REESE LIBRARY 
 
 OF THK 
 
 UNIVERSITY OF CALIFORNIA. 
 
 , igo . 
 Accession No. 85424 . class No . 
 
ELECTRICAL TEACTION. 
 
ARNOLD'S PRACTICAL SCIENCE MANUALS. 
 GENERAL EDITOR: PROF. R. MELDOLA, F.R.S. 
 
 ELECTKICAL TRACTION. 
 
 BY 
 
 ERNEST WILSON, WHIT. SCH., M.I.E.E. 
 * 
 
 ASSISTANT PROFESSOR OF ELECTRICAL ENGINEERING IN THE SIEMENS LABORATORY', 
 KING'S COLLEGE, LONDON. 
 
 EDWARD ARNOLD 
 LONDON: NEW YORK 
 
 37, BEDFORD STREET. 70, FIFTH AVENUE. 
 
 1897. 
 

PKEFACE. 
 
 IN the following pages I have endeavoured to deal with 
 the principles relating to "Electrical Traction," and I 
 have, I hope, incorporated sufficient descriptive matter for 
 the purpose of illustrating the subjects treated of. Much 
 of the information with regard to the Leeds Tramways 
 which will be found in the text is given for the first time. 
 I have also included descriptions of certain experiments 
 which I have made for the purpose of this work. Descrip- 
 tive matter and illustrations have been obtained from 
 several sources, especially from the Journals of the Institu- 
 tions of Civil and Electrical Engineers, and the Northern 
 Society of Electrical Engineers, Engineering, the Railway 
 World, and a Report of the Corporation of Sheffield on 
 " Tramway Traction." 
 
 I have received much valuable information and many 
 illustrations from the Westinghouse and the British Thom- 
 son-Houston Companies, for which I wish to express my 
 obligations and thanks. Mr. E. W. Blackwell has also 
 kindly lent me some illustrations. 
 
 85424 
 
vi PREFACE. 
 
 I take this opportunity of acknowledging my indebted- 
 ness to Dr. John Hopkinson for the information he has 
 furnished me with in the preparation of this work, and for 
 the great kindness and assistance I have always received 
 from him during the last seven years. My colleague, 
 Mr. E. F. Herroun, and Mr. C. J. Evans, have kindly read 
 through the proofs; and Mr. J. R. Salter and Mr. C. J. 
 Evans have given me much assistance in the clerical part 
 of the work. I have received help from Messrs. Tomkins, 
 Woodfield, Strutt, Simpson, Greenbank, and Davey, past and 
 present Student-Demonstrators in the Siemens Laboratory. 
 To all these gentlemen I tender my thanks. 
 
 ERNEST WILSON. 
 
 KING'S COLLEGE, LONDON, 
 1897. 
 
CONTENTS. 
 
 CHAPTER PAGE 
 
 I. INTRODUCTORY ... ... ... ... ... 1 
 
 II. THE DIRECT-CURRENT MOTOR, AND ITS CONTROL ... 13 
 
 III. OVERHEAD AND OVERGROUND CONDUCTOR SYSTEMS ... 47 
 
 IV. THE TRACK AND ROLLING-STOCK ... 80 
 
 V. THE SLOTTED CONDUIT SYSTEM ... ... ... 93 
 
 VI. THE SURFACE-CONTACT SYSTEM ... ... 108 
 
 VII. STORAGE-CELLS ... ... ... ... ... 126 
 
 VIII. ALTERNATE AND DIRECT CURRENTS ... ... 158 
 
 IX. EFFICIENCY ... ... ... ... ... 183 
 
 X. THE POWER-HOUSE ... ... ... 204 
 
 APPENDIX ... ... ... ... ... 225 
 
 INDEX ... ... 246 
 
ELECTRICAL TRACTION, 
 
 CHAPTER I. 
 
 INTRODUCTORY. 
 
 General Remarks. Before considering the methods and 
 means for the utilization of electrical energy for traction 
 purposes, it will perhaps be well to briefly consider the agents 
 used on railways or tramways, and for our present purpose 
 we will divide the whole subject of line locomotion into two 
 branches, namely, the main trunk lines of a country, and the 
 shorter lines which include elevated and underground rail- 
 ways and street railways, or tramways. 
 
