rP6 

"^ N 



1 



330 FORMULA FOR RATING KEROSENE ENGINES 

per 11,000 cu. in. piston displacement per min. and a few engines 
have for short periods of time developed 1 b.hp. for every 9800 cu. 
in. piston displacement per min. Therefore, making allowances for 
general wear, mishandling and improper adjustment, the formulae 
given in Par. 11 will be seen to be conser\'ative and well adapted for 
the rating of this type of engine. 

DISCUSSION 

Harry F. Smith thought that it would be altogether out of 
place to attempt to create any arbitrarj'^ standard of cubic-inch 
displacement for the determination of the horsepower of the kero- 
sene-oil engine. He said he knew of tests recently made on sub- 
stantially identical kerosene-oil engines in which the b.hp. developed 
differed by 100 per cent, due to change in cylinder design. That 
meant that the thermal efficiency of one engine was 100 per cent 
better than the other. It seemed to him, therefore, that the manu- 
facturer who was in a position to double the thermal eflfieiency of 
the kerosene oil engine ought to be entitled to whatever benefits 
might accrue thereby, and not be handicapped by the action of the 
Society or of State legi>hitures in rating his engine at a certain horse- 
power according to its i^ize or tlie numlxT of pounds of cast iron put 
into it. 

Stafford Montgomery inquired whether the rating of kerosene 
engine builders from 8000 to 20.000 cu. in. per b.hp. was limited 
to kerosene engines or whether it applied also to gasoline engines. 
This question was referred to the meeting by the chairman. 

Fredrik Ottesen protested against manufacturers of large 
gas engines having to adopt the formula set forth in Mr. Arnold's 

f)a|>er. 

William T. MACiRVDER remarked that in The Ohio State Uni- 
vci>itv School of Militarv Aeronautics thev had formulated a state- 

^ *r » 

iiicnt which was fairly accurate, that instead of using 13,000 or 
14,fKK), 10,(MK) cu. in. represented (luite clost'ly the horsepower of an 
aeroplane (•ll^ine up to 1(K) per cent capacity, and increased speed 
after that did not jjive correspoudingh' increased horsepower. He 
^aid that the horsepower of a steam tractor is unknown until tested 
U}T the rea>on that its overload capacity is not like that of a boiler, 
one or two or three hundred |K'r cent, but from four to five hundred 



DISCUSSION 331 

per cent; and that in tests performed at the laboratories of the uni- 
versity, tractors nominally rated at 20 hp. gave over 100 hp. and 
kept it up; on the other hand, in actual tests of kerosene and gaso- 
line tractors on blocks, either in field work or in comparative b.hp. 
tests, many of the tractors fail to operate at their rated horsepower 
and not even their own experts could always get from them what the 
nameplates indicated they should deliver. Such being the con- 
ditions in practice, it appeared to him that by the standardization 
of horsepower, misunderstandings arising from the erroneous inter- 
pretation of the discrepancy between rated and actual horsepower 
could be avoided. He further observed that the question was boimd 
to be taken up soon by the various legislatures and, as Mr. Arnold 
said in his paper, the conditions at present existing in the gas tractor 
and the kerosene tractor would be dupUcated in the automobile if 
the Society did not take prompt and effective steps to translate 
the idea of a horsepower so as to make it intelligible to the average 
man who knows nothing about machinery or about mechanics. 

John Chucan assented to the remarks made by Professor 
Magruder and asserted that he had actually used for several years 
a formula identical with that proposed by Mr. Arnold and had foimd 
it to be very conservative. The formula, as he understood it, was 
intended only for tractors and not for automobile or any other 
engines. 

The Author. There seems to have been some argument and 
objection to the proposed formula though it seems to me to result 
from a failure to realize the necessity of such formula. 

First: It must be understood that any formula that can be 
devised at this time is purely empirical and is established for a 
basis of comparison. 

Second: That having established a standard basis by which 
engines of this type shall be rated, each engine can then be com- 
pared in relation to the per cent of maximum overload they can 
sustain under actual brake test for a given period of time. 

Third: This formula applies equally to gasoline as well as 
kerosene engines, the only diflference being that under certain con- 
ditions the gasoline will probably sustain a greater maximum. 



I 



No. 1700 

STANDARDS OF CARBURETOR PERFORMANCE 

By O. C. Berry,* Lafayette, Ind. 
Non-Member 

The type of test which a carhureiar is most frequently given to establish its merit 
or lack cf merit is carried out in the foUowing manner: An engine in good mechani- 
cal condition is tested^ using a carburetor which bears a high repiUation. These tests 
are then repeated, using the new carburetor , and its merits are thus reported in terms 
of the comparative performances of the engine. The results of these tests are valua^- 
ble and convincing^ and while they wtU always be the final criterion of carburetor 
performance f they fall a little short of the ideal in thai they faU to show the reasons 
for the differences. Several writers have recently pointed this out and suggested that the 
performance of a carburetor should also be expressed in terms of its ability to per- 
form those functions which are essential to proper carburation. In order to do this 
it wiU be necessary to determine these essentials. Each function must then be studied 
to see when and how it operates, its comparative importance, and the conditions for 
the best results. This is a difficidt task, but its accomplishment should prove of 
great benefit. In this paper a list of the essential factors, as now understood, has 
accordingly been made. Some experimental data are also presented which it is hoped 
wiU help to establish some of the standards of performance. These data cover the 
following points: 

(a) The efficiency and power capacity of an engine as affected by the rich- 
ness of the mixture 

(Jb) The effect of the speed of an engine upon its mixture requirements 

(c) The effect of the torque produced by the engine upon its mixture require- 
ments 

{d) The effect of the dryness of the mixture upon the mixture requirements, 
power and efficiency of the engine: 

1 The heat for drying the mixture coming from the sensible heat in the air 

2 The heat far drying the mixture coming from a "hot spot** in the intake 

manifold. 
Further research is now under way which is expected to throw light on other phases 
of the problem. 

Satisfactory carburation of a liquid fuel depends jointly 

upon the carburetor, the intake manifold, and the temperature 

of the combustion-chamber walls. These parts may therefore be 

considered the carburating apparatus and the headings under which 

* Professor of Mechanical Engineering, Purdue University. 

Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Society of Mechanical Engineers. 

333 



334 STANDARDS OF CARBUBETOB PEBFOBICANCB 

are grouped the tests to determine the degree of excellence with 
which each of these parts performs its respective functions may all 
be listed together. These headings expressed in the general order 
of their importance are as follows: 

a The range of flow-rate capacities, or, in other words, the 
maximum and minimum number of cubic feet of air per 
minute that can be handled 

h The richness of the mixture as afifected by: the rate of 
flow of the air through the carburetor; sudden changes 
in the rate of flow; the amount the throttle is open 

c The pressure drop through the carburetor at different 
rates of flow 

d The thoroughness and uniformity with which the fuel 
and air are mixed 

e The uniformity of the richness of the mixture furnished to 
the different cylinders 

/ The temperature and dryness of the mixture entering the 
cylinders 

g The temperature of the combustion-chamber walls, par- 
ticularly the piston head. 

The relative importance of these items is largely a matter of opinion 
and will be different under different conditions. For example, the 
capacity for handling large amounts of air is much more important 
in passenger cars than in trucks and tractors, while in tractors using 
kerosene the temperature of the mixture and combustion-chamber 
walls becomes comparatively much more imp)ortant than it is in 
passenger cars burning gasoline. 

a RANGE OF FLOW-RATE CAPACITIES 

2 In passenger-car engines great importance is attached to 
floxil)ility, and the buying public would not consider a car that 
was notably defective in this respect. The power developed by 
an engine varies almost directly with the weight of air used per 
minute. The carburetor must therefore be able to supply this 
air, properly mixed with fuel, through a wide range of flow rates. 
As in practice the carburetor is frequently found to l)e one of the 
main factors limiting the flexibility of the engine, the range of flow- 
rate capacities is placed at the head of the list as probabl}' the most 
important item by which the carburetor is to be judged. 



X 



O. C. BERRT 335 

3 Carburetors are usually rated as to size by the size of their 
outlet flanges. No account is taken of the maximum or minimum 
rate of flow, though in practice there is a marked lack of uniformity 
in the flow-rate capacities of carburetors of the same rated size. 
Nevertheless these carburetors are supposed to be interchangeable. 
The range of flow-rate requirements of various engines also differs 
widely, being large for the passenger-car engine and small for the 
truck and tractor engines. 

4 The passenger car is expected to idle down to one or two 
miles per hour on high gear and at the same time have a maximum 
speed capacity well above 50 miles per hour. Few carburetors are 
capable of furnishing a proper mixture over so wide a range, although 
most carburetors are sold imder claims of wonderful flexibiUty. The 
passenger-car engineer can only be sure of a proper carburetor by 
specifying the exact flow rates required to secure the desired flexi- 
bility. Similar specifications with the less exacting truck and trac- 
tor engines, which are usually governor-controlled, will frequently 
result in a marked saving through the substitution of a cheaper 
carburetor of more rugged design. 

5 A carburetor is truly suited to an engine only when its flow- 
rate capacities correspond to the requirements of the engine, and 
this condition will be attained with the greatest certainty when the 
requirements of the engine and the capacities of the carburetor are 
both given in cubic feet of air per minute. It is therefore suggested 
that the flow-rate capacities of all carburetors should be stated defi- 
nitely in cubic feet per minute, and that this information should 
always accompany the statement of the size of the carburetor flange. 

6 The air required by an engine may be computed as follows: 
With the carburetor adjusted so that the engine carries its full torque 
with open throttle, the air required per brake horsepower per minute 
will remain nearly constant irrespective of the amount of gasoline 
used or the speed of the engine. For the usual passenger-car type 
of engine with a compression ratio of about 4 to 1 this constant is 
about 2.1 cu. ft. per min. at full power, and seldom exceeds 2.3 cu. ft. 
per min. The air used when idling at any speed is almost exactly 
one-quaHer of that used under full load at the same speed. With 
these data, which are based upon the performance of the engines 
tested at Purdue, it is easy to compute the air requirements for any 
engine of the same type, for any torque and any speed. 



336 STANDARDS OF CARBURETOR PERFORMANCE 

b RICHNESS OF THE MIXTURE 

7 The thermal efficiency of an engine at any speed and load is 
affected more by the richness of the explosive mixture than by any 
other factor. It is therefore highly important that the carburetor 
deliver the mixture to the engine in the proper proportions, at all 
speeds and loads. In order to be sure that this is being accomplished, 
it will be necessary to do several things. In the first place, the rich- 
ness of mixture giving the best power and the one giving the best 
efficiency must each be experimentally determined. This must be 
done for various speeds and loads, so as to learn what effect a change 
in either will have on the mixture requirements of the engine. The 
temperature and the dryness of the mixture must also be varied to 
determine whether or not they affect the power or efficiency of the 
engine. After thus estabUshing the results desired of the carburetor, 
it will be necessary to determine the types of failures observed in 
carburetors in meeting this requirement. Tests may then be out- 
lined that will establish the degree of excellence of the carburetor 
with respect to the richness of the mixture furnished to the engine. 

8 In the Purdue tests the richness of the mixture is expressed 
in pounds of gasoline per pound of dry air. With a dry mixture at 
half load and 1000 r.p.m., regular firing may be obtained with mix- 
tures between 0.0575 and 0.12. The best efficiency imder the same 
conditions accompanies a mixture of about 0.067, and the best power, 
0.08. The point at which the engine begins to miss is hard to deter- 
mine as the missing appears so gradually. Likewise it is hard to 
choose the point of highest power or highest efficiency, as the curves 
are both flat on top. These points were carefully estimated, how- 
ever, and the figures reported are close approximations. The method 
used in obtaining these results and a few of the characteristic curves 
will be presented at the end of this paper. 

C PRESSURE DROP THROUGH THE CARBURETOR 

9 The ideal condition of perfect volumetric efficiency is not 
attiiiniiblc, since the air and ga»soline nuLst be caused to flow into 
the cylindei-s. It is always necessary to have the gasoline in the 
float chamber of the carburetor at a lower level than the delivery 
orifice in order to prevent overflowing when the engine Ls not run- 
ning, and the suction in the carburetor must \yc great enough to 
overcome this safety head before any gasoUne will be deUvered. The 



\ 



O. C. BERRT 337 

best attainable condition, therefore, will be to have just enough 
vacuum in the carburetor to cause a satisfactory flow of fuel and 
air, and no more. Because of the influence of volumetric efficiency 
on engine capacity the importance of a small carburetor vacuum is 
very great. For this reason designing engineers should insist on data 
showing the pressmre drop through the carburetor necessary to give 
the desired rates of flow. This drop in pressure should be measured 
at the throttle on the carburetor side and the method of making the 
measiu^ment should be specified very carefully. The demand for 
this information would soon induce carburetor manufacturers to 
publish guaranteed vacuum-air-flow cm^es, thus making possible 
intelligent selection of a carburetor for a given service. 

d THOROUGHNESS OF THE MIXING 

10 The thoroughness and uniformity with which the fuel is 
mixed with the air is important. One of the greatest helps in reduc- 
ing the fuel to a gas is to atomize it thoroughly and mix it with the 
air. This gives the fuel a large amoimt of exposed surface and 
helps to bring all parts of the mixture up to the same degree of 
saturation. When the fuel leaves the carburetor as a Uquid not 
well mixed with air, it flows toward the cylinders more slowly than 
the air, and consequently collects in the manifold and arrives at the 
cylinders in waves, causing the mixture to vary widely in richness. 
It is very difficult to design a manifold that will distribute the liquid 
evenly to all of the cylinders. All of these facts add to the imp)or- 
tance which we must attach to the thoroughness with which the 
fuel is atomized and mixed with the air. 

11 The mixture in all parts of each individual cylinder must 
be uniform and the fuel reduced to a gas at the end of the compres- 
sion stroke or else the combustion will not be complete, thus lower- 
ing the power and efficiency of the engine and causing uneven run- 
ning. Therefore, if the engine runs with a regular and even exhaust, 
the thoroughness of the mixing is probably good along with all of 
the other factors influencing engine performance. The best direct 
test of the quaUty of the mixing is to have the carburetor discharge 
into a section containing glass placed between the carburetor and 
manifold. Fig. 4 shows the design used in the Purdue tests. The 
best dry mixtures appear as colorless and dry as pure air, while 
wet mixtures resemble a fog, and in most cases streams of liquid 
fuel are seen following a spiral path along the walls of the manifold. 



338 STANDARDS OF CARBURETOR PERFORMANCE 

e UNIFORMITY OF MIXTURE TO ALL CYLINDERS 

12 The richness of the mixture entering the various cylinders 
often differs widely, especially when very wet mixtures are used. 
Some manifold designs also aggravate this condition. As dry mix- 
tures not only are seldom used but are also of questionable desira- 
bility, it would seem that the problem resolves itself largely into 
one of manifold design. A direct quantitative test of the richness 
of mixture to each cylinder is impracticable, but the results at- 
tained by any given manifold may be tested by removing the 
exhaust manifold and observing the flames from the exhaust open- 
ings. This can be done to best advantage in comparative darkness. 
By adjusting the carburetor for continuously leaner mixtures the 
impoverished cylinders will be caused to miss, and then by gradually 
enriching the mixture the yellowish flame will indicate the cylinder 
with the rich mixture. Uniformity is of course the desired goal. 

I 

/ TEMPERATURE AND DRYNESS OF THE MIXTURE 

13 With the rapidly increasing difficulty in vaporizing the com- 
mercial liquid fuels, the temperature of the mixture becomes a more 
and more important consideration. Before the fuel can be burned 
it must be vaporized, and this requires both heat and time. The 
heat may come from the sensible heat in the air entering the car- 
buretor, or from hot surfaces in the carburetor, intake manifold or 
combustion chamber. Often the temperature and vapor pressure 
of the fuel are high enough to maintain a dry mixture, once it is 
established, but the time element is lacking and the mixtures are 
consequently quite wet. There is a difference between requiring a 
mixture to be dry as it enters the cj-linder, and dry at the end of 
the compression stroke. The heat due to compression raises the 
teniporature of the mixture well up toward the ignition point by 
th(» end of the compression stroke, so that any fuel suspended in the 
form of a fog will tend to be vaporized. The temperature of the 
piston lu'ad is also high enough to aid materially in flashing into 
a gas any of the fuel which may enter the cylinder as a liquid. The 
ol)j(»cts to be accomplished are to have the mixture drj' and ihor- 
oiighly mixed at the end of the compression stroke in order to get 
p;oo(l combust ion, and to have the gas temperature as low as possi- 
l)le at the end of \\w suction stroke, in order to keep up the volu- 
m(*tric cflicicMicy. It is therefore desirable to make the fullest 
possii^le use of the heat in the combustion-chamber walls, piston head 



V 



O. C. BEBRT 339 

and compression^ and to introduce the mixture into the cylinder as 
wet as possible and still be sure of having it dry before it is burned. 

14 Interesting light is thrown on this point by some of the 
Purdue tests. A series of tests was run at 1000 r.p.m. and a half- 
load throttle setting with all of the conditions constant except the 
temperature of the air to the carbiu^tor. Under these conditions 
an air temperature of about 300 deg. fahr. was necessary in order 
to have the mixture in the manifold dry. With the air at 300 deg. 
fahr. and the mixture dry the highest power was 12.5 b.hp. and the 
highest eflSciency 18 per cent. As the temperature of the air was 
lowered the power was increased, imtil at 80 deg. fahr. the best power 
was 16 b.hp. or an increase of 3.5 hp. Below 80 deg. fahr. the firing 
of the engine became irregular and the power dropped ofif. The 
richness of the mixture giving the best efficiency remained the same, 
as the air temperatures were lowered to 150 deg. fahr., indicating 
that the mixture was dry when combustion started and the com- 
bustion complete. Below 150 deg. fahr. the mixture for best effi- 
ciency commenced to grow richer, indicating incomplete combus- 
tion and a possibility that the fuel was not completely vaporized 
before combustion took place. At 80 deg. fahr. this tendency had 
not become sufficiently pronounced, however, to counterbalance the 
increase in p)ower, so that the efficiencies increased slightly, even 
down to 80 deg. fahr. At this point the best efficiency was 18.8 per 
cent, or an increase of 0.8 per cent. The engine ran smoothly through- 
out the entire temperature range when good pulling mixtures were 
used and it was found that with high temperatures a leaner mixture 
could be fired. 

15 It is possible to compute approximately the change in power 
that will accompany a definite change in the temperature of the 
mixture furnished to the engine. The computation is based upon 
the fact that the density of a gas varies inversely as its absolute tem- 
perature. Since the power generated in the cylinder varies almost 
in direct proportion to the weight per minute of the mixture used, 
it will vary almost in the inverse ratio of the absolute temperatures 
of the mixtures. As an illustration, suppose the engine will develop 
12.5 hp. when the mixture has a temperature of 250 deg. fahr. What 
will it develop when the temperature is 60 deg. fahr.? The power 
times the inverse ratio of the mixture temperatures is as follows: 

12.5 X '^^±^ = 17.06 
460+60 



340 STANDARDS OF CARBURETOR PERFOBMANGB 

This indicates that 17.06 hp. is to be expected with the 60 d^. fahr. 
mixture. As a matter of fact, the temperature in the cylinders at 
the end of the suction stroke rather than the intake manifold tem- 
perature determines the density of the charge, but this temperature 
cannot be measured. Again, the indicated rather than the brake 
horsepower is the one that varies with the density of the charge, 
while the brake horsepower is the one that is usually measured. 
These are the main reasons why the results obtained are only ap- 
proximately correct. The figures given are from two Purdue tests 
run at the same speed with the same throttle orifice, and the power 
was increased from 12.5 to 16.0 instead of 17.06 b.hp. 

16 The task of determining the best method of introducing 
into the mixture the heat for vaporizing the fuel is both important 
and difficult, and will require considerable careful experimentation. 
The Purdue tests with the *'hot spot" warrant the conclusion that 
it is superior to any method of preheating the air, in that it dries 
the mixture sufficiently at lower temperatures and, therefore, does 
not decrease the power of the engine so much. These tests also 
indicate that the design of the hot spot is still in need of experi- 
mental development. 

g TEMPERATURE OF THE COMBUSTION CHAMBER 

17 When the mixture is dry as it enters the cylinders, or the 
fuel so well atomized that it remains suspended in the air and is 
entirely vaporized during the compression stroke, the heat absorbed 
from the cojubust ion-chamber walls docs not improve the carbura- 
tion, but decreases the power capacity of the engine without im- 
proving either its efficiency or the way it nms. These conditions, 
however, arc rare. A considerable portion of the fuel usually enters 
the cylinders as a liquid, which collects on the piston head. Under 
those* conditions the tomix^rature of the comlmst ion-chamber walls, 
ospocially the* piston head, U^comos voiy important. The reasons 
for this may he oxplaincd as follows: The piston head is usually 
at a toniiKTalure two or three hundred degrees al>ove the cylinder 
walls. If the temperature of the walls is 200 deg. fahr. the pb- 
ton \wiu\ will therefore l»e lK»tween 400 and 500 deg. fahr. If the 
wall tciniKM'aturo is lowered to 100 deg. fahr., the piston-head tern- 
jXTature will drop to hetw(»en 300 and 400 deg. fahr. These tern- 
|)oraturos ap|)ly to passonger-car engines running under ordinary 
conditions. Several tests have l)een run at Purdue to determine the 



\ 



O. C. BERRY 341 

rate at which Red Crown power gasoline will evap)orate from the 
surface of a hot iron plate. The maximum rate of evap)oration seems 
to occur when the metal is about 450 deg. fahr. If the evaporation 
in oimces of fuel per square inch of metal per second be taken as 
100 per cent at 450 deg. fahr., then the evaporation at 400 deg. fahr. 
is about 40 per cent, at 350 deg. fahr. about 9 per cent, and at 300 
deg. fahr. about 1.8 per cent. It is therefore imp)ortant that the 
jacket-water tempferatiu^ be kept high when the piston head is 
depended upon to flash any considerable amount of Uquid fuel into 
a gas. This conclusion is borne out by the engine tests. With the 
air entering the carburetor at 70 deg. fahr. and the jacket water 
maintained at 110 deg., the engine would not fire regularly with 
any richness of mixture. When the jacket-water temperature was 
raised to 200 deg. fahr. the engine would fire some of the richer mix- 
tures regularly and by raising the air temperature to 80 deg. fahr., 
the engine developed full power and efficiency and would fire a wide 
range of mixtures. 

THE PURDUE TESTS 

18 The tests at Purdue University were carried out on a Haynes 
Light Six and a Willys-Knight four-cylinder engine, mounted on 
a Diehl electric dynamometer, the majority of the tests, however, 
being on the latter. Fig. 1 shows the Knight engine mounted ready 
for a test. Fig. 2 gives a more detailed view of the scales and the 
electrical apparatus for starting and stopping the tests. The supply 
of gasoline is piped from tank A, Fig. 1, into a 2-qt. glass vessel 
placed in one of the scale pans. This is shown at A , Fig. 2. In order 
to give the balances freedom of motion, the gasoline was siphoned 
from this vessel to the carburetor. The balances J5, Fig. 2, were 
capable of weighing the gasoline to the one-hundredth part of an 
ounce, and were equipped with wires dipping into mercury cups 
C, Fig. 2, which completed an electric circuit just when a balance 
was reached. The hand on the dial of the air meter was equipped 
with an electrically operated clutch and brake, the stop watch was 
operated by a strong magnet D, Fig. 2, and the revolution counter 
on the end of the dynamometer shaft was electrically operated. By 
these means, when the scales came to a balance they would start 
the stop watch, the revolution counter and the recording hand on 
the air meter and ring a gong. Weights corresponding to the gaso- 
line to be used were then removed from the scale pan, and when 



342 BTANDARDB OF CABBUBBTOB PEBFOBUANCB 

the scales again came to a balance the recording apparatus was 
electrically stopped, making it possible to time all of the readings 
tc^ther and make the record at leisure from instruments that were 
standing still. 

19 The air was metered through an Emeo No. 4 gas meter 
B, Fig. 1, reading to cubic feet, so that tenths of a cubic foot could 
be estimated. The drop in pressure through the meter was indi- 
cated by the water manometer C, Fig. 1. The barometer and the 
wet- and dry-bulb readings on a hygrometer were taken periodically, 
in order to determine the pounds of dry air used in each instance. 



FiQ. 1 Willts-Knioht Engine and Eouipubnt roa TKsTiMa 
ITS Carburetor 

.4, GmrcUne tuikl 5. sir mstcr; C, msDomclcr: D. hnlcr; B, (u jeti; 
F, thcrmoDietei; G, water Isnk 

20 'J'he me((^i' wan connected to a heater D, Fig. 1, by means 
of niliiier tubing, which wiw kept from colkii)sirig by a coil of wire 
iiiwide <if it. The heater wju-* made up of 4 ft. of 3-iii. wrought-iron 
pi|)e, surnmrided by an astwritos cylinder. Between the pipe and 
asbcjitos w;iM a s|>aco large enough to allow the i)ii»e to be heated 
l>y the llaiiic.-; from llic gas burners K. The tem|X'raturc of the air 
IciLving llic hcattT was indicated by the thermometer F. This ther- 
nionu'lcr passed through a jMicking gland in the air line, so that 
its bulb was exposed directly to the air inside of the pipe. The 
tcmiK'ratun' af llic air could l>e regulated within close limits by 
careful aiijiistmcnt of the gas flumes. Betw'cen the healer and tbo 



O. C. BERBT 343 

engine the line was covered by a thick layer of hair cloth to pre- 
vent radiation. In some instances a section containing glass was 
inserted between the carburetor and engine.' The glass tubing was 
of the same inside diameter as the rest of the line and was carefully 
packed in a section of wrougbt-iron pipe. The sides of the pipe 
were milled away so as to ofFer ample opportunity to observe the 
mixture in the tube. By holding an electric light behind the glass 
as shown, very aatisfactoty observations could be made of the char- 
acter and behavior of the mixture inside. 

21 In the Haynes set-up the temperature of the cooling water 



Fia. 2 Scales for WBiaHiNO Gasounb, and ELBtTTRicAi. 
Equipment roB SrARTiNa and Stoppino Tests 

jl.Guoliiie beaker; B. bslsDces; C, mercury cupg; D, ningaot 

for the engine was kept between 125 and 135 deg. fahr. Circula- 
tion through the jackets of the engine was induced by the standard 
Haynes engine-driven pump. In the Willys-Knight tests the tank 
G, Fig. I, was used, but only for a time, it being replaced by a radia- 
tor from a Liljcrty B, truck. Water wju*? also introduced directly 
from the service lines of the laboratory. In all of these tests the 
temperature of the cooling water wiis carefully recorded. 

22 The speed of the engines was r'jkI on a tachometer as well 
as being computed from the stop-watch and revolution-counter 
readings, giving a good check on this important factor. The torque 
developed by the dynamometer was weighed by means of a sensi- 



344 STANDABDS OF CARBDRETOR PERFOBMANCE 

tive set of Fairbanks scales. This is the same as the brake load on 
the engine, thus making it possible to compute the power developed 

by the engine to a satisfactory degree of accuracy. 

MBTHOD OF CONDUCTING THE TESTS 

23 The object of the first series of tests was to determine the 
effect of changing the richness of the mixture on the performance 



ao5 ooe 007 ooa oce oio on an qi3 

ftnjnd&of6a&oline per Pound of Air in Mi«ture 
Kio. 3 Power as'd Efficibnct Curves or Willtb-K might ENatKS 

ThrollU' Bct for ImU-lond. Speed, 1000 rpm. Cdoling-watet Uoipentun. 200 itf. lahr. 

of :iii eiLgint'. It was planiicii to dotcnninc the mixture that would 
Kivi- the Ih'si [Hjwcr, the one for the Ijest efficiency and the range of 
itiixlnrcs th:tt •niiUI Ih; fired regularly. This \vu.s done as follows; 
A tliin siwl pliite was placed l>ctwc(?n the carburetor and intake 
mariifdid, an<l tlie throttle rfinove<i. In this plate was drilled a 
liolu of such r'hii Hint when a jKiwerful mixture was uimhI the engine 
would licvcloji tJic dttfinid [X)wcr at the desired epeed. A scries of 
lists was Itii'n run with this orifice and at the given speed, and a 



O. C. BERBT 345 

set of power and efficiency curves plotted similar to Fig. 3. In this 
case the speed waa 1000 r.p.m. and the orifice was for half load. In 
the first test in this series the weight of gasoline per pound of air 
in the mixture was computed, together with the power and efficiency 
developed by the engine. This gave the first point on each of the 
curves. The mixture was then made slightly richer by opening 
the gasoline needle and the brake load was adjusted to bring the 
speed back to 1000 r.p.m. A second test was made imder these con- 
ditions, and another point on each of the curves determined. This 
was repeated imtil the engine would miss and could not carry any- 
where near its original load. The amount of gasoline was then 
gradually decreased each time and a series of tests made until the 
mixture was so lean that the engine would not perform properly. 
This process was repeated, making the mixture alternately richer 
and leaner until a large number of determinations had been made, 
and the results when plotted gave the desired information with 
satisfactory clearness. An attempt was made during all of the tests 
to keep the temperatures of the jacket water and carburetor air 
constant, as well as the spark setting. 

ACCURACY OF RESULTS 

24 When a single series of tests is run in a continuous sequence 
from the lean to the rich end of the mixture range, the plotted points 
tend to fall in a consistent line and the results may seem to have 
an accuracy that they do not possess. However, if the series is 
repeated back and forth a number of times, the points will tend to 
vary somewhat from the original line and show clearly the degree 
of accuracy being attained. An engine running under what seem 
to be constant conditions will vary in its performance, and the 
best apparatus is liable to a certain amount of error, so that before 
any far-reaching conclusions are drawn from experimental results 
it is desirable to know their limits of accuracy. For this reason the 
points in Fig. 3 were obtained as suggested above, and since this 
work was done at the beginning of the entire series of tests, before 
the men became as expert as they were later, it is felt that the errors 
are never greater than is indicated by these curves. 

24 The vertical line in Fig. 3 (at 0.0672) represents the chemi- 
cally perfect mixture, or the one in which there is just enough oxygen 
in the air to bum the fuel and no excess of either fuel or air exists. 
The curves show that the engine will run with a mixture as lean as 



346 STANDARDS OF CABBUBETOB PEBFOBICANCE 

0.05 lb. of gasoline per lb. of air, but will not pull well with so lean 
a mixture. The test log shows that it misses frequently at this 
power, but that the performance becomes better as the mixture is 
made richer, until at 0.055 it fires every cylinder regularly. The 
best efficiency is obtained at 0.063, when the engine is developing 
91 per cent of its maximum power capacity with this orifice at 1000 
r.p.m. The best power accompanies a mixture of about 0.08, at 
which point the thermal efficiency has dropped from 17.25 to 14.8 
per cent. The richest mixture that can be fired regularly is about 
0.1275, but the engine will run with mixtures as rich as 0.135. Nearly 
full load can be carried with a mixture as lean as 0.065, or as rich as 
0.115. In other words, a carburetor can be adjusted with as lean a 
mixture as can be used to carry full load, and the amount of gaso- 
Une can be nearly doubled, without greatly affecting the power 
developed or the smoothness of running of the engine. It is prac- 
tically impossible to stand by the side of an engine mounted on a 
test block and distinguish any difference in its performance as the 
mixture is being changed through this range. 

25 In applying the information obtained from these tests to 
other conditions it must be remembered that these results are for 
half -load at 1000 r.p.m., when a warm mixture was used that was 
dry as it left the intake manifold. Before applying the conclusions 
to other conditions, one must learn the effect of the load carried, 
the speed and the temperature and dryness of the mixture upon the 
mixture requirements, power and efficiency of an engine. 

EFFECT OF LOAD UPON MIXTURE REQUIREMENTS 

20 Figs. 4 and 5 show the power and efficiency curves taken 
from the Willys-Knight engine running at 1000 r.p.m., but with dif- 
ferent throttle orifices. The figures given on each curve indicate 
the largest brake load carried with that particular orifice at 1000 
r.p.m. During these tests the temperature of the air entering the 
carlmretor was kept at about 150 deg. fahr., and the cooling water 
at about 120 deg. fahr. ThLs was true of all of the curves excepting 
the one for 92.8 lb., in which ctise the mixture was heated up to 125 
d(»j2:. fahr. in a ''hot spot" and the cooling-water temperature was 
1()0 deg. fahr. Fig. 4 shows that the mixture giving the "best power 
at a fixed throttle setting does not vary with the brake load carried, 
i)ut remains constant at about 0.08. At light loads the engine will 
not oiKTati; well with as wide a range of mixtures as it can use when 



O. C. BBBBT 347 

cuTying more nearly its full capacity. Fig. 5 shows that with light- 
load throttle Bettings the mixtures for best power and best efficiency 
tend to coincide, but as the brake load is increased the mixture for 
best efficiency becomes continuously leaner, until at full load it is 
0.062. In the case of the higher brake loads the engine will hit 
regularly and run smoothly with the lean mixtures which give the 
high efficiencies, but the power developed is reduced considerably 
below the hi^est attainable at that throttle setting. The most 



satisfactory mixture for general iLse at or near full load will, there- 
fore, be approximately 0.067, giving almost full power and nearly 
the best efficiency. For lighter pulling conditions the mixture had 
better be caused to approach 0.08, the one for best pulling. 

EFFECT OF SPEED UPON MIXTURE BEQUIBEMENTS 

27 In Fig. 6 is shown a set of curves taken from tests on the 
Haynes Light Six engine running at different speeds from 400 to 
1600 r.p.m., and in each case with a throttle orifice givii^ about 



-J TT 



ZSFORMANCE 



:v:re giving the Ixj.st power 
:: that at high speeds the 

re as much excess fuel as 
:hat the mixture for tlif 



riMPKRATlRES 

-■■.:? taken from the Willv- 
' .'.:. :i half-load orifice ;is a 




: : zi 

*  1 « ^ 

• "N'.^i *. :v.;Hr Six Kxgivk 



.. ». '^ -^ ."V 



•x 



. '.'..o oarluirctor al tem|N*ra- 
Tliev sliow that SO deg. 



0. C. BBRBT 349 

mixture as lean as 0.055. Fig. 8 gives the efficiency curves. The 
best efficiencies at 80 and 125 deg. fabr. are nearly exactly the same, 
18.75 per cent, but at 80 deg. fahr. it accompanies a mixture slightly 
richer than for 125 deg. fahr. Each increase in temperature between 
125 and 275 deg. fahr. decreases the efficiency slightly. At 150 deg. 
fahr. the mixture for the highest efficiency reaches its leanest point, 
0.063, and remains the same up to 275 deg. fahr. The effects of 
increasing the temperature of the air entering the carburetor above 
80 deg. fahr. are therefore to decrease the power capacity of the 
engine considerably and its efficiency slightly, but to make it fire 
regularly when using leaner mixtures. 



FiQS. 7 AND 8 Heated--\ir Test Curves, Willys-Khiobt Engine 

Throttle Ht for tiif-loBd. Sp«s<l. 1000 r.p.to. Cooling-wster Umpersture, 200 deg. t»lif. 

29 Figs. 9 and 10 show the curves for the same series of tests, 
but with the colder air temperatures. Fig. 9 shows that with an air 
temperature of 61 deg. fahr. the engine could use only the compara- 
tively rich mixtures and its power capacity was greatly reduced. 
When the air temperature was increased to 71 deg. fabr. the power 
was brought nearly up to maximum, but still the engine could not 
fire the leaner mixtures with regularity. Fig. 10 shows that the 
efficiency for the 61 deg. fahr. air is very poor, while the 71 deg. fahr. 
air was much better but the engine stopped before it reached a lean 
enough mixture to give the best results. It may therefore be seen 
that 80 deg, fahr. is about the lowest temperature at which the air 



350 STANDABDB O? CABBUXBTOB FBRFOBUANCE 

may be drawn into a carburetor when using Red Crown power gaso- 
line, to get good carburation in an engine having a 200 deg. fahr. 
cooling-water temperature. If the cooling-water temperature is 
lowered the temperature of the air will have to be raised corre- 
spondingly, while raising the quality of the gasoline will improve the 
performance with cold air, cool water or both. 



THE HOT-SPOT TESTS 

30 Figs. 11 and 12 show curves taken from the Willys-Knight 
engine using a so-called "hot spot" between the carburetor and 



Pound! o* 6o.alifLei«r Pou-d rf «ir ir-Hi-tu-* Poiindt ol &ato>ii<* ptr ^uiid st »if i« llrit»n 

FiQS. AND 10 Cold-Air Tests, WiLLYs-KNiairr Enoikb 

Tbrottlo gal For hall-loul. Speed, IQOO r.p.ni. Coolinc-inlar IcmpentuR. 200 dc«. tab. 

the intake manifold. This hot spot was designed to flash the liquid 
fuel into a gus while heating the air as little as possible. The curves 
show that by u»ing the hot spot the engine is able to Arc the lean 
mixtures without in turn lowering cither its power or its efficiency. 
The explanation is that the factor contrullinf; the power of the engine 
is the weight of air that can pass through the throttle orifice in a 
given time. The air always being cold when it passes this point, the 
[K>wor is not affected by the heat. This is always true for a throttled 
(■ondition, but a.s the throttle is opened and the manifold or valves 
bofonie the limiting factors in the production of power, the advan- 
tage of tho hot siKit decreases, but it does not disappear. It is there- 



O. C. BEBRT 



clear that the hot-spot method of introducing heat into the 
ture is always superior to the hot-air method, and particularly 
rben the engine is throttled down. 

























s 
-1 


F« 




57 










m 


*^ 


5^ 


^ 


f^ 






f\ 










^ 


1^^ 


/ 


1 
















1 

















AND 12 "Hot Spot" Tests, Willts-Knioht Enoine 

a hBlf-Lood. Speed, 1000 r p.rn. CoDling-ntu («mpentun. 200 de(. faht. 



CONCLUSIONS 

il It may be well to again point out the importance of judging 
merit of a carhurating system in terms of the degree of thor- 
iness with which the system performs those fmictions which 
necessary to proper carburation. It is hoped that a discussion 
he topic will result in establishing specifications for these tests, 



352 STANDARDS OF CARBURETOR PERFORMANCE 

and that many may join in the task of carrying out the experiments 
that will determine the standards of performance in each case. 
Tests are needed showing the richness of mixture that will give the 
best power and the one for best efficiency when using liquid fuels 
other than gasoline. It would also be interesting to extend the gaso- 
line tests to include wider ranges of speed and higher compression. 

32 The problem of determining the best method of introducing 
into the mixture the heat necessarj^ to vaporize the fuel is one of great 
imix)rtance, esiwcially in connection with the heavier fuels. As a 
part of this problem it will be interesting to determine the actual 
temperatures found in the metal of the piston head and cylinder 
walls of automobile, truck and tractor engines, and how these tern- 
IH^aturos affect the carburation. 

33 The fullest discussion of these problems and the methods 
employed at Purdue in attacking them is earnestly solicited, and 
any suggestions in connection with carrying out this line of investi- 
gations will Ih» nn'oivi^l ma*<t grat^^fuUy by those in charge of the 
Purilue Engineering ExixTiment Station. 

ACKNOWLEDGMENTS 

34 The exiH*rimontal work presentetl in the forc^ing paragraphs 
was carried out in the laln^ratories of l^r^lue University under the 
auspices of the Punlue Engineering Ex|H>riment St^ition, and has 
riHvived thnnighout the jx^rsonal attention and encouragement of 
Pn^f. Ci. A. Ymmg. Heail of the So1uh>1 of Mechanical Elngineering. 
Most of tlie ti^sts were carriinl out and the computations made by 
Mr. r. S. Kegonvis. Research Asv<istant. It is hoped that the Elxperi- 
mont Statii>n will have a bulletin on carbuniiion ready for distribu- 
tion by early winter. The plan is to make this a detailed report of 
rht* work whicli is summari/od in this |vi{vr. together with such 
ni\wv d.i-.ix as an* tlien available for piibliration. iVpies of this bulle- 
tin in:iy Iv o'ttaiiiovl l\v adilrx^ssiiic i\ H. IVnjamin. Dirwtorof the 
l'::.::i!'iT v.::^: l\:H'riiiu ::t Station. Puriiuo TniYtTsiry. I^fayette 



I • • " . * ' 



I 



nisnssu^N 

V ■:;.: ^i r * \k:*; 1 1 »^ :v!ri:i:*Mv: ti:.i: :\x \.\xrh\:rv*oT \U^i^ 

:*r'. i \M'.!-. !:\o ^:ruoii:ra'. iio:a'.*< of tho engine, 
\\\*\. ' '■'■ • '" . i" ivx:,»:: :\ir:s. ;i:\: t<:xvi.-^*\\ w.rh ihe form 



1 « • • * • « . 



