Skip to main content

Full text of "Electrical boats and navigation"

See other formats

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

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

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

Usage guidelines 

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

We also ask that you: 

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

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

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

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

About Google Book Search 

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

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


of the Harvard Colkgi Lihrarif 

This book is 


and circulates only with permission. 
^ Please handle with care 

and consult a staff member 
before photocopying. 


Thanks for your help in preserving 
Harvard's library collections. 













Patt Preaideni American Institute Electrical Engineers, Editor The Electrical Engineer » 


Member N. Y. Electrical Society and Associate Member A. /. E» E, 

New York 

C. C. SheivI^ey, Pubwsher, 

lo & 12 Coi,i,EGK Place. 


--^T'.o -^ 

jyo^t> '"'4-. 


,. ^^ 

M 7 


Copyright, 1894: 

Chablbs C. Shbli<by, 

New York. 


It is believed that the contents of this volume will be of interest 
to a wide circle of readers, many of whom have not hitherto paid 
attention to the rapid advances made in the application of electric 
power to the purposes of navigation. The industry of equipping 
and operating electrical boats is still new ; but, as will be seen, a 
great deal of instructive experimental work has been done, and in 
many instances marked practical success has been attained. No 
attempt has been made here to give the details of every electric 
craft built up to date, but a certain number are dealt with as 
illustrative of the history of the art and of its evolution. The book 
also embodies a large quantity of data on the subject not brought 
together before. It has sometimes been found that an art passing 
through its earlier stages of perfection is helped by the appearance 
of literature on its new problems and conditions ; and should any 
stimulus to electrical navigation be given by this book, the authors 
will feel rewarded for their trouble. 

Mr. Martin has taken charge of the preparation of the book as 
a whole. Mr. Sachs has particularly applied himself to the uses of 
electricity in canal boat propulsion and similar work, including the 
data on screws, resistance, etc. The canal section is, in fact, prim- 
arily based upon a lecture delivered by Mr. Sachs before the New 
York Electrical Society. 

Fuller details could have been given with regard to some of 
the classes of boats here discussed, whether automobile or deriving 


live current by wire from a central source of supply ; but it is 
hoped that the book, as it stands, may subserve the needs of 
the present moment, while the art is still in a transition state. 
References are given in various places to authorities that may 
be consulted with benefit, as to detail apparatus. 

The authors wish to acknowledge their indebtedness to 
Mr. J. C. Chamberlain and Mr. F. Reckenzaun for suggestions and 
assistance ; and to Mr. Joseph Wetzler for collaboration in passing 
the book through the press. 

Thomas Commebpobd Mabtin. 
Joseph Sachs. 

Nbw York City. September, 1894. 



Electrical Boats : Historical and Intboductobt — ^Pbimast 
Battery Boats. 


Storage Battery Boats : Preliminary ; Single Launches. 


Storage Launch Fleets and Passenger Boats. 


Special Features op Storage Launch Operation and 


Special Electrical Craft — Rowboats, Catamarans and 
Paddlewheel Boats. 


Submarine Electbic Tobpedo Boats. 


DiBiGiBLE Electric "Tobpedoes" for Waefaee and Life 


Some Genebal Consideeations on Electric Launch Requibe- 



Canal Boat Propulsion : Histobioal — Ebib Canal. 


Conditions Entssing into Canal Boat Pbopulsion. 


Methods op Applying Electbicity to Canal Boat Pbopuxsion 
— Boats Equipped with Motobs. 


Methods op Elboteic Canal Boat Pbopulsion with Motob 
ExTEBiOB TO Boat. 

Genebating Plant and Distbibution. 


Resistance op Canal Boats — Compabison op Cost, Pbopelleb 
vs. Hauleb. 


Pbopulsion : Resistance op Boats and Pbopellebs ; Paddle- 
wheels AND SOBEWS. 


Miscellaneous Uses op Electbical Powee, 

Storage Battebibs, Motobs and Dynamotoes. 

Electrical Boats and Navigation. 


Electrical Boats : Historical and Introductory. — 
Primary Battery Boats. 

1. The success of electric locomotion on land is now well 
established, and thousands of electric cars are to-day in 
use on the streets, on elevated roads, in tunnels beneath 
cities, and across country. But while electric traction has 
thus been advancing rapidly, electrical navigation has 
also made progress more quietly, as an art and an indus- 
try ; and it is believed that the time has arrived when the 
subject of electrical boats, for a large variety of services, 
may be discussed with pleasure and profit. As the suc- 
ceeding pages will show, electrical navigation divides under 
two main heads, namely, that which includes automobile 
or independent boats carrying their own power, and that 
which includes dirigible or dependent boats deriving cur- 
rent all the time from a fixed point on shore. In some 
instances the characteristics of the two groups interlap, but 
the arrangement of the chapters in this volume has been 
made in accordance with the classification suggested by 
such a natural difference in function and performance. 

Nearly sixty years ago — 1838 — Prof. Jacobi, a distin- 
guished physicist and electrician residing at St. Peters- 
burg, made a demonstration on the Neva of what could be 
accomplished with the electric motor as a means of propul- 
sion for boats. Towards the expenses of the experiment, 


the Emperor Nicholas contributed $12,000. Jacobi em- 
ployed a form of motor wMcli had already attracted con- 
siderable notice, having been brought before the Academie 
des Sciences of Paris in 1834. This attempt is the first of 
the kind on record, and is another proof of the originality 
and versatility in electrical investigation that made Jacobi 
one of the discoverers of the great modern art of electro- 
plating. Just what led the Emperor to take an interest in 
the application of electricity to marine propulsion and 
spend this money on it is not known, but it is supposed 
that these tests, like others, then, in Europe, were due to 

Fig. 1. — Jacobi's Elbctbic Boat Motob op 1838-9. 

the exhibitions made in London by Thomas Davenport, the 
Vermont blacksmith, of his electric motors. The Jacobi 
boat on the Neva derived its current from primary bat- 
teries, and the motor propelled it at a speed which never 
exceeded 3 miles an hour. Both Grove and Daniell cells 
were used in the first and succeeding experiments. The 
motor (Fig. 1) was geared to paddle wheels. The boat was 
28 feet long, 7 feet wide and 2 feet 8 inches in draught. 
She carried as many as 14 passengers. 

Since that time and the later work in England of Robert 
Hunt; of G. E. Bering, in 1856; and the Count de Moulins in 
France in 1866, there have been a great many other attempts 


Fig. 2. — TeouvIj'b Elbcteic Boat on the Seine. 

Fig. 2a. — Teouvb'b Boat foe Five Passbngees. 


to apply primary batteries to the propulsion of boats, but 
the experiments have nowhere resulted in lasting success. 
Perhaps the most notable instance is the work done by the 
ingenious French inventor, Trouve, who at the Paris Expo- 
sition of 1881 showed a small boat (Fig. 2) of somewhat 
novel construction. He employed a motor, placed imme« 

Fig. 2b. — TbouvJbj's Motob and Scbew. 

diately over the rudder and driving by means of sprocket 
cogwheel and chains a three-bladed screw, carried in the 
rudder frame (Fig* 2b). Current was conveyed to this motor 
by means of two little flexible cables from the batteries ; 
and the cables served also as yoke lines. The batteries 
were of the bichromate of potash type, with the plates so 


arranged as to be raised from or lowered into the solution. 
Trouve also tried Plante storage cells, but gave them np. 
With a load of four or five passengers, this boat would 
attain a speed of about Si miles an hour. Fig. 2a shows 
a similar Trouve boat of the same period. 

Other workers here and in Europe have made various 
boats for primary battery use, and some are probably run- 
ning at this moment. There is in reality no reason why 

Fig. 3. — The Pbimaby Battery Launch "Electric" on the 


they should not be numerous in a day when every newborn 
genius dabbles in electricity and when any form of electric 
locomotion has undoubted fascination for the public. An 
ordinary primary battery is not difficult to manage, and if 
a good form of solution is used, such as the bichromate of 
potash, a long run, say of four or five consecutive hours, 
may be made by a light boat, at a cost for motive power 
not exceeding 10 cents an hour, and even less. 

2. In order to show what may be done with primary 
battery boats, we will illustrate and describe in passing, 



one with which many trips have been made on the Potomac 
at Washington, and one whose size marks her as the most 
pretentious of her class. 

This launch (Fig. 3) which was built by the Naphtha 
Launch Co., of Morris Dock, N. Y., and A. Glose & Son, 
New York, is 21 feet in length and 5 feet beam. She 
draws 18 inches of water with battery and motor, which are 
thus, it will be observed, barely sufficient to serve as ballast, 
instead of being an encumbering dead weight. The 
propeller is 17 inches in diameter with three blades, 
and makes 250 revolutions per minute. To the propeller 
is geared a small Riker motor wound for 60 volts and 30 
amperes, and weighing 240 pounds. The gear wheels are 
respectively 2i inches and 6 inches. The launch has aver- 
aged 7 miles an hour. 

Fig. 4. — Plan op Pbimaby Batteby Launch " Electbic." 

The battery equipment is very interesting. It consists 
of 60 Hanson cells grouped as shown in the engraving. 
Fig. 4, with 30 cells on each side, and so connected that 
half the required current comes from each row of cells. 
Each battery box holds two cells. Each cell is 11 x 12 x 3 
inches and in each there is a porous cup 9x11x1 inch, in 
which is suspended a carbon plate 7xllxi inch. Out- 
side the porous cup are suspended two zinc plates each 
7xl0xi inch, the surface being thoroughly exposed to 
the action of the solution. In the porous cup are three 
pints of strong solution and in the outer cell are six pints 
of weak solution. The total cost of the solution, including 
depolarizer, for the 60 cells, is $1.60. From one charge of 
this battery, the yield has, it is said, been 180 ampere 
hours. The cell gives an average of 1| volts, and will yield 
50 to 60 amperes on short circuit. The zincs are amalga- 


mated before using, and again after 24 hours' use. The 
average consumption of zinc is less than one pound per 
hour when the battery is in use ; and when it is not in use 
the local action is not sufficient to render it necessary to 
remove the zincs. Each double cell weighs 40 pounds, or 
1,200 pounds for the 60 cells, when the zincs are new. The 
battery cells are of wood treated with a special preparation 
making them acid-proof. 

From one charge of the solution, the boat gets a good 
12 hours' run. In the tank shown at the bow of the boat, 
is carried extra solution enough for three more charges. 

Figs. 6 and 5a — Vauhan-Shbbein Primaby Battbby Boat. 

One man can recharge the whole 60 cells in li hours. Of 
course the 12 hours of service can be extended over several 
days, a few hours at a time. The current is thrown on or 
off by a simple switch that anyone can control. The 
"Electric" seats 12 persons easily. 

3. One of the most noteworthy of the primary battery 
boats built in England is that of Mr. Vauhan-Sherrin, 
(Figs. 5 and 6a) launched a few years ago and tried during 
1890-91, although no recent reports have been made public. 
The primary battery used by Mr. Sherrin is a two-fluid 
form, in which the anodes are of sheet zinc, and the 


cathodes are of carbon, specially prepared. In each cell 
there are three fixed cathodes and two replaceable anodes. 
Very light plates are used, and the particular construction 
adopted permits these to be placed very close together, so 
that the internal resistance is small. The oater cells are of 
gutta-percha, and in them are embedded the porous cells 
which surround the anodes. The liquid used in the inner 
cells is simply water ; that placed in the outer cells around 
the carbon cathodes is a depolarizing liquid of special com- 
position, capable of being produced at a low cost. In one 
of the tests made by Professor Silvanus P. Thompson, one 
of these cells gave out an average current of 8.75 amperes 
for five consecutive hours, with an average electromotive 
force of 1.88 volts, although the cell was only about half 
filled at starting. Professor Thompson says that he knows 
of no battery, primary or secondary, which, for a given 
gross weight of cell, will yield as great an output, while 
the economy of zinc is remarkable, the consumption being 
close to the theoretical limit. The net cost of electric 
energy from such cells is estimated at 18 to 20 cents per 
Board of Trade unit of 1,000 watt hours. 

The motor is a modified two pole Gramme machine, hav- 
ing the field magnets constructed in a special manner, 
which, while maintaining great mechanical strength, ad- 
mits of perfect lamination of the iron. It is well designed 
and constructed, and when properly set, is free from spark- 
ing at the commutator. It is also light, a one horse-power 
motor representing only 62 lbs. of dead weight. 

The inventor can carry his battery composition in the 
form of a paste, which, by merely mixing with water, 
forms at once a fresh charge for his batteries. One writer 
describing this boat reports having seen a launch 40 feet 
long, belonging to Mr. Sherrin, and fitted under his system 
with a capacity for a 600 mile continuous run. 

4. The statement has been made of another primary bat- 
tery boat built in 1892, to run on the Schuylkill River, that 
she has made a trip from New York to Philadelphia, or 
about 90 miles, on one charge of solution ; but the report 
lacks "corroborative detail." In the meantime, in view 
of the many attempts still likely to be made in this direc- 


tion, it is worth while to point out that any figures given 
as to the marvelous capacity of primary batteries to do a 
great deal of work for nothing and to yield a profit on the 
residual by-products should be received with a fair amount 
of caution. A valuable discussion of this subject will be 
found in a paper by Prof. Francis B. Crocker, of Columbia 
College, read before the American Institute of Electrical 
Engineers, in 1888, on ''The Possibilities and Limitations 
of Chemical Generators of Electricity." This paper has 
never been used as part of the prospectus for floating the 
stock of a new primary battery company. We may quote 
the conclusion of Prof. Crocker's remarks : — "I am by no 
means a skeptic in regard to chemical generators of elec- 
tricity ; the possibilities are very, very great as I have 
shown. But these possibilities do not seem to have been 
brought to reality in a very perfect manner as yet. But 
batteries even in the imperfect state in which they exist 
to-day have their useful and legitimate function. A 
Leclanch6 cell is exceedingly well adapted to ringing elec- 
tric bells intermittently and to telephone work; and 
gravity batteries have long done good service for tele- 
graphic purposes. 

"But when it comes to developing any considerable 
amount of actual power, then the limitations become appar- 
ent. When we remember that battery electricity will cer- 
tainly cost in practice 50 cents per horse-power hour, since 
the materials alone cost 26 cents, and that dynamo elec- 
tricity only costs 2 cents per horse-power hour, the claims of 
some primary battery electric lighting promoters show up 
in their true light. As a luxury, of course, it makes very 
little difference what it costs, but even then people soon 
tire of paying very high prices for that kind of a luxury. 
For small electric lighting and small power in special cases, 
batteries are useful, particularly where no other source of 
current is available. A physician or dentist to whom a 
horse-power hour may be worth hundreds of dollars could 
almost afford to use a chloride of silver battery and throw 
away the silver." 

5. These cautionary remarks may fitly close a chapter 
on primary battery boats. If a good battfery be obtained 



for which no extravagant claims are made, and if the boat 
be equipped with almost any one of the large number of 
small motors on the market, its owner should be able to 
derive many pleasant hours from its possession, with abso- 
lute, or almost entire, respite from the fatigue of rowing or 
the care of tending sails. To invalids and lazy people this 
is quite a consideration. Any ordinary rowboat can be 
converted into an electric boat. The apparatus complete 
for a 12 foot boat, to make a speed not greatly exceeding 
4 miles an hour will run from $160 to $250, placing the 

Figs. 6 and 6a. — Teoitve's Raft Battery fob Sea-Going Ships. 

equipment within the power of persons of moderate means. 
To this sum must be added the cost of renewal of zincs and 
solution, for which 60 cents a day would be a fair allowance. 
It will be gathered from what has been said that the pri- 
mary battery boat is looked upon as one for individual 
and casual use rather than for such units and heavy service 
as would constitute in magnitude and importance a new 
development of electrical engineering. At the same time, 
we must not forget to mention that, returning to ideas ad- 
vanced at least fifty years ago, M. Trouve has suggested that 
large electric boats for sea-going purposes may be propelled 


by primary batteries, the elements of which are immersed in 
the salt water of the ocean as an electrolyte. For the sake 
of the curious, we here reproduce sketches (Figs. 6 and 6a) 
from the history of M. Trouv6's inventions/ showing an 
electric ship to which is attached a "raft-battery," the 
cables from which carry the current to the motor on the 
ship, the raft consisting of copper-zinc couples, so grouped 
as to give the desired voltage and amperage. The detail 
drawings illustrate other plans of supporting the elements, 
with arrangements for greater or less submersion in the 
sea. M. Trouve assumes that from such a battery he might 
get 60 watts per square metre of surface of zinc, both sides 
of the plates being utilized ; and that a raft 100 metres 
long and 16 metres wide would by a proper disposition of 
the plates, dipping 4 metres into the sea, yield him 5,120 
horse-power (French) on the same basis of calculation'. On 
arriving in port, it would be much easier, he thinks, to 
renew this "beautiful magazine of energy," by simply tak- 
ing another "raft" oflf the wharf or out of dry dock, than 
to recoal a steamship of corresponding' capacity. It is 
natural to express the wish that some experiments might 
be made in this direction, at least on a small scale. That 
we shall ever see a large ship laboriously hauling an acre 
or two of such batteries through the surf, is almost too 
much to expect. Many persons would prefer to try gener- 
ation of current by thermopiles or pyromagnetic means. 

1. Hiatoire d'un Inventeur. By Qeorges Darral. Paris, 1891, p. 461. 

2. The French h. p. is 82,649 foot pounds per minute, or slightly less than the English. 



Storage Battery Boats. — ^Preliminary : Single 

6. The electrical propulsion of boats was one of the many 
industries which had to await the perfection of means of 
generating current cheaply. Its success depended also 
upon improvements in motors and in no small degree upon 
the introduction of the modern storage battery. Not only 
did the fumes from Jacobi's batteries drive away the spec- 
tators on the banks, but his apparatus was as a whole the 
most costly and inefficient that could well be conceived as 
likely to drive away capital, although at the time of its 
use it was practically the best to be had. It is not less 
true that each of the later experiments down to 1881 repre- 
sented the hopes raised by some new departure in the 
electric arts, hopes that could not be realized, however, 
under the conditions that had to be encountered. 

The great advance made in the production of the 
dynamo-electric machine had an immediate influence upon 
the electric motor, while at the same time the fact that 
large generating machines could be run by engines and 
turbines stimulated experiment with storage batteries in 
which some of their current might be accumulated. One 
of the most noteworthy results of this interaction and par- 
allel evolution is the storage battery boat of to-day, which 
is already so successful and which promises to add so 
greatly to the comfort, convenience and welfare of all 
whose pleasure or business takes them upon the water. 

In his boat at Paris, Trouve bridged the gap between the 
old practice and the new, but the first serious attempt in 
the direction of utilizing stored current was made in 1882, 
when the late Anthony Reckenzaun, the brilliant young 
Austrian electrical engineer, who died all too soon, designed 
a launch called "Electricity," for the Electric Power 



Storage Company, of London. She was a boat of good 
size, being 25 feet in length and about 6 feet in beam. Her 
draught of water was 1 foot 9 inches forward, and 2 feet 6 
inches aft-. During the earlier experiments with her, cur- 
rent was furnished from 45 accumulators of the Sellon- 
Volckmar (Faure) type. These were connected to two 
Siemens motors which operated singly or jointly, and 
which communicated power by countershaf ting to a 22-inch 
propeller screw. The boat would carry 12 passengers, and 

Fig. 7. — ^Thk "Volta" Electbic Launch. 

made as high a speed, it is said, as 8 knots an hour against 
the Thames tide. 

7. The next important step was taken in 1883, when the 
Yarrows, of England, exhibited at the Vienna Electrical 
Exhibition a launch 46 feet in length, capable of carrying 
50 passengers, and able to maintain a speed of 8 or 9 miles 
an hour. Current was furnished by 70 accumulators 
stowed away under the floor of the launch. The motor, of 
the Siemens type, drove directly, the shaft of the armature 
being continued as the spindle of the screw. The cost of 
the launch complete was placed at $3,000. From that time 
onward, advaijces in electric launch work were made, 


chiefly through the energy of Mr. Reckenzaun ; but the 
attention of the public was not arrested until a remark- 
able trip was made across the English Channel in 1887, 
when on September 13, Mr. Reckenzaun voyaged from 
Dover to Calais and back. His launch '' Volta" (Fig. 7) 
was a boat 37 feet long, 6 feet 10 inches beam and 3^ feet 
draught, and was built of galvanized steel plate. Her bat- 
tery consisted of 61 accumulators placed under the floor, and 
she was propelled by two Reckenzaun motors. The trip 
was made on a single charge of the batteries. The launch 
carried 7 passengers, and made a speed ranging between 
6 and 12 miles an hour. The Channel was fortunately not 
as ''choppy" as usual on this occasion, and the progress 
of the boat was so stealthy and quiet that one of the party 
captured with his hands a sleeping gannet afloat on the 

8. As a result of this very effective demonstration of the 
capabilities of electric power in navigation a great many 
boats were built in Europe, and their number has been 
steadily on the increase ever since. In England, the 
Immisch electric launch service was started, for example, 
in 1888, and its fleet to-day includes several very flne boats 
operated by the General Electric Power and Traction Co., 
Limited. The best known of these boats is the ' ' Viscountess 
Bury," which is 66 feet long, by 10 feet beam, with a 
mean draught of 2 feet 9 inches. She has often accommo- 
dated 70 passengers, so that in this country she would 
carry at least 140. Her dining saloon will seat 24 passen- 
gers. She has a battery of 180 E. P. S. accumulators which 
can be arranged in two or four parallels, as desired. Her 
motor will easily develop 10 horse-power at 1,000 revolu- 
tions. It drives a two bladed screw of phosphor bronze, 
19i inches in diameter, with a pitch of 18 inches. Her 
speeds are 4i and 6 miles an hour. Further mention will 
be made of this class of boats, belonging to fleets in regular 
service, in a later chapter. 

9. The flrst storage battery launch put in commission 
in America was that owned by the brothers Anthony and 
Frederick Reckenzaun, called the ''Magnet" (Fig. 8). She 


was built by the latter, at Newark, N. J., in 1888, and was 
28 feet long, with 6 feet beam and 3 feet depth amidships, 
and drew 2 feet 6 inches at the stern. Her Reckenzaun 
motor drove directly a gun metal two bladed screw of 18 
inches diameter. The motor weighed 420 pounds and fitted 
snugly into the curved lines of the hull. The boat was 
equipped with 56 cells of the Electrical Accumulator Com- 
pany' s make, disposed under the floor along the keel, each 
cell weighing a little over 40 pounds and the total weight 
being about 2,400 pounds. The motor and cells were con- 
trolled by two switches placed near the pilot's seat in the 
stern. One of these started, stopped or reversed the motor. 
The other connected the battery in a series of 66 cells or 
grouped it in two parallel series of 28 cells each. The total 
weight of motor, cells, switches, wiring, screw, etc., was 
about 3,000 pounds. With the cells in parallel, the motor 
and screw made about 540 revolutions per minute, consum- 
ing an average current of 33 amperes, or nearly 2i electrical 
horse-power and driving the boat at a speed of from 6 to 8 
miles an hour. One charge was good for a 10 hours' run, 
covering a distance of from 60 to 75 miles. With the cells 
all in series, a speed of 10 to 12 miles could be made. But 
in this case both the electromotive force and the amperes 
were about doubled, and the net result wa& that while 
speed was gained, the duration of the discharge was cut 
down to about one-quarter of the normal time and the 
actual mileage to about one-half. Besides supplying cur- 
rent to the motor, the cells lit up seven 16 candle-power 
incandescent lamps and a 100 candle-power lamp placed in 
a reflector. The cells were charged from the factory plant 
of the Electrical Accumulator Co. at Newark, over about 
1,600 feet of serial line of No. 8 B. & S. wire, the current 
being from 20 to 30 amperes at from 140 to 150 volts. 

This boat which would carry about 20 passengers sitting 
back to back along the centre under an awning, was sent 
some time ago to California, but not until after she had 
done a great deal of work on the Passaic River, Newark 
Bay, the Staten Island Kills, and even in New York Bay 
and the Hudson River. It was the good fortune of the 
present writer to make a long trip in her, in 1888, from 
New York to Newark, across the path of the ocean 



steamers, the whole journey back and forth between the 
two points representing from 60 to 60 miles at least. 
Figs. 8a and 8b show the boat in plan and longitudinal 
section. She is still in use at San Francisco. 

10. As in Europe, so here, the last few years have seen 
a very general adoption of electric launches for private 
and special use. Reference will be made in a succeeding 
chapter to the use of such launches at the World's Fair ; 
in the present chapter we propose simply to take note of 
individual boats of importance. One of the foremost 
American patrons of this type of craft has been Mr. John 

Fig. 8. — The Rbckenzaun Boat "Magnet," on the 
Passaic, N. J. 

Jacob Astor, who has long made electricity a study and 
whose personal interest in electrical navigation has given 
the art a most valuable and helpful impetus in this coun- 
try. During 1893 he put into commission an electric 
launch called the "Corcyra," illustrated (Fig. 9), page 18, 
built in accordance with his designs, and' those of Mr. J. C. 
Chamberlain, whom he consulted. She was so immediately 
successful that her lines and general plan were adopted by 
the American and Russian naval authorities. 

While being entertained by Mr. Astor at Rhinebeck-on- 
the-Hudson, the Grand Duke Alexander, of Russia, was 
much pleased with this launch and learned that a similar 















launch had just been completed by the General Electric 
Launch Company for the U. S. Government cruiser "New 
York," to be used as the Cg^ptain's gig. He shortly after- 
wards had an opportunity to inspect this electric gig also, 
and took such a fancy to it that, at the request of the 
Department of State, the Navy Department directed the 
contractors to deliver the launch to him and to begin 
the construction of a duplicate for the cruiser "New 

Fig. 10. — Electbic Gig fob the Gband Duke Alexandeb. 

This electric gig, which is illustrated in the accompany- 
ing engraving (Fig. 10), is 30 feet long, 6 feet 8 inches beam 
and 22 inches draft. It is equipped with 66 storage bat- 
teries capable of delivering 3 horse-power to the motor for 
a period of 10 hours with one charge. The motor normally 
has a speed of 650 revolutions at 3 horse-power and is able 
to propel the boat at a speed of 6.4 miles per hour. The 
motor, however, is capable of developing 12 horse-power 
for a spurt of 5 or 10 minutes, increasing the speed of the 
boat to about 10 miles an hour. The batteries are placed 


tmdemeatli the seats and beneath the flooring, leaving the 
entire boat space for passengers. The motor is placed 
near the centre of the boat and underneath the flooring, 
being directly connected to the propeller shaft. The con- 
troller for regulating the speed is located near the wheel. 

The launch will be used hereafter for the personal service 
of the Grand Duke in his cruises around the Russian bays 
and waters. 

Mr. Astor has more recently had built for his personal 
use a twin screw 46-foot cruising launch called the '^Pro- 
gresso," which is shown in Fig. 11. She is undoubtedly 

Fig. 11. — ^Mb. J. J, Astob's 46 ft. Twin Screw Cabin Obuising 
Launch "Pbogbbsso." 

the largest electric boat afloat in America, but it is under- 
stood that her owner has even more extensive plans under 
consideration. It is to be hoped that they may soon be 
realized, for the benefit of the art. The " Progresso" has 
twin propellers of 18 inch diameter, and each motor makes 
800 revolutions when working up to capacity, viz., 4 horse- 
power each. She has 136 cells of battery, and makes from 
8 to 12 miles an hour easily. 

11. The Universal Electric Launch Company, with works 
at Nyack, N. Y., made recently an interesting test of a 
new 40-foot launch embodying several improvements. The 
boat is 40 feet long, 6 feet 6 inches in beam and normally 
draws 2 feet of water, giving a displacement of about three 






'' iV 











tons. It is equipped with. 72 cells of battery, weighing 
3,000 pounds and having a capacity of 150 ampere hours. 

The accompanying illustrations (Figs. 12 and 12a) show 
a sectional and plan view of the boat. The motor, a 10 
horse-power Riker machine running at 600 revolutions per 
minute, is, as will be seen, placed in the bow. The enlarged 
view (Fig. 13) gives a clear idea of its arrangement and the 
method of setting. 

The general form of the machine, that of a letter V, 
admits of its being placed under the forward deck in small 

Fig. 13. — ^MoTOB of Bikeb Electric Launches. 

boats or under the flooring in larger boats if desirable, 
where it is out of the way, yet easily accessible. The 
toothed armature is very low and near the keel, to which 
the machine is firmly, bolted, and there is no vibration. 
The cells are arranged to be instantly changed from 
series to parallel or xice versa by a pull switch, as normal 
or high speed is required. During the test the launch 
made 5J miles an hour with the batteries in parallel, 
giving a current of 20 amperes at 70 volts, and spurts of 
eight miles an hour with the series arrangement, by which 
43 amperes at 140 volts are delivered to the motor. The 



test of speed was considered very satisfactory, especially 
in view of the fact that the boat had been in the water all 
summer and the hull was consequently very dirty. 

This launch acquitted herself most gallantly during the 
storm that raged along the Atlantic coast, August, 1893. 
She was at that time lying at anchor off Stamford, Conn., 
directly to leeward of a large naphtha launch, which, in 
spite of all the power of her engine running at full speed 
to take the strain from the cable, dragged her anchor and 
drifted down upon the Riker boat. In order to escape a 

Figs. 14 and 14a. — ^Pinnace fob Steamships and Men-of-Wab. 

collision the latter was forced to slip her cable and rely 
upon her motive power alone. This manoeuvre was suc- 
cessfully executed. The launch forced her way directly 
against the heavy sea and the gale, weathered a dangerous 
reef and then, turning in the trough of the sea, ran out 
of danger. 

12. One more boat deserves inclusion in this group, 
namely, the electric pinnace shown in 1891, by the Electric 
Power and Traction Company at the Naval Exhibition in 
London ; and here illustrated in Figs. 14 and 14a. It is 


designed on the lines of the English Admiralty steam 
pinnaces and is 36 feet long, 7 feet beam, with a maximum 
draught of about 2 feet. It is built in pine with oak stem, 
stem, and stem posts, and is bright all over. The accumu- 
lators, 50 in number, are arranged in teak boxes under the 
seats, as shown in section, and are so placed as to be easily 
removable when necessary. The whole is so strongly built 
that the pinnace is slung on davits with all electrical equip- 
ment ready for use. The cells are charged either with the 
pinnace slung in the davits or moored alongside the ship. 
The motor is much more powerful than those usually sup- 
plied by the company for use in the launches on the 
Thames and at Windermere, and a speed of 11 miles per 
hour is obtained. Since these boats are principally used 
for shore purposes, speed and power have been considered 
of more importance than duration of run. For shore and 
harbor pinnaces electricity possesses many advantages over 
steam ; and they will no doubt be largely in demand before 
long. One of these pinnaces, 40 feet long, 7 feet 9 inches 
beam, and 2 feet draught, with a handsome teak cabin, 
has been built to the order of the naval department of the 
Russian government. 



Storage Launch Fleets and Passenger Boats. 

13. It is evident that storage battery launches oflEer many 
advantages for the purposes of miscellaneous work on a 
large scale, and that if a good charging station be provided 
there is little limit to the number of boats that can be 
operated from it. A peculiar feature of attractiveness is 
that the boat can be put in the hands of any intelligent 
person and that it is free from many if not most of the 
objections that attend the employment of manual labor, 
steam power, or sails ; while it has some excellencies that 
even the convenient naphtha launch does not share. With 
an electric boat, there is an absolute freedom from noise, 
heat, smoke, gas, cinders, ashes, etc. There is nothing 
to explode ; the current used is of a perfectly safe low 
pressure free from any risk of giving shock; and if the 
boat is properly wired there should be no danger whatever 
of fire. The boats are necessarily stable, the centre of 
gravity being low; the batteries are under the floor or 
seats and can be disposed as desired. This also increases 
the carrying capacity enormously, as the entire deck is 
free for passenger occupancy and there is no need for any 
movement of the crew, should the boat be manned. An 
element of economy is that this form of motive power dis- 
penses with a trained crew or engineer, and that when the 
boat is needed, a simple movement of the switch renders 
part or all of her concealed power immediately available. 
As soon as the boat stops, all consumption of energy or 
fuel ceases. 

14. These are some of the reasons that have led to the 
equipment of storage battery fleets, which, so far, have 
been of two kinds, namely, fleets all the boats in which are 
intended to perform a common duty in the nature of f erri- 


age ; and, secondly, fleets any boat of which can be hired 
individually for the day, week, or season. Before passing 
on to describe both classes, it will be interesting to quote 
here some figures prepared in 1890 by Mr. Fred. Recken- 
zaun, at the request of the writer, to show how an 
investment in such a fleet might be made and what were 
the possibilities of return. The calculations and separate 
items may be taken as fairly approximate at the present 
time, subject to special conditions and to the steady lower- 
ing of the price of all electrical apparatus from year to 

Estimate of Cost 

of a fleet of 12 electric launches, each 28 feet long, 6 feet 
beam, carrying one ton of storage batteries, to run 6 miles 
per hour for 60 miles with one charge : 

12 hulls complete, with interior fittings (bat- 
tery troughs, seats and lockers), fixed roofs, 
shades, flag staffs, steering wheels, etc., - $6,600 

12 tons storage batteries (cap. 16,240 watt hours 
per ton) at $560 per ton, ... - 6,720 

12 motors, at $400, --.-.-- 4,800 

12 screw propellers, shafts, couplings, thrust 
bearings and stuffing boxes, - - - 1,200 

Switches, wires, incandescent lamps (4 per 
boat), with fittings, - - - - - 480 

Acid and labor of placing electric outfit, - 1,200 

Seat cushions, ropes, boat hooks, tools, pumps, 
etc., - 300 

Total. 12 boats complete, in running order, $21,300 
or $1,776 each. 

CHABGiNa Stations. 

Land and buildings (on suburban water front), 

say, $4,000 

Steam plant, 60 horse-power complete, erected, 4,000 
Dynamo, capacity 40,000 watts, with accesso- 
ries, erected, 2,000 

Charging circuits and appliances, erected, - 250 

Mooring facilities, tools, etc., - - - - 500 

Total cost of station, say, - - - $10,750 
Grand total cost of 12 launches with charging 
facilities and real estate, - - - - 32,060 

storage launch fleets and passenger boats. 27 

Estimated Cost op Operation. 

It is assumed that each, of the 12 launches makes a daily 
run of 60 miles, divided into 6 trips of 10 miles each 
(3 round trips) during 5 months in the year : 

12 pilots at $2.60 per day each, for 5 months, $4,600.00 

1 station engineer, at $3 per day for 6 
months, - - - - - - - 450.00 

1 station fireman at $2 per day for 5 months, 300.00 

1 station laborer at $1.76 per day for 5 
months, - - - - - - - 262.60 

Coal (4 lbs. per horse-power hour, 60 horse- 
power for 7 hours daily), 112i tons (for 5 
months) at $4 per ton, - - - - 460.00 

Oil, waste, miscellaneous supplies and inci-- 
dentals for 5 months, say, - - - 200.00 

Labor, etc., putting boats in running, order, 
at beginning and storing same at end of 
season, say, 360.00 

Depreciation, per annum, on boats and pro- 
pelling apparatus, at 10% on $21,300, - 2,130.00 

Depreciation of station machinery and ap- 
pliances, at 6% per annum on $4,750, - 285.00 

Interest, per annum, at 6%, on invest, of 
$32,050, 1,923.00 

Total operating expense, depreciation 

and interest, . - . . $10,860.50 

or $905.04 per boat per annum. 
Total mileage run per boat per month 

(60 per day), 1,800 miles. 