 The former branch is universally worked by steam, which 
 will probably hold its own for many years. The Heilmann 
 locomotive has, however, been proposed for this branch of 
 locomotion. The idea is a simple one. The locomotive 
 consists of a boiler and engine, the engine driving a dynamo 
 or dynamos to generate electric current, which current is 
 used for driving motors on the wheel axles. The advantages 
 which have been claimed for this form of locomotive are 
 that each axle of the locomotive, or indeed, of the whole 
 train, can be a driving axle having motors upon them, which 
 are supplied with current from the dynamo, and that hence 
 a greater tractive force can be obtained ; and further, that 
 the motors can run at any speed in relation to the speed 
 of the dynamo and steam engine on the locomotive ; thus, 
 
2 ELECTRICAL TRACTION. 
 
 in ascending a steep gradient, the engine can be run at 
 full speed with the maximum economy of steam, and the 
 connection with the axle is such that the speed is reduced 
 and consequently power gained, whereas in ordinary loco- 
 motives the power is proportional to the speed of the 
 engine if we neglect variations of the cut-off and wire- 
 drawing. But it must be borne in mind that although 
 these advantages may or may not be real, they are obtained 
 only at the expense of the extra cost and weight of the 
 electrical apparatus. Moreover, there is the efficiency of 
 the combination to be considered. We have first the con- 
 version of mechanical into electrical energy, and then the 
 conversion of electrical energy back again into mechanical 
 energy, and this, as is well known, entails a loss of about 
 15 per cent. 
 
 With regard to the latter branch, let us first of all con- 
 sider the tractive agents used on street railways or tramways. 
 We have 
 
 Horse, 
 
 Steam, 
 
 Cable, and 
 
 Electricity. 
 
 It is not our intention to enter deeply into the relative 
 merits of these agencies, but one may say roughly, with 
 regard to horse traction, the great drawback is that for 
 tramway purposes the life of the horses is only from three 
 to four years, and of course horseflesh is expensive. It is 
 only where a great number of horses are used in a large 
 scheme that any substantial profits can be looked for. 
 
 With regard to steam, the locomotive makes a good 
 deal of noise and smoke ; it is uneconomical, and must be 
 supplied with good fuel. Steam locomotives are still run- 
 ning in many places ; for instance, Leeds and Birmingham, 
 and both these cities have taken up the propulsion of 
 tramcars by electricity. 
 
INTRODUCTORY. 3 
 
 The propulsion of traincars by means of an underground 
 cable is in some cases undoubtedly most efficient ; for instance, 
 where the traffic is extremely heavy and the line fairly 
 straight, or where there are very stsep gradients. But 
 naturally the maximum speed of the car can be but the 
 speed of the cable, and the amount of power absorbed by 
 the cable itself and its maintenance are serious items in the 
 expenses of working the system. It is to be remarked that, 
 whereas propulsion both by steam and cable has been in 
 existence for a considerable time, propulsion by electricity 
 has practically superseded these systems in the United States, 
 that it has been considerably extended on the Continent, 
 and is now making great strides in this country. 
 
 As regards electricity, the case is an extremely good one. 
 Electricity may be looked upon as a form of gearing which 
 gives extremely convenient connection between fixed engines 
 and boilers or turbines, which generate the power, and the 
 tramcars, which are propelled by it. Some of its advan- 
 tages over steam are that it is noiseless and clean, and that 
 in the station inferior fuel can be used where there are 
 proper appliances ; and, since the whole of the power is 
 produced on the same spot, this can be carried out much 
 more economically than in the case of a locomotive 
 having to produce its power as it goes along, as well as to 
 carry the weight of the fuel. One of its great advantages 
 over the cable system is its extreme flexibility ; for instance, 
 the power station can be situate some distance from the track, 
 at a site where water may be easily and cheaply obtained, 
 where the cost for the cartage of fuel, etc., is at a minimum, 
 and where land is also cheap, or where there exists suitable 
 water power ; and on the overhead trolley system the con- 
 ductors or feeders can be taken almost anywhere without 
 much trouble or inconvenience, whereas with the cable 
 system enormous trouble and expense is necessarily incurred 
 in large towns and cities where there are numerous gas, 
 water, and service pipes which may have to be moved. 
 
4 ELECTRICAL TRACTION. 
 
 Perhaps the best basis of comparison is to compare the 
 actual cost of driving a tram-car one mile under the different 
 systems. The following figures represent roughly the 
 average results obtained in practice. 
 
 Horse 9d to 12d. 
 
 Steam 12df. 
 
 Cable 9d. 
 
 Electricity 5dL 
 
 The ratio of working expenses to receipts with electrical 
 traction averages 50 to 60 per cent. 
 