DISCUSSION 363 

of the manifold. The real problem of carburetor design, as he saw 
it, was to get the combination of a carburetor and inlet port in an 
engine which would give a uniformly suspended mixture of minute 
droplets in air at the instant the spark fires the charge. But 
this would not lead, say, to the knowledge of the best kind of a 
combination carburetor, manifold and engine which would enable an 
automobile to climb a good stiff hill at ten or fifteen miles per hour, 
and at the same time give a speed of three miles per hour in a 
crowded traffic street and a speed of fifty miles per hour on a 
smooth, clear road. 

Mr. Cardullo further observed that a rich mixture could be 
secured at the instant of acceleration more by properly designing 
the form of the manifold than by providing methods for increas- 
ing the mixture. Also that by reason of the separator action of 
the manifold it would be impossible to get a representative sample 
of the mixture entering the cylinder, and he beUeved the proper 
thing to do would be to analyze the exhaust of each cylinder for 
the proportion of carbon dioxide in order to determine whether 
each cylinder was getting its due proportion of gasolme in the mix- 
ture provided. 

Thomas J. Litle, Jr.,^ observed that in automobile practice 
many diflSculties are encountered which do not occur in stationary 
engine work. With reference to the statement in the paper that 
the best performance is obtained with a temperature of 80 deg., 
he preferred to take the temperature of the mixture entering the 
block, after it passes through the carburetor intake manifold. He 
claimed that the temperature of the charge as it entered the block, 
after passing through the carburetor and intake manifold, was the 
controlling factor in motor performance. He suggested as a future 
line of research work tapping into the intake passage just at the end 
of the supply pipe and analyzing a sample of the mixture taken at 
that point. To a question of Chairman Magruder inquiring how 
that sample of air and gasoline vapor was going to be obtained, Mr. 
Litle explained that he had succeeded in drawing it off by lifting 
the valve of a tall gasometer with a long water still. 

The Author. The definition which Professor Cardullo gives 
of the object to be accomplished in carburetion is one that probably 
will not satisfy all automotive engineers. The temperature attained 

* Research Engineer, Lincoln Motor Company, Detroit, Mich. 



354 STANDARDS OF CABBURETOR PEBFOBMANCE 

at the end of the compression stroke will usually be high enough 
to vaporize all of the minute droplets of the liquid fuel which were 
suspended in the mixture at the end of the suction stroke. The ideal 
condition seems to be to have the mixture dry and uniformly mked 
at the end of the compression stroke or at the instant the spark fires 
the charge. There are two main objections to a wet mixture in the 
manifold. The first is that it tends to wet any surface it comes in 
contact with and thus gives up a large part of its liquid content on 
its way to the cylinder and this deposited liquid is very hard to dis- 
tribute satisfactorily. The second is that in many cases this Uquid 
is not vaporized in time to be mixed with the air and burned, and 
is consequently wasted. 

In carrying out these tests it was decided to avoid the difficult 
task of obtaining a true sample of a wet mixture, as Mr. Lytic sug- 
gests, but to use a mixture that was hot enough to be entirely dry. 
A part of the tests here reported were accordingly run under these 
conditions. Furthermore, the tests seem to indicate that the tem- 
perature and dryness of the mixture at the end of the compression 
stroke are the important factors rather than the temperature of 
either the air entering the carburetor or the mixture in the intake 
manifold. 

In the ^'hot spot" tests no single test was continued long enough 
to give any considerable carbon deposite. With the heavier fuels 
some depa<^it will probably collect at any temperature that is high 
enough to vaporize it rapidly, but the Purdue tests offer no infor- 
mation on this point. 



No. 1701 

PULVERIZED COAL AS A FUEL 

By N. C. Harrison, Atlanta, Ga. 
Member of the Society 

The author first renews same of the uses of puberized coal in the industries — 
such as the cemeni, steel and copper industries, after which he gives a technical definir 
tion of pulverized coal, describes the process by which it is prepared for use and fur- 
nishes tables of costs of preparation. The pidverized-coalrbuming open-hearth steel 
plant of the American Iron & Steel Manufacturing Co., Lebarum, Pa., is described 
and the advantages observed in the plant of pulverized coal, compared with producer 
gas as a fuel for openrhearth furnaces, are listed. 

The use of pulverized coal in stationary boiler plants is discussed, five deter- 
mining factors in the successful operation of such a plant being taken up in detail. 
As compared unth mechanical^stoker plants the advantages of the pulverizedrcoal 
plant are enumerated and certain precautions to be observed with the latter type of 
plant are brought out. 

A report of a test of a 468-hp. Edge Moor boiler with pulverized-coal equip-^ 
ment is included in the paper and the efficiency obtained is compared with the effi- 
ciency of a stoker-fed boiler in the same plant, a greater net efficiency being found in 
the pulverized-coal plant. 

The paper concludes with a statement of some advantages obtained in the pul- 
verized-coal plant, 

nPHE peculiar conditions as they exist today, on account of the 
war and for other reasons, such as the gradual disappearance 
of sources of fuel, like natural gas, and the shortage in the supply 
of crude oils, which have become of too great value for ordinary 
fuel purposes, have compelled those interested to consider the possi- 
bility of the adoption of pulverized coal as a fuel, to replace their 
present methods of operation. Pulverized coal was first used in 
the United States about 26 years ago for the economical burning of 
the cement rock in the rotary kilns of the portland-cement industry. 
2 Pulverized coal was first applied successfully for economical 
reasons m connection with the bummg of portland cement. The 
growth of the portland-cement industry also had a great bearing 
on the development and use of pulverized coal, in that, it is in this 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of Thb 
American Society of Mechanical Engineers. 
For discussion see p. 392. 

355 




356 PULVERIZED COAL AS A FUEL 

industry that pulverizing machines were brought to the present high 
.state of development, for in the manufacturing of cement not only 
the coal is pulverized but for each barrel of cement manufactured 
weighing 380 lb. there are required about 600 lb. of raw material 
such as limestone shale or cement rock as well as the 380 lb. of 
clinker produced by the kilns which must be pulverized in order to 
make the finished product, so that in the neighborhood of 1100 lb. 
of raw materials, clinker and coal must be ground to produce one 
barrel of portland cement. As there are a hundred million barrels 
of Portland cement made in this country annually, these figures 
will give one a reason why pulverizing machines have been so highly 
developed during the last few years. Fine grinding of the raw 
material means reduction in the quantity of fuel required and also 
makes possible the highest quality of the finished product, so far 
as the chemical analysis or combination is concerned. Fine grind- 
ing of the clinker means increased strength for the reason that the 
hydraulically active units in cement are in direct proportion to the 
percentage of fine or im{)alpable iK)wder in the finished product. 

3 This statement is made to impress upon j-ou that equipment 
for preparing and handling pulverized coal has long since passed 
the experimental stage and has now l)een developed to a high state 
of efficiency and is readily obtainable. Somewhere between 30 and 
50 million tons of pulverized coal have been used to date in the manu- 
facture of cement alone. There is now being used approximately 
6 million tons annuallv. 

4 The apphcation of this fonn of fuel has been gradually taken 
up by engineers connected with other industries, who have speedily 
reoognizovl its value to such an extent that the steel industry today 
is using in the noighl>orlKKHl of two milUon tons of pulverized coal 
annually in various tyix^s of furnaces such as open-hearth, heating, 
puddling, soaking pits, continuous-heating, reheating, annealing, 
fonrinc furnaoos. and furnaivs of practically everj' description where 
Iu\U is ivquiivvl. 

5 Ti:o I'opjvr industry is using l»otwoon one and two million 
t.'!i< ivr yoar in oi\^rv\istini: furnaces. rovcrl^nitor>- and copper-melt- 
ii.ii :";irn:uos oi :ill tvjvs. Larco amounts of pulverized ooal are 
;>•: : ::: :o:;trv kilns other than tho ooniont industr\0 for the de- 
>■.;'.:':.::■. :::'.c itv.d vo:i>\:nc oi v:^.rious i:r:iars oi ores: for nodulinng 
':'...-:-:.:::.. 1 r ::;i' •:■.;<: so ;i> :o :v.;iko ;iv;-.:l:i:-o provlucts heretofore 
\i:\ i\:\:.-.\o :*.» ri\^'\i:-. :\r ':u:;i:::t: '.i::.f 'o oxide of lime for 
•.>i ■.::  >:■.-":.»..::'•■. :*-.:::-;;vi > . :\^r : -.'.rr.iv.i: vii»\'mito for opeil-Jiearth 



N. C. HARBISON 357 

furnaces; and for the calcining (or driving off of CO2 gas and water 
of crystallization) of various minerals, from which are obtainable 
such conmiodities as plaster of paris, stucco, potash, etc. A total 
of approximately ten million tons of pulverized coal are burned 
annually in the United States in the above industries. 

6 A still further and very important development is now going 
on, which will, when it attains its growth, require more pulverized 
coal than probably all of the other industries combined, and that 
is in its application to locomotives, particularly in the West. This 
application is now being developed. There is still another field in 
which enormous quantities of this fuel will be used and a field in 
which we are all concerned, and that is in the generation of power 
in stationary power houses. 

7 Practically any coal can be burned in pulverized form with 
a proper furnace and burning equipment. Each application 
however must necessarily be governed by the quality of the fuel 
available in the district in which it is made. Generally speaking, 
however, the coals which would give the most satisfactory results 
would be those in which the ash content would be less than 10 per 
cent, the volatile averaging between 30 and 40 per cent and the fixed 
carbon between 40 and 50 per cent. The sulphur content should 
be low, although coal with a sulphur content running as high as 
4J to 5 per cent is being burned in pulverized form under boilers 
without any detrimental results. The ash should have a high melt- 
ing point. These statements, however, are tentative, as most ex- 
cellent results have been obtained from all sorts of coals, differing 
widely from the ideal analysis stated. 

8 It is very apparent that the development in this method of 
burning coal has brought coals, from which heretofore very ineffi- 
cient results have been obtained, within reach of a great many con- 
sumers. For instance, from Texas to Edmonton, Alberta, the 
coimtry is underlaid with various grades of lignites, low-grade 
mineral coals with high moisture content and of such a nature that 
the ash would melt or flow down on the grates, thereby preventing 
the highest efliciency from being obtained. They are of such a 
nature that their use in gas producers is not very satisfactory, so 
that until the development and burning of these coals in pulverized 
form was an assured success these coals were not used in as large 
quantities as is now possible. The largest deposits of Ugnite and 
mineral coals appear to be in the Northwest awaiting future de- 
velopment when proper means are at hand for obtaining the highest 



358 PULVERIZED COAL AS A FUEL 

possible economy from their combustion, and the location of these 
large deposits will now be of great value to the districts in which 
they are located. 

9 Around steel plants there are large quantities of waste fuel 
such as coke breeze. This fuel is being used to a certain extent on 
some forms of grates, with forced draft, but it can be burned in 
pulverized form under boilers for the generation of power, and pos- 
sibly in the open-hearth furnaces for making steel. In the anthracite 
field there are large quantities of coal daily pumped back into the 
mines, which coal is a result of the washing and crushing operation, 
for bringing the coal to commercial sizes. This silt or washery waste 
coal carries as high heat value normally as the coals which have 
been marketed. I understand that eight to ten million tons annually 
of this silt are allowed to be pumped back into the mines to fill up 
old workings. 

10 A number of the coal companies are now carefully investi- 
gating the application of pulverized anthracite or low volatile coab 
with a view to using this waste coal in pulverized form so as to 
obtain power from its use, making available coals of higher grade 
for the market, which they are firing at the present time. A very 
successful installation of this kind is in operation at Lykens, Pa. 

11 The above statements have been of a rather geneml nature 
so as to bring out forcibly the fact that coal in pulverized form is 
going to become one of the most important fuels. The results thus 
far obtained have shown that with installations properly designed 
and installed, that from an operating standpoint it is not only a 
desirable fuel but one which will eventually become necessary on 
account of its economy. 

WHAT IS PULVERIZED COAL? 

12 The average man will tell you that pulverized coal is coal 
ground to a powder. Any coal which is ground or powdered from 
his i)oint of view is pulverized coal. From a technical standpoint 
pulverized coal is that coal which is properly dried, crushed and 
pulverized so that the product contains the highest percentage of 
impalpable powder. Merely powdering coal does not fulfill the 
requirements. Coal nmst be pulverized so that at least 95 per cent 
will pass through a 100-mesh sieve having 10,000 openings to the 
square inch, or in terms of dimension 95 per cent must be less than 
one two-hundredth of an inch cube. 

13 The average person does not fully realize to what a high 



N. C. HABBI80M 359 

degree of fineness it is possible to reduce the coal today by pulveri- 
zation. The finer the coal is pulverized the more efficiently it can 
be burned and the more readily it will be diffused when mixed with 
the air for combustion and fed into the furnaces. A cubic inch 
of coal pulverized so that 95 per cent will pass through a 100-mesh 
sieve will contain over two himdred million particles, none of which 
will be greater than one one-himdredth of an inch cube, and a large 
percentage will be less than one six-himdredth of an inch cube. 
A cubic inch of coal has a superficial area of 6 sq. in., but the com- 
bined area of these multitudes of small particles shows that when 
the coal is ground to above mentioned degree of fineness the super- 
ficial area will increase to nearly 30 sq. ft. or an increase in area of 
approximately 700 times. This increase in area permits perfect 
and instantaneous combustion. Its rapidity depends directly upon 
the surface exposure. This is one of the reasons for the grinding. 

14 Pulverizing certainly does not change the nature of the 
coal. We do however change the form of the coal to a certain ex- 
tent in pulverizing it, in that we change it from a soUd into a fuel 
having Uquid properties. As the coal is pulverized it is mixed with 
air, and when handled in the conveyor it flows like water; when 
fed to the furnaces it is more or less like a gas, and the furnaces 
must be designed to bum a gaseous mixture. 

DESCRIPTION OF A COAL-PULVERIZING PLANT 

15 A pulverizing plant consists of three main units: a crusher, 
a crushed-coal drier and a pulverizer. The number of each one of 
these three main units will depend on the size of the plant. The coal 
is dumped in a track hopp>er and conveyed by either a belt or apron 
conveyor into hopper, feeding single-roll coal crusher, or where 
slack coal is at hand, direct to elevator pit. After being crushed to 
about one inch in diameter, it passes by gravity to elevator pit, 
where it is taken by bucket elevators to the drier storage bin. Au- 
tomatic weighing scales may be installed if desired before the drier 
storage bin. Magnetic-separator pulleys are also installed, where 
a belt conveyor is used from the track hopper to remove iron or 
steel scrap in the shape of nuts, bolts, pick-points, wedges and 
such foreign matter, which would interfere with the pulverization. 

16 From the drier storage bin the coal passes to a coal drier. 
This drier must be of a size to deliver the required quantity con- 
tinuously, thoroughly dried. The drier is heated either by hand- 
firing on grates, or by pulverized coal, so arranged in either case to 



360 PULVERIZED COAL AS A FUEL 

avoid igniting the drying coal. The cylinder of the drier is rotated 
by power, either a small motor or Une shaft being used. The dried 
coal falls from the drier through a chute into the pit of an elevator. 
In this chute the coal passes over another magnetic separator to 
remove any final pieces of metal which might be left in the coal and 
which were not caught on the magnetic separator spoken of above. 

17 This elevator carries it to a storage bin set aloft for supply- 
ing the pulverizer. By spouts and gates the coal is permitted to 
enter the pulverizer as desired. This pulverizer grinds the coal to 
the fineness required. From the pulverizer the coal is conveyed in 
various ways to the pulverized-coal bins. With the type of mill used 
at our plant, it is carried by spouts from the mill to the pit of an 
elevator, which carries it aloft to the screw conveyor, which feeds 
the pulverized-coal bins. In another type of mill the fine pulverized 
coal is conveyed by suction fan from the mill to a cyclone separator, 
properly located over the pulverized-coal storage bins, or over the 
screw conveyor to the pulverized-coal storage bin. This separator 
will allow the coarser particles to fall back to the mill, to be r^round, 
while the fine dust passes to the storage bin. No fine-dust elevator 
is necessary with this mill. 

18 These pulverized-coal storage bins are of a capacity propor- 
tional to the service and hold a supply in excess of the amount re- 
quired in the intervals when the grinding is not going on. Thus the 
mills may supply in eight to ten hours all that the furnaces may use 
in 24 hours, by making provision therefor. 

19 If the pulverizing plant is located within 200 ft. of the fur- 
naces and the furnaces are of large capacity, these storage bins are 
located directly at the furnaces. But if there are numerous small 
furnaces located at a considerable distance from the pulverijung 
plant, we then locate the above storage bins at some central point 
and convey from these points to the various furnaces by means of 
one of three methods: first, screw conveyor; second, in a mass by 
means of compressed air; third, in suspension in a current of air. 
We now liave the pulverized coal at the furnaces ready for use. 

COSTS OF PrLVKUIZIX(; COAL 

20 Th(^ cost of tlic operation of pulverizing coal dej^ends upon 
four items: first, the amount of moisture that must \)C expelled 
from the coal l)('f(>ro pulvorizinp; second, the cost of labor; third, 
the cost of coal di^livcred at the pulverizing plant; fourth, the cost of 



N. C. HABRISON 



361 



electricity. Table 1, issued by a pulverized-coal engineering company, 
gives the cost of pulverizing plants, including buildings, and costs 
of pulverizing coal in plants of capacity from 10 to 250 tons per day. 
These figures include all costs, except interest and depreciation, in 
the pulverizing plant proper, and deliver the dust from the top of 



TABLE 1 



COSTS OF COAL-PXTLVERIZINQ PLANTS AND CX)STS OF PULVERIZ- 
ING COAL PER TON NET 



Tods daily 


ToUleost 

pulTerisinc 

per ton, doOan 


Cost of pUnt 

including buildinci 

doUan 


Labor, 
houra 


Labor, 

cost 

per ton, doUan 


10 


0.56 


31.000 


10 


0.80 


20 


0.51 


31.000 


20 


0.25 


30 


0.49 


31.000 


30 


0.23 


40 


0.49 


31.000 


40 


0.23 


50 


0.39 


37,000 


26 


0.13 


60 


0.39 


37.000 


30 


0.13 


70 


0.39 


37.000 


40 


0.13 


80 


0.39 


37.000 


40 


0.13 


90 


0.39 


37.000 


46 

• 


0.13 


100 


0.34 


45.000 


34 


0.09 


110 


0.34 


45.000 


37 


0.09 


120 


0.33 


45.000 


40 


0.06 


130 


0.33 


45.000 


44 


0.08 


140 


0.32 


50,000 


45 


0.06 « 


150 


0.32 


50.000 


47 


0.06 


160 


0.32 


50.000 


50 


0.06 


170 


0.32 


50.000 


54 


0.06 


180 


0.32 


50.000 


57 


0.06 


190 


0.30 


62.000 


48 


0.04 ^ 


200 


0.30 


62.000 


51 


0.04 


210 


0.30 


62,000 


53 


0.04 


220 


0.30 


62.000 


! 56 


0.04 


230 


0.30 


62.000 


59 


0.04 


240 


0.30 


62,000 


61 


0.04 


250 


0.30 


62.000 


63 


0.04 



Data on which Tabls 1 is Babcd 

Labor rate: Millers, 30 cents per hr.; drier firemen, 20 cents per hr.; common labor. 20 cents. 

Cost of drier fuel: 6 cents per net ton. based on 7 x>er cent moisture. Coal at 85 per ton 
delivered. 

Evaporation: 6 lb. per lb. of coal burned or 26 lb. of coal per ton. 

Repairs: 7 cents per net ton. This includes whole pulverising plant, all machinery. 

Power has been based on 12.7 cents per ton pulverised, and a consumption of 17 hp-hr. per 
ton pulverised at 1 cent per kw-hr. or about $54 per hp. per annum. 



the last elevator to the screw conveyor, which feeds the pulverized- 
coal storage bins. 

21 We have taken from Table 1 figures relating to those plants 
having daily capacities approximately the same as our plant, and 



N. C. HABBI80N 363 

into consideration the knowledge obtained from the experience at 
the above-mentioned .plant, and also that obtained from other 
installations, it has been found that furnaces can be successfully 
operated by various methods of applying this fuel. Each tjrpe of 
metallurgical furnace presents different requirements as to the kind 

TABLE 3 CX)ST8 OF PULVERIZING COAh PER NET TON AT THE 

ATLANTIC STEEL CX)MPANY 



Daily Output 80 Tons pbb Dat 

Cost per ton 

Labor SO. 22 

Repairs 0.19 

Power 0.134 

Drier coal 0.0218 

Total Cost 10. 5658 

Daily Output 90 Tons pbb Dat 

Labor 10.195 

Repairs 0.19 

Power 0.184 

Drier coal .0218 

Total Cost 10.6408 

Daily Output 100 Tons pbb Dat 

Labor 80. 176 

Repairs 0. 19 

Power 0. 184 

Drier coal 0.0218 

Total Cost 80.6218 

Total coal pulverized Jan. and Feb. 1919 6276 tons 

Power: 17.9 kw-hr. per ton coal pulv. at | cent 80. 134 per ton 

Labor: 1 man 16 hr. at 80.40 8 6.40 

2 men 16 hr. at 80.35 11.20 

Daily cost 817.60 

(B 80.22 for 80 tons output. 80.195 for 90 tons, 80.176 for 100 tons.) 
Drier coal: Cost of drier fuel 2.8 cents per ton, based on 2.62 per oent moisture. Coal 
at 85 per ton, 6 lb. evaporated per lb. of ooal burned or 8.7 lb. per ton of 
coal pulv. 

Repairs: Total repairs for Jan. and Feb 81412. 16 

Credit 409.69 

Charged 81002.47 (- 80.19 per ton pulv.) 



of burners to be used. Probably the greatest recent development 
in its use has been as a fuel for open-hearth furnaces and boilers. 
Its application to boilers will be taken up later. 

DESCRIPTION OF OPEN-HEARTH FURNACE USING PULVERIZED COAL 

25 All open-hearth furnaces using pulverized coal as a fuel are 
of the reversing type. There has been only one exception to thi8« 



364 PXTLYEBIZED COAL AS A FUEL 

as far as I know, in this country and that exception was at the 
plant of the American Iron & Steel Manufacturing Co., Lebanon, 
Pa., where they fired their open-hearth furnaces from one end only. 
On the other end they installed waste-heat boilers and economizers. 
As an open-hearth proposition this turned out to be a failure, but as 
a waste-heat boiler proposition it was a wonderful success. During 
1918 they remodeled these furnaces and fired them from both ends. 

26 The pulverized coal is delivered into storage bins located 
at each end of the furnace. On the bottom of these bins are screw 
feeders, driven by variable-speed motors for supplying the amount 
of coal desired. This carries the coal by gravity into the burner 
pipe. These burners are usually a combination of compressed air 
at from 60 to 80 lb. pressure and fan air at about 8 oz. pressure. 
In some cases compressed air alone is used as the medium for con- 
veying this coal into the furnaces. The hearth of a pulverized coal 
open-hearth furnace is practically the same as the hearth, of any 
other open-hearth furnace. The uptakes, slag pockets and checker 
chambers are entirely different from other furnaces. The uptakes 
are made as small as possible so as to hold the gases in the furnace 
as long as possible without blowing, and the slag pockets are made 
as large as possible so that the gases will have a sjow velocity going 
through them, thereby depositing a large percentage of the heavy 
particles that arc in the outgoing gases. On account of this heavy 
deposit, removable slag pockets, or ver>' deep stationary pockets, 
should be used, so as to collect thi? accumulation over the run of 
th(^ furnace. Where removable slag pockets are used, they are 
takiMi out and I'leaned and replaced about evcrj- two weeks. 

*J7 Onlv one checker chamber is needed on each end of the 
furnace. If the checker chamber is large enough, these chambers 
should he built u)) with lar^e tiles and laid in such a manner as to 
ftnin vertical Ihjes, having openings of at least 6x9 in. or better 
9 \ 1 1 in. In some easels, no checkers at all are used but the cham- 
bers an* IIIKmI with batlle walls witii openings from the outside, so 
tli.it the aci-unnilatit>n lM»t\\een these ballle walls can l>e raked out. 
.\11 passages from slag pockets to stack nuist be as straight as possi- 
ble and whenever any bends nnist be made, some agitating de\'ice 
should 1k» installed at these points. The reversing valves are asually 
of t\\o nnishroom and damjHT-slide type. 

2S Oil and ^as an* tlie ideal natural fuels. The increase in the 
price of oil has ma<le its use as a fuel praotirally prohibitive. The 
natund-gas supply is rapi<Ily I>eing exhausted, even in those parts 



N. C. HABRISON 365 

of our country which have enjoyed its use for years. Consequently 
we will eliminate these two fuels from our discussion (except in 
Table 4) and compare the use of pulverized coal with producer gas. 

29 The best coal for use in pulverized form in open-hearth 
practice is a bituminous coal as high in volatUe matter as possible, 
and preferably low in ash. It should never contain below 32 per 
cent of volatile, nor more than 8 per cent of ash. For open-hearth 
furnace use it is necessary that the coal be as finely ground as possi- 
ble and it should be so fine that about 97 per cent will pass through 
the 100-mesh sieve, preferably 90 to 93 per cent and not less than 
85 per cent through the 200-mesh sieve, and from 70 to 75 per cent 
through the 300-mesh sieve. 

30 This very fine pulverization is necessary for quick combus- 
tion and for the removal of sulphur in the coal. By this very fine 
pulverization we attempt to have complete combustion before the 
flame strikes the bath, thereby burning out the sulphur in the coal 
to SOt gas, which passes up the stack. In order to get this com- 
plete combustion before striking the bath, some 6 or 8 ft. are 
necessary from the end of the burners to the bath. 

31 The advantages and disadvantages from the use of pulver- 
ized coal, as compared to gas producers, as a fuel for open-hearth 
furnaces, from observation of its use up to date are as follows: 

a Since the coal is of a more even chemical composition all 
the heat units are consumed in the furnace, while in the case of 
the gas producer from 18 to 25 per cent of the heat units are lost 
in the producer itself when converting the coal into gas. This will 
result in a greater number of heats per week. 

b Open-hearth furnaces using powdered fuel operate on a very 
low fuel consumption equal to the best producer-gas practice, and 
much better than the average of the older plants in this country; 
at our plant about 50 per cent less. 

c Coal can be pulverized in plants of about 100 tons daily 
capacity and delivered to the furnace for approximately 50 cents 
per ton, which is about the same as the costs for gasifying coal 
in gas producers. 

d Although the use of this fuel in metalliu'gical furnaces has 
been developed only about 75 per cent, we believe that this devel- 
opment is steadily increasing. In our plant the pulverized-coal 
open-hearth furnace has been shut down oftener than our producer- 
gas furnace of the same size, due to checkers and slag pockets 
filling up with cinders and slag, after about 80 heats; we are 



366 PTTLVEBIZED COAL AS A FUEL 

gradually overcoming these troubles by decreasing the size of our 
uptakes and enlarging the slag pockets, thereby holding the gases 
in the furnace longer and passing them slowly through the large 
slag pockets, so that the heavy particles can settle, and now only 
the fine particles are going to our checkers, which particles are being 
blown off daily by compressed air. By these means, we expect to 
get much longer life out of our checkers, and consequently longer 
runs out of the furnace, since the filUng up of the checkers has always 
been the deciding factor in the length of run of the furnace. On 
account of this continued development, we beUeve that inside of 
six months we will show a 25 per cent increase in production over 
our gas-producer furnaces of the same size. 

e Sulphur does not give us any trouble as long as we have 
good draft and the furnace is working hot, as we are now using over 
1 per cent sulphur in our coal and getting good results, although 
when checkers get clogged up and the furnace begins to blow, due 
to lack of draft, we have trouble with the bath taking up sulphur. 
This takes place during the last week's run of the furnace, just be- 
fore it goes down for repairs. 

/ The pulverized-coal open-hearth furnace is under complete 
control of the first helper as to the amount of coal being used at 
all times, air blast and temperature. 

g The flame, using the same coal as on gas producers, is hotter, 
which allows us to use a greater percentage of scrap per ton of steel, 
thus reducing the consumption of high-priced pig iron. 

h The finished steel is quieter in the molds, due to not being 
overly oxidized, as the coal coming directly in contact with the 
bath has a greater reducing action. We feel reasonably certain that 
the oxidation losses are less with pulverized coal than with producer 
gas, consequently the per cent of product is greater. All gas-house 
troubles are eliminated (cleaning fires, burning out flues, etc.) 
although the pulverizing plant must be given attention as to dryness 
and fineness of coal. 

i Up to date, our refractory costs have been very much greater 
on our furnace using pulverized coal than on our gas-producer fur- 
naces and wiis almost twice as great a year or so ago, although I 
believe on account of the steadily increasing development of the 
use of this fuel, these refractory costs will be steadily decreased. 
As a fair sample, we will consider the life of the roof. When 
originally installed, we got only about 100 heats per roof, but on 
the last run of this furnace we obtained 232 heats. I believe that, 



N. C. HARRISON 



367 



in time, the refractory cost will be nearly as good as gas-producer 
furnaces, but never any better. 

32 Table 4 shows a comparison of fuel costs for all fuels now 
used on open-hearth fiunaces and it will be seen that natural gas 
is not only the ideal fuel but is the cheapest fuel used. As a 
direct comparison in our plant, we will compare the figures shown 
in Table 4 under gas-producer and pulverized fuels, using all cold 
metal, or items 4 and 7. 

33 These figures show 500 lb. coal per ton of ingots using pul- 
verised coal as a fuel, against 739 lb. using producer gas. We have 
also used ^.40 per net ton as the price of coal for both kinds of 
fuel, while we invariably use a larger per cent of run of mine at a 

TABLE 4 FUEL COSTS FOR OPEN-HEARTH FURNACES 



Kind of f od 



1 
2 
3 

4 
5 
6 
7 



Remarks 



Natural gas 

Natural gaa 

Producer gas ' Hot metal 

Producer gas ^ i Cold metal 

Fuel oil 

Tar' 

Pulverised coal 



8 Electric power. 



Amount x>er 
ton steel 



6000 eu. ft. 
6000 cu. ft. 
510 lb. coal 
739 lb. coal 
40 gal. 
40 gal. 
500 lb. coal 
500 kw-hr. 



Rate 

cost f ud, 

dollars 



0.04 per M 

0.12 per M 

3.40 

3.40 

0.02 

0.025 

3.40 

0.0075 



Cost of 

fuel and 

labor, 

dollars 



03 
93 



90 



Cost per 

ton steel, 

dollars 



0.24 

0.72 

1.00 

1.46 

0.80 

1.00 

0.975 

3.75 



Above includes handling cost 

1 Our plant 

' Tar is a waste product at some plants and has to be burned. 



cheaper price on the pulverized-coal furnace. The cost per ton of 
pulverizing coal is 50 cents, against 53 cents for gasifying. This 
gives us a total cost of fuel per ton of ingot steel of 97^ cents using 
pulverized coal against $1.46 using producer gas. 



BOILERS 

34 Many engineers who attempted to bum coal in pulverized 
form obtained unsatisfactory results, and concluded it was "im- 
possible." In many of the earlier trials to burn pulverized coal 
under boilers the usual method was to install coal-feeding devices 
of some kind in the furnace as it stood, with the result that the fire 
bricks melted down and the tubes were plastered with unconsumed 
carbon, ashes and soot. So destructive were the results that we can 



368 PULVERIZED COAL AS A FUEL 

hardly blame those making the tests, from arriving at the conclusion 
they did. How close some were to success was not fully realised. 
Conditions were not ripe. Today, results are being obtained that 
are of sufficient importance to warrant careful investigation and 
consideration. 

35 For the proper combustion of coal under boilers, there are 
five main points which must be given serious consideration, other- 
wise the burning of this fuel will not be a success. These five points 
are: 

a Coal fineness 
b Size of combustion chamber 
c Necessary air opening 
d Proper damper regulation 
e Clean tubes 

36 Coal Fineness. The pulverized coal should run about 96 
per cent through the 100-mesh screen and about 85 per cent through 
the 200-mesh screen. If it nms below 80 per cent through the 200- 
mesh screen, we notice particles of carbon flying through the air 
inside of the combustion chamber, and these particles are not com- 
pletely burned when the gases reach the tubes, and will then deposit 
themselves on the tubes. Also, these heavy particles in the pulver- 
ized-coal mixture will sometimes settle on the bottom of the combus- 
tion chamber and will soon build up in the shape of stalactites. This 
accumulation will continue to build until the bottom of the com- 
bustion chamber continues to be raised, until it comes in contact 
with the flame. These built-up particles will then fuse into a solid 
mass, which, in a very short time, will cause a shut down of the 
boiler to dig this fused mass out. It is, therefore, necessary, for the 
successful burning of this pulverized fuel under boilers, to have 
this coal as finely pulverized as possible. 

37 Size of Combustion Chamber. Before installing, or consider- 
ing the use of pulverized coal under boilers, we must first make a 
study of our boiler-house installation and know at exactly what rating 
we wish to oix^rate these boilers, considering any pxjak load which 
may develop. We must then design our combustion chamber large 
enough to take care of the maximum loads which will be developed 
from our boilers at any time. After this maximum rating has been 
determined, we then figure our combustion chaml>er of a cubical 
capacity equivalent to approximately 50 cu. ft. per lb. of coal burned 
per niin., or approximately 2^ cu. ft. per hp. developed by boUer. 



K. C. HABRI80N 369 

iTwe decide to run this boiler at 150 per cent of this rating and design 
our combustion chamber accordingly, the efficiency will not be 
decreased perceptibly if the boiler is run imder this 150 per cent rat- 
ing, but if we develop over a 150 per cent of the rating, we run into 
serious difficulties. We have to admit more coal and air to the 
boiler to develop the greater rating, consequently we need more 
combustion space, but not having this combustion space, the flames 
impinge on the brick work and cut it away very rapidly. Also, 
combustion is not complete at the time the gases strike the bottom 
row of tubes and consequently the gases will pass up the stack un- 
bumed. Efficiency is then decreased. If the gases are not com- 
pletely binned by the time they reach the first row of tubes, they 
will not bum later on in the boiler. The size of the combustion 
chamber should also be so designed that the velocity of the gases 
should not pass through this combustion chamber at a speed of 
more than 6 ft. per sec. The mixture of air and coal entering the 
combustion chamber, as stated above, should be at as low a pressure 
as is possible to bring this mixture in in suspension; or, in other 
words, breathe it in. 

38 Proper Air Openings. The pressure at which the pulverized 
coal is admitted to the furnace is as low a pressure as can be used 
to carry this fine coal in suspension, and is, I should say, about half 
an oimce pressure at the nozzle. In some installations the coal 
falls by gravity from the variable-speed screw conveyor, located 
on the bottom of the pulverized coal bin, into a fan air line 
which carries it into the furnace and also supplies the necessary 
air for combustion. Some few openings are placed in the front 
wall of the boiler to give any additional air which may be needed, 
and a few may also be placed on the side walls of the boiler to 
protect the brickwork at times. 

39 In other installations the amount of air necessary to con- 
vey the coal into the furnace varies, according to the rate at which 
the boiler is being operated. The balance of the air to bum the 
coal properly is admitted through adjustable air openings in the 
front, sides and bottom of the combustion chamber. These open- 
ings are made adjustable and are placed on all sides of the combus- 
tion chamber to take care of the various grades of coal which may 
be used in the boiler plant. By properly observing the combustion 
in this chamber, by a little experience the fireman knows at exactly 
what points to give more, or less air needed for combustion. 

40 Damper Regulation, In order to give the proper velocities 



370 PULVERIZED COAL AS A FUEL 

of gases passing through the combustion chamber, it is necessary 
that we have very accurate damper regulation to take care of the 
various load conditions which the boiler is to supply. The damper 
should be so regulated that we should practically have a balanced 
draft inside of the combustion chamber and only a slight vacuum in 
the first pass, while at the damper itself we do not want more than 
0.10 to 0.15 in. If we have more vacuum than this, it pulls 
our gases through the combustion chamber too fast, causing them 
to be unburned before reaching the first row of tubes and will then 
build up very fast on the outside of these tubes. This very small 
draft needed at the base of the stack will allow us to operate boilers 
using pulverized coal with stacks of about 30 to 35 ft. in height. 

41 Clean Tubes. In order to get the maximum evaporation 
from any boiler it is necessary that the tubes be kept clean, both 
inside and outside. The keeping of the tubes clean inside is a ques- 
tion of the proper quahty of water and has nothing to do with pul- 
verized coal. The keeping clean of the outside of the tubes is very 
necessary with the use of pulverized coal as a fuel, and they should 
be blown by means of mechanical soot blowers at least every six 
hours, and oftener if it is necessary. Also, once every 24 hours by 
means of a hand-lance steam jet we should blow the bottom of the 
first row of tubes. These are the tubes in which the gases come into 
contact first after leaving the combustion chamber. This material 
can be blown off easily if the combustion chamber has been 
properly constructed and if they are blown regularly as needed. 
But if they are not blown regularly this material builds up and 
accumulates very fast and in time will become fused and cannot 
be blown off. 

42 Another item which might have been included in the five 
main points above, and might have been called the sixth point, is 
the removal of ash which deposits at the bottom of the combustion 
chamber. As spoken of above, this ash should be removed at regit- 
lar intervals, which intervals will be determined by the amount of 
ash in the original coal. If we do not remove this ash regularly, it 
will build u[) until it comes in contact with the flame, when it becomes 
fused and lias to be dug out. But if removed at regular intervals, 
it can be (\isily raked out with the ordinarj' boiler-room ash rake 
and will not consume more than half an hour per 24 hours, and will 
not interfere with the operation of the boiler while this is being done. 

43 The following are some advantages of pulverized coal as a 
fuel for boilers over stokers: 



N. C. HABRISON 371 

a Much wider variation in the quaUty of the coal usable is 
obtained when burning coal in pulverized form. Practically any 
and all grades of coal can be burned in this form with economy. No 
stoker will satisfactorily handle all grades of coal. Therefore the 
use of pulverized coal will largely overcome most troubles due to 
poor coal, and it is particularly desirable for this reason alone. 

b The ability to take care of peak loads almost instantaneously. 
In other words, a pulverized-coal burning system is much more 
flexible than a stoker installation. Its flexibiUty approaches that of 
oil or natural gas. 

c The amount of coal that can be burned per sq. ft. of grate 
surface on stokers is limited so that for increased capacity the boiler 
setting must be spread out to cover more area. When using pulver- 
ized coal, this condition does not exist, for proper furnace conditions 
can be obtained by increasing the height of the boiler setting or 
the depth of the combustion chamber. 

d By throwing a switch the entire firing operation ceases; an 
advantage in case of- accident or emergency. 

e Ash is in much better condition to handle. The ash is in the 
form of a dust or slag depending upon its melting point. This helps 
to maintain constant furnace temperature as there are no interrup- 
tions to firing conditions on account of cleaning fires. 