Total mileage run per boat in 5 months, 9,000 '' 
Total mileage run, 12 boats, at 9,000 
miles each, - - - - - 108,000 " 

Cost of operation, including running expenses, deprecia- 
tion and interest, as per above estimate, = lOyf^^ cents per 
boat mile. The boats assumed can seat 20 passengers and 
over. If an average of only one-half of this number is 
constantly carried, paying fare at the rate of one cent per 
mile each, the receipts will equal the operating expense, 
depreciation and interest on investment, as above. 

The boats, in this instance, run at intervals of about 17 
or 18 minutes (allowing for short stops), If miles apart, 
along the entire distance of 10 miles. 

The cost and operating expense of electric launches will, 



of course, vary with different sizes and speeds, which 
the conditions and requirements of each distinct case 
contribute to determine. 

15. One of the earliest, largest and finest fleets :^or public 
hire is that which was established in 1888, as already noted, 
on the Thames, using Immisch motors, and is now carried 
on by the General Electric Power and Traction Co., Lim- 
ited, with headquarters at Piatt's Eyot, Hampton. Accord- 
ing to the latest information received, the Company now 
possesses a fleet of 22 launches, 5 floating charging stations 
and 4 permanent charging stations. The largest boat is the 
"Viscountess Bury" (Pig. 15), to which reference has 
already been made. She can be hired, it is stated, for twelve 

Fia. 16, — ^Thb "Viscountess Buby." 

guineas ($60) per day, that sum including captain and assist- 
ant, lock dues, etc. As she will accommodate 70 passengers, 
it is evident that the charge per head is very small. The 
"Omicron" is also 65 feet long, like the "Viscountess 
Bury," but is only 7 feet beam. Then come a number of 
boats ranging in size from 45 feet long to 35 feet long, and 
another batch running up to 27 feet long, the charges for 
hire ranging proportionately from $30 per day down to $15, 
the sum named including wages for a deckhand, lock dues, 
etc. These launches have long been a familiar sight on the 
upper Thames, and are largely availed of by parties which 
wish to enjoy its placid beauties at a minimum of cost, 
fatigue or inconvenience. The passengers they have car- 
ried run up into the thousands. The boats ranging from 30 


to 45 feet have motors weighing 350 pounds, and at full 
speed develop about 3 brake h. p. at 760 revolutions per 
minute. They are equipped with from 40 to 60 cells. The 
rules of the Thames Conservancy require a low rate of 
speed as a precaution against ''wash," and the boats are 
therefore arranged to run at a normal rate of 6 miles an 
hour, or at a "half speed" of 4i miles. At the higher 
speed they take 28 amperes at 85 volts and at the lower 21 
amperes at 43 volts. The "Viscountess Bury" has 180 

16. Another picturesque spot that has welcomed the 
noiseless, smooth-gliding electric launch has been Lake 
Windermere, with whose scenes the names of Wordsworth, 
Southey, Ruskin and others are so closely associated. It 
can easily be imagined that in such a place these launches 
were found preferable to steamboats. The charging station 
is driven by water power, an ideal source of energy in such 
surroundings. An old mill had been partially destroyed 
by fire, but as the water wheel mechanism remained intact, 
a Glasgow firm of boat builders conceived the happy idea 
of putting dynamos in the ruined building at Cockshott 
Point. The conditions were not such as to ensure a high 
efficiency for the plant, but water power is generally cheap, 
and a good rate of hire for the boats has been obtainable. 

17. A boat that belongs in the category of passenger 
craft is the "Electric" transport, built for the British 
(government for the special purpose of conveying troops to 
and fro along the River Medway, between Chatham and 
Sheerness, two important military and naval depots. This 
boat is- open, 48 feet 6 inches long, with a beam of 8 feet 
6 inches, and built chiefly of mahogany and teak. She 
derives her power from 70 cells of 19 plates each. She has 
a draught of 2 feet 3 inches and will carry about 40 men. 

18. A very interesting and important experiment in the 
utilization of storage launch fleets was made in Scotland, in 
1890, at the Edinburgh International Exhibition, when the 
enterprise of the General Electric Power and Traction Co. 
led it to place a flotilla on the Union Canal. These boats 



proved a great success, and an interesting account of them 
was read before the Royal Scottish Society of Arts, by 

Mr. A. R. Bennett, M.I.E.E., to 
whom we are indebted for some of 
the details. 

These launches numbered four, 
and were built of steel, the hulls 
being designed by Morton & 
Williamson, of Glasgow. They 
measured 40 feet long over all, by 
6 feet beam, 3 feet 1 inch from 
gunwale to keel, and drew 2 feet 
1 inch when empty. Equipped 
with motor and cells, they weighed 
Si tons out of the water. Each 
boat had a capacity of 40 passen- 
gers, with ample elbow room. They 
carried 50 cells of the E. P. S. boat 
type, each cell weighing 58 pounds 
complete. The cells were of ebo- 
nite, standing upon glass insul- 
ators filled with resin oil. These 
cells were ranged along the sides 
of each launch in two rows, 25 on 
each side ; and the lids, covered 
by cushions, served as seats. The 
motors were of a modified Immisch 
type and weighed 350 pounds each. 
The propellers were coupled direct 
to the motor shaft. A ball-bearing 
thrust block was attached to the 
motor bed and constructed in com- 
bination with the plain bearing of 
the motor. Fig. 16 shows the 
launch in longitudinal section ; 
Fig. 16a a midship section ; and 
Fig. 16b shows the motor and the 
ball bearings. The motor had four 
field coils with a resistance of .18 
ohm when hot. These were in series with the drum arma- 
tore, which had 48 coils and a hot resistance of .3 ohm. At 



















its most favorable speed this motor had an efficiency of 85 
per cent. When the launches traveled at the rate of 4^ 
miles an hour the motor efficiency was cut down to 75 per 

Fig. 16a. — Midship Section. 

The controlling switch is shown in Fig. 16c. It had three 
levers. The current from the accumulators was led to the 
switch through a lead fuse which opened at 42 amperes. 
For the first instant there was nothing but the ordinary 
conductor resistance of the coils to overcome, and a rush of 
current would have occurred but for the interposition of 
the series of resistances shown. The beginning of the 
movement admitted current through a resistance of 2 ohms, 

Fig. 16b. — Immisch Launch Motor. 

which was gradually diminished as the lever passed the 
second and third contacts, until it reached the fourth, by 
which time the armature had started up and was develop- 
ing its counter electromotive force, when full current was 
on. The second lever in the cpntrolling__switch set the 



speed at 6 or 4i miles as desired by throwing the cells 
either all in series or in two equal parallel groups of 25 ; 
while the third lever was used for reversing, and simply 
changed the direction of the current through the armature. 
Neither the second nor the third lever could be shifted 
while the current was flowing, the first lever locking the 
other two mechanically while it was in the "on" position. 
In this manner the motor was protected from harm, and the 
current was guarded against waste. 

Fig. 16d. — Edinbubgh Charging Station and One op the 


The propeller shaft was bolted directly to the armature 
spindle, and passed through an ordinary water-tight gland 
at the stern, the back and forward thrust being taken up at 
the armature by means of the ball bearings already men- 
tioned. Propellers of various patterns were tried. At a 
speed of 4^ miles an hour, a three bladed propeller made 
too much wash, and a phosphor bronze two bladed form 
was adopted, having a diameter of 19i inches and a pitch 
of 14 inches. 



19. These four launches were charged during the night, 
so that all of them were available for service throughout the 
day. The charging plant, Fig. 16d, consisted of an Immisch 
shunt- wound dynamo, running at 750 revolutions and 
delivering 120 amperes at 130 volts, or about 21 horse- 








power. The dynamo was driven by a 25 horse-power 
engine. For charging, the cells in each boat were put in 
series, and then the four boats were grouped in parallel. 
Each launch was charged with current suflBicient to run it 
easily from 10 to 12 hours. The longest run made by any 


Th ^ 1 

1 lie 



iToa.ia— .81 .oil 


boat was from the Exhibition to 
Linlithgow and back, a distance 
of 40 miles. The current was also 
nsed for signal bells and incan- 
descent lamps. 

The boats plied regularly along 
the Union Canal, and a four cent 
fare was charged. Between May 
31 and October 11, the boats car- 
ried no fewer than 71,075 paying 
passengers, besides season ticket 
holders, officials and others enti- 
tled to ride free. On the busiest 
day, 2,560 passengers were 

20. It will be remembered that 
the World's Columbian Exposi- 
tion in 1893, at Jackson Park, 
Chicago, was laid out by Mr. F. 
L. Olmsted so as to include a 
series of lagoons and canals, com- 
municating with Lake Michigan 
and serving as a means of ap- 
proach to nearly all the import- 
ant buildings of the Fair. A 
course of about three miles was 
thus open to profitable use, and 
the competition for so valuable 
a concession was very active. 
Several concerns were bidders for 
the exclusive franchise, and each 
of them was required to build a 
sample boat to be submitted to 
trial tests in order to determine 
which was most suitable for the 
requirements of passenger traffic 
on the Exposition waterways. 
Steam, naphtha and electric 
launches were tested, and the 
privilege was awarded to the 









III '" 




Electric Launch & Navigation Company of New York, who 
thereupon equipped a fleet of 50 launches and placed them 
upon the waters of the Fair, while the Exposition equipped 
four boats for its own special use. The launches were fitted 
with Beckenzaun motors modified and constructed by the 
General Electric Company of New York, and with accumu- 
lators made by the Consolidated Electric Storage Company 
of New York. One of the boats is shown in Pig. 17, in 
perspective ; in section in Fig. 18, which also shows the 
motor and its bearings. 

Each boat was 35 feet 10 inches over all in length and 31 
feet 6 inches on the water line. The beam was 6 feet 2^ 
inches and the draught 27 inches. The hulls of these boats 
were constructed of white oak frames, with white cedar 
planking. The inner paneling, decks and other parts were 
mahogany. All the woodwork was finished in its natural 
color, thus giving a very rich appearance. The charging 
station was at the southeastern comer of the Agricultural 
building, in the South Pond. When the boats were to be 
charged they were laid up there, and whenever a boat 
needed repairs it was hauled up in its berth. The charging 
was all done at night, so as not to interfere with regular 
trips during the day. 

The motor was protected by a box which rose flush with 
the main deck of the boat, and was so set that all working 
and wearing parts could be readily reached. The storage 
batteries were placed around the sides of the boat, under 
the seats and entirely out of sight. The motor was nomin- 
ally of four horse-power, and was coupled direct to the 
propeller shaft, with a thrust ball bearing, in which the 
shaft ran. By this combination of direct coupling and 
thrust bearing all gearing and loss of power, as well as 
unnecessary noise and jar, were done away with. 

The batteries used were of 150 ampere hours' capacity. 
Each boat had 66 cells, and these cells could be connected 
in three groups of 22 cells in series or in two groups of 33 
cells in series. Fig. 18a shows the plan of wiring adopted 
for the boats, although in the diagram here presented 78 
cells are enumerated. For the regular and moderate 
speeds, the batteries were grouped in three divisions, in 
series, with or without a resistance in circuit. The other 


grouping consisted of putting the cells in two groups in 
series, with or without resistance. The controller con- 
sisted of magnetically controlled switches, combined with 
mercury cups, operated by a lever at the pilot's right 
hand. Four speeds ahead and two astern were obtainable, 
and proved adequate for every purpose. A current of 18 

Figs. 18b and 18c. — Arrangement of the Seats, Batteries, 

Motor, Etc. 

amperes per group was generally used to charge after a 
run of 50 or 60 miles at nominal speed, and from six to 
seven hours was required. In case of necessity, however, 
a current of 30 amperes was used, when the batteries were 
charged in four or five hours. 
Not one ot the pilots or guards who managed the fleet 



had ever before handled an electric lannch, yet they ex- 
perienced no trouble whatever from the first. The control 
was by means of a small lever switch at one side of the 

Fig. 18d. — Map of World's Fair Grounds, Indicating Interior 
Basins and Lagoons upon which the Electric Launches 
WERE used. 

steering wheel, located in the forward part of the boat. 
This lever allowed of four speeds forward and two back- 
ward. The nominal speed at which the boats were run 


was from six to seven miles an hour, but they had a reserve 
speed of from eight to ten miles. At the ordinary speed, 
the launches at the Exposition were in constant use from 
12 to 14 hours a day on one charging, and the cost of this 
charging never exceeded 60 cents per day per boat. All 
the electrical parts except the switch were protected, so 
that there was no possible danger of shock, and there was 
no report at the Exposition of any mishap of this kind. 

The greatest test the launches had during the period of 
the Exposition was on Chicago Day, when 622 trips, each 
trip of three miles, were made by the fifty boats. Six of 
these boats averaged 50 miles each, while 20 of them aver- 
aged over 40 miles, carrying on each trip about 40 people. 
There were no fewer than 25,000 passengers carried by the 
launches on that day, and entirely without accident. In 
fact, very few accidents occurred to any of the launches, 
and almost all that did happen were caused by the pro- 
peller becoming entangled in wires or other debris in the 
bottom of the lagoons, thus springing the shaft or bending 
the propeller blades. One bunch of wire was shown in the 
office of the company that was gathered up by a propeller 
and that would require a bushel basket to contain it. 

21. During the month of October, 1893, the writer had 
frequent opportunities to watch the performance of these 
boats and found it most satisfactorily to the public and to 
all concerned. Believing that definite data would be valu- 
able he secured from Mr. R. N. Chamberlain, the electrical 
engineer of the service, the following detailed figures up to 
that time : 

April 13 to October 1, 1893. 

Mileage and Passenger Traffic. 

Passenger trips of three miles each, - - 47,787 

World's Fair special launch trips, - - 6,750 

Special trips of regular launches, - - - 1,220 

Trial trips, 270 

Experimental trips, 180 

Total number of trips, - - - - 56,207 


At three miles per trip, total miles, - - 168,621 

Average miles per launcli to October 1, - 3,122 

The total number of days the 54 electric 

launches have been in service on the lagoons 

of the World's Fair was, - ~ - - 6,694 
Therefore the general average of miles per 

launch per day was, 25.67 

Minimum miles per launch per day, - - 14 

Maximum " '' " - • 37* 

Maximum miles, one launch, one day, - - 54 

Total number of passengers carried from May 1 

to October 1, 801,000 

Maximum passengers carried in one day by 

one launch, 464 

Maximum number of people carried by one 

launch for one round trip, - - . 40 

Operating Cost. 

Average cost per launch per day for charging, 
at 3 cents per electric horse-power, - - 55ic. 

Average cost per launch per day for care and 
repair of shafting, propellers, 54 motors, 162 
packing boxes, 3,524 storage batteries, in- 
cluding labor for charging 54 controllers — 
all the above being gone over every 24 
hours, 43c. 

Renewals of batteries per launch per day, - 41c. 

Renewals and repair material for all else per 
launch per day, 9c. 

Total cost per launch per day, - - - $1.48 J 
Average cost per launch mile for labor and 
material, exclusive of office expenses, - 53c. 

The Fair lasted a month after October 1, and the number 
of passengers reached about 1,000,000, before the service 
was suspended. When it is understood that even at the 
time these figures were compiled the launches had run on 
an average f 6ur times the number of miles a launch owned 
by a private individual would ordinarily cover during a 
regular season, and that the operating expenses were 


decidedly heavier than under ordinary circumstances, it 
will be acknowledged that the cost per launch mile is 
exceedingly low. Judging from the experience of six 
months with the 54 launches, Mr. Chamberlain believed 
that the expense can and will be, in the near future, 



% ■ 

■I I 


00 ^ 



brought down to as low a figure as three cents per electric 
launch mile where at least 30 launches are operated under 
like conditions. 

The original cost of these launches complete was about 
$3,000. At the end of the Fair the fleet was placed on sale, 
and several of the boats have been bought for use in differ- 


ent parts of the country by persons who had witnessed 
their operation and had become desirons to introduce them 
on local waters for private pleasure or for public hire. The 
money-making ability of the fleet may be inferred from the 
fact that it had earned a total of $314,000, on the above 
showing, the round fare for the trip of three miles being 
50 cents. 

22. In competition with these launches at the Fair, were 
a number of gondolas, which were a graceful and pictur- 
esque element, but which were certainly not more poetic in 
ease and smoothness of motion than the electric boats. In 
.fact the contrast in many practical features was so marked 
that a syndicate of Italians purchased one of the electric 
launches in September, 1893, and sent it direct to Venice to 
serve as the nucleus of a fleet on the historic canals. 
Under the fitting name of ''Venezia," the little craft, 
which had been so busy at the Fair, made her reappearance 
in the waters so long associated with the song and oar of 
Adria's gondolier. It is needless to say that she created a 
sensation, for her quiet and graceful performance, so 
different from the noisy, fussy movements of the steam 
launch, gave immediate proof that in the electric boat the 
time-honored gondola had at last met a rival against 
which no halo of poetr j^ and romance would help it greatly. 
Fig. 19 shows her at anchor opposite the Doge's Palace. 
She now has 68 cells in 4 groups, governed by a mechanical 
controller. In order to protect her hull against the weeds 
and action of the canal water, it has been sheathed with 
copper. The success of the boat has been indubitable, 
but it is understood that the fleet of which she is the proto- 
type will all be longer, so as to afford greater passenger 
capacity. Visitors to the World's Fair will note that her 
deck housing is slightly changed, it being now made to 
conform to the general style of the quaint cabin-top seen on 
the gondolas, in Venice, but not very much in evidence on 
the gondolas used at Chicago. 

In the succeeding chapter reference is made to other 
launches and launch fleets that have been started, and 
details are given as to methods of operation in use or likely 
to be generally adopted for such service. 




Special Features of Storage Launch Operation and 


Fig. 22.— Charg- 
ing Station. 

23. There are, all over this country, elec- 
tric railways running through picturesque 
suburbs, watering places and summer re- 
sorts, skirting navigable streams or termi- 
nating at the shores of lakes and bays. 
Thousands of pleasure seekers and rest 
seekers are carried daily to these charming 
spots where the railway people turn them 
loose to fall into the hands of the boatmen 
for the rest of the day. One railway man- 
ager, however, conceived the idea of being 
his own boatman and keeping the public 
himself. He carried out his plan with what- 
ever old apparatus he had at hand, or could 
secure ; operated two boats all last summer 
and is so well satisfied with the success of his combina- 
tion plant that this year he intends to enlarge his fleet 
materially and equip it with improved apparatus. 

The Seashore Electric Kailway at Asbury Park, New 
Jersey, runs for some distance at the northern end of its 
route near the shore of Deal Lake, a very pretty little body 
of water separated from the ocean only by a low range of 
bare sand dunes and reaching inland in labyrinthine fash- 
ion, its bare banks changing with surprising rapidity to 
cool mossy terraces covered with trees and undergrowth. 
At the ocean end, the road makes a turn within about 50 
yards of the lake. This point was selected for the " charg- 
ing station," a view of which appears in Fig. 22. Before 
discussing this structure, however, it will be well to speak 
of the boats themselves, as they are unique among electric 
craft. The flagship, ''Bonaventure," is a, flat-bottomed 


boat 36 feet long, 7 feet in beam, draws 18 inches of water 
and will carry 36 people at the rate of 8 miles an hour. 
It is equipped with 110 cells of Julien battery in series, 
modified to suit existing conditions, as will be explained, 
and a Crocker- Wheeler motor of 2 horse-power running at 
a potential of 220 volts (Fig. 21). The motor shaft is 1 inch 
in diameter and drives an 18 inch propeller. The boat has 
run for 12 hours on a single charge. The '' Dart" (Fig. 20) 
is 32 feet long and 7^ feet in beam with 2i feet draught. 
Unlike the "Bonaventure," it is clinker built and has a 
round bottom. Fifty-five cells connected in series supply 

Fig. 20. — Charging the "Dart." 

current at 110 volts to a Crocker- Wheeler motor in the 
centre of the boat directly connected by a 1 inch shaft to a 
22 inch two-bladed screw, making 600 revolutions a minute. 
The "Dart" carries 25 people at an average speed of 10 
miles an hour and can make, it is said, 16 miles an hour at 
a spurt. A Crocker- Wheeler rheostat of enameled wire 
coiled about cylinders of asbestos paper to prevent vibra- 
tion, controls the motor and is provided with stops for 7 

The batteries, as stated, are old Julien cells formerly 
used on street cars and, for some time before being assigned 
to their present work, occupied a conspicuous place in the 



scrap heap of the railway company. When it was decided 
to try them on the launches, each alternate plate was 
removed, increasing the distance between the remaining 
ones to three-eighths of an inch. This, of course, results in 
a much higher internal resistance, but, on the other hand, 
the cells do not give out quite so often, and the company 
preferred resistance, to investment, in new cells during the 
experimental stage. 


Fig. 21. — Cbockeb- Wheeler Electric Launch Motor. 

But the chief novelty is the method of charging. The 
street railway feeder, 50 yards away, is simply tapped and 
a line run down to what has been called the "charging 
station," but which is in reality a small box nailed to a 
post and containing a cut-out switch, an ammeter and a 50- 
ampere fuse (Figs. 20 and 22). The railway takes a current 
with a potential of 500 volts and this pressure is reduced 
for the launches by an amusingly simple process. After 
passing the switch, fuse and ammeter in the box, and the 
batteries in the boat, the circuit leads to an old iron pot 


sunk in the mud at the end of the pier. A section of iron 
pipe is stuck loosely in the dry sand above high water mark 
and this is connected to the railway return circuit. It will 
thus be seen that to vary the potential it is only necessary 
to push the iron pipe farther down into moist sand or pull 
it up a little where it is drier. The batteries are charged 
with a current of 15 amperes at about 140 volts. 

Mr. A. S. Hickley, general manager of the Seashore 
Electric Railway Co., with whom the plan originated, is 
much pleased at the result of his first season's work, and, 
as stated, intends improving it this summer and putting it 
upon a really commercial footing. The fact that with such 
apparatus — all old except the motors — ^two old boats could 
be run for three months and make money without charging 
exorbitant fares, certainly shows that thought, work and 
capital expended in perfecting such a system can be well 

24. So crude a method as this is here described and illus- 
trated chiefly with the idea of showing that the combina- 
tion of storage launch work with electric railway work is a 
new and promising field, and deserves attention. At the 
present time many electric railway companies in this coun- 
try are interested in suburban property and in rural 
pleasure grounds at the outer end of the roads. Several 
of the companies have gone so far as to improve existing 
pools and streams and have even laid out extensive sheets 
of artificial lake. Hence the opportunity already exists, 
but the chief difficulty of the work proposed appears to 
lie in the high voltage of the railway circuit — ^f rom 400 to 
500 volts. It does not seem feasible to install in small 
launches for casual, or even for regular, use a number of 
cells sufficient to represent that voltage, even if two or 
three boats were charged with all the cells in series. But 
it is obvious that an easy way to meet the problem is to 
Install a dynamotor, or motor-generator, at the charg- 
ing point. By this means, the voltage of the railway cir- 
cuit can be transformed down to that of the groups of 
cells, and a very successful use made of the railway cur- 
rent whenever it may be needed, either day or night. If 
desired, the motor-generator could be put on board a boat, 


SO that wherever it went it would obtain current from any 
railway circuit passing the shores of the waters it hap- 
pened to be navigating. 

Up to the present time, no demand for current has been 
of sufficient importance to make a special study of the 
question desirable in America, but the time has arrived 
when the increase in the number of electric launches 
depends on the frequency and accessibility of charging 
stations. A great many hotels along the sea shores, the 
rivers and the lakes have incandescent lighting plants, but 
as a general rule no arrangement has been made by any 
of them for running a tap circuit to a dock or landing 
where launches may be charged. Yet it would seem that 
money could be earned by the maintenance at most of 
these resorts, of one or two launches to be hired by the 
guests at a reasonable rate or for ferriage purposes ; while 
in course of time the sale of current to visiting launches 
sailed by private owners might also be made a source of 
income. On the Thames, the General Electric Power and 
Traction Company has a tariflE for the sale of its current to 
private launches. This tariJBf is based on a day's run and 
is estimated by the carrying capacity of the craft. Thus 
for a boat capable of carrying 30 passengers the charge has 
been about $4 (17 shillings) ; for 24 passengers, $3.75 (15 
shillings) ; for 15, $3.25 (13 shillings and 6 pence) ; and for 
10 passengers, $2.75 (11 shillings and 6 pence). These rates 
are fair and remunerative. 

25. Leaving aside the isolated plants, it is safe to say that 
no large stream in this country near a town or city is far 
from a central station. As it is the business of any central 
to develop the sale of current, it is evident that the field of 
storage launch work is worthy the attention of local plant 
managers, either in the way of charging cells or in carry- 
ing leads to the water's edge, so that boats may come 
alongside to obtain supplies of current. This was begun 
last year in Boston by the local Edison Co., whose station 
backs on the Harbor. At Newport this year, the electric 
launches of Messrs. Astor and Vanderbilt will be charged 
from the local lighting station. 

Where launches are attached to steam yachts or steam- 


ships it is easy to charge them hanging in the davits while 
the plant on board is running, and even to use the storage 
launch as a means of current supply during the day, if the 
demand is not large. If the launch is a big boat and no 
outside supply of current is obtainable, it should be pos- 
sible to have a small charging plant on board, driven by an 
oil or naphtha engine, and to charge during the night or 
while at anchor. This arrangement is not complicated or 
expensive and might often prove advantageous. Mr. Astor 
has such a plan under consideration. 

26. The Milwaukee, Wis., Electric Launch Company was 
formed recently with the intention of operating on the Mil- 
waukee River, which is dammed at the city limits. Above 
that point, for a distance of about two miles, range a 
number of summer resorts, reached hitherto by steam 
launches. This new company has bought a fleet of electric 
launches to be run both on pleasure trips and as a regular 
passenger line, taking one hour to make the round trip. 
The electric street car line ends at the dam, and it is the 
intention, this summer, to charge the launches from the 
600- volt circuit by putting three launches in series, with 
the object of avoiding any loss by resistance. The boats 
are already very successful there. 

Another company which has entered the same field is the 
Altoona (Pa.) and Logan Valley Electric Railway Com- 
pany, of Altoona, which has an artificial water of 13 acres 
in Lakemont Park, at the end of its road, and upon which 
it has placed a 26-foot General Electric launch. With this 
boat, passengers are carried for 10 cents, the round trip 
lasting 7 minutes. The current in this instance also is 
derived from the street railway circuit. A group of ger- 
man silver coils has been hung upon poles in the open air, 
for the purpose, but it is understood that this is only 

In Boston, the necessity of making better provision for 
the crowds seeking to obtain recreation and amusement on 
the waters in the public parks, has led to the acceptance 
by the Park Commissioners of a proposition under which a 
fleet of no fewer than 260 boats of all sizes, rigs and origin 
will be put afloat, in charge of Gen. C. H. Barney, who 


managed the electric fleet at the World's Fair. The boats 
will be distributed upon the waters of some eight parks, 
and will be either handled by attendants, or hired inde- 
pendently. It is intended to make electric launches a 
special feature of the work. As a matter of fact this is 
not the first time that public park authorities have taken 
up the subject, as a movement in this direction was made 
as long ago as 1892, when the Southport corporation, Eng- 
land, started a special carvel built electric launch, with 
remarkable success. This boat, "The Bonnie Southport," 
will seat 40 persons, and is 38 feet 5 inches long, 7 feet 6 
inches beam, and draws only 2 feet 3 inches of water. She 
is driven by a 6 horse-power motor. 

New Haven, Conn., is also the scene of the utilization of 
electric launches in connection with the trolley system, and 
similar work has been begun at Rochester and Buffalo. 
About three or four miles from New Haven is a very pretty 
sheet of water called Lake Saltonstall, to which the trolley 
road runs. Mr. Ot. H. Townsend, the owner, has begun the 
equipment of a fleet for the purpose of carrying passengers 
around the Lake, which is about 3i miles long. Twenty 
cents is charged for the round trip. The current is received 
by the batteries through a resistance, consisting of a tank 
of salt water in which two large iron plates are immersed. 
The resistance is regulated by varying the distance between 
the plates in the salt water solution, a handle being attached 
to them for the purpose. 

An interesting experiment is, it is said, soon to be tried 
on the Gulf of Mexico, with a line of electric launches ply- 
ing between a number of towns and ports along the coast, 
and carrying both passengers and freight. The boats 
required for such service are necessarily small, and hence 
owing to the heat of the climate, steamboats are not in 
favor. Several electric light stations exist along the coast 
and from one or more of these, the launches will obtain 
their supply of current. Moreover, the boats are easy to 
handle and the fact that no part of the deck space is occu- 
pied by machinery enables better provision to be made for 
passenger service, as well as allows larger space for a 
variety of small freight. 

Another method of charging that has been resorted to is 


that in use by the Earl of Aberdeen, Gtovemor General of 
Canada, for his electric launch at Ottawa. The method 
employed there, very successfully, under the supervision of 
Mr. Osmond Higman, is that of taking the current from a 
250 volt motor circuit, owned by the Standard Electric Co., 
a rheostat being used to reduce the current to the proi)er 
proportions. The charging current is 40 amperes under a 
pressure of 80 volts. 

A novel use was made of an electric launch in 1890 by the 
London Daily Oraphic^ whose boat followed close behind 
the racing crews in the very fast Oxford and Cambridge 
boat race, and from which carrier pigeons, with sketches 
by special artists on board, were dispatched frequently 
while the struggle lasted. 




Special Electrical Craft — Rowboats, Catamarans, 
XKD Paddle Wheel Boats. 

27. From various statements already made, it will have 
been evident that the range of applicability of electrical 
boats is by no means small, and that a number of very dif- 
ferent types are possible. A few specific references to 
modifications or novelties will be helpful and instructive. 
The first example here shown is an electric rowboat, the 
design of Mr. J. C. Chamberlain. We have recently had the 
opportunity of testing her merits on the Harlem River, and 
are of the belief that in many respects she would prove very 
useful, wherever current is available. It is obvious that 
primary batteries can be used, but in this case the boat is 
fitted up with storage batteries, which are preferable for 
many reasons, and which will doubtless be used when pos- 
sible. We show the boat in Figs. 23 and 23a, and cannot 
do better than quote figures supplied by Mr. Chamberlain 
as to dimensions, cost, etc. ^e boat illustrated has the 
No. 2 equipment, being 17 feet long and 46 inches beam. 

Boat fitted with equipment 


Length ..... 

Beam ...... 

Draught ..... 

Freeboard with fullload 

Floating capacity (8 to 0-inch draught of hull) 


. pounds 














Speed— Miles per hour for ) 8-4 hours 

continuous run of f 6-7 " 
Speed rate for short spurt 

No. of boxes of batteries 
Tvpo of motor and oontooUer . 
Charging current— volts 

No. hours for full 
Cost per hour-> cents 
Totalweight, boat with equipment 
Price complete, boxed and crated, 

f . o. b. N. Y. dty. 


▼amished boat 







































Weight of equipment onlj 

Price of same, boxed, f . o. b. New York City 









It may be added tliat each boat has regulating and re- 
versing speed controller. She has no sail, but carries oars. 
The boat is specially desirable as a tender to larger craft, 
for hunting or fishing up shallow creeks or rivers, but is, of 







course, equally convenient for any casual use of business 
or pleasure afloat, where it is deemed advisable to supple- 
ment, or diminish, the labor of rowing. As an adjunct for 
fishing purposes, it is easy to drop overboard from this 
boat, in circuit with the batteries, a little incandescent 



lamp. It is well known that such a submerged light is an 
excellent lure for fish. 

28. We may next consider an ingenious and novel varia- 
tion in electrical boats, in the shape of a catamaran. The 
boat here shown (Pig. 24) was buUt and run very success- 
fully in 1888 by Mr. Louis S. Clarke, of Pittsburgh, Pa., 
who sailed her upon the Conemaugh Lake that broke loose 
and so terribly devastated Johnstown lying below. This 
boat was built on the catamaran plan in the belief that less 
resistance might thus be offered to passage through the 
water, and that she might prove steadier than another boat 
of equal capacity of the ordinary build. The twin hulls 

Pig. 24. — ^Electbio Catamaran oit Lake Conemaugh, Pa. 

were made of galvanized sheet iron. They were 22 feet 
long and 14 inches in diameter, and displaced about 2,600 
pounds of water. A light platform 6 by 7 feet was con- 
structed on the hulls, which were separated a distance of 
4i feet. On this platform stood the storage battery box, 
which served as a seat and contained 26 cells of storage 
battery weighing 400 pounds. They were charged from an 
8-light dynamo on Mr. Clarke's steam launch. The motor 
to drive the catamaran was placed forward in a small box. 
The screw shaft stood at an angle of about 8 degrees from 
the water and ran clear to the stem. The little motor was 
of Mr. Clarke's own design, having a Grramme ring arma- 
ture, developing about six-tenths of a horse-power and 



driving a 10-inch screw at about 460 revolutions per minute. 
The motions of starting, stopping, reversing and regulat- 
ing the speed were governed by two small switches handily 
placed, and the steering was done by a small tiller in front 
of the seat. At night a little search light was used, or the 
craft could be lit up by one or two 60-volt incandescent 
lamps. The " Spark,'' as she was called, made a speed of 
about four miles an hour. 

29. While it is not, perhaps, a distinct type of boat, the 
"paddle wheel" form deserves special mention, in spite of 

Fig. 25. — ^TBOUvii Paddle- Wheel Boat. 

the fact that so few examples are known. We illustrate 
here a " stem- wheeler " built by the ingenious M. Trouv6, 
who has supplied more than one such boat for use on rivers 
or waters where aquatic plants and weeds are abundant 
and might clog a screw. Fig. 26 is a perspective view of 
the boat, and Fig. 26 shows the method of mounting which 
enables the wheels to operate together or singly, and re- 

Quite a novel piece of apparatus was constructed a few 
years ago at the works of the Freeman Electric Manufac- 
turing Company, of New York City. It consisted of a boat 



mounted upon four wheels, and provided with an electrical 
equipment for its propulsion. The apparatus comprised a 
reciprocating motor which connected with the axle upon 
which the wheels were mounted. The motor itself weighed 
87 pounds and the batteries 64 pounds, consisting of 20 
storage cells. The vehicle, it was said, could be run at 10 
miles an hour. Mrs. P. A. Truax, the designer of this 
apparatus, had in mind not only a vehicle for land travel, 
but one which also could be used for water. For this pur- 
pose the body was in the shape of a boat, and the wheels 

Fig. 26, — ^Tbouvb Method of Connecting up the 
Paddle- Wheels. 

attached to the motor were also provided with paddles. 
Thus, upon entering the water, the apparatus was at once 
transformed into a boat with paddle-wheels. The wood- 
work of the boat was remarkably light, weighing about 33 
pounds. The wheels weighed 24 pounds each, and the boat 
itself, with a load of 1,300 pounds, drew but 5 inches of 
water. No records as to the performance of this curious 
craft have been obtainable. 