 If we look at the growth of electricity as applied to 
 traction we find that it has been very rapid. The year 
 1888 marks an epoch of great activity in the United States 
 of America, since in that year about 200 miles of horse 
 tramways were converted into electrical. Since then 
 electricity has continued to make such headway, that in 
 1896 over 12,000 miles of electrical tramway system were 
 actually laid, employing 40,000 cars, and it is computed 
 that 50,000,000 sterling has been expended in electrical 
 tramway work in the United States. "With regard to 
 Europe, probably 1000 miles covers everything at present, 
 but it is developing with great rapidity. Although we 
 have in the United Kingdom some of the very earliest 
 instances of electricity being applied to traction, yet we can 
 boast of having done very little in this direction ; and 
 probably one reason which has conduced to this is that in 
 the United Kingdom the electrical engineer has not had 
 such a free hand as in America, since the legislation is 
 totally different and very severe. The rail or track itself 
 has to be very carefully laid and constructed, owing to 
 the Board of Trade regulations with regard to leakage 
 currents ; in fact, in every direction we see that the British 
 tramway engineer has had great difficulties to contend 
 with. 
 
 There is no doubt that electric traction in the United 
 Kingdom is beginning to make very great progress, and 
 
INTRODUCTORY. 5 
 
 the Light Railways Act, which was passed last year, will, 
 no doubt, facilitate this, since it provides that " light rail- 
 ways " can be worked by electrical traction.* 
 
 Before the passing of this Act the method of procuring 
 powers was very expensive, and the regulations with regard 
 to construction severe. No district requiring facilities of 
 transportation need be deprived of its "light railway" at 
 the present time. The sixteen approved schemes up to 
 June, 1897, that is within a year from the passing of the 
 Act, and the twenty-eight further applications for " Orders " 
 lodged in May last, show that this Act is being appreciated ; 
 and, although no line has yet been constructed under it, 
 there can be no doubt but that it is a thoroughly practical 
 measure. 
 
 Steam and electricity will be the motive powers brought 
 into competition ; and, although in some cases steam is 
 undoubtedly the power to adopt, the relative advantages 
 of these agents are generally pretty evenly balanced. Out 
 of the sixteen schemes approved, four are to be worked by 
 electrical traction, and the rest by steam. Of the twenty- 
 eight new applications, electricity is proposed for seven, 
 with three cases in which electricity is an alternative. 
 
 General Regulations. It will be well to briefly review 
 a few of the general regulations of the Board of Trade 
 with regard to tramways, and also the general considerations 
 with regard to speed and the forces to be dealt with. 
 
 The Board of Trade have power to issue, and do issue 
 from time to time, regulations for securing the public 
 safety, and bye-laws with regard to particular tramways 
 worked by electrical traction. For example, regulations 
 dated llth of February, 1897, were issued regulating the 
 use of electrical power on the Blackpool Tramways (see 
 Appendix, pp. 225-230). Amongst other requirements, 
 these regulations state that every motor carriage shall be 
 
 * See " The Light Railways Act, 1896," with notes, etc., by Evans 
 Austin, M.A ,LL.D., Barrister-at-law. London : Reeves and Turner. 
 
6 ELECTRICAL TRACTION. 
 
 fitted with a governor which cannot be tampered with by 
 the driver, and which shall operate so as to cut off all 
 electric current from the motors whenever the speed 
 exceeds ten miles an hour. The speed at which carriages 
 shall be driven or propelled along tramways shall not 
 exceed the rate of eight miles an hour, and the speed at 
 which carriages shall pass through facing-points, whether 
 fixed or movable, shall not exceed the rate of four miles an 
 hour. The speed shall not exceed the rate of four miles 
 an hour at certain places such as crossings, or where traffic 
 is heavy, etc. ; that no two carriages shall enter any 
 passing place from opposite directions at one and the same 
 time. Further regulations are that the wheels shall be 
 fitted with brake appliances, which can be applied by a 
 screw or treadle, or by other means, and there shall be in 
 addition an adequate electric brake (see p. 91). The 
 regulations above referred to further provide that the 
 motor-car shall be fitted with a suitable fender which will 
 act efficiently as a life protector, and that it shall be free 
 from the clatter of machinery ; and any machinery under 
 the carriage shall be concealed from view at all points above 
 four inches from the level of the rails. Every train and 
 carriage shall be fitted with a brake, and in addition to 
 the conductor, there shall be a man on the front platform 
 of the trailer car, whose sole duty it is to attend to the 
 brake. 
 
 It is well known that it may require a supply from the 
 mains of as much as three times the power to get a tramcar 
 into motion as is required to keep it in motion when once 
 started. In the case of railways, definite stopping-places 
 are provided, and on single-line tramways definite stopping- 
 or passing-places are imperative. But as usually worked, a 
 tramcar stops at any point to pick up passengers. There is 
 no doubt that greater economy will be secured by stopping 
 at definite places only. To encourage this, if a fare of 
 say \d. or Id. were charged for carrying a passenger from 
 
ADHESION AND TRACTIVE FORCE. 7 
 
 one stopping-place to the next, and that if a passenger 
 stopped a car in an intermediate position he were charged 
 the full fare, he would rather walk to the nearest stopping 
 place in order to save the extra fare which he would 
 otherwise have to pay. 
 