/ Since there are no grates used when the fuel is burned in pul- 
verized form, we experience no clinkering of grates, as in the case 
of stokers, particularly after operating at maximum rating. 

g Pulverized coal is fired dry, containing less than 1 per cent 
of free moisture, whereas coal burned on stokers may vary any- 
where from 1 to 10 per cent, of free moisture as fired. 

h Considerable less excess air is necessary for complete com- 
bustion. This item is of the utmost impoi-tance when making com- 
parisons. Less excess air means less power for furnishing air supply 
particularly where forced draft is used. With less excess air the 
stack losses are less. Ix)wer grades of coal fired on stokers require 
more excess air as it is quite difficult for the ox>'gen to get in close 
contact with the com})ustible. An air supply sufficient to furnish 
all the air for combustion should be available, although at times 
only 50 per cent of the air is necessary' to be injected into the fur- 
nace with the coal, the balance being supplied by the induction 
action of the burner or drawn in by the stack draft through the 
various adjustable openings in front and sides of combustion cham- 
ber. The air going into the furnace should be under control to per- 



372 PULVERIZED COAL AS A FUEL 

mit close regulation under all conditions of firing. Less draft is 
required for pulverized-coal-fired furnaces. 

i All the combustible in the coal is consumed when it is burned 
in pulverized form, providing the furnace capacity is not exceeded. 
None of the combustible goes out into the ash pile and therefore fires 
are eliminated in the ash pile. 

j There is less erosion from sulphiu* on the boilers due to less 
moisture in the coal as fired, therefore highnsulphur coals can be 
burned more readily and without serious results. 

k With furnace properly proportioned and with properly 
designed burning equipment smokeless operation may be main- 
tained indefinitely. This is due to complete combustion of all the 
particles of coal before coming in contact with the cold surface of 
the tubes of the boiler. 

44 The following few points must be kept in mind for the suc- 
cessful burning of pulverized coal imder boilers: 

a A boiler furnace using pulverized coal should have as few 
burners as is possible consistent with good regulation. The burners 
must be proportioned for the maximum rating of the boilers, and 
they must be adjustable. SimpUcity of design is desirable. It has 
been found much more desirable to introduce coal into the furnace 
as far away from the side walls as possible so that the rapid continu- 
ous expansion of the gases will not develop high velocities in dose 
contact with the furnace refractories. Fiunaces under boilers should 
be proportioned so that the velocity of the gases should not be 
excessive, particularly at the smallest cross-sectional area of the 
furnace. Vertical baffles should replace all horizontal baffles. 

b A boiler of any size can be fired successfully with pulverised 
coal. Various designs and makes of boilers can be readily arranged 
for pulverized-coal firing, but those containing the smaller percent- 
age of space for the lodgment of ash are preferable. 

c Feeders for regulating the flow of pulverized coal to the fur- 
nace must be designed so that at all times the variation in quantity 
will be directly proportional to the speed of the screw and no flood- 
ing allowed. The speed of the feeder should be so regulated tibat 
operating at its maximum r.p.m. the supply of pulverized coal to 
the furnace will not exceed the capacity of the furnace. Soot blowers 
should be installed in settings where pulverized coal is used. 

d The equipment for using pulverized coal is standard for any 
grade of coal so far as handling, preparing and delivering to the 
furnace is concerned. Only a slight change is necessary in the fur- 



N. C. HABRISON 373 

nace to take care of coals of very low volatile content, such as an- 
thracite, culm and coke breeze, and increased dr3ang capacity is 
desirable when lignite coal is used. With stokers this is not the 
case as the varying quality of coals require different type stokers 
to obtain highest efficiency. 

e The labor required to operate a pulverized-coal installation 
may be of higher class but the number of men required will be less 
in the larger installations than that required for a stoker installa- 
tion, thereby affecting a saving in the labor charge in favor of pul- 
verized coal. 

45 The following is a report of a test made on a 468-hp. boiler 
using pulverized coal as fuel. This installation is noteworthy not 
only by reason of the high efficiency obtained, but also because of 
the fact that it has made clear some of the conditions necessary 
for the successful operation of boilers utilizing powdered fuel. 

46 When the boiler was first put into operation, a number of 
undesirable conditions resulted. An insufficient air supply caused 
high-furnace temperatures resulting in fusion of the ash particles 
and a consequent accumulation of slag between the tubes, on 
the furnace walls and in the ashpit. The removal of the molten 
slag presented considerable difficulty. It was also found that the 
combustion chamber was of insufficient size. High gas velocities 
resulting from insufficient air in the chamber tended toward de- 
struction of the refractory surfaces of the furnace. 

47 A new furnace was, therefore, designed. The combustion 
chamber was enlarged and a regulated air supply was provided for 
by means of a number of auxiliary air openings equipped \\dth damp- 
ers. The accumulation of slag in the pit was prevented by raising 
the point of admission of the fuel into the furnace. As a result the 
flame path has been raised above the base of the pit, hence particles 
of ash dropping from the flame are not fused. The ash, therefore, can 
be drawn from the pit in the form of a powder and small slugs of slag. 
Analysis has shown that the ash contains practically no carbon. 

48 Having established satisfactory furnace-operating condi- 
tions, a series of efficiency and capacity tests were conducted pre- 
liminary to proving the contract guarantees. The brickwork was 
then given a thorough trial by carrjdng the boiler at a continuous 
rating of 180 per cent over a period of several days. On August 12 
and 13 a final efficiency test, the results of which are given below, 
was run. The boiler is a three-pass water-tube boiler, equipped 
with a superheater. 



374 



PULVERIZED COAL AS A FUEL 



TABLES LOG OF TEST OF A PULVERIZED-FUEL-BURNINQ STATIONARY 

BOILER. DATE AUGUST 12-13, 1918 



Make of boiler Edc6 Moor 

Rated hp 468 

Heating surface, sq. ft 4685 

Time fired or test started 11 . 15 A.m. 8/12/18 

Time fire out or test finished 11 . 15 a-m. 8/13/18 



Duration of test 

Maximum Minimum 

Temperature of boiler room (deg. fahr.) 99 85 

Temperature of feedwater 168 135 

Temperature of steam (deg. fahr.) 477 427 

Barometer in. of mercury 29.35 29.20 

Temperature of flue gases (deg. fahr.) 515 455 

Average boiler pressure, lb 

Atmospheric pressure, lb 

Temperature of steam, deg. fahr. . . 

Superheat, deg. fahr 

Safety valve set for, lb 

Fuel fired per hr., lb 

Total fuel, lb 

Total water, lb 

Water apparently evaporated per hr., lb 

Water apparently evaporated per lb. of coal, lb 

Factor of evaporation 

Water evaporated from and at 212 deg. fahr. per lb. of coal, lb 

Maximum Minimum 

Carbon dioxide (COt) per cent 15.4 12.2 

Oxygen (O) per cent 5.6 3.2 

Carbon monoxide (CO) 

Fuel used Bituminoui 



24 hr. 
Average 
93.3 
157.2 
448.7 
29.25 
495.3 
167.0 
14.4 
373.8 
74.9 
175 
1.990.6 
47.775 
893.168 
16.393.0 
8.23 
1.150 
9.47 
Awagc 
13.85 
4.38 
None 



Fuel analysis No. 1 

Amount of coal represented by each sample, lb. . . 19,775 



No. 2 

20.000 
41.1 
11.0 
36.96 
49.13 
13.91 
2.06 

10,763 

12,093 



No. 3 
8 ,000 
1 6.9 
9.7 
38.77 
48.29 
12.94 
2.12 
11.263 
12.473 



ATsragG 



Per cent of total 41 .3 

Moisture (per cent) 10.3 

Volatile (per cent) 33 .81 

Fixed carbon (per cent) 50 .43 

Ash (per cent) 14 .36 

Sulphur (per cent) 1 .90 

B.t.u. as received 10,600 

B.t.u. dry 11,817 

Vacuum in burner, in 

Vacuum under primary arch, in 

Vacuum in combustion chamber, in 

Vacuum in first pass, in 

Vacuum in second pass, in 

Vacuum in breeching, in 

Feoder speed, r.p.m (No. 1), 53.6; (No. 2); 50.7 

Coal per rev. of screw, lb 0.81S 

Acruniulation of slag on tubes NoiM 

Flues blown during test 6 tjmn 

Ofxyration of furnace Very 

Pupation ff{ 



10.49 

35.96 

49.58 

13.98 
2.04 
10.779 

12.045 
0.000 
0.000 
0.000 
0.000 
0.0057 
0.09 



Condition of smoke li^t 

Heat effect on brick Nc 

Back loMh of flame in burner N4 

Pounds of steam i>or hr. from and at 212 deg. fahr 18,842.0 

Horsepower 546. S 

Per cent of rating 116.7 

Boiler eflliciency, per cent 85.39 



N. C. HABRISON 375 

MononuidA — Fuel-preparation deduotion: ^ 

Coal uaed in drier, lb 1.140 

Motor c^eration 449 .3 kw-hr. 

Coal eqi^valent at 3 lb. per kw-hr., lb 1.348 

Total deduction, lb.. 2.488 

Reniltinc net efiEidency, per cent 81 . 1 

^ No deduotion made for stand-by loeses in drier. 

49 At this same plant are other boilers fired by one of the most 
efficient types of underfeed stokers. A comparison is made between 
results of above test and tests made on the stoker-fired boilers. 



PULVERIZED COAL VS. MECHANICAL STOKERS 

50 Under this heading fuel-preparation costs will first be con- 
sidered. In the case of powdered coal this can be classed under 
three general divisions: 

a The cost of crushing the coal. This expense is the same for 
pulverized-coal equipment as for stokers. 

b The cost of drying and pulverizing the coal. Although no 
cost records s,ve available at present, it is estimated that 32 cents 
per ton will cover this preparation cost on a 200-ton-per-24-hr. plant 
using bituminous coal containing about 12 per cent moisture. 

c The maintenance costs of the drying and pulverizing plant. 
This imit has not been determined from actual experience; however, 
it is estimated that 3 cents per ton will cover the maintenance. In 
stoker practice the maintenance cost per ton of fuel fired is close 
to 5 cents per ton. 

51 Summarizing the above facts it is evident that, with fuel at 
$5 per ton, the gross efficiency shown by the pulverized-fuel boilers 
will have to exceed that shown by the mechanical-stoker-fired 
boilers by 6 per cent in order to offset coal-preparation costs. A 6 
per cent deduction from a gross efficiency of 85.22 per cent results 
in a net efficiency of 79.22 per cent for the powdered-coal burner. 
In stoker practice the maximum attainable gross efficiency at any 
of our plants has been 80.54 per cent. Deducting the 2.5 per cent 
for auxiUary uses, the resulting net efficiency is 78.04 per cent, 
which is lower by 1.18 per cent than the figure obtained in pulver- 
ized-fuel practice. 

52 Other advantages resulting from the use of pulverized fuel 
are summarized herewith: 

a Continuous boiler operation at a uniform rating as well as a 
constant efficiency is made possible. At no time is there a loss in 



376 PULVERIZED COAL AS A FUEL 

capacity due to the clinkering of coal on the grates or the cleaning 
of fires, as is the case in stoker practice. 

b Heavy overloads can be taken on or dropped off in a very 
brief time through adjustment of the coal feeders and the furnace 
drafts. 

c From 97 to 98 per cent of the combustible in the coal is utilized, 
regardless of the quality of the fuel. 

d The ash-handling costs are reduced to a minimmn due to the 
reduced volume. 

e The banking conditions when operating with pulverized coal 
are somewhat different from those obtained in stoker practice. By 
stopping the fuel supply and closing up all dampers and auxiliary 
air inlets a boiler can be held up to pressure for about 10 hours. 
The furnace brick work having been heated to incandescence during 
operation gives off a radiant heat which is absorbed by the boiler 
rather than being sent out through the stack. The ease of controll- 
ing the fuel, feed and drafts, the abiUty to take on heavy overloads 
in a brief time, the thorough combustion of the coal and the uniform 
high efficiency obtainable imder normal operating make pulverized 
coal a most satisfactory form of fiiel for central station uses. 

53 The full story of maintenance expense is only partly known 
as yet, however. Indications are that no unusual difficulties will 
be met. The cost of fuel preparation and labor for operating a 
boiler room fully equipped. with pulverized-coal-buming boilers will 
be a question for the engineer to decide for himself according to his 
particular conditions. If properly installed with respect to capacity 
of storage, size of drier and pulverizers, and on a sufficient number 
of boilers to properly and fully employ the minimum number of 
men, the pulverized-fuel installation will undoubtedly be more ad- 
vantageous. The main item that must be borne in mind by engi- 
neers is that the ease with which a high efficiency is obtained and 
the constant nature of that efficiency, as compared to the lack of 
constancy of efficiency in a stoker-fired boiler, unless very doeely 
supervised, is the one factor about the burning of pulverized fuel 
which justifies its use. There is no doubt that with a well-equipped 
plant burning pulverized fuel, having all the necessary recording 
and indicating instruments to guide the operators in maintaining 
the proiK?r conditions, a lower cost of generating steam will be 
[)ossil)le than has heretofore been the case in any type of equipment. 



No. 1702 

PULVERIZED COAL FOR STATIONARY 

BOILERS 

By Fred'k a. Scheffler, New York, N.Y. 

and 

H. G. Babnhubst, Allentown, Pa. 
Members of the Society 

It is the purpose of this paper to present those facts which the authors believe 
indicate the coming general adoption of pvloerized coal as a fud for boilers. Since 
stoker firing is the most efficient method when solid fuels are used, a comparison of 
stoker and putotrized-fuel plants is given, with particular reference to reliability, 
cost, adaptability and efficiency. The cost of pulverizing coal and the cost of stoker 
operation are discussed in detail and tables given showing results of tests on pulver^ 
ized-fud plants and data regarding boiler installations using pulverized coal as a 
fuel. 

T^HE authors, being of the opinion that best present-day prac- 
tice of firing boilers in power plants of moderate and large 
capacity has attained the maximum efficiency that might be ex- 
pected, believe that if we are still further to conserve the coal fields 
of the coimtry and in addition reduce the operating costs in the use 
of fuel and labor, some other method of firing boilers by coal must 
be adopted. 

2 The principal increased costs of power-plant operation dur- 
ing the past three or four years have been due to the increased cost 
of coal and labor, which in many cases have alone added 100 per 
cent to the cost of operating the boiler plant. It is also safe to say 
that the improvements in the turbo-generators, condensing equip- 
ment and large-size steam units now being used with increasing 
boiler pressure and high superheat may possibly not permit of a 
still lower water rate. We must therefore turn to the boiler-furnace 
equipment for further reduction in operating costs, and the authors 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
AifEBiCAN Society of Mechanical Engineers. 

377 



378 PULVERIZED COAL FOB STATIONARY BOILERS 

respectfully submit the present discussion on the use of pulverized 
or powdered coal with a view of demonstrating a practical com- 
mercial solution of this highly important question. 

3 Although a very great interest has been aroused throughout 
this country and abroad in the adaptation of pulverized coal to 
boiler furnaces, it is remarkable how little is known, in a practical 
way, of what is actually being done, how it is accomplished, and 
what results have been obtained. It is hoped, therefore, that the 
data and illustrations presented will prove of interest and will call 
forth discussion which will serve to develop further this study of 
the best method of reducing power-operating costs. 

COMPARISON OF STOKER WITH PULVERI ZED-FUEL PLANTS 

4 Since the purpose of this paper is to present the facts which 
indicate the coming general adoption of pulverized coal as a fuel 
for boilers, the discussion is presented in the form of a comparison 
with stoker firing, the latter being the most efficient method in 
general use for burning sohd fuel under boilers. 

5 The ultimate adoption of a new method depends entirely on 
its overall commercial efficiency. In the generation of power, over- 
all efficiency may be considered as composed of the following fac- 
tors: reliabihty, cost and adaptabihty. A method may acquire a 
wide field if it shows improvement in any one. or two of these points. 
Improvement in all three points leads to the general superseding of 
other methods. 

6 Reliability. Let us compare the reliability of a pulverixed- 
coal installation with that of stokers. This factor depends on two 
items: apparatus for preparing and presenting the fuel for combus- 
tion, and continuity of operation of the furnace itself. In a stoker 
installation the first of these includes the stoker itself. Neglecting 
the inherent defects of any system that presents a metal mechanism 
to the action of high temperatures, it may be admitted readily that 
the stoker system is satisfactorily reliable, with respect to its appara- 
tus, for preparing and presenting the coal for combustion. 

7 The corresponding mechanisms for pulverized fuel are equally 
reliable. This fact is proved by their widespread use for years 
in the cement industry and more recently in an ever-increasing 
variety of industries. It should be recognized that these mecha- 
nisms are not innovations, but are the result of years of development 
under oix^atin^ conditions. Proi)er design of equipment by engi- 



W. A. BGHSFTLBR AND H. G. BARNHXJB8T 379 

neers of standing who are specialists in this Une has made negUgible 
the danger of dust explosions, the occasional occurrences of which 
in years past have furnished ammunition to the opponents of pul- 
verized fuel. 

8 The second condition for reUability is the continuity of opera- 
tion of the furnace. Here again we find an apparent balance be- 
tween stoker and pulverized-fuel installations during operation. 
The advantage Ues with pulverized fuel, however, for several rea- 
sons. The mechanism is altogether outside the furnace, hence 
cleaning and adjustment and the making of the few repairs required 
need not interrupt the operation of the boiler. In case of sudden 
necessity the fire may be ignited and quickly brought to full inten- 
sity, or it may be extinguished almost instantly. Greater uni- 
formity of flame and temperature is conducive to longer Ufe of the 
furnace lining in a properly designed furnace, and to the minimum 
variation in furnace efficiency. Finally, the pulverized-fuel installa- 
tion reUeves the power plant from dependence upon the availabiUty 
of a certain grade of coal. Stokers will not handle all grades of coal. 

9 Cost, This second fac.tor refers to the cost per B.t.u. de- 
livered to the boiler. The various items entering into this cost by 
the stoker system comprise power, repairs and maintenance, labor, 
interest on investment, depreciation, insurance and taxes. 

10 With pulverized-coal equipment the cost of fuel for the 
drier should be added to the preceding items. A moment's considera- 
tion will show that this item must be taken care of in the furnace of 
a stoker-fired boiler and that it is clearly cheaper to remove the ex- 
cess moisture content from the coal in a drier, from which the gases 
leave at very low temperature, than in the furnace itself, where the 
evaporation of the moisture damps the fire, increases the content 
of inert gases and at the same time carries off a very perceptible 
amount of heat. 

11 Returning to the balanced cost items, it appears that these 
show a saving in favor of pulverized fuel in a large power plant and 
for the stoker in a small power plant. The figures in question are 
discussed further on in detail. It should be noted that when central 
pulverizing plants are built, they will relieve small power plants 
of the necessity for maintaining pulverizing equipment and make 
pulverized fuel considerably cheaper than stoker-fed fuel, regardless 
of the size of installation. This feature is already being carried out 
successfully. 

12 The final factor of the cost, furnace efficiency, which gov- 



380 PULVERIZED COAL FOB STATIONARY BOILERS 

erns all the others, results in all respects to the advantage of pul- 
verized fuel for the following reasons: 

13 First: The fuel enters the combustion chamber in a finely 
divided state, being introduced with air at low pressure, and is ap- 
proximately perfectly mixed with the air for theoretically perfect 
combustion. Therefore no excess air is required for complete com- 
bustion. The imits of heat taken up by heating excess air reduce 
the combined boiler and furnace efficiency. It is impossible to get 
a uniform fuel bed on a stoker or a grate and, therefore, impossible 
to approach complete combustion without introducing excess air. 
Should it be desired for other reasons to introduce excess air with 
pulverized fuel, it can be done in exact amounts, evenly distributed, 
without affecting the uniform nature of the flame and flue gases. 
This uniformity, which cannot be obtained in either grate or stoker 
installations, means maximum efficiency in all parts of the furnace 
and a maximum rate of heat transference to the boiler throughout 
its exposed area. It also means that flue-gas analy^ gives an accu- 
rate determination of conditions in the furnace, and that control of 
coal delivered and air supply can be adjusted with great accuracy. 

14 Second : In pulverized form all of the combustible is burned, 
a consummation certainly impossible in lump-coal firing by either 
hand or stoker. It is not unusual to find 20 to 30 per cent of carbon 
in ash refuse from grate- or stoker-fired boilers. 

15 Third: With pulverized fuel there are no standby losses 
with change of load or when shutting down, such as banked fires, etc. 

16 Fourth: With properly designed pulverized-fuel apparatus 
nothing of a mechanical nature takes place in the furnace. In 
stoker and grate firing not only is the mixing with the air done in 
the furnace, but the presentation of fresh surfaces of combustible 
to the air supply must take place by the removal of the ash and 
its discharge through the grate bars, or the pressure must be great 
enough to force the air supply through the ash bed. 

17 AdaptabilUy. Let us now consider the third factor of overall 
efficien(!y, which is adaptability. Pulverized fuel is here preeminent. 
The primary feature is the possibility of burning all grades of fuel 
without affecting the efficiency of the furnace. To bum anthracite 
and very low grades of fuel requires a furnace allowing a return flow 
of the flame past the incoming flame, to heat up the incoming fuel, 
and in a furnace of this iypQ fuel containing over 50 per cent ash 
has been burned with high efficiency. The stoker is very much 
restricted in comparison. 



F. A. SGHSFTLBB AND H. Q. BARNHT7B8T 381 

18 The flexibility in the use of pulverized fuel is perfect, and 
the fire may be instantly adjusted to suit any condition of overload 
or lower load, including the cutting in and out of the boilers. The 
paramount importance of this feature and the utter impossibility 
of approaching it with stoker or grate firing is readily evident. 
Furthermore, the operation and the determination of conditions for 
complete combustion may be made automatic, the result being a 
smokeless and sootless boiler plant, which is essential in modem 
cities. 

19 Furnace Design, A few words on the design of furnaces 
for pulverized fuel may be of interest. The primary requisite for 
good results is to maintain low velocities in the furnace. The com- 
bustion is no less perfect with high velocities, but this will result 
in damage to the linings and in their erosion. A furnace cubical in 
shape usually gives the most satisfactory results. 

20 The burners should inject the coal under low pressure and 
should permit of varying the density of the mixture in the burner 
itself. Their location and number will depend upon the size of the 
boiler and rating required, and also may be varied to suit the grade 
of fuel. High boiler ratings such as are used in modem boiler prac- 
tice can be obtained when desired, and such overratings should be 
predetermined and the furnace volume designed accordingly. 

21 It will be noted that pulverized coal behaves more nearly 
like liquid and gas fuels than it does like lump coal and that it is 
in the ideal state for burning with the highest possible eflBciency. 
It has been shown that it is superior to lump coal as regards all three 
factors of overall efficiency and these statements are susceptible 
of proof upon investigation. The novelty of the pulverized-fuel 
plant is rapidly beginning to disappear, and on accoimt of the fact 
that all obtainable coals are apparently becoming more inferior in 
quaUty, the interest in the use of pulverized fuel is very general 
throughout the United States and other coimtries. 

22 In Table 1 will be found an itemized statement of the costs 
of pulverizing coal, and in Pars. 34 and 35 some statements as to 
the cost of stoker operation for comparison purposes. Table 2 gives 
a list of boiler installations using pulverized coal and Table 3 reports 
of preliminary tests made on some of the pulverized-fuel installa- 
tions now in operation. WTiile these do not show the maximum 
efficiency to be expected with the further development of the art, 
they nevertheless indicate that the inherent difficulties have been 
solved and that at the present moment pulverized fuel is in a posi- 



382 



PULVERIZED COAL FOB 8TATI0NABT BOILERS 



tioii to coiui)etc advantageously with any other method of burning 
Holid fuel under boilers. 



COMPARISON ON AN EFFICIENCY BASIS 

2Ii Olio of the most prominent engineers in this country, a 
nioiulxT of the Society, has stated that the combined boiler and 
furniKH* (»(ru!ion(5y by the month, day in and day out, of a modem 
stokor-lirod power pbmt with the best average plant operation is 

TABLIO 1 COST OF DELIVERING PULVERIZED FUEL TO BOILERS 



100-ton plant, I 1000-ton pUnt, 



Puwvr At \ cent prr kw-!ir. and 17 kw-hr. per net ton . . . . • 

I<«bor At 40 oonta per hr 

Drier ct»al At $5 iH>r net ton lielivered 

Ht^lHkin*: ; 

Total Actual o«.>«it of pulveriiinK l>cr net ton i 

Interest at t» i>er i^nt 

IVprtviation 

'rAxea Ami iiwutAnee 

Total cviat \^t net ton 



doUara per 


doUara per 


net ton 


net ton 

1 


$0.1275 


1 

> $0.1275 


O.U 


0.04 


0.06 


0.06 


0.07 


0.07 


0.3975 


0.2975 

t 


0.105 


0.039 


0.12 


0.04 


0.035 


0.013 


0.6575 


0.3895 



not Ivttor than fnnu (Vi to l>o |vr wnt, although a carefully con- 
iluotiHl tost on one Innlor and furnace might show during several 
hours' run To ^xt ivnt otlioionoy. This statement has been oon- 
tirnu\l by other engineers. 

24 The n^suUs with pulverizevl fuel would lx> totally different. 
Thon^ is lu^ apivvriMit n\iSvH\ why a cvnubiiuxi furnace and boiler 
ethoioiu\v of ?."> |vr vvnt. and even hii::her, oi^uKl not be 
thrv^ucluHit t!io year, as the ojvnition of the plant would be 
tu\:»Hy oviuivalotit to that of a fuel-^nl installation, in which stand-by 
IvVvms. l\»:.ki\i ti:x^s. oto . ar\^ alnKv>t entirely eUiuinatt\l. Unque»> 
:-ov..iM\ li.oiv >':,o;f,vi Iv a scwiuj:^ under ihos^^ eireunistancM, of 
rj tv^ 1 "» :vr o\ :.: v^f t!ie total o\\d vvusumption in fa\\>r of pulvrriaed 
v\\»*. :»v.vi :i.> rwi.utior,. on a Kvsis of even ^ ilXX^xnlor-hp. plant, 
w;** >1..'\\ ., \ory fair r\:uru v>n the in\>^suuent. ne^livtio^ ibe fact 
:V.a: a *.vn\t r-cr.uio anvi cheaivr cvvU cvmld U* uianI. 



F. A. 8CHSFFLEB AND H. G. BARNHUBST 383 

COST OF PULVERIZING COAL 

25 The cost of pulverizing the coal is of prime importance as 
low costs are essential for success and are achieved when the quan- 
tity used per day of 24 hours exceeds 100 tons. The cost of pul- 
verizing is made up of a number of items as follows: 

Power Interest 

Repairs and maintenance Depreciation 

Coal for drying Insurance and taxes 
Labor 

26 Power. The power required in an up-to-date pulverized-coal 
plant is from 12 to 13 kw-hr. per net ton of coal crushed, dried and 
pulverized. The additional power required for transferring the coal 
to the point of use and feeding it to the boilers will vary consider- 
ably, depending upon the distance transported, the size and number 
of the boilers, and the conditions under which they operate. The 
power required for this latter purpose varies between 4 and 6 kw-hr. 
per net ton, so that the total power for the entire process from the 
track and storage deUvered to the boilers is 17 or 18 kw-hr. per net 
ton. In the following paragraphs the cost of power has been as- 
sumed at f cent per kw-hr. 

27 Repairs. The item of repairs, including material, labor and 
general upkeep of the plant or maintenance, for the entire pulveriz- 
ing plant and burning equipment will vary from 7 to 10 cents per 
net ton of coal handled. The figures depend upon local conditions, 
and the size and general arrangement of the entire installation. 

28 Drier Fuel, The item of coal for dr>ang depends directly 
upon the percentage of moisture and upon the price of coal. Ordi- 
narily only from 1 to IJ p)er cent of the total amount of coal used is 
required for drying. Assuming coal to have an average of 7 per 
cent moisture as received and the cost to be S2.50 per net ton, the 
cost per net ton of drying the coal will be 3 cents. At $5 per net 
ton the cost of the drier coal will be 6 cents. 

29 Labor. This item is the greatest variable in connection 
with the pulverizing of coal, due to the increased output that can 
be obtained in larger plants per man employed. It is also subject 
to local rates of wages. For example, assuming labor at 40 cents 
per hour, a plant of 100 tons daily capacity, properly designed and 
equipped, will require approximately 34 labor hours to prepare the 
fuel and deliver it to the conveyors, whereas in a plant having a daily 



384 



PULVERIZED COAL FOR STATIONARY BOILERS 



1^ 

8 



1^ 
O 

z 

QQ 
P 

o 
o 






-a 

o 
O 



i! 
i! 



s 



&*»: 



I 

II 



o 

n 



•ss 



o 



I 



a 

a 

8 



s 



o 

«8 

-a 



P 



6 



I 
- a 

■21 






s 



s 

I 

3 



I 

o 






I 
S 

1 

a 
O 



o 

•c 






i 






9 

1 



^.9 



^3 



i 






ss s s 



s 



M S 




» 



CO 



•a 

00 



CO 



•1 C« iH 



CI 






1I 



9 



3 

P 



I 1 

o o 



f 

.a 



<3S 



a 



S 

V 

•mm 

& 

a 

< 



> 

GO 



« 

a 

« o 

§ ? 

•H ** 

» a 

— ^ '-' « 



n n n pq II 4 g • 



lO «0 



1 

I 



ca 






•d 

a 

2d 
•I 



o 
>!» 

■Si 

•§3 

A > 



GO 
X CQ 



a 
*> .2 

«. JS 

-1 






<D 00 



o 5 



00 

I 



00 

Ok 

i 



00 



o 






^ SI 

2S 



00 

Ok 
K 

3 



r. A. SCHEFFLBR AND H. O. BABNHCR8T 



385 



I 



CO 

Z 

o 

H 

-< 
iJ 
•J 

•J 

o 

n 

e« 




:g 



S 



o 
o 

(g ! O '2 K Ph 

Pk 
o 

QQ 



^•s a 

I a • 

— •? < 
< O M 



o 
O 

o — 

"S r5 O 

o X O 
o 
je *« 2 

o e* u 

^ »5 Ph 



C4 



0) 



a 

GQ 



eo 



^*S 



<« « 



» 



PQ CQ 00 



I I 



M 



GO 
M 



>% 

2H 

^ «< o 

CI M CI 

d t^ ^-1 ^4 

CI d «6 CI 



I 

! 



f^ w^ o CO '^ 



CI 



CI 



w 



CI 



PU 

S 



o 

O 

e 
o 



o 
O 

flS 

c 

c 

ii 

j= 

cr 



Ok 






o 
O 

•a 

o 






J3 
4 



o 



73 
CI 



o 
O 

C 



SB 
o 
CD -r 

OS 

o 



1 


1 


1 


U3 

a 


& 


PC 


^ 


^ 


« 


« 


« 




•^J 


3 


*« 


'«■> 


«s 


flS 


08 


08 


^ 


(^ 


(^ 


S 



1 

•i 
? 
1 

c 

3 

a 

a 

«>• 

'** 
t 

i 

S 

s 






6 

QQ 

CO 



s 

3 

o 

08 
'*' 
08 

z 



'u 

O 

*^ 

s 

08 

-a 

♦^ 

OD 



o 

o 

"oS 

o 
O 

a 

o 
O 

o 

to 

c 

08 

a. 



A 9 00 
9 0) 



CC 
O 



o 






QO 



00 



00 

l-t 



00 



^ o a> 

c8 C C 

^33 



> 

o 



« 

a 

3 



08 



a 



a a 



a 



386 



PXTLVEBIZEO COAL FOB STATIONABT BOILBBS 



capacity of 1000 tons, approximately 115 labor hours are required. 
Therefore the labor cost would be 14 cents per net ton in a 100-ton 
plant, only 4 cents per net ton in a 1000-ton plant, and as low as 
2i cents per net ton in a plant of 5000 tons daily capacity. 

30 Interest, The interest item is based on 6 per cent of the 
entire investment, and the cost of the pulverized-coal plant and 



TABLE 3 REPORT OF PRELIMINARY TESTS MADE ON PULVERIZED- 

FUEL PLANTS 









1 


Effici- 








Date of 
test 


Location 
of plant 

Seattle, Wash. 


Duration, 
hr. 

14.5 


Coal used 


ency 
main- 
tained, 
per 
oent 

77 


B.Lu. 
per lb. 
of coal 
as fired 


Adi. 
per 
cent 


Rating, 

per 

oent 


Apr. 16, 1917 


Renton buckwheat 


10,000 


11.60 


122 


Dec. 4. 1917 


Chanute, Kan. 


5 


Kansas bituminous ! 


72 


11.996 


17.7 


125 


Deo. 12, 1917 


Chanute, Kan. 


5 


Kansas bituminous 


83.94 


12,500 


18.25 


125 


Jan. 28. 1918 


Chanute, Kan. 


(25 days) 


Kansas bituminous 


78.1 


11,435 





100 


Apr. 26, 1918 


Parsons, Kan. 


6 




80.3 


12,900 


17.49 


• • • « 


Apr. 28. 1918 


Parsons, Kan. 


6 


Kansas bituminous ' 


80.9 


12.289 


17.49 


130.8 


June 14, 1918 


Milwaukee, Wis. 


12 


Illinois and Indi-! 
ana screenings ' 


83.3 


10,897 


15.80 


117.7 


Nov. 5. 1918 


Lykens, Pa. 


10 


Lykens No. 3 buck-! 
wheat anthracite 


84.2 


12.530 


16.02 


135 


Nov. 16, 1918 


Lykens, Pa. 


5 


Lykens slush buck- 
wheat anthracite [ 


81.2 


13.653 


11.00 


142 


Nov. 22, 1918 


Lykens, Pa. 


5 


Lykens slush buck- 
















wheat anthracite 1 


85 


12,753 


18.04 


146 


Nov. 23, 1918 


Lykens, Pa. 


5 


No. 3 buckwheat | 
anthracite 


72.7 


12,530 


16.01 


116 


Dec. 2, 1918 


Lykens, Pa. 


5 


Lytle slush anthra- 
cite 


75.3 


12.758 


33.02 


18S 


Feb. 1. 1919 


Seattle, Wash. 


24 


Issaquah screenings 


78.95 


11,660 


14.81 


186 


Feb. 2, 1919 


Lykens, Pa. 


4 


1 No. 3 buckwheat 
















anthracite 


78.9 


13.067 


14.08 


in 


Apr. 7, 1919 


Vancouver, B. C. 


4 


Nanaimo slack 


83.3 


0.364 


28.4 


125 


Apr. 17, 1919 


Vancouver, B. C. 


1 5 


' Nanaimo slack 


77.1 


10.050 


84.8 


160 


Feb. 3, 1919JLykens, Pa. 


5.5 


: No. 3 buckwheat 


78.9 


12,530 


14.00 


 • • 








anthracite 










Sept. 24, 1918 


• Verde, Aris. 

1 


(6 days) 


Gallup, New Mexico 

1 


79.5 


10,680 


14J1 


166 



burning equipment will of course vary considfTably with the condi- 
tions undtT which the plant is installed. Roughly speakingi how- 
evtT, the actual investment will vary from S12.80 per kw, output 
in a 5000-k\v. plant down to $L80 per kvv. in a 50,000-kw. plant 
and $4.12 in a 100,000-kw. plant (assuming a turbo-generator water 
rate* of 10 lb. and continuous boiler and furnace efRciency of 75 
pcT cent). 



r. A. BCHBFFUIH AND H. Q. BAJINHTIRST 387 

31 All these figures in relation to cost are based on the present 
high prices. The investment reqiiired for a 5000-kw. plant using 
100 tons of pulverized coal daily is approximately $64,000 and for 
a 50,00&iw. plant using 1000 tons of pulverized coal daily, approxi- 
mately $240,000, so that, on a basis of 6 per cent and allowing for 
365 days' continuous operation, the interest item will vary from 



lOj cents per net ton in a 100-ton plant down to 3.9 cents per ton 
in a 1000-ton plant. 

32 Depredation. Depreciation in a coal-pulverizing plant is 
usually calculated as follows: The life of the building is considered 
as 40 years, of the coal driers as 15 years and of the balance of the 
equipment as 20 j-ears. With a 100-ton pulverized-coal plant and 
burning equipment the depreciation item will be approximately 12 



388 PULVERIZED COAL FOR STATIONARY BOILERS 

cents per net ton, and in a plant of 1000 tons daily capacity it will 
be approximately 4 cents per net ton. 

33 Taxes and Insurance. Taxes and insurance are based on 
2 per cent of the entire investment and for a 100-ton plant this item is 
approximately 3 J cents per ton and for a 1000-ton plant, 1.3 cents 
per ton. Summarizing, the foregoing results show that the total 
cost of pulverizing and delivering pulverized coal to boilers is ap- 
proximately as given in Table 1. The cost of the pulverizing equip- 
ment complete compares favorably with the stoker equipment when 
everything, such as coal- and ash-conveying machinery, etc., is taken 
into consideration, and in large plants it is considerably less. 

COST OF STOKER OPERATION 

34 The equipment required for a first-class stoker installation 
must necessarily be taken into consideration when making a com- 
parison of the costs of the different installations. The cost of pul- 
verizing is an item of expense which must be included with the cost 
of the fuel, and since it includes the complete handling of the coal, 
the expense of crushing, handling, power, repairs, maintenance, 
interest, taxes and insurance covering the stoker equipment must 
also be considered when making comparisons. 

35 Stoker installations and operation are expensive and the 
investment is as great, if not greater, than that required for a pul- 
verizing equipment in plants of 10,000 kw. and upward. For 
example, in a plant using 1000 tons in 24 hours the cost of operation 
will be approximately as follows: 

Power for stoker, 2 per cent of the total boiler hp. developed . . $180.00 

Power for fans, 2 j)er cent of the boiler hp. develo|)ed 180.00 

Coal handling, 100 kw. at } cent per kw-hr 18.00 

Lal>or for coal handling, 2 men per shift and 3 shifts at 40 cents 

per hour 19.20 

Repairs for stokers at 30 cents per lK)iler hp. per annum 17.50 

Hcpjiirs for coal-handling equii)ment 10.00 

$424.70 

Total cost iMT net ton $0,425 

To tiiis must also Ix* added the cost of fuel u.^cd to heat the moisture 
in the coal, interest, depreciation, in.^urance ancl taxes, showing 
that even on a basis of e(nial efTieieney the cost of operating a pul- 
verized-coal eciuiprnent is eoiisiderahly less than the cost of operat- 
ing^ an cfjuivahMit stoker in>tallation. It sliould be stated that the 
figures just pven are based on present average results in both cases. 



r. A.. BCBBFTLEB AND H. O. BARMHUBST 



i 

1 

II 



5": 

4 •* 



PTILVBRIZED COAl, TOR STATIONART BOILEBB 



F. A. 8CHBPFLBB AND H. Q. BABNHDBST 



3D2 PULVERIZED FUEL 

DISCUSSION 

The following discussion on pulverized fuel applies to the preced- 
ing papers, Nos. 1701 and 1702, which were discussed jointly at 
the meeting. 



Thomas A. Marsh (written). The writer would preface his 
(liHcussion by the statement that the general proposition of tak- 
ing raw coal with all of its by-products and preparing it to the most 
(»xp(»nsiv(» size (pulverized) and then burning the raw coal and losing 
th(» by-products, appears to be a move very much in the wrong 
direction. Pulverized coal has a field, no doubt; for instance, such 
fuels ns lignit(*s which cannot be burned on stokers. This field is 
linnl(»d, iiowevcr, for chain grates are now developed to bum lignites 
conlaining from lU) to 1^5 per cent moisture, which embraces practi- 
cally all but (hose of the North Dakota field. 

l*ulv(Mi/.cd fu(*l is adaptable to certain processes as the cement 
pro(«cf4?4 and (HMtain metallurgical processes. In substantiation of 
iIm ndiiplnbilily for steam boilers, however, the paper by Messrs. 
SclicllliM" nnd HHrhhui*st n^pivsiMits pulverized fuel at its best, hints 
a( no( M diawl»Hck or tnuible, submits tests of four or five hours' 
duialion and evidently not made under the A.S.M.E. code, and 
iMintiantn rth»kciM o( the most exiHMisive type to operate as to power 
III \\\\\v }\\\k\ m.1 lo power for fans. If it is necessary to resort to such 
l^^h^lHlU^-. hi make a ciuupanible showing for pulverized fuel, it 
i.ihht'l It '.\\\\ ilijil it is on a com|H^titive basis with modem auto- 
hull it -iMkri'. :\\\\\ furnaces. 