Submarine Electric Torpedo Boats. 

30. We have thus far been dealing with boats to navi- 
gate the surface of the water ; but electricity has also been 
extensively tried in the propulsion and manoBuvering of 
submergible boats intended chiefly, if not, indeed, wholly, 
for purposes of warfare. These boats constitute a distinct 
and important class. Without going into the history of 
submarine boats in general, or touching on the work of 
Van Drebble, Bushnell and Pulton, it may be stated that 
in 1887 the U. S. Navy issued a circular showing the 
requirements to be broadly fulfilled in the design and trial 
of a proposed steel submarine torpedo boat for the Ameri- 
can navy. Some plans were actually submitted; but, so 
far as is known, the navy still remains without any such 
boat, whether electrical or otherwise. The main features 
in the stipulated requirements were that the boat should 
be able to make a speed of at least 15 knots an hour when 
running on the surface; 12 knots when "covered" by 3 
feet of water ; and 8 knots when submerged and offering 
no view of the object of attack other than one through 
water. She was also to be able to run for about 30 hours 
at full power, on the surface or covered, without detriment 
to her power for use under water; and when submerged 
she was to be capable of running at least 2 hours at 8 
knots mean speed. If intended for "covered" and "sub- 
merged" work only, without using air draught, she was to 
be capable of running in that condition about 30 hours at 
full power. She was to be able to turn in a circle of a dia- 
meter not greater than four times her length, without 
reversing engines, and was to be able to pass from the sur- 
face to the plane below in 30 seconds. The shell of the 
boat was to be strong enough to withstand an external 
water pressure due to a submergence of at least 160 feet. 





Against the bottom of a ship, 
running at speed, she was 
to deliver with reasonable 
certainty torpedoes carrying 
charges equal in minimum 
effect to 100 pounds of gun 
cotton. Besides this, she was 
to have means of obtaining 
an all-round view ; of purify- 
ing the air for the crew so as 
to allow of at least 12 hours' 
submersion ; of keeping the 
temperature within the boat 
down to 100 degrees Fah. ; 
and of getting away from ob- 
structions, lighting the in- 
terior, etc. The Navy Depart- 
ment limited the maximum 
displacement to 200 tons, 
when the vessel was sub- 
merged, but thought that 
about 90 tons would give the 
best results. 

Before passing on to note 
what has been done abroad, 
in this field of work, mention 
is in order of the paper read 
by Lieut. Hovgaard, of the 
Royal Danish Navy, in 1888, 
before the British Institution 
of Naval Architects, on the 
' ' diving boat ' ' for which bids 
had thus been invited by the 
American government. His 
boat then described is shown 
in Figs. 27 and 28. She has a 
fish shaped hull, 122 feet long 
and 12 feet beam, with a low 
superstructure surmounted 
by two conning towers, li feet 
high. Her displacement at 







light draught was intended to be 171 tons ; and, submerged, 
196 tons. The ordinary power of the boat was derivable 
from vertical compound, surface condensing engines cap- 
able of developing 600 i. h. p. ; but during submersion she 
was to be driven by an electric motor developing 35 i. h. p. 
taking current from a storage battery capable of giving a 
speed of nearly 6 knots for five consecutive hours. Of stor- 
age cells she was to carry no fewer than 540, of the lead grid 
type, each weighing 79 pounds and capable of delivering 35 
amperes for 6 hours, the average electromotive force of each 
being 2 volts. The major portion of these cells — 490 — ^were 
to be placed in a separate battery room. AH movements 
up and down were to be brought about by a small propeller 
placed near the centre of buoyancy in a vertical well going 
right through the vessel, and driven by a 6 horse-power 
motor also operated by storage battery current. Provision 
was also made for electric lighting, and for pumping out 
the water ballast by electric motor in about 30 minutes. 
The shafting of the propeller was so arranged as to shift 
over, when needed, from the steam engines to the electric 

31. Turning to what has actually been attempted in 
Europe, the first electrical boat we will describe is that 
designed and experimented with in 1888, and since, by Mr. 
J. F. Waddington, of Seacombe, near Liverpool, England. 

The length of this vessel (shown in Pigs. 29 and 30) is 37 
feet over all, and her diameter amidships is 6 feet, tapering 
in a curve at each end. On the top of the vessel is the 
conning tower a, provided with ports at the sides and a 
water tight scuttle | on top for entering the interior. The 
vessel is divided by the bulkheads bb into three compart- 
ments, of which the two end ones co are used for storing 
the supply of compressed air, while the centre one is occu- 
pied by the driving, pumping and steering apparatus. 
The electricity used for working the motor is stored in 45 
large accumulator cells d, which have a capacity of 660 
ampere hours each, and are connected in series to an elec- 
tric motor E, which drives the propeller F direct at about 
750 revolutions per minute. The motor when working at 
full speed uses a current of about 66 amperes and 90 volts, 














developing 7.96 electrical horse-power, which would drive 
the vessel for 10 hours at a rate of eight miles an hour, 
thereby enabling the vessel to go eighty miles at full speed 
without replenishing the accumulators, while at half speed 
she would travel 110, and at slow speed 150 miles. The 
great advantages the inventor claimed in using electricity 
are that when once charged the vessel is ready to start at a 
moment's notice ; that when stopped, the motive power is 
not subject to waste, as in the case of steam power ; and 
that no heat or poisonous gases are given off to pollute the 
air in the vessel, which is an important feature in sub- 
marine vessels ; also that a small sized class of these ves- 
sels can be carried on a man-of-war's davits, ready for 
instant use, the same as any ordinary launch. All the 
levers for manoeuvering the boat are arranged amidships, 
within easy reach of the steersman, so that he can control 
the movements of the vessel, while looking out of the con- 
ning tower. A second man accompanies the steersman, to 
assist in case of emergency, so that the crew of the boat 
consists of two men only. On either side of the vessel 
large water ballast tanks g are provided, by filling which 
the buoyancy of the vessel may be reduced by several 
hundredweight when preparing to dive. At the after end 
are arranged two vertical rudders hh and two horizontal 
ones II, the latter ones being used for keeping the vessel on 
an ''even keel" when moving under water. It was found 
necessary to make the rudders ii act automatically, as a 
man might lose his presence of mind when most required. 
The inventor at first provided a pendulum suspended from 
the top of the vessel and connected to the rudders ii ; but 
it did not answer readily enough to the action of the vessel, 
and he has since adopted an electric motor k working into 
worm gearing, and arranged in such a manner that the 
slightest cant up or down causes the rudders to be put over 
to meet it, and then stop in the same manner as a steam 
steering gear. At each side of the vessel balanced side 
planes ll are provided, which may be inclined at different 
angles by the lever m. Close to the bulkheads at each end 
of the compartments cc a vertical tube, open at each end, 
extends right through the vessel so as to leave a clear 
course for the water from the vertical propellers nn, which 


are worked by separate motors, and may be used separately 
or together, as the occasion may require. At the bottom 
of the vessel a heavy weight is attached in such a manner 
that it can be released from the inside in case of an emer- 
gency, such as a sudden leak, etc. The mode of working 
the vessel is as follows : The water tanks aa are filled so 
that only the conning tower is above water, represent- 
ing only several hundredweights of buoyancy, the motor 
started, and a speed of say four to five miles an hour 
attained. Then the side planes incline by means of the lever 
M, and the vessel is thereby submerged. By regulating 
the angle of the side planes and the speed, any reasonable 
depth may be maintained. If it be desirable to submerge 
the vessel without forward motion, only the vertical pro- 
pellers are used, and any reasonable depth may be main- 
tained by regulating their speed. These vertical propellers 
may also be used for regulating the depth when the vessel 
is moving onward, but the side planes are considered pre- 
ferable in that case. The fresh air is drawn from the two 
end compartments cc, and the bad air escapes through a 
valve as soon as the pressure inside becomes greater than 
that outside. But the compressed air need only be used 
when staying below a lengthened period, as the centre 
compartment contains air sufficient for two men for six 
hours without drawing upon the supply of compressed air. 
The vessel carries two automobile torpedoes o, one on each 
side, secured by grips, which can be opened from the in- 
side of the vessel. By releasing these grips the propelling 
motor of the torpedo is started, and it shoots ahead of the 
submarine vessel. A mine torpedo p, is also provided for 
attacking any vessel at anchor with the torpedo nettings 
down, and which can be fired from a distance by an elec- 
tric wire paid out from the submarine vessel. The dotted 
lines indicate portable guard rails and stanchions, which 
are put in place when the boat is above water — a very 
necessary precaution in a seaway. 

32. Considerable notoriety was achieved in 1888 by the 
electric submarine boat ''Peral," named after her inventor, 
Lieut. Peral of the Spanish navy. She is shown in Pigs. 
32 and 33. Her lines resembles those of the famous White- 




head torpedo, but with an ogival cross-section. Her length 
was 72 feet ; maximum diameter 9 feet 6 inches ; displace- 
ment, 86 tons : draught on water when operating at the 
surface, 2 feet 11 inches. She was equipped with two pro- 
pellers, and with an Immisch electric motor driving each 
one ; while the pumps were also worked by electric motor. 
She carried 613 cells of Julien storage battery, of which 
126 were allotted to each propeller motor and 100 to the 
pump motors. The other cells were utilized for a pro- 
jector, for inside lighting and for various other features 
about which the inventor and the Spanish officials were 
very reticent. Her nominal speed at the surface was 11 
knots, and when submerged lOi knots. Besides her two 
screws for propulsion, the '^ Peral" had two others for im- 
mersion. If any accident occurred to the motors and the 
immersion screws were stopped, the boat rose at once to 
the surface without further aid, though, of course, the 
emptying of the water compartments hastened the ascen- 
sion. To avoid an undue expenditure of power in connec- 
tion with the immersion screws, the water compartments 
were filled to an amount which enabled a very slight 
motion of the screws to sink the boat to the required depth 
and maintain it there. The automatic apparatus which 
regulated the depth at which the boat was to work was 
designed on a principle somewhat similar to that of the 
aneroid barometer. A curved tube of elliptical section was 
placed in connection with the sea, and its deformations due 
to the alterations of pressure actuated a switch, by which 
the strength of the current going to the immersion screws 
was varied. The positions of the contacts of the swit<5h 
were altered to suit the particular depth at which it is 
required to work. A very sensitive automatic electrical 
device was also employed to keep the vessel in a horizontal 
position. The apparatus consisted of a pendulum playing 
between two contacts. If the boat was not perfectly level 
from stem to stern, the pendulum touched one of the two 
contacts, and the result was that the corresponding vertical 
screw was actuated and the boat was righted^ Sundry 
trials were made with this boat, but the silence in which 
she is now submerged would indicate that she was not an 
ultimate success. 




















33. Another electric submarine 
boat of considerable fame is the 
"Gymnote" (Figs. 34 and 35), 
which was designed by M. Zed6, 
the engineer-in-chief of the Forges 
et Chantiers de la Mediterranee, 
whose plans were accepted for the 
French government by Admiral 
Aube. She was launched in Sep- 
tember, 1888. Her shape is that 
of the Whitehead torpedo. Her 
length is 59 feet, greatest diameter 
6 feet 11 inches ; displacement 29.5 
tons. She has two horizontal rud- 
ders worked by hydrostatic pres- 
sure or at will, and two vertical 
rudders operated by the usual 
appliances. The motive power is 
electricity. The motor is of the 
multipolar type, designed by Capt. 
Krebs, well known from his work 
in ferial navigation, and weighs 
4,400 pounds. It drives a 4-blade 
propeller, 4 feet 10 inches in di- 
ameter, connected directly to the 
armature shaft, and revolving at 
200 turns per minute. The esti- 
mated speed is 10 knots for six 
hours, current being taken from a 
battery of 564 Desmazures alka- 
line accumulators, each cell weigh- 
ing 38.5 pounds ; or a total weight 
of 9.66 tons. The boat is also 
steered and lighted by current 
from the cells. On one trial made 
in 1888, the boat dived 23 feet and 
traveled 1,600 feet at a speed of 
about 4 knots per hour, everything 
working well, although certain 
improvements at once suggested 



The Frencli Minister of Marine reported in 1891 that the 
" Gymnote " had ''solved in a satisfactory manner the pro- 
blem of submarine navigation, that is to say, the running 








of a direct course under water, toward any determined 
point. This was proved last year at Toulon, when the 
^Gymnote' ran outside of the harbor and directed her 



course under the water towards the 'Couroime' without 
deviating from the straight line for an instant." The 
Minister remarked, however, that such a boat must have 
weapons to fight with, and that these, with improvements. 








were to be furnished and embodied in a new boat called 
the "Gustave Zede," to cost $223,800. 

34. Another French submarine boat, of which it is said 
that the Russian navy has adopted it "provisionally," is 
the "Goubet " (Figs. 36 and 37), built on the lines of the 


ordinary fish torpedo. She is about 18 feet long, 6 feet in 
diameter, and has a total weight of 7 tons. Her hull is of 
cannon bronze. During the first trial made of her at Cher- 
bourg, in 1889, to test her system of aeration, the two men 
who constituted her crew were hermetically sealed in the 
boat for 8 hours at a depth of 33 feet. At the end of the 
experiment, the men when they came up to get some coffee, 
after their dinner, were perfectly comfortable and it was 
found that there remained sufficient oxygen to have pro- 
longed the experiment for 25 hours. The oxygen supplied 
for breathing was compressed at a pressure of 70 atmos- 
pheres in steel tubes 3.93 feet long, 4.72 inches in dia- 
meter, weighing 66 pounds. Minor advantages alleged to 
exist in the case of the " Goubet" are the possible substi- 
tution or supplementing of the electric power by ''subma- 
rine oars ; " the provision by means of a heavy mass of lead 
attached to the keel and detachable at the touch of a but- 
ton, for the instant coming to the surface, in the case of 
any accident ; and the ease of steering, which is done by 
means of the driving propeller. The report of the French 
authorities has not been at all favorable as to the capabili- 
ties of this boat for actual work. 

35. In 1888, the French government commissioned the 
Compagnie Generale des Bateaux Parisiens to build a sub- 
marine boat to be used in destroying submarine mines. 
Her greatest length is stated at 14.9 feet ; her diameter at 5 
feet 4 inches ; her complement, two men. She is driven by 
an electric motor taking current from a primary battery 
and is lighted by electricity. No reports of her perform- 
ance have fallen under the writer's eye, and this remark 
applies equally to some of the submarine boats built by 
Russia. There is no desire on the part of the various 
governments to make public their tests in this direction, 
whether success or failure attend them ; and hence the diffi- 
culty in furnishing even the simplest details of construc- 
tion. It is evident, however, that any great naval war would 
bring to light what has been done secretly in this field. 

36. One of the most interesting submarine electrical 
boats is that designed by Mr. George C. Baker, of Chicago, 













and tried at Detroit, Mich., in 1892. She is shown in 
section in Fig. 38, and afloat in Fig. 39, She is built of 
wood, and has about 75 tons displacement, divided as 
follows : — Hull, 20 tons ; ballast, 30 tons ; storage battery- 
cells, 10 tons ; engine, boiler and gearing, 8 tons ; and motor, 
3 tons ; leaving 4 tons buoyancy. At normal draught, 
about 2 feet of the crown of the hull remains above water. 
The shell proper is six inches thick with a sheathing an 
inch thick. It is built of strips of 3-inch oak, 6 inches 
wide, nailed flatside together, with 9-inch spikes. The 
boat is braced horizontally across its centre line by eight 

Fig. 39. — Appeabance of Bakes Boat when Ready for 
Going Undeb Water. 

6 by 6-inch oak beams. The interior of the shell at the 
centre is 13 feet deep by 8 feet wide, and the outside length 
of the boat is 40 feet. The arrangement of the parts and 
mechanism is shown in the diagram. It is the plan of the 
designer to employ the steam plant to drive the dynamo 
and thus charge the storage battery. When the charging 
has taken place at any convenient point, the smokestack is 
drawn down into the boat and the fire is extinguished by 
closing the air-tight furnace doors. The cells may then be 
turned on again to the circuit, but this time to run the 
dynamo reversed, as a motor. The cap of the smokestack 
is a valve, which when the stack has been pulled in closes 
the only opening to the boat other than the manhole. 
















The pilot stands on an elevated platform in the centre of 
the boat, his head being surrounded by the dome turret 
with five plate glass windows, four at the sides and one on 
top. Entrance and exit are made through the turret as a 
manhole. The turret cap is provided on its under edge 
with a gasket of rubber tubing. To seal the vessel, the lid 
has only to be swung around on the stud at the side, which 
forms its pivot or hinge ; and then by turning a nut on the 
lower and inner end of the stud, it is let down and screwed 
tightly into place. 

The side screws are so geared as to regulate the depth of 
submersion as well as to propel the boat. Mr. Baker's 
theory in regard to such boats has been that a boat should 
be forced under water by her screws rather than be sunk 
entirely by means of an added weight, such as water drawn 
into reservoirs on board. In other words, by the intro- 
duction of water into reservoirs, in addition to ballast 
already on board, the vessel's buoyancy could, at will, be 
reduced to a minimum ; and when thus suspended or bal- 
anced in the water, the screws can be run at the proper 
angle to force the boat under. By driving the screws on 
an angle, the boat could thus be navigated at any reason- 
able desired depth. Another theory relates to the proper 
location of the propellers. Mr. Baker's belief is that the 
best plan is to employ two screws, each so arranged that it 
can be set from within the boat to revolve at any angle in 
a plane parallel with the vertical centre plane of the boat. 
These screws are therefore connected so that they may 
propel the boat backward or forward, on the surface or 
below, force it down for complete submersion or bring it to 
the surface. By placing the propellers also at the point of 
the boat's centre of gravity, Mr. Baker has sought to 
secure greater stability and to maintain the craft, under 
all circumstances, with its keel parallel to the surface of 
the water. 

The equipment of the boat deserves brief mention in 
detail. The Willard engine is of the marine type, rever- 
sible, with links, etc., of nominal 35 horse-power. The 
Roberts boiler is rated at 60 horse-power, oi the ordinary 
marine pipe type, and tested up to 220 pounds pressure. 
In connection therewith is a Worthington duplex pump 











4J by 2i by 4 inclies, employed to feed the boiler and to 
draw the water ballast out of the reservoirs shown in the 
keel. No filling of the reservoirs by pump is necessary, as 
the water runs in of its own accord. 

The electrical plant comprises a 50 horse-power Jenney 
motor and 232 Woodward storage cells. The motor is built 
for an electromotive force of 200 volts and runs at a maxi- 
mum speed of 900 revolutions. It is geared to the two pro- 
pelling screws, which are four bladed, and 24 inches in 
diameter from tip to tip of blade ; and the screws run at 
300 revolutions, which gives an estimated speed of about 8 
or 9 miles an hour. When run by the engine as a genera- 
tor of current, the dynamo works at 1,025 revolutions per 
minute, and at a charging pressure of 220 volts. The cells 
are grouped in four sets of 58 each, and are discharged in 
two sets of 116 cells each. At the top of the boat is a con- 
venient controlling switch connected with galvanized sheet 
iron resistance coils placed in the forward end of the boat. 
By means of this switch and the circuit breaker, the appa- 
ratus is under control and variations in speed are obtained. 
The pumps as well as screws can be geared to the motor. 

The boat carries ordinarily two men, who have remained 
in her 1 hour and 45 minutes without inconvenience. 
One of these men acts as pilot ; the other as his assistant 
and relief. There are two wheels to handle ; one for the 
rudder, the other for changing the angle of the screws. 
Mr. F. L. Perry, of the Western Electrician^ in the course 
of a most interesting article in that journal,^ reports 
spending 35 minutes in the boat, closed up, and chiefly 
under water ; and mentions a test on May 24, 1892, when 
Mr. Baker and his assistant were sealed in the boat, partly 
on the surface and partly below it, 2 hours and 44 minutes. 
The official ^' Notes on the Year's Naval Progress," for 
1892, issued by the TJ. S. Navy Department, in speaking of 
this boat says : ^' She was frequently submerged, retaining 
an even keel below the surface, and answering readily to 
the requirements of the pilot. * * * * It is the 
opinion of the Bureau that the problem of submarine navi- 
gation and attack is approaching a solution and will play 
an important part in naval defensive warfare of the future." 

1. WegUm Electrician^ June 4, 1892. 


In April, 1892, the submarine electric boat ^'Audace" 
was launched by Migliardi Brothers, of Odnito-Vene, at 
Foce, Italy, to the order of the Roman Company, for fish- 
ing and the recovery of treasure, etc., from the bottom of 
the sea. She measures 8.50 metres long, 8.50 metres in 
height, and is 2.16 metres maximum beam. She is divided 
into compartments and is built entirely of steel, having an 
ovoid form in a transverse direction. She is driven by an 
electric motor, actuating a screw propeller. It is stated 
that she can descend more than 300 feet ; but the manner in 
which she does this, as well as the various other features 
of her operation have been kept a profound secret. The 
boat will accommodate four or five passengers, and can 
remain under water with them for six consecutive hours. 
She has one of her compartments fitted with a door in such 
wise that through it divers may carry on work in recover- 
ing treasure, repairing hulls, fishing for pearl oysters, and 
other like pursuits. 



Dirigible Electric ^'Torpedoes" for Warfare anb 


37. While they cannot be classed strictly as boats, elec- 
tric dirigible torpedoes are not to be overlooked in a gen- 
eral treatment of the subject of electrical navigation, as 
they are, practically, electric craft without a crew on board. 
The advantages of being able to control these torpedoes 
from the shore or shipboard, are obvious. Some idea of the 
havoc possible with floating torpedoes may be formed from 
the fact that toward the close of our Civil War, in 1865, in 
Mobile Bay, within the brief space of two weeks, no fewer 
than five Federal gunboats, two of them heavy double-tur- 
reted monitors, were totally destroyed by coming in contact 
with buoyant torpedoes ; and that a large launch was also 
blown to pieces with its crew. If such effects were possible 
with uncontrolled torpedoes, it would seem reasonable to 
infer that infinitely greater injury could be inflicted on the 
largest ships of an enemy if the torpedoes were controlled 
or dirigible, and were also automobile. Such types are now 
extant and successful. The earlier forms were electrical 
only in the sense of having their steering mechanisms gov- 
erned electrically. The Lay-Haight torpedo has its rudder 
actuated by an electrically controlled gas engine, wires 
running from the operating station to the torpedo through 
an insulated cable. The Patrick torpedo is also governed 
in its actions by electricity, by means of a two-wire cable 
used with 80 Bunsen cells in series. It is our object, how- 
ever, to deal with the types in which electricity is also the 
motive power. That which is illustrated in Figs. 40 and 41 
is known as the Sims-Edison. The engraving shows the 
torpedo in section. It consists of a cylindrical hull of cop- 
per, with conical ends, and is supplied with a small screw 
and a rudder. The hull, carrying the dynamite section, 








controlling cable, electric motor, and steering gear, is 
supported at a submerged depth by an indestructible iloat 

attached to the hull by an upright 

steel stanchion. As the boat moves, 

only the top shell of this float is 

visible, carrying little rods showing 

signal flags or balls. This float, 

against which the Gatling and Hotch- 

kiss guns of the enemy would be 

. vigorously directed, is said to be 

g entirely impenetrable. The hull and 

I § float are protected from cables, ropes, 

^ or other obstructions by a sharp steel 

f » blade set at such an angle as to make 

B the boat dive under or cut away the 

S obstacle. 

Q The electric current from the 

2 dynamo on shore is conveyed to the 
I torpedo by a cable stored in one of 
S its sections, which is paid out as the 
^ torpedo proceeds on its errand. The 

operator from his station on shore or 
S on shipboard can at will start, stop, 
^ or steer the torpedo to port or star- 
J board and explode the charge, which 
^ can also be arranged to explode by 
^ contact if desired, and he receives 

1 notice when the hull or blade meet 
5 with any obstruction, together with 
^ the magnitude of the same, thus 
^ making sure of the proper moment 

for explosion. 

Steering is effected by a powerful 
electromagnet, into which is 
switched the main current by means 
of a polarized relay actuated by the 
current of the shore battery. Two 
keys control the relay in the boat, 

and the rudder is thus thrown from side to side. 
The diflEerent sections are connected together by gun 

metal rings with wedge-locking pieces ; in other words, 


they screw together with the help of large spanners, and 
the joints are made water tight with India rubber rings. 
The act of joining the sections makes electric communica- 
tion where necessary, by means of spring contacts, so that 
no jointing of wires is required. 

In No. 1 section is seen the charge, occupying the space 
from the nose as far as a supplementary bulkhead, through 
which the primer case is screwed into the centre of the 
charge. At this bulkhead there is a joint, which is, how- 
ever, seldom broken. The charge carried varies, but may 
be taken as about 500 pounds of gun cotton or other 
explosive, which is fired by means of an electric detonator, 
and not by percussion. No. 2 section is fitted with a lid, 

.-^:^, *_ r^ -^ 

Fig. 41, — Tbial of Sims-Edison Tobpbdo. — Launchbd fbom 


which takes right oflf, so as to allow of putting in the coil 
of cable, and a tube comes away from it under the torpedo, 
through which the cable is carried away clear of the 
propeller. This section is necessarily open to the water. 
The cable, of which about 7,000 feet is carried, is wound 
closely on a spindle, layer over layer, and when this is done 
the spindle is withdrawn, leaving the coil held between 
two end plates which are held together by four rods. 
The inner plate is then drawn out, and the operation is 
repeated in such a manner that after the second winding, 
the cable on being drawn out from the centre, comes out 
straight, pliable and free of kinks. It is now placed in its 
receptacle, the leads in the end are connected through the 


after bulkhead, and the inner end is rove through the tube 
and connected when required to the dynamo lead through 
the switchboard. The cable, which is very flexible, is 
about i inch diameter over all, and contains two concentric 
conductors highly insulated. The outer conductor carries 
the dynamo current for working the motor, and the inner 
carries current from a secondary battery, for operating 
relays in connection with the steering gear. 

The after part of section 2 is bulkheaded off to form a 
watertight compartment, which contains a relay for send- 
ing, the motor current through to the charge. Section 3 
contains a two-pole series-wound motor of Edison make, 
which runs at about 1,600 revolutions per minute, and with 
a current of 25 amperes at 1,150 volts develops about 33 
horse-power. At the after end the shafting is geared 
down so as to give about 800 revolutions for the propeller. 
N"o. 4 section contains the shaft, which turns in ^'metaline" 
bearings, so that no oil is required in any part of the 
torpedo. A simple clutch connects this shaft with that in 
section 3, but they are insulated from one another by vul- 
canized fibre. This section also contains the steering 
mechanism, which consists of two electro-magnets for 
working the rudder one way or the other, and a relay for 
sending the main current from the motor to either of these 
electro-magnets on its road to the frame. On the top is 
carried the rudder, which, when not drawn over either 
way, is kept straight by the motion of the torpedo through 
the water. On the end of the shaft is keyed a gun metal 
right handed propeller 30 inches in. diameter, which for 
ship use is fitted with a guard to prevent fouling the cable 
after launching. The sloping stay at the bow is made per- 
fectly sharp so as to cut through obstructions, but failing 
to do this, the torpedo dives under, and when clear comes 
again to its former level. The vertical rods on the float are 
so constructed as to hinge back in such a case, and to 
spring up again when clear. The total weight of the 
torpedo ready for service may be taken as li tons, and the 
floating power is all in the float, which for service is filled 
with cotton, so that if riddled with shot it would still leave 
a sufficient margin of buoyancy. 

The current for operating the motor is produced by a 


continuous current dynamo, shunt wound, capable of pro- 
ducing a current of 32 amperes at 1,200 to 1,300 volts. The 
current for working the steering relay is taken from a 
small box of secondary cells, giving current at 50 volts. 
The torpedo is worked by sending in any given direction 
the dynamo current, which can be switched on or oflf as 
desired, and also reduced or increased, by means of a set of 
resistances in the shunt. When the charge is to be fired, 
the main current is reversed by means of a suitable switch. 
This acts on the relay in the after end of section 2 and the 
current goes through to the charge. Just before switch- 
ing over, the current is reduced by means of the resist- 
ances. For steering, the current from the secondary bat- 
tery is sent through in one direction or the other by means 
of a suitable switch, and so acts on the relay as to send the 
main current to either of the two electro-magnets which 
work the rudder. 

The return for both currents is by water through the 
frame of the torpedo. To avoid a chance of premature 
explosion, a safety plug is attached to the switch, which 
must be taken out before it can be reversed, and in later 
torpedoes there is an arrangement to prevent the circuit 
being completed through the primer until the motor has 
made a certain number of revolutions. Near the switch- 
board at the directing station are placed a voltmeter 
and ammeter, to show what current is going away to the 
torpedo, and they also indicate at once when the torpedo 
has met with any obstruction, owing to the sudden extra 
work thrown on the motor. When working the torpedo 
from a ship in motion, the cable is to be paid out from the 
vessel as well as from the torpedo as required, so that in 
no case will there be any drag on the cable. 

This torpedo has been adopted by the U. S. Army for 
coast defence. The army reception trial at Willet's Point 
in 1891, was made with a torpedo 31 feet long, 26 inches in 
diameter, carrying 2 miles of cable, and having a capacity 
for an explosive charge of 500 pounds. It was manoeuvred 
at will, attaining a speed of 20 knots an hour or two knots 
in excess of the contract, although it is said^ that in more 

1. Annual of the Offlce of Naval Intelligence. Gen. Inform. Series, No. XI., July, 1892, 
p. 188. 



recent tests, the same high rate had not been maintained. 
This shortcoming is attributed to poor insulation of the 
wires, a defect it should be easy to overcome. 

This torpedo has also attracted considerable attention in 
Europe ; and Fig. 42 illustrates some interesting experi- 
ments made in England to launch and manipulate it from a 
vessel under way as well as from a fixed point on shore. A 
tumbler frame fitted on rollers on an overhead track holds 
the torpedo, and the detaching is accomplished automati- 
cally when the torpedo is clear of the ship's side, it entering 
the water with an initial impetus which assists it to clear the 

Fig. 42. — Trial of Sims-Edison Tobpedo. — Launched from 


vessel. The final trials took place in Stokes Bay, off Ports- 
mouth, in 1892, when the *' Drudge " running at 4 knots an 
hour launched the torpedo quite successfully. The torpedo 
was then accurately manoeuvred from the ship and ran out 
7,000 feet of cable in 4 minutes and 10 seconds, or at the 
rate of 18.1 knots an hour. 

38. Another electric dirigible torpedo is the Nordenf elt, 
of which a successful trial in England was reported in 1888. 
It is cigar shaped, moves 8 feet below the surface, and has 
two floats indicating its position to the manipulator. Its 



^ — « 


length is 85 feet ; maximnm diameter 29 inches ; total 
weight ready for action, 6,200 pounds ; explosive charge 800 
to 600 pounds. It carries its own motive power, propelling 
and steering apparatus and cable. The motive power is 
furnished by 120 cells of storage battery, which will develop 
18 horse-power. The motor weighs 780 pounds and drives a 
screw at 1,100 revolutions per minute. The speed obtained 
is 14i knots. The steering is done by a balanced rudder, 
manipulated from the shore through a three-core cable, 
3,000 to 4,000 yards in length. When tested, the torpedo 
has run nearly 2 miles in three successive trips. Mr. Nor- 
denf elt had in hand more recently a similar torpedo, to carry 
180 storage cells, developing 84 horse-power, and giving a 
speed of 16 knots at 1,500 revolutions per minute ; but the 
writer has not seen any record of its performance. 

Fig. 43 shows details of the Nordenfelt, while Pig. 44 
contrasts the general design of the three types, the Sims- 
Edison, the Lay-Haight or Patrick and the Nordenfelt. 
In Pig. 43, A is the charge chamber ; c, c, o, c, the storage 
batteries; d, the cable chamber; e, the electric motor 
chamber ; f, the controlling instrument chamber ; g, the 
steering power chamber ; t, t, the fins ; 1 1 the fin plates ; 
N, K, the directing points, which contain electric lights 
for service in night operations. 

39. An interesting but apparently complicated torpedo 
of the dirigible type is that brought to general notice in 
1890 by Mr. Read Murphy, an Australian, and called by 
him the ^'Victoria." It is intended to be operated either 
from shore or from shipboard. In all essential respects 
the weapon is a Whitehead torpedo. It is of the same 
general shape and construction, and is propelled by com- 
pressed air in the same way. But to the Whitehead 
equipment there are added some new features of great 
importance by which it is controlled. The steering de- 
vices are found in both types of weapon, but in the shore 
torpedo there are provided means for stopping, starting, 
and exploding at the will of the operator. The shore tor- 
pedo is the larger and more important of the two. It is 24 
feet long and 21 inches in diameter at the largest part. 
The head carries the charge ; next comes the compressed 



air chamber ; behind this is a chamber in which are coiled 
1,200 yards of electric cable inclosing the insulated copper 
strands ; further aft again is a chamber in which are three 
electrically controlled spring motors, one for working the 
vertical rudder, the second for the air valve which controls 
the propelling engine and allows the cable to pay out from 
the torpedo when at full speed, and the third for explod- 
ing the torpedo and also for bringing it to the surface. It 
is through the agency of the cable and the motors that the 

Fig. 44. — The Sims-Edison, Lay-Haight and Nobdenpelt 
Types of Dibigible Tobpedoes. 

torpedo is controlled. The 1,200 yards coiled within it, 
however, do not represent its range. An additional quan- 
tity is coiled at the place where the operator is situated, 
and it is intended that this shall be drawn upon first. As 
the speed of the weapon is under control, it is seldom 
advisable to launch it at full velocity. The amount of 
power required to drive it at full speed for, say, half a 
mile, would proi)el it at several miles at half speed, and 
hence it is usually wiser to send it away at a moderate rate 
until within easy shooting distance, when full speed can 


be given if desired. During the first part of the time the 
torpedo would be dragging its cable behind it, not a very 
serious matter, since it only weighs 41.110 grammes per 
yard in the air, and when greased it will leave the torpedo 
22.312 in water. But when the air valve is opened wide to 
give full power, the coiled cable is released and pays out 
from the body of the torpedo, thus obviating the drag, and 
giving 1,200 yards free run. There is also a device pro- 
vided whereby, should the shore end of the cable become 
fouled, or offer too much resistance to the motion of the 
torpedo, the clip which holds it is tripped, and the cable 
within the torpedo is paid out. In the present case the 
course of this torpedo is intended to be shown by Holmes' 
compound, the gas from which is forced to the surface by 
the rush of water through a tube. 