 Adhesion and Tractive Force. The energy supplied to 
 the rail and the resisting forces have special interest for 
 the traction engineer. The actual force in pounds neces- 
 sary to start a motor-car or a train consisting of a loco- 
 motive and carriages, is a most important quantity, since 
 the motors must be able easily to exert this force. Equally 
 important is the question of adhesion between rails and 
 wheels ; for, if the motors do exert the requisite force, 
 and skidding is the result, the load cannot be started. 
 Perhaps one of the worst conditions is when the rails are 
 wet and snowy. In this case, the adhesion or frictional 
 resistance between rail and wheels is found to be as 
 low as one-tenth the total weight on the driving wheels. 
 The locomotives, with coaches and passengers, on the City 
 and South London Railway, require about 140 amperes 
 to start them. This current corresponds to a pull of 
 about 216 Ibs. per ton of locomotive, or 73 Ibs. per 
 ton of train, the weight of the locomotive being taken as 
 13*5 tons, and the complete train as 40 tons. In such a 
 subway the conditions are not the same as in the open ; 
 but if 200 Ibs. per ton on the driving wheels be taken as 
 the basis for the lowest limit of adhesion, it will not be 
 far wrong. No exact figure can be given, since the problem 
 is complicated on account of gradients ; but, unless abso- 
 lutely necessary, cars should not be stopped on steep in- 
 clines, and should be supplied with sand in the event of 
 their doing so. As an alternative to the use of sand the 
 wheels might be magnetized, such method serving also as 
 a brake. 
 
 Once the motor-car or train is in motion, the rate at 
 which work must be done on the rails by the motors is to 
 
 I UNIVERSITY } 
 
8 ELECTRICAL TRACTION. 
 
 be accounted for in three ways, all of which may be present 
 at the same time. 
 
 First, the train has to be accelerated ; that is, in a 
 certain time it must acquire a certain velocity. If the line 
 be level, or rising in the direction of motion, this work 
 must come from the source of supply. Matters can be 
 greatly helped, as on the South London and Central London 
 Railways, by making the trains store up energy by ascend- 
 ing to the station, and having this energy returned when 
 descending on departure from the station. 
 
 Second, there is the frictional resistance offered by the 
 rails, wind, etc., which is variable since these are influenced 
 by the state of the rails which, on ordinary tramways, 
 sometimes get filled with mud ; and in the open, head 
 winds have to be overcome. 
 
 Third, work must be done against the force of gravity 
 if ascending an incline, whilst gravity does work on the 
 train if descending. 
 
 Experience shows that it is not well to make motors 
 which are only just able to do the work required ; they 
 should do this amply. A careful examination of the plans 
 will enable the engineer to judge as to what must be the 
 power of his motors ; for instance, in the case of a railway, 
 the distance between the stations, the gradients, the pro- 
 bable worst condition of the rails and the service, that is, 
 the time required to get from station to station, and the 
 maximum velocity are given. We might take as an ex- 
 ample the City and South London Railway, which was 
 electrically equipped by Messrs. Mather and Platt, of Man- 
 chester.* An examination shows that, perhaps, the worst 
 section is between the Oval and Stockweil stations, for here 
 on the down journey the gradient is comparatively severe. 
 We may assume it to be 1 in 130 for the whole distance 
 of about 4000 feet, which is comparatively great. We 
 
 * See papers by Hopkinson and Greathead, "Proc. Institution 
 Civil Engineers," vol. cxii. p. 209, and vol. cxxiii. p. 39. 
 
LOCOMOTIVE POWER. 9 
 
 require the weight of the train necessary to carry, say, 
 100 passengers ; suppose it to be made up as follows : 
 
 Tons. 
 