I III iMJlowmk; ^liM-UNsiou and questions are submitted in the 
l»i.|«r \\\\\ \\w lull an.NWcis thereto will provide facts on which to 
!•.( M nnh lu--h»h^ m av\'oul;mee \\itl\ the Society's jxist standards. 

/»•..'.•••''. I lulei the item of " Keliabilitv'* in the above- 
nieiiiiPKtil |Mpei it i-. ^tnted unqualitiedly that the mechanisms 
l«M pnl\»n mi', hul :u\' e\iuallv reliablt* with tluK*?e for crushing 
xx^\\ um loKri'. The e\Ha e»itupuuMit neci^»*iir>' for this exceed- 
hu'.In eUl'oi lie )Mrp:u uion \A the fuel is another link to the mechan- 
i^iu .:ni«l I liu'.h peed oiu^ M [\\aO , aud I think no engineer will 
:uMee \\\a\ ih»' »vMnl«in u umi oI' a eiusher systeni and a pulverizing 
^x-^t m i>^ e. ixliil'le e \ xi\{^\u'v alone. Kven, if this complex 
v\^t.?ii N\,!e ;i . tviijl'l*' e. \ iMu^hM. wliich it is not. the station 
.l.j>, ihi- ni .Ml |Mil\,iied inel \\\»\ild sutTer shutdowns unless the 
rMi-h« ! ;n!.l iMiKrti ei e»iin|Mi\eiit Wert* duplicattnl. The writer 



DISCUSSION 393 

understands that cement plants are usually equipped with duplicate 
pulverizing systems. This will all have a bearing on investment 
charges. 

Further, it is stated in the paper that all grades of coal can be 
used. There are stokers which have limitations only on anthracite 
or coke breeze and in this connection the question arises as to whether 
a pulverized-coal furnace and burner designed for, say, bituminous 
35 per cent volatile will bum anthracite or coke breeze. I ask this 
because in one of our other technical societies it is a matter of recent 
record that for burning low-grade anthracite or coke breeze in pul- 
verized form, reversal of the flame is necessary, a firing method not 
suitable for the richer bituminous coals. It was also brought out 
that with coke breeze or anthracite culm the fire was unresponsive 
and the load had to be varied to suit the fire rather than the fire to 
suit the boiler load. What occurs when a rich high-grade fuel is 
used in a furnace and burner designed for low-grade fuels? 

Adaptability. Under this heading the authors state that the 
primary feature is the possibility of burning all grades of fuel with- 
out affecting the efficiency of the furnace. In order for this to be 
true, it is necessary to bum the 50 per cent ash coal with no more 
excess air than the 3 per cent ash coal and suffer no more ashpit 
loss in the former case than in the latter. The writer therefore 
specifically asks if the 50 per cent ash coal requires in practice no 
more air than the 3 per cent ash coal. When 50 per cent ash coal 
is used the sensible heat of this amount of refuse leaving the furnace 
chamber is in itself an item which will affect the efficiency of the fur- 
nace. Can efficiencies be independent of these items? The tests show 
wide variation in efficiencies. What causes these wide variations? 

Chain grates are burning coals containing 35 per cent ash and 
producing some excellent results, and have been for years. It would 
be desirable along with this statement by the authors regarding 
high-ash fuels for them to submit figures of tests with over 28.4 
per cent ash, which is rather a low percentage as compared with 
chain-grate practice with western fuels. 

As to flexibility with pulverized coal, the writer has seen cases 
where the furnace became unresponsive when forced. There is 
certainly no e\'idence in the paper to indicate that the pulverized- 
fuel burner is quickly responsive over a wide range of loads; in 
fact the evidence is much to the contrary, for of all the tests referred 
to, the maximum overload indicated is 188 per cent, which is not 
considered a high overload in modern specifications and practice. 



394 1>UL\'ERIZED FUEL 

Furnace Design, Under this heading the tests of Table 3 are 
referred to. Tests of four or five hours' duration should not be 
included in an A.S.M.E. paper. Efficiencies of from 70 to 80 per cent 
are shown here. Starting and stopping conditions, surging of water 
in the boiler and other items are so uncertain as to make these short 
tests absolutely unreUable. Under similar conditions, tests of 100 
per cent efficiency and even higher are frequently made. 

To be more specific, the writer would inquire if in the test of 
April 16, 1917, at Seattle, the heating value of the coal was deter- 
mined by a calorimeter or was it assumed. The even figure of 10,000 
simply suggests estimated figures. 

What grt^at diiTerence brought about the marked increase in 
efficiency betwet»n the December 4 and December 12, 1917, tests 
at Chanute, Kan.? This wide difference is inconsistent with the 
8tat(»m(>nt of uniform efficiencies, regardless of operatives, coal or 
oth(T itA^ms. 

On the January 28, 1918, test at Chanute, Kan., of 25 da3rs' 
duration, wert» the water and coal weighed? Did the boiler operate 
thin long without the blow-off l)eing opened, or if the boiler was 
blown d(»wn. how was the blow-off discharge measured? 

Tt\sts of April 26 and 28 at Parsons, Kan. Some tests were 
iniidr \\\ this plant simply by measuring the coal in a spiral con- 
vi\vi»r. Was this done in this instance or were the coal and water 
prnpiM-ly weighed? The writer has a record of other tests made at 
this plant, as follows: 

CombiiKHl Boiler and Furnace Efficiency 
ll«iint^|Mi\MT l)i'vrh)p*Ml Per cent 

ISO 53.9 

l\H 73.5 

'2M) 67.8 

•JIT 67.4 

.MM*. 61 

;i.-.C» 59.0 

•j;jo.:. 71.5 

'J'hcsr nsuhs iiuliciitr \\\v possibility of low efficiencies on teste 
with piilvtMi/(Ml furl. What, then, may we exjKTt in daily opera- 
tion? Why slniuld we pi<'k only favorabh* tests for comparative 
fi^un's*.' 

On the t(\^t of A|>ril 20, shouhl not the ixTcentage of rating be 
given? 



DISCUSSION 395 

The writer is surprised at the high calorific value of the slush 
anthracite in all tests reported, as it has the highest heat value 
of any coal in the tabulation. We should not confuse this with 
low-grade anthracite, which contains less than 9000 B.t.u. 

The complete data of temperatures, draft and CO2, together with 
methods of measuring fuel and water, should be given for these tests. 

Referring to the statement that a modem stoker-fired power 
plant with the best average plant operation obtains an efficiency 
of not better than 63 to 65 per cent, while carefully conducted tests 
show 75 per cent efficiency, the writer suggests that the authority 
for this statement be mentioned so that full credit can be given. 
This may be true with some types of stokers not completely auto- 
matic but dependent largely on the firemen's skill. With the chain- 
grate type, however, it is far from true, in fact the chain-grate 
plant operates within 2 or 3 per cent of the test figures, and test 
figures with modem chain-grate furnaces run frequently in excess 
of 75 per cent net efficiency. It has always been a feature of chain, 
grate performance that daily operating results approach very closely 
to maximum test efficiencies with no deductions to be made for fans 
or other auxiliaries. The fuel costs for modem chain-grate power 
stations may be adduced to confirm this fact. 

The authors have selected for their comparison those types of 
stokers with which it is most difficult to get high operating efficiencies, 
whereas if the comparison were made with completely automatic 
stokers such as chain grates, the statement made of the operating 
advantage becomes void, as chain grates are evidently more capable 
of reproducing test results in daily operation than the pulverized 
fuel furnace. 

Cost of Pulverizing Coal. Under the item of Drier Fuel the 
writer notes that all costs are based on only 7 per cent moisture 
and yet many of the fuels mentioned in the tests contain more 
than 7 per cent. Comparatively few fuels have as low as 7 per 
cent moisture, particularly west of Pittsburgh. The figures run 
from 10 to 40 per cent in the western section of the country, figures 
of 15 per cent predominating. The writer suggests therefore that 
the drying coal cost should be doubled. 

Interest. As mentioned earlier, the pulverizing equipment 
should be duplicated to prevent shutdowns, which would, of course, 
raise the equipment's interest charge. There are some uncertainties 
regarding the cost of this equipment, as some of the plants men- 
tioned are reported to have cost several times the figures quoted 



396 



PULVERIZED FUEL 



in the paper. The writer would therefore safeguard this figure by 
making it at least three times that used by the authors. 

Depreciation. In figuring depreciation, everything should be 
limited to 10 years, inasmuch as we have all seen three or four com- 
plete changes in power-plant equipment in the last 30 years and 
are likely to see the same in the next 30 years. To figure that the 
fuel-pulverizing building has a useful life of 40 years is making a 
pretty strong effort to show up pulverized fuel in a favorable Ught. 

The writer would therefore recalculate Table 1 as given below. 

TABLE 1 COST OF DELIVERING PULVERIZED FUEL TO BOILER 





100-ton plant, 

dollars per 

net ton 


lOOO-ton plant, 
net ton 


Power at f cent per kw-hr. and 17 kw-hr. per ton 
Labor at 60 centa per hour (40 cents too low). . . . 

Drier coal at $6 ton delivered (16% HiO) 

Rei>air8 (no df^tn — &nc«pt &iithom') . . . . , 


0.1276 
0.175 
0.12 
0.07 


0.1275 
0.05 
0.12 
0.07 






Total coat of pulverizing per net ton » 

Interest at 6 per cent. (Triple the investment) . . 
Depreciation (lO-yr. basi«); Multiply authors' 

figures by 3 

Taxes and insurance (no data — accept authors' 

figures) 


$0.4926 
0.316 

0.36 

0.036 


$0.3575 
0.117 

0.12 

0.013 




$1.2025 


$0.6175 



* In existing plants of 600 to 6(X) tons daily capacity this figure is reported to aetiially be 
$1 or more per ton, indicating that in actual practice and operation many incidental itema of 
cost enter which are nut shown in a theoretical estimate. 



Cost of Stoker Operation. In making this comparison, the authors 
have again selected those types of stokers which show highest operat- 
ing costs and cause the pulverized-fuel installation to benefit by 
contrast. 

The p()W(M' to drive some typ<\s of stokers may l)c as high as 
2 [)er cc^nt. ( -hain grates re(iiiire but iV <>f 1 P<^r cent. Power for 
fans with tlio.sc* types of stokers recjuiring fans may be 2 per cent of 
th(* power dt^velopcvl. Chain grates have no fans, so this item is 
eliminated. Coal handling would se(»m to be an item which would 
cancel out, as it should cost no more to deliver coal to the stok^ 
hoppers than to the pulverizer. However, accepting these figures 
on the authors' basis, we havt^ as follows per 1000 tons of coal: 



DISCUSSION 397 

Power for stokers, 1 hp. per 1000 developed; chain-grate 

practice, 20 cents per 1000 lb. of steam $ 4. 00 

Power for fans, standard chain-grate practice 0.0 

Coal-handling power (accept authors* figures) 18. 00 

Labor for coal handling (accept authors' figures) 19.20 

Repairs for stokers (accept authors' figures) 17. 50 

Repairs for coal-handling equipment (accept authors' 

figures) 10.00 

$68.70 
Total cost per net ton 10. 0687 

We have, therefore, an additional cost of from 55 cents to 
$1.13 per ton against the pulverized-fuel plant, a figure that cannot 
be made up by increased efiiciency, for it must be realized that many 
stoker installations have given performances within 5 per cent of 
the highest theoretically possible. 

Now as to some of the items not mentioned by Messrs. Scheffler 
and Bamhurst in their paper. 

Fine Ash. The writer understands that from 60 to 80 per cent 
of the ash carries over to lodge in tubes and combustion chambers, 
but mostly goes out of the chimney to scatter over the surrounding 
country. This may by comparison not be objectionable in a cement 
mill or in some of the localities mentioned in the list of installations, 
but it is evident that if the Commonwealth Edison Company of 
Chicago were to scatter fine ash from the 5(X)0 or more tons of coal 
they bum daily, say 6(X) to 8(X) tons of ash, it would create a condi- 
tion compared to which the worst smoke fog would seem Uke a para- 
dise. 

Slag. Some installations have encountered trouble due to the 
fine ash slagging in the tubes. The writer would inquire how prev- 
alent this is, how it is overcome or prevented and what influence 
high boiler ratings seem to exert on it. 

The refuse from pulverized-fuel firing contains some combusti- 
ble. It has been reported that the slag containing this combustible 
causes damage to ashpits. Is this a serious item of maintenance? 

It is reported that pulverized fviel if stored will pack and be- 
come difficult to handle. Where the container is jarred or vibrated 
this action is reporUnl to increase. How serious is this action? 
Does it interfere witli the use of stored powdered coal? Does it 
preclude its use on shipboiird? 

Turning now to Mr. Harrison's paper, the coals he states as 
most suitable seem to have (a) less than 10 per cent ash, (6) volatile 



398 



PULVERIZED FUEL 



30 to 40 per cent, (c) low sulphur, and (d) high melting point of 
ash. This, of course, means a very choice bituminous coal. Such 
fuel commands a high price in any market, much higher per 1000 
B.t.u. than the high-sulphur, high-ash fuels frequently competing- 
For instance, there are in Illinois and Indiana coals containing 18 to 
20 per cent ash, 30 per cent volatile, 6 to 8 per cent sulphur, with 
the fusion point of ash at 1900 deg. fahr. The writ<^r would inquii*e as 
to whether in the present state of the art this fuel could be com- 
mercially used to develop, say, 150 per cent rating from a boiler by 
means of pulverization. 

It seems to be a prevalent opinion that lignites cannot be burned 
on stokers. The writer would rectify this at once. Lignites and 
sub-bituminous coals are being burned and producing high efficiencies 
and capacities. The following results of a few tests will substan- 
tiate this statement. 



Fuel 

Moiflture, per cent 

Volatile, iKjr cent 

Fixed carbon, per cent 

Ash, per cent 

B.t.u. (conwn.) 

B.t.u. (dry) 

Type of Btoker 

Type of boiler 

Coal per aq. ft. per hr., lb. , . 

Furnace draft, in 

CCh at daniiM'r, iH»r rent .... 
Per cent, rating developed . . 
Combined etlioicncy. jht rent 



Colo. 
Lig. 


Tex. 
Lig. 


Mont. 
Sub- 
Bit. 


Mont. 
Sub- 
Bit. 


Mont. 
Sub- 
Bit. 


23.73 


29.93 


20.73 


15.28 


15 27 


;r).25 


30.96 


33.44 


37.56 


41.24 


:i3.23 


25.45 


32.50 


38.07 


30.01 


7.79 


11.66 


13.28 


15.09 


12.58 


8511 


7124 


8772 


0118 


9553 


11,110 


10,314 


11,073 


10,763 


11^6 


Green 


CIreen 


Green 


Green 


Green 


li & w 


B&W 


B A W 


B & W 


BA W 


28 


32 


28.91 


38.06 


30.46 


. 2A2 


0.32 


0.167 


0.221 


0.17 


12 


11. SI 


10.30 


12.97 


11.77 


128 




l-ir>.8 


209.1 


181.5 


CD 7 


♦19 


72 


75 


77.3 



Mont. 
Sub- 
Bit. 



20 52 
31 23 
39.00 
9 25 
9096 
11.444 
Green 
B4kW 
35.31 
0.196 
12.62 
195.09 
78.04 



Th(^ ability to burn those coals is a developiiu»nt of the last few 
years, ami has opc^iu^d up a woiKlcM-fiil fuel tieUl to cliain grates. 
Some of tlu'st' furls contain as hi^h as 30 j)er cent mositure, in which 
cast* the drying cost when pulverized fuel is used would l>e 4f times 
tliosi* j»;ivcii ill Taivlc 1 and would luakt* it necessary for the pul- 
vcrizcd-fucl inslallation to obtain .^oincwlien* in the neighborhood 
of 100 per criit clIici<Micv to Im* on a coinpctitvc ba.^is. 

Morcov(M', ihr.<c fuels »!<» not at this tiiiit* coiiiinand a price any- 
where near s.'> jut ttni. so that tin* cost of drying and pulverizing 
represents probably al>oiit 2"> p«'r ceni of the fuel cost, a figure not 
to be rei;:iiiit'd l»v iucrea.M'd i»ni<'iencv. 



DISCUSSION 399 

Reference is made to the use of low-grade refuse around the 
mines, as at Lykens, Pa. Mr. Sheffler and Mr. Bamhurst also 
refer to this same fact in their paper. The coals they mention, how- 
ever, have calorific values not only far beyond the average idea of 
culm piles (12,500 to 15,400 B.t.u.), but are good enough to be used 
to advertise the coal. Certainly no western fuels and few mid- 
western fuels will equal these supposedly refuse coals. 

It should be imderstood that when a chain-grate engineer refers 
to low-grade coal, he certainly has in mind fuel containing less 
than 9000 B.t.u., dry basis. No such fuels seem to be discussed 
even as refuse fuels in pulverized-fuel practice. 

Mr. Harrison bases his costs of pulverizing on pre-war labor 
prices (30 cents per hour for millers, 20 cents for firemen and com- 
mon labor), war coal prices (which is not consistent with 20-cent 
labor), and 7 per cent moisture in the fuel. His figures for labor 
should be doubled. For average fuels the moisture percentage should 
be doubled, and for the real field as mentioned by the author at 
least quadrupled for coal with 30 to 35 per cent moisture. 

Referring* now to (a) under Pulverized Coal vs. Stokers, the 
writer would inquire if a furnace suitable for lignite will handle 
anthracite. High overloads are spoken of freely and he would 
also inquire just what ratings have been sustained or even reached 
to confirm this. 

Stokers are criticised in (c) as requiring more area for high over- 
loads than pulverized-fuel burners. Has this been proved to be 
commercially true? With chain grates and coals such as mentioned 
in the average pulverized-fuel test, ratings of over 250 per cent are 
sustained for long periods, and peaks swimg for short periods beyond 
this rating. Have any pulverized-fuel installations so far equaled 
this? 

Referring to (e), the writer would state that not only is ash, more 
easy to handle, but there is only about one-fourth of the usual amount, 
the remainder going out the chimney. This is one item of decreased 
cost that pulverized-fuel exponents seem loath to claim. We cer- 
tainly must overcome this difficulty if we are to make pulverized- 
coal plants commercially successful, as our city and health ordinances 
will soon put the ban on this wide distribution of fine ash. 

The writer would correct item (/) to say that coal burned on 
stokers may contain from 1 to 35 per cent of free moisture as fired. 
As a point of interest, he would inquire as tb how readily pulverized 
coal picks up moisture when stored. 



400 PULVERIZED FUEL 

Under (d) in the following paragraph certain changes in the 
furnace are mentioned to suit various coals, also rather extensive 
changes in drying equipment. This is contrasted with stoker prac- 
• tice. The author presents as a favorable argument for pulverized- 
fuel burners that, with only a slight change in the furnace and some 
increased drying capacity, the equipment may be changed from one 
suitable for anthracite to one suitable for lignite. The writer would 
inquire in what commercial fuel market or location is this a valid 
benefit. What market receives both anthracite and lignite as steam 
coals? 

Referring to the comparison of costs given under the heading 
Pulverized Fuel vs. Mechanical Stokers, the writer would question 
the estimated figure of 32 cents and would state that since the in- 
stallation had its acceptance test ten months ago, the costs should 
not be concealed, but should be made part of the paper. 

Reference to Table 1 in the complete paper would indicate 
that pulverizing cost would be 

(a) For pulverizing and drying 12 per cent moisture coal. . .$0,516 
(6) Labor (use 2 times tabulated cost) 0.08 

$0,596 

Maintenance? (use author's estimate) 0.03 

$0,626 

As a matter of fact, in existing installations of from 500 to 600 
tons daily capacity, the cost of drying and pulverizing fuel is over 
$1 per ton. This makes pulverized fuel prohibitive for steam-making 
purposes. 

From all the figures presented, some good efficiencies are indi- 
cated when high-grade coals are burned (no tests are presented 
on low-grade coals). The advantage in efficiency, however, is in- 
sufficient to offset the increased cost of preparing the fuel, even 
on a basis of ^5 fuel. This caust»s the pulverized-fuel burner to 
operate at a loss, regardless of the efficienr,y obtained. Fuel costs 
will have to Ix^ from 50 to UK) jXT cent higher before the increased 
cost o{ pulverizing will be justified, unless pulverizing and drying 
costs decrease niateriallv. 

Furtiier, no modern ratings are recorded in the paper; furnace 
anil slag troubles are minimized; quick pickups of boiler load over 
a wide range of rating ari' not provcnl and seem very problematic; 
ash from chimneys is at present a prohibitive feature in cities. 



\ 



DISCUSSION 401 

Fred'k a. Scheffler (written). I cannot agree with Mr 
Harrison in his statement that when a boiler is being run with pul- 
verized coal it is not necessary- to have more than 0.10 to 0.15 in 
draft at the damper of the boiler, and consequently the boiler could 
be operated with stacks about 30 to 35 ft. in height. It is a well- 
known fact that it is necessary to have at least 0. 10 in. in the furnace, 
and that the average water-tube boilers have a frictional or draft 
loss through the boiler of about 0.30 in. w hen run at rating, and when 
run at 200 per cent of rating this draft loss is almost double, or 
0.60 in. Consequently, the stack would have to be at least 125 
ft. in height in order to be sure that there is sufficient draft to over- 
come the frictional resistance through the boiler and allow a suction 
of at least 0.10 in. in the furnace. 

W. N. Best (written). I should like to inquire how many open- 
hearth furnaces there are in the United States that are successfully 
burning pulverized coal, and for how long a period they have been 
operated. 

The Boiler Test Code Committee of the A.S.M.E. would Uke 
to know of large power plants that are and have been successfully 
burning pulverized coal for a period of, say, 4 years. We have en- 
deavored to locate such plants for some time in order to examine 
same and secure some data for the Society. I am aware that many 
plants have experimented with pulverized coal in the generating of 
steam, but results were very disappointing. 

At the beginning of Mr. Harrison's paper, he states: "... 
and the shortage in the supply of crude oils which have become 
of too great value for ordinary' fuel purposes." This, I think, is 
misleading, for statistics prove that the production of oil has never 
before been so great, and the demands for it never so great, as to- 
day. In Mexico alone there are many reservoirs as large as lakes 
filled with crude oil awaiting transportation, and hundreds of wells 
are capped awaiting to be turned on to meet the demand. 

The increased price of coal and the liability of further increases 
in its price have compelled many boiler plants along the Eastern 
Coast to change to oil as fuel, and many more contemplate changing 
soon. The cost of oil is now ver\' attractive in boiler plants owing 
to the fact that it only requires 147 gal. of oil to represent a long ton 
(2240 lb.) of bituminous coal, the coal having a calorific value of 
14,000 B.t.u. per lb. One man can fire and w^ater-tend twelve 300-hp. 
boilers. Oil is so attractive at the present time as a fuel along the 



402 PULVERIZED FUEL 

Atlantic Coast that I believe it will only be a matter of a year or a 
year and a half until all the larger power plants will use Mexican 
oil as a fuel in their boilers. 

It is my opinion from close observation and study that the two 
distinct fields for pulverized coal are the furnaces of rotary kilns 
and copper matting furnaces. Both of these are constructed so 
that the building is quite open, and there is practically no liability 
of explosions being caused by spontaneous combustion. I believe 
the large quantities of poor coal and lignite that we have in our 
country can be successfully burned in combination with oil, the 
coal and Ugnite referred to being, of course, in the pulverized state. 
By this combination practically perfect combustion can be attained 
and maintained at all times, and this combination could be of value 
owing to the low calorific value of the coal and lignite, and the high 
calorific value of the oil. The process of burning this combination 
of fuels would not be by mixing them together, but each fuel should 
be delivered separately to the furnace. 

W. G. DiMAN (written). So far as the better grades of bitu- 
minous coals are concerned, I think that the cost of drying, pulveriz- 
ing, conveying, feeding and the first cost of the apparatus, together 
with the fact that all grades of fuel can be burned with a tolerable 
degree of smokelessness and efficiency in the regular stoker apparatus 
will restrict the use of pulverized coal for boiler purposes to special 
cases. If low-grade waste coal can be utilized for pulverized fuel 
there is a great fi(»ld for its use. There must be some limit, however, 
in the use of low-grade fuel, for the cost of grinding would eat up 
any advantage in the economy. WhcTc^ low-grade fuels are high in 
ash and slate there should be a limit to the proiwrtion that one can 
afford to grind. The use of high-ash coal will be bad, as it will 
accumulate rapidly and be difficult to remove, esi)ecially if it slags 
to any extent. This would occur more where the ash deposits are 
within the limit of the flame. With anthracite, es[)ecially culm and 
of that nature, it nii^ht be difficult to pulverize and nmst be finely 
pulverized; it n(M'ds a higher temiKTatun* for combustion, bums 
more slowly, and in certain boilers like a H.R.T. it would require 
plj'nty of brickwork and a large combustion chamlKT to avoid the 
('hilliiijz: etlVct and to maintain combustion. Thr best roal to use 
is one lii^h in volatile' and low in ash. Such a coal \voul<l not be 
a very satisfai'tory on(^ for outside storage in large (juantities due to 
the likelihood of sj)ontaneous combustion. In order to successfully 



DISCUSSION 403 

use the pulverLEed fuel, I am of the opinion that the best results can 
be obtained by designing the furnace and equipment to meet a 
specific grade of fuel. I do not think that the average cost per ton 
for getting the coal from the car into the boiler is as low as stated, 
and if the overhead is also taken into consideration it will run still 
higher. 

E. H. Peabody (written). In Mr. Harrison's paper I note 
that among other important matters he mentions three points 
which occur to me as particularly significant in the use of pulverized 
coal for boiler purposes: 

1 Very large furnace volume required, or, in other words, the 

fuel must be burned under conditions which imply a very 
low rate of combustion per cubic foot of furnace volume; 

2 The continual effort necessary to keep the boiler tubes, and, 

to a less degree, the furnace itself, free from slag and 
clinker caused by the refuse in the coal ; and 

3 The very high and apparently inaccurate, boiler efficiency 

reported in the tests of August 12-13, 1918, which appears 
to be due to crediting the boiler with work done by the 
coal driers. 

If, as seems to me proper, the boiler efficiency is figured on the 
basis of the heat value of the dry fuel, the result should be 76.3 
instead of 85.2. The degree of efficiency obtained in the drier itself 
would in no way affect the results obtained in the boiler in actually 
transferring the heat in the fuel to the steam. 

Oil fuel is imdoubtedly superior to all others for boiler pur- 
poses. The furnace volume required for this fuel to give satisfactory 
results is about one-quarter the furnace volume specified in the paper 
for pulverized coal. 

The similarity of action in many features between oil fuel and 
pulverized coal appears to me to constitute one of the principal 
attractions of the latter. It would seem, however, that the large 
furnace, with its extra first cost and extra radiation loss, together 
with the difficulties due to slag and their effect on operation, would 
offset the desirability of pulverized coal to a very large extent. 

Albert A. Gary (written). To many who have not had occa- 
sion to keep informed concerning the previous use and past ap- 
plications of powdered coal as a fuel, the idea seems to be prevalent 



404 PULVERIZED FUEL 

that such fuel is a recent development, holding almost unlimited 
possibilities for all kinds of furnace applications and capable of 
easily and cheaply accomplishing phenomenal results. 

As a matter of fact, inventors have been struggling for almost 
a century to make powdered coal a practical and successful fuel. 

In 1831 an English patent was issued to J. S. Daws for a process 
for burning powdered coal, and this was rapidly followed by a large 
number of patents relating to this subject in England, the Unitod 
States, Germany and elsewhere. 

In 1881 a United States patent was issued to C. H. Palmer 
describing the means for feeding fine coal to a locomotive boiler 
with an air blast, while in 1870 Whelpley & Storer began to take 
out a series of patents for using pulverized coal in reverberatory 
metallurgical furnaces. In 1876 an elaborate series of tests was 
conducted by the Bureau of Steam Engineering of the Navy De- 
partment under the direction of B. W. Isherwood — using the 
Whelpley & Storer pulverized-coal system applied to a steam boiler. 

Thus we are able to understand that the preparation and use 
of pulverized coal fuel is by no means a recent development, and 
notwithstanding the considerable expenditure of money, effort and 
ingenuity by many inventors, including men whose past experiences 
qualified them to carry on such work; most of these older produc- 
tions have been scrapped so that today we have left merely the 
benefit of their varied experiences, which has, nevertheless, proved 
a vahiable source of information for the more recent developments 
in the preparation and use of pulverized coal. 

Turning now from this n^cord to the more recent developments, 
it will recjuire but litth* investigation to find that during the last 
decade (or even fur a much shorter i)erio(l) there have been many 
pow(l<' red-coal installations which hav(» either bt^^n n*jected or 
found to pivc* poor satisfaction. 

After the presentation of such facts, I may be accused of con- 
(liMnniiiji the us(» of pulv(M*iz(Ml fu(^l as a practical and (efficient fuel. 

On the contrary, in the lipht of my experience and with the 
evi<lence j)resented by many siitisfactory e(juipments now in oper- 
ation, I am an unciualititMl advocate of tlie use of such fuel, but I 
un\>\ liinil my endorsement to applications when* a desirable grade 
of file! is availai)le - wiiere proper j)re])aration of the fuel and proper 
furnace conditions can 1k» obtained and where other methods for 
burning the available fuel are inferior to accomplish the desired 
h(>at itit; 



DISCUSSION 405 

About 24 years ago I was called upon to assist in the develop- 
ment of a pulverized coal equipment for a cement plant, and since 
that time I have been called upon to test, investigate and design 
a number of pulverized coal equipments for cement kilns, boilers^ 
metallurgical and other industrial furnaces and thus have had an 
opportunity to follow the development of this art since the early 
days of its practical commercial adoption in this country. 

Pidverized Coal in Cement Plants. Mr. Harrison has very 
properly given first place in his paper to the use of pulverized-coal 
fuel in the rotary kilns of our cement plants, as not only do we owe 
the present development of our pulverized-coal equipments to the 
pioneer work done in their development at such plants, but the very 
form ol these long, cylindrical, refractory-lined furnaces furnishes 
us with the ideal construction for the use of this form and kind of 
fuel. 

They permit the use of the long, air-projected current of finely 
powdered fuel, giving it ample time to complete its combustion 
while being held in suspension and without direct flame and ash 
impingement upon any impeding furnace wall. 

The simplest form of burner and fuel-feeding device can be 
used with such furnaces, and the ash resulting from the combus- 
tion of this finely groimd coal causes little or no trouble, provid- 
ing the ash content and its quality remains nearly constant, as this 
ash drops down and mingles with the burned clinker without ma- 
terially affecting the quality of the final product. 

Pulverized Coal in Metallurgical Furnaces. Next in order of 
desirabiUty in the way of furnace design for the use of pulverized 
coal, we have our reverberatory metallurgical furnaces of great 
length (100 to 150 ft.), such as are used for copper smelting, where 
the flame is projected from the front toward the rear of these furnaces 
over the charge of ore on the hearth. 

Without the exceptional facilities for taking care of the ash 
resulting from the combustion of the pulverized coal as found in 
the rotary cement kilns, ash troubles proved to be a pretty serious 
matter with the early pulverized-coal installations appUed to these 
furnaces about a dozen years ago. 

The ash fused and formed a slag blanket over the top of the 
ore charge and stuck to the interior of the flue outlets in a way to 
block these passages; but these and other associated troubles were 
finally overcome by stopping the infiltration of large amoimts of 



406 PULVEBIZED FUEL 

cold air, by stopping the practice of charging large quantities of 
cold ore into the furnace, by using a better design of biuner and coal- 
feeding device, and by a more careful preparation of the pulverized 
coal. 

By these means a much higher temperature was maintained in 
the furnace, a very much higher ratio of charge to fuel used was 
obtained, and a better and more constant regulation was secured 
at the burner. Then pulverized coal began to be recognized as a 
most excellent reverberatory fuel. 

When pulverized coal was used as a fuel for shorter and smaller 
reverberatory furnaces and directly fired into their interiors, the 
above-named troubles were accentuated and due to the imperfect 
absorption of the heat in these furnaces, the use of waste-heat boilers 
generally became a necessity in order to obtain efficient results. 
Accumulations of more or less fused ash piled up rapidly on the 
heating surface of these boilers necessitating frequent cleanings. 
Ebcperience has shown that the reversing type of fiunace for the 
production of open-hearth steel, such as described in Mr. Harrison's 
paper, is the best design of pulverized-coal furnace for that purpo.se. 
Checkerwork and baffie walls must be specially designed for use of 
this fuel. 

A modification of the above-described reverberatory furnace 
consisting of an extension furnace in front of the main furnace 
chamber, so proportioned as to retain a large percentage of the 
ash within this chamber, makes a highly efficient equipment. It 
requires a burner equipment permitting a close regulation of its 
coal and air supply, and one which will produce a short, brush- 
like flame dose up to the burner, which means a very rapid com- 
bustion and a high temi)orature in the combustion chamber. 

In some recently-built extension furnaces a large percentage 
of the ash is scparatt^d from the body of the flame and drops into 
a comparatively cool part of thc^ chaml>er as a fine, dr}' ash mingled 
with but a small pern^ntajjo of fused ash nodules, which do not 
iiit(Mf(Mv with the easy removal of the refuse from the furnace 
bottom. 

With otluM' nM*(Mit designs of (extension furnaces, where very 
hij?h tt'inixTatuH's are maintained at the burner outlet, the rapid 
fusinjr of the asli is oxiK'dited, which refuse drops to the slag pit 
below in a molten mass and under projx^r conditions may be drawn 
olT tliroujijh taj) holes from th(» bottom of the extension furnace. 

Surh a d(*sij:n of extension furnace has lx>en installed by the 



DISCUSSION 407 

Lopuloo Company in a steel plant near Pittsburgh, in connection 
with their heating furnaces, where it has given excellent satisfaction. 

Application to Stationary Boiler Plants. The application of 
pulvenzed-coal fuel to steam boiler furnaces meets with many 
challenging difficulties, which, however, recent developments have 
done much to overcome. A number of such installations are in use, 
some of which are showing very desirable economy and producing a 
material increase in the steaming capacity of the boilers. 

Great care must be taken to prevent any of the unconsumed 
coal from coining in contact with the chilling water-containing 
surfaces of the boiler, which, therefore, demands the use of an ex- 
tension furnace in which the complete combustion of the coal is 
accomplished, and in which the proper handling of the fine ash 
(and its resulting slag ) is provided for. 

All that has previously been said concerning the requirements 
needed for extension furnaces apphed to short reverberatory furnaces 
apply with equal force here. 

With the intense heat generated in burning pulverized* coal 
much trouble has been experienced in improperly designed boiler 
furnaces due to the rapid melting down of the refractory linings, 
which may also be subjected to the scouring or cutting action of 
impinging ash, thus requiring frequent and expensive relining. 

Process of Combustion of Pulverized Cool, Pulverized coal 
musi, be burned while suspended in the current of air which accom- 
panies it into the furnace. 

As the finely divided coal enters the highly-heated furnace, any 
moisture contained by each minute particle is first driven out, 
and if this is excessive, it will form a steam cloud around the particle 
and suppress its further rapid combustion, thus defeating to a greater 
or less extent the purpose for which the coal was prepared. 

One of the necessities for supplying the furnace with very dry 
pulverized coal is thus appreciated. 

With Uttle or no moisture present, the next effect of the very 
high surrounding temperature in the furnace is to almost instantly 
distil oflF the volatile matter occluded in the coal as this gas is liberated 
at a comparatively low temperature and thus, with properly de- 
signed furnace and burner, the dust cloud entering the combustion 
chamber suddenly becomes thoroughly saturated with a highly 
inflammable gas, mixed with an ample supply of air. 

Under these conditions the gas is raised to its ignition tempera- 



408 PULVERIZED FUEL 

tuie with great rapidity, and the flanie produced is propagated 
with intense speed throughout the atmosphere of fuel and air enter- 
ing the furnace; all of which is simultaneously subjected to the 
same action. 

With the intense degree of heat thus generated, not only is the 
temperature of the furnace maintained to stimulate further com- 
bustion, but the particles of fixed carbon (or flecks of coke) left 
behind, after tlie volatile gases have been driven out of tlie coal, 
have their temperature raised with great rapidity to their tennx»ra- 
ture of ignition and so are most speedily burned, leaving behind the 
non-combustible matter (which we have called ash) to be more or 
less effectively taken care of in a properly designed furnace, or to 
give great trouble in furnaces where its importance has been ignored. 

Mr. Harrison very tritely states that "The finer the coal is 
pulverized, the more efficiently it can be burned." 

With the above analysis of the process of combustion before 
us, we can more readily understand the reason for this statement. 
Not only does the greatly increased fuel surface (presented to the 
highly heated interior of the furnace) facilitate the more rapid dis- 
tilling off of the volatile gases, but with the volume of the particle 
itself greatly reduced, penetration of the heat to its interior takes 
place ill a very much shorter period of time. Aside from these 
advantages the important fact stands out that the I-aw of Ma.<s 
Action applies to such combustion (which is akin to the action of 
explosives) by which we find that the amount of the reaction in a 
unit of tune is proportional to the active mass, or, in other words, 
the smaller the particle of coal, the smaller the number of gram- 
molecules of combustible matter found in the unit mass. 

Under these most favorable coiulitioiis, the velocity of combus- 
tion is greatly accelerated, which is the principal object striven 
for when pulverized coal is chosen for a fuel. 

A iiK Client's thought will recall the fact that the volume of a 
mass varies as the cube of its diameter and thus a particle of coal 
\\.^ ill. in diameter has eight times the volume that is found in 
aimther particle tliat is .}n„ in. in diameter. 

The etTeet of tlie decreased surface in proportion to the mass 
results in a very ^reat retardation of tlie combustion of the larger 
l>articles ms is very plainly shown in the photographic stutly of the 
conihiistion of coal \\\\<\ in Bulletin No. 102 of the Bureau of Mines, 
entitled 'i'lie Intlaininability of Illinois Coal Dusts on Plate IV. B 
and Plate V. B. 



DISCUSSION 409 

Unifonnity in the size of pulverized particles is also very de- 
sirable, as we can readily understand that if we have mixed a con- 
siderable percentage of each of the above-named sizes of coal fed 
to our furnace, the combustion of the larger size will proceed much 
slower than the burning of the smaller size, which condition not 
only reduces the velocity of combustion of the total mass, but it 
prevents the production of that "gas-Uke flame," which Mr. Harrison 
refers to. 

When such grades of coal, averaging from 30 to 40 per cent of 
volatile matter, as described above, are pulverized so that over 
90 per cent will pass through a 200-me8h screen, the finer coal ac- 
companying it does not seem to materially affect its even rapid 
burning effect in general furnace practice; and further, as Mr. 
Harrison has stated, with this very fine pulverization, which in- 
sures rapid and high temperature combustion, the sulphur in the 
coal bums rapidly to SO2 gas, which passes to the stack without 
affecting the product of the reverberatory furnace. 

When a coal, running high in fixed carbon, is used as pulverized 
fuel, we have a different and less desirable set of furnace conditions. 

Lacking the production of an ample supply of volatile com- 
bustible gases given off by the coal at a comparatively low tem- 
perature, and also lacking the heating effect which the burning of 
these gases produces, to hasten the combustion of the associated 
fixed carbon, we must necessarily maintain a higher furnace tem- 
perature to ignite this fuel, and to obtain the best possible results, 
the coal should be ground finer, as it will be found that this fuel 
is slower to ignite. Much of such coal carries a higher ash content, 
which still further complicates matters and lowers the heat value 
of the coal. 

Mr. Harrison gives a table purporting to give the cost of coal 
pulverizing equipments and the cost of preparing the coal per net ton. 
I question the reliability of these figures. Nothing is said of the 
pulverized-coal system upon which these figures are based nor 
the class of machinery or equipment included. 

There is a considerable range in cost of equipments as furnished 
by different concerns and the figures presented in this table appear 
low for the better class of equipment. 