For shore stations, the "Victoria" torpedo, instead of 
being launched in the usual manner, may be deposited 
with a buoy in a cage under water, it may be a mile or 
more, from the shore, and is there left until the enemy 
appears, when it is released. On being released, the buoy 
ascends a given distance, and the torpedo starts on ite 
journey, pulling the cable from the buoy as it would from 
land. As the buoy contains the cable that would be other- 
wise wound at the sending station, the torpedo has its run 
of 2i miles from the position of its cage, and is worked 
from that point exactly as it would be from land. 

An ordinary cable connects the torpedo cage with the 
operating station on shore. This cable contains the three 
strands already described, of an extra cross-section to carry 
the extra current of electricity ; and two extra strands, one 
of which enables the torpedo and its buoy to be released at 
pleasure, and when not required for this purpose is con- 
nected with an ordinary electric bell. The second strand is 
connected with electric cells, so that if the torpedo or the 
cable is interfered with by the enemy, or hurt by mis- 
chance, the water will connect the two strands so that the 
bell will ring, and the officer instantly be apprised. If, 
however, all goes right, he can, by a touch of the key, open 
the cage and liberate the torpedo, which will rise as des- 
cribed ; its engines can then be set in motion, and it can be 
controlled and steered at will. The feature of interest, 



electrically speaking, is that each spring motor in these 
torpedoes is connected with a small electric motor, by means 
of pulleys and clutches. No record of any use of the 
" Victoria " dirigible torpedoes is at hand. Their mechan- 
ism seems to be most unnecessarily complicated. 

40. We have thus far been considering dirigible craft of 
the torpedo character simply as weapons of naval warfare, 
but it is obvious that they constitute an admirable means 
of saving life at sea either as a projectile with life lines to 
aim at a wrecked or stranded ship or as an apparatus that 
can be directed towards an overturned boat or to persons 
struggling in the water and needing a buoy. Indeed an 
electric life boat of this description has been devised in 
England by Mr. J. Hibberd, whose plan is illustrated in 

Fig. 45. — HiBBEBD Life-Saving Dirigible Electbic 
Torpedo Float. 

Fig. 45. A, A are the air chambers ; b, the electric motor ; 
c, sand or a grapnel ; d, rod to open the bottom to release 
sand, etc.; f, an electric light ; o, o, guide lines, and h, 
communication line. This float can be sent from shore to 
ship or vice versa. The object of loading one of the spaces 
with sand is to keep the float under water till it reaches 
the object aimed at, when, the sand being dropped, the 
float rises, and can be secured. 

An electrically lighted life buoy has been invented 
lately by Capt. Melter, and some trials were lately made 
with it at Kiel on board the German war vessel '' Worth." 
The buoy was thrown overboard when the vessel was pro- 
ceeding at a speed of about 16 knots, and for about 12 


seconds it was lost in the eddy current caused by the twin 
screws of the vessel, but then reappeared. It is stated that 
the exi)eriments resulted so successfully that it is probable 
the new life buoy will be adopted generally in the Gterman 
navy, and there seems no doubt it will be found of great 
value at night time. It is evident that if such a buoy were 
dirigible, as it might easily be, its value would be incalcul- 
ably increased, not only for use on ships but for utilization 
by shore patrols. 



Some Genebal Considerations on Electric Launch 

41. In the course of the preceding chapters a great many 
features in the construction, operation and maintenance of 
electric launches and boats have been touched upon ; note 
has been made of many of their advantages and conveni- 
ences, and some suggestions have been offered of means 
for improving and developing such work as the service 
involves. We now supplement such information by two 
tables furnished by Mr. J. C. Chamberlain, E. E., giving 
data, in harmonious relation and of an interesting nature 
as to size, capacity, speed, cost, etc. The first category 
includes yacht tenders, for yachts that have a charging 
plant, and pleasure boats to be charged for regular 
stations : — 

Lensrth over all— ft. and in. 










5 2 

6 4H 

5 6 

6 2 


5 8 


6 9 


1 1 

1 2h 

1 4 

1 1 

1 8 



1 10 

1 11^ 



8 1 

8 8 

8 ?• 

Speed-Miles per) S^hrs. 
nourforoontln-v 6-7 " 








uousnmof ) 8-10" 

4I4 1 







Speed rate for tahort sport . 







No. of batteries stand, type 









Charging corrent— Tolts . 


















No. hours to recharge . 









Cost per hour to recharge 

batteries—cents . 









Seating capad^ 









Prices-Stand, type, planked 
with selected cedar, decks 

of mahogany, entire huU 
and flniwings handsomely 
polished andyamished . 









Same type of boat, painted 
hull, pine decks, calked. 

handsomely finished 











The table is, it is thought, the more valuable for th^ 
prices it contains, as it affords an idea of the utmost 
exi)enditure involved in buying and operating electric 
boats. The next table, from the same authority, deals 
with regular electric launches, running from 18 up to 70 
feet in length. 

^ i II «i5| i 


oS5 » 

o g 

^•«i § I gsJg I I 

»??S s 


^^ 2 li ^§53 

^fc-§ S 

6 £84*2 

55 '^ 

•2 5 

11 ss^Ss 


gt.^1 I d ss5| I 

>^<o2 8 

d SSiSs 

?«o^ 2 

d 8g5^ 

S5 '" 

i ^ 

?o2 s 



8 jd ss2? i I 

Note will be taken of the fact that the greatest length 
here provided for is 70 feet ; and the greatest number of 
cells, 264. These are figures far beyond the ordinary to-day, 
in electric launch work, but they were actually exceeded in 
the remodeled and enlarged "Electron," rebuilt on the 
Hudson, in 1890, by Mr. James Bigler, from 36^ feet to 76 


feet. Her original equipment of batteries was 200 cells, 
but in the rebuilt boat this number was increased to no 
fewer than 376. This boat had decks and a cabin, and 
was operated at Atlantic City, carrying passengers for 
fare from shore to sea, two or three miles, and back. But 
her use was attended with patent and other litigation, 
pending the settlement of which she has been more lately 
operated as a steamboat. Her return to the ranks of elec- 
trical boats is awaited with interest, and it is to be regretted 
that in this case as in so many others of late years, 
where storage batteries have been concerned, fruitless legal 
squabbles have been allowed to interfere with the legitimate 
advance of the art. 

42. It will interest many readers to know what are the 
elements entering into the design and construction of such 
boats, and a few points dealing with such topics are now 
given, with the assistance of Mr. F. Reckenzaun, who has 
so long devoted his attention to this subject.^ The first 
item to be taken into account is the hull, which, however, 
need not materially differ from that of a steam or naphtha 
launch. The shipbuilder's work remains the same, only 
the interior or joiner's work requiring adaptation to the 
different nature of the equipment. By this is not meant, 
however, that any sort of hull will give satisfaction. 
Before we have it built it is well to know what we are 
going to put into it — the kind and size of motor and the 
number and size of battery cells. Space is limited on all 
sides, and the features of the hull, motor and battery 
should be carefully considered in their mutual relations. 

Let us assume that we wish to design our own boat. 
After we have roughly modeled the lines, either in accord- 
ance with the shipbuilder's practice, or to suit a special 
fancy, it will be of advantage to lay down upon paper the 
cross sections, taken at suitable distances apart from bow 
to stem, a longitudinal section and a plan. We have but 
a single straight line in an ordinary launch hull, and that 
is the keel. The rest are all curves of various character. 
These curves and the dimensions of the motive power outfit 

1. The accompanjlng sectioiis are largely based upon an admirable contribution of 
Mr. Reckenzaun to Tlie Electriccd Engineer, New York, of Aug. 18, 1890. 


should mutually agree, or else they are likely to prove 
awkward when the apparatus is being put in place, and a 
sacrifice of some kind would be the probable consequence, 
which, perhaps, a very slight deviation from the lines of 
the model might have avoided. The hull should be sub- 
stantial throughout to withstand the strain from the weight 
of battery and motor when in rough water. The joiner's 
work will include a battery receptacle or receptacles of the 
required dimensions. A trough placed directly over and 
along the keelson, with the seats arranged on top of it, has 
the advantage of giving the boat maximum stability, since 
the centre of gravity will then fall near the keel, below the 
water line. The passengers sitting back to back in two 
rows along the centre, will add to this advantage, while at 
the same time they may have an unobstructed view in front 
of them. Another method consists in placing the battery 
in a similar trough, laying the floor over it, and arranging 
the seats above this floor along the sides or across the hull. 
Access to the battery can be obtained in the first instance 
by removing the top of the seats, and in the latter through 
trap doors in the floor. Again, another way of distributing 
the cells consists in arranging the battery receptacles along 
the sides of the hull, with seats on top of them. A combi- 
nation of the above methods may be effected to suit prefer- 
ences. It is well to remember, however, that one of the 
advantages of the electric launch is that its stability may 
be made to exceed that of any other launch by a judicious 
distribution of the weight of the propelling apparatus. 

As to the material of the hull, wood is preferable to steel 
or iron when it is considered that acid is to be carried on 
board, although by a suitable construction of the cells and 
receptacles leakage or spilling can be prevented under 
ordinary conditions. These remarks apply, however, to 
ordinary boats. It has already been pointed out that 
submarine boats should have steel huUs, while some have 
been built of phosphor bronze. 

43. The next element to consider is the motor, which 
must embody high efficiency, with special compactness, 
low speed and reasonably light weight. 

It is usually desirable to put as large a power outfit into 


a boat as can be conveniently placed there and looked after, 
and since the battery requires the largest amount of space, 
the most suitable place for the motor is in the stem, as far 
back as possible. The shape of this space then will gener- 
ally determine the selection or construction of the motor. 
The cross section in an ordinary launch hull resembles the 
shape of the lower part of a heart, diminishing in Avidth 
toward the stempost. The available base area for the 
motor being triangular, the base must be narrow. With an 
armature of about double the length of its diameter and the 
field magnets crowded around it to suit the lines of the 
hull, we have a motor that can be placed without unneces- 
sarily encroaching upon space desirable for batteries or 
passengers. A low armature speed (say 500 to 800 revolu- 
tions per minute) will admit of coupling the motor shaft 
directly on to the screw shaft without necessitating exces- 
sive fineness of pitch in the screw. We have here an ideal 
method of transmission. There is no lateral strain on the 
motor bearings, while the thrust bearing, interposed 
between the motor and the screw, takes up the longitudinal 

The conditions of load in a launch motor are analogous 
in the main to those of a fan motor, but more particularly 
to those that would be encountered in an air ship or flying 
machine, the following characteristics being observed : The 
load consists in the resistance offered by the water to the 
motion of the screw. The movement of the latter is inde- 
pendent of its support — the boat in the present case — and 
its effect will be either to set in motion the medium (water) 
in which it moves, if the support is fixed, or to propel the 
latter if it is free, in consequence of the inertia of the 
medium. Hence the force required to start a boat is merely 
that required to overcome the inertia of the body of water 
affected by the screw proper and is independent of the in- 
ertia of the boat. ^ The motion, being first imparted to the 
water, is gradually transferred from the latter to the boat, 
until a point of equilibrium is reached, determined by the 
resistance encountered. It will be seen from this that even 
if we throw the full load upon the motor at once (as is usually 
done), the difference between the starting effort and that 

1. The sabject of screws is treated in a later chapter at more length. 


required to maintain the final speed of the boat is so small 
as to be entirely negligible in the determination of the 
capacity of the motor. Speed once being reached, the load 
will remain constant on a straight run, while it will slightly 
increase on turning the vessel about, until a straight course 
is resumed. In running against a current, the load of the 
motor is smaller than in still water, while it is greater on 
running with the current. Under the conditions ordinarily 
met with, this difference is, however, but slight. Auto^ 
matic governing devices are obviously not required, unless 
we were to consider an equipment for a large sea-going 
vessel. If it is desired to get two or more different rates of 
speed, the battery may be split up into a corresponding 
number of sections, by means of a special switch con- 
nected in series or parallel. The brushes should admit of 
reversing the direction of the armature movement, but their 
"lead" is best adjusted for forward motion of the boat, 
unless a double set of brushes is employed, with reversing 
lever to engage one or the other as required. A suitable 
switch inserted between motor and battery serves for 
starting, stopping and reversing the motor. 

44. The storage battery, on account of its superior fitness, 
is universally employed in connection with electric launches 
at present. Without entering here upon details of con- 
struction, we will consider the features to be dealt with in 
its application. The first question confronting us is that 
of bulk and weight. The manner of disposing of the bat- 
tery has already beeij touched upon in considering the hull. 
Being composed of a number of small units, there need be 
no difficulty in distributing it. An ordinary launch hull 
can well carry all the weight corresponding to the hulk of 
battery which can conveniently be placed with due regard 
to accessibility. We will, for convenience, take into con- 
sideration a battery of a well-known type,i designed for 
portable and locomotive purposes, occupying, per cell, 0.23 
(solid) cubic foot of space (box, plates and all) and having a 
capacity of 150 ampere-hours, or about 290 watt-hours, at a 

1. The cell here dealt with has ahwady undergone considerable improvement, but the 
ilgure will serve as a conservative estimate. The subject of batteries is resumed in a later 


discharge rate (normal) of 25 amperes or an average of about 
48.32 watts ; its weight being about 40 pounds. Reduced 
to unit cubic foot, we have : Weight, about 175 pounds ; 
capacity, about 1,260 watt-hours, with normal discharge 
rate of 210.1 watts per cubic foot. The displacement per 
cubic foot of battery will then be equivalent to (H.V =) 2.8 
cubic feet of (pure) water; at this rate, allowance, for weight 
of battery, must be made in determining the water line. 
With ordinary launch hulls, the average battery load that 
can be carried conveniently represents about one-third of 
the total actual displacement in tons, including passenger 
load. Any smaller proportion may, of course, be applied, 
with correspondingly reduced results in capacity. The 
capacity of the battery and motor are considered mutually; 
for maximum effect the former guides the calculation. 
With the type of battery above assumed, if it is to be 
worked at *' normal" rate, the capacity in electrical horse- 
power ef the motor required will be equal to the number of 
cells multiplied by 0.0647, the working rate in electric 
horse-power per cell, or to the number of cubic feet of bat- 
tery multiplied by 0.2816, the corresponding constant per 
cubic foot. For a rough preliminary calculation, on the 
basis that the weight of battery represents one-third of the 
total weight, we have. 

Capacity = -q" X 3.604 electrical horse-power; 

where D = total displacement (weight) in tons and 8.604 
the working rate of battery in electrical horse-power per 
ton (12.8 cubic feet). The duration, T^ of the run in hours 
for one charge of battery will be : 

^ = -^ hours, C denoting capacity of battery in watt 

hours and R denoting rate of delivery in watts. 

Since the power required to propel a vessel varies as the 
cube of the speed, and since the duration of the run varies 
inversely as the power (rate of delivery), it follows that 
the mileage covered by one charge of battery will vary 
inversely as the square of the speed. In practice, due 
allowance is to be made for the characteristics of the motor 
and for a falling off in the total output of the battery when 


pushed to a high rate of delivery. Where a maximum of 
speed is to be effected, the battery should have a maximum 
of active surface and a minimum of internal resistance, to 
facilitate a heavy discharge without an excessive drop of 
potential. Special care should be taken to render the cells 
acid tight, by the use of suitable covers, etc. Spilling may 
also be avoided by preparing the electrolyte in a suitable 
manner. The jelly electrolyte invented by Dr. P. Schoop 
offers in this respect a remarkable advantage.' It is also 
advisable to line the battery receptacle with some acid- 
proof material, preferably an insulator, and to provide a 
bed for the cells to stand on containing a substance capable 
of absorbing and neutralizing acid. All wires or cables 
employed about the boat should have a good acid and salt- 
waterproof insulation. 

The question of charging facilities has already been 
pretty fully discussed, but cannot be too exhaustively con- 
sidered by the owner of an electric launch. It is believed 
that the introduction of dynamotors (or continuous current 
transformers), by means of which any direct current, as 
from railway or lighting circuits can be raised or lowered 
to the potential required by the battery, is destined to 
prove of incalculable service in promoting the electric 
launch industry. But let us assume, in order to present 
the necessary calculations, that we have a 40-cell launch, 
cells being of 150 ampere hours capacity, and that 110 volt 
current is available from some everyday incandescent 
lighting plant. The difference of potential required at 
battery terminals (cells in series) would then be 40x2.5 
= 100 volts at the finish — if a constant current, 25 amperes 
in this case, is to be maintained. To reduce the initial 
electromotive force of the circuit, we must then introduce 
a resistance of 

n 110-80 ^ „ , 
B = ^g = 1.2 ohm 

at the start and gradually reduce the same as charging 
goes on to the final minimum of (approximately) 

o 110-100 ^. , 
Ji = — ^r^ — = 0.4 ohm. 

1. See later chapter on Storage Batteries, etc. 


Or, we may apply a constant electromotive force of about 
90 volts (2.25 volts per cell) by inserting a constant resist- 
ance of 

o 110-90 ^Q , 
R = ^^ = 0.8 ohm, 

in which case we will receive a heavier current at the start, 
reducing itself (in consequence of the increasing counter 
electromotive force of the battery) gradually to a minimum 
at the end, the average being the same as in the other case. 
The latter method may be preferable; the current will 
decrease in proportion to the facility with which the gases 
can be absorbed by the plates, while the results in time 
and efficiency remain practically the same, and constant 
attention is rendered unnecessary. If a boat is to be 
charged, the battery of which, connected in series, requires 
a higher electromotive force than that available, we merely 
need to split it up into two or more equal series to get 
within the required limit and then charge these in parallel 
with a proportional current, adjusted as above. 

The battery may, of course, be charged either on the 
boat or may be removed for that purpose. While the 
former method is ordinarily practiced, it is obvious that in 
order to avoid delay, a freshly charged battery may be 
substituted for the exhausted one. With suitable facili- 
ties for handling the batteries, such as a hoisting crane, or 
equivalent device for lifting and lowering the cells into 
and out of the boat, tables to receive, the cells for charging, 
suitable cell crates with connections and lifting attach- 
ments, etc., the work of exchanging the batteries could be 
effected promptly and efficiently for a whole fleet engaged 
in continuous traffic. It is evident, of course, that a 
dynamotor would dispense with all such handling of the 



Canal Boat Propulsion : Historical. — Erie Canal. 

45. The invention of Canals or artificial waterways can 
be carried back to the earliest ages of civilization. Various 
historians credit the ancient Egyptians with having first 
devised and constructed artificial interior waterways as 
early as 1500 B. C. The Chinese were also among the first 
to build canals, and the great Imperial Canal of China 
about 1,000 miles in length is still in existence. The 
canals constructed by the ancients did not, however, 
resemble our modem canals ; they were in fact simply 
large ditches dug through level stretches of country. 
When it was necessary to pass from one level to another, 
the boats, which were comparatively small, were raised or 
lowered by means of inclined planes. The Chinese were" 
perhaps the first to make use of such devices. 

Although various canals were built throughout Europe 
by the Romans, canal building did not make any great 
progress until the invention of the canal lock as used at 
the present day. With this improvement the canal became 
a most simple, effective and economical source of interior 
transportation. In fact, until the advent of the railroad, 
the canal was the great carrier of all classes of products and 
goods through level countries where no natural waterways 
existed. A very large number of canals have therefore 
been constructed and remain in use throughout the world 
at present. 

Exact statistics as to the cost, mileage and use made, of 
the canals of the world are not obtainable. The latest 
authentic figures we can compile give Europe between 
12,000 and 15,000 miles of canal, including England with 
4,700 miles ; France, 3,000 ; Germany, 1,250 ; Holland, 930 ; 
Belgium, 540. In the United States, according to the 
report made by Special Agent T. C. Purdy for the Census 


of 1880, there are 4,468 miles of canal that have cost 
$214,000,000. Of these, 1,953 miles were abandoned, in 
1880, and of the remaining 2,515 miles, quite a large pro- 
portion was not paying expenses. 

46. Navigable canals may be divided into two classes : 
Ship Canals, which can be navigated by seagoing vessels 
of large draught and tonnage ; and Barge Canals, which are 
generally rather shallow and narrow and only permit the 
use of barges or lighters thereon. The first class is con- 
fined to a few examples, such as the Suez, Manchester, 
Welland, Corinth and various ship canals in Holland. 
They have been of immense value to commerce and will 
grow in number. 

It is with the second cla^ss, or barge canals^ that we shall 
particularly deal. Of this class there are very many in 
existence, but to a certain extent the barge canal has 
unfortunately fallen into disuse in those countries where 
the railroad has become a serious competitor. Few have 
been built within the last fifty years, while thousands and 
thousands of miles of railroad have been constructed. 
Although such canals have to a great extent been super- 
seded by the railroad as carriers of freight, it is acknowl- 
edged that they remain one of the cheapest and simplest 
methods of transportation for large and bulky products. 

It is true that the railroad is many times as rapid as the 
canal, but it must be remembered that a single canal boat 
of from 100 to 250 tons, such as used on most modern 
canals, will carry as much material as can be transported 
by from 10 to 20 freight cars ; and the cost of operating a 
railroad train, carrying a load equal to that carried by a 
canal boat or number of boats, is so very many times in 
excess of the cost of transporting the same load by canal 
boats, that even considering the time element, canal trans- 
portation is cheaper for very heavy and bulky loads. To 
enable the canal to be better able to compete with the rail- 
road, however, it is primarily necessary to improve the 
present methods of propulsion. 

47. Upon the early canals, the boats were pulled by 
slaves ; gradually the horse or mule was substituted for 


man power and with the advent of steam various forms of 
this source of energy were adapted to canal boat propul- 
sion. The majority of the present barges on canals are 
still hauled by horses or mules. 

The rapid advance of electricity and the many things 
that have been accomplished with this form of energy, 
have caused it to be suggested at various times as a source 
of power for canal boat propulsion. It is particularly 
within the last two or three years that the electrical pro- 
pulsion of canal boats has become a leading topic of dis- 
cussion. In this country various state legislatures are 
about to pass or have already passed bills looking to the 
improvement of the state canals by the adoption of some 
form of electrical motive power. Private corporations 
owning canals have also taken steps to investigate this 
substitute for the mule, and experiments have been tried, 
both in this country and abroad, with one or two forms 
of electrical canal boat propulsion. During the latter part 
of 1893 experiments were made upon the Erie canal near 
Kochester with an electrical canal boat. Experiments 
have also been made in Prance recently. Of these trials 
more will be said in later chapters. 

48. The largest, most important and most valuable of our 
American canals, and one which will serve as a good 
example, is the Erie, connecting the Great Lakes with 
the Hudson. It has materially helped to build up the 
internal commerce of our country. This canal was started 
in 1816 and finished in 1825. Boats were first operated 
upon it in 1826. It practically cuts the State of New York 
in two, extending from Troy to Buffalo. As originally 
built, this canal was about 363 miles long with the follow- 
ing dimensions : Width of bottom 28 feet, width at top 40 
feet, depth 4 feet. Eighty-four locks were used along the 
route, each 90 feet long and 15 feet wide. The boats used 
were 78 feet 8 inches long, 14 feet 5 inches beam and 3 feet 
5 inches draught when loaded to 80 tons. Each boat was 
towed by one mule. 

As, however, the value of the canal became apparent and 
it served as the great outlet for the products of the west, 
these dimensions were found inadequate, and it was gradu- 



ally enlarged to the present dimensions : Width of top 70 
feet, bottom B2i feet, depth 7 feet, and the length shortened 
to 352 miles. 

The locks and structures were enlarged and the boats 
gradually increased to the present size : 98 feet long, 17i 

Fig. 46. — Shobt and Long Lock on Ebib Canal. 

feet beam and 6 feet draught when loaded to 240 tons. 
There are at present 71 locks having an average lift of 8 
feet and most of them having a length of nearly 220 feet so 
as to permit two boats coupled tandem to pass through at 
one time. Some of the locks have not been lengthened and 
are only 110 feet long, but these will all be changed to the 
greater length. 

Several large aqueducts and embankments on which the 
canal is carried over rivers, etc., are located along the route. 
It may also be well to state that there are about 207 bridges 

- I ' 19 ', ,r u^ 

Fig. 47. — Sbction op Ebie Canal. 

of various kinds crossing the present canal and that the 
towing path, which is from 10 to 15 feet wide, crosses from 
one side to the other very frequently. The illustrations, 
reproduced from Seaboard^ slightly reduced, show a 
standard section of the canal and a general plan of the 
present locks. (Figs. 46 and 47.) 


The Erie Canal cost originally about $7,500,000, but 
with the improvements this has been increased to about 
$51,500,000. The canal is only in operation about two- 
thirds of the year ; generally opening from the 1st to the 
5th of May and closing between November 30th and Jan- 
uary 3d. The average navigable season lasts about 215 

49. Various methods of boat propulsion have been tried 
and used on the Erie Canal. The original method, how- 
ever, consisted in towing each single boat by a mule on the 
tow path. This mode of propulsion was used for some 
time. Afterwards two boats were coupled tandem (called 
double-headers) and hauled by two or three mules. This 
is the present method and with the existing large boats 


Fig. 48. — Obdinaby Mule Hauling on Ebie Canal. 

three mules are used (Fig. 48) and an average speed of li 
miles per hour is obtained through water. Over 95 per 
cent, of the boats are moved in this way at present. They 
take about 10 days to go the entire length, including lock- 

The necessity for an improved and more rapid method 
has caused various forms of mechanical power to be tried, 
and in 1871, the Legislature of New York passed an act 
offering a prize for an application of steam to canal boat 
propulsion on the Erie. This brought forth various types 
of steam propeller canal boats some of which have con- 
tinued in use to the present day. In 1873 a system of 
Belgian cable towing was also installed and tried on a 
level between Buffalo and Lockport, but it was discon- 


tinned as it proved unsuccessful. Various other plans 
were also tried. 

There are in use on the Erie Canal about 60 steam pro- 
peller canal boats. These steamers are about the size of an 
ordinary barge, but are of somewhat finer lines, and when 
loaded to 6 feet draught carry about 180 tons in addition 
to machinery. They are generally equipped with boiler 
and simple condensing upright engine, revolving a screw 
at the stern 6 feet 6 inches in diameter at about 100 revolu- 
tions per minute. 

The usual modus operandi of propulsion by this method 
is for the steamer to push one barge, to which it is rigidly 
coupled, and tow two or more (generally two) boats by a 
tow line. The boats are all coupled in pairs or double 
headers. Such fleets or trains attain a speed of about 2i 

Fig. 49. — Steam Canal Boat and Consobt. 

miles per hour and make the trip in a little over seven 
days. (Fig. 49.) 

The trip of a canaler is, however, not finished when the 
Hudson is reached, but the freight must be transported to 
New York. The horse or mule boats are made up into 
large tows and are taken down the Hudson by tugs. The 
steam canaler attended by its consorts, proceeds down the 
river under its own power. 

There are, therefore, two methods of propulsion on the 
Erie Canal at present, mule towing and steam propeller. 
The latter has proved the cheaper method and although 
the mule boats are in the great majority, it may be said 
that they do not pay their owners any very great profits. 
Unless the season's traffic is unusually good, the canaler 
does not reap a large income in spite of some very hard 


50. In 1893 the New York State Legislatnre passed an act 
authorizing experiments to be made on the Erie Canal with 
a view of secnring a method of electrical propulsion of 
canal boats that would be sux>erior to the present methods ; 
and, as has been stated, one method has already been 
tried. This mode of applying electric power consisted in 
substituting an electric motor for the steam engine used in 
the present steamer and conveying current to the motor by 
suitable wires fed from a central station. 

Electrical canal boat propulsion is, however, not confined 
to this particular method. There are various other ways of 
applying electrical energy to the propulsion of boats in 
narrow and shallow waterways. Although the methods 
hereafter described are particularly considered with a view 
to their application to the Erie Canal, their employment is 
not in any way limited to any particular canaL 



Conditions Entering Into Canal Boat Propulsion. 

61. Before taking up the various py steins of electrical 
canal boat propulsion, it will, perhaps, be well to consider 
what features and advantages any new system of propul- 
sion should possess, and to discuss, generally, the applica- 
tion of such an improved method of boat propulsion on the 
Erie and similar canals. 

The prime object is to cheapen transportation. This can 
be done in two ways : (1) by increasing the speed of pro- 
pulsion, or (2) by decreasing the cost of propelling power. 
The first depends on the canal and the method of propul- 
sion ; the latter, upon the nature or form of the propelling 
energy used. In regard to electricity, it has been claimed 
that both results could be accomplished by its adoption. 
Although the adoption of electrical motive power in some 
form or other will perhaps h& more economical than the 
present methods, increased speed cannot be obtained upon 
the Erie Canal with any system without changing the pres- 
ent dimensions. With the present depth of barely 7 feet 
and the boats loaded to 6 feet draught, it happens very 
frequently that the boats, when going at 2i to 3 miles per 
hour, come in contact with the bottom of the canal, which 
sometimes causes great injury to the boats. 

Whether electricity or any other method of propulsion is 
adopted on the Erie Canal, it is absolutely necessary to 
deepen the canal before any increased speed can be obtained. 

Various plans have been suggested for deepening, the 
present channel of the Erie Canal. Among the most note- 
worthy, is a plan advocated by Ex. -State Engineer Sweet 
for raising the banks and structures of the channel one foot 
and thereby securing a depth of water of 8 feet. Another 
plan is advocated by Ex. -State Engineer Horatio Seymour, 
to bring the depth of water to 9 feet by excavating one foot 


and raising the banks and structures one foot. The excava- 
tion of the bottom would necessitate the replacing of the 
impervious material composing the present floor. The 
simplest and perhaps cheapest plan to gain additional 
depth of water, would, therefore, be to raise the banks as 
much as possible. It may be noted that a bill has passed 
the last New York Legislature which authorizes the present 
depth to be increased to from 8 to 9 feet throughout. 

52. There is, however, another consideration. It has been 
found most economical to operate the boats on the Erie in 
pairs. This makes the steering of the boats quite difficult, 
and it will be found impracticable to steer at a very greatly 
increased speed; but the present arrangement could np 
doubt be continued at a speed of perhaps from 3 to 4 miles 
per hour. 

The increase of the depth of water would either permit 
an increase in the present speed with the present boats and 
methods, or would necessitate less power to propel the 
present boats at the present speed. Either plan would cer- 
tainly lower the cost of transportation. The total propel- 
ling power necessary for a certain speed would, however, 
be greatly dependent also upon the mode of applying the 
propelling power and the cost of furnishing a horse-power- 
hour of energy at the boat. 

All practical electrical systems contemplate the genera- 
tion of current in central stations along the route of the 
canal and a distribution of the power over suitable wires 
to the motor devices. This will necessitate quite a number 
of transformations from the prime mover to the motor 

That Niagara will some day furnish electrical energy as 
far as the Hudson is not to be doubted, and the subject is 
treated fully in a later chapter. It has also been suggested 
to use water power and the waste weirs along the route 
of the canal. The want of sufficient water power adjacent 
to the canal makes this inadequate to the requirements, 
although no doubt it can be utilized to some extent. 

Contemplating merely the various transformations neces- 
sary, from the burning of coal, in the boiler, to the delivery 
of power available for propulsion on canal boats, it seems 


hardly probable that a horse-power-hour of energy can be 
furnished cheaper than, if as cheap as, the same amount 
of power as at present generated by the use of boiler and 
engine directly upon the boat where a consumption of 
from 3 to 6 pounds of coal gives a horse-power-hour at the 

63. There are, however, numerous other considerations 
in regard to the application of various forms of electrical 
propulsion to canals. The various features that a practical 
system should possess are : 

First — The system of propulsion should not in any way 
injure or aflfect the structures or banks of the present canal • 
in any way. 

Second. — It should not necessitate equipping the canal 
boat with expensive machinery which can only be used 
where there are corresponding systems of electric transmis- 
sion available, and which always takes up valuable room. 

Third. — The chance of being stalled by breaking down 
of machinery or lack of current supply must be reduced to 
a minimum. 

Fourth. — Any new method must permit of a continuance 
of the present methods of propulsion. 

Fifth. — If any structure along the canal is necessary, it 
should be strong and not easily injured. 

Sixth. — ^The labor should, if possible, be less than that 
at present required. 

54. All electric systems of whatever nature, receiving- 
their current supply from a central source of generation or 
distribution, can only utilize their propelling machinery 
where there is a source of supply of current. It has, there- 
fore, been stated that the application of electrical power to 
canal boat propulsion on the Erie would not be complete 
as it would not allow the boat to continue down the 
Hudson by the same power. 

This is certainly true for the present, and would be an 
objection to any system contemplating the equipping of 
the freight-carrying boat with electrical machinery receiv- 
ing current from a central generating station. By the use 
of storage batteries on the boat, furnishing current to an 


electric motor revolving a screw propeller, this would not 
be the case, as the boat would then be self-contained. The 
application of the storage battery to canal boat propulsion 
is, however, not yet in a practical shape. The self- 
contained storage battery boat is a possibility, but will not 
here be considered at any length. 

If, however, some appUcation of the electric motor to the 
propulsion of these mule barges will lower the cost of pro- 
pelling them below that of any other means of propulsion, 
the application of such cheaper method would unquestion- 
ably effect great economy in their operation along the 
entire route. The canal is over two-thirds of the entire 
•distance of the trip from Buffalo to New York. 

The application of electricity to the propulsion of canal 
boats is in no way limited to placing the motor directly on 
the boat, nor should it necessitate in any way changing 
or modifying the present barges. There are various other 
methods of applying this power ; in fact, the most feasible 
and practical plan would seem to be some method in which 
the ordinary canal boat is propelled by an exterior motor, 
or a tug. 



Methods of Applying Electricity to Canal Boat 
Propulsion. — Boats Equipped with Motors. 

55. Electrical canal boat propulsion is divisible under 
two distinct heads : — ^Propellers, in wMch the water forms 
part of the system of propulsion ; and hauling or towing 
systems. There are, however, five methods of applying 
these. They may be classified as follows : 

A. — Propellers. 

B. — ^Flexible submerged cable or chain towing. 
(7. — Rigid rail or rack haulage. 
B. — Movable cable haulage. 
E. — Motor locomotive haulage. 

These five classes cover nearly all possible forms of 
practically operating canal boats by electric or other power. 

Propellers. — Class A. 

56. This class includes all forms of boat propellers 
operated by electric motor on the boat. Although the jet, 
paddle wheel and screw propellers are included, we shall 
only consider the screw. The other mechanisms have been 
found more or less impracticable for boat propulsion on 

As generally suggested, this method consists in placing 
an electric motor directly on the freight-carrying boat hav- 
ing at its stem a screw propeller which is revolved by the 
motor. The motor receives current from suitable contact 
wires suspended either over the canal waters or on the 
banks. As the boat has more or less lateral movement, the 
contact arrangement must be flexible, and as the canal 
(being fresh water) cannot well be used as a return, a 
double metallic circuit must be used. This will necessi- 
tate two wires for boats going in each direction. With 



any such system it will be found necessary to use some 
form of double overrunning trolley carriage as shown in 
the illustration (Fig. 50), and suitable switches and 
turnouts must be arranged in the contact wires. 