 Dead weight of passenger stock at 0'2 ton per passenger 20 
 100 passengers at 0'06 ton each ... ... ... 6 
 
 Locomotive ... ... ... ... ... 14 
 
 Total 40 
 
 Assume, as a programme of time, that the train requires 
 half a minute to attain a speed of say 20 miles per hour, 
 that is, nearly 30 feet per second ; assume further, that 
 it runs at this speed for 1 minutes, and then begins to 
 slacken, and one minute is taken to slacken and stop at 
 the station. It is clear that the maximum horse-power 
 would be required during the first half -minute on the section 
 we are considering. First, if acceleration be constant, we 
 have during the first half-minute to do work at the rate of 
 *m* = 40 x (BO)* = = 
 
 time 2 X 32'2 X \ 
 
 H.P. Second, we have frictional resistance ; for this we 
 take 15 Ibs. per ton of train. This is got by allowing 
 10 Ibs. per ton of the train for the straight, and 5 Ibs. 
 extra per ton for curves. This is a fair average figure 
 to take for railway lines. On account of this we must 
 
 do work at the rate of 40 X 15 X 44Q = 16 H.P., since 
 
 33,000 x i 
 
 the distance travelled is 440 feet. Third, we have to raise 
 the whole weight of the train against the force of gravity 
 through a distance of 3*5 feet, since the gradient is 1 in 
 130, and the horizontal distance 440 feet. The rate at 
 
 which work is done on this account is 
 
 19 H.P. The total rate at which work must be done during 
 this first half-minute is then 111 H.P., and the motors must 
 be capable of doing this. In actual working on this subway, 
 the maximum speed would appear to be from 15 to 20 
 
io ELECTRICAL TRACTION. 
 
 miles an hour, whilst the average, exclusive of stops, is 
 stated to be 13 miles an hour. We see that, if a velocity 
 of 20 miles per hour has been obtained, it requires only 
 35 H.P. to keep up this velocity on this incline, that is, 
 about one-third of the maximum power. 
 
 The Application of mv 2 . When a force / is applied 
 
 to a mass m, the acceleration = -L , and the velocity acquired 
 when the body has moved through a distance s, with 
 this acceleration is given by v* = 2s-- ; it follows that fs 
 
 = fynv* : fs is the work done in foot-pounds, say, while the 
 force / acts through distance s, and \mi^ is the energy of 
 the mass m moving with velocity v. If the body has 
 initially the velocity v when the force / begins to be 
 applied it has already energy -g-wiv 8 , and the gain of kinetic 
 energy, equal to/s, is \mv? - %mv 9 where v 1 is the velocity 
 after the force /has acted through distance s. This is the 
 
 same as stating that t\ 2 v 2 is equal to 2s---. Consider a 
 
 train moving down an incline a distance I under the influ- 
 ence of gravity alone. Then since it falls vertically through 
 a certain distance ^, which is given at once when I and 
 the gradient are known, it follows that mgli is the actual 
 work done on the train by gravity where g is the earth's 
 acceleration. We can conveniently divide this into two 
 quantities, namely, that portion of the total work which 
 is taken to overcome frictional resistance of the line, 
 which we may say is equal to 12 Ibs. per ton of train, 
 multiplied by the distance I in feet if we express work in 
 foot-pounds, and the other portion which is the energy of 
 the train in virtue of its velocity, and is equal to fynif*, 
 v being its final velocity. If we want to stop the train, 
 clearly this work \mv* must be done upon something, 
 whether it be a brake in the ordinary sense of the word, 
 or a brake action due to the motors being converted into 
 
A PPL1CA TION OF \mv\ 1 1 
 
 generators, and returning the whole of this energy to the 
 system except that part dissipated in the transformation ; 
 or the energy can be returned to the system during the 
 time that the train is descending. This question will be 
 seen to have an important bearing in electric traction since 
 the dynamo is a reversible engine, and on steep inclines 
 the subject is of great importance. Say that from rest 
 to a speed of 20 miles an hour the tractive force varies 
 from 3000 to 1500 ; the weight of the train is say 40 tons, 
 and allowing 12 Ibs. per ton for frictional resistance, we 
 have 2520 effective for acceleration and work against 
 gravity if necessary. Say we wish only to accelerate ; then 
 
 2520 = - - x the acceleration ; acceleration equals 
 o2"2/ 
 
 0*91 feet per second per second. Assume this goes on until 
 v equals say 8 miles an hour, that is 1T7 feet per second. 
 Then s equals 75*2 feet, and the time taken equals nearly 
 13 seconds. Suppose we now descend an incline, such that 
 we can allow say one foot per second on account of gravity, 
 and that the tractive force is about 1500, which allows 
 1000 Ibs. for acceleration ; we have 0*36 due to the 
 motors; then, if this goes on until the velocity equals 
 30 feet per second, we have 900 = 2 x 1-36* + (ll'7) a ; 
 s= 280 feet. Also since v 1 = v 4- at where a is the ac- 
 celeration, we have/ =13*5 seconds. If the length of the 
 train be considerable as compared with the distance one 
 has to deal with, account should be taken of it. "We 
 see, therefore, that the total time from commencing is 
 26 seconds, the total distance travelled is 355 feet, the 
 final velocity is 30 feet per second, and the train has kinetic 
 
 energy * 1 * = l- 25 x 10- foot-pounds. 
 
 o2 "2 
 
 Suppose we stop the train from 30 feet per second, and 
 assume, say, that one-half of this energy is returned to the 
 system ; to get at the figure accurately we have the dis- 
 sipation of energy due to friction and motors to consider. 
 