Much depends upon the class of equipment one is willing to in- 
stall, which has much to do with results obtained including cost of 
future upkeep. 

There are many elements entering into the cost of such equip- 



410 PULVERIZED FUEL 

ments. For example: should one be obliged to handle very wet 
coal in the coal driers, the normal capacity of the drier would be 
reduced and either larger driers or more driers would be required 
to obtain a fixed amount of dried coal, which would also require 
a greater building space. 

Again, should one decide to grind the coal very fine to obtain 
maximum furnace results, the output capacity of the pulverizer 
would be materially reduced below that needed for coarser pulveri- 
zation and more or larger mills would be needed, and thus I might 
continue this list so as to include most of the contained units. 

Wliereas, Mr. Harrison's total cost for pulverizing 80 and 90 
tons of coal per day agrees pretty well with the cost estimate given 
in this table, his cost for producing 100 tons per day runs 21 per 
cent higher than the figure shown in the table. 

In looking over the log of the boiler test quoted by Mr. Har- 
rison I note what appear to be a few discrepancies, but lacking 
full and complete data, I have been unable to check them up as 
carefully as I would otherwise do. 

Taking his analysis of furnace gases, he gives the maximum 
result obtained as 15.4 per cent of CO2 and 5.6 per cent of C with no 
CO. 

This is consistent and possible. 

He then gives his minimum result obtained as 12.2 per cent of 
CO2 and 3.2 per cent of with no CO; the sum of these gases giving 
15.4 per cent with 84.6 per cent of N by difference. 

There is evidently some discrepancy here, as I do not see how 
it is possible to burn such a coal as he indicates so as to obtain such 
an analysis. 

The percentage of boiler efficiency given in this table as 85.22 
per cent secerns entirely inconsistent and I cannot find sufficient 
indication contained in the other data submitted to assure me that 
the total losses in the performance of this test amounted to only 
14.78 per cent. 

( )n(* very inU^resting question in coimection with this test is, 
What nutans wore used for accurately weighing the pulverised 
coal used? As a niatttT of secondary importiince I might also ask, 
What means were usimI for ascertaining the weight of water fed to 
tlir boiler? 

J. K. MrnLKKLi) (written). Mr. Harrison's excellent paper 
reviews <latM \\\i\{ eoiifonn in general to my experience during the 



DISCUSSION 411 

past few years in the development of sjBtems for the more effective 
and economical utilization of coals and lignites in pulverized form, 
for stationary, locomotive and marine boilers, and for metallurgical 
and chemical heating furnaces, and the information that he has 
given will be of great assistance to existing and prospective users of 
this method of firing and burning soUd fuels. 

Until recently power-plant capacity and eflSciency have been 
dependent upon combustion possibilities. Fortunately, this has 
now changed and the problem is in boiler design and construction. 

In the log of test on a pulverized-fuel-buming stationary boiler 
given by Mr. Harrison, the operating capacity deductions on ac- 
count of the power and other items necessary for the preparation 
of the coal in pulverized form should be 3.17 per cent, in place of 
4.22 per cent, as used in arriving at the net efficiency. This figure 
of 3.17 per cent compares with the power necessary to operate 
mechanical stokers, and which, in an installation of this kind, would 
amount to from 2^ to 5 per cent, varying with the boiler load carried. 

In connection with this log of test data it may be of interest 
to give the heat balance for the same 24-hour run, which is as given 
in Table 2. 

Furthermore, to show the definite control over stand-by losses 



TABLE 2 HEAT BALANCE FOR BOILER TEST IN TABLE 3 OF MR. HARRISON'S 

PAPER 

Ultimate analyab of pulverUed Indiana and Illinoia screeningB as fired, given in percentages: 

1 Moisture 3.20 

2 Carbon 56.66 

3 Hydrogen 4.47 

4 Oxygen 19.77 

6 Sulphur 1.97 

6 Nitrogen 1.45 

7 Ash 13.48 

100.00 

The distribution of the calorific value of one pound of coal as fired among the several items 
of heat utilised and lost is as follows: 

B.t.u. Per cent 

Heat absorbed by boiler 9,934 85.22 

due to evaporation by moisture in coal 41 . 35 

due to heat carried away by steam formed by burning of hydrogen 493 4 . 22 

Loes due to heat carried away in dry flue gases 975 8 . 37 

Loes due to carbon monoxide 0.00 

Loss due to combustible in ash and refuse 51 0.44 

Loeses due to heating moisture, in consumed hydrogen, hydrocarbon, 

radiation and unaccounted for 163 1.40 

Calorific value of fuel ss fired 11,657 100.00 



412 



PULVERIZED FUEL 



by the use of the same pulverized-fuel system, at the same plant, 
the data in Table 3 are of interest. 

TABLE 3 SHOWING CONTROL OVER STAND-BY LOSS 

Date August 19. 1918 

Hoiler No. 5 Edge Moor. Rated 

468 nominal hp. 

Furl f(>o(l 8hut off, uptake damper closed and auxiliary air inletii cloeed. ... 9.00 p.m. 

Boiler strain outlet to header closed and 175 lb. steam on boiler 9.20 p.m. 

Safety valves released about one minute at the following times (p.m) 9.40. 9.55, 

10.08. 10.15. 10.25, 10.38. 10.43. 10.52, 11.02, 11.00, 11.18. 11.28» 11.38. 11.48, 11.52. 
Steam on boiler 158 lb. when fuel feed started and boiler steam outlet to 

header opened 7.00 a.m. 

Drop of steam'prcssurc in boiler, from 9 p.m. until 7 a.m., or during 10 hour* 

while fuel feed was off and during which time safety valves popped 15 

times, for one minute each, or a total of about 15 minutes 20 lb. 

Time required to bring boiler from 155 to 175 lb 4 minute* 



To further show the positive control over the combustion, the 
l)erformance of one of these boilers in Table 4 is of interest. Dur- 
ing the 14-hour run, from 6 p.m., February 1, to 10 a.m., February 



TABLE 4 SHOWING POSITIVE CONTROL OVER COMBUSTION 



Date 



February 1. 1919. . . 



IVhruary 2. Ill 10, 




5.00 p.m. 



• • • 



6.30 p.m. 



9.00 p.m. 
10.00 a.m. 



13.00 


6.0 


13.6 


4.2 


13.2 


4.2 


13.0 


6.0 


13.8 


• • • 


14.2 


8.2 


14.4 


• • • 


14.6 


8.4 


14.8 


4.6 


13.0 


5.0 



CO 









0.2 



1 







2, for which iHTiod the boiler was si^uUhI and operated at about 150 
IHT cvnt rating, it will bo iu>tod that there w:is a change of only 
0.2 \M^T ci'iit in CO.. and 0.4 ixt ct^nt in Oj, and no slag was produced. 

i>bs<Tvaiions made of the stai'k from the outside of the power- 
plant buildiiii:. wlion ojHTaiini: the pulverizod-fuel and the sbd^er- 
oquipivd boilrrs in I'onibinaiion and iiulojx^ndently have also demon-' 
strau\l that no smoke was pnuhuH'd from the pulverized-fuel-fired 
iM.ilrrs. 

The four pulvtMi/ed-turl-tH|uipped boilers at this plant 



DISCUSSION 413 

cut in on ihe main line over 90 per cent of the time during the months 
of February and March 1919, and when line banked no fuel was 
required. 

Referring to various information as set forth in Mr. Harrison's 
paper, I would like to point out the following: 

I can hardly agree with Mr. Harrison that the equipment for 
preparing pulverized coal has been developed past the experimental 
stage. There is still considerable to be done in this direction, par- 
ticularly with respect to the eUmination of the reabsorption of 
moisture and of entrained moisture, which results in condensation 
in the dry- or pulverized-fuel bins and the consequential trouble 
that it causes. The type and operation of drier and pulverizer has 
a great deal to do with these factors, and experience has demon- 
strated that certain fuels require special treatment in that regard 
or trouble is boimd to occur. 

From my experience, when pulverized fuel is used for steam 
generation the detrimental efifect from the ash and sulphur content 
is nil, regardless as to the melting point of the former. 

The economic use of coke breeze is at this time problematical, 
as the expense of preparing the fuel will, in a large number of cases, 
offset the fuel value of this by-product. A study should be made of 
each specific appUcation before the use of coke breeze is gone into. 

The utiUzation of the full thermal value of anthracite silt has 
been developed to a point where there are no commercial obstacles 
in its way, and this in itself will release an equivalent amount of 
conmiercial size of coal. The shipment of silt to a considerable 
distance from the mines, however, may not be economical, but the 
power consumption by the mining and nearby industries is suf- 
ficient for the full utiUzation of this by-product, which should make 
it unnecessary to ship the silt to other points. 

The author's specific statement as to how the coal should be 
pulverized is not in conformity with my experience on this sub- 
ject, as the degree of fineness should vary with the character of the 
fuel. The results of careful analyses and tests have demonstrated 
that this degree of fineness for effective and economical use cannot 
be arbitrarily set down, as it varies with the volatile content and 
the combustion characteristics of the fuel handled. 

Commercially the expense of pulverizing increases with the 
degree of fineness, and consequently the coarser the particles of 
the fuel the less the cost chargeable against the pulverizing process, 
for which reason it is advantageous to use coarse pulverization 
whenever the character of the fuel is such that this is feasible. 



414 PULVERIZED FUEL 

From my experience the author's tabulated costs for pulverizing 
are somewhat misleading, as the cost for preparation will vary- 
over wide ranges. While the information as set forth is valuable 
from a general standpoint, it must be used with care for the reason 
that it is not susceptible to specific adaptations. 

In the case of open-hearth practice the system to be used for 
burning the pulverized fuel determines the degree of fineness. The 
high-velocity method of burning, such as described, will, of course, 
require much finer grinding than a low-velocity method such as 
obtains with other systems in which the time limit is increased, 
thereby permitting combustion to take place more slowly with the 
same thermal results. 

By using a low-velocity flame the length of from 6 to 8 ft. men- 
tioned by the author can be decreased. 

Some of the steel companies have reduced their pulverized-coal 
consumi)tion per ton of output to a little less than 450 lb. instead 
of 500 lb. as given in the paper. 

It has been found in operating practice that pulverized-fuel- 
burning furnaces, under the best system of combustion, can be n*- 
duced to a little less than 1 cu. ft. of volume per b.hp. developeil, 
and that the rating can be increased up to the capacity of the furnace 
to keep the lower tubes of the boiler free from honeycomb and still 
maintain effective combustion below the lK)iler-tube line. 

The l)est svstem of combustion necessitates — 

1 Projx^r preparation of the fuel 

2 Effective means for feeding 

3 Furnace pro|XTly designed and eciuipi^ed with auxiliar>' air 

supply to insure a distinct hot zone and a distinct cold 

zone, and gas an^as and baffles an<l draft coordinating 

with the fon»going. 

The pressure at which pulverized fuel can be admitted to the 

furna<v has hiHMi found, in stTvioo practice to l»e not more than 

phis (>.().'> in. of wMtor at the burner ntizzU*. 

In my opinion a InultT cannot W successfully o|)erated at any 
i':i|>ai'ity wiih the natural draft obtaini'd with a X(>- to H5-ft. stack, 
tho HMson lu'ini: that the frii'tion li>sses of the prases passing through 
ilir hoilrr tul»i's. and the cas areas prodtUMMl by the baffles, will 
ro-lu*"!' 0:r' draft, i\\w to this low lu*ii:lii. Whm oj'NTating at full 
i:ip:ti'ir\ ai'fM! JM<) ill. t>f wator draft in tlu' furna»v, plus the fric- 
tion. !,><v,N iliioMiih the boiltT si'tiini:. is i lie mininunn uniler which 
an> pu!\i ri/iii-furl I'oiltT installation should be ojHTaied AD\'thing 



DISCUSSION 415 

less than this will limit the factor of regulation and will tend to 
cause constriction of heat to the furnace chamber and result in slag, 
which latter, in all cases, is to be avoided. Any pulverized-fuel- 
buming installation, in combination with a steam generator, in 
which the formation of slag is of any considerable consequence, can 
be taken as a failure from the standpoint of effective and economical 
resulits. 

Under proper combustion conditions the use of a hand-lance 
steam jet to clean the bottom tubes as recommended by the author 
is unnecessary. The top of the tubes should, of course, be kept free 
from the accumulation of ash, which acts as an insulator and which 
tends to carry the heat in the ash and gases through the boiler to the 
stack. 

While any ash accumulation in the bottom of the furnace 
should be periodically removed, at the same time with a proper 
design of hot-zone and cold-zone furnace any ordinary accumula- 
tion should not convert into slag. 

In comparing pulverized equipment with stoker equipment, the 
author states that considerably less excess air is required with the 
former. The amount of air injected with the fuel into the furnace, 
however, should be a negligible percentage of that required for com- 
bustion, the only function of that air being a conveying and com- 
mingling medium. In the Edge Moor installation to which the 
author refers, the percentage of air entering with the fuel is about 
1^ per cent of that necessary for combustion, the remaining 98J 
per cent being induced by the stack. 

Pulverized Fuel for Locomotives, From a study made of the coal 
measures in the southern part of Brazil during 1904-06, it was 
concluded that the native coal was unsuitable for economic use. 
Later, during 1915-16, at the direction of Dr. Miguel Arrojado 
Lisboa, then Director of Government-Operated Railways, an in- 
vestigation was made of the use of pulverized fuel in the United 
States, with the result that the Central Railway of Brazil decided 
to install at 15-ton-per-hour capacity fuel-preparing and coaling 
plant and a stationary boiler equipment at Barra do Pirahy, an 
engine house and shop terminal about 65 miles north of Rio. Plans 
and specifications were prepared and installation was made of the 
**Lopulco^' system by the International Pulverized Fuel Corpora- 
tion of New York. Arrangements were also made for the equipping 
of 250 existing and new locomotives with the same system and by 
the same company, and since that time twelve ten-wheel-type and 



416 PULVERIZED FUEL 

two Consolidation-type locomotives have been newly built and so 
equipped, by the American Locomotive Company, and put into 
regular use on the Central Railway of Brazil. 

The first official run with Brazilian native coal, pulverized, was 
made on September 9, 1917, with a special train tliat transported 
Dr. Wenceslao Braz, President of the Republic of Brazil, and his 
staff. T(*n-wheel-type locomotive No. 282 handled the President's 
special train from Barra do Pirahy to Cruzeiro, a distance of about 
90 miles, and during the greater part of the trip President Braz 
remained in the locomotive cab and fired the locomotive, on which 
the steam pressure was fully maintained throughout, without any 
smoke. 

As the result of this performance, President Braz sent a tele- 
gram to the Minister of Public Works, as follows: 

From Barra do Pirahy to Vargem Alegre, I traveled on ten- 
wheel locomotive No. 282, fitted for the use of pulverized fuel, witli 
excellent results. The trip was made with a velocity of 63 kilomet(»rs 
per hour, having a train of 210 units behind it. I take great pleasun* 
to give you this communication which I am certain will 1x5 receivcMl 
by all Brazilians interested as solution of one of your most impor- 
tant national problems. Salutation. 

Wkxceslao Braz. 
Barra Mensa, September 9, 1917. 

The Central Railway of Brazil locomotives equipped with 
pulverized-fuel-burning equipment are oiK*rating in fast-passenger, 
mixed-passi^nger and freight, and freight service, and all are giving 
excellent results. 

In tests rec(»ntl3' conducted with Brazilian native coal and lig- 
nite the distance traveled during trials was 118 miles and the boiler 
j^ressun* remained almost constantly at 175 11)., which is working 
pressun*. TIk* results of th(\*^e t(»sts may be found in Table 5. 

The only difficulty inrt with has bi»en in instructing the engine- 
men, who were not ac(iuainted with this method of combustion, 
and for this purpose an illustrated instruction Imok has been issued 
to ('ach man. 

Only by adopting this pulverized-fud sy^?tem hiis the problem 
of \hv utilization of Brazilian fut^l. which cannot l)e bumed prac- 
lirally or iMMHioinically on ^ratfs or in n't(»rts. or utilized to good 
a«lvantai:i' r»>r the produ«'tion of j^n Mincer tjas. lu'en solved, and the 
drvclupnu'iit <>t' thr hativr coal fu'Ms of the country is now in process 
tlihHmh the c>tal)1i>hincnt of steamship and niilway means of trans- 
poriatjiin t'lciin tin- iiiinr^. and in the actual mining developments. 



DISCUSSION 



417 



In the United States the development work in connection with 
the use of pulverized anthracite and bituminous coals and Ugnite 
for steam locomotives has been carried out by making applica- 



PERCENTAGE ANALYSES OF FUELS USED 



Name 



Jacuhy 

Santa Catharina 
S. Jeronymo . . . . 
Cacapava 



Kind 



Bituminous 

Bituminous 

Bituminous 

Lignite 



Moisture 



6.10 
12.60 

3.00 
19.00 



Volatile 



22.40 
36.00 
31.00 
36.00 



Fixed 
Carbon 



45.70 
42.10 
39.30 
19.20 



Ash 



19.80 

9.00 

26.70 

25.80 



B.t.u. 
per lb. 



10,851 

10.259 

9,565 

5.249 



PERFORMANCE DATA 



Fuel 


B.t.u. 
per lb. 


Quantity 

burned 

per trip, 

net tons 


Evapora- 
tion, lb. 
water per 
lb. <A fuel 


Ash found 
in firebox 


Name 


Kind 


and pan, 
lb. 


Jftc^ihy 


Bituminoiis 

Bituminous 

Bituminous 

Lignite 


10.851 
10.259 
9.565 
5.249 


4.19 
3.5 
5.419 
4.41 


7.2 
7.1 

7.1 
7.3 


176 


Santa Catharina 

S. Jeronymo 


176 
198- 


Cacapava 


220 







tion to single locomotives of different types which were distributed 
on five different railroads of the country in order to determine upon 
a composite and interchangeable pulverized-fuel-feeding, burning 
and furnace equipment that would be adaptable to any kind or 
size of steam locomotive, as well as to all possible fuels or combina- 
tion of fuels locally available, and which at the same time would 
permit of the quick conversion from pulverized fuel to fuel oil, and 
vice versa. 

When it is taken into consideration how many modifications of 
firebox, grate, ashpan, brick arch, smokebox draft appUances, ex- 
haust nozzle and stack designs and equipments are required to 
adapt steam locomotives to the various anthracite and soft bitu- 
minous coals and lignites as used for fuel, even on a single railway, 
it can readily be imagined what the development of a single pul- 
verized-fuel-firing mechanism and furnace arrangement for the 
entire United States has involved, particularly to make it adaptable 



418 



PULVERIZED FUEL 



to existing us well cOS new designs of locomotives. For example, 
the time rociuired for the development and the practical use of fuel 
oil and of a Siitisfactory superheater is comparable. 

During the past year the financial, labor and material condi- 
tions on steam railways, brought about by the war, have prevented 
any appropriations being made for the equipping of operating ter- 
minals and divisions in the United States for the extended use of 
pulverized fuel, but the result of what has obtained may be summed 
up in the following data applying to The Delaware and Hudson 
Company and the Atchison, Topeka & Santa F6 Railway: 

On The Delaware and Hudson Company a newly built Con- 
solidation tyjH^ of freight locomotive, No. 1200, with a tractive power 
of from 01,400 to 04,000 lb., was equipped for experimental purposes, 
frt>m March 1916 to August 1917, and operated in road freight 
service betwinni Carbondale and Plymouth, Pa., and Oneonta, N.Y., 
on runs of frwn 37 to 94 miles one way. Pulverized fuel was supplied 
fn>m The Hudson Coal Company's stationary-boiler experimental 
pulverizing plant at Olyphant, Pa. 

This locomotive was designed for a working steam pressure of 
195 lb., but the Innler was designed to carry 215 lb. steam pres- 
sun\ Witli 195 lb. working pressure the cylinder horsepower rating 
is 23t>S and the Innler hoi*sei>ower rating 2540, giving a 107.2 per 
cent Innler. 

Pulverizeil-fuel tests wen* made with the following adjustments: 



Atlju!it incut 






Boiler 


Tractive 


Fmetaroi 




piYMurc, lb. 


power. lb. 


adhenoo 


R«suIU 


195 


61.400 


4.36 


O.K. 


1\X) 


63.000 


4.24 


O.K. 


LXV» 


64.600 


4.14 


O.K. 


i 210 

I 

1 


m:2oo 


4.03 


O.K. 



Tho raw oivtl which was supplied for those tests analyzed about 
:\s follows: 

Pag* 418, bottom table should read; 



Content 



Anthracite 
Slush 



Moisture . 
Volatile . 
Fixed Carbon 

Ash 
B.t.u 



1.30 

6.34 

65.70 

27.96 

10.500 



Anthracite 
Birdseye 



0.50 

8.44 

62.66 

28.40 

10.100 



Bituminous 
Slack 



1.67 
23.63 
65.16 
13.21 
13,671 



DISCUSSION 



419 



This raw coal was mixed in the proportion of 60 per cent an- 
thracite and 40 per cent bituminous, which, after drying and pul- 
verizing, produced a fuel of from 15 to 20 per cent volatile content. 
This was entirely satisfactory for locomotive purposes and yielded 
an average of one boiler horsepower for each 1.4 sq. ft. of combined 
firebox and tube heating surface. 

Dynamometer-car tests conducted to determine sustained pull- 
ing capacity on heavy grades and at starting gave the following 
results: 



MftTlmuiP 


Speed 


Reverse 








dynamometer 


miles 


lever 


Throttle 


Boiler 


Grade on 


drawbar 


per 


cutK)flF, 


opening. 


pressure. 


line. 


pull, lb. 


hour 


per cent 


per cent 


lb. 


per cent 


64.000 


At start 


Full 


75 


200 


1.65 


59.000 


6 


66 


FuU 


205 


1.65 


58.000 


8 


66 


FuU 


205 


0.72 


56,000 


lOJ 


66 


FuU 


205 


0.72 



During these tests a fuel mixture of 60 per cent anthracite 
birdseye and 40 per cent bituminous slack was used, and the ap- 
parent evaporation ranged from 7.3 to 9.3 lb. of water per lb. of 
coal consumed. The coal fired per 1000 ton-miles averaged 202 lb. 

In heavy-tonnage-service runs — over ruling grades of from 0.72 
to 1.65 per cent — for a distance of 37 miles the following data 
show typical performance: 



Item 



Miles run 

Speed, average, miles per hour . . . 

Ton-miles, actual 

Ton-miles, adjusted 

Coal consumed per 1000 ton-miles 
Steam pressure, average, lb 



Trip No. 1 


Trip No. 2 


37 


37 


14.5 


13.1 


83.147 


85.758 


88.553 


90. U3 


186 


202 


199 


200 



When in hcavy-mine-run service between Carbondale and 
Olyphant, Pa., for the three months' period, March 13 to June 12, 
1917, the performance of the No. 1200 was as follows: 



420 



PULVEBIZED FUEL 



Period 


Days in 
road service 


Hours ill 


From 


To 


road service 


1917 
March 13 
April 13 
May 13 


1917 
April 12 
May 12 
June 12 


28 
27 
26 


301 hr. 3 min. 
301 hr. 30 miu. 
273 hr. 10 min. 


Total .... 




80 

• 


875 hr. 43 min. 









After the day^s work, upon arrival at the Carbondale engine 
tenninal, the locomotive would be run directly into the house, no 
fire, track or ashpit delays or work being required. 

On the Atchison, Topeka & Santa F6 Railway an existing 
Mikado type of freight locomotive, No. 3111, with a tractive power 
of 59,600 lb., was equipped for experimental purposes — from May 
1917 to July 1918, and operated in road freight service between 
Fort Madison, Iowa, and Marceline, Mo., on runs of 112.7 miles one 
way. Pulverized fuel was supplied from the company's experimental 
pulverizing plants at these points. 

Dynamometer-car tests were run with the following average 
results, using Frontenac, Kan., run-of-mine bituminous coal, averag- 
ing in analysis when pulverized as follows: 

Moisture, per cent 1 . 05 

Volatile, per cent 32.67 

Fixed carbon, per cent 51 . 57 

Ash, per cent 14. 71 

Sulphur, per cent 3. 95 

B.t.u. per lb 12,022 

Per (rent through ll)0-niuah 97.8 

Per criit through 20()-inosh S2.C 

The p'lH'nil ])CTfonnanco uf the locomotive ecjuipix'd with the 
Lopulco pulv(M'izc(l-fm*l system was as follows: 



4 



DISCUSSION 421 



DuOaafniiM Mar. 4 to Mar. 22, 1018 

Total tripa nm (112.7 milea eaeh) 14 

Total ndlea nm 1678 

ATwace nmoinc tinM 5hr. 6 min. 

AToracBqMMl, milaaper hour 22.3 

Avvrace train tonnace, net tona 2273 

AToaoB gnMB 1000 toiHnika 256.5 

AvwacB ooal per grooi 1000 ton-miles, lb 82 .4 

Averace water per groee 1000 ton-milee, lb 666 

ATeraoB boiler preeeore, indicated, lb 188 

ATerace feedwater t«nperatare, dec. ' ^br 48 

Arerace fine-giM temperature, deg fahr 553 

ATeraae emokebox draft, inebes of water 11.3 

ATvacB finboz draft, inobeeof water 1.3 

Arerace <iaality of steam, per oent dry 96 .0 

Arerace superbeat in steam, deg. fahr 233 

Awage lb. of coal per hour of running time, per equivalent sq. ft. of grate area 71.3 

Average lb. of ooal per hour of running time, per sq. ft. of boiler heating surface 1 .01 

FUBL PBUrORMANCB 

EUiuivalent evaporation, lb. of water from and at 212 deg. fahr. per lb. of coal for boiler 

and superheater 9.22 

Per boiler horsepoww for boiler and superheater 11.15 

Combined efBciency for boiler and superheater, per oent 74 . 5 

Thermal efficiency for locomotive, per cent 4 . 19 



An actual evaporation (not corrected for the quality of steam) 
showed at the rate of 8.46 lb. per sq. ft. of boiler heating surface. 

The combined boiler and superheater efficiency showed a gain 
of 23.2 per cent for pulverized fuel as compared with hand-firing. 

Based on the hand-firing performance, the use of pulverized 
fuel showed a saving of 22.5 per cent in fuel. The combustion was 
practicaUy smokeless and the pulverized-fuel operating mechanism 
gave no trouble. 

Fig. 1 shows a typical application of a pulverized-fuel-buming 
furnace as adapted to a modem steam-locomotive type of boiler 
for the use of any soUd fuel having 15 per cent or higher volatile* 
or for fuel oil, regardless as to the other chemical characteristics. 

Pulverized Fuel in Marine Service. During September 1918 
quite a number of tests were made on the USS Gem (SP-41) on Long 
Island Soimd to determine what results could be obtained from the 
use of Navy Department specification coal in pulverized form as 
compared with oil and other fuels. 

One of the two Normand type of water-tube boilers was equipped 
with two pulverized-fuel feeders and burners, the furnace being 



422 



PULVERIZED FUEL 



fitted up with firebrick in a manner that would enable the use of 
either pulverized coal or fuel oil, or a combination of both. On 
account of the boiler not being equipped with induced draft the 
pulverized-fuel induced-air burners were connected to the regular 
air-blast system instead of being open to the atmosphere. 




Costdtu ITn 
, ^ /^rfo'vfrd..... . 



L /^rfo'Vf 



Half Section 
on B-B 




Ha !f Section 
cr AA 



Fia. 1 Typical Application of a Pulverized-Fuel-Burning Furnace 

TO A Steam-Locomotive Boiler 

The characteristics of the coal used, after pulverization, were 
as follows: 

M(>i^<t ure, per cent 1 . 02 

Volatile, per cent 18. 70 

Fixed carbon, per cent 75. 10 

Ash, per cent 5. 18 

Sulphur, per cent 0. 65 

B.t.u. per lb 14,975 

Fineness: 

Percentage through a 2(K)-!nesh s<Teen 83.6 

Percentjvge through a 100-inesh s(Te<»n 92.0 



On September 18, 1918, a four-hour run was made, from 8.30 
A.M. to 12.30 P.M., during the last two hours of which the mo6t 
economical speed for the ship, i.e., 16 knots per hour, obtained, and 
which onal)l(^d the enjriues to take the steam as fast as generated. 
This sjMH'd was niaintaiiu'd for the two-hour period, and .then, as 
(luriiijz; tlir I'litin^ four-liour ti'st run, the furnace operation was good 
there was no heat etlect on the refractory —and there was either 
no smoke, or it was very lijrht ; there was no accumulation of slag 
or ash on llu^ boiler tubes or settings, and had lh(^ boiler been equipped 
with induced draft the etlieitMiey results would hav(» been Ix^tter and 
(he light srnok(» that was pnxlueed would havt' b(M»n entirely elimi- 
natetl. 



DISCUSSION 



423 



The log of the test for the last two hours of this four-hour con- 
tinuous run was as follows: 



TEST NO. 23. SEPTEMBER 18, 1918 



Time fired up 8.30 a.m. 

Time of test 10.30 a.m. to 12.30 p.m. 

Duration of test, houn 2 

Arenge speed, knots 16 approx. 

Average boilo* pressure, lb 210 

Average superheat, deg. fahr 67.5 (max. 86) 

Average flue-gas temperature, deg. fahr 541 (max. 570, min. 500) 

Average indicated horsepower 751 

Average revolutions per minute 261 . 5 

Total pulverised fuel fired, lb 3,235 

Coal per Lhp.-hr., lb 2 . 15 

Total water evaporated from and at 212 deg. (on basis of 25 lb. 

per i.hp -hr ), lb 37.550 

Water evaporated per lb. of coal from and at 212 deg., lb 11.6 

Boiler eflBciency, per cent 75 approx. 

COj per cent Avg, 13.5, max. 14, min. 13 

Coal Pocahontas Bituminous 

Lepulco system equipment operation Good 

Lcpulco system furnace operation Good 

Smoke Light 

Brickwork heat eflFect None 



This same boiler when using fuel oil of about 18,500 B.t.u. 
and when operating at a speed of 14 knots, develops an indicated 
horsepower-hour on 1.68 lb. ©f such fuel. Therefore, with coal of 
14,975 B.t.u., in order to give equivalent results it should use 2.08 
lb. per i.hp-hr.; whereas the performance at 16 knots shows 2.15 
lb. per i.hp-hr. 

A comparison of superheat as obtained with straight pulverized 
fuel and with fuel oil, respectively, on various test trips follows: 



Trip 
No. 



22 
17 
23 



Pulverized Fuel 



Average 

knutd 



10 
12 

16 



Average 
superheat 
deg. fahr. 



43 
73 
66 



Trip 
No. 



14 
16 
24 



FuelOU 



Average 
knots 



10 
12 
14 



Average 
superheat 
deg. fahr. 



45 
70 
58 



424 PULVERIZED FUEL 

Edwin Lundgren, who opened the oral discusfflon, called at- 
tention to the large combuation space specified in Mr. Hanison'n 
paper for a boiler burning pulverized fuel ae comparea to a stoker- 
fired boiler, especially when the boiler was to be operated at, say, 
300 per cent of rating. In the matter of draft, he thought Mr. Har- 
rison had not taken the resistance of the boiler into consideration in 
specifying but 0.10 to 0.15 in. For the ordinary type of boiler a 
draft of about 0.15 in. was required at 100 per cent rating, which 
increased to about 0.7 in. for 300 per cent rating. A stack 100 ft. 
high would give about 0.6 in. draft, so that a height of 36 ft, was out 
of the question. 

As to the low efficiency (63 to 65 per cent) of stoker-fired plants 
mentioned in the paper by Messrs. Scheffler and Bamhurst, he 
would cite the Detroit Edison plant where an average efficiency of 

t 
1. 



fyCwt Boil»r Rating. 

Fia. 2 BouiER EFnctEKCT Curves, Cincinnati Gas and Himcimic Co. 
76 per cent was maintained throughout the year, and where Dr. 
Jacobus' series of 24-hour tests showed eSiciencies of 80 per cent or 
better. Also, the Cincinnati Gas & Electric Co. maintained a com- 
bined boiler and furnace efficiency of about 80 per cent in their 
plant constructed to maintain 365 per cent maximum capacity and 
300 per cent continuous capacity. 

The efficiency curves are shown in Fig. 2. 

With regard to combustible in the aali, he said, the plant of the 
Boston Edison Co. had a record of as low average percentage of 
nimbustible ns 8 and 10 per cent and the Detroit Edison plant as 
Iinv as 11 per cent, under daily operating conditions. 

Then- w:is no ijuostinn but what pulverized- fuel equipment 
would be useful in burning some of the rhonpcr grades of fuel, tike 
some of the western fuels, for which it pcemed no real type of stoker 
!i;nt hcfu dcsigneil. He did not believe, however, that it would be 



DISCUSSION 425 

m 

generally applied to modem central stations, for one reason because 
the pulveri2ed-fuel-eq\iipment buildings, to judge from the slides 
shown, appeared to be about as large as the power stations proper. 

C. F. HiRSHFELD said that in modem boiler plants the chief 
losses were due to carbon in the ash and to heat escaping up the 
stack. The Detroit Edison Co. operated under good conditions with 
10 per cent of carbon in the ash, or a very small per cent of the fuel 
originally fired, and with but 10 to 12 per cent of excess air. If use 
of powdered fuel would make it possible to bum all the carbon in the 
coal — which was contrary to his experience — and with no greater 
amount of excess air, the saving effected would still be of little con- 
sequence. He believed in using powdered coal where conditions 
were such that it was distinctly economical to do so, but these did 
not appear to obtain at the Detroit plants. In his opinion there was 
greater chan<!e for improvement in the turbine end of the plant than 
in the boiler room. 

As to greater uniformity of flame and temperature, which con- 
duced to longer life of the fumace lining, he believed that the ad- 
vantage was with the stoker with its smooth bed of incandescent 
fuel. While powdered fuel might be more adaptable when a wider 
range of fuels had to be used, the stoker seemed just as adaptable 
so far as the art had developed. 

In the paper by Messrs. Scheffler and Bamhurst the power for 
operating a stoker was stated to be 2 per cent of the total boiler hp. 
In the Detroit Edison Co/s plant, however, it was much less than 1 
per cent with the boilers operated at 200 per cent of rating. 

H. Wade Hibbard said that in a larger plant he had in mind an 
increased eflBciency resulted from adding moisture to coal that had 
dried out too much before it reached the stokers. He therefore 
asked whether drying was necessary in order that the coal might be 
thoroughly pulverized, and also whether there had been any experi- 
ments performed in introducing steam along with the powdered coal. 
Afl5rmative replies were respectively made to these queries by 
Messrs. ScheflBer and Bamhurst, the latter adding that he did not 
know whether the introduction of moisture had increased the effi- 
ciency. 

P. A. PoppENHUSEN, referring to a statement made by Mr. 
Lundgren, said that his company had successfully adapted the 



426 PULVERIZED FUEL 

chain-grate stoker to the burning of lignites of all grades and con- 
taining as high as 30 per cent of moisture, and that 24-hoiir tests 
had shown eflBcicncies of from 76 to 78 per cent. The power to 
operate a stoker was insignificant — less than 0.5 hp. per unit, 
whether a 200-hp. or a 600-hp. unit. He thouglit Mr. ScheflBer's 
figure of $64,000 low for the cost of a pulverizing plant of 100 tons 
daily capacity, and asked how large a plant should be in order that 
this cost would not be prohibitive. 

John Van Brunt, calling attention to Mr. Harrison's refer- 
ence to coke breeze, said he could state that over 75,000 hp. of 
boilers equipped with traveling-grate stokers were now satisfactorily 
operating on this waste fuel in the large steel plants and gas works 
of the country. 

Gilbert A. Young spoke of successful experimental work car- 
ried on for the past two years at Purdue University on a furnace 
for burning crushed, undried coal containing as high as 20 per cent 
moisture. The particles after crushing ranged from the size of a 
match head down to powder passing a 200-mesh screen. He hoped 
to submit a paper later giving comparative tests of the crushed-coal 
furnace and several types of stokers, all operating under the same 
conditions. 

W. P. Frey ^ said that his company burned 1,000,000 tons of 
coal a year for steam and electric power, and that he was inter- 
ested in pulverized fuel as a possible means of utilizing the twenty- 
odd million tons of small coal lying on the ground in the anthra- 
cite regions. But as it was said to take twice as much power to 
pulverize anthracite as bituminous coal, and as the minimum wage 
for the anthracite union laborer was 43 cents i>er hour, this would 
make the cost of pulverizing the $1 coal in question about 75 cents, 
and he did not sec how pulverizing would make up for this enormous 
difference. Firing a boiler at a high rating with anthracite required 
far more draft at the damper than the 0.10 to 0.15 in. mentioned in 
Mr. Harrison's paper, and in consequence stacks much higher than 
40 ft. The iiro-hrick i>rol)lem was a very important one, and he 
would he obliged if Mr. Sclieffler or Mr. Barnhurst would give 
ppcciiications for bricks that in the lonK run would stand the steady 
temperature of over 3000 (lej]j. that they would have with their 

* V\U'\ I^nginccr, Lchigli Coal & Navigation Co., I^AZisford, Pa. 



i 



DISCUSSION 427 

low supply of air, unless they reduced the temperature in the com- 
bustion chamber. 

W. L. WoTHERSPOON referred to tests he had made ten years 
ago in South Africa on a Bettington boiler fired with pulverized 
fuel and the comparative results obtained between it and a B. & W. 
boiler with a chain-grate stoker. With boilers of about 1000 hp. the 
overall efficiencies obtained were practically equal. Low-grade 
fuels of about 9500 B.t.u., of which large quantities were available 
in the Transvaal, were used, and the results obtained were such that 
some of the Bettington boilers installed at that time are still in use. 

There are several in use in England, and two or three have been 
in continuous use in Nova Scotia, Canada, for about five years. 

More recently he had been paying attention to the use of pul- 
verized fuel in connection with smelting in blast furnaces, where 
combustion takes place under pressure. Development work had 
reached a state where it had been able to carry on a continuous test 
for eight days and smelt the refractory copper-nickel ores of the 
Sudbury district with 50 per cent of the normal coke used replaced 
by pulverized coal. 

At the smelter where these tests were made coke cost approxi- 
mately twice that of bituminous coal. He also referred to tests at 
The Tennessee Copper Company ^s smelter, where pulverized coal 
had been used in the smelting of copper ores in the standard blast 
furnace. 

Continuous tests had been made for periods of ten to twelve 
days with practically all the coke replaced by pulverized coal and, 
in addition, it was found the total weight of fuel used was about 
25 per cent less. 

Regarding the cost for drying and pulverizing coal, the pre- 
war cost at The International Nickel Company^s smelter in Ontario, 
Canada, was 40 cents per ton, but owing to the increased cost of 
labor, repairs and supplies due to the abnormal conditions now 
existent, the costs were practically double. 

Referring to the transmission of pulverized fuel, the screw-con- 
veyor had been commonly adopted in the past, but compressed air 
was now being used to a greater extent, and he had recently carried 
out experiments in Canada on the transmission of fuel by com- 
pressed air and found it possible to convey 2^ tons in five minutes 
through a 3-in. pipe, over a horizontal distance of 1200 ft.^and then 
elevated 60 ft. These experiments were made in order to prove the 



428 PULVERIZED FUEL 

practibility of utilizing compressed air for transmitting about 50 
tons of fuel per day from the pulverized fuel plant at the reverbera- 
tory furnace department to the blast furnace in a separate building 
at considerable distance. 

The air pressure used was 70 lb. and the results were quite satis- 
factory and were in use for several months. This method might 
prove advantageous for the transmission of pulverized fuel from a 
central pulverizing plant to several isolated power plants by means 
of underground pipe lines. 

W. E. Snyder felt that while pulverized coal was an ideal fuel 
in the cement plant, there were, nevertheless, many problems to be 
solved in regard to its use in steel furnaces and under boilers. The 
best modem stoker-fired plants showed very high efficiencies, and 
with the same expenditure of intelligence the pulverized-coal plant 
might be brought to equal them. 