The first boat so operated in this country was the '' Frank 
W. Hawley," on the Erie Canal, near Rochester, during the 
latter part of 1893 (Figs. 51, 51a, 51b, 61c and 51d). She 
was named after the gentleman whose energy and enter- 
prise led to the making of this noteworthy experiment the 
first of its kind in America. 

This boat was an ordinary steam canal boat equipped 

Fig. 60. — Overhead Double Trolley. 

with what is known as a dish-pan screw. The engine was 
disconnected from the shaft, and two Westinghouse electric 
motors of 25 h. p., street railway type, were substituted and 
directly connected to the screw. The motors received cur- 
rent from a pair of wires suspended over the canal through 
two ordinary underbearing trolley poles, as shown in the 
illustration (Fig. 51c). This arrangement was crude and 
caused a great deal of trouble as the lateral movement of 
the boat continually caused the trolley wheels to run off 
the wires. An arrangement of trolley carriage, somewhat 
similar to that shown in Fig. 50, was afterwards substi- 



tuted. The boat was run along the canal at quite a high 
speed and its passage caused a great deal of excitement. 

Various plans have been devised by which the boats are 
permitted to move laterally without losing contact with 
the trolley wire and by which two boats can pass each other 

1 ^^^^m^^^^K" ^"^^^^^^^^^^^^^^^^^^^^^^^Ml 


1 ^^ 



^ E3 


" o 


when going in the same direction. Prominent among these 
is a method devised by Mr. S. W. Gear, of Buffalo, which 
consists in making the trolley contact and wire laterally 
movable. A general plan of this arrangement is shown in 
the illustrations (Figs. 52 and 52a). The trolley wire is 



suspended on small carriages that are adapted to travel 
laterally on the span wire stretched across the canal. The 

Fig. 61a. — ^Propelleb op the "Frank W. Hawley." 

contact carriage on the boat is also laterally movable as 
shown. This makes apparently a complicated arrangement 

Fig. 51b. — ^Thb Electrical Boat, "Frank W. Hawley" — 
AT Canal Bank. 

for which there is really no necessity as the same results 
can be obtained with the simple method previously shown. 


57. The placing of a motor directly or permanently on 
the freight-carrying boat is not the most feasible form of 
applying the propeller to canal boat propulsion. There are 
two others ; (1) by the use of a false stem, rudder or port- 
able motor which can be attached to the boat when desired 
and which contains the motor and screw, and (2) by the use 
of small propeller tugs. The illustration (Fig. 53) shows 
quite a novel method devised by Mr. Samuel H. Jones, of 
Newark, N". J. The motor is placed in a box or casing and 

Fig. 61c. — Deck of the " Frank W. Hawley," Showing 
THE Trolley Poles. 

revolves a propeller on the outside. This arrangement is 
either permanently or temporarily fastened to a rudder 
post, and the boat is steered by changing the position of 
the propeller and motor, a represents the rear end of the 
boat, B the casing, c the motor, d the screw, e the rudder 
post and tiller, and f are the wires going to the motor. 

Various forms of applying a detachable motor and screw 
similar to the Jones plan can be devised. Another plan 












is to have a false stem rigidly attached and holding the 
rudder, screw and motor. 

58. The tug method can also be varied. The illustration 
(Fig. 54) shows a combination of the two which possesses 

Fig. 62. — Gear's Laterally Movable Trolley Wire 
FOR Canal Boats. 

several points of advantage over other methods of apply- 
ing the screw, a represents an ordinary canal barge ; b 
is a small boat about 35 feet long, 8 to 10 feet wide, and 5 

Fig. 52a, — Laterally Movable Trolley Wire and Hanger. 

feet draught, which holds the motor and screw, and which 
is rigidly attached to the freight-carrying boat. Contact 
is made with the trolley wire by the trolley carriage, c, 
which is connected with the boat by flexible connections. 
Such an arrangement can be applied to any number of 



boats, pushing one or two and towing one or more pairs, 
or the tug may simply tow boats. 

Fig. 56 shows an arrangement of propelling mechanism 
for canal boats for which a patent has recently been issued 

Fig. 53. — Jones' Detachable Propelleb and Rudder. 

to B. C. Scott, of Belgium. The object of this device is to 
prevent the loss of power caused by the screw being but 
partly immersed. The screw, a, is enclosed in a chamber, b, 
at the stern of the boat. Water enters the chamber from 
the opening, c, at the bottom of the boat and is expelled 

Fig. 54. — Electric Ttjg Propeller System. 

through the opening, d, at the stern. Gates or valves are 
located at both openings so that the chamber can be closed. 
When necessary the chamber can be exhausted through 
the valve, e. This will cause the water to rise in the 



chamber and immerse the screw completely when the inlet 
valve or gate is opened, and by opening the valve, d, the 
boat can be propelled with its screw completely immersed, 
without reference to the immersion of the boat. An 
electric motor is shown here revolving the propeller screw, 
and current is supplied by overhead wires as in the other 
propeller methods. 

If the propeller is used it T^ill readily be seen that some 
form of separate arrangement, either false stern or tug, 
must be adopted. With such an arrangement some of the 
requirements are met with, but it is doubtful whether the 

Fig. 55. — Scott Combination op Jet and Pbopbller Operated 
BY Electric Motor. 

propeller is the most efficient and feasible means of canal 
boat propulsion. 

The placing of a motor on the boat itself would cause but 
a very slight saving if any over the steamer. We could, 
perhaps, use a screw of small pitch turning at a high rate 
of speed, but the additional surface friction would make 
the total loss in the screw but very little, if any, less than 
with those at present in use; and as has been stated, the 
delivery of power at the motor shaft may be but very little 
if at all cheaper than as furnished at present by the engine. 
Such an arrangement, therefore, would really not be of any 
material advantage over the present steam propeller 


59. Aside, however, from the equipment of the boat and 
the efficiency of the motor and propeller, the nature of the 
canal itself seems to be the principal objection to such a 
method of propulsion. It must be remembered that the 
canal is simply a large ditch, and no provision has been 
made in its structure to prevent the injurious action that a 
propeller would have on the walls and banks of the canal. 
The objection raised, when the steam propeller was first 
tried, was that the wash from the screw would ruin the 
banks of the canal. The screw agitates the water and this 
agitation or wash would, if there were enough of it, be very 
injurious to the structure of the walls. With the present 
steamers (and they are in the great minority) on the canals, 
at their low speed of between 2 and 2J miles per hour, the 
banks of the canal are but slightly affected, but what 
would be the consequence if all boats on the canal were 
propelled in this way and at a Tiigh speed ? 

There can be no doubt, that a general adoption of high 
speed electric screw propellers upon the Erie Canal for all 
craft would cause great injury to the walls and banks. It 
would apparently be necessary to rebuild the present 
walls if the screw were universally resorted to, even though 
the present speed were not increased. 

It would appear, therefore, that although a system of 
electric trolley propellers is perfectly practical and oper- 
ative mechanically, its adoption, on the Erie Canal at least, 
would not cause any great improvement over the present 
steam propeller, as regards economy or so far as relates to 
the preservation of the canal structure. 

Flexible Submerged Gable or Chain Towing. — Glass B. 

60. This system is extensively used on some of the canals 
and shallow canalized rivers throughout Europe. The 
steam engine has hitherto been used as a motor. Experi- 
ments have, however, been made recently with electricity. 

The apparatus consists, in part, of a cable or chain laid at 
the bottom of the canal or waterway. It is lifted and 
passed over rollers or wheels on the boat, which firmly grip 
the chain or cable. These rollers are rotated by a motor on 
the boat and as they revolve they pull the boat along the 



waterway on the cable. The general arrangement of wires, 
motors, etc., would be very much the same as in the 
propeller method, but instead of turning a propeller, the 
motor would revolve the hauling drums or machinery. 

Although it would be quite feasible to place the motor 
directly on each canal barge, it would be most advisable to 
employ separate towing boats ; not only for the room gained 
but for the other reasons previously given. The arrange- 
ment of distributing and contact wires and appliances 
would be about the same as in the previous class. Instead 
of having a double contact wire and double contact trolley 
carriage, the cable could be used as the return conductor. 
As this would necessitate only one contact wire, the con- 
tact arrangement and switching devices would be simple. 

In France, Germany and Belgium, where the cable or 

Fig. 66. — Oebman Method of Submerged Cable Haulage, 
Showing Steamer. 

chain towing method is used, it has been found necessary 
to employ a separate towing boat on which the boiler, 
engine and necessary hauling machinery is installed. This 
is necessary because the machinery used has been bulky, 
occupying a great deal of room ; and, the arrangement for 
taking up the chain or cable generally necessitates a 
separate boat. The European cable towing steamers are 
generally quite large and designed to take very heavy tows. 
There are various methods of taking up and grasping 
the chain or cable. The illustration (Fig. 56) shows the 
plan used very extensively on the river Elbe, in Germany. 
In this method a flexible cable is used which passes over 
















the top of the boat and around the hauling drums that are 
revolved by a steam engine on the boat. 

In 1873, a cable system of hauling was tried on the Erie 
Canal, but it did not prove successful. The method was 
similar to that in use in Belgium. The illustration (Fig. 
67) shows the cable towing steamer ''Gov. Clinton" which 
was operated on the Erie Canal. It will readily be seen 
from the cut how the cable is raised and passed over the 
central wheel or sheave which, by a peculiar arrangement, 
is so arranged as to grip it firmly. The boat was equipped 
with boiler and engine. Two such boats were operated for 
some time between 1873 and 1880, but they became so 

Fig. 58. — Canal Boat "Ampere," France. 

objectionable and interfered with other boats to so great an 
extent that their use was discontinued. 

There is great difficulty in properly gripping the chain 
or cable with any of these methods, and various plans have 
been devised for obtaining a good hold between the hauling 
drum or wheel and the chain or cable. Among these is a 
plan devised by a French inventor in which a sprocket 
wheel is substituted for the friction drum or pulley and the 
chain used is so arranged as to engage with the sprocket 
wheel. This plan is more or less impractical as the chain 
very soon gets out of adjustment. 

In France, on the Seine, a peculiar and novel system of 
chain towing has been in use for some time. Although 
electricity does not furnish the actual propelling power, it 



plays quite an important part in the general method. A 
chain is used instead of a flexible cable, and in order to 
obtain the necessary grip upon the chain, the drum over 

which it passes is mag- 
netized by an electric 
current and the neces- 
sary grip obtained by 
magnetic adhesion. 
(Fig. 58.) In the illus- 
tration (Fig. 58a) there 
are shown transverse 
and longitudinal sec- 
tions and a plan of the 
towing steamer. This 
method was devised by 
M. De Bovet and con- 
structed by the Com- 
pagnie de Touage de la 
Basse Seine et I'Oise. 
The boat called the 
"Ampere," can be pro- 
pelled either by haul- 
ing on the chain or by 
means of the propeller 
at the stern as will be 
seen. Either or both 
can be used at once. 
The engines are of 150 
horse-power. The chain 
passes over the bow, 
around the towing pul- 
ley A, and is guided by 
a non-magnetic guide 
pulley, B, and passes out 
over guide pulley c. 
The latter is very mas- 
sive, and is made of iron 
for the reason that if at that point the guide wheel is put 
in contact it gives to the magnetic lines of force an easier 
passage than that offered by the chain, and the latter no 
longer serving to close the magnetic circuit, releases itself 


more easily under the action of the weak tension of the 
slack towards the rear. A finger of non-metallic metal is 
arranged above the pulley a, so as to make sure of the 
disengagement of the chain. 

In this method of towing it is sometimes necessary to be 
able to control the paying out of the chain at the stem of 
the vessel, and for that purpose, after passing over the 
grip pulley a, part of the chain is deposited loosely in the 
pit p, provided for that purpose ; keeping it there when 
there is sufficient slack and paying it out again when more 
slack is needed. The paying-out chain at the rear of the 
vessel is, therefore, also magnetized like the towing pulley, 
but not so strongly, as the braking effort is far less than 
is required for traction. 

When the chain has little slack, the action of the pulley 
c is insufficient to produce the disengagement of the chain 
from the main pulley a, and then by magnetizing the 
pulley D, it assists the pulley o in pulling off the chain 
from the main driving pulley. 

The construction of the magnetic pulley is shown in sec- 
tion. The magnetic friction being a function of the 
intensity of the current, it is evident that the frictional 
force can be regulated to a nicety, and sudden strains put 
upon the chain merely allow it to slip a trifle without 
causing rupture. 

61. The substitution of the electric motor for the steam 
engine and boiler, as used in the European methods, would 
without question be perfectly feasible, and the economizing 
of space effected might make it possible to locate the 
towing apparatus directly on the canal boat, if such were 
desirable. The illustrations show a method of applying 
such towing apparatus directly to canal boats. Fig. 59 
shows a plan of this method on a canal. Fig. 69a is a view 
of the motor on the boat with the hauling machinery. 

France possesses a large number of canals which render 
water communication very easy between different parts of 
the country. Unfortunately, up to the present time, the 
hauling has been done most generally by men or animals. 
The time occupied was frequently very long, and, as a 
result, railway transportation has been preferred. It was, 



therefore, necessary to find a remedy for this backward 
state of affairs. The first investigation with this end in 
view goes back to the year 1888, when M. Maurice L^vy 
made a number of experiments with cable hauling in the 
neighborhood of Paris, at the junction of the St. Maur and 
St. Maurice canals. The system was based on the employ- 




•ff}/f>}fOn>>>'^ - 


Figs. 59 and 59a. — Electbic Motob Submebged Cable 
Towing System. 

ment of an endless wire cable on each bank, carried on sup- 
ports provided with pulleys. The cable was operated by a 
fixed motor, and to it the boats were attached by means of 
hauling ropes. This system gave a speed of four kilo- 
meters per hour. It constituted a considerable improve- 
ment on the existing methods but did not give entire 
satisfaction. Further reference will be made to it. 

In 1890 there was talk of employing electric traction, 
and comparative plans were drawn up for steam and 



electricity. The first plans for steam were rejected as 
being too costly and impracticable ; as for electricity one 
could utilize as motor power the water falls, found 
on the route of the canals themselves. This idea was 
worked at for several years and finally, towards the end of 
1893, an installation of the submerged flexible cable type 
(Fig. 60) was put in on the Bourgogne Canal, which unites 
the river Tonne and the river Saone. The Tonne being an 
affluent of the Seine, this canal establishes communication 
between the English Channel and the Mediterranean. On 
this canal the hauling machinery is now operated by an 

Fig. 60. — Electrical Canal Boat, Bourgogne Canal, France. 
(Submerged Flexible Cable Haulage.) 

electric motor on the boat, receiving current from an 
overhead trolley circuit. 

The installation has been established by the Department 
of Bridges and Roads for a length of six kilometres close 
to an underground tunnel 3.3 kilometres long. The motive 
power is obtained from water falls having a fall of 7.5 
metres in two successive locks. There are thus two gener- 
ating stations established at the two ends of this junction 
canal. In each the turbine drives a Gramme dynamo, one 
running at 1,200 revolutions per minute and 380 volts, and 
the other at 900 revolutions and 270 volts. Each of the 
dynamos is appropriate to the conditions existing at the 
locks and the corresponding turbine. The dynamos are 



connected after the manner shown in the accompanying 
diagram (Fig. 60a). a and b are the two dynamos connected 
to a bronze overhead feeder of 8 to 10 mm. diameter, held 
by porcelain insulators mounted on poles. The dynamos 
are coupled in series, d d are the trolley wires, and e the 
towing motor which drives the chain drum. 

A storage battery of 250 cells, having a capacity of 160 
ampere hours and a discharging current of 15 amperes, is 
coupled in parallel. These are Chloride batteries made by 
the Soci^te pour le Travail Electrique des Metaux. The 




Fig. 60a. — Diagram of Cibcuits op French Electric Canal 


mean rate of discharge varies, according to the load, from 
10 to 25 amperes at the potential above given. The stretch 
of six kilometres is covered in less than an hour. 

Special arrangements have been made for the automatic 
regulation of the turbine. The water gates are opened or 
closed according to the power furnished by the machine. 
The tunnel of 3.3 kilometres mentioned is lighted by incan- 
descent lamps which are branched in multiple on the power 
circuit. It may be said that this electric canal installation 
has given such satisfaction that it is certain that before 
long similar methods will be largely- employed in France, 
where canal locks are numerous. It is the first to be 
operated on a practical, commercial basis. 



62. However, the preferable way to operate any electric 
propelling system, would appear to be in having no machin- 
ery on the freight carrying canal barge. Some form of 
separate towing or hauling boat or some arrangement of 
motor detachable from the canal boat would be more 
feasible than placing the motor directly on the boat. 

The first plan would correspond to the European methods, 
with the exception that the towing or hauling boat would 
be comparatively small. The second plan would seemingly 
embody more or less complication and is, perhaps, not as 
practical as the first. The illustration (Fig. 61) shows a 



Tig. 61. — BussEB Method op Submebgbd Flexible Cable 


plan devised by Mr. Otto Biisser, a German engineer, in 
ivhich a detachable prow is used holding the motor and 
the hauling machinery. The general operation is similar 
to that in other cable towing methods with the exception 
that the motor, instead of being fixed on the canal boat, is 
placed with the hauling machinery on a platform which 
can be attached to the gunwales of the boat. Current is 
fed by a trolley. 

Although a system of electric cable or chain towing 
would be more efficient than the propeller, and free from 
some of its objections and would hardly cost very much 
more to install, it has various difficulties which make its 


practical adoptioii, on the Erie Canal, for example, quite 
donbtfol. Such systems would, however, be most satisfac- 
tory on comparatively short, level and shallow waterways, 
and where a very large number of boats could be hauled by 
one towing boat. With regard, however, to the Erie it 
may suffice to say that it has been tried and found 

Amongst the objections to the application of a method 
of cable or chain towing or haulage by electric motor, 
various troubles may be mentioned : 

The first is the difficulty in rounding curves and in steer- 
ing the boat. As the chain or cable lies loosely at the 
bottom it frequently comes very close to the banks, and as 
the boat must follow the chain or cable, it runs into the 
banks and sustains damage. 

The boats cannot be steered, and frequently cause injury 
to other boats propelled by mules. 

If two cables or chains are used, one for each direction, 
it is difficult to keep them from interfering with one 

The many locks on the Erie Canal would necessitate the 
taking up and dropping of the cable a great number of 
times, which would cause considerable difficulty and loss 
of time. 

As the cable must be long enough to be raised by the 
different towing boats, a large amount of slack is necessary 
which must be pulled in by each boat. Slipping of the 
cable on the hauling drum or sheave would also be 
experienced in operating heavy tows. 

Rigid Rail or Rack Haulage. — Class C. 

63. This method consists in placing a rigid rail or rack 
parallel to the canal. The rail or rack may either be 
placed on the banks or over the canal or may be sub- 
merged. Engaging the rail or rack are rollers or pinions, 
extending out from the boat on suitable arms or supports, 
connected with a motor on the boat by which they are 

The boat is propelled by the rotation of these rollers or 
pinions on the rail or rack by the motor on the boat. The 



motor receives current from a similar contact arrangement 
as in the previous system (one wire only being used each 
way). If the exterior rail is used, the wires can be 
supported by the structure holding the rail. 

This plan has never been practically tried, although 
various methods have been devised of operating in this 
manner with a rail or rack exterior to the canal and an 
engine and boiler on the boat. A plan devised by Mr. 
N. P. Otis, of Tonkers, in which a rack and pinion is used, 
is shown in Fig. 62. The rack a is placed along the bank 
of the canal. On the canal boat, b, is placed an engine 
which revolves the pinion, c, at the side of the boat and 
which engages with the rack. At the other end of the 

Fig. 62. — Otis Mbthod of Pbopellino Canal Boats. 

boat is another pinion, d, engaging the rack, but not 
rotated, to keep the boat in line. Both of these pinions 
slide in guides to allow for the varying draught of the 
boat. It will readily be seen that by rotating the pinion c 
by the engine, a direct pull is obtained on the rack, and 
the boat pulled along the waterway as the pinion revolves. 
It would be very simple to substitute an electric motor for 
the engine. 

64. Instead of placing the electric motor directly on the 
barge, it may be placed on a separate towing or hauling 
boat as in either of the previous methods. A novel and 
original plan of rail haulage is shown in Fig. 63. The rail 
is suspended over the canal and a separate hauling boat is 



used. The cut shows a transverse section of the canal with 
the towing boat only. Suspended from the cable, a, strung 
across the canal, are the double headed rails, one for each 
direction. The contact wire, c, is suspended below the rail 
and supported by insulators fastened to the rail which 
forms the return. On the boat, d (which is connected to 

"T' *" 

Pig. 63. — Ovekhkad Rigid Rail Haulage. 

the barges), is placed an electric motor receiving current 
through the contact arm, e. The rollers, f, engage each side 
of the rail and are supported as shown. As the motor 
devolves the rollers, the boat is pulled along by the trac- 
tion of the rollers on the rail. Arrangements must be 
made to give the rollers the necessary pressure on the rail, 
and some means is necessary for readily disconnecting the 
boat from the rail. Instead of using rails, a rack and 
pinion can be used. The hauling boat may also be equipped 

with a screw so that it may be propelled through locks. 
A different contact arrangement is necessary if this is done. 

65. The placing of the rail or rack at the bottom of the 
canal and operating, in a manner similar to the previous, 
is also a somewhat new and novel suggestion. The illus- 
tration (Fig. 64) shows such a method, which is practically 



the reverse of the one last described. The cut shows a 
transverse section of a canal with the hauling boat. Over 
the canal are suspended two trolley wires, a, one for each 
direction, but they may also be supported on posts along 
the banks. The boat, b, is equipped with an electric motor 
receiving current through a contact arrangement as shown. 
This motor revolves the two rollers, c, below the bottom of 
the boat. The rail, d, is anchored at the bottom of the 
canal, and the rollers engage with it and propel the boat. 

Fig. 65 shows a general arrangement of the circuit con- 
nections and propelling mechanism that could be used 



Fig. 65. — ^Elkctbic Motok, Gearing and Friction Drums 
TJsBD with Rigid Rail. 

in either of the preceding methods of rigid rail or rack 

66. As most canals have slope walls, some form of over- 
head construction would be used in rack or rail methods 
unless the walls were changed ; or it might be possible to 
have arms extending out from the hauling boat to the rack 
or rail on the bank. 

In each of the classes already described it is necessary to 
place upon the propelling boat more or less stationary or 
detachable machinery. It is true that this may be a 
separate towing boat carrying no freight, but such boats 


would interfere to some degree with the present methods 
of locking and, if there were a large number, would cause 
more or less trouble. Over 95 per cent, of the present 
barges are towed by mule power from the banks. There is 
no objection to this method were it not for the large cost of 
operation. Such a plan has various advantages which wiU 
readily be seen. 

In the other methods to be described, this plan is 
adhered to. The present canal barges are left as they are 
and are propelled by power exterior to the waterway, as 
in the case of mule haulage. 



Methods of Electric Canal Boat Propulsiok with 
Motor Exterior to Boat. 

Movable Cable Haulage. — Cflass D. 

67. This system, although not very practical or feasible 
on the Brie, could be successfully applied on some other 
canals to boat propulsion. The propelling power is fur- 
nished by a moving cable operated parallel to the canal. 
Extending from the boat, are suitable arms, to which 
devices are attached, adapted to grip and hold the moving 
cable. The boat is pulled along the waterway by the 
movement of the cable, and can be started or stopped, by 
connecting or disconnecting the grip with tbe cable by 
suitable devices on the boat. The boat is therefore oper- 
ated mechanically. Messrs. Oriole and Maurice Levy, in 
1888, proposed and experimented with, this method on 
some of the French canals, as already stated. The cable 
stations were, however, operated by steam. 

The first experiments in this direction. Figs. 66, 67 and 
68, were undertaken at the junction of the Saint Maur and 
Saint Maurice canals. This point was especially selected 
because the canals meet at right angles and hence present 
peculiar difficulties in turning the boats. An endless 
metallic cable is installed upon each bank, a few yards 
from the edge, in order to leave the tow path free. It is 
supported here and there by channeled pulleys, which, are 
loose upon metallic supports, from six to ten feet in height. 
On a straight line, these pulleys are vertical, and are ten 
inches in diameter. On curves they are more or less 
inclined and have a diameter of five feet. At its starting 
point, the cable passes over three large pulleys, actuated 
by a steam engine, placed in a small power house at the 
edge of the canal. To the right, there is a fourth, pulley 



carried by a small car provided with a counterpoise which 
serves to keep the cable uniformly taut. The cable is pro- 
vided with links to which is affixed the rope that hauls the 











boat along. These links, fixed between rings, are capable of 
revolving freely upon the cable so as to avoid troubles from 
the latter' s tension. In order to prevent the cable from 



jumping out of the pulley channels, it is kept in place by 
a small overriding roller, and the flanges of the pulleys are 
notched, so as to allow the hauling links to go by. The 
cable runs at 2i miles an hour. 

The system herein proposed is, however, electrical, as 
the cable lengths are operated by large electric motors 
located along the line of the canal and receive current 
from one or two very large generating stations located at a 
cheap source of power such as a water-fall. No distribut- 
ing or contact wires are used along the canal excepting the 

Fig. 67. 

power wires from the large central generating station to 
the various electric cable stations. Fig. 69 shows a general 
arrangement of this method. As will be seen, electricity 
is generated by means of large alternating dynamos, which 
are operated by water-power. The current is transmitted, at 
high pressure, to the various cable operating plants along 
the canal where it is reduced in pressure and synchronous 
or induction motors are operated. The motors move the 
cable machinery, which is similar to that of any cable 
railway plant. The cable would be operated in lengths of 
about 5 miles, each way, from the power station, the out- 
going cable being used for boats in one direction and the 

















§ g 






'7) 1 iii'H 

return for those in the 
other. Fig. 69a shows a 
transverse section of a 
canal equipped with 
this method. Along the 
bank of the canal a 
track is laid, upon which 
the grip carriage runs, 
which is suitably held 
to the rails, by the rol- 
lers, as shown. The 
moving cable is grasped 
by the grip, b, on the 
carriage. The grip, b, is 
operated by a connection, 
c, from the boat. A rigid 
connection, d, extends 
from the boat to the 
carriage. As will readily 
be seen, there are some 
difficulties in such a 
plan, but judging from 
the experiments made in 
France, this plan would 
appear to be a promis- 
ing method. Upon very 
short level canals, a plan 
of cable haulage such as 
shown would certainly 
operate successfully. 
The current must, how- 
ever, be generated where 
power is very cheap. 

Motor LocoTnotive Haul- 
age. — Glass E, 

68. In this plan, the 
old and primitive method 
of haulage or towing as 
by mules is adhered to ; 


but, for the slow and uneconomical animal, an efficient 
electro-mechanical mule is substituted. The boats are not 
equipped with electric machinery, but any ordinary boat 
can be towed or hauled without any previous preparation. 
Such systems of operating canal boats were among the 
first proposed. In the earlier forms, the steam locomotive 
was used instead of the electric motor. This was run 
upon a track on the banks of the canal and towed the boats 
connected therewith. For the steam engine an electric 
locomotive can very readily be substituted and current can 
be supplied in any suitable way. There would practically 

Fig. 10. — Davis' Electbic Canal Boat Towing System. 

be an electric railway along the banks ; the boats being 
"trailers," connected by tow line to the motor cars. 

69. Among the plans that have been devised for operat- 
ing electric locomotives on suitable tracks along the banks 
of the canal and towing the boats in various fashion, is the 
method of Mr. T. D. Davis, of Syracuse, N. Y., as shown in 
the illustration (Figs. 70, 70a, 70b, 70c, 70d). The plan con- 
templates the laying a narrow gauge rail-track on each 
bank of the canal and moving the boats in trains or tows 
of four or six boats each by means of a small car furnished 



with a device for gripping the rail, to be driven by an elec- 
tric motor from an overhead trolley line. The boats are 
connected with the inner rail, laid near the water's edge, 
and require no steering. In some of the larger cities on 
the line of the canal where it might be impossible to lay a 
track on the berm bank a double track may be laid on the 
tow-path and the outside tows steered as they now are. 

Figs. 70 and 70a show the principle of the device for tow- 
ing the boats, aaa are three horizontal wheels secured to 
vertical shafts carrying on their upper ends three cog wheels, 
BBB, by which they are geared together, and all have a uni- 
form speed. Two of the wheels, a, are on one side of the 
rail. The third wheel is placed in a movable frame on the 


Fig. VOa. — Davis' Electbig Canal Boat Towing System. 

opposite side and is forced against the rail by the spring, d, 
surrounding a shaft, on one end of which is a threaded 
extension, e, connected by gearing with a hand wheel 
within easy reach of the motorman. The object of this is 
to secure an elastic grip of any desired pressure of the 
wheels on the rail. The shaft of one of the wheels, a, ex- 
tends above the cog wheel, b, and is provided with a bevel 
wheel, 0, which is driven by the motor, g ; the whole carried 
on a small platform car and driven by a trolley connection. 
This device insures a positive grip on the rail and utilizes 
the entire amount of power received from the motor, with 
the exception of a small amount of loss by unavoidable 



Fig. 70b shows a light triangular wooden frame. TMs is 
provided at its wide end with a pair of T-hinges by 
which it is connected to the side of a boat as shown at 
Fig. 70c. A pair of slotted irons about four feet long, 
secured to the side of the boat, have their upper end open 
for the reception of the T-hinges which are secured at any- 
desired height by pins. This frame is provided with an 
iron bai:, b, moving freely in the frame and provided near 
its outer end with three small wheels ; one vertical wheel, c, 
rolling on the top of the rail, e, carrying the weight of the 
free portion of the frame, and the two horizontal wheels, dd, 
rolling, one on each side of the rail. 

Figs. 70b and 70c. — Davis' Electric Canal Boat Towing 


The bar, b, is provided with two spiral springs as shown 
and also with an extension or guard at its extreme end. 
The spring will compensate for any sudden jar or move- 
ment of the boat, thus relieving the boat and rail and 
securing an elastic, instead of a rigid, connection. 

When a boat enters a lock, the free end of the frame, a, is 
raised by a lanyard and turned over on the deck, as shown 
by the boat leaving the lock at the left of Fig. 70d. 

It will readily be seen that an ordinary electric hauling 
locomotive, as used in mining or factory work, with the 
necessary trolley and contact wire, and so arranged as to 
resist the side pull of the boats, could be used in the Davis 
system. There have also been various other suggestions 




similar to this, nearly all necessitating the use of tracks on 

one or both banks of the canal (preferably both) on which 

runs an electric hauling locomo- 
tive, which is suitably held to the 
track, either by its own weight or 
by peculiarly constructed track, 
and is operated by a person rid- 
ing on the hauling locomotive. 
Perhaps a very feasible form of 
this method of hauling would be 
to erect along the banks of the 
canal, a form of telpher system 
with suitable devices for holding 
the car to the cable or rails. 

An ingenious form of such a 
system as that referred to in the 
foregoing paragraph, has been 
invented and put in operation by 
Mr. Richard Lamb, C. E. ; and our 
illustration (Fig. 71) shows it in 
use on the Raritan Canal at Tren- 
ton, N. J. This system is intended, 
also, for the haulage of logs in 
lumber camps, to which it has 
been applied with considerable 

As shown in Figs. 71 and 71a 
the system consists of a cableway 
along the towpath supported by 
brackets from 100 to 150 feet apart, 
placed upon trees or substantial 
columns and insulated from the 
ground. The main, or bearing 
cable has a copper core, having 
large steel wires surrounding it, 
giving ample strength as well as 
being a good conductor for the 
current, and presenting a good 

wearing surface for truck wheels to roll upon. 
The cable rests in steel saddles on the bracket, having 

wedge clamps upon them to hold the cable rigidly to the 
















bracket; these saddles are insulated from the brackets. 
Under the bracket proper is an auxiliary bracket having a 
V-shaped saddle with its mouth opened at the top, and 
having jaws projecting inwardly to engage the i-inch trac- 
tion cable, preventing it from coming out of the saddle, 
except when temporarily lifted when the car passes a 
bracket. Owing to this feature, the cable is not confined 
to a straight line, but can be located with right and left 
curves. The bracket is easily removed and in canal-boat 
towing a simple permanent bracket is used. This i-inch 
cable is attached firmly at either end and grounded. In 
practice this cable does not require to be very taut. 

The motor is made with a truck, having grooved wheels 
to run upon the cable. A horizontal axle on the frame of 
the truck, located below the centre and between the wheels, 
supports a hanging-frame. An eUiptically grooved sheave 
is attached to this frame, which is revolved by means of a 
newly patented worm, or wedge-gearing, driven by a 6 K. w. 
Lundell electric motor, with vertical shaft. 

By taking a couple of turns of the i-inch traction cable, 
around the eUiptically grooved sheave, when the electric 
motor revolves the gearing, the sheave winds up, and at 
the same time pays out on the i-inch cable, the action 
being the same as that of any elevator, only the direction 
of the pull of this motor is horizontal instead of vertical. 
The direction of the pull on the traction cable is parallel 
with the bearing cable; in consequence, the motor can 
easily climb steep grades. 

The hanging-frame is divided just below the level of the 
cable, and is bolted together, with provision for thorough 
insulation of the upper from the lower part of the hanging 
frame. An insulated wire is connected with the upper 
part of the frame ; this extends to the rheostat, which in 
turn is connected with the motor. The return current is 
allowed to pass through the frame to the elliptically- 
grooved sheave, then through the ^-inch cable to the 

In canal-boat service the towing hawser contains insulated 
wires. This cable is rigidly attached to an eye-bolt, just 
below the point of insulation on the hanging-frame, and 
from the end of the rope, the various wires are connected 



with their respective connections. The bight of the rope is 
connected with a clamp made of non-conducting material ; 
the socket of this clamp, and the pin which engages in it, 
which contains the wires leading to the reversing switch 
and rheostat, are made irregular in shape, in such wise 
that corresponding wires are obliged to come in contact, 
when the clamp is connected. 

In operating, a canal boat will apply for a motor ; if the 
boat does not own a rheostat one will be placed on board. 

Fig. 71a. — The Lamb Elbctbic Cablbway. 

The towing rope will be attached to the samson-post, 
leaving the end of the rope free. The wires in the cable 
will be connected with their respective wires in the rheostat 
by the clamp, as described, and the boat proceeds. 

On approaching another boat coming in the opposite 
direction, motors are stopped, cables are disconnected, and 
boats exchange cables and, consequently, motors, and pro- 
ceed. Of course, an extra cable, one above the other, or 
one on both sides of the banks, would obviate the necessity 
of exchanging motors. 

The trial plant illustrated, is operated with a 15 k. w. Edi- 


son dynamo, at 220 volts. The motor is a 6 k. w. Lnndell, 
provided with a metal bonnet to protect it from rain, but 
the bonnet is not shown in the illustration. The ease with 
which heavy logs can be pulled in from a distance on either 
side of the cable, and lifted and carried upon the car, is 
remarkable. The large scow shown is put in motion 
quickly ; boats can be pulled at the rate of six miles per 
hour with the motor now made. 