12 ELECTRICAL TRACTION. 
 
 Then if the time be, say, 12 seconds, the average rate will 
 
 be - x X 74G x 10 6 - 70,100 watts, which at 500 
 12 550 
 
 volts would give an average current of 140 amperes ; or 
 since the work returned equals 0'62 x 10 6 foot-pounds, and 
 the total weight of the train is 40 tons, this is equivalent 
 to a fall of about 7 feet. 
 
 In changing from one service to another, say from 2-| to 
 2 minutes, it must be remembered that since the work is 
 nearly all done in starting, with frequent stopping-places, 
 the tractive force will vary more nearly as the square of 
 the speed if the change is effected by accelerating the 
 velocity of each car instead of increasing the number of 
 cars, and therefore heavier locomotives would be required. 
 
CHAPTER II. 
 
 THE DIRECT-CURRENT MOTOR AND ITS CONTROL. 
 
 General Remarks on Tooth and Smooth Core Armatures. 
 There are two types of armatures which may be said to 
 divide dynamos more into two distinct classes than the 
 shape and number of magnet limbs, since these, after all, 
 serve only to complete the magnetic circuit of the armature ; 
 these are known as slotted, or toothed, and smooth-core arma- 
 tures. The original Paccinotti ring serves to illustrate the 
 former, and the Siemens modern drum armature, the latter. 
 Both classes are largely used ; but, generally speaking, the 
 smooth-core armature may be said to be favoured in Europe, 
 and the toothed-core in America. Each has its advantages 
 and disadvantages when compared with the other. 
 
 It has been experimentally demonstrated * that in non- 
 synchronous alternate-current motors, the shorter the air- 
 space, the more efficient the motor, since the power factor 
 is increased ; and, magnetically speaking, the direct-current 
 machine would be improved by diminishing the length 
 of the air-space. The toothed armature lends itself to 
 diminished air-space ; but this cannot be wholly taken ad- 
 vantage of in the case of direct-current dynamos, owing 
 to consequent great armature re-action and sparking, 
 and the excessive stresses due to magnetic pull if the 
 armature gets slightly out of centre. So that the total 
 line integral of magnetizing force is not much different 
 in machines with the two classes of armature : in fact, 
 with toothed-core armatures the air-space must have a 
 
 * See Electrician, 28th August, 1896. 
 
i 4 ELECTRICAL TRACTION. 
 
 greater length than is mechanically necessary on account of 
 this. On the other hand, the smooth core gives the best 
 results obtainable as regards sparking, unless specially 
 prepared carbon brushes are used ; but it introduces other 
 disadvantages, which are, in whole or in part, not shared 
 by the toothed core. It is, however, to be noted that 
 the splitting up of the magnetic field into two parts adds 
 reluctance to the cross-magnetizing force due to the arma- 
 ture, thereby diminishing its reaction.'"* Take the case of an 
 alternate-current motor, in which the armature conductors 
 are threaded through holes, punched out in a circle round 
 the edge of the core discs. With fairly high permeability 
 in the iron, one may say that the field within such passages 
 is practically void of magnetism, and such a conductor could 
 be moved about in such field without experiencing any con- 
 siderable force. Yet force is transmitted to the armature 
 proper, and the conductors play an important part. "We 
 may say, without entering into a full discussion of the 
 subject, that force is transmitted to such armature, due, 
 in part, to the current in such conductors. In ordinary 
 toothed armatures we find slots which are sometimes par- 
 tially closed in on the periphery, and although the con- 
 ductors in such slots are not so perfectly shielded as is the 
 case in an induction motor, yet the whole force acting upon 
 the armature cannot act upon such partially shielded con- 
 ductors. In the tooth armature, then, the force t on the 
 conductors is not so severe, and they need not be supported 
 with such special care, as regards driving, as is the case in 
 smooth -core armatures. It is also well known that in the 
 smooth-core armature, the conductors have to be stranded, 
 or other precautions taken, to suppress induced currents in 
 the copper itself. This is not the case with tooth armatures, 
 
 * See Journal Institution Electrical Engineers, vol. xxvi. p. 608. 
 
 t Since the above was written, a paper dealing with this subject 
 has been read by Mr. Mordey. See Journal Institution Electrical 
 Engineers, vol. xxvi. 
 