He had experienced difficulty in obtaining definite facts regard- 
ing performances of pulvcrized-fuel installations, and was skeptical 
as to the accuracy obtained in measuring the coal by observing the 
revolutions of the feed screw, for it did not always give the same 
quantity per revolution. Apropos of acciu*ate measurementSy Mr. 
Snyder caused considerable amusement by telling of the man in 
charge of a small power plant who had succeeded in overcoming 
troubles he had experienced with a venturi meter by merely boring 
out the narrow part! 

Narrowing down the comparison of the two methods of getting 
coal into the boiler furnace, lie said, it resolved itself into a ver\" 
simple matter of reducing waste in the ashpit and up the stack, and 
a possible saving in repairs and maintenance of the means employed 
to handle the fuel. 

W. F. Vkrnkr gave estimates that had l>cen prepared covering 
the cDinparative costs of ojHTatinp: six 2400-hp. boilers at the Ford 
Motor Company's blast-furnace plant with pulverized fuel and with 
stokcT \\\v\. Tho ostiinatod cost for the pulverizer equipment and 
luiiltliniis was SCOl.(KK) ami for a corresponding stoker plant 
S47r>.(K)0. Tlio rost of pulvorizing per ton, including fixed charges, 
powiT. in;iinttMi:iiu'0. luhrirants. I'to.. anvl labor (at 88 per day) was 
S(K70; \ov transiTiissioTi irom pulvorizinc building to boiler room, 
S(V2.">: an«l for boiliT room. ?l.i;>: or a total of S2.14. For a cor- 
n'spomllTii: sti^krr plant tlio liiruros wore: for transmission from 



DISCUSSION 429 

breaker building, $0^4; boiler room, $1.66; total, $1.90. For a 
plant with twelve 2400-hp. boilers the total for the pulverized-coal 
installation was $1.63, and for the stoker equipment, $1.49. These 
figures, in connection with the higher estimated efl5ciency of the pul- 
verized-fuel plant, indicated for it a saving of 4 per cent over the 
stoker plant. 

One important point in favor of pulverized-fuel plants was that 
the stand-by losses were reduced to a minimum as compared with 
stoker installations, where the fires had to be banked over the shut- 
down periods. An additional reason for installing the former was 
that blast-furnace gas would be available for use under the boilers 
and in quantity suflBcient to carry the light loads. 

John A. Stevens thought that the matter of stand-by losses 
brought out by Mr. Verner a very important one. In New England 
factories running on one shift of 9 hours a day, the stand-by coal 
required in the best hand-fired plants ranged from 20 to 25 per 
cent of the daily consumption. This, of course, decreases with 
greater number of shifts, or with continuous rims of say 24 hours, 
as in paper mills. With stoker fired boilers there is a loss in run- 
ning full load right up to quitting time, and also in burning a bank 
of coal. 

R. Sanford Riley said that the papers and discussion were 
valuable in directing attention to the problem involved in all com- 
bustion — ^the carburization of air. The final question regarding the 
two systems of burning fuel imder consideration was which would 
prove the more satisfactory imder the long, hard test of average 
conditions, and he felt that in the long run the simpler apparatus 
would win out. 

Edward N. Trump spoke of experiments he had made about 
25 years ago in burning powdered fuel under a 300-hp. B. & W. 
boiler. The furnace used had too small a combustion chamber, and 
the ash accumulated and blocked the passages between the tubes 
in less than a week. This ash had taken up tar from the combustion 
products and clung to the tubes, and it proved very diflBcult to scrape 
off. It was important, therefore, to see that the combustion chamber 
was large enough, and, as noted in Mr. Harrison's paper, that the 
boiler tubes and surfaces were carefully and frequently blown. 

H. G. Barn HURST, in closing the oral discussion, said that there 
were certain arguments in favor of pulverized coal for certain lo- 



430 PULVERIZED FUEL 

calities and for certain grades of fuel which could hardly be over- 
come by any other method of burning. One point which had not 
been brought out was the fact that by pulverizing coal and burning 
it in a furnace the boiler installation was made entirely inde- 
pendent of any particular quality of fuel. In the course of experi- 
menting his company had burned coal with as low as 2 per cent of 
volatile matter and up to 40 per cent with an ash content running 
as high as 51 per cent, just to show that the percentage of ash did 
not aflfect or interfere with the combustion conditions. Mr. Trump's 
main troubles, he thought, were due not only to the small com- 
bustion chamber used, but also to the fact that his coal was not 
finely pulverized. He wished again to emphasize the fact that the 
coal must be pulverized to a very high degree of fineness in order to 
get good results — two or three hundred million particles to the 
cubic inch. This increased the surface exposed to the air about 700 
times over that which it was in the form of lumps, and made it very 
much easier to effect combustion. Moreover, various grades of 
coal could be burned which could not be handled on stokers. 

As to tlie volume of a furnace per pound of coal burned per 
minute, he would not care to assign any particular value, because 
all combustion chambers had different shapes. The essential fea- 
ture was low velocity of the gases, for the brick became plastic at 
high temperatures, and if subjected to the action of gases at a high 
velocity, the under rows of lining would be destroyed. 

The question of furnace firebrick was, of course, very impor- 
tant, and the higher the grade used the better would be the results. 
The ash accumulation in the bottom of the furnace by further re- 
finements might be cared for by movable hearths so that it would 
not build up in the furnace and interfere with the operation. The 
ash going out of the stack, in some cases, was now being recovered 
to a certain extent, and there was certainly a field for recovering 
ash not only from pulverized-roal-fired furnaces, but also from 
stoker-tired furnaces, because the percentage of ash from many 
stoker inst:tllMtions, under forced-draft conditions, rose very high. 

Frkt>'k a. Si'Mi .ffleu and TI. G. Barnhurst. Mr. Marah 
pn^faces his rnnnrks with a n^ference to the desirability of utiliiing 
the }\v-produrt> from con]. It appears to the authors that this is 
n(»f pertiTinit to tlu* <lisciission and that it will not nniuirc considera- 
tion until it In^'oTnt^s conuntm practice to remove the by-products 
from roal U^forr firini; on hand jirates or stokers. The inference is 
that pulvi^izt'ii roal cannot W burned'after removing the volatile 



DISCUSSION 431 

constituents, a supposition which is not home out by the facts. 
His statements that pulverized coal is restricted to a field comprising 
forms that cannot be burned on stokers is an admission that pul- 
verized fuels increase the latitude of combustibility and is not sup- 
ported by evidence. 

In reply to Mr. Marsh's objections to the tests submitted, it 
should be noted that with pulverized fuels uniformity of opera- 
tion is obtained and maintained in a short time after firing the 
furnaces, and the variation in the figures need not be expected by 
prolonging the tests. In particular in the test made May 18, 1919, 
on one of the 600-hp. B. & W. boilers at the plant of the Puget 
Soimd Traction, Light and Power Co. at Seattle, during which both 
the coal and water were weighed, the duration of the test being 13f 
hours, the efficiency obtained was 78.99 per cent with coal contain- 
ing 10,106 B.t.u. per lb. as fired. 

In regard to the figures assumed for stoker efficiency, it is sub- 
mitted that the best information we have been able to obtain shows 
that an average efficiency of 65 per cent may be expected in modern 
power houses. Where power is not produced for sale this figure 
appears to be too high. Mr. Marsh does not substantiate his objec- 
tions to these figures. 

In regard to reliability, it is customary to duplicate a sufficient 
proportion of the pulverizing equipment to obviate irregular opera- 
tion in pulverizing plants. This is taken care of in the figures on 
investment charges given in the paper and makes unnecessary Mr. 
Marsh's assumption that the investment charges quoted are too low. 

In reply to Mr. Marsh's question as to the latitude of pulverized- 
coal furnaces, it should be stated that a furnace built for high volatile 
bituminous coal will not handle anthracite or coke breeze. Where 
a considerable variation in the grade of available fuels is to be ex- 
pected, however, the furnace may be designed to burn anthracite 
or coke breeze and will then be able to handle any grade of bitumi- 
nous fuel with slight changes in the arrangement of the burners. 
It is a matter of surprise to the authors that the tests mentioned by 
Mr. Marsh show tliat the fire was unresponsive with low-volatile 
coals, since it has been their experience that the flexibility of all 
grades of fuels when pulverized exceeds that obtained when burned 
in lump form. The burner for pulverized fuel is very simple and has 
a wide range of adjustments, easily meeting any requirements made 
upon it by a change in the fuel. 

In regard to adaptability, the effect of ash in coal is to carry 



432 PULVERIZED FUEL 

away a small amount of heat as the percent-age increases. The 
excess air required to burn fuels of the highest ash percentage is. 
however, negligible. It may be stated in regard to the varied effi- 
ciencies shown in the tests that pulverized-f uel furnaces require proper 
design as well as any other type, and that tests in many instances 
were a part of the experimental development and variations were 
to be expected. 

In regard to the burning of high-ash coal on chain grates, it 
should be stat<)d that the 28.4 per cent ash fuel mentioned by the 
authors was tried out on chain grates and was a failure. The stokers 
were taken out after they had been proved incapable and return 
was made to liand firing. 

In regard to flexibility it is readily admitted that pulverized 
fuel requires large furnace volumes to burn successfully. It should 
be noted, however, that other types of fuels, including oil, are burned 
more efficiently with large furnace volumes, and that the best new 
stoker installations have furnace volumes as large as required for 
pulverized coal. The amount of heat that may be developed in a 
furnace with pulverized fuel is restricted only by the resisting powers 
of the refractories up to this maximum. The flexibiUty of control 
far exceeds that obtained with any other method of burning solid 
fuels. The upper limit is merely a matter of design and rating. 

Under the head of furnace design Mr. Marsh again protests 
against the inclusion of tests of four or five liours' duration. This 
has been answered above. 

Mr. Marsh objects to the figures used for moisture percentage in 
coals. It may be stated that our tables were based on a general 
average moisture content for the entire country. Further, the 
figures given for pulverizing cost are th(» result of present information 
and the best obtainable, '^rin'y will not vary gn^atly except under 
unusual conditions. 

We must take exception to Mr. Marshes statement tliat three- 
quart ei*s of the asih goes out of the ehiiniK'V. (lenerally speaking, 
20 to 2it per cent will remain in the eoinbustion ehaml>er and 30 
to U) prr ceTit in the stHMtnd and third pass, aiul the balance in the 
HiU's and out «»f tin* staeks. NCry lilt If si-tllin^ takes place after 
tli«* pisrs lr;ivr tlir bnilrr. W ith I'oal iMmtMiniiii; 10 t«» 15 per cent 
asli till* stark «li-rliara' i?* nnt nhjt'cti«iiiaMi'. rvrii in citit^. With 
hiiilirr a>li euiitriit it niiuhl lu' imti's^^miv lo nri»vi»li' dust collectors 
n th«' \hu'< in city j>n\vrr plants. 

In answer to Mr. Marsh's qur>tion. j»u!\iMi/i'il roal is stored 



DISCUSSION 433 

in closed tanks, absorbs moisture very slowly, and under the methods 
of installation in use does not stay in storage for any considerable 
period. 

Mr*. Marsh criticizes the authors' claims for range of fuels, and 
follows by asking of what value the capacity to handle both an- 
thracite and Ugnite may be. It is self-evident that with such a 
range as this many of the intermediate fuels may be burned. His 
statement that the cost of drying and pulverizing fuel in installa- 
tions of from 500 to 600 tons daily capacity is over $1 per ton is at 
variance with figures obtained from hundreds of plants and upon 
which our estimates are based. 

Mr. Diman is correct in stating that there is a limit to the 
quality of low-grade fuels that can be used in pulverized form. 
There is no doubt that if the coal is too low in combustible the cost 
of grinding would overcome any advantage obtained in economy 
of burning these coals in pulverized form. Success in burning an- 
thracite depends upon the form of the furnace. As the coal lacks, 
volatile, increased temperature must be generated near the burner, 
and this is accomplished by the use of some of the products of com- 
bustion from the furnace itself. This means designing the furnace 
so that there is a returning flame to ignite the incoming fuel. With 
this arrangement very satisfactory operating efficiencies are being 
obtained, and there is no trouble in starting a furnace so designed 
in but a little more time than that required with pulverized bitumi- 
nous coal. 

In Mr. Peabody's discussion of Mr. Harrison's paper he states 
that oil is imdoubtedly superior to all other fuels for boiler pur- 
poses, and that the furnace volume required for oil fuel is only 
about one-quarter of the furnace volume specified in the paper for 
pulverized coal. Mr. Harrison recommends a larger combustion 
chamber than we know is actually required, but we have found 
that where oil is fired in pulverized-coal furnaces much better ef- 
ficiency is obtained than is obtained in furnaces designed for oil 
alone. Larger combustion chambers are constantly being installed 
under all boilers for the purpose of obtaining better combustion 
conditions. This is due to the fact that the best mixtures of the 
air and carbon or coal cannot be made with the present method of 
oil or lump-coal firing. 

The slight extra cost of the larger furnaces required is more 
than overcome by the increase in efficiency obtained over other 
methods of firing. It has been mentioned previously that com- 



434 PULVERIZED FUEL 

bustion chambers recommended for pulverized coal are no larger 
than those now being used in a great many stoker installations. 

Mr. Hirshfcld has probably not examined very many pul- 
verizcd-coal-fired furnaces, or he would have found that there is 
hardly any trace of combustible in the ash in the furnace. We 
do not agree with his statement that a more uniform flame and 
temperature can be obtained with stoker practice due to the smooth 
bed of incandescent fuel. A fuel bed is a very uncertain quantity in 
tlie average furnace. It cannot remain of constant thickness or of 
constant quality. The conditions of the fuel bed are beyond con- 
trol, and the coal must frequently be redistributed by the fireman. 
Certain types of stokers must be watched constantly and the holes 
in the fire bed kept covered with fuel. 

In reply to Mr. Hibbard, the authors would state that dr>-ing 
of the coal is necessary for the purpose of properly pulverizing it. 
The injection of water or steam into the firebox is a dead loss, and 
is resorted to only as a means to overcome defects in the method of 
burning fuels. There is no occasion for this with pulverized fuel. 

The exix^riments mentioned by Professor Young have no con- 
nection with the subject of pulverized-coal burning, because the 
quaUt}' of the material with which he has been experimenting is 
thousands of times coarser than the standard pulverized coal as 
now being used for commercial purposes. 

Mr. Frey takes up the cost of pulverizing coal, basing his state- 
ments on data obtained from a first-class power house. We are 
doubtful, however, even in this power house, as to whether the 
evaporation from the coal that he is using would average 6 lb. of 
water on the average over a year's operation, and if the evapora- 
tion was increased only 50 jx^r cent, certainly the increased efficienc}' 
obtained from the fuel would more than counterbalance the extracost 
due to pulverizing unless the coal is obtained for practically nothing. 

It is evident that the lower the price of the coal the larger must 
be \\\c plant and the lower the cost of preparation to show economies 
warranting its application, but the cost of coal is going up steadily 
on :u'i'o\u\{ of the increasi^l demands for fuel for manufacturing 
purjH^S(»s. 

Mr. \\ otluM-sptHHi's remarks had no bearing on the subject of 
tin* appliiMtinii of pulveri/i'd I'l^al to boiliT-tiring work, with the 
exri'jMi«»n ni* liis nftM'iMU't^ Xo tlio Hi^ttinirtiMi b\>ikT ejcix'riments which 
wt'H' 111. nit* a Tuimbrr of yi\irs aijo. aiul whii'h were an entire failure 
so !":ir as prartical opiM'ation was t-onoerned. 



j.^i 



DISCUSSION 435 

Mr. Snyder's remarks are noted with interest. The question 
of the adoption of pulverized coal to power houses is only war- 
ranted where the cost of operation would show a fair interest on 
the investment. We admit that stoker plants are operated with 
high eflSciency under careful management, but we beUeve that 
pulverized coal will show increased efficiency and lower cost of 
operation than the average stoker installation, particularly in plants 
using a large enough quantity of coal to permit obtaining a low cost 
of preparation. 

Mr. Riley's remarks are particularly gratifying in that he fully 
realizes that the question of the adoption of pulverized coal is a 
matter of dollars and cents, and that its appUcation in any installa- 
tion is a question of which system will give the best returns for the 
money invested. 

Mr. Trump spoke of experiments made in past years, and the 
writers particularly wish to point out that the furnace was too 
small for the boiler which he used, or else the quantity of coal fired 
to the furnace was beyond its capacity for satisfactory service. 
Furthermore, in the earlier days the fineness of pulverization was so 
much coarser than that now used in standard practice that there is 
no comparison, and the results obtained could not possibly have 
been satisfactory enough for commercial purposes. 

Mr. Gary is correct in his statement that the idea of the use 
of pulverized coal has been in the minds of engineers for many 
years. Many experiments and thousands of dollars have been 
spent in an endeavor to accomplish this purpose. It has, however, 
been an evolution, and the earlier disastrous results were strictly 
due to the fact that sufficient knowledge of the subject was not at 
hand. The engineers did not realize that it was not the pulverizing 
of the coal that caused the trouble, but the fact that they tried to 
bum this kind of fuel, which is nearly in the form of a gas, in a furnace 
designed for lump fuels. 

We would like to mention in regard to Mr. Gary's statements 
that today the fineness of pulverized coal is such that it practically 
all passes through the 50-mesh sieve which means that the largest 
particles are less than 0.01 in. in diameter. The velocities are slow 
enough to insure the complete combustion of the particles of this 
size, hence it is not necessary to go to the extra cost of pulverizing 
to any higher degree of fineness than is now practically the standard, 
which equals 95 per cent through a 100-mesh sieve and 85 per cent 
through a 100-mesh sieve, all passing through the 50-mesh. 



' L 



436 PULVERIZED FUEL 

In closing, the authors wish to express their appreciation of the 
attention and interest accorded them, as evidenced by the discus- 
sion. It is not their wish to present the pulverizing of fuels as a 
modem panacea, but merely to lay before the Society a statement 
of its possibilities and to remove some of the fallacies in regard to 
it which have had more or less general circulation. 

N. C. Harrison. I note that Mr. Marsh states in his discussion 
that I have used pre-war prices for the cost of pulverizing. He has 
cited the cost used in Table 1, whereas he should have used the cost 
shown in Table 3, or rates of 40 cents for millers and 35 cents for 
laborers and S5 a ton for coal. These are not pre-war prices, but 
are the present-day prices as paid by the Atlantic Steel Company. 

Mr. Marsh also speaks about pulverized coal picking up mois- 
ture when stored. I thoroughly agree with him on this, but every 
one who uses pulverized coal has found by experience that this fuel 
must not be stored. It must be kept moving at all times in the bins, 
consequently there is very Uttle moisture picked up by the pulverized 
coal from the air. 

I note in several discussions that considerable conmient has 
been made regarding the height of stacks necessary for the use of 
pulverized coal on boilers, as spoken of in my paper. During Feb- 
ruary I visited one plant installed in Kansas, and saw boOers work- 
ing there very satisfactorily with a stack of about 35 ft. in height. 

Most of the comment has been that we should have a draft of 
0.10 in. in the combustion chamber and a draft of from 0.20 in. 
to 0.60 in. at the damper. This has not been my experience for 
the best burning of pulverized coal under boilers. I beUeve that we 
should have practically a balanced draft in the combustion cham- 
ber and first passage through the tubes, and from 0.10 in. to 0.20 
in. at the damper. These figures are based on 150 per cent of rated 
rapacity. Of course, if the boiler is intended to be worked at a 
larger rating than this, it would be necessary to have larger stacks. 



No. 1703 

ECONOMY OF CERTAIN ARIZONA STEAM- 
ELECTRIC POWER PLANTS 
USING OIL FUEL 

By C. R. Wbtmouth,* San Francisco, Cal. 
Member of the Society 

In certain ateam-eleciric power plants the combination of load factor and high 
fud cost not only necessitates but also makes possible the attainment of high econo- 
mies in operation. These partiadar conditions exist in three Arizona plants 
known and located as follows: Inspiration Consolidated Copper Company, Inspires 
tioUf Ariz,; New Cornelia Copper Company, Ajo, Ariz, and Arizona Power Com- 
pany, Clarkdale, Ariz, All of these plants embody many similar feattares and in 
this paper the author discusses the economy of operation cf each. Tables and curves 
cf operating characteristics are also given and these show that the best economy per 
bbl, of oil obtained at the Inspiration Plant was 294-5 kw-kr., at the New Cornelia 
Plant S26.S kw-hr., and at the Arizona plant SSS.3 kvj-kr. AU these values were 
obtained during the winter months when conditions were most favorable. The paper 
concludes with a section dealing with the personnel responsible for the installations, 

TN certain Arizona steam power plants the combination of favor- 
able load factor and high fuel cost has not only necessitated 
but has also made possible the attainment of high fuel economy, 
and even in plants where cooling ponds are used for condensing 
purposes. This paper refers to the performance of three such power 
plants, namely, those of the Inspiration Consolidated Copper Com- 
pany^ Inspiration, Ariz.; the New Cornelia Copper Company, 
Ajo, Ariz.; and the Arizona Power Company, Clarkdale, Ariz. 
These plants embody many similar features. They differ, however, 
in methods of condensing, as cooling ponds are used at the Inspira- 
tion and New Cornelia plants, and water from the Verde River at 
the Arizona Power Company^s plant. 

2 The Inspiration plant was designed in the winter of 1913-1914. 
The International Smelting and Refining Company's smelter adjoins 
the Inspiration power-plant site, and the steam generated in waste- 

* Chief Engineer, Chas. C. Moore & Co. Engineers. 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
American Society of Mechanical Engineers. 

437 



438 ARIZONA OIL-FIRED POWER PLANTS 

heat boilers in the smelter is utilized for the operation of the recipro- 
cating blowinp; engines, which are designed for about 175 lb. steam 
pressure. These l)lowing engines are located in the same powor- 
j)lant building as the steam turbines, and since the waste-heat 
boilers are connected to the same steam header as the oil-fired boilers 
their steam pressure, and the economy of the turbine plant have 
been limited by the common steam pressure of from 175 to 185 lb. 
The New Cornelia Copper Company's plant was designed in the 
winter of 1915-1916, and l>eing independent of blowing engines, the 
boilers wore selected for 250 lb. pressure. While a higher steam pres- 
sure would have been possible, the remote location of the plant 
and experience at the date of design led to the lower boiler pressure 
being selectiHl. The Arizona Power Company's plant was designed 
in the winter of 1916-1917, and has boilers for 250 lb. pressure. 

3 The Inspiration plant is 25-cycle, 3-phase, 6500 volts. The 
New Cornelia plant and the Arizona Power Company's plant are 
both 60-cycle 3-phase, 2300 volts. The maximum load at the In- 
spiration plant was estimated to be 12,000 kw.; three 6000-kw. 
Curtiss tmbines were therefore selectetl, thus giving one spare tur- 
bine. For the New Cornelia Copper Company's plant the load was 
estimated to be 75(K) kw., and this led to the selection of two 7500- 
kw. turbines, one imit InMng a spare. The Arizona Power Com- 
pany's plant was designed as an aiLxiliar}- to a hydroelectric system, 
ami intended to carry a peak load of 5(X)0 kw. Owing to the quick 
shipment retiuiivil. however, a turbine previously ordered for another 
comi>any. and rated at 6000 kw., was installed, whereas the remain- 
ing eqtiipment was selected only for a 5(H)0-kw. load. 

1 All lliive plants have Stirling steel-<»nca.<eil l>oiler8, with 
PoalHuly-liannuol oil t'urnares. (uven fuel economizers, Moore 
autoinatii* fuel-oil reg\ilatiiiii systems. Wluvler surface condensers, 
WIuvKt dry vamuni pumps, rontrifiiiral hotwt^U pmnjis, direct-con- 
niMi^l i'\riirr<. >u\»in-drivrn boiler-frrd jninips. et«'. Superheaters 
:i:"i' i:i-'.;'!r.i :!i :i!l p':-.:;--. ^-i-r.ititvl :«• i:nr h>«^ ilr*:. superheat for the 
b>*-.- •.. :. :•'..• • :.v.! \'^o r.m -^iirr; !.»;.: •'.•:■ lM>t!i iln' New Cornelia 
 :  \-  ' i\\\. • y ^ :■ :•:.:>'- ;\:.!"-. :." :m;jm::-, -i at the lx)iler, 
..■:-•■ •■ : . . r ■■> ' ' '' •  » '" \'- '•■*' rl;ii:ts also have 

• •• _   •  ^ • :•■ •■-. •• •■ 1: -:■•:••. ' :-:•:: < ttinc turbine- 

-■••'■ ''.::•'. ' •'• 1 :-:■:■;:■ i.-n Copper 

» ■•■•..•■■■ .V ■..-• . . • •.■ -«:':.:> -•.Mild com- 

'-'■■•■' i ..* i i-^,-.:''- ;■  : :\rr:incement. 



C. B. WEYMOUTH 439 

The actual vacuum shown by the operation of this plants however, 
has been a disappointment, as a pond of insufficient area was in- 
stalled. Since the design of the Inspiration plant this detail has been 
corrected by the addition of cooling towers. The condensate is 
returned to the oil-fired boilers, and make-up water for the pond is 
purified by a Booth water softener. A Cochrane hot-process puri- 
fier purifies the make-up for the oil-fired boilers if needed, but it is 
primarily for purifying boiler feed for the waste-heat boilers installed 
at the smelter. The blowing engines operate at slightly lower 
vacuum than the turbines, and advantage is taken of this fact to 
heat slightly the turbine condensate by passing it through the Volz 
heaters in the surface condensers of the blowing engines. 

6 Cole-Bergman water weighers and Lea recorders are installed 
for feedwater measurement, for computing the steam supplied from 
the waste-heat boilers and the steam required by blowing engines, 
as well as the steam consumption of the turbines. Steam-flow 
meters are also used for checking purposes. By this means a separ- 
ate record is kept of the economy of the turbine plant on the basis 
of operation independent of the blowing engines. 

7 The feedwater for the New Cornelia plant contains from 30 
to 50 grains of impurities per U. S. gallon. The water is low in cal- 
cium sulphate and carbonate, high in sodium sulphate, and very 
high in sodium chloride. It is not practicable to purify this water 
by chemical treatment. The condensate from the condensers is 
returned to the boilers, and raw make-up water is used for the pond 
and boiler-feed make-up purposes. Frequent blowing down is re- 
quired for the cooling pond, as well as for the boilers. Scale is also 
formed both in the condensers and the boilers, and this requires 
frequent cleaning, as condenser scale tends to impair the vacuum. 

8 The Arizona Power Company uses Verde River water for 
condensing purposes, this being taken through a flume at such a 
point that the pumping head is reduced by gravity flow. The con- 
densate is returned to the boilers, and the raw make-up water is 
purified in a Cochrane hot-process purifier. 

ECONOMY OF INSPIRATION CONSOLIDATED COPPER COMPANY'S 

PLANT 

9 I'pon the completion of the Inspiration plant it was placed 
in regular service. The performance of the individual pieces of 
apparatus was investigated in order to make certain that everj-thing 
was working to the best advantage, particular attention being 



440 



ARIZONA OIL-FIRED POWER PLANTS 



paid to the efficiency of boilers, the adjustment of burners, furnaces, 
and the automatic firing system. A number of uniform load tests 
were made of boilers for checking purposes, and the results obtained 
are given in Table 1 and the curv^es of Figs. 1 and 2. These tests 
show a fairly high efficiency at rating, and, rather contrary to the 
usual results obtained with non-casing-set boilers, a higher efficiency 
at fractional loads than at rating. The higher efficiency is due to 
the tightness and the insulating efficiency of the steel casing. As 
a result of these tests, instructions were given the operators to 
divide the load equally among all boilers, and this, of course, was 
done automatically by the firing system. The operators, however, 
were instructed to keep as many boilers on the line at light loads as 
could be properly fired, maintaining a fire in each of the three 



0»5 QA 

J t TO 

50 
40 



Sttam fo Burner Pressure - Oil to Burner Pressure f3Z 



I 



c c 







10 



20 



Dash Lint, indkatts Reiafionsbipof Steam • 
and Oil Pnssure at iht Burners for a Rrnimet 
ed^tiflZtnaf Boiler Rating. [ 

This Relationship mau be DtffemMottm'Bifints  
in ihe Steam andOfftoBurnerUmsandatStiam 
tSiBurner Rcgulah r due to FhcfwtBk 
40 50 



30 



60 



Pressure in Oil Line at Burner in Lb. per Sq. In. 
Fig. 1 Steam and Oil Pressures at Burner Lines 

burnei*s i>er boiler; below this load boilers were cut off the line, and 
only refiri'il when the load again increased. 

10 The curves of Figs. 1 and 2 show the relation between the 
steam prossui*e of the atomizing steam and the oil pressure, both 
me:u«unHl in the supply pi|>es at the individual burner, between the 
throttle valve and burner. From data obtained from these curves 
the sieam-to-burner regulator was set to give the proper pressure 
of aloiuizini: steam, based on the momentary oil pressure. A com- 
plete desrription of the automatic oil-firing system in use in these 
plants will U* fi»un<l in a pajK^r by the writer entitled Unnecessary 
l.t»<MS in Hurnini: (^il Fuel and an .Vutomatic System for Tlieir 
Kliminat it'll, pam- T',»7. Vol. oO, of Tuansaitions. and the system 
u-f.i in th«-r ]>l:ii;i< diiYrrs fn»in that desi'iibed in the |xiper merely 
in thr ii-f ni" ;i iliaphraiinv pump iiovi'inor to maintain a constant 
p!«-i.:. r:L!:.i-.: n:a\iiiiuni pn'>suri' at tin' o\\ pumi>s; the oil-to- 
I'urn-r r. pi'.at-r tlnn opt-ratis a throtiK* valvt^ lo give the desired 
pn^ssiirr' at tii'- "il burni*n>. 



C. B. WETVOCTR 441 

11 Under variable load the damper controller has been able to 
maintain COt readings varying from 12 to 14.5 per cent COi, for 
which the corresponding excess air for normal conditions is 28 per 
cent and 6 per cent, respectively. 

12 After instructing the operators, and while the plant was still 
under the control of the engineers, an average economy was obtained 
for the month of September 1915, as follows: 

Average number of turbine units in operation 1 

Avoage dailr load, 24-hour basis, kw G9S0 

Average steam pressure at boilers, lb. per sq. in 178 

Average vacuum in condenaera, in. of mercury, abeolute 2.66 

Average rating on boilera, per cent 95 

Groea boiler efficiency, per cent 80.9 

Average economy, kw-hr. per bbl. of oil as fired 289 

Average economy B.t.u. per kw-hr 21,500 



t > 



11 



PerCtntof Rated Copoci+y if Boiler 

Fio. 2 PREsacEEs AND Efficiencies at VAiuon8 Loads 

In the subsequent operation of this plant t>y the owners the econ- 
omy has been maintained practically equal to that shown under 
the direction of the engineers, but during the winter, due to colder 
circulating water, the economy is even better than indicated above. 
On the other hand, during the summer months, with the warmer 
circulating water and falling off in vacuum, the economy naturally 
drops to a lower figure than that given for the month of September. 
13 This plant o|)erates in conjimction with the hydroelectric 
plant at the Roosevelt Dam, and at periods of the year preference 
is given to hyilroclcctric power. As a result, there is a fractional 
load, or partial j^hutdown of the steam plant, and for certain months 
this in turn has naturally resulted in a reduced economy. 



ARIZONA OtL-FIRBD POWER PLANTS 



W-HR.FDRE4HR. 



PERCENT OF RATED LOAD ON TURBINE 



IN INCHES OF MERCUm ABSOLUTE 



inipLu 



1 : , , I i '  ^i-~-^i 1 1 










', 












t" 




u 










'  .  ' 1 . '■ ' 








! 



PLANT STEAM COHSUHPTION.LB.PCRKW-HR. 




I\-.sp:h*thin- Power Plant 



C. B. WBTMOUTH 443 

14 The operating results for the plant, furnished by Mr. W. W. 
Jourdin, Mem.Am.Soc.M.E., Chief Engineer of the Inspiration 
power plant, are given by the curves of Fig. 3. It will be noted that 
the best monthly economy for winter conditions has been 294.5 
kw-hr. per bbl. of oil, or 20,910 B.t.u. per kw-hr.; and the poorest 
economy for summer conditions for the normal load has been 257.5 
kw-hr. per bbl. of oil, or 23,700 B.t.u. per kw-hr., although for the 
month of September 1917, due to the very light load, the economy 
was only 237 kw-hr. per bbl. or 25,970 B.t.u. per kw-hr. As pre- 
viously stated, this plant, in comparison with non-cooUng-pond 
plants, is subject to an accumulation of scale in the condensers, 
and as a result there is a sUght loss in vacuum. Since purified feed- 
water is used, the plant is not subject to troubles from scale for- 
mation in the oil-fired boilers, nor does any loss of fuel result from 
shutting down boilers for cleaning or boiler blow-off. 

15 The curves of Fig. 3 also give the operating records for com- 
bined boiler and economizer efficiency, atomizing steam deducted, 
this being the form in which the power-plant records are kept. A 
comparison of this result with the boiler efficiency tests would seem 
inconsistent without the explanation that the economizers at this 
plant heat the feedwater through a temperature range of from 40 
deg. to 45 deg., whereas it will be noted that results for the Arizona 
Power Company's economizers, given in Table 2, indicate a tem- 
perature rise in the economizer of 91 deg. fahr. In proportioning 
the Inspiration economizers with reference to the investment and 
fuel saving it was assumed that the average period of operation 
would be less than half the year, owing to the use of hydroelectric 
power; it was also assumed that the fixed charges would be very 
high. Tliis combination of circumstances, together with high 
freight rates and construction costs in Arizona, resulted in the 
selection of an economizer of comparatively small surface, giving a 
favorable return on the investment but a result considerably less 
favorable than that attained in the average economizer with oil 
fuel, when measured only by temperature rise. 

ECONOMY OF THE NEW CORNELIA COPPER COMPANY'S PLANT 

16 Following the installation of the New Cornelia plant in 
November 1917, an attempt was made by the designing engineers 
to check the economy of the station, but owing to the war and a 
scarcity of labor this work had to be abandoned before completion. 
While it has never been possible to show the best performance of 



444 



ARIZONA OIL-FIRED POWER PLANTS 



TABLE 1 RESULTS OF EFFICIENCY TESTS ON STIRLING ^ATER-TUBB 
BOILER IN INSPIRATION CONSOLIDATED COPPER COMPANY'S PLANT 

Class M No. 26 battery Stirling water-tube boiler with steel casing. Heating surface per 
boiler, 7129 sq. ft. Rated boiler hp., 712.9 (based on 10 sq. ft. per hp.). BoUer-room flow elera- 
tion, 3615 ft. Normal barometer, 26.13 in. 



Test at Test at Test at 
100 per cent 80 per centj60 per cent 
rating ^ rating * rating 



Temperature of flue gases, 
deg. fahr., measured 
across breeching outlet 
and in rotation 



Date of test, 1915 

Number of test boiler in plant 

Duration of test, hours 

Temperature of feedwater entering boiler,dcg. fahr. 

Temperature of superheated steam, dpg. fahr 

Deg. fahr. superheat 

Temperature of steam to burner, deg. fahr 

Temperature of oil to burner, deg. fahr 

No. 1 

No. 2 

No. 3 

No. 4 

No. 5 

No. 6 

No. 7 

Temperature of outside air, deg. fahr 

Temperature of air entering aslipit, deg. fahr 

Steam pressure, lb., gage 

Pressure in oil line before burners, in lb., gage. . . . 
Pressure in steam line / Before burner valves .... 

to burner, lb., gage \ After burner valves 

I Top of first pass No. 1 
Draft, inches of water i Bottom of 2nd pass No. 2 

[ Front of damper No. 3 
Draft, inches of water (i>ower-plant instrument), 

front of Damper 

Water, per cent (by centrifuge) .... 

Sand 

Sulphur in per cent of dry oil 

Carbon in per cent of dry oil 

Hydrogen in per cent of dry oil ' 

Net B.t.u. per lb. of oil as fired. . . . 
^ Sulphur corrected in B.t.u. per lb. oil 

Total water actually evaporated, lb 

Lb. water actually evaporated per hour 

Factor uf evaiK>ration 

Tutal water evaporated from and at 212 deg. 

fahr., lb 

Lb. watiT f'va|H>ratO(l from and at 212 deg. fahr. 

I>cr hr 

Tofjil oil fire<l, lb 

( til tin'd JK.T huiir, II > 

Htiilrr liorsc'iKtutT Jfv«'lni»etl 

VvT rvut of ratod rapHi-ily, ba.H'd on work done by 

wiiliT-lieatiiiK surfarf 

Lb. water actually rvapuratt-d {kt lb. of oil 

LIm. water ova|K)iar(Ml |nt lb. uf ml fruiii and at 
2\-2 dig. fahr 



Analjrsis of 
fuel oil by 
Smith Em« 
ery A Co. 



May 29 

5 

10 

210.15 

503.9 

122.1 

504.2 

187.8 

466.5 

469.0 

470.1 

473.3 

472.5 

470.8 

471.3 

95.0 

93.1 

187.4 

22.0 

56.6 

0.014 
0.062 
0.076 

0.065 
0.480 
Trace 
1.06 
85.58 
12.87 
18,510 
85 
232,730 
23.273.6 
1.124 

261,594 

2»i.l59 
16,743 
1,074.3 
758 

106.7 
13.90 

15.62 



June 1 
5 
6 
200.0 
502.4 
122.0 
502.7 
186.9 
423.7 
427.5 
434.7 
434.7 
432.5 
429.0 
427.0 i 
90.2 i 
100.3 ' 
184.03 
15.0 
46.8 

 • 

+0.016 
0.080 
0.070 

0.074 
0.490 
Trace 
1.06 
85.58 
12.87 
18,540 
85 
104,394 
17,399 

1.134 



June 2 

5 

6 
191.0 
491.0 
111.0 

• • 

186.0 

389.0 

394.0 

389.0 

394.0 

391.0 

390.0 

392.0 

86.5 

95.0 

182.3 

9.+ 

38.0 

• • 

0.013 
0.032 

o.oao 

0.033 

0.490 

Trace { 

1.06 

85.58 

12.87 

18.540 

85 

76,857 I 

12.809 

1.137J 



Teat at 

125 par cent 

rating 



June 4 

5 

10 

209.5 

517.3 

135.5 

•  

185.0 

495.0 

5Q2.0 

507.5 

508.0 

507.5 

510.5 

506.0 

69.0 

75.9 

187.3 

36.8 

67.0 

• • 

0.005 
+0.076 
+0.091 

0.110 
0.40 

1.00 
8S.5S 

12.S7 
18,6«0 
86 



26^996.3 

i.iai 



118.383 87.386 805^28 



19,730.5 

7.498.5 

1.249.7 

572 

80.3 
13.92 

1.5.73 



14.564 
5.458.5 
909.5 
422 

59.2 
14.08 

16.00 



19,907 
1.990.7 



124 
18^ 

18184 



C. B. WEYMOUTH 



445 



TABLE 1 BESULTS OP EFFICIENCY TESTS ON STIRLING WATER-TUBE 
BOILEB, INSPIBATION CONSOLIDATED COPPER CO/S PLANT iCont.) 





1 
Test at Test at • Test at 

100 per cent 80 per cent 00 per cent 

rating^ rating rating* 

81.76 82.54 83.74 

15.5 15.17 15.1 

15.6 15.41 15.0 
0.8 1.11 . 1.4 

Trace 
83.6 83.48 83.6 
14.5 . . 14.5 
2.3 2.1 

Trace |  
83.2 I 83.4 

4.19 . 4.19 

1 


Test at 

125 per cent 

rating 


Eflidmugr of boiler based on groes eraporation 
per oent. 


80.29 


Per cent COs in flue gas at top of first pass 

[COf 


16.1 
14.4 


Peroentacie analysis of cases at 


1.94 


bottom of seoond pass CO 





N 


83.66 


CO« 


14.3 


P^roentace analysis of cases at 


2.11 


front of damper CO 





N 


83.59 


V *^ 

Per eent excess air over chemical requirements at 

front of damper 


6.14 







> Apfmitai discrepancy in the draft readings oi power-plant instrument and thermometers 
are due to the different location of the noules and to alight leaks in flue-gas piping. 