70. The method of hauling canal boats in trains by a 
large locomotive running along the banks would perhaps 
be somewhat objectionable on the Erie Canal. The track 
—either surface or elevated — ^and the hauling locomotives 
would be quite large, and would interfere with the present 
methods of proj)elling by horses. Such an arrangement 
would also take up both banks of the canal and would be 
costly to install. 

There would seem to be a number of objections to plac- 
ing a track upon the banks of a canal and hauling the 
boats in trains by large locomotives running on the track. 
There is, however, no particular reason for adhering strictly 
to this plan. Instead of operating large hauling loco- 
motives in the manner described, a small motor could be 
supported by a light structure running parallel to the 
canal and one or two boats could be pulled by each motor. 
As the motorman would be dispensed with in such an 
arrangement, provision must be made to regulate the motor . 
in some other maimer. The supporting structure should 
be light and strong and not interfere with the present 
methods of propulsion. 

Such a plan has been devised and proposed by Mr. 
Joseph Sachs. It contemplates erecting a suitable struc- 
ture along the canal and supporting a rail or similar de- 
vice thereon. This structure is arranged so as to permit a 
continuance of the present methods. The hauling motor 
runs upon the rail so that it cannot be displaced and is 
connected to the boat by a tow line or otherwise. The dis- 
tinctive fea^tures of this plan consist in regulating the 
hauling motor from the boat to which it is connected 
and preferably pulling each boat by a single motor, 
whether the boats are connected or not. This admits of 




















maMng the motors comparatively small, and of erecting a 
small, light structure. 

The current is supplied by wires supported on the struc- 
ture which also forms the return circuit. It will readily 
appear that a great variety of supporting rails and struc- 
tures for the motor could be devised, and several different 
forms are illustrated. Fig. 72 shows a type of duplex 
structure which permits motors going in opposite direc- 
tions to run upon a structure on only one side of the canal. 
Ranging along one bank of the canal is the supporting 
structure to which the rails, upon which the hauling motor 

Figs. 72a and 72b. — "Bicycle" Motor Hauler. 

runs, are attached. This structure consists of iron or 
wooden posts, a, standing about 8 to 10 feet above ground 
and very solidly planted to a depth of at least 6 feet. 
These posts are erected from 15 to 30 feet apart, as may be 
found necessary. Fastened to the upper portion of each 
post are the yokes, b, which face the canal and to the three 
arms of which the double headed rails, o, are fastened. On 
the opposite side of the supporting posts or pillars are 
suitable arms and insulators for various wires, and fastened 
to the central rail is the contact rail which furnishes cur- 
rent to the motors above or below. These motors or haul- 


ers, it will be seen, are small "bicycle " electric locomotives 
which are run between the middle and upper and middle 
and lower rails. Traction is obtained by pressing the 
lower flanged rollers or wheels of the hauler against the 
lower rail upon, which the motor runs. This is accom- 
plished by causing the upper or pressure roller to exert an 
upward pressure, thereby forcing the lower traction wheels 
into intimate contact with the lower rail. The upper and 
lower motors travel in opposite directions and are regu- 
lated from the boats to which they are connected by an 
ordinary tow line. 

It is unnecessary to go into the detailed construction of 
this particular plan. It will readily appear that this struc- 

FiG. 72o. — *< Bicycle" Motor Haulage, with Regulator 
ON Carriage. 

ture can be suitably constructed and erected along a canal. 
A view of the motor and carriage is given in Figs. 72a and 

The current is taken by the trolley arm from the supply 
rail and is brought to one terminal of the motor, the other 
being connected with the rheostat, which is in this case 
placed on the boat ; and the other wire from the rheostat 
goes to the frame and to the rails which are used as a 
return. A reversing switch may also be placed on the 
boat. In fact any ordinary motor regulator can be used, 
it being merely a question of putting the necessary wires 
into the cable connecting the motor with the regulator. It 
may, however, be found preferable to leave the regulator 
upon the motor. In fact, if any form of series parallel con- 
troller is used, it would be somewhat troublesome to run a 



cable containing a large number of wires from the motor to 
the boat. If such should be used, the regulator is placed as 
shown in the illustration, and the handle connected by a 
cord or chain to the boat (Fig. 72c). 

71. The illustration (Fig. 73) shows another form of 
duplex construction. In this form there are two rails, 
side by side, used for each motor, and the wheels of the 
motor are so arranged as to clamp or grip these rails; the 
upper being the driving wheels. The general idea of the 
previous plan, however, is adhered to, viz.: duplex struc- 

FiG. 73. — Double Rail Duplex Stbuctuke. 

ture on one bank of the canal, single contact rail for both 
motors, regulation from boat and structure or ground 

72. By the duplex structure arrangement, as shown, only 
one bank of the canal is used and the other is left free. As 
the towing path on the Erie crosses from one bank to the 
other, such an arrangement would meet with some diffi- 
culties. Instead of operating motors to run in both 
directions in this manner, a rail or support could be placed 
on each side of the canal on suitable elevated supports, so 
as to clear the towing path, and several such methods of 



construction are shown in the illustrations (Pigs. 74 and 
74a). The general operation and regulation would be the 
same ; and the plan in Fig. 74b shows the arrangement 
common to all these methods. 

We can, however, also support the rail upon which the 
motor runs, over the canal. Such a method has some 
advantages in affording a direct pull, but the cost of con- 
struction would perhaps be larger than for a structure on 
the banks. The illustrations show plans of supporting the 
motor over the canal by cables which would be a practical 
form of construction. 

Figs. 74 and 74a. 

In this method of canal boat propulsion, the light hauling 
motors run upon a rail, suspended by suitable wire cables, 
over the canal. Figs. 75 and 75a show transverse and 
longitudinal sections of a canal equipped with this method. 
Spanning the canal at intervals of from 50 to 75 feet are 
heavy wire cables supported from poles erected upon the 
banks. These poles are similar to the ordinary trolley 
poles used for electric railways, but of stronger construc- 
tion and stand about 8 to 10 feet above ground. Suspended 
longitudinally along the route of the canal are two cables 






fit I Ml," 




























which are fastened to, and supported by, the transverse 
spans. The rails upon which the motors are run are sup- 
ported by each of these longitudinal cables. 

To understand better how the motor is supported, Figs. 
75b, 75c, 75d and 75e are shown. Fig. 75b gives an end 
view of a hauling motor supported upon the suspended 
rail. A shows the transverse cable which supports the 
longitudinal cable, b. To this longitudinal cable the rail, c, 
is fastened at intervals of from 6 to 10 feet as shown. This 
rail weighs about 50 pounds per yard. The contact wire is 
supported beneath the rail and the circuit is composed of 
the wire and rail. The motor carriage, d, is supported by 
means of the rollers, e, which run upon the flange of the 
rail, 0, on each side of the supporting web. The carriage is 

Figs. 76b, and 76c. 

propelled by the flanged rollers, p f', which firmly grasp the 
sides of the rail and are rotated by the motors on the car- 
riage. More or less traction power can be obtained by 
altering the pressure of the flanged rollers, f p', against the 
rail. Instead of using an arrangement as shown, a rack 
and pinion may be used for the smooth roller and rail. 

Fig. 75o gives a plan view of the hauler carriage with the 
web of the rail in section. Fig. 75d shows a side view of 
the hauler carriage, rail and supporting cable. Fig. 75e 
shows the motors and gearing within the carriage. The 
gearing is so arranged that either or both motors can be 
used. Various other methods of transmitting power from 
the motor to the propelling rollers can, however, be used, 
and it would, perhaps, simplify matters to use but one 
motor instead of two. 



As will readily be seen, this method of motor haulage has 
various advantages. Such an arrangement would not in 
any way interfere with the present operation of the canal 
and could be economically constructed and maintained. It 
will be noted that the propeller method can also be operated 
without any addition to the structure shown. The trolley 

Figs. 75d and 75e. 

carriage would be so arranged as to run upon the rail, and 
contact would also be made with the wire. Current would 
be taken to the motor on the boat by means of a flexible 

73. An arrangement shown in Fig. 76 would perhaps be 
possible on some narrow, continuously level canals. As will 


be seen, a structure is built over the canal on which a high 
speed electric railroad can be operated, while the hauling 
motors for the boats in the canal are operated below the 
tracks as indicated. 

Instead of using the arches over the canal, as shown, the 
cable method of suspension could be used and would be 
particularly applicable here. 

Such an arrangement would cost a large amount of money 
to construct, but when its double use is considered and it 
is remembered that no grading, excavating or other work, 
required with a surface road, need be done, such a plan 
appears, to some, within practicability. 

A great variety of structures and supporting rails could 
be devised, but some form of overhead cable suspension or 
single rail structure on the banks would, no doubt, be a 
desirable arrangement. With any form of structure 
used, it would be most advisable to operate the boats as 
shown in Fig. 74b. It will be seen here, that there are 
two boats coupled together as in the present mule method. 
Each boat is pulled by a single motor and the motors can 
either be regulated separately on each boat, or from one 
controller on the forward boat. Such an arrangement 
permits of erecting a light structure just strong enough to 
withstand the strain of one boat. 

The chief difficulties in installing such a system of haul- 
age, as described, would be met with in switching, so as to 
permit the boats to pass one another going in the same 
direction and in passing under obstructions, such as bridges. 
These requirements can, however, be met with by suit- 
able switching devices and by erecting the motor support- 
ing structure in such manner as to permit of its being 
attached to the bridge. The passing of boats going in the 
same direction, does not, however, seem to be an absolute 
necessity as the boats would be continually in motion at a 
constant speed. Although it would, perhaps, be an advis- 
able arrangement on the Erie, it would not be so necessary 
on private canals. Suitable switching stations could be 
placed along the route so as to enable boats to pass one 
another going in the same direction. 

Such a system of motor haulers possesses a great many 
advantages. It does not in any way necessitate any addi- 























tional machinery on the present canal boats nor are any 
extra towing boats required which take np room in the 
canal and interfere with the locking. 

The method should prove more economical to operate 
than other systems, whether hauling or propelling. 

Any speed permissible by the canal structure can be 

There is no disturbance of the water as with the propel- 
ler, and injury to the banks is prevented to a very great 

It would not interfere with the present methods of pro- 
pulsion, as the structure could be so arranged as to clear 
the towing path. 

As a whole, this method of hauling canal boats by small 
motors which are supported on a light structure and con- 
trolled from the boat, seems to have various points of 
superiority. It is true, it also has various diflBiculties, but 
these should all be overcome by proper attention to design 
and construction. 

It is true that the cost of the structure would be in 
excess of that necessary for the propeller and also of that 
of one or two of the hauling systems. It would, however, 
appear that the various other advantages and sui)erior 
eflBiciency of such a system would warrant a larger cost of 

74. In all of the methods described, the storage battery 
has not, in any way, been considered. It could be applied 
to classes 1, 2, 3 and 5. Instead of taking power from a 
central source of supply, the barge, towing boat or motor 
locomotive might be supplied with storage batteries from 
which the current could be taken. Suitable charging and 
changing plants would be situated along the line. The 
immense cost of plant and operation, and diflBiculties in 
such a method, make it necessary to consider the central 
station and distributing wires as a source of supply for any 
practical system. 

It is true that a boat equipped with a storage battery 
and having a motor revolving a screw, would be entirely 
self-contained and could proceed anywhere without any 
assistance. The same thing could, however, be done by a 


steamer, either barge or tug. Therefore, the only possible 
advantage would be one of cost, and in that respect the 
superiority would api)ear to remain in favor of the steamer. 

Of the various methods of applying electricity to canal 
boat propulsion, the struggle for sui)eriority seems to lie 
between some form of propeller and one of the various 
hauling methods. From the requirements necessary for a 
practical system, it would appear that a separate motor 
tug would be the best form of applying the propeller. 

Of the various hauling methods, the operation of a small 
hauling motor on an elevated structure adjacent to the 
canal, towing or hauling the boat in the water and regu- 
lated from the boat, seems to be the most practical and 
feasible. Some form of cable suspension over the canal or 
single rail on each bank, would be, perhaps, the most 
practical form of structure to use ; although the duplex 
structure possesses many points in its favor. 

Aside from the question of economy, the exterior motor 
method certainly seems to have various advantages over 
thie propeller, as will readily be seen by carefully consider- 
ing both. When, however, the cost of operation of each of 
the two methods mentioned is also considered, the hauler 
method again seems to be superior to the propeller. In 
the following chapters an approximate estimate of the 
cost of installation and operation of both systems will be 



Generating Plant and Dibtribution. 

75. Before comparing further the methods of propellers 
versus motor haulage, let us look into the questions of gen- 
erating and distributing power for canal boat propulsion. 
The practicability of any electric system will greatly 
depend upon the cost of operating the prime mover, and 
the efficiency of transforming the energy into electricity 
and distributing it to the points of consumption, where it 
is re-transformed into mechanical energy. 

It is believed by many that the working pressure or 
voltage should not exceed 600 volts, and that preferably 
a simple two- wire system (either alternating or direct cur- 
rent) should be used. It may also be found that a direct 
current will prove most practical for the working circuit 
unless some efficient and simple single phase alternating 
motor should offer. The general electrical arrangement of 
the working or feeding circuit and the motors would be 
the same as in ordinary street railroad practice. 

We can, however, use various systems of distribution to 
secure this working pressure, which will greatly be gov- 
erned by the size of the canal, its location, kind of power 
available, etc. Whatever method of generation and dis- 
tribution is used, it must permit of continuous operation 
24 hours of the day. 

76. One of the simplest forms of generation and distri- 
bution would be to use the methods employed in ordinary 
electric railway systems, by generating and distributing 
the electric current directly. As the distributing distance 
would be limited by the cost of copper, it would be necessary 
to erect numerous stations along the route of canals that are 
of any length. The generating machinery could be operated 
by either steam or water power, as found most practical. 


77. Upon long canals, where it would be necessary to 
erect nnmerous stations, other methods of distribution 
could, however, be employed. Current could be generated 
in one or more large central stations situated at some 
cheap source of power, distributed at high tension, and 
reduced to the working pressure of the motors by trans- 
formers along the route of the canal. This plan permits 
of various modifications. We could, for instance, dis- 
tribute direct current at high pressure and reduce to the 
working voltage by motor transformers along the route, 
or a single or multiphase alternating current could be used 
and transformed to a direct current of lower pressure at 
the rotary transformers. 

In districts where cheap power and labor can be obtained 
along the route of the canal, the direct method of genera- 
tion and distribution would, no doubt, be used. Where, 
however, cheap power could not be obtained along the 
entire route some modification of the high tension method 
would be adopted. 

78. As the Erie and other canals in the higher latitudes 
are closed during part of the year, the most economical 
operation of plant could not be obtained if power were 
only furnished for canal boat propulsion. It would, there- 
fore, be most practical and advisable to arrange the dis- 
tributing and generating plant so as to operate at all 
times ; power being furnished for other purposes as well 
as canal boats. In the case of the Erie and some other 
canals, an electric railway could be operated parallel to 
the canal or over it ; and light, heat and power could be 
furnished by the same plant used to operate the boats. If 
an electric system is used it should also be applied to the 
operation of lock gates, etc., along the canal. 

79. In comparing the two methods of electrical canal 
boat propulsion, we shall roughly assume, for the sake of 
simplicity, that we are distributing power by the direct 
method with stations located along the route of the canal 
about 20 miles apart (distributing 10 miles each way from 
the station), and using steam as a motive power. This wiU 
necessitate about 18 stations along the route of the Erie 


canal. The station machineiy, engines and dynamos, will 
be arranged in units of say, two or three hundred horse- 
power each and one-third of the generating plant will be 
surplus over that actually necessary to furnish power for 
boat propulsion ; which will be necessary as the system 
must be kept constantly in operation. We shall only 
consider the use of the plant for canal boat propulsion. 

The cost of a combined steam and electric generating plant 
may be roughly placed at $100 per horse-power and coal to 
be consumed at the rate of 2i pounds per horse-power hour 
at dynamo terminals ; 20 per cent, will be allowed for loss 
in distribution. It next becomes proper to determine how 
far the expense will now be governed by the adoption of 
either the propeller or the motor haulage method, but 
before passing on to deal with that question, a comparison 
should be made between the steam driven electric genera- 
tors, just referred to, and the supply of power from Niagara 
to the canal. Messrs. E. J. Houston and A. E. Kennelly, 
electrical experts of high standing, have recently con- 
tributed to The Electrical Bngineer^ of New York, the 
subjoined data on the subject of Niagara power as com- 
pared with the highest class of compound condensing 

80. One of the most important questions of the day is, 
How far is it commercially possible to transmit water 
power electrically? It is recognized that the limiting 
commercial distance depends upon two associated inter- 
dependent factors ; viz., cost and electric pressure. 

(1.) Cost^ including the purchase and maintenance of the 
necessary machinery and wires together with the annual 
interest chargeable upon such expenditure. 
^ (2.) Electric Pressure. The pressure or voltage at which 
the line transmitting the power can be operated with con- 
tinued security to life, and assurance of permanence of 
supply, and permanent protection to the lines of conductor 
from lightning, weather and all disturbances. 

It is clear that if reliable machinery could be purchased 
cheaply enough, and the conductors could be safely oper- 
ated at sufficiently high pressure, the Falls of Niagara 
could to-day stop steam engines in New Orleans, La., by 


underselling their power. There must be a certain radius 
from Niagara, within which electrical power will carry- 
death and displacement to the ordinary steam engine, and 
it is desirable to ascertain how far this radius may be 
likely to extend. 

We shall first assume that a steady transmission of 
power is to be provided for, from Niagara Falls to various 
cities along the banks of the Erie Canal, as shown in the 
accompanying table. 

Distfuioe. Miles. k. w. h. p. 

To Buffalo 15* 32,500 30,160 

ToSyracuse 164 7,500 10,060 

To Schenectady 800 7,500 10,060 

To Albany 380 15,000 20,110 

To various points along canal, for barge pro- 
pulsion 15,000 20,110 

67,500 90,500 

* To outskirts of Buffalo city. 

We shall assume that the engineering diflBiculties can all 
be overcome by bare, overhead, tri-phase wires, at 85,350 
volts receiving pressure, with step-up and step-down 
transformers at each end of the lines. It remains now to 
consider only the question of cost. 


We require to estimate the annual cost of producing one 
kilowatt steadily at the turbine shafts at Niagara Falls. 
It has been stated that the estimated cost of the hydraulic 
works at Niagara for a total of 119,000 h. p., 

Was $ia50per H. p. or $14070 per K. w. 

If we add as much as 20 per cent, for any 

possible miscalculation 3.10 " " 3.814 « 

And also add cost of turbines at 5.00 ** " &70S " 

$17.60 $2a587 

Allowing 5 per cent, for interest on capital, 2i per cent, 
for superintendence and repairs, and 2i per cent, for 
depreciation, the total annual cost is 10 per cent, on invest- 
ment, or $1.76 per h. p., or $2,359 per kilowatt, assuming 
that the turbines are always running at full load. This, how- 
ever, is more than can reasonably be expected. The aver- 
. age load under favorable conditions can hardly exceed 60 
per cent, of the maximum load handle, joaaking the annual 


cost $2,982 per h. p. or $3,931 per kilowatt of average annual 

Another line of reasoning is as follows: — It has been 
stated by Prof. Forbes that he did not continue to add cost 
to his generators, when $60 of extra expenditure in them 
did not increase their electrical output by 1 h. p. Prof. 
Forbes also mentions that he took 5 per cent, as interest 
on investment, so that it seems clear that his estimate of 
the annual cost of producing one horse-power at generator 
terminals is 5 per cent, of $60, i. e., $3 per h. p. or $4,021 
per K. w. ; and eliminating the cost of generators, this 
would be about $3.72 per x. w. at turbine shafts. How- 
ever, as our purpose is to be conservative, it may be safer, 
in the absence of actual assurance, to take $4.00 per x. w. 
a« the cost of power of turbine shafts per annum ($2,984 
V^x H. p.). 


The cost of alternators, as is well known, varies greatly 
with their size and capacity. In sizes below 3 x. w. their 
purchasing cost may be $100 per x. w., while in sizes of, 
aay, 60 x. w. they may cost $45 per x. w. Finally in very 
large sizes, say of 4,000 x. w., their purchasing cost would 
probably be reduced to $8,616 per x. w. ($6,352 per h. p.), 
including exciters, and all station apparatus. 


The cost of motors may be expected to average slightly 
higher than that of generators, for the reason that they 
will frequently require to be made in smaller sizes (say, 
for 1,000 X. w.) than the generators. The purchasing cost 
per kilowatt may be taken as $9,581 per x. w. ($7,146 per 
H. p.) with exciters and apparatus included. 


The step-up and step-down transformers would probably 
not cost more than $5,157 per x. w. ($3,847 per h. p.). 
Prof. Forbes has stated that offers as low as $3.52 per h. p. 
($4.72 per x. w.) have been made to deliver such trans- 
formers. These transformers may be conveniently regarded 
as forming part of the generators or motors at each end of 
the lines. In this regard, allowing 0.96 as the efficiency of 


geneiufors or motors, the purchasing cost of generators and 
all accessories becomes $13,888 per k. w., or $10,863 pef 
H. p. Purchasing cost of motors $14,963 per K. w. or 
$11,164 per H. p. 

The efficiency of complete generator and motor plants 
becomes 0.94 at full load, allowing 0.98 as efficiency of 


It is obvious that the right amount of capital to expend 
in line conductors and construction, after the voltage and 
frequency have been decided upon, is that such additional 
expenditure just ceases to save its value in interest, by 
reason of the consequent reduced loss of power in the line. 
In cases where the expenditure in the line and construc- 
tion is in proportion to the weight of copper in the line, 
this economical point must be attained when the annual 
charges of interest, repairs and depreciation on the whole 
line, are equal to the money value of the power lost in 
transmission along the line, which is Kelvin's law, and is 
applicable to alternating as well as to continuous currents 
and whether such currents have lag or not. Usually, how- 
ever, the expenditure upon the line does not increase 
directly with the weight of conductor, and the first 
statement is, therefore, more generally applicable than 
Kelvin's law. It can be shown that with the prices above 
mentioned, the most economical weight of copper to 
employ on each main tri-phase conductor, will be about 
20i pounds per ampere per mile. 

Allowing for the best quality of oil insulators, and bare 
wires, not exceeding No. 000 A. W. G., a first-class ordin- 
ary pole line can be erected for $1,000 per mile of line; and 
if this allowance be made for the line to each city, the 
combination will admit of a substantial and permanent 
structure in wood or in iron, capable of holding all the 
wires safely, and in a manner practically safe from light- 
ning, yet readily accessible to linemen for repairs. By prop- 
erly grading the distances between conductors, the antag- 
onistic effects in the lines of capacity and inductance can 
usually be rendered negligible, so that the fall of pressure 
in the lines becomes that simply due to ohmic resistance. 


By systematically transposing the wires of all circuits, say 
at every mile, their mutual inductance would be negligibly 
small and their capability of disturbing one another would 
be consequently annulled. 


The following staflP might be expected to handle the 

One general superintendent and aasistantB $90,000 per annum. 

One chief engineer 20,000 " 

Twelve assistant engineers 36,000 ** 

Forty dynamo assistants 40,000 " 

Forty linemen and assistants •••••••••• 40,000 " 

Office expenses, salaries, taxes, etc 34,000 ** 

Total. $200,000 " 

Or $a00 per H. p. of total capacity, or $2,681 per k. w. 

Distance, 830 Miles ; Maxdcum Deuvxbt, 15,000 k. w. 

For a delivery with full load at motor shafts of 1.000 k. w. 

The delivery at line fcerminals 1.064 '' 

And with a power factor of 0.9, .the received current on 

each conductor is 0. 01931 amp. 

Copper per mile on each conductor 0.3958 lb. 

Total copper in line, per K. w. delivered 391.9 lbs.* 

Approximate loss of energy in line 0.862k. w. 

Approximateenergy delivered at Niagara terminals-... 1.926 " 

Approximate energy delivered at Niagara turbines 2l048 ** 

Cost of 1.926 K. w. generator at Niagara at $1&888 per 

K. w. $26,750 per K. w. 

Cost of 391.9 lbs. of copper in lines 4a987 " " 

One Idlowatt motor at Albany at $14.953 14958 '< « 

$90,690 " " 
Cost of 330 miles of line at $1,000 for 15,000 K. w 22.000 " " 

Total investment per k. w. delivered $112,690 " " 

Annual cost of interest, depreciation and repairs at 10 

percent 11.269 " " 

Cost of 2.048 K. w. at turbine shafts at $4. &192 " " 

$19,461 " " 
Labor, superintendence and general expense. 2.681 " " 

Annual cost of delivery of one K. w. at sustained full load. $22. 142 " " 
At average of 60 per cent of full load, $27.53 per h. p. or $36,903 ** " 

* The number of wires in the line to Albany would be nine, about No. 000 B. & S. gauge* 
three on each main conductor. 


They next give corresponding calcnlations for Syracuse 
and Buffalo, and then summarize as follows : 


Cost of delivering a max. of 22,500 K. w. at Buffalo. . .at $ia 694 1240,610 
Cost of delivering a max. of 7,500 E. w. at Syracuse, .at IT. 223 129,170 

Schenectady 7,500 k. w at 21.260 159,450 

• At Albany 15,000.K. w.... at 22.142 832,130 

Different points 15,000k. w at 17.830 267,450 

67,000 $1,128,810 

Average cost of delivery, full load $13,475 per h. p. $16,723 per k. w. 

Average cost of delivery 0.6 full load. . .$20i792 per H. p. $27,872 per K. w. 
By Emery's tables the cost of generating 

steam power per f^Tmnnn with coal at 

$3 per ton is for 308 days of ten hours. $SS6,27 pern. p. and $3a88 per k. w. 
365 days of 20 hours. $44.43 per H. p. and $59.56 per K.W. 

This is with large triple expansion compound engines. 

The foregoing results indicate that on the basis of prices 
and voltages assumed and detailed, the power of Niagara 
Falls can be transmitted to a radius of 200 miles cheaper 
than it can be produced at any point within that range by 
steam engines of the most economical type with coal at $3 
per ton. That Niagara power can maintain at Albany a 
large day and night output cheaper than steam engines at 
Albany can develop it, but that for power taken at Albany 
for 10 hours per diem the best steam engines have some- 
what the advantage over Niagara unless exceptionally 
favorable conditions of load could be secured for Niagara 

These conclusions are, of course, entirely dependent upon 
the reliability of the prices, voltages and estimates as 
detailed above. The prices, however, appear to be con- 
servative, and are probably in excess of the latest market 
values. This would also appear from the statement recently 
published that the Cataract Construction Company has con- 
tracted for the privilege of right of way for conducting 
lines along the Erie Canal banks, by supplying power for 
barges at the annual rate of $20 per h. p. ($26.81 per kilo- 
watt) whereas the above figures make the average cost of 
production $20,792 per h. p. ($27,872 per k. w.) on the basis 
of an average output of 0.6 full load ; so it would appear 
that the Niagara Power Company are in the possession of 


better prices. Moreover, if the voltage delivery on the 
long lines, here taken as S6,860 at the receiving end, can 
ever be safely increased, a marked reduction in the cost of 
delivery would naturally result. 

The broad conclusion to which an inquiry of this nature 
inevitably leads, is that while under ordinary conditions 
the commercial limit of electrical transmission of power 
from water powers of less than 600 kilowatts can hardly 
exceed fifty miles, the radius at which it will be profitable 
with good fortune and management, to electrically trans- 
mit a water-power aggregating 50,000 k. w., or more, is, 
I)erhaps, to-day, two hundred miles, and that it might be 
commercially advantageous for such a large water power 
to undersell large steam powers at twice this distance with 
no profit, in order to reduce the general expense upon 
delivery nearer home. The reason for this difference in 
the transmission radius between small and large water 
powers, lies obviously in the fact that electrical and 
hydraulic machines can be built and purchased much more 
economically in large sizes than in small, so that the cost 
of producing and of maintaining one kilowatt is very much 
less for large than for small water powers. 

We may add here that these interesting calculations 
have been variously discussed and criticised, and that 
several dissenting opinions of weight have been expressed. 
Messrs. Houston & Kennelly have nevertheless adhered to 
the figures presented, maintaining their accuracy, and our 
readers will probably soon have the opportunity of seeing 
in how far this confidence is justified. 



Be8istakcb of Canal Boats. — Compabison of Cost, 
Propeller m. Hauler. 

81. In order to compare the two methods of propelling 
canal boats — ^the propeller and hauler methods — we shall 
consider the installation of an electric system of boat pro- 
pulsion on the Erie Canal, assuming 350 miles as the total 
length. We shall assume that the canal is open for 215 
days of the year and that we have at least one foot more in 
depth of water throughout the canal, giving a standard 
depth of 8 feet, and that the locks are all lengthened so that 
two boats can pass through at one lockage. (See page 101.) 

Upon this improved canal we shall assume the operation 
of 600 barges (300 going in each direction), each carrying 
340 tons of freight. For the sake of simplicity, we shall 
assume that these boats are continuously moving at an 
average speed of 3 miles per hour. Such would not acta-' 
ally be the case, as a number of boats would be waiting at 
the various locks and, therefore, taking no power. The 
percentage of all the boats on the canal that would be 
locking at one time would be small. The total amount of 
energy required to operate each boat the entire length of 
the canal, would not be seriously affected by the locking; 
but to allow for the lockages and also for the smaller 
number of boats at the beginning and ending of the season, 
we shall assume that all the boats are taking power for 200 
days. The full number of boats would gradually be 
reached at the opening, and decrease at the closing, of the 
canal. It will, therefore, be necessary to consider the 
operation of the power plant for the fuU time of 215 days. 

Resistance of OanaZ Boats. 

82. As we are going to operate the ordinary barges a* 
present in use (98 feet long, 17i feet wide and 6 feet 



draught, loaded to 240 tons) by electric methods, let us 
look into the resistance met with by these boats on the 
Erie Canal under different conditions. The law of boat 
propulsion which could be applied to boats in a broad 
expanse of water is not applicable here, and the contracted 
waterway enters greatly into the computation. Ex-State 
Engineer Elnathan Sweet has given in a paper read before 
the American Society of Civil Engineers, March, 1890, a 
table compiled from various experiments made by him on 
the Erie Canal. This table is shown below. 

Traction Exfebimsnts on Ebie Canal, 1878, Nbab Canajo- 
HABiE, WITH Canal Boat Henby L. Pubdy. 








Observied. Computed. 


H. P. 





8 00 









7 ft. 


8 ft. 

8 ft. 

8 ft. 

6 ft. 

6 ft. 

1.5 ft. 

7 ft. 
































These experiments were made in the fall of 1878, with 
the boat "Henry L. Pnrdy," which fairly represents the 
average canal boat on the Erie Canal. The boat was towed 
by horses and the resistances were measured by a dynamo- 
meter and are shown in column marked " observed." The 
resistances shown in column marked "computed" were 

deduced from the equation R = 

0.10303 s v^ 

units being 

r — 0.597 

in feet, pounds and seconds. The slight differences be- 
tween the observed and computed resistances were caused 
by the rapidly varying power exerted by th^ animals. 


The boats similar to the one experimented with, have a 
horizontal cross-section of a little over 1,600 square feet 
above the 6 foot immersion line, and carry 25 tons for 
every half foot submersion beyond 6 feet. The submerged 
surface of these boats when drawing 6 feet is 2,870 square 
feet ; when drawing Qi feet, 2,980 square feet ; and 3,090 
square feet when drawing 7 feet of water. The resistances 
at various speeds for boats of similar model can readily be 
computed from the data obtained from the table and by 
using the formula (also given in the paper) of Scott Rus- 
sell and Du Buat for the resistance of boats in narrow 

channels of water, viz.: Ji = — ; — =- in which 

r + B 

R = Resistance of boat. 

s = The submerged surface of boat. 

V = The velocity of boat. 

r = The ratio of the channel's water section to immersed 
section of boat. 

A and B constants det/ermined by experiment. 

Mr. Sweet found the most probable values of A and 5, 
for the model of boat used, to be, 

A = 0.10303. B = — 0.597. 

The equation therefore reads B = 0-^0303 s v' ^ ^^ 

T — 0.697 

used in the table ; the units being in feet, pounds and 
For velocities in miles per hour, this becomes, 

J. _ 0.2215 ^t?« 
r — 0.597' 

Using the latter equation,, the table on page 172 has been 
compiled showing the resistance of the ordinary canal boat, 
as used by Mr. Sweet, at one mile per hour with various 
boat immersion and canal prism sections and depths of 



BxnsTAHOB or Cakai, Boats at Onx Mils Pxb Hova; 


i2 = 

0.2216 9 9* 
r — 0.597 



Lbs. H. P. 


7 ft 

7 ft. 













It will be noticed that the resistance is about 15 per cent, 
less in 8 feet depth of water than in 7 feet. 

Taking the resistances shown in this table at one mUe 
per hour, we can readily find the resistance for any other 
speed for the same conditions. As the resistance of boats 
through water increases as the square of the speed, 
Ji = L v^ when H = the resistance, L = pounds resistance 
at one mile, v = velocity in miles per hour. As the power 
necessary to overcome this resistance increases as the cube 
of the speed, ff. P. = P v^ when ff. P. = -horse-power to 
overcome resistance of boat, P = horse-power necessary at 
one mile per hour, and v = velocity in miles per hour. 

We can now very readily get the resistance of the 
ordinary canal boat as used on the Erie under the condi- 
tions previously named ; viz. : 8 feet depth of water, 6 feet 
immersion of boat, speed 3 miles per hour. By taking the 
resistance at one mile per hour under these conditions at 

146 pounds or -—- horse-power, as shown in the table, we 

find that at 3 miles per hour, the boat would encounter a 
resistance of 1,305 pounds, or it would take 10.4 horse- 
power to move the boat. Let us say, in round figures, 10 


83. It has been found most economical, in operating boats 
on. the Erie Canal, to connect them in pairs, as already 
described. We shall, therefore, endeavor to adhere to this 
general plan and arrange the electric methods so as to 
operate the boats in pairs and, also, so aa to facilitate 

For the propeller method we shall consider the plan 
•hown in Fig. 54, in which a small tug is used, pushing 
one freight barge and pulling two, making a fleet of three 
barges each carrying 240 to 250 tons, with a total of 750 

For the hauler method we shall consider the plan as out- 
lined in Fig. 72, in which a small motor is operated on a 
light structure adjacent to the canal, each motor pulling a 
single boat and being regulated from the boat. It will be 
more advisable to operate only two boats by this plan. 

Propeller Method. 

84. As we are going to oi)erate 600 boats (300 each way), 
we will have 200 three-boat fleets or trains. Allowing 10* 
horse-power as the power necessary to overcome the resist- 
ance of each boat, and because the resistance increases 
with the surface, we will have 30 horse-power as the total 
power necessary to overcome the resistance of each fleet of 
three boats (excluding the tug). 