THE DIRECT-CURRENT MOTOR. 15 
 
 since the shielding effect is so great, and therefore we find 
 solid copper conductors. This fact tends towards cheapness 
 in manufacture as well as increased efficiency, always assum- 
 ing that the dissipation of energy avoided does not make 
 any equivalent appearance elsewhere. It is usual in such 
 dynamos to work with the teeth highly magnetized. For 
 motors, perhaps the greatest advantage gained by teeth is 
 the protection they give to the conductors against mechanical 
 injury. One knows well enough how carefully smooth-core 
 armatures have to be handled, whilst teeth give great pro- 
 tection. There is no doubt, also, that the tooth armature 
 can better get rid of the energy dissipated in its core, since 
 there is a good heat conductor in the tooth, connecting the 
 air outside with the core itself. It is, however, usual to 
 ventilate both tooth and smooth cores (see p. 41) ; that is, 
 to provide internal air passages, so that the air in them, 
 when acted upon by centrifugal force, is driven out and 
 takes the heat away with it. Further than this, the motor 
 is better off than the generator, since, ceteris paribus, arma- 
 ture reaction is not so great in a motor as in a generator, 
 the magnetizing force due to induced currents and magnetic 
 hysteresis being the reverse in a motor to what they are in 
 a generator, so far as the current in the armature conductors 
 is concerned. With regard to the fields of both motors and 
 generators, cast steel has played an important part, and has, 
 one may say, developed the multipolar dynamo. The 
 ordinary tramway motor is usually a four-pole consequent 
 type, and perfectly closed in and watertight. It will be well 
 to discuss briefly the theory underlying the action of direct- 
 current motors in general. 
 
 The Theory and Practice of Direct-Current Motors. In 
 an ordinary drum-armature, with two poles, let I be the 
 total induction in C.G-.S. units, m the number of effective 
 convolutions, and E the electromotive force in volts induced 
 by the rotation of the armature in this field. Then we 
 have the well-known equation 
 
1 6 ELECTRICAL TRACTION. 
 
 E - 2lm rem 1Q-8 . (l) 
 
 second 
 
 The torque exerted by the armature is equal to a certain 
 force acting at a certain radius. Let T be this torque, then 
 
 revs 
 
 T27T - is the rate at which work is being done mechani- 
 secoud 
 
 cally by the armature, and is equal and opposite to the rate 
 at which work is being done electrically on the armature, 
 viz. EC where C is the current in the armature. We have, 
 
 therefore, T2^r i. = EC x 10 7 where T is expressed 
 second 
 
 in C.G.S. units and C in amperes. Inserting the value of 
 E, obtained above, we get 
 
 T = -O.GLS. units , (2) 
 
 7T 
 
 The engineer generally requires this torque in foot- 
 pounds. Since one foot-pound equals 1*356 x 10 7 ergs, it 
 
 T m Pi 0~ 8 
 
 follows that T = ~ foot-pounds. The whole of the 
 
 IT T356 
 
 power EC is not available at the shaft of the motor, since 
 this includes the rate at which energy is being dissipated by 
 journal, brush, and wind friction, and also by induced 
 currents and magnetic hysteresis. It is, therefore, necessary 
 to define what we mean by efficiency. We have (1) the 
 efficiency of conversion as being the ratio of the mechanical 
 power, given at the motor shaft to the electrical power EC ; 
 it is less than unity, on account of the dissipation of energy 
 due to bearing, brush, and wind friction ; induced currents 
 and magnetic hysteresis. This efficiency should, in a well- 
 designed motor of fair size, come out high. In the case 
 of a Siemens machine, intended to give 11 kilo-watts at 
 880 revolutions per minute as a motor, the efficiency of 
 conversion is 93 per cent, at full load." 
 
 * See Engineering, 24th March, 1893. 
 
THE DIRECT-CURRENT MOTOR. 17 
 
 (2) The electrical efficiency, being the ratio of the 
 electrical power, EC, to the electrical power applied to the 
 motor ; it is less than unity, on account of the dissipation 
 of energy in the armature and field-coil resistances. In the 
 machine above referred to, this efficiency is 93 per cent. 
 
 (3) The commercial efficiency, being the ratio of the 
 mechanical power given out by the motor shaft, to the total 
 electrical power absorbed by the motor. In the Siemens 
 machine, above alluded to, this efficiency is about 87 per 
 cent. It must be remembered that this takes no account of 
 loss due to gearing, which would exist in an ordinary tram- 
 way motor. 
 
 A subsequent test on the Siemens machines, above men- 
 tioned, running as motors, gives the results in Table I. 
 
 TABLE I. 
 