* On account of the wast»4ieat boilers and intermittent firing of the other boilers the load was 
very jerky. On this account the damper was fixed to a low point on the draft and the ashpit door 
was regulated during the test. 

which the plant is capable, the operating crew have, for the most 
part, been very efficient in handling it, except during an illness of 
the chief engineer, when the economy of the plant fell oflf to a dis- 
appointing figure. This occurred during the summer and fall of 1918, 
and the figures for economy for that period are thus hardly fair 
to the plant. It should also be bonie in mind, in connection with 
performance data, that the feedwater condition at this plant is such 
that the frequent shutdowns for boiler cleaning and the large 
amount of hot water which is blown off from the boilers affect to an 
appreciable extent its economy. 

17 This plant, due to its location in the southwestern portion 
of Arizona, is subject to more intense summer heat than probably 
any other power plant in the western territory, and this in turn gives 
rise to considerably less favorable cooling-pond and condenser per- 
formance, with respect to vacuum, for the summer months than for 
the winter months. The vacuum performance is also influenced by 
the accumulation of scale \\'ithin the condensers between the periods 
of cleaning condensers, due to concentration of salts in the cooling 
pond. 

18 The best economy for this station, known to the writer, is 
for the month of January 1918, the average performance for four 
successive days being as follows: 



446 



ARIZONA OIL-FIBBD POWER PLANTS 



Date 
1918 



Jan. 26 
Jan. 27 
Jan. 28 
Jan. 29 



Average load, 
kw. 



7921 
7830 
7726 
7800 



E^nomy, kw-hr. 
per bbl. of oil 



321.7 
324.1 
324.2 
326.2 



B.t.u. per 
kw-hr. 



18.975 
18.833 
18.829 
18.711 



19 The poorest economy for this station during the summer 
of 1917 was for the month of July, namely 293.5 kw-hr. per bbl. 
the average vacuum was 1.66 in. absolute and the average load 
550 kw. The poorest economy during the summer of 1918 was also 
for the month of July, or 287.5 kw-hr. per bbl. The average vacuum 
was 2.14 in. absolute, the average load 5850 kw., and consequently 
the economy for the summer of 1918 was abnormally low. 

20 The monthly report for December 1918, gives the following: 
Average load, 7800 kw.; average economy, 312 kw. per bbl. and 
19,913 B.t.u. per kw-hr. The monthly report for January 1919, 
gives: average load, 7790 kw.; average economy, 317.9 kw. per 
bbl., and 19,535 B.t.u. i)er kw-hr. The improvement for the 
month of January over December is due to a straightening out of 
the aforementioned difficulties experienced in the summer of 1918, 
and it is the writer's belief that the station will soon be operating at 
its best previous economy for the corresponding season. 

21 It will be noted that the economy of the New Cornelia 
plant is materially bettor than that of the Inspiration. This is due 
somewhat to the larger turbine units installed, but in the main to 
the higher steam pressure and to the improved design of cooling 
pond, which results in better vacuum. In comparing the economy 
of these cooling-pond stations with that obtained in tidewater 
plants, allowance should be made for the size of turbine units, the 
obtainable vacua under operating conditions, the increased head 
on circulating pump due to the greater quantity of water which 
must be handled through the condensers, and the increaaed pump- 
ing head due to the cooling-pond nozzles and longer lengths of cir- 
culating-water line. 

ECONOMY OF THE ARIZONA POWER COMPANY'S PLANT 

22 This plant was completed in September 1917, and due to 
the war conditions in the mining region it was difficult to assemble 
a skilled operating crew. Tiie i)lant was furnished under a contract 




C. R* WEYMOUTH 447 

covering a complete plant-economy guarantee at 5000-kw. load, and 
the final test, covering 48 hours' operation in r^ular conunercial 
service, under variable load, was concerned mostly with the economy 
at this load, although a rim was made at 6000-kw. load, which is 
the rated capacity of the turbine The results for the final test are 

TABLE 2 RESULTS OF TEST OF THE ARIZONA POWER COMPANY'S 

STEAM PLANT AT TAPCO, ARIZ. 



Duration of teat, houn 6 

Boiler premire, lb. per sq. in., g»c* 250 

Steam preaBure at .turbine throttle, lb. per sq. in. face 238.5 

Avg. temp, of superheated steam at boikis. deg. fahr &I6 

Temp, of superheated steam at turbine, deg. fahr 516 

Temp, of feedwater entering boilers, deg. fahr 207 

Temp, of water leaving feedwater heater, deg. fahr. 116 

Temp, of circxUating water from condenser, deg. fahr 62 .4 

Temp, of circulating water to condenser, deg. fahr 45 . 7 

Room temperature, deg. fahr 83.2 

Barometer, inches of mercury 26.773 

Vacuum in condenser, inches o( mercury 25.783 

.Absolute vacuum, inches oi mercury 0.900 

Temp, of flue gases leaving economiier No. 3, deg. fahr 276 

.\ verage load, kw. per hr 5,815 

Power factor by power-plant indicator 0.93 

Electrical Output by Integrating Meters: 

Gross kw. generated 34390 

Groas kw-hr 5315 

Auxiliary power, k. w 277 

Net kw. output 34,613 

Net kw-hr. output 5,769 

Oil Measurements: 

Total oil weighed, lb 34,516 

Correction due to diff. in temp, at start and finish, lb 46 

OU actually used, lb 34,470 

Average gravity of oil (analysed by Smith Emery A Co.), deg. B.. 17.65 

Weight of oU per bbL of 42 gal., lb 332 

Heat value of oil (anab'sed by Smith Emery A Co.), B.t.u. per lb. 18,703 

Economy: 

Fuel used per groas kw-hr., lb , 0.988 

Fuel used per net kw-hr, lb 0.996 

Kw-hr. per bbl. of oil. groes 336.0 

Kw-hr. per bbl. of oU net 333.3 

B.t.u. per kw-hr. groas 18,478 

B.t.u. per kw-hr. net 18,628 



given in Table 2. At all times during this test the plant was subject 
to a variable load, due to the regulation of the hydroelectric sj'stem. 
The oil was oarofully weighed and the electrical output was meas- 
ured by calibrated meters. The electrical output given is the net 
useful output for the station at the 2300-volt bus, deduction having 



448 ARIZONA OIL-FIRED POWER PLANTS 

been made for the power consumption of electric auxiliaries, includ- 
ing lighting for the operators* cottages, circulating water pump, 
deep-well pump and air washer. The average electrical auxiliary 
load was 46 kw., which is somewhat smaller than would have been 
the case had the entire head on the circulating water been overcome 
by pumping; against this condition is the fact that during the test 
the quantity of circulating water was somewhat less than specified, 
so that, roughly speaking, the one condition nearly offsets the other. 

23 It is not possible to give daily operating results for this plant, 
at the load for which it was designed, for since its installation it has 
been maintained only for reserve purposes, carrying occasional 
peaks but the majority of the time a very light load, and for a num- 
ber of hours during an average day, with the turbine at standstill. 
Of course, favorable economy is not possible under such conditions, 
as the fuel losses due to keeping hot boilers, piping, etc., the dead- 
load losses for the operation of auxiliaries, and the zero-load steam 
consumption of turbine result in a zero-load fuel consimiption of the 
plant which is an appreciable percentage of the full-load fuel con- 
sumption. With the foregoing explanation, the results given below 
for the month of February 1918, are as follows: 

Total kw-hr 1,235,680 

Total bbl. of fuel oil used 4783.0 

Total hours of operation 444 

Average kw. for operating period 2800 

Kw-hr. per bbl. of oil (delivered to lines, net) 258.50 

Operating period load factor 0.467 

Average kw. for monthly period 1838 

Monthly period load factor 0.306 

24 It is of interest to note that the turbine at the Arisona 
Power Companj-'s plant is practically a duplicate of those at the 
Inspiration Cop|XT Compan^^'s plant, having the same number of 
stages, the difTerence in the ojxjrating economy of the turbines being 
larjrcly due to steam pressure. Here, again, is a marked increase 
in plant ocononi}' due to an increase in steam pressure, and also 
by reason of the colder river water at the Arizona Power Company's 
plant as compared with the co()ling-]K)nd water at the Inspiration 
{)1an1 and the eonseiiuent improvement in vacuum. All economy 
iiuiire^ ^iven for these plants are l>ase<l on oil as firod, without deduc- 
tion or eorrection for moisture, sulphur or silt. 

•J.") The en.i:in(»ers have en<leavored to instill in the minds of 
the o{HTators of these plants and to show by example that higli 



C. B. WEYMOUTH 449 

economies need not merely be looked for during the test period, but 
can be maintained during the operating period. The Inspiration 
plant is large enough to permit the employment of a boiler-room 
engineer, but, due to their smaller size, such an engineer is not main- 
tained at the other two plants. This plant was designed on the 
assiunption that the average load would be maintained for a period 
of six months only during the year, and that oil delivered would 
cost about $1.45 per bbl. The New Cornelia plant was designed 
on the assumption that oil would cost $1.25 per bbl., but since the 
date of design of these plants the cost of oil has materially increased. 

PEBSONNEL 

26 The selection of the principal equipment for the Inspiration 
plant and its general layout were made jointly by John Langton, 
Consulting Engineer for the Inspiration Consolidated Copper Com- 
pany, and Chas. C. Moore and Co. Engineers, which firm was 
also responsible for the detailed designs, installation and tuning up 
of the plant. 

27 For the New Cornelia plant the entire work was in the 
hands of Chas. C. Moore and Co. Engineers, with the approval of 
A. G. McGregor, Consulting Engineer for the New Cornelia 
Copper Company. 

28 The Arizona Power Company's plant was designed and 
built by Chas. C. Moore and Co. Engineers, with the approval 
of R. S. Masson, Chief Engineer for the Arizona Power Company. 

29 A considerable portion of the testing work on the Inspira- 
tion plant was handled by A. G. Budge, imder the writer's direction, 
and for the New Cornelia and Arizona Power Company plants by 
T. B. Paulson, also under the writer's direction. 



DISCUSSION 

C. H. Delany (written). The Pacific Gas and Electric Com- 
pany is operating at Sacramento, Cal., an oil-burning plant which 
is similar in some respects to the three plants discussed in Mr. Wey- 
mouth's paper. This plant was built in 1911 and is equipped with 
one 5000-kw. Curtiss turbine, Stirling boilers, Peabody Hammel oil 
furnaces, and a surface condenser taking water from the Sacramento 
River. This plant was designed for 175 lb. pressure and 100 deg. 
superheat at the turbine throttle, and is therefore similar in this 



450 ABIZONA OIL-FIBED POWER PIANTS 

respect to the Inspiration Copper Co.'s plant. The plant, however, 
being essentiaUy a stand-by plant, is not equipped with economizers, 
steel casings for the boilers, or automatic fuel-oil regulators. 

In comparing the operation of different steam plants operating 
at variable loads, it is convenient to plot on a diagram the kilowatt- 
hours generated against the total oil burned, each point plotted 
representing the average results of a month or a day as the case 
may be. If a sufficient number of points at different loads is ob- 
tained, it is possible to draw a line through them representing their 
average, and thb line in most cases is found to be practically a straight 
line. Such a diagram for the Sacramento station for the year 1918 



'O M *) 60 80 K»IZ0 1« ieOt60200:«12«!«)2SOMO 
Thousond Kw-trr per Day 

Fia. 4. Diagram Showiko Relation Betwee.^ Barrels or On. Busned pkr 
DAT and Kilowatt-Hours Generated 

is given in Fig. 4 and to it have boon added points taken frmn Mr. 
Weymouth's paper for the three Arizona plants. The points for 
the Inspiration Copper Co.'s plant arc sufficiently numerous to 
enable the line AC to be drawn through them, correHponding to the 
line .4^ for the Sacramento plant. It will be noted that these two 
linos interject the ordinate of zero load at almost the same point, 
.4, indicating al)out the same quantity of oil (that is, in the neigh- 
borhood of 25 or 30 bbl. per daj') required to operate either {daot 
at no load. 

TluTf- are nut sufficient points to enable a similar line to be 
drawn for cillicr the New Corneliii Copper Co. or the Arizona Power 
Co. plants. Iliiwcver, :issuniing the |x>int A for zero load to be 



DISCUSSION 451 

the same as for the Inspiration Copper Co., the line AD has been 
drawn as an approximate average for these two plants. 

This diagram brings out more clearly than a mere statement 
of the kilowatt-hours per barrel of oil, though perhaps somewhat 
crudely, the essential differences between the various plants. Thus 
a comparison of the lines AD and AC shows at a glance the effect of 
the higher steam pressure, higher superheat and better vacuum car- 
ried at the New Cornelia Copper and Arizona Power Companies' 
plants. The line AC compared with the line AB shows the economy 
resulting from the use of economizers, steel casings, fuel-oil regu- 
lators, and other minor differences such as the greater age and smaller 
size of the turbine in the Sacramento plant. 

W. W. JouRDiN (written). Referring to Par. 4 of the paper, 
the Inspiration power plant is equipped with engine-driven circu- 
lating units and barometric hot well, with sealed sump and lift pumps 
located on a hillside at a suflBcient distance below the condensers to 
h'ecure positive drainage of the condensate therefrom. 

The high-pressure cylinders of the blowing engines are being 
replaced with poppet-valved cylinders good for 250 lb. pressure, 
and all new equipment, including boilers recently installed, is de- 
signed for the higher pressure. However, imtil such time as the 
original boilers may be disposed of, a maximum of only 200 lb. pres- 
sure will be available at the throttles. 

The cooling tower was not put into service until November 1918, 
therefore the plant operated under the handicap of poor vacuum, 
due to insufficient cooling capacity, throughout the period covered 
by the curves in Fig. 3. 

Figs. 1 and 2 do not represent present practice — in fact, there 
has been a very considerable departure in numerous operating fea- 
tures from the original instructions, although these in principle were 
good. Owing to labor troubles, the plant was shut down from July 
2 to September 20, 1917, inclusive, and in the absence of detailed 
explanation concerning conditions, the curves would better be broken 
between June and October 1917. 

The heat recovery is low in the economizers not so much on 
account of inadequate heating siuface as because the recovery by 
the boilers is so high, compared with coal-burning practice, that 
little heat available for warming feedwater is contained in the gases 
to the economizers. 

The Inspiration plant is jointly owned by two companies, and 



452 ARIZONA OIL-FIRED POWER PLANTS 

the money involved is so large, due to high fuel costs, that it is 
essential that the measurement of all quantities shall be so made 
as to secure the highest possible accuracy. The metering equip- 
ment is quite elaborate, and the accounting is on the basis of contin- 
uous overall plant test; consequently the operators are enabled to 
check economic performance more closely than is usually possible 
even in the most highly developed plants. 

The economy reported Ls based on net electrical production, and 
no deduction is made for fuel consumed in warming up and stand-by 
due to changes in electrical load. 

W. D. Ennis (written). The maintenance of high overall effi- 
ciency in regular operation is a feature to be expected with well-de- 
signed plants using oil fuel. In discussing the reasons for the 
superiority of the New Cornelia plant over the Inspiration the author 
omits mention of the higher superheat used. Allowing for differ- 
ences in pressure and superheat, the steam temperatures in the two 
cases are 471 deg. and 551 deg. The load factor may have been of 
weight also, but there seems to be nothing in the table dealing with 
this plant to show whether or not both turbines were running. If 
they were, the New Cornelia load factor was around 0.50 as com- 
pared with a value of about 0.70 for the Inspiration plant. 

A most interesting feature of this paper is the indication of 
maximum boiler efficiency at about 0.60 rating (Fig. 2). The author 
attributes this to the tightness and insulating efficiency of the sheet- 
metal casing. It is regrettable that the l)oiler proportions are not 
given in some detail. The curve of boiler efficiency against rating 
cannot be considered a normal curve. The (actual) rate of evapo- 
ration at normal rating works out 3.26 lb., or at 60 per cent rating, 
less than 2 lb. 

W. N. Best fwritten). There is one particular point to be 
emphasized in the use of oil fuel in |K)wer plants, and that is the 
imiM)rlance, wherever jHJssible and wherever plants are of suffi- 
cient size, of the company's employing an engineer whose duties 
will l)e to scH.' that the oil is pro]3erly burned in order that the strictest 
economy and highest efficiency may be attained and maintained at 
all tinu»s. The writ<»r has always recommended this in jwwcr plants 
and in forge shops. an<l he is glad to learn that there is one plant in the 
United States that really has placed a man in charge, whose duty it 
is to s(»e that the oil is properly burned. He can save his wages many 



DISCUSSION 453 

times by the economy effected in fuel by his attention to his specific 
duties. 

W. J. Davis, Jr. (written) . We have to thank Mr. Weymouth 
for presenting us with some very interesting and useful information 
on the possibilities of obtaining imusually high economies from com- 
paratively small steam power plants operating in districts where 
lack of sufiicient circulating water for the condensers makes neces- 
sary the use of cooling ponds. 

The monthly average performance of the Inspiration Consoli- 
dated Copper Company^s plant of 21;500 B.t.u. per kw-hr. is es- 
pecially commendable when we consider the plant conditions, 
namely, units of 6000 kw. capacity, steam pressure of 178 lb. gage 
at boilers and vacuum of 2.66 in. absolute back pressure. If cor- 
rection is made for the low vacuum at the Inspiration plant as- 
suming 1 in. average back pressure in the condensers, a condition 
frequently obtained in modem Eastern plants, the average economy 
would be reduced to less than 20,000 B.t.u. per kw-hr. 

This figure is very good and compares favorably with the 
economy of plants of modern design using much larger units and 
operating at higher steam pressures and superheats. 

While the performance of the Inspiration plant may be held to 
be unusually good, that of the New Cornelia is considerably better 
due mainly to the remarkably high vacuum conditions resulting 
from the improvements in the design of the cooling pond and also 
to the use of higher steam pressures and superheats. 

These plants were both designed to meet unusual conditions of 
high cost of fuel and an inadequate supply of cooling water. The 
high economy maintained as shown by the monthly plant records 
should prove highly gratifying to the designers of the plants as well 
as to those in charge of operating them. 

A questionnaire on power conservation was undertaken by the 
Cliairman of the Engineering Committee of the Pacific Coast Sec- 
tion of the National Electric Light Association, the results of which 
appeared in the Journal of Electricity for April 15, 1918, and which 
showed the economies obtained in steam plants where oil was the 
fuel. The subject of power plant losses was also undertaken by R. 
J. C. Wood, General Superintendent of the Long Beach Plant of the 
Southern California Edison Company and the results of his in- 
vestigation were published in the Journal of Electricity for April 
1, 1918. 



454 ARIZONA OIL-FIRED POWER PLANTS 

Robert Sibley (written) . In the Spring of 1918 the questions 
of the conservation of fuel oil and of its utilization at the highest 
degree of eflSciency became matters of prime importance throughout 
the nation and especially west of the Rocky Mountains where there 
was a shortage of hydroelectric power and an urgent necessity for 
the economical operation of steam plants. 

In both of these articles the average economy of the great steam- 
electric power plants of the West was shown to be from 200 to 240 
kw-hr. per bbl. of oil. There was widespread interest among engi- 
neers, therefore, when it became known that Mr. Weymouth had 
secured such results as are mentioned in his paper. An average 
economy of 289 kw-hr. per bbl. of oil as fired was attained for the 
month of September 1915, at the Inspiration plant while it was 
still under the control of the engineers. At the New Cornelia plant 
the monthly report for September 1918, showed 312 kw-hr. per 
bbl. of oil, and in January 1919, the average economy rose to 317.9 
kw-hr. per bbl. of oil. At the plant of the Arizona Power Company, 
during September 1917, the final test covering 48 hours' operation 
in regular commercial service under variable load due to the regula- 
tion of the hydroelectric system showed the remarkable record of 
333.3 kw-hr. per bbl. of oil. 

Such instances as these indicate the remarkable attainments 
that are possible in modern high-pressure steam generation and the 
economic performance of automatically-controlled oil-burning fur- 
naces. 

The only adverse comment which can be made of the excellent 
results obtained by Mr. Wejnnouth is that the tests were conducted 
under abnormal conditions wherein engineers, expert in this par- 
ticular field, were employed so that possibly under ordinary operat- 
ing conditions these efficiencies could not be obtained. The fact 
remains, however, that a high goal of accomplishment has been 
established which shows clearly tluit the engineer, properly trained, 
can increiise efficiencies far beyond those now prevailing in our 
power plants. It would also indicate? that our power plant managers 
should give more attention to this important subject and see that 
men who have in cliarge the oixjration of the plant are qualified to 
develop the highest efficiency attainable. 

Mr. We\in()uth*s paper brings to our attention again the ad- 
visiibility of the estabhshment by the Society through its Power 
Test Codes Committee of sUmdard test codes for oil-fired boilers. 
While it is true that the present boiler test code may hi) adapted 



DISCUSSION 455 

to this purpose, many definite rulings should be made by the Com- 
mittee. The subject of definition of eflSciency, for instance, is one 
of misunderstanding where oil is used as a fuel, due to the fact that 
a certain amount of the steam generated is used for atomizing 
purposes. 

W. L. Du Moulin (written). Besults of tests on power plant 
equipment arc always of interest. They show the performance of 
the equipment under the most favorable conditions. Information 
of this nature is particularly of interest from an engineering stand- 
point, and desirable to have. However, test conditions are usually 
special, and consequently, actual average operating results over a 
period of time are of more practical benefit. The data given of the 
economy of the power plant of the New Cornelia Copper Company 
are an indication of what may be attained under operating condi- 
tions by a plant subject to the extreme summer heat of Southwest- 
ern Arizona, and located where circulating water for condensing 
purposes is a serious problem. The operating records of this plant 
demonstrate that economies can be obtained by a well-designed 
plant operating under favorable conditions approaching those ob- 
tained under test conditions. 

The operating conditions of the New Cornelia plant are excep- 
tionally favorable, both because of the high power factor and the 
relatively high and steady load factor. About 72 per cent of our 
total power generated is converted from alternating to direct current 
by four motor-generator sets with synchronous motors. These 
motor-generator sets supply the direct current required by the electro- 
l^'tic tankhouse, and their load is, therefore, uniform. 

During the last eleven days of January 1919, the average net 
economy in kw-hr. per bbl. was 323.1; average load 8085 kw.; aver- 
age vacuum 1.28 in. absolute. January is the most favorable month 
in the year, due to the better vacuum obtainable on account of the 
cooler condenser circulating water, and July is about the most un- 
favorable month, because of the poorer vacuum. The average net 
economy in kw-hr. per bbl. for July 1919, was 302.4; average load 
7587 kw.; average vacuum 2.08 in. absolute. On January 26 and 27, 
1919, the net economy in kw-hr. per bbl. was 327.3 and 326.5; aver- 
age load 8091 and 8012 kw. respectively. While the records of such 
individual isolated days are interesting, yet the average records 
covering longer periods of time are the ones that will be of more 
practical value, as the influence of the personal equation is reduced 






c.owir.z v.-^i..-::*^ :: \ll^ ^.-^zzi.-z.'. \irrr:rT>^i stnc-'^.y wiiti tr^e 

rrnii'V-ri ::!- 1 iiL-'^i"--: ir. *:.-- r-ii^ining r-iilrrf >:- is :o nrmove the 
s*>:-: ::.::: *:.- f':>;:r.-u*:rr-: t%-'::::':i r^u'tr-i in incr^SLang the suwr- 
hri: i'- :: l'- i-cr*-. I" i- ''z.^ ttj:--: i ■:rirJvn thai mechanics! 
s-*-* ? 1 -.v. > rr r*r.v ir.-':-/.!--': ::. .i" *• il-r^ w.:-il*i material! v ai.i 

• • • « 

ir. ::.:!::.• ?.i:.:r.i: *h- "'— * - r. rr.y. Ir. Vi'-r^ : th^ size c»Mitaineti 
ir. 'hi- r-*..-.:-*. *h-- :- r. vi:.;^ .-: ^-.- '-~-;*ivv!y fr* n: Sailer and sup^^r- 
h«=':i*'=^: r:-- :-.:.:.• "* :-. ::.: ->h-i ^:h i hini Lai:«7e. The writer 
• i»^:rv> :■-".- *  ::.: L:-lz'- *?.- vi/iv o: o>-rj:ir.g the ec*>nociiier tube 
5."nir- :- ::.  :.:.••■•! :. •.^:*:. 'jil-ir*.--: >:i^r? in or».ier to maintain 
grH>i ... :.■ !:.y. A!--. •!.•: ::.-r*ci'.I:r.:or. ' : :V-:r:w;»>?r regulators will 
\ovA t" :i r-- i'l ■*: r. ::. orKrrri •::.£; 'iiff:-?*:l:i-:<. 

Ti.» :i:r«::::i*:; tirir.g r^ff-iia^^'r ^-.i? *>--n :ir. appreciable factor 
in m:ii:.*:ii::ir;2 "•:: f;C'»r.or:iKs. A ?r^-:ir.. r^r^-jk^Mre has been main- 
tainovi with :i vuriatior; of nor ov<^r 3 !': . ;.r- i tho amount of steam 
roqiiirt.d i«»r :ir"i:Jzine thu- f-K-1 oil v-ry ■■'■.<■> r>"£riLittHi. 

Wl.il.^ ihr rx'Vi.-ii.i-illy :";:'.' r:i- '• « TV n-.T.i.i: conditions have 
a vc n* i« • ! 1 s i - i». r:t M» • '■*■:•. ri r. i: f •: . • : . - t:- -. - i t • •  r . • •: » i y \ %^ in e obtained , 
vot tli»' rr»ii*r r«.«.':ii i»l:iv- ;; ::..-: iii.T^r::-.:;* iv-.r. t^wciallv in as 
i<ol:iTo«i ;i v!:;'V- ri< Aj". wi'A \i\ ;i ■•■.::.* r>- whon.* fu»l i< ix>>tly, as large 
Uv<sr< ::■;:>• •" *:: '\vr^ . T--- ::.•!• i. «:ir>^ .Mi.i-.iit U^ exercised to 
do:i'r:vi:-.v '":.■ vt"> ' :•■•::::>.' ••• t'-T:-Mi-]i in ».»r-it^r Ti^ «^btain the best 
:\'^-.:'> v.v. i :* 'l.- '':>T:;- iiii: i--.»:;'ii:iMn-. Tht- fMiia-r room requires 
v*':>: .*:: ."':.::•:. :v:;i iv»>o :ipi>li«':itinn to ilrtail to maintain good 
ii .^v.oTr.y. 

1\ ".. \\::.u :- *!« :ivtr:i.i:e Ih^Uit pmhh ilata onwrinir a rrprf^senta- 



DISCUSSION 457 

Temperature of feedwater entering economizers, deg. fahr 175 

Ave. temperature of feedwater leaving economizers, deg. fahr 234 

Temperature of superheated steam, deg. fahr 508 

Degrees of superheat, fahr 100 

Temperature of oil to burners, deg. fahr 190 

Steam pressure, lb. gage 255 

Pressure in oil line to burners, lb. gage 10 to 15 

Pressure in steam line to burners, lb. gage 30 to 45 

Lb. of water actually evaporated per lb. of oil 13. 7 

Factor of evaporation 1 . 097 

Water evaporated from and a't 212 deg. fahr. per lb. oil 15. 02 

Boiler horsepower developed 3500 

Per cent of rated capacity developed 106§ 

Boiler eflficiency, Gross, per cent 78. 4 

Net (deducting steam to burners) per cent 77. 6 

Boiler and economizer efficiency, per cent 82. 8 

By consistent work, the boiler room performances as indicated 
may be improved. 

All economies given in Mr. Weymouth's paper and in this dis- 
cussion are net. 

D. S. Jacobus (written). Information respecting plant effi- 
ciencies such as that given in the author's paper and by those who 
have discussed it is of great value in placing on record the progress 
in the art of power generation. Saving of fuel was brought promi- 
nently before the country during the war and is still prominently be- 
fore the pubUc in view of the increased cost of fuel and the general 
movement that has been organized to avoid excess waste. Higher 
and higher boiler-room and power-plant efficiencies are being ob- 
tained and reliable tests are most useful in a study of the problem and 
in inspiring those in charge of power plants to secure the best results. 

Mr. Sibley states that the only adverse criticism that can be 
made is the fact that the author's test was, perhaps, conducted under 
abnormal conditions, in that engineer experts were employed and 
that under ordinary operating conditions the efficiencies which the 
author gives could not be attained. The tests given in the paper 
bear on the results secured in the boiler room and not in the plant 
as a whole. All of the plant economies given in the paper are operat- 
ing figures and Mr. Sibley's statement therefore appUes only in 
part to the figures given in the paper. The general idea involved in 
Mr. Sibley's statement is certainly correct for most cases, but there 
is no reason why there should be any difference between test and 
operating results with the very best operation. In some plants every 
detail is attended to in the same way as in a continuous test — in 



458 



ARIZONA OIL-FIRED POWER PLANTS 



fact the plant is supervised by high-class experts and subjected to a 
continuous test; and there are instances where the operating results 
are as high as any test results. More and more attention is being 
given to securing engineers of the right type for supervising the opera- 
tion of the boiler and engine room with the result that higher and 
higher efficiencies are being obtained. It is a mistake to assume that 
as high a type of man is not required in the boiler room as in the 
engine room, for as I have said on many occasions, there is more to 
\)G lost or gtiined through the operation in the toiler room than in 
any other part of the plant. 

The results secured at the Inspiration plant, taking into account 
the conditions outlined in Mr. Jourdin's discussion, show what can 
be done with high-class attendance and operation. The same ap- 
plies to the figures for the New Cornelia Copper Company plant 
which are amplified in Mr. Du Moulin's discussion. To obtain high 
efficiencies of the sort there must be a proper equipment, but aside 
from this there must be the highest class of operation. 

It is interesting to compare the efficiencies secured in the oil- 
buniing plants with those that have been obtained with coal fuel. 
The Ix^st maintained economy for the Inspiration plant is given :us 
20,910 H.t.u. i)er kw-hr. and the best performance of the New Cor- 
nelia plant is about 18,000 B.t.u. per kw-hr. Mr. Alex Dow in his 
discussion of a pajx^r presented by Mr. Richard H. Rice on Recent 
Installations of Ijirge l\irbo-Generators at the New York meeting 
of th(^ American Iron and Steel Institute, 1917, gave the following 
flguix\s for the results secured from the Comiers Creek Power House 
of the I )etroit Edison Company. 



Kw-hr. out put 

MHxiiiiuiii ilrinuud ('M niinutt*fi) 

Avi*IU|,(l- l(iH(l 

I.oimI fnrtor 

( 'tiiil |Mr kw-lir.. Ill 

It t li. (H-r kw hr 



12 mouths 

ending 

June 30. 1016 



12r>, 158.800 

35,000 

14.300 

.400 

1.44 

19,700 



12 month* 

ending 

Dec. 31, 1016 



162.117.600 

36.000 

18,500 

.514 

1.45 

10.800 



8 montha 
NUi«h31.1017 



54.654.000 

46.000 

25,800 

.562 

1.56 

20300 



2(),(KM)-kw. turbines are employed at Mr. Dow's plant. 
Mr. Dow states that the difference between the July to June, 
12 months, and the January to December, 12 months, was due to 



V 



DISCUSSION 459 

disturbances of coal supplies and that the same cause, together with 
the increased use of heat in the buildings during the winter months 
affected the three months period given in the final column. 

Mr. Dow estimated what might be done in securing efficiency 
through making certain changes in the plant, his idea in this connec- 
tion being as follows: 

Were we designing today for fuel at $5.00 — which seems to be the proba- 
bility — we would buy turbines of still higher rotative speed, which would 
require about 9 per cent less steam than our Conners Creek turbines, and which 
would be nearly as reUable; we would install economizers, for which we have 
room, but which we have not heretofore thought desirable, and thereby bring 
our maintained boiler-room efficiency up from 76 per cent to, say, 81 per cent; 
and we would make certain other refinements in our heat balance which might 
save 1 per cent of oiu' total fuel. The result of these changes would be a reduc- 
tion from a normal use of 19,700 heat units per kw-hr. to something like 17,000 
per kw-hr. of net output. 

The high efficiencies secured in Mr. Dow's plant are an instance 
where the operating results are as high as the test results, as the 
plant is fitted with all the apparatus necessary for making a con- 
tinuous test, and a continuous test is actually conducted with the 
plant operated with the highest degree of intelligence. 

Regarding the limit which will be reached in the economy of 
large power plants, the writer recently made the following statement: 

A large steam-turbine plant of the best modem design can be built to 
generate a kw-hr. with a heat consumption based on the heat in the fuel of 17,000 
B.t.u. per kw-hr. This is a round figure for plants of the best modem construc- 
tion throughout with a load factor of, say, 60 per cent and steam pressure of 300 
lb. per sq. in. By increasing the steam pressure and raising the superheat the 
figures could no doubt be reduced to the neighborhood of 15,000 B.t.u. per kw-hr. 

To secure these higher efficiencies we must install plants in- 
voKing more complication than the older plants. The day is fast 
passing where simplicity is considered of first importance irrespec- 
tive of the economy. Where modem power plants have been in- 
stalled with the more complicated apparatus it has been found that 
with the right sort of instruments the same class of men that operated 
the more simple plants, after proper training, effectively operated 
the more compUcated plants and experience has demonstrated that 
the sa\'ing in fuel far offsets the increased cost of investment and the 
cost of providing better expert supervision. 

The Author. The straight line diagram representing barrels 
of oil and kilowatt output, submitted by Mr. Delany, is very interest- 



460 ARIZONA OIL-FIRED POWER PLANTS 

ing, and the writer has used this extensively when investigating the 
economy of single-unit plants at fractional loads.* It is a fact that 
a line of the character shown is very nearly a straight line through- 
out its rated capacity for a single-unit turbine plant, but when two 
or more units are in operation at the peak load, and ^ith a reduc- 
tion in the number of units in operation at the hghter loads, that is 
turbines, boilers and auxiliaries, the line becomes a jagged line, in- 
stead of a straight line, and the line which Mr. Delany has shown 
for the Inspiration plant is, therefore, approximate only. Further, 
as the Arizona Power and New Cornelia Co.'s plants have different 
characteristics, and different rates of economy, it is not accurate to 
draw one line as Mr. Delany has done, to represent the performance 
of the two stations. 

Tlie comparison of lines is Ukely to lead to serious errors. For 
instance, the highest load recorded for the Sacramento plant is 
145,000 kw-hr. i)er day, corresi)onding to which the average load 
was 6041 kw. and the economy 238 kw-hr. per barrel of oil, yet Mr. 
Delany has extended this as a straight line corresponding to average 
loads of about 8000 kw. with a higher apparent economy. This 
overload, on a 5000-kw. unit, with boilers and connected auxiliaries, 
if such a load were possible, would result in an upward turn of the 
hne beyond the economical rating of the plant. 

ITic graphic method, when plotted with respect to the total 
load, gives a distorted comparison. For example, a plant of best 
modem design, having a 30,000-kw. unit, operating with 300 lb. 
pressure boilers, 150 deg. superheat, and tidewater vacuum, would 
give, at full load, a test economy of 309 kw. ixir barrel of oil, or 
16,772 B.t.u. per kw-hr. If the characteristic line for this plant 
had the same ratio of deiul-load to live-load fuel consumption as the 
line for the /Vrizona Power Co., as plotted by Mr. I>lany, it would 
show an economy, at 5000-kw. load, of al)out 245 kw-hr. per barrel 
of oil, and when plotted on the same sheet with line ''B*' shown by 
Mr. Delany, would indicate al)out the same economy at the nOOO-kw. 
lojul; and yet, comparing the economies at rate<l loads, there is a 
^rc'Mt Mip<'riority for the larger plant. The comparison of zero-load 
eeonoiiiy is iriMccurMte since the Ins|)iration plant, with two units in 
ojM'ration jil zrro load, would show practically double the fuel con- 
sumption for a siiiLcle unit. 

Although a sin;:;le line gives an approximate idea of the per- 
fc»nn:uue, ;it various loads, of a single-unit tidewater plant, a family 
of lines is neeessjiiy to repivsent tlie jXirfonnance of a plant such as 



DISCUBSldN 461 

that of the Inspiration Copper Co., having a cooling pond for con- 
densing purposes, due to the effect, on the economy of the plant, of 
the large variation in vacuum for winter and summer conditions. 
While it is necessary to point out these limitations with the grai^c 
method of comparison, we are indebted to Mr. Delany for pointing 
out the usefulness of this method in diecking up tiie day4o-day 
economy of a given plant, at various loads. 

Referring to Mr. Jourdin's discussion: The boiler efficiencies 
given in Figg. 1 and 2 were the result of a series ci uniform load tests 
of boilers. Mr. Jourdin does not state whether these results have 
been improved in subsequent operation, and no doubt he means that 
in variable load operation the efficiencies are necessarily somewhat 
different than imder constant load test conditicms. It wiU be noted 
that Fig. 3 gives the combined boiler and economizer efficiency, with 
atomizing steam deducted, taken from the power plant logs, this 
data being supplied to the writer by Mr. Jourdin, and showing a 
very favorable performance. Referring to the performance of 
economizers: This is relatively poorer than in other stations oper- 
ated with equal bofler efficiency, partly because of the less econo- 
mizer surface per boiler horsepower; this is diown by a comparison 
of results at the Inspiration and Arizona Power Co.'s plants, and 
the writer is unable to agree with Mr. Jourdin's explanation of 
this detail. 

Mr. Ennis states that maintenance of hi^ boiler efficiencies is 
to be expected in regular operation with weD designed oil-burning 
plants, but he apparently is not aware that many well designed 
plants are not showing as high overall efficiencies as those given in 
the writer's paper. The operating efficiencies for the New Cornelia 
plant were based on one unit only, and full information is given to 
determine the load factor. The curve of boiler efficiency against 
rating is a normal curve for an oil-fired steel-incased boiler. If the 
writer understands it correctly, the inference from Mr. Ennis' dis- 
cussion is that if these oil-burning plants had been burning coal, a 
considerably poorer economy would have resulted. It is regrettable 
that no economies have been given by Mr. Ennis for record per- 
formances of modem coal-burning plants, having units of the size 
given in the writer's paper. Happily the conversion of a number of 
eastern coal-burning plants to oil will soon afford ccnnparative data. 
From the information as to operating boiler efficiencies given for the 
Inspiration plant, and in Mr. Du Moulin's discussion for the New 
Cornelia plant, also in Dr. Jacobus' discusBi<m for coal4mming plants, 



462 ARIZONA OIL-FIRED POWER PLANTS 

it will be seen that the operating efficiency with oil fuel is practically 
the same as in the best fired stoker plants. It may be interesting to 
point out the fact that in oil-burning plants there is a deduction for 
the amount of steam for atomizing the oil, for heating, and for 
pumping oil. While it is true that good firing with oil fuel permits 
a very small percentage of excess air over chemical requirements, 
the large hydrogen content of the oil decreases the actual efficiency 
of the boiler. Were it possible to bum Uquid carbon with the same 
minimum excess air as with oil fuel, boiler efficiencies would be 
obtained of nearly 90 per cent; and the difference between this 
figure and the actual efficiency is mainly due to the large hydrogen 
content, and the formation of steam from the combustion of hydro- 
gen, with its large latent heat, superheated to the temperature of the 
escaping gases. In coal fired plants, having economizers, there is 
much greater heat recovery in the economizers owing to the greater 
weight of gases per boiler horsepower, and in turn the higher tem- 
perature of escaping gases. Therefore, it is a fair statement that 
the net combined boiler and economizer efficiency, after deducting 
steam for atomizing, heating, and pumping oil, is but very little 
better in oil-burning plants than that obtained in the best coal- 
burning plants, taking as a basis for coal the figures given in Dr. 
Jacobus' discussion, which the writer understands were obtained 
from actual performances at the plant of the Detroit Edison Co. 