There will, however, be power used above this which is 
necessary to propel the tug, and that which is lost in the 
slip and friction of the screw. As is well known, the loss 
of power in a screw propeller, in ordinary cases, amounts 
to from 15 to 30 per cent, of the power applied at the shaft. 
With the steam canal boats used on the Erie Canal at 
present, this loss is much larger, amounting to from 50 to 80 
per cent, (with an average of about 65 per cent.). By sub- 

* The aUowanoe of 10 h. p. per boat necessaiy to propel it at three mJles per hour is low. 
As has been shown previously, this is the power necessary to OTercome the actual resistance 
of the boat at the above speed and does not take into consideration the additional resistance 
caused by- the propeller. The augmented resistance due to the screw propeller would, no 
doubt, add greatly to the actual hauling resistance which is considered here. The prop^ler 
increases the resistance of the forward boat by preventing the water from fully reacting 
upon its stem, and causes an increased resistance in the rear boats l^ the stream that is 
thrown astern by it. 


stituting an electric motor for the steam engine on the 
canal boat, this loss could perhaps be somewhat reduced. 
The ordinary slippage of a screw propeller increases when 
other boats are towed, and it is doubtful whether the losses 
in slip and friction of the propeller screw could be reduced 
to any great extent in an electrical method if the screw 
and motor are located on the canal boat. 

If, however, the propeller screw is placed upon a small 
tug, as in this case, and revolved at a higher speed, thereby 
allowing the use of a smaller screw which is kept con- 
stantly immersed, this loss could be somewhat reduced. 
It will perhaps be fair to assume that in the case under 
consideration, the combined losses in the screw and power 
to propel the tug, will amount to 45 per cent, of the power 
at the motor shaft. We must, therefore, have 45 per cent, 
more power at the motor shaft than is finally necessary, in 
order to overcome the resistance of the canal barges; and 
as this is 30 horse-power, the power necessary at the motor 
shaft on the tug would be 55 horse-power. Allowing a 
loss of about 10 per cent, in the motor, we have, in round 
numbers, 60 electrical horse-power necessary at the motor 
terminals. This would be the power taken by the motor 
when the boats are loaded to a depth of 6 feet immersion, 
each carrying 240 tons. 

Upon the Erie, the boats going east are generally fully 
loaded, while those going west carry but a small amount 
of freight. We may fairly assume that the average power 
necessary at the motor terminals for westward fleets will 
amount to about 20 horse-power. Although this may not 
be strictly correct, it will answer in the present calculation. 

As there are 100 fleets going in each direction we would 
require a total of 8,000 horse-power delivered at motor 
terminals. Allowing 20 per cent, loss in distribution, we 
find that it would be necessary to have 10,000 horse-power 
at the dynamo terminals at the generating stations in order 
to propel these 600 boats. This will give an efficiency of 
about 40 per cent, for the system from the dynamo to the 
actual power used to propel the boats. If we have 18 sta- 
tions along the canal, each station would have an output 
of about 655 horse-power. The machinery installed, how- 
ever, would be about 50 per cent, in excess of this, or about 


800 horse-power. The entire generating capacity would be 
about 16,000 horse-power. 

85. Taking the coal consumption at 2^ pounds per elec- 
tric horse-power hour delivered at the dynamo terminals, 
we get a total coal consumption of 26,000 pounds per hour. 
At this rate we get about 3.6 per horse-power hour at the 
motor shaft and about 6i pounds per effective horse-power 

As a rough approximation of a propeller plant under the 
above conditions, with a steam generating plant, we may, 
therefore, take the following : 

Cost of Electric Propeller Plant. 

Distributing system (poles, wires, feeders, 
etc.), 850 miles at $6,000 per mile, - - $2,100,000 

Generating plant (buildings, steam and 
electric), 16,000 h. p. at $100 per h. p., 1,500,000 

Propelling tugs, 200 at $4,000, - - - 800,000 

Superintendence, engineering and inci- 
dentals, 200,000 

Total cost, $4,600,000 

We can also roughly approximate the cost of operating 
this plant for 216 days. 

Cost of Operation. 

Labor to operate tugs, one man per tug, - $76,000 

Labor (station, line, superintendence and 

office), - - 100,000 

Coal, 60^000 tons at $3 per ton, - - - 180,000 

Interest on investment, 4 per cent., - - 184,000 
Maintenance, depreciation, etc., ] 005 ooo 

oil, water, waste, insurance, etc., ) ' 


Let us call this $766,000. 

At this approximate figure, power will cost about 2 cents 
per horse-power hour delivered at the motor shaft on the 


tag. ThiB will make the cost of propelling a fleet of three 
boats, going at three miles "per hour, each loaded to 240 
tons by the method described, as follows : 
86.6 cents per mile, 
12.2 " " boat mile, 

$128.10 for entire trip, Buffalo to Albanjr (loaded). 
Going in the opposite direction (lightly loaded) propul- 
sion will cost : 

12 cents per mile, 
4 " " boat mile, 

$42.00 for the entire trip west from Albany to Buffalo. 
This would make an entire round trip under these 
conditions cost about $170. 

The cost of propulsion for each of these boats for an 
entire round trip would be about $57. 

Hauler Method. 

86. Let us now compare with the above the method of 
hauling as outlined in Fig. 72. In this method we shall 
operate but two boats in each fleet ; each boat pulled by a 
single motor. As we have 600 boats, we will have 300 two- 
boat fleets or trains, 150 going in each direction. As each 
boat encounters a resistance of about 1,305 pounds (10 
horse-power at three miles per hour) the motors must 
exert an effective puU to overcome this. Allowing a loss 
of 20 per cent, in the gearing of the hauler, we find that 
each motor must furnish 12J horse-power at the shaft. 
Each of the motors would, however, be of about 15 horse- 
power capacity. Taking 10 per cent, loss in motor, as 
before, we get 14 horse-power of electrical energy necessary 
at the terminals of each motor. 

This would be for loaded boats. For light boats going 
in the opposite direction, we shall assume, as in the previ- 
ous case, about one-third of this or about 6 horse-power. 

As we have 300 boats going in each direction, and as 
each boat is operated by a single motor, it will be neces- 
sary to furnish a total of 6,700 horse-power at the motor 
terminals. Allowing 20 per cent, loss in distribution, the 
power needed at the terminals of the generating dynamos 
will be 7,125 horse-power. The entire capacity of the gen- 
erating plant would be about 10,700 horse-power. The 


efficiency of the entire system from dynamo terminals to 
power actually used to propel the boats is about 68 per 

Although we could operate this plant with a smaller 
number of stations than in the previous case, we shall 
assume that the same number of stations are to be used, 
but of smaller capacity. Each station would, therefore, 
have an output of about 400 horse-power and a capacity of 
about 600 horse-power. 

At the rate of 2i pounds of coal per horse-power hour 
delivered at the dynamo terminals we also get the same as 
before, 3^ pounds per horse-power hour at motor shaft, but 
the coal consumption per effective horse-power hour would 
be less than in the previous case ; amounting to 4i pounds 
per horse-power hour. 

We may consider the following as a crude approxima- 
tion of the cost of the hauler method. The supporting 
structure considered is a light, single support on each 
bank of the canal, or else some form of cable suspension 
as shown in Fig. 76, 

Cost op Electric Haijler Method. 

Structure and distributing system, 360 
miles, at $10,000 per mile, - - - $3,600,000 

Generating plant, 10,700 horse-power at $100 
per horse-power, 1,070,000 

Motor haulers, 600 at $1,000 each, - - 600,000 

Superintendence, engineering, etc., - 200,000 

Total cost, $5,370,000 

The cost of operating this plant under the same con- 
ditions as before would approximate about as follows : 

Cost op Operation. 

Labor (station, line, superintendence and 

office), $100,000 

Coal, 42,750 tons at $3.00 per ton, - - 128,250 
Interest on investment at 4 per cent., - - 214,800 
Maintenance, depreciation, etc., oil, waste, 
insurance, etc., 200,000 



In round figures, $640,000. 

Taking these figures we may form an estimate of the 
cost of propulsion by the hauler method. The cost per 
horse-power hour delivered at motor shaft will be about 
2.6 cents. 

It will, therefore, cost 10.8 cents per boat per mile (21.6 
cents for each pair of boats or fieet). This will make the 
cost of the entire trip east of two boats, each loaded to 240 
tons, going at 3 miles i)er hour, about $76.50. Going in the 
opposite direction the cost of propulsion would be about 
one-third, or about $25.00. The entire round trip of two 
boats by this method would cost about $100.00; or a 
continuous round trip of a single boat would cost $50.00. 

Comparison: Propeller vs. Hauler. 

87. From these crude estimates we have, therefore, the 
following : 

Propeller. Haulage. 

Cost per boat mUe (east), - - 12.2 cents. 10.8 cents. 

" " " trip (east), - - $43.00 $38.00 
" '' " " (east and west), 57.00 50.00 

or as a comparison of the total operating expense of 600 
boats for 200 days at 3 miles i)er hour, we have, 

Propeller, $765,000 ; Hauler, $645,000 ; 

making a difference of $120,000, or 15 per cent, in favor of 
the hauler. 

We have considered in both cases a very crude arrange- 
ment of generating and distributing plant, and in the latter 
method particularly, the general plan would be somewhat 
uneconomical. In fact if the plant were properly planned 
a larger saving could be shown in favor of the hauler 
method. It must also be remembered that we have con- 
sidered that the resistance of the boats is the same in both 
cases. This is not a fa<3t, as the resistance of the boats 
operated by the propeller tug would be greater than by 
the hauler method. If we take these facts into considera- 
tion, and furthermore, if such plant were installed by the 
State and no profit was expected on the invested capital. 


it will readily be seen that the advantage in favor of the 
hauler method would be still greater. 

Comparison With Present Methods. 

88. With this rather rough approximation it would 
hardly be fair to make any depreciatory comparison with 
the present methods. With the actual plant that would 
be installed with either method, the cost of propulsion 
would no doubt be brought down to quite some extent. It 
must also be remembered that but a very small and limited 
output has been considered. An increase in the number of 
boats propelled and a continuous operation of the gen- 
erating plant — power being also used for other purposes — 
would also greatly decrease the cost of propulsion. An 
approximation of the cost of propulsion by the present 
methods would, however, be interesting. 

At the present speed of 2i miles per hour in 7 feet of 
water, we may take the following as the cost of propulsion 
per boat mile by tjie present steamer-and-consort method. 
Each barge carries 240 tons and steamer 180 tons. 



Coal, 25 pounds at $4.25 ton, - - - 5 cents. 
Engineer's pay and board, - - - 2 

Oil, waste, etc., 1.5 

Interest, repairs, insurance, etc., on propel- 
ling machinery, - - - - 1.6 ** 

Total cost per boat mile, - - - 10 cents. 

We cannot, however, compare this but must reduce it to 
the same conditions as in the electric method, namely 8 
feet of water and 3 miles per hour. 

Referring to the table on page 172, we find that there is 
a saving of about 15 per cent, in the resistance of a boat 
loaded to 6 feet when moving in 8 feet of water instead of 
in 7 feet. Taking the coal consumption in 7 feet at 25 
pounds per boat mile, at 2J miles per hour, we find that in 
8 feet it would be 21i pounds. As the power necessary to 
propel a boat increases as the cube of the speed, at three 
miles per hour, in 8 feet, we would get about 37 pounds per 


boat mile. As the other conditions will alter but slightly, 
we will assume them to be the same ; and we now have 

Coal, 37 pounds at $4.26 per ton, - - 8 cents. 

Engineer's pay, board, etc., - - - 2 " 

OH, waste, etc., 1.6 " 

Interest, repairs, insurance, etc., - - 1.6 " 

Total cost per boat mile, - - 13 cents. 

This figure is, however, more likely to be about 15 cents 
per boat mile. As the cost of towing by horse or mule is 
higher than by steam it will be unnecessary to consider 
it in such a comparison. 

Taking then the various methods, we have the following 
as an approximate estimate of the cost of propelling the 
ordinary canal boat on the Erie Canal at three miles in 8 
feet of water, loaded to 240 tons. 

steam propeller, 

13 cents per boat mile. 

Electric " - - 

- 12 

" hauler, 

10.6 " " 

It would appear from the above, and from the fact that 
the estimates are very crude and that the cost of operation 
can be decreased to a large extent by an economical and cor- 
rect design of plant, that an electric system can be economi- 
cally operated. It would also appear that the hauler 
method had the advantage in all directions. It is difficult 
to make any positive statement as to ultimate cost of oper- 
ating such a method, but the cost could perhaps be 
decreased 20 per cent, below the figures given even if we 
limit ourselves simply to numerous generating stations 
using coal, and not to the cheap power from Niagara, which, 
if the calculations of Messrs. Houston and Kennelly be cor- 
rect, can be delivered even in Albany on terms of more 
than an equality with the energy from the best triple 
expansion engines on the spot. 



Propulsion: Resistance op Boats and Propellers; 
Paddlewheels and Screws. 

89. The subjects treated in this chapter relate not only 
to electric launches, but to the other classes of boats dealt 
with in the present volume, including canal boats. It has 
been deemed proper, to discuss them broadly, leaving 
fuller treatment to books on the specific topics that are 
here passed in review. As will have been noted, some 
data on the subject, as applied to canal boats, has been 
given in the previous chapter. 

Boats can be propelled through water in two ways ; first, 
by exerting a direct pull or strain on the boat sufficient to 
overcome its resistance ; second, by the use of a propelling 
instrument on the boat which acts upon the surrounding 
water in an opposite direction to that in which the boat is 
to be moved. 

The first plan has been illustrated in the previous chap- 
ters by the cable towing and motor hauling methods on 
canals. Such methods of propulsion are limited to canals 
and rivers. The second plan of propulsion is, perhaps, the 
more important and, in fact, the only way of giving motion 
to a boat through water, excepting as above. The principle 
involved in any marine propeller is the projection of a 
quantity of water in a direction opposite to that in which 
the boat or vessel is to move, and of such mass as is neces- 
sary to overcome the resistance of the boat. 

The resistance of a body, such as a boat, moving through 
a broad expanse of water depends upon quite a number of 
conditions which are quite variable. The total resistance 
encountered by a boat is made up of the following : 

i^i>^^.— Erictional resistance due to the surface friction 
between the water and the skin or surface of the boat. 
- Sec(md.-T-WQ.Ye making resistance due. to the cutting of 


the water by the bow of the boat which sets up waves 
called waves of displacement. 

The f rictional resistance is by far the most important in 
boats of ordinary good lines and moderate speed. It 
depends upon the area of immersed surface and its degree 
of roughness, and varies about as the square of the speed. 
It is also affected to some extent by the length of the boat, 
but is not affected by its form or proportions except in 
extreme or unusual cases. Even with high speeds, the 
frictional or skin resistance is the largest part of the total 
resistance encountered by the boat. 

The wave making resistance depends upon various ele- 
ments, but particularly on the form and proportions of the 
boat. With a boat of good model and fairly fine lines, at 
fair speed, the resistance caused from waves would amount 
to but a very small percentage of the total resistance. 

There is, however, a limiting speed for a boat of any 
particular model, above which the resistance increases 
very rapidly. In some large poorly built boats a slight 
additional resistance is caused by eddies formed at the 
stern. In boats of good lines, however, this is quite 

Various methods have been used to deduce the resist- 
ance of a boat of any particular model or form at a certain 
speed, all of which are based for their accuracy upon 
certain practical data obtained from experience with other 
boats of similar model. It is quite difficult to lay down a 
general formula of resistance for all models of boats, as the 
wave resistance of a boat of narrow beam going at a 
certain speed would be less than that of a boat of the same 
displacement and midship section going at the same 
speed, but of broad beam. It will, therefore, be seen that 
although the immersed surface of the two boats may be 
about the same, their resistance at a certain, speed may 
be different. 

It must also be understood that wind, tide and waves in 
rough water, increase the resistance of the boat. The 
resistance from these causes cannot be approximated and 
can only be arrived at from actual results and experience 
under the various conditions. 

The resistance of ordinary boats of good model at 


moderate speed, in fairly smooth water, varies as the 
square of the speed ; and the power necessary to overcome 
this resistance, as the cube of the speed. This would 
apply to boats without reference to the mode of propulsion, 
except in the case of certain propellers, such as the screw, 
which have a tendency to augment the resistance under 
certain conditions. 

The resistance of a boat, however, also depends upon 
whether the boat is traveling through a broad expanse of 
water or along a confined waterway — such as a narrow 
river or canal. The resistance encountered by a boat, in a 
confined and narrow waterway, is very much greater than 
that experienced by the same boat in an unlimited expanse 
of water and depends upon the' relation between the section 
of the boat and that of the waterway through which 
it is passing. It will readily be understood that if the 
channel section is small as compared with that of the boat 
the resistance encountered by the boat will be greatly 
increased even at very low speed. At comparatively high 
speeds the resistance caused by the contracted waterway is 
experienced to a very much greater extent. 

90. We shall not endeavor to go into all the details of 
boat resistance under various conditions. It may, however, 
be well to give an empirical rule used to approximate 
indicated horse-power required for screw, propelled steam 
boats or vessels. It has been found from practice that an 
allowance of about 5 horse-power per 100 square feet of 
wetted surface at a speed of 10 knots per hour will approxi- 
mately give the indicated horse-power necessary at the 
engine cylinder. This allowance can, however, only be 
made with boats of fair model and good lines, and at 
moderate speed in free waterways. 

The above has been used in connection with larger boats 
than would generally be found among the types considered 
here. For boats of small size the allowance of horse-power 
for the given speed and surface would no doubt be slightly 
larger. The proper allowance can no doubt be easily 
deduced from experiment. 

With some alteration this rule could be applied to 
the electrical energy necessary under similar conditions. 


We may assume that the losses from the engine to the 
screw amount to 25 per cent, of the total indicated power. 
With an electric motor the losses will probably be half of 
this, or, let us say, 15 per cent. We may, therefore, deduce 
the electrical energy necessary at the motor terminals as 
follows : 


= horse-power electrical energy at motor. 

\ 10 / ^100 

Where >8^ = Speed in knots per hour. 

A = Area wetted surface of boat in square feet and 
allowing 4 horse-power per 100 feet of immersed wetted 
surface of boat at 10 knots per hour. 

In order to reduce this approximation of the energy 
necessary to propel a certain boat, to the resistance met 
with by the boat (leaving out the losses in the engine and 
screw), we may assume that the loss between the indicated 
horse-power at the driving engine and the real effective 
propulsive power used to overcome the resistance of the 
boat is from 60 to 60 per cent. This would, therefore, give 
us a rate of about 2i horse-power per 100 feet of wetted 
surface at a speed of 10 knots per hour. Reducing this to 
pounds we get approximately a resistance of about 80 
pounds per 100 feet of wetted surface at 10 knots per hour. 
Therefore, to get the resistance in pounds approximately : 

= H (resistance in pounds) ; 

^^' = effective horse-power. 


JS = Speed in knots per hour. 

A = Area in feet of wetted surface. 

j5^ = Speed in feet per minute. 

The allowance made for each 100 feet of immersed surface 
of the boat is not at all constant, but would vary with the 
surface and model of boat. With the ordinary model, such 
as used in electric launches and boats, it would form a rough 
approximation of the power required. We are assuming 
here, however, that the boat is propelled by means of a 
screw. With boats that are hauled, as in systems of canal 
boat towing, this would not apply. In such cases the 
absolute resistance of the boat in passing through the 





80 A 



X I' 


surrounding water must be deduced in a manner similar to 
that in a previous chapter, and the horse-power necessary 
to overcome the resistance will vary as the cube of the 

91. Under all ordinary conditions, excepting as above, 
boats are propelled by means of a propelling instrument or 
mechanism located on the boat. The propeller may be 
either a paddle, a screw or a jet, but the action of the 
propelling instrument upon the water is the same in all 
cases. It will readily be seen that an amount of work 
must be done in an opposite direction to the direction of 
the boat in order to overcome the resistance of the boat 
and cause it to move forward. As has been previously 
stated, all propellers work on the principle of projecting 
a certain mass of water, equivalent to the resistance of 
the boat, in an opposite direction to its motion. 

The velocity of the projected stream is always greater 
than the actual velocity imparted to the boat. As the 
water is a yielding medium it will readily be understood 
that there must always be a certain slip of the projected 
stream. Therefore if V be the velocity of the stream pro- 
jected by the propelling instrument, and v be the velocity of 
the boat, then V — v = velocity of the stream with respect 
to the water. This is called the slip of the propeller. The 
stream thrown from the stem of the boat is the slip. If 
the entire speed of the boat were equal to that of the 
projected column of water there would be no stream 
thrown astern and no slip. 

Although the jet, the paddle or screw propellers could 
be utilized for the propulsion of electric boats, the screw is 
particularly adapted to the propulsion of such boats on 
account of the high speeds obtainable with the electric 
motor. It has also been shown by practice that the screw 
has various advantages for such craft. We shall particu- 
larly deal with the screw propeller, but it will be well to 
also give a general description of the two other forms of 
marine propellers. 

92. The jet propeller is, perhaps, the simplest in principle 
and operation, but it is more or less impractical and very 


inefficient. It operates by throwing stemward from the 
boat a stream of water which has been drawn in at the bow 
or sides and passes through pipes in the boat. Between 
the inlet and outlet a pump, turbine or other device is 
located for imparting a velocity to the water, which, being 
in an opposite direction to the direction of the boat, causes 
a certain thrust which must be equivalent to the resistance 
of the boat. 

93. The paddle in its simplest form, consists of a number 
of flat boards or paddles, radially secured to a circular 
frame, revolved at its centre, so that, as the paddle wheel 
is revolved, the paddles push the water back in an opposite 
direction to that in which the ship is moved. If the water 
were unyielding, its action would be analogous to that of a 
rack and pinion, and the distance traveled would be ^ X the 
pitch circle of the paddles or circle of centres of pressure. 
As the water yields, however, the action is the reverse to 
that of an undershot water wheel. 

This form of propeller was the first to be used for 
mechanical boat propulsion. It is confined now chiefly to 
boats for inland navigation. There is another form of 
paddle wheel which, although acting in the same manner, 
is much more efficient. This wheel is called the feathering 
paddle and is so arranged that the floats, instead of being 
bolted as in an ordinary paddle, are pivoted on an axis 
parallel to the axis of the wheel. The paddles are con- 
nected by rods with an eccentric which is so set, with regard 
to the axis of the revolving wheel, as to cause the paddle 
or float to be nearly vertical when entering the water. By 
this means the floats always have a direct sternward action 
on the water and not an oblique action as is the case with 
the radial paddle. 

The feathering paddle is, perhaps-, the most efficient 
marine propeller and is used to-day to a great extent for 
inland and shallow water navigation. It has, however,^ 
various objections, particularly that of bulkiness; and for 
rough water it is entirely impractical. 

94. The screw is the most generally used marine pro- 
peller and is the only practical and efficient propelling 



instrument for the great majority of the boats described 
in this book. The action of the screw is more complex 
than either of the other forms of boat propellers. 

The general principle of the action of the screw pro- 
peller can be compared to that of a bolt and nut — ^the 
water represents the nut. If the nut be fixed and the bolt 
be revolved, a forward or backward motion will be im- 
parted to the bolt at right angles to the direction of 
rotation. If a blade of metal be wound endwise upon a 
cylinder or shaft so as to form a screw thread upon the 
shaft and revolved in water surrounding it, the same effect 
will be produced; that is, the water will take the place of 
the nut and a motion parallel to the axis of the screw will 
ensue. All screw propellers are based upon this principle. 


It is not necessary, however, that the blade should make 
a complete revolution around the shaft. If a slice be cut 
from such a thread, as shown in Fig. 77, a piece of blade 
will remain. If two such threads are wound around the 
shaft, two blades will remain, and so on. Such a slice 
with two or more blades forms a screw propeller. Each 
blade represents a thread of the screw. Two or more slices 
of threads or blades are used in order to decrease the 
surface friction between the surface of the blade and 
the water. 

The pitch of a screw propeller is the distance measured 
along the axis of one complete revolution of the screw 
blade. The pitch of ordinary screws varies from one to 
two and one-half times the diameter. 


The disc area is the circle swept by the edges of the 
blades as they revolve. 

The thrust of a screw propeller is the power exerted in 
pounds on a line parallel to the axis. 

The developed area or blade surface is the area of aU the 
blades added together. 

Fig. 78. 

The projected blade surface is the sum of all the blades 
measured on a plane at right angles to the axis. 

The length of the screw is the length of the slice cut 
from the whole thread, measured on a line parallel to the 

The ''leading edge" of a screw propeller is the edge 
that strikes the water first ; the next being the "following 

When the blade is twisted so that the pitch varies along 
the blade, the screw is said to have a varying pitch. If 
the pitch increases toward the periphery of the blade, the 
propeller is said to have an expanding pitch. If it is 
larger at the hub, the pitch is said to increase radially. 

X A 

Fig. 79. 

Several forms of propeller screws are shown in Figs. 78, 
79, 79a and 79b. 

If the screw propeller were worked in a solid medium, 
as in the case of the nut and bolt, the velocity of the boat 
driven by screw would equal P (the pitch) x H (the 
revolutions) of the screw. But, as will readily be underr 



stood, the boat does not generally advance at this rate. 
With the screw, as with other marine propellers, there is 
always a certain slip of the projected column of water 
which is lost in a stream thrown astern by the propeller. 

Fig. 79a. 

Let P be the pitch of screw and R number of revolutions, 
then V (speed of propeller) = P X JR. 
If now the speed of the boat be «, then. 
V — V = slip of screw ; 


X 100 = per cent, of slip. 

Fig. 79b. 

This is, however, the apparent slip of the screw, that is, 
the slip relative to still or quiet water. As, however, the 
boat causes a current to follow in her wake, this current 


must be taken into consideration in getting at the real slip. 
In some cases it has been found that v was larger than F, 
caused by the following current being so strong as to 
entirely counteract the apparent slip. This is called 
negative slip. 

Although it would appear that this would be a case of 
added efficiency, such is not the case. To create the fol- 
lowing wake, a certain expenditure of energy has been 
necessitated which can only be partly returned. It is 
very seldom met with in well proportioned boats of the 
class here treated of. It must, however, be remembered 
that in calculating the real slip, the* velocity of the 
following wake must be taken into consideration. The 
slip generally spoken of is the apparent slip. 

The real slip represents the true value of the backward 
velocity impressed on the water by the propeller, and is 
not to be regarded as an evil characteristic but the con- 
trary. Absence of real slip is a sign of inefficiency. It 
will readily be seen that the thrust transmitted to the 
boat by the screw shaft will vary with the velocity of 
the water thrown astern. 

The laws of slip of screw propellers have been deduced 
from experiment as follows : 

First. — ^For the same screw, slip increases with the 
resistance of the boat. 

Second. — ^The slip increases slightly as the revolutions 
of the screw are increased. 

Third. — The slip increases with the pitch. Therefore 
screws of fine pitch should be most efficient. 

Fourth. — The screw should be made as large as possible 
with relation to the mid-section of the boat. 

Fifth. — The slip decreases as the length of the screw or 
as the area of the blades with a fixed diameter increases. 

Screws of smaller pitch have generally been found to 
have less slip than those of coarse pitch. As such screws 
generally revolve at a high speed, the loss from surface 
friction between screw blades and water becomes greater. 
Screws to work at high speeds, however, can have less 
area than those at low speeds. With electric launches 
and most other boats (herein described), high speed, low 
pitch, screws would be used. 


In designing a screw it must, however, be borne in mind 
that in order to get the best results the screw must be kept 
immersed. If the propeller screw breaks through the 
surface of the water, it carries down with it a certain 
amount of air which is a cause of inefficiency. This is 
particularly the case with large screws. The number of 
blades does not seem to affect the working of the screw 
except as above stated. Three or four, however, give less 
vibration than two. 

95. It is particularly imperative that the lines of the 
boat should be such as to permit a free flow of water to 
the screw. It has also been found that the action of the 
screw increases the ordinary resistance of the boat by 
preventing the water that has been displaced by the bow 
from reacting on the stem as it should. This augmented 
resistance amounts to but a small percentage in well 
designed boats. Energy is also wasted in throwing water 
radially from the propeller instead of backward. It is to 
prevent this centrifugal action that the propeller blades on 
some screws are inclined astern. Thin blades and small 
pitch are preferable for such boats as electric launches. 
It has been pointed out that the slip of a screw propeller 
decreases with the pitch, but it must also be kept in mind, 
that as the screw revolves, the friction between the blade 
surface and the water must increase with higher speeds 
with the same blade area. With large screws revolving at 
high speeds this surface friction has been found to amount 
to several per cent, of the total energy used by the screw 
in propelling. 

The action of the screw is to drive sternward a column 
of water whose area is equivalent to the area of the screw 
disc minus the boss or hub. The velocity with which this 
area is projected by the propeller — P X B. As has been 
shown, however, this cannot be the velocity imparted to the 
boat by the propeller ; but the velocity imparted to the boat 
by the propeller will be the speed of the propeller less the 
slip. Let V = velocity of boat in f eot per second, V = speed 
of propeller or velocity of projected stream per second, 
A = area of the column of water projected in feet. Then 
JL X y = volume in cubic feet of water projected per 


second. Taking the weight of a cubic foot of sea water at 
64 pounds and gravity at 32 pounds, we get : 

{T) Momentum of stream = ^ ^ Z,^ ^ ( V— t?) = 

2 Ax V{V—v). 

This is the thrust of the propeller exerted along the 
shaft of the screw in pounds, and varies as 

Ax{P X ay. 

To move the boat this thrust must overcome the resist- 
ance of the boat, and the speed at which the boat will then 
move will be as previously shown. The thrust x the speed 
of propeller = total work, and resistance of boat x speed 
of boat = useful work. The horse-power necessary at the 
screw shaft outside of the boat varies as ^ x (P x -H)^ or 

TT p _ Tx 8 

' "" 33000 ' 

8 = speed of propeller per minute. 

It will readily be understood that the velocity of the 
following wake, if there be any, must always be taken into 
consideration when the velocity of the projected stream is 
to be deduced. 

It will readily be seen, also, from the above f ormulse 
and general principles, how the diameter, pitch, thrust and 
horse-power of a screw propeller can be approximated 
for a certain resistance of boat. In order to get the 
energy necessary at the motor terminals, the losses in 
transforming the electrical into mechanical energy and 
transmitting it to the propeller must be considered. If 
a constant resistance be deduced for a certain model of 
boat at a certain area and speed of propeller and for a 
standard area of wetted surface, the propellers necessary 
for other boats of the same model can be roughly deduced 
from the above data. As, however, screws of different 
diameter and pitch have a varying percentage of slip, V—v 
will vary. As has been shown, the resistance of boats 
varies with conditions, and although the screw propeller 
has been investigated by many, we are still compelled to 
resort to experimental tests with screws upon boats of 
various model to get constants to be used in a computation. 


The formula given, although quite rough and approximate, 
will, however, allow reasonable deductions to be made 
within the limits stated. 

Those who are particularly interested and desire to 
further investigate the propulsion of boats and propellers, 
will find of use various papers read before the Institution of 
Naval Architects, particularly by Dr. Froude ; the book on 
" Marine Propellers," by Sydney M. Bamaby ; and various 
other books and papers on naval architecture and marine 
engineering, which treat of the subject fully. Among 
these also are the writings of Prof. Rankine, A. E. Seaton, 
W. H. White, Scott Russell and Du Buat. 



The Miscellaneous Uses of Electrical Power for 
Boat Propulsion and Canals. 

96. The electrical propulsion of boats from some central 
source of generation or distribution is not limited to the 
classes already discussed. Such methods can also be 
applied to boats on other waterways. In the illustration 
(Fig. 80) a feasible plan for an electric ferry is shown, in 
which a motor on the boat receives current from overhead 
wires suspended transversely over the river or other chan- 
nel. The trolley or contact wire is suspended from a cable 
spanned across the river attached to the towers or poles on 
each bank. Where a railroad bridge, for example, crosses 
the river adjacent to the ferry, the wire can be supported 
directly beneath the bridge. These contact or trolley wires 
are arranged so as to permit a small trolley carriage to run 
thereon, and means are provided to keep the carriage secure- 
ly upon the wire at all times. The boat would be connected 
with the carriage by a flexible cable which would permit 
sufficient lateral motion and would be connected with the 
motor revolving the propeller. The boat could be started, 
stopped and regulated in speed and direction by any suit- 
able form of controlling or regulating device on the boat. 

Such an electric ferry could, no doubt, be readily con- 
structed across ordinary rivers and operated in conjunction 
with an electric railway. Its operation might be somewhat 
difficult across rivers having strong currents, but with 
sufficient slack cable this difficulty could be overcome. 
Although such an arrangement might possibly be operated 
beneath the Brooklyn Bridge, it is not intended to be 
operated on so large a scale ; but it certainly would seem to 
be a feasible arrangement for some small ferries that are 
located in the vicinity of electric railways from which 
current could be taken. 



97. Another arrangement is shown in Fig. 81. In this 
plan current is supplied to the boat by a submerged 

electric cable which is 
connected with the 
source of electric supply 
on the shore. As the 
boat proceeds, the cable 
is let out, and taken in 
again as the boat returns. 
The reel for winding the 
cable is preferably lo- 
cated on the shore, as 
shown. In some cases, 
however, there would be 
no necessity for winding 
the cable. This form of 
electrical boat could, no 
doubt, also be practically 
operated under certain 

At this point may be 
mentioned the plan 
patented by Mr. H. P. 
Wellman, of Catletts- 
burg, Ky., for the elec- 
tromagnetic mooring of 
ferry boats, canal boats, 
etc. Broadly stated this 
plan consists in placing 
upon the boat and the 
wharf, alike, a series of 
mooring surfaces, as in- 
dicated in Figs. 82 and 
83, one of the surfaces 
serving as a ''keeper" 
to the other, so that the 
boat is held at her wharf 
upon approach and is 
released when ready for 
departure. The scheme does away with ropes, snubbing 
posts, etc., and should prevent bumping. It would answer 















best in still waters, and where there is little rise and fall 
in tide. 

Instead of holding the boat to its mooring, quay or dock 
by suitable hawsers, as has been the practice, electro- 

FiG. 81. — Elbctbic Fkbby Boat Opebated by a Submerged 


magnetic means are used to accomplish this from the boat 
itself. In one of the modified forms of this method which 
is here* shown (Fig. 82) electro-magnets are placed upon 
the boat with their poles flush with the mooring side of 
the boat which in the illustration is the bow. The wharf 
or dock at which the boat lands is provided with a sheet 
or apron of iron which is so placed as to abut and come 
in contact with the poles of the magnets upon the boat 
when a landing is to be made. Ordinarily the electro- 
magnets upon the boat are not energized ; but when it is 

Fig. 82. 

desired to moor the boat, the captain or pilot can at once 
energize the magnets by connecting them with a source of 
electricity, which may be the electric light dynamo on the 
boat. As the magnets on the boat will abut the apron on 
the dock, which will act as a keeper for the magnets, the 
attraction between the magnets and keeper will be sufficient 


to hold the boat. Instead of using magnets upon the boat 
only, they may also be placed upon the dock and supplied 
with suitable current, or the boat may be equipped with 
an iron keeper at its mooring portion and the magnets 
placed upon the dock. It would appear that this novel 
application of electricity would promise practical success 
particularly where frequent landings are made, as in 
ferries, etc. 

98. Along a canal or river where an electric system of 
propulsion is used with distributing and contact wires 
paralleling the route of the waterway, the various bridges, 
lock gates, lifts, hoists and other machinery requiring 

power should also be electrically operated. When canals 
pass through large cities, movable bridges are generally 
constructed across the canal. Such bridges, whether swing 
or lift, should and can all be operated electrically, and such 
apparatus is already in use in various parts of the world. 
Where a canal passes through a country of varying level, 
either locks or other devices must be provided to pass the 
boats from one level of the canal to another. Where locks 
are used, it is usual, in most cases, to use power to operate 
the lock gates. The electric motor would certainly be a 
most satisfactory substitute for any of the present methods. 
The boats could also be hauled into the locks by electrically 
operated hauling machinery. On some canals, such as the 
Morris or the Delaware and Hudson, the boats are trans- 
ferred from one level to another on cars or cradles which 



are moved by means of an inclined track by means of a 
chain. In Fig. 84 a cradle, car and boats are shown, sls 
used on the Morris canal, New Jersey. The car is raised. 

Fig. 84.— Inclined Plane of the Morris Canal, at 
Bloomfield, N. J. 

and lowered by means of the chain to which it is attached. 

The application of electricity to this work is very simple 

and satisfactory. Direct lifts or elevators are also used 


upon some canals, which might just as easily be operated 
electrically as by steam or hydraulic power. 

99. Upon the canal system extending from Lake Biwa, 
Japan, to the ancient city of Kyoto, the boats are hauled 
up and down an incline on cradle cars by means of a 
Sprague 50 horse-power motor driving the cable drum. 
This inclined railway connects the two canal sections, 
between which there is a fall of 120 feet, and upon which 
it was first intended to use locks. When the incline was 
adopted, it was intended, also, to use water-power direct, 
with a wheel at the foot, but in the present plan, the 
power-house is driven by Pelton wheels, and the Sprague 
motor is placed at the head of the incline. The total 
length of the inclined railroad is 1,800 feet with a grade of 
1 in 15. The same power from the canal is being utilized 
for industries in the old city of Kyoto. ^ 

100. A very large field of utility for the electric motor 
will also be found in the loading an4 unloading of boats 
and barges upon a canal operated electrically. 

The electrical illumination of the towpaths, locks, 
etc., along a canal will also be a decided step in advance 
over the existing methods. The navigation of the Suez 
Canal at night by means of portable arc light plants has 
increased the facilities for traffic, and the income, of that 
canal enormously ; and where current is already used for 
power purposes the employment of a part of it at night is 
highly feasible. As an example of what can be done in 
this direction, it may be mentioned that the Baltic Canal 
vdll hereafter be illuminated by means of 25,000 incandes- 
cent lamps arranged along both banks. Each lock is also 
to have 12 arc lights, and electric lights are to be used for 
signaling purposes. The generating plant is to be placed 
in special power-houses at Holtenau and Brunsbuttel. 
This represents a plant of from 2,000 to 2,500 horse-power. 

The Manchester, England, ship canal is electrically 
illuminated, and the proposed ship channel from the Gulf 
of Mexico to the city of Mobile is to be similarly lighted. 

1. See " Electricity in the Far East." By W. Stuart Smith ; Elec. World. VoL ttyju. 
No. 1, Jan. 6, 1894, p. 6 et ^eq. 


In the latter case the channel is 80 miles long, and the 
alternating current has been adopted for the work. 

101. An electrical equipment has recently been furnished 
by the Canadian Cteneral Electric Company, of Toronto, to 
the Dominion Government, for the operation of the Sault 
Ste. Marie Canal locks. It consists of a pair of ' ' W. P. 50 " 
motors for each pair of gates and each pair of valves, 
operated form either side of the canal. In the case of the 
gates, the mechanical arrangement is the same as in a 
hydraulic equipment, with the exception that the hydraulic 
cylinder and piston are replaced by a pair of screws 
connected to the crosshead and operated by the motor. In 
the case of the valves, which are of the horizontal butterfly 
type, the operation is performed by means of a crank on 
the end of the valve shaft, which is capable of being 
operated through an arc of 45 degrees, by means of a 
vertical screw driven by the motor. The prime power for 
the generating plant is supplied by horizontal turbines, 
the water being taken from the upper level and discharged 
into the lower level* in the usual maimer. As might be 
expected, a lighting plant is included in the system. 
Such work would soon become common with the general 
adoption of electricity for canal boat propulsion. 



Storage Batteries, Motors and Dynamotors. 

102. Many references have been made in this volume to 
storage batteries, and as their use is important for the vari- 
ous purposes of electrical navigation, a few remarks upon 
some general characteristics and a few details as to special 
forms employed in this branch of work vdll not be out of 

Reference has also been made to some of the diiferences 
between primary and secondary batteries, in favor of the 
latter, and the reader may consult again Chapters I. and 
II. It should be pointed oilt here, for the benefit of those 
who have not before made a study of electricity, that a 
primary battery is one which when exhausted can only be 
replenished, refreshed and renovated by a new supply of 
some of its constituents, such as the zinc or the solution in 
which the solid elements are immersed. These renewals 
are not only expensive — ^zinc, for example, being many 
times as costly as coal — ^but to effect them is generally a 
great nuisance and annoyance. To-day there are millions 
of such primary cells in use for telegraph, telephone, 
alarm, and other purposes ; but the whole tendency of the 
times is to abolish them and to substitute either storage 
battery current or current generated by means of a small 
dynamo or motor-generator. From this obvious state of 
facts it inay fairly be inferred that where heavy currents 
are necessary, to drive motors, for example, the primary 
cell is not likely to prove more desirable and efficient than 
in the departments where it is now less and less employed 
to furnish small currents for very light work. At the same 
time, it is not true that while primary batteries need fre- 
quent renewals, the storage battery is free from them, 
or needs only a charge of current to keep it in a condition 
to do its duty. Plates "pasted" orunpasted have been 


known to drop to pieces; buckling and sulphating are not 
unfamiliar idiosyncracies; and the acid solution evapor- 
ates. All these and other qualities or tendencies of decay 
are less noticeable now than they once were, and can be 
minimized by a little intelligent care; but to talk as 
though they were unknown, and as though storage batteries 
were free from the weaknesses of youth or the ailments 
of old age would be criminally to deceive the unwary. 

At the present time storage batteries are in use success- 
fully in hundreds of central stations and isolated lighting 
plants, where they are merely placed on shelves or floors 
and charged or discharged under favorable conditions of 
quiet and repose. Their use in street cars has been much 
less successful, owing, as a rule, to the great jolting or 
"washing" they undergo, the sudden strains due to 
abnormal calls for current and incessant starting or 
stopping, and also to the frequent handling in the transfer 
from power plant to car body and back again. In launch 
work we have to deal with conditions midway between 
those of station use and of car service. On a boat the 
batteries are retained under the seats or along the hull 
all the time and are charged in situ. They are hardly 
subjected to such heavy strains in starting the vehicle 
up from a dead rest, and when once the boat is started it 
is far less apt to make frequent stops, having usually a 
free course with definite landing places wide apart. 
Moreover, the motion of a boat through quiet water is far 
more tranquil and smooth than that of a car traversing 
even the best track that was ever laid. For these reasons, 
a storage cell might easily succeed afloat that had proved 
a failure under all the disintegrating influences ex- 
perienced on land ; and for these reasons also, the storage 
battery has an enormous field of usefulness before it in 
launch propulsion. In canal boat propulsion also, the 
availability of the storage battery as an adjunct, either on 
the canal bank or upon the propeller, is not to be over- 
looked, and we must bear in mind the many uses to which 
current, cheaply generated, can be put along a canal, as 
evidenced, for example, by its use upon the Canal de 
Bourgogne already referred to. * 

1. Chap. XI., page 125. 


Modem practical storage battery use dates from the 
work in France of Plante, who used plain lead plates 
immersed in dilute sulphuric acid. These plates, however, 
as is well known, were very slowly '^ formed," so as to be 
fit for service ; and it was not until Faure introduced the 
practice of applying the "active" material to the face of 
the plates, and thus prepared the battery for use by 

Fig. 85. 

mechanical means instead of those strictly electrical, that 
the storage battery became really useful. The Plante and 
Faure types still remain the most prominent ; but there are 
endless varieties and modifications of the lead battery form, 
as well as many others that are quite distinct and different. 
Only a few forms will be noted here, as examples, and for 
further information and data, our readers are referred to 
books devoted specially to the subject of storage batteries. 



103. There are still not a few persons entitled to con- 
sideration who believe that the broad Plant6 method is 
the correct one, if properly and perfectly employed, upon 
which to construct storage batteries. The weak points in 
the original Plants type of cell are well known, and it was 
to overcome these and to increase the capacity as well as 

Fig. 86. 

the life of the cell that the Electric Power Storage Co., 
of New York, some time ago undertook a series of 
experiments carried out by Dr. Leonard Paget. In order 
to obtain a large surface for the action of the electrolyte a 
construction was adopted, as shown in Pig. 85, which 
represents a section of a plate with the upper connecting 
strip and connection plug; while Fig. 86 shows the cell 
complete and the manner of connecting the plates. As 


will be seen, the plates are built np of lead strips held in 
place at their upper and lower ends by horizontal bars 
which are cast over the ends of the lead strips, which are 
dovetailed for that purpose. The original form of the 
lead strips is a continuous ribbon, Vr of an inch thick, 
1 inch wide and weighing 1 pound per square foot. This 
is run through a machine which cuts it oif to the required 
length and dovetails the ends. At the same time, the lead 
ribbon is nicked at intervals of i of an inch, as shown in 
Fig. 85. When the strips are assembled to form a battery- 
plate, these nicks, about ^ inch in depth, form an oflFset 
which leaves a space between adjacent strips so as to leave 
free play for the circulation of the electrolyte. 

The initial formation of the lead plates so made up is 
effected in a bath which converts the lead into lead oxide 
of lamellar structure, as distinguished from the usual 
granular form, with the result that it clings closely and 
intimately to the plate, so that no movement or buckling 
of the plate causes a dropping oflf of the active material. 
The plates are also subjected to a treatment by which it is 
claimed that the lead cores of the strips remaining after 
the formation are made impervious to action by the battery 

The top of each plate into which the lead strips are cast 
is made of a composition not attacked by the acid and 
carries a dowel or taper stud which fits into a corresponding 
hole in the strip connecting the positive and negative 
plates of each cell. When the connecting strip is put in 
place a drop of solder is run into each connecting point 
and the joints thus made give additional conductivity. 
Each cell is connected to its neighbor by a connector bar 
which fits over a similar taper stud on the connecting 

320 Amfkke-Hour Gbue^ 

DischarKO Bate. 

P. D. 


80 amps. 

for 10 h. 


per cent. 


2.05 to 1.78 


46 " 

« 6 " 




tt it 


63 " 

U ^ it 




tt tt 


105 " 

<( 2 " 




tt tt 


106 « 

« 1 (( 




tt tt 

6 ampere-hours per pomid of Pb. 

These cells are built in various sizes and capacities from 
70 to 6,000 ampere hours, and the above table gives the 



results of a capacity test at different rates of discliarge. 
At the normal rate the cell has a capacity of 6 ampere- 
hours per pound of lead. The 280 ampere-hour cell with 
connectors has a resistance of 0.00416 ohm. The internal 
resistance of the cell alone does not exceed 0.0001 ohm. 
No records are available of the use of this cell in navi- 
gation, but it would appear to be well adapted for the 

104. The best known battery of the distinctive Fanre 
type in use in America is that made by the Consolidated 
Electric Storage Company, of New York and Philadelphia, 
whose battery was used with such eminent success in the 
electric launch work at the World's Pair.^ The plates in 
thi^ battery are of the familiar perforated grid form, the 
positive plate being filled with red lead which is turned 
into peroxide by the current ; and the negative plate with 
litharge, which is reduced to spongy metallic lead. In 
this form some marked improvements have been seen of 
late years. We give herewith an illustration of the cell 
(Fig. 87), and the subjoined table will show the capacity of 
the cell in various sizes : — 

Tablb of Ampebe-Houb Capacities at Vabious Rates of 
Discharge, and Time in Hours. 




17 8.. 





















12 5 

12 5 






105. We come next to what is known as the " Chloride " 
accumulator, with which, already, considerable launch 
work has been done in Europe, and which has been 

1. For full descriptioa of this see Chapter III., pages 35-42. 


similarly tried in this country, at Buflfalo, and elsewhere. 
The makers are the Electric Storage Battery Company, of 
Philadelphia. It is shown in Figs. 88 and 89. It derives 
its name from the fact that the plates are made up of 

Fig. 87. 

tablets cast from fused chloride of lead and zinc, which are 
held rigidly by a frame of antimonious lead. When so 
cast, however, they are not ready to be used, as the material, 
in this condition, is unfit to become active in a secondary 


battery. To make the plates active a chemical change is 
effected in the chloride tablets by means of a bath of 
chloride of zinc, in which the plates are immersed in 
connection with a slab of metallic zinc. The arrangement 
forms, in fact, a primary battery, the zinc acting as a 
positive and the tablets as a negative element. The 
electro-chemical action which results draws the chloride of 
zinc from the tablets by simple solution in the bath and 
also withdraws the chlorine from the chloride of lead and 

Fig. 88. 

fixes it with the zinc, forming chloride of zinc. The latter 
is then washed out of the plate, leaving the mass of 
crystallized metallic lead, which is immediately available 
as active material in a storage battery. 

Our engraving (Fig. 88) shows the Chloride cell as now 
constructed. As will be seen, it consists of a negative 
plate with round tablets of active material, which are 
perforated in order to permit of the free circulation of the 
battery fluid. The negative is separated from the positive 


plate, first by a separator, made of wood, soaked in 
insulating compound, and perforated to correspond with 
the location of the tablets in the plate. The perforations, 
it will be noted, are also connected by vertical grooves 
which permit of the circulation of the liquid, and also 
allow any gas which may be generated to escape. The 
positive plate, which is made considerably heavier than the 
negative, is surrounded by asbestos cloth which prevents 
any active material, which may become loose, from falling 
out and causing short circuits between the plates. The 
asbestos cloth, it will be noted, encircles the bottom of the 


# 4^ 9 ^-^ # # 

<& ^^ # # ^ # # 
^ # # # #^ # # 

Fig. 89. 

plates as well as the sides so that no material can fall to the 
bottom. Fig. 89 shows the plate complete in perspective. 
The capacity of the Chloride cells is from 5 to 6 ampere 
hours per pound, with a discharge rate of one-half ampere 
for each pound of plate — si very high rate. Notwith- 
standing this high capacity and high rate of discharge the 
efficiency of the cell is very high, the loss in current being 
less than 10 per cent, and of watt efficiency from 75 to 85 
per cent. Experiments have shown that at the rate of 
one-half ampere over three-quarters of the capacity is 
obtained above 2 volts. For electric launch work this 
feature is evidently a very valuable one. 



106. A very compact form of storage battery possessing, 
it would seem, various points of advantage in electrical 
boat work is that designed by Donate Tommasi, of Paris, 
and here shown in Figs. 90 and 90a. Each electrode is 
composed of a perforated tube of lead, ebonite, porcelain 
or celluloid, the bottom of which is closed by a plate of 
ebonite, in the centre of which is fixed a rod of lead, which 
acts as a conductor. The space between the central rod 
and the walls of the tube-electrode is filled with the oxide 
of lead. Metallic contacts connect respectively the rods of 

Fig. 90. — ToMMAsi's Multitubular Storage Battery. 

the positive tubes with the rods of the negative tubes, so 
that the current, in order to pass from one to the other, is 
obliged to spread over the entire active mass and thus 
produce a chemical circuit without loss and with uniform 
action throughout the active material. 

The tubular electrode from which the best results have 
been obtained is in the form of a rectangular cylinder, as 
shown in the accompanying illustration, and, in this form 
of the central lead rod, is provided with a number of 
wing-like projections. Special precautions are, of course, 
taken to prevent the coming in contact of electrodes of 


diflEerent polarity. As a result of this arrangement, the 
active matter, and hence the capacity of the cell, is greatly 
increased, and its weight is said to be from two to six 
times less, and its volume four to eight times less, than 
that of the accumulators at present in use. M. Tommasi 
also claims that in forming or charging his multitubular 
battery a current of 60 amperes per kilogramme of electrode 
may be employed, as against one ampere employed in 
present practice. On account, also, of the absence of 
all soldered joints in the connections between the different 

Fig. 90a. — Tommasi's Multitubular Stobagb Battbby. 

elements, all interruptions in service are prevented. This 
type of cell also is free from expansion of the tube, and the 
active matter, being entirely enclosed, does not fall, and 
hence a short circuit cannot take place. The illustration, 
(Fig. 90a) shows a set of these cells connected up for work. 
The Tommasi accumulator includes 67 per cent, of active 
matter, the ratio of active matter to that of lead in weight 
being about 2.1 to 1 ; that is, for 100 grammes of lead there 
are 210 grammes of active matter. The following figures 
give the principal electrical details of the cell : e. m. f., 


2.4 volts ; capacity, per kilogramme of electrode, 16 ampere 
hours ; current efficiency, 95 per cent. ; watt efficiency, 80 
per cent. 

107. A few years ago M. Emile Reynier, of France, 
brought out a cell intended to furnish a high voltage in 
and of itself, particularly for such work as boat propulsion. 
It is shown in Fig. 91, and consists of 16 plates mounted in 
flexible pockets, so as to have a certain amount of elasticity. 
These elements are placed flat, one against the other, and 
compressed between two end plates of wood by means of 

Fig. 91. — ^RBTNiisE's Electric High Potential Accumulatob. 

rubber spring bands. A bridge, consisting of hard wood 
impregnated with a waterproofing material, carries the 
whole, which may be suspended or rest upon its base, as 
desired. The spring arrangement gives to the active solid 
matter an artificial elasticity, which results in large specific 
power and storing capacity. The continuous compression 
of the plates, insulators and flexible pockets insures for 
these thorough protection against shaking and rough 
handling. Each of the pockets into which the plates are 
inserted is closed on top by means of a flexible and 
insulating stopper. 


We give below the principal figures relative to the cell, 
which has 16 couples and which is known as the horse- 
power-hour type. 

E. M. F 32 volts. 

Available fall of potential 28 volts. 

Current discharge 3 to 6 amperes. 

Normal power, about 150 watts. 

Capacity 30 ampere hours. 

Available useful energy 740 watt hours. 

{Length 0.40 metre. 
Breadth 0.80 " 
Height 0.30 « 

Contents, without containing cell 36 cub. decim. 

Total weight without cell 50 kilogr. 

Weight per kilowatt 330 " 

" " kilowatt-hour 67 " 

Volume " kilowatt 240 cub. decimetres. 

« « kilowatt-hour 49 « « 

108. Mention may be made of the use of ' ' Lithanode, ' ' for 
launch work. Its employment is due to the investigations 
of D. G. Fitz-Grerald, who has advocated its advantages as 
an active material for the anodes or positive plates. It is 
peroxide of lead in a dense, coherent, and highly 
conductive form. Its composition is almost the same as 
that of active material in general, but it differs somewhat 
in molecular construction and in its freedom from local 
action. The electromotive force developed by lithanode 
in conjunction with spongy lead is 2 volts. With a 
combination of lithanode and zinc, an electromotive force 
of 2.5 volts is obtained. The electrolyte is a solution of 
sulphuric acid and water, of a density about 1,220. In 
cells for heavy work, the elements are made up of a 
number of small slabs of lithanode, whose outer edges are 
V-shaped, and which are at once very hard and very 
porous. Around these, in the casting frame, molten lead 
or lead alloy is run, so as to fill the interstices and make a 
strong, complete plate out of which the pellets cannot fall. 
The negatives plates are built upon a corresponding plan ; 
and where lightness is desirable a copper gauze spongy 


lead negative is used. A marked reduction of weight is 
said to be obtained in all types of the lithanode cell. 

109. Reference has already been made to the use of a 
semi-solid electrolyte, which could not splash or spill. 
Dr. Paul Schoop,! of Switzerland, has been an active 
worker in this field, and to him we owe a successful 
gelatinous electrolyte, obtained by adding one volume of 
dilute sodium silicate — water glass — density 1.18 to two 
volumes of dilute sulphuric acid 1.250 density. Similar 
work has been done in England by Mr. Barber-Starkey, 
who has interposed a mixture of plaster of Paris and 

Fig. 92. 

sawdust between the plates ; and in America by Mr. 
Pumpelly, who has thus used cellulose or wood pulp. 

110. We do not know of any present instance in electrical 
boat work, in which the motor is not directly connected 
with the shaft of the screw, thus avoiding gearing. An 
example of opposing practice is that here illustrated in 
Figs. 92 and 93, which show the ^'Electricity" already 
referred to,i the first electric launch, we believe, ever 
propelled upon the English Thames. Fig. 93 shows two 
motors connected by belts to an overhead countershaft and 
arranged with a friction clutch, by means of which either 
motor could be thrown in or out of gear. Prom the 
countershaft, the belting passed down to a pulley on the 

1. See Chapter vm, page 96. 
1. Chapter U, page 12. 


axis of the screw. Such a method was, however, only 
tried to be abandoned, and we do not know of its revival. 
The modem motors, of which several have already been 
illustrated and described in these pages, ^ may be taken as 
types of machines many of which are obtainable to-day, 
admirably suited for direct driving in boat service. One 
type, however, that we have not yet included, but whose 
shape adapts it remarkably to the narrowed, converging 

Fig. 03. 

sides of a boat is the Storey, a form of which is shown in 
Fig. 94. It is the motor illustrated in the Chamberlain 
rowboat* where the drawing gives an excellent idea of the 
manner in which the sides of the boat cradle the motor, 
and at the same time permit it to rest very close to the 
keel, thus bringing the propeller shaft very low. The 

1. See pages 22, 81, 34, 45. 

2. See Chapter V, page 52. 



motor is a perfect cylinder in shape, is ironclad, and is 

111. A reversion to anterior methods has been made 
recently in this country, in trying again the plan of Trouve 
of mounting the motor on the rudder. ^ There is now in 
operation on the Schuylkill River, at Fairmount Park, 
Philadelphia, an interesting electric launch, built by Mr. 
F. A. La Roche, of that city. It measures 16 feet on the 
water line, 3 feet 10 inches beam, and draws about li feet 

Fig. 94. — Type of Storey Motor for Launch Work. 

of water. It will be seen from Fig. 95 that the entire 
mechanism is set on the stern post in exactly the same 
manner as a rudder is hung on an ordinary boat or launch 
and may be removed at pleasure and set on any other boat 
without alteration, provided, of course, that the rudder 
hinges are the same distance apart on both. 

The accumulators, six in number, are of the La Roche 
type, each having twelve plates 6 by 6 inches and each cell 
weighing 25 pounds, making the entire weight of all six 
(encased in two wooden boxes, which may be set in any 
convenient part of the boat) only 175 pounds. The floor 

1. Chapter I., page 4. 




space occupied by the boxes is exactly 3 square feet. 
Current is conveyed to the motor through the rheostat and 
a reversing switch by an ingenious arrangement of the 
brass railing on the gunwale. 

Pigs. 96 and 97 show two types of motor and their 
mechanism. The latter is a small i horse-power machine 
capable of developing in the launch described, and with 

Figs. 96 and 97. — La Roche Electbio Launch Motors. 

six cells of battery, a speed of four miles an hour, while 
the former is rated at i horse-power, but will give nearly 
double that amount for a short time if desired. All parts 
are made of aluminum, and the motor is of the multipolar 
type running at 400 revolutions per minute, and can drive 
the 16 foot launch at the rate of 8 miles per hour. In 
practice the motor is covered with a water-tight sheet 
iron cap. Mr. A. L. Biker has lately tried similar designs 
of motor on rudder in this country. 


112. The use of a dynamotor has more than once been 
touched upon in these pages. It will have been evident 
that there are many kinds of current available for the 

Fig. 98. 

charging of storage batteries, such as street railway current, 
power circuits, central station, isolated plant, etc., — ^but 
it is also plain that frequently the current will have to 

Tablb op Dynamotor Dimensions and Capacity. 

Dimensions Oyer All. 





. Pounds. 


























































* Long Commutator for Large Currents of Low Voltage. 

be transformed to render it suitable for charging. We 
illustrate here, in Fig. 98, a Crocker- Wheeler standard 


dynamotor, or direct cnrrent transformer, and supplement 
it by a table which will give an idea of dimensions. Such 
machines may be kept ashore or carried on the boat ; their 
flexibility in stepping a cnrrent up or down or changing 
in nature, is practically unlimited; and the authors of 
this volume believe that their utility will be strikingly 
demonstrated in the field of electrical navigation, although 
it is to be expected that other methods will also develop as 
the art advances. 



Altoona Launch Fleet 48 

" Ampere " Towing Steamer on Seine - - - - 122 

" Audace " Submarine Boat 76 

Astor Launch " Corcyra " 16 

Astor Launch " Progresso " - 20 

Baker Submarine Boat 69 

Baltic Canal Illumination .-.-.--, 199 

Batteries for Launches - 94 

Bicycle Motor Hauler 148 

Boat Propulsion on Erie Canal^ 102 

Boat Resistance 183 

" Bonaventure " Storage Launch 43 

Bourgogne Canal Electric Haulage 125 

" Bonnie Southport " Storage Launch - - - - 49 

Boston Launch Fleet - - - - l - - - 48 

BUsser System of Detachable Prpw - - - - 127 

Cable Haulage on River Elbe - 119 

Canalsy Ship and Barge 99 

Canal Boat Propellers 109 

Canal Generating Plant and Distribution - - - 160 

Canal History 98 

Canal Liclines with Electric Motors - - - - 197 

Catamaran, Electric - - - 62 

Chamberlain Electric Bowboat - - - - - 61 

Charging, Methods of - 47 

Chicago World's Fair Launch Fleet ... - 35 

" Chloride " Accumulator 206 

Combination of Railway and Canal Boat Haulage - - 155 

Comparison of Canal Boat Propulsion Costs - - - 173 

" Corcyra " Storage Launch - - - - - - 16 

Cost of Erie Canal Steam Towage 179 

Cost of Electric Hauler Method 177 

Cost of Electric Propeller Method 173 

Crocker, Prof, on Primary Batteries . - . - 9 

222 INDEX. 

Crocker- Wheeler Dynamotor - - - - 219 

Crocker- Wheeler Launch Motor 45 

"Daily Qraphio" Storage Launch 50 

"Dart" Storage Launch 44 

Davis Motor Locomotive Haulage 139 

Deal Lake Launch Fleet 43 

Deepening the Erie Canal 105 

Double Rail Duplex Stucture for Haulage - - 150 

Edinburgh Exhibition Launch Fleet - . . . 29 

" Electron'* Electric Launch 90 

"Electricity" Electric Launch 214 

" Electric '* Government Transport - - - - * - 29 

" Electric " Launch, Primary 6 

Erie Canal 100 

Estimates for Launch Fleet 26 

Faure Battery of Consol. Co., N. Y., .... 206 

Ferry Boats, Electrical 194 

Ferry Boat with Cable 195 

Flexible Submerged Cable Haulage 118 

"Frank W. Hawley" on Erie Canal - - - - 110 

Gear's System of Canal Trolley Ill 

Gig for Grand Duke Alexander 19 

" Goubet " Submarine Boat 68 

" Gov. Clinton " Canal Towing Steamer - - - - 121 

Gulf of Mexico Launch Fleet 49 

" Gymnote " Submarine Boat -* 66 

Hanson Primary Battery - - 6 

Haulage on St. Maur Canal 133 

Houston & Kennelly Niagara Power Estimates - - 163 

Hibberd Dirigible Life Saver 87 

Hovgaard Submarine Boat 57 

Hull Lines of Launches 91 

Immisch Launch Motor 31 

Jacobi Experiment 1 

Jacobi Motor 2 

Jet Propellers 185 

Jones Canal Boat with Detachable Motor - - - 113 

Lamb Electric Cableway Haulage 142 

Lake Biwa, Japan, Canal 199 

Lake Windermere Launch Fleet 29 

La Roche Motor on Rudder Post 216 

La Roche Launch 217 

Launch of Cruiser " New York " - ... - 19 

INDEX. 223 


Launch Requirements - - - - - 89 

Laws of " Slip " of Screw Propellers . . . . 190 

Lay-Haight Dirigible Torpedo - - - - - - 86 

" Lithanode " Storage Batteries 213 

" Magnet " Storage Launch -.--.. 14 

Manchester Canal Illumination - - - - - 199 

Melter Electric Life Buoy ------ 88 

Milwaukee Launch Fleet - ----- 48 

Mobile Canal Illumination - - - - - - 199 

Morris Canal, N. J., with Incline ----- 198 

Motor Locomotive Haulage - - - - - -138 

Motors for Launches - - 92 

Movable Cable Haulage - - - - - - - 133 

Mooring of Boats Electromagnetically - - - - 195 

New Haven Launch Fleet ------ 49 

Nordenfelt Dirigible Torpedo ----- 82 

Otis Rack and Pinion Haulage - - - - - - 129 

Overhead Double Trolley for Canals - - - - 110 

Paddle wheels - - - - - - - - - 186 

Peral Submarine Boat ------- 62 

Plants Battery of E. P. S. Co., N. Y. . - - . 204 

Pinnace with Storage Batteries ----- 24 

Rail at Bottom of Canal - - 130 

Reckenzaun Channel Trip 14 

Reckenzaun, F., on Launches 91 

Reckenzaun Launch Motor 15 

Resistance of Boats and Propellers - - - - - 181 

Resistance of Canal Boats 169 

Reynier's High Potential Accumulator - - - - 212 

Rigid Rail Canal Haulage 128 

Riker Launch Motor 22 

Riker Motor on Rudder Post 218 

Sachs Motor Haulage System - 146 

Sault Ste. Marie Canal with Electrically Operated Locks 200 

Schoop Semi-Solid Electrolyte ------ 214 

Scott Combination of Jet and Propeller - - - - 116 

Screw Propellers - - - - - - - 186 

Siemens Launch Motor 13 

Sims-Edison Dirigible Torpedo 77 

Storage Batteries on Canal Boats 158 

Storage Batteries on Boats 16 

Storage Battery Utilization 201 

Storey Motor for Launches 216 

224 INDEX. 


Suez Cftiud ninmination 199 

Tariff for Launch Hire -..---- 47 

Tommasi Moltitubular Storage Battery .... 210 

Towing by Magnetic Adhesion 121 

Traction Experiments on Erie Canal .... 170 

Trouv6 Battery Raft 10 

Trouvfi Motor 4 

TroavS Paddlewheel Boat 54 

Trouv6 Primary Battery Boats 3 

Trouv6 Storage Launch 12 

Trouv* Sea-Gk)ing Electric Ship 11 

Truax Electric Vehide-Boat 56 

Tug Propeller, Electric 116 

Universal Electric Launch 20 

U. S. Navy Proposed Submarine Boat - - - . 66 

Yauhan-Sherrin Boat, Primary 7 

Yauhan-Sherrin Primary Battery 8 

Yenice Launch Fleet 42 

" Yenezia " Storage Launch 42 

"Victoria" Dirigible Torpedo 84 

"Viscountess Bury" Storage Launch - - - -14,28 

"Volta" Storage Launch 13 

Waddington Submarine Boat 69 

Yarrow Vienna Storage Launch 13 


The Eleotpjo Storage Ba ttery Company 

las acquired all the patents and patent rights 
concerning the manufacture of electric storage batteries 
of all of the various important types heretofore developed, 
as well as the protection of every decision heretofore 
rendered by the United States courts in the interpreta- 
tion of electric storage battery patents. 

They have also acquired the control of the General 
Electric Launch Company and the Electric Launch and 
Navigation Company, and are prepared to furnish elec- 
tric launches and boats of every character complete, 
or will furnish batteries independently, of any capacity 

Catalogue containing full description of the various 
types of elements, weights, dimensions, capacities and 
price list, mailed upon application. 


Drexel Building, PHILADELPHIA. 

The Book of Books. 

- - for - - 

Wiremen, Electrical Engineers, Contractors, 
Constructing Engineers, Architects and Students 

■■■'■■■■ fa ' g"" " " ' 


How To -^ 
Wire Buildings 


Cloth, 8vo, illustrated, i6a pages. Price, $1.50. 

This is one of the most important practical books on electrical work 
that has ever been issued. It is written by a past master in the art of 
interior wiring; by one who has probably wired up more buildings and 
lamps than anybody else in the world. Mr. Noll began the practice of 
his profession at the very commencement of electric lighting by incan- 
descent lamps, and he has been an active leader in all the later develop- 
ments of recent years. The book abounds in solid, definite practical 
instructions, rules, suggestions and advice, and is liberally illustrated 
with drawings, diagrams, tables, etc. 

Chap. 1.— Introduction; Chap. 2. General Considerations; Chap. 8.»Location of Con- 
ductora; Chap. 4 —Division of CSrcuits and Distribution of Current; Chap. 5.— Loss of 
Electrical Ener^ in Conductors; Chap. 6.— Plans; Chap. 7.— Conduit Wiring; Chap. 8.— 
Switchboards; (%ap. 9.— Appliances and Connections; Chap. 10.— Converter work; Chap. 
11.— Overhead WiriM; Chap. 12.— Fuse Wire; Chap. 18.— Insulation; Chap. 14.— Electrolysis; 
Chap. 16.— Adverse wiring Conditions; Chap. 16.— Theatre and Stage Lighting; Chap. 17.— 
Plans of Distribution; Chap. 18.— Distribution of Light; Chap. 10.— Distribution of Labor 
and Hints to Foremen; Chap. 20.— Prelimimuy to Rules, Electrical Data, etc.; Chap. 21.- 
Rules for Ascertaining Required Sizes of wire; Chap. 22.— Energy-Power; Chap. 28.— 
I^rnamosand Motors; Chap. 24. -Pulleys; Chap. 25.— Belting; Chap. 26.— Engines; Chap. 
27.— Conclusion. 



%*Sent, postage free, to any address, on receipt of price, by 


Publisher, - — ^ 

la Coffege Place, NEW YORK. 



(central \Otatu 


♦ ♦ ♦ 

Ivlanageinent and J^inance, 


American Institute Electrical Engineers : S] 
Agent for Central Stations). 

Published in one Volume, - - Svo., at $1.50, 

(Member of American Institute Electrical Engineers : Special Census 
Agent for Central Stations). 

C. C. SH ELLEY, 12 College Place, NEW YORK. 

Sent, postage free, to any address 
"^^-on receipt of price. :: :: ::- 





bng b05S.94V 

Electrical lK»ts mtd frnvigokm. 

CatJOt Scieno« 0059SMS4 

3 2044 092 002 922