Mr. Sibley, no doubt in error, has referred to the economies 
given in the writer's paper as test economies under the care of ex- 
perts, instead of operating economies. A more careful reading of 
the writer's paper would have indicated to Mr. Sibley that the per- 
formances for the Inspiration plant, given in Fig. 3, are strictly 
operating records, under the care of the regular power plant opera- 
tors, extending over a number of years, and not in the hands of the 
engineers. Also, as the writer stated, the engineers never developed 
the economy of the New Cornelia plant, and the figures given are 
strictly operating results, no complete plant tests ever having been 
made by the engineers. The economy for the Arizona Power Co.'s 
plant is adniittedlj^ a test result, but one which can Ix; practically 
maintained, since it has lx)en proven that in the operation of the 
Inspiration plant, subsequent to the engineers' tests, actual operat- 
ing efiiciencies wore vsecured fully equal to the test results. As has 
l)een i)ointed out by Dr. Jacobus, modem ix)wer plant operation is 
only at its best when each day's run is regarded by the regular 
oix^rators as a test ixM-formance. Wiile engineers have frequently 



V 



DISCUSSION 463 

excused the rather ordinary performances of then* stations by claim- 
ing that superior results were obtained imder test conditions, the 
writer is quite sure that this is not the spirit of Mr. Sibley's discus- 
sion. The men responsible for the economy of the Inspiration and 
New Cornelia plants are very clever men, but no better than the 
type of men that should be employed in every modem steam-electric 
power station. Mr. Sibley's statement of economies of other plants 
on the Pacific Coast is hardly fair, since many of these operate in 
conjunction with hydro-electric plants and are not on as favorable 
a load factor basis as the Inspiration and New Cornelia plants. 
While the writer does not know of any performances of other Pacific 
Coast plants equal to those given in his paper, there are several 
plants which have shown very high efficiency, higher than indicated 
by Mr. Sibley. The writer is pleased to endorse Mr. Sibley's recom- 
mendation that the Power Test Code Committee give especial at- 
tention to codes for oil fired boilers. 

The figures given by Dr. Jacobus afford an interesting compari- 
son of the efficiencies of coal-burning and oil-burning plants, but al- 
lowance should be made for the fact that the plants referred to in 
this paper are those having much smaller imits than the coal-burning 
plants quoted by Dr. Jacobus. Dr. Jacobus' statements as to the 
ultimate economies now possible for large plants are of great interest 
to engineers. 

When considering the boiler efficiencies given by Mr. Du MouUn 
for the New ComeUa plant, and comparing these with the efficiencies 
given by Mr. Jourdin for the Inspiration plant, due allowance should 
be made for the fact that in the New Cornelia plant there is a con- 
siderable heat waste due to the necessary, but abnormal amount of 
boiler blow-ofif, and as well the sUghtly higher temperature of escap- 
ing gases, due to the higher steam pressure of boilers and higher 
temperature of water within boilers. 



No. 1704 

ELEMENTS OF A GENERAL THEORY OF 
AIRPLANE-WING DESIGN 

Bt Walter C. Durfee, Boston, Mass. 
Member of the Society 

This paper presents in brief outline form ten suJbjects which have reference to 
the theory of fluid motion around the wings of airplanes. These are: the vortex 
theory of lift; the theory of initial motion around wings; vortex theory of shape; 
hydrodynamic-electromagnetic analogy; action of vortices with reference to each 
other; action on vortices with reference to their images; influence of the local wind; 
laws of energy content in trailing vortex; friction and head resistance; and explosion 
of eddies. These various subjects are not discussed but are merely brought forward 
for the purpose of providing a starting point for discussion. 

T^HERE are several subjects which seem so interesting in con- 
nection with a study of the action of wings upon the air that 
the writer has thought it valuable to the Society to place them on 
record, in brief outline and in such a way as to provide a starting 
point for discussion and the addition of any data which members of 
the Society may wish to contribute. These subjects which have 
reference to the theory of fluid motions around the wings of air- 
planes are as follows: 

a The vortex theory of lift, which states that the air which 
passes the wing of an airplane, or the blade of a propeller, 
contains a component of circulation around that wing 
or blade, in such a direction that there is a comparatively 
high velocity and low pressure on the upper surface of 
a wing; and a comparatively low velocity and high 
pressure on the under surface. 
b A theory which states that an imperfect fluid will act 
like a perfect one momentarily; from which it may be 
inferred that the circulation around a wing cannot exist 
at the first moment or beginning of its motion of advance 
but must develop at some time after the first beginning 



Presented at the Spring Meeting, Detroit, Mich., June 1919, of The 
Akxbican Societt of Mechanical ENGiNEERa 

465 



466 AIRPLANE-WING DESIGN 

of the motion, since there is no circulation in the begin- 
ning. 

c The vortex theory of shape, which treats of a soUd body 
in motion as being somewhat similar to the core or kernel 
of a group of vortices. 

d The hydrodynamic-electromagnetic analogy, which states 
that distributions of fluid motion are very similar to 
distributions of magnetic flux; so that one may cal- 
culate the fluid motion around a supposed vortex or group 
of vortices mechanically, by arranging electric currents 
or groups of currents in a manner analogous to the sup- 
posed vortices, and measuring the magnetic forces which 
result. 

e The laws of vortex motion with reference to the action 
of vortices on each other, by which it seems possible to 
estimate the circulation or strength of the trailing vortex 
loop which is generated by a wing in flight. 

/ The laws of the actions of vortices with reference to their 
images in solid surfaces combined with the laws, so far 
as known, concerning the generation of eddies and vor- 
tices by friction, especially near sharp edges. 

g The concept of a local direction of the wind as due to the 
eflfects of all vortices existing in the neighborhood of a 
wing — such as its own trailing vortices and the influ- 
ence of neighboring circulations. 

h Laws concerning the energy contained in various distri- 
butions of vortex motion by which one may estimate 
favorable arrangements of the trailing vortex systems 
in t^nns of the load carried by various parts of the wing 
span, and from which the drag might be estimated. 

I (Coefficients of friction and head resistance representing 
losses of cnerg}' which oixn l)e adde<l to the losses attrib- 
uted to the energy of the tniiling vortices. 

j KxixTience concerning the explosion of eddies and vor- 
tices and the causes and effects of such disturbances. 
2 It is the writer's \xA\iA that there arc engineers, mathemati- 
cians and rxixM-inicntcrs in the Society who can give illuminating 
and interesting statt^ments concerning the subjects mentioned; and 
that a jiroup of such statements assembled in the fonn of a discus- 
sion would constitute almost a complete and classical theory of the 
action of wings in st^\idy flight. This i^iijxjr, therefore, outlines in 



WALTER C. DURFEB 467 

a preliminary way the bearing of these various theories and indi- 
cates their approximate exactitude. 



a THE VORTEX THEORY OF LIFT 

3 It is not difficult to believe that a component of circulation 
exists around a wing in flight. If it is granted that the wing carries 
any load at all, as wings evidently do, there is certainly a differ- 
ence of air pressure between the lower and the upper surface. Con- 
sequently there are accelerations in the neighboring fluid from the 
under surface around in various circuits to the top surface; corre- 
sponding to the fall in pressure from one surface to the other. The 
quiet or still air into which a wing advances, experiencing these 
accelerations, must accumulate an upward velocity in front of the 
wing, and disturbances of a similar nature evidently must occur not 
only in front of the wing but also to the right hand and to the left 
hand. Since an upward motion in one region involves a downward 
motion in another, there must be a downward motion somewhere. 
Actually after the beginning of the flight there is a sort of circulation 
up in front and down behind; and consequently to the rear above 
and in a forward direction below. 

4 In practice such motions can sometimes be seen in the form 
of little jerks or jumps of a fluid in the neighborhood of a model wing 
passing through it. It is not difficult to believe that this disturb- 
ance around a wing is rather similar in arrangement to the distribu- 
tion of velocity around a vortex or group of vortices — their axes 
parallel to the span of the wing, and perpendicular to the direction 
of advance. The intensity of motion is very likely greatest near the 
seat of the disturbance. 

5 According to mathematical theory the lifting force would be 
in proportion to the strength of the vortex and to its rate of advance, 
just as the lifting force on a wire in the armature of an electric motor 
is in proportion to the strength of the current and to the intensity 
of the magnetic flux from the pole pieces. A formula is given in the 
Encyclopedia Britannica for the theoretical action of a vortex which 
surrounds a circular rod which is projected sideways. A force de- 
velops perpendicular to the axis of the rod and vortex and at right 
angles to the motion of advance. This is very similar to the lift of 
a wing in flight. 

6 Practical examples, however, suitable for mathematical an- 
alysis seem to be very rare, nevertheless the writer found one case 



468 AIRPLANE-WING DESIGN 

which seemed to be reasonably free from objectionable complica- 
tions. This was a wing tested by Eiffel (Eiffel No. 8, at 9 deg. center 
section). From the measured pressures in this case the probable 
approximate velocities of the air near the wing surface were esti- 
mated, using Bernoulli's theorem. From these velocities a calcula- 
tion was made of the circulation around the wing, which is the line 
integral of the tangential component of the velocity vector in a circuit 
around the wing. The result agrees with the theoretical formula for 
lift, or L = p Vml, in which L = force perpendicular to advance; 
p = density of air in absolute units; V = velocity of advance; m = 
circulation of the vortex; and I = mean span of sustaining vortex. 
It would be interesting to have more measurements of the circulation 
around wings. 




Fig. 1 Diagram to Illustrate Theory of Initial Motion 

Around Wings 

b THEORY OF INITIAL MOTION AROUND WINGS 

7 It is very evident that no circulation exists around a wing 
when it is standing still in quiet air on the ground. Mathematical 
theory further declares that circulation cannot be expected to de- 
velop immediately at the beginning of the advance. In the first 
instant of motion conditions are supposed to be very much as they 

C C Axfs of Trailhg ^^irfktt A A' 





Axfs of Trailing Vbrf fees. ^ Ajtkof 

otStorf. 

Fir.. *J Diagram further Illustrating the Theory or Initial 

Motion Around Wings 

would l)(' in a (MMfrut fluid. The amount of circulation if lero at 
the start would he zero inuncdiately aften\'ard. For example, 
imagine, for simplicity's sake, an inclined plane moving from the 
|M)sition A A' to BB' in Fig. 1, starting suddenly from rest. The 
volume (Miuivalent to the space AA'-BB' can be expected to be dis- 
placed and to go around the edges in such a manner that there is 
no n(^t circulation around the section. Verj' soon after this b^;iii* 
niiig of motion, however, in the case of a real wing suddenly started 



WALTER C. DURFEE 469 

in a real fluid, a very violent eddy or vortex is left behind at A'B'. 
When the wing has advanced to a further position CC the condi- 
tions are as sketched in perspective in Fig. 2. There is a vortex 
loop stretching rearward from near the wing tips and joined to- 
gether by the eddy generated at the starting point. This vortex 
circuit is completed by the sustaining vortex which circulates 
around the instantaneous position of the wing. 

C VORTEX THEORY OF SHAPE 

8 Suppose that the axes of a number of equal-sized vortices 
are arranged as the circles in Fig. 3. Then the direction of motion 
in the fluid due to their combined action is almost exactly^ in the 
directions indicated by the full-line arrows. Suppose that a cer- 
tain motion of translations is added to this particular arrangement 
of vortex motion. Then the resultant velocity of total result may 




— >. 



Fir,. 3 Diagram to Illustrate Hydrodynamic-Electromagnetic Theory 

be in the directions indicated by the dotted arrows. This particu- 
l(ir kind of fluid motion corresponds to a certain velocity of motion 
added to a certain arrangement of vortices. 

9 Now it is a fact that the shape of the curved hne of circles 
shown in the diagram is, as nearly as may be judged, the effective 
shape of a cross-section of a certain real wing (Eiffel No. 8 at 9 deg.), 
deducting an allowance for the thickness of the section. Also it is 
a fact that the spacing of the circles indicates the actual distribu- 
tion of lifting force experienced by that wing between the front 
and the rear of the wing section. Also the component of horizontal 
velocity added to produce the dotted arrows is the velocity used in 

* "Almost exactly," because the diagram contains an almost impercepti- 
ble allowance for the termination of the sustaining vortices within the length 
of the wing span and for the influence of their rearward extensions as trailing 
vortices. (Compare heading g.) 



470 AIBPLANE-WING DESIGN 

the published tests of this real wing, and the vortex strength used in 
preparing the diagram agrees with the vortex theory of lift (Par. ^) 
and the estimated circulation around the wing mentioned. It is 
evident from these figures that there is a close connection between 
the factors of distribution of fluid motion, and shape. 

d HYDRODYNAMIC-ELECTROMAGNETIC ANALOGY 

10 Diagrams like Fig. 3 are easily obtained, not by mathe- 
matical calculation, but by arrangements of electric currents and 
magnetic fields representing vortices and velocities, choosing any 
desired amount of flux to represent a standard velocity. 

e ACTION OF VORTICES WITH REFERENCE TO EACH OTHER 

11 The Encyclopedia Britannica gives formulae for the action 
of groups of parallel colunmar vortices upon each other, in terms 
of the strength of these vortices and their distance apart. It is inter- 
esting to estimate the strength of the trailing vortices from a bi- 
plane by observing their actions on each other. The pairs from the 
right-hand wing tips arc of one kind or direction and revolve around 
each other in approximate circles. Those from th^ left-hand tips 
also revolve around each other, but in the reverse direction from 
the first-mentioned pair. Very careful experiment would be re- 
quired to detect any error in the vortex theory of flight in terms of 
the action of these pairs of trailing vortices. The vortices can be 
seen in a smoky atmosphere when moving models are used. 

/ ACTION ox VORTICES WITH REFERENCE TO THEIR IMAGES 

12 Many peculiarities of fluid motion are roughly explainable 
in terms of the action of eddies as if under the influence of their 
images in solid surfaces. For example, there is a remarkable dif- 
ference in the circumstances surrounding the eddies formed at B and 
li' in Fig. 1 on the upper surface of the plane. The one at the rear 
B' should tend to pass off if considen^d as under the influence of its 
image. Conversely, the one at A should tend to remain with the 
plane. 

(J INFLl'ENCE OF THE LOCAL WIND 

13 \'ortices, although regarded as having their axes in some 
particular location, are usually considered as having an influence 
through the fluid in which they exist, just as the magnetic effect of 



WALTER C. DUBTEB 471 

an dectiic enrrrait is considered as having an effect at remote dis- 
tances. It is int^esting to calculate the effect of the trailing 
vortices on a wing of short span. A wing in horiiontal flight does 
not act as if encountering a horiiontal onrush of the atmosphere. 
It acts as if ih&e were a downward component of motion in the 
air around tiie wing, of very much the same amount that mig^t be 
calculated from the strength of its own trailing vortices. This 
is manifest espeaaSy in the case of short wings by a correspondingly 
poor lift, as if scmiething were reducing the an^e of attadc, and in 
a greater resistance as if climbing up through a descending wind. 
The hydrodynamic-electromagnetic analogy is capable of yielding 
interesting information in this connection. 

h LAWS OF ENERGY CONTENT IN TRAILINQ VOBTEX 

14 Mathematically it would appear to be as easy to calculate 
the energy of vortex-motion lift in the wake of an airplane as it is 
to calculate electrical self-4nductanoe. The arrangement of trailing 
vortices behind an airplane evidently depends considerably on the 
distribution of the loading along the wing span^ because a wing can 
terminate in effect considerably short of the actual tip by an easing 
up of the lift. Calculations concerning the best arrangements would 
be interesting. The writer has made some approximate computa- 
tion by assuming the trailing vortices to be a group of parallel colum- 
nar vortices: a sort of sheet of vortices constituting the wake of the 
wing. This method of calculation gives the usui^l values of the 
lift-drag ratio, when friction is taken into account. 

t FRICTION AND HEAD RESISTANCE 

15 In a practical way friction is a large item and it would be 
interesting to have separate tests for the friction losses of wingps. 
Tests might be made of the resistance of a hoop or endless ring 
having the cross-section of a good wing. 

j EXPLOSION OF EDDIES 

16 Frequently the low pressure at the center of a vortex or 
eddy in real air appears to be penetrated by a rush of air along the 
axis. Knowledge about this, especially with reference to the effect, 
cause and control of such disturbances in the wake of wings, would 
be interesting. 



472 AIRPLANE WING DESIGN 



DISCUSSION 

John R. Freeman (written). The writer is hardly competent 
to discuss Mr. Durfee's paper in the language and symbols of mathe- 
matics, but a possible line of investigation of these phenomena 
occurs to him which may perhaps be useful. 

About ten years ago when members of our Society were guests 
at a meeting in England of the Institute of Mechanical Engineers, 
the eminent mathematician, Dr. Hele-Shaw, presented an illas- 
trated discussion on stream-line problems in air currents relating to 
airplanes, in which he showed the disturbing effect upon the stream- 
line of models designed to represent various forms of wings. The 
fluid in that case was a Uquid and the stream lines were repre- 
sented by ingeniously colored liquid filaments. The investigation 
was along hnes of previous investigations of the distribution of 
velocities in flowing Uquids containing colored liquid threads which 
showed the eddy currents caused by such obstructions as bridge 
piers. 

At that time it seemed to the writer that the difference in com- 
pressibility of air and water, and also the difference in inertia effect, 
impaired the analogy, and on further thought he laid out a line of 
experiments for tracing the motion of liquids around obstructions 
in channels by a combination of methods borrowed from the ultra- 
microscope and the moving-picture machine, although he has never 
found time to carry out the experiments. This method seems 
admirably adapted to experiments on air currents in connection 
with airplanes. 

The method in brief is to make an optical cross-section in any 
desired plane by means of a thin, broad beam of intense light put 
into appropriate shape and parallel rays by proper condensing 
lenses, and an optical slit, analogous to that used with the ultra- 
mi(Toscoi>e. By means of dust particles of proper density intro- 
(hurd in the air current, one can render visible the direction and 
velocity (»f the currents set up somewhat as he sees the air currents 
in his liviiiji: rooms made eviilent by a sunlx^ain acting upon the 
susiKMHJed (lust particles. 

Tiuj narrow slit of lij^ht reveals only the motions in one plane 
and simplifies the observation by rendering the particles visible 
only while traveling in this optical plane. 

In the case of the airplane, these motions would mostly be too 



DISCUSSION 473 

rapid for the eye to follow, but within limits it is possible to observe 
them and to record them by a motion-picture apparatus which can 
be so constructed as to make 30 or 50 exposures per second instead 
of the customary 16. It is indeed possible to obtain exposures of 
much shorter frequency by means analogous to the shutter-testing 
device developed in Dr. Mees' Research Laboratory at Kodak 
Park, in which a rapidly revolving polygon of mirrors serves to 
catch and record the fleeting image. Also there have been devised 
and patent^ed means of revolving polygonal prisms of glass, so ad- 
justed that their reaction holds the image approximately stationary 
on the screen or film for the fraction of a second. 

Such a series of photographs recorded in a short reel of film 
can be rotated for purposes of study at a much slower speed than 
that at which they were taken. 

By these means, the writer believes the actual pathway of 
the particles of air as they pass either wing plane or propeller can 
be made evident and precisely recorded at velocities far higher 
than can be observed in any other known way, and a series of optical 
sections, analogous to those of the ultra-microscope or the sun- 
beam will simplify the hopeless complexity of a dense mass of par- 
otides traveling in various directions. 

Edward P. Warren (written). It is manifest that any con- 
siderable development of the theory of wing action beyond the 
point already reached must be conditional on the use of new and 
more powerful and logical methods of attack. In most of the work 
so far done, whether by the simple assumption of plane impact and 
reflection or by such more elaborate methods as that of Kirchhoff 
and Helmholtz, the continuity of the air has been ignored, and the 
results have consequently been far from the truth. 

The work of Lanchester, Kutta, and others on a vortex theory 
of sustentation seems to offer the most promising path to an analy- 
sis of wing action which shall be of real practical use. It leads to 
the only method which takes due account of the fact that there can 
be no actual acquisition of downward momentum by the air as a 
whole, since the center of gravity of the atmosphere cannot shift, 
and any downward motion imparted to the air in the neighborhood 
of the wing must be counterbalanced by an equal upward motion 
imparted to an equal mass at some other point. 

Promising as the vortex theory is, however, it should not be 
overrated. There are many factors in the action of wings for which 



474 AIRPLANE WING DESIGN 

it docs not appear to account, and the mathematical weapons are 
not at hand for applying it, except in the simplest cases. The elec- 
tromagnetic analogy proposed by Mr. Durfee is very interesting, 
but it must be handled with care, particularly in connection with 
thick wings, where the air-flow changes from stream-hne to turbu- 
lent types and back again with the greatest suddenness and in 
response to the minutest alterations of wing form or conditions of 
operation. It is doubtful if this analogy could be extended to any 
cases beyond those of the flat plate and the simplest forms of thin, 
cambered sections. 

George de Bothezat^ (written). The statement "a" of 
Mr. Durfee^s paper constitutes in reality the well-known "Kutta 
theorem," discussed by Kutta himself (Illustrirte AeronaiUische 
Mitteilungenj 1902; Sitzungsberichte der Koniglichen Bayerischen 
Akadcmie der Winsenschaftcn, 1910 and 1911); by Joukowski 
{Ahodynamiquey Paris, 1916) and Dr. de Bothezat (Report No. 28, 
Note I, from Fourth Annual lleport. National Advisory Committee 
for Aeronautics). 

The statement *'c" was first made by Ix)rd Kelvin with refer- 
ence to an example actually classical (the so-called atmosphere 
around a system of two rectilinear and parallel vortices rotating in 
inverse sense). 

The statement *'d" of the hydrodynamical-electromagnetic 
analogy is well known. But the suggestion to study the flow around 
an airfoil by this method is of interest, and such experiments con- 
ducted in a suitable manner could bring valuable results. 

A solution of the question proposed in statement "e" is directly 
obtained by the successive application of the Kutta theorem, Ix>rd 
Kelvin theorem on the constancy of circulation and the Stokes 
tlKH)rem connecting circulation with vortex intensity. (Sec Report 
No. 2S of the Fourth Aiuuial.) 

The statement **f/*' is not (juite clear; if local wind means only 
tiie instantaneous value of the thiid velocity at a given point around 
the airfoil, it is only a regular conception. 

The statements **/<*' and "T' demand very careful considera- 
tion, because it seems that in the case of hydrodynamieal phenomena 
some spci'ial conditions niMv occur which we ilo not meet in electro* 
niai^nclic phcni>nuMia. 

' AonnlyiKuiiiial lAjvrt. N:itii»n:il Advisory iommittoo for Aefonautics, 



DISCUSSION 475 

F. W. Caldwell ^ (written). This paper is very timely, par- 
ticularly in view of the growing tendency among aeronautical engi- 
neers to regard the classical coefficients Kg and K^ as inadequate. 

It has been almost universally the practice to write L = —KySV^ 
and D = -KJSV^ where L is the lift, D the drag, S the area of the sup- 

•7 

porting surface, V the velocity of advance, p the density of the air 

in weight units, g the acceleration due to gravity, hence - the density 

of the air in slugs. 

It is well known as the result of experience that the values of 
Ky and Ks vary somewhat with velocity and also with the size of the 
surface imder consideration. If { represents one of the linear dimen- 
sions of the surface it is assumed that the values of Ky and Kx are 
functions of the product VL This is known as the scaUng effect. 




Fig. 4 Diagrammatic Representation of Flow 

Information on the scaUng efifect is very meagre. In the case 
of propeller sections we have been forced to make use of characteris- 
tics obtained at a speed of 30 miles per hour and apply them to 
conditions where the speed obtained is as great as 600 miles per 
hour. While the use of scaling rules and empirical factors worked 
out in practice have enabled us to produce .very fair results, the 
need for accurate data has been pressing. 

The high-speed wind tunnel operated by the Technical Section 
of the Department of Mihtary Aeronautics is of the venturi type 
and shows an exceptionally uniform flow at all speeds up to about 500 
miles per hour. The flow is produced by means of an especially 
designed 24-blade propeller which produces a suction of about 16 
inches of water at the large end of the venturi. 

An extensive series of experiments has been started on propeller 
airfoils in order to determine the effect of speed on the lift and 
drag coefficients. 

It is desired to call attention to the fact that it is not sufficient 
to understand and be able to calculate the circulation about a hori- 

* Aeronautical Engineer, 330 Eklgewood Ave., Dayton, O. 



47G AIRPLANE WING DESIGN 

zontal axis perpendicular to the direction of motion; there axe also 
very important vortices about axes parallel to the direction of 
motion. These are particularly apparent in the form of tip vortices. 
Their intensity might be estimated similarly to the calculation for 
fore and aft circulation. 

The flight vortices have been visualized for the first time in the 
army wind tunnel and a moving picture of the phenomenon has been 
made. While the moving pictures show up very well on the screen 
it has, unfortunately, been impossible to make them show up well 
in print. 

Fig. 4 shows the type of flow corresponding to normal flight 
conditions. It is supposed to represent a view of the plane taken 
from upstream. 

F. E. Cardullo. As I understand the theory which Mr. 
Durfee is attempting to develop, instead of considering the reac- 
tions due to the acceleration produ(»ed under the action of the 
motion through the air of a plate, which may be straight or curved, 
he considers these reactions from the standpoint of the Bernoulli 
theory, on the basis that differences in velocity of air relative to 
the plate will exist on its two sides, and that in consequence to these 
differences there will be a difference of pressure and a lifting force. 
He points out that there are certain vortex motions at the ends and 
at the trailing edges of the plate. The best method of attack 
would seem to me to be that of testing wing sections either by the 
emission of smoke from a fine orifice or with threads attached to 
needles. This method gives an opportunity for studying the vor- 
tices which represent irregular motion and lost energy. Mr. Durfee 
prefers to attack the "problem from the standpoint of the investiga- 
tion of these stream lines and velocities rather than the development 
of the theory of the reaction on the surface produced by accelera- 
tion. I do not know that there is nmch choice in the mathematics 
of the two methods, but in either case the mathematics is too diffi- 
cult to offer a practical solution. 

The Author. Mr. Warner's remarks with reference to tlie 
sensitive fluctuations l>etween one type of motion and another in- 
dicate the probable cause of much uncertainty and doubt connected 
with aerodynamic science. Although the action of a first-class wing 
is comj)arativcly simple and easy of analysis the action of a poor 
wing is likely to be intricate and diflicult of analysis, even though 
the shape may l)e simple in the last case and refined in the other. 



DISCUSSION 477 

At least three or four types of motion might be named and de- 
scribed. There is first the type of motion which occurs theoretically 
in the case of a perfect fluid and which may be observed, according 
to Lord Kelvin, in any fluid at the instant when the wing or other 
object is first put in motion. Radical modification of this ideal 
system of motion may take place owing to the formation of eddies. 
This may result almost immediately in an alteration of the effective 
shape of the body by means of eddies which are carried along by 
the object in question, perhaps on the under side of a wing, sharply 
curved, and on the upper side of a wing, slightly curved, or in the 
rear of objects bluntly shaped. Either with or without the above 
mentioned modification (type two) , a third type of motion is pro- 
duced in the case of successful wings. This is a circulation aroimd 
the instantaneous position of the wing in such a manner that the 
atmosphere below the surface moves forward (with reference to the 
still air through which the wing is flying) and the atmosphere 
above it moves backward. The creation, existence, and preservation 
of this circulation is intimately connected with the vortices which 
Mr. Caldwell mentions as parallel to the direction of motion. As 
it is evident that such trailing vortices must decay through friction 
and other causes, and that the eddies mentioned under the second 
type of motion must also decay or suffer violent destruction or be 
swept away, it follows that violent fluctuations in the type of air 
flow are likely to occur in an irregular manner. The causes which 
determine which types or type of flow will occur or to what extent 
each of several will occur in combination probably have much to 
do with the scaling effect mentioned by Mr. Caldwell. The scaling 
effect is probably influenced also by the fact that the drag of a 
wing is composed partly of friction and partly of a rearward com- 
ponent in the reaction of the sustaining vortex. The following table 
shows this. 

{absolute unit » 0.5 (units of force per unit density and area at unit velocity) 
French " = 0.0625 (kilograms per square meter at one meter per second) 
English " a 0.(X)255 (pounds per square foot at one mile per hour) 

Corresponding strength of vortex » i (breadth of wing) X (velocity of 

wind) 

Corresponding mean velocity over top of wing with refer- >■ | (velocity of wind) plus (Uiickness 
ence to wing correction) 

Corresponding mean velocity over bottom of wing with — } (velocity of wind) plus (thickness 
reference to wing correction) 

Cf., approximate coefficient of friction in absolute unite — 0.005 (or 0.0025 for each side) 

C/M, aerodynamic drag coefficient or coefficient of drag due 
to vortex of reasonably good distribution and for aspect 
ratio = 6, and C^ = 0.5 -» 0.030 

C , coefficient of drag in absolute units » Cp + Cj^ » 0.035 

Calculated reasonably good lift drag ratio for aq>ect ratio » 
6 at value of C^ = 0.5 ^Cj^-i- {Cf. + C^ - 14.3 



478 AIRPLANE WING DESIGN 

Friction is an important part of the drag especially at low angles 
when the drag due to lift is small. But the coeflScient of friction 
is not a constant at different speeds when it is defined like the 
other coeflBcients as a force divided by area, density and squared 
velocity. For this reason one may expect that high velocities and 
large dimensions will result in better lift-drag ratios. 

The analysis with reference to the direction of the reaction of 
the sustaining vortex is improved by the concept which I have 
spoken of under the heading of the local wind. This concept is not 
exactly with reference to the instantaneous value of the fluid ve- 
locities at a given point but is with reference to conditions which 
surround the wing or blade. For example, in the case of the upper 
wing of a bi-plane there is a circulation of the atmosphere around 
the lower wing in such a direction as to increase slightly the in- 
tensity of the wind which acts against the upper wing. It is the 
apparent velocity and direction corrected by such influence which 
I would speak of as the local wind with reference to this upper plane. 
In this particular case this local wind with reference to the upper 
plane contains also certain components of nearly vertical velocity 
due to the existence of the four vortices which trail behind the four 
tips of the wings. In other words a wing appears to suffer from the 
downward motions due to the existence of the trailing vortices in the 
same sense that tlie tail-plane suffers; but in lesser degree. In this 
manner it is possible to understand the peculiarities of aspect ratio, 
since by lengthening the wing a greater proportion of the span is 
placed in a region remote from the tip vortices so as to be influenced 
less severely. Presumably the blades of a propeller or fan may be 
similarly considered as moving in a local wind of an intensity and 
direction determined by the complicated vortex system which con- 
stitutes the blast. The local wind with reference to the blade then is 
not precisely the velocity at that point because that velocity is in 
part made up of the circulation around the blade itself. By local 
wind I mean tlie conditions which in effect act on that blade. A 
preci.^e dcfhiition of this concept might be to say that the local wind 
is indicated l)y the velocity and direction in which the sustaining 
vurtex would itself move if the wing or bh\de should vanish. 

Mr. Warner's remarks as to the care with which the electro- 
magnetic analog>' must be used are very nmch to the point. The 
experiments must be conducted, as Doctor de Bothezat remarks, 
*'in a suitable manner." The air will not follow the desired and 



DISCUSSION 479 

ideal type of motion if conditions are permitted to exist more favor- 
able to another kind of motion. Confining one's self to arrangements 
of vortices, all of the desired kinds as shown in Fig. 3, it is not 
difficult to postulate oiu- arrangement (by crowding the vortices 
nearer the leading edge) so that the fluid motion will conform in 
theory to the shape of a flat plane, but the resulting diagram is of 
such a character as to suggest the probable formation through 
friction of an intense eddy at the leading edge. This appears to be 
the case in practice; and the effective shape of the wing appears to 
be changed by the presence of this eddy and the electromagnetic 
analogy impaired by the difficulty or impossibility of producing 
such circumstances. On the other hand, by attempting to rearrange 
the electromagnetic model so as to avoid unfavorable conditions one 
soon discovers a series of shapes and typical designs which are 
well kno\Mi to be good. Futhermore, the analogy can furnish useful 
information with reference to the local wind at various points. It 
is not altogether inapplicable to moderately thick shapes because, 
according to Lord Kehan, there is considerable similarity between 
the fluid motion around a solid body and aroimd a group of vortices. 

Mr. CaldwelFs remarks as to the intensity of the tip vortices 
are of interest. These vortices may be observed in a smoky atmos- 
phere very readily by using a model which is driven through the 
air guided by a cable trolley. The intensity, so far as I have been 
able to measure it, is the same as the theoretical intensity of the 
sustaining vortex. The measurement of intensity is made by using 
a small biplane and observing the rotation of the pairs of trailing 
vortices around each other; for example, by observing the rotation 
of these from the right-hand tips. It is interesting to inquire in 
what manner the low pressure is preserved along the axis of these 
trailing vortices. This inquiry leads one to become interested in 
the vortex which joins the trailing vortices together at the rear as 
shown in Fig. 2. Presumably this vortex is renewed from time to 
time so fast as the original system breaks down. 

Mr. Freeman's remarks with reference to a method of observa- 
tions derived from the ultra-microscope and the moving picture 
machine are very attractive. By making the observation in still 
air with a moving model it would seem possible to learn very much 
concerning the formation and decay and internal arrangements of 
the entire vortex loop. 

In answer to a question by Professor Cardullo as to whether or 



ittf) AIRFlJkNE WING DESIGN 

not ihf: mtdhcmuiutHl mcthcKl was the only practical one, the author 
ninUul ififtt he tH;liovf;^l that there were many useful experimental 
uit'ihtxlH. Tlu; prcfWint paper was mainly an analysis of the various 
iih'frM'fifH in thf; action of a gorxl wing, intended to be useful as a 
Kuidf! in irinny lim*H of inventigation. 




No. 1705 

AIR FANS FOR DRIVING ELECTRIC 
GENERATORS ON AIRPLANES 

By Capt. G. Francis Gray*, U. S. A. 
Lt. John W. Reed*, U. S. A. 

AND 

P. N. Elderkin* 
Non-Members 

In this paper the authors briefly describe the method employed by the Radio 
Development Section of the War Department in testing air fans iised for driving the 
electric generators tistiaUy installed on airplanes for radio communication. They 
also discuss at some length the variotis types of air fans and present numerous 
photographs and curves clearly Hhistrating the construction of the fans and their 
operating characteristics. 

The difficulty of the problem lay in designing a fan which would turn at constant 
speed in the air streams of widely varying speed set up by the airplane in flight. 
The various types of fans tested were: Pixedrblade fans of special blade shape; 
fixed-Hade fans with wind brakes. cerUrifugaUy regulated; fixed-blade fans using a 
friction clutch or a friction brake centrifugaUy regulated, and pivotedrblade fans in 
which the pitch is centrifugaUy regulated. 

DURING the war extensive use was made of radio telegraph 
and telephone apparatus on military airplanes, and the prob- 
lem of power supply for such equipment received a great deal of 
at t cation. The possible sources of energy may be listed as follows: 

a Storage batteries or dry batteries 

6 Generators driven from the airplane engine, with or with- 
out floating storage batteries, and supplying the radio 
sets directly or through dynamotors 

c Generators driven by separate gasoline engines 

d Generators driven by air fans or "windmills," placed in 
the air stream outside the airplane fuselage. 

2 From an economical point of view, method b is preferable. 

^ Engineering and Research Division, Radio Development Section, War 
Department, Washington, D. C. 

Presented at the Spring Meeting, Detroit, Mich., June, 1919, of Thb 
American Society of Mechanical Engineebs. 

481 



4S2 AIB FAN!! FOB AIBTLAXE CCXEBATOBc 

It wa% i^rrioii'-lv ':'.ti.«i'i*rr^ biit involved cojperaikai between t 
izntli/jti-. tioniiniiy ofifrnitijig iridfrfjerirleDTly &Dd ns i^jpooti waf 
'i«rUyt;*l. Mf^afi'A'lii]"; fJi»: pra^K* in our Army fol]ow€*i lisai of <:*.ir 
AJIi':=^, pri(i'ri(jally tJi*: Frfrnch. in ihe u.«e of method d. tDOustii^ 
tJift ((''ft'-'^*'''' '"iti^idft thf: iiii7>lane fiLsela^t and driiin^ it with in 
iiir fun. 

3 'I^iL* f/ajxrr n;vi»;wh the work done on ibe development of iir 
fart- f'/f thi-r ■J-rvi'.'*; a« eaiTie<i out Vjv the Radio DeveloproMit Sec- 
tion of tin; HiffjiiiJ f'orjjs. Aft the work was done under the press of 
iciilitary neir'rr-ity it was directe*! entirely by utilitarian coDsider&- 



Fni. 1 Am Vas ash Tkmt (Jknbkatoh Moivted in Wikd Tmnnu. 

lioiiH, wiw often riitKnii'iil ar>-, and ncRlertcd invostigatiooB, the 
need of wliicli WHS realized hut for whirh limn and personnel were 
lint availfiMc. This n;c<ini is pnsteiitMl with the work still in un- 
fiiiislicil form in the lioix; tliat remiltu obtained may be useful to 
lliose w)io may liiive occasion to carr>' out further investigations on 
I lie iirolilcm. 

(iiMitrioNS KOIt WIIKII TUK AIK FANK WERE DESIGNED 

-1 Two si/cs of lEoncralorH were to !« driven by the air fans it 
w!is ilcsinii io develop. The ewential data on tho?e arc as foUom: 



, F. ORAT, I. W. BBED AND P. N. ELDEBEIN 





forR«iio 
TBlecnpbSeti 


Osnentoi 

forBxlio 
TelsphoneSeU 






200 
330 


4H 


Genemtor output, w 























METHODS OF TEST 



5 Practically all of the tests recorded in this paper were made 
with the wind tuimel of the Bureau of Standards, and the manage- 
ment and personnel of the Bureau aided materially in expediting 



Fia. 2 Test Genehator with Streamline Casino 

them. A special testing generator was mounted in the wind tunnel 
as shown in Fig. 1, and the fan to be tested was attached to it. This 
generator was provided with a magnetic tachometer, a separately 
excited field, and convenient means for applying load to the arma- 
ture circuit. The external shape and size of the machine were made 
identical with those of the radio generators, with which the air fans 
were to operate in service, by the addition of the molded micarta 
streamline casing shown in Fig. 2. With this generator and the 
regular wind-tunnel equipment for measuring wind velocity, tests 
could be made rapidly and accurately. 

6 All of the air fans considered in this development have normal 
speeds of 4000 r.p.m., or above, and the centrifugal forces are very 



484 AIR FANS FOR AIRPLANE GENERATORS 

considerable. It was therefore found necessary (o pruvide an 
overspeed test to precede the wind tunnel test. For this purpose a 
30-hp. motor operating througii a 43 to 3 De I^aval siiecd-inereas- 
ing gear to a suitable shaft extension for the fan was set up. 
This device was capable of driving the air fans at speeds up to 
14,000 r.p.m. and to prevent danmge to gciirw or ticarinps in case 
of failure of the fan, u "weakest point" was provided by a ri>- 
placcable notched-shaft extension. Failuit! at this jwint inon-ly 
resulted in the breaking of the rcplawablo sliaft extension, and of 
course the destruction of the fan, without uijury to liearinps or geai-s. 
Heavy guards prevcntwl the flying fragments from causing any 



I'lu. ',i I'HK-W.^K AiK Fans 

damage. This i<|uipment wsis tle-signeitl.yC'apl. 1'. K. rernnt,()ffi(vr in 
I "harge of the Signal ( 'uriis Ijiboratmies at the Mim'au of Standanls. 

7 'I'herv is one >enons obji'i-iiim to (liif iiietlioil of overspwd 
li'stiiiiT in that it al>s<irl>s a lar^e iliikimiiI nf |H.\t'<-r and iiuts a}>- 
iiiirrnal lhrii>i >1r:iitis on tlii> fniis, A biillli'. io ))revenl air How 
nialiTJally n-diic-i-d tlii-: •'tV<-i-l, and latri' i'\|irn''nn' indieatetl thai 
il would