 
 Watts given 
 
 Electrical 
 
 Efficiency of 
 
 Commercial 
 
 Speed. 
 
 to motor at 
 
 efficiency 
 
 conversion 
 
 efficiency 
 
 
 104 volts. 
 
 per cent. 
 
 per cent. 
 
 per cent. 
 
 620 
 
 13,016 
 
 91-4 
 
 951 
 
 86-9 
 
 613 
 
 10,686 
 
 91-7 
 
 95-3 
 
 87-4 
 
 618 
 
 7,691 
 
 90-6 
 
 93-5 
 
 84-7 
 
 630 
 
 4,826 
 
 87-7 
 
 89-3 
 
 78-3 
 
 In Mr. Greathead's paper, before the Institution of Civil 
 Engineers,'"' a drawing is given of the motors on the 
 Siemens locomotives, and is reproduced in Figs. 1 and 2. 
 The motors MM (Fig. 3) are worked permanently in series 
 with one another and an extra starting resistance, R. The 
 magnet windings WW are placed in parallel, there being 
 two on each motor. C is a current meter in the circuit, 
 and S a reversing switch. The type of motor is the 
 ordinary Siemens bi-polar drum machine with horseshoe 
 magnets. The armature has a smooth core, which is fixed 
 
 * See Proceedings Institution Civil Engineers, vol. cxxiii. p. 39. 
 
 c 
 
f8 
 
 ELECTRICAL TRACTION. 
 
 FIG. 1. 
 
 FIG. 2. 
 
ELECTRICAL LOCOMOTIVE. t) 
 
 rigidly to the driving axle. The diameter of the driving- 
 wheel is 2*25 feet, and the wheel base 6 feet ; the shortest 
 radius on the line is 140 feet. There are two motors on 
 each locomotive, the magnets being supported from the 
 axle by gun-metal bearings, and from the locomotive body 
 by springs. The resistance of the four magnet windings 
 
 FIG. 3. 
 
 in parallel is 0*24 ohm, and each armature has a resistance 
 of 0*18 ohm.* Therefore the total ohmic resistance, if 
 K = 0, is 0'6 ohm. 
 
 In Fig. 4 the current through the motors is plotted 
 horizontally in amperes. The characteristic curve of one 
 motor is given. This curve therefore gives the total in- 
 duction in the armature in terms of the current in the 
 armature and magnet circuit. f Let Y be the potential 
 difference in volts applied to the locomotive. This is 
 assumed to be 450, and is supposed constant, and gives the 
 
 * See Proceedings Institution Civil Engineers, vol. cxii. p. 257. 
 f For construction of characteristic curve see Hopkinson's "Dynamo 
 Machinery," and Thompson's " Dynamo Electric Machinery." 
 
20 ELECTRICAL TRACTION. 
 
 straight line V. If, when the motors are not rotating, the 
 circuit be suddenly made, the extra resistance R is such 
 that the current is 140 amperes. This, then, is the maximum 
 
 FIG. 4. 
 
 current we have to deal with. Suppose for the moment that 
 R = 0, then E, the electromotive force of the motor, differs 
 from V by 0*6, multiplied by the current at any time. 
 Assume this to hold from no current to 100 amperes, 
 
THE DIRECT-CURRENT MOTOR. 21 
 
 and that then the curve for E be drawn as shown in Fig. 4. 
 E is obviously zero at 140 amperes, and the curve of E 
 between 100 and 140 amperes is determined by the value of 
 R. Taking this curve E as drawn, and the characteristic 
 curve for I, we can easily find the speed in revolutions per 
 minute, say, of the motors for any current. This we obtain 
 
 from equation (1) ; for revolutions per minute = -r - > 
 
 the 4m because there are two armatures in series. The 
 brashes are rigidly fixed on the neutral line, and armature 
 reaction is neglected. The revolutions par minute have 
 been worked out, and give the curve in Fig. 4 ; for which 
 the vertical scale must be multiplied by 200. It is obvious 
 that the path of the E curve, as also the revolutions per 
 minute, are under the control of the motor-man. The 
 economy of power at starting depends upon the control 
 of R, whilst the demand for current depends upon the track 
 conditions and the time required to accelerate. 
 
 From equation (2) we see that the torque depends upon 
 
 I and C for given motors. In the equation T = 
 
 TT x 1'356 
 
 foot-pounds, insert corresponding values for I and 0, and 
 we obtain the curve T in foot-pounds, Fig. 4. The rate 
 
 at which the motors do work is obviously T T ^^-* - -^ 
 
 mm. 33,000 
 
 English horse-power, if T be expressed in foot-pounds. But 
 this must be multiplied by the efficiency of conversion to 
 get at the rate at which work is usefully done external to 
 the motors, there being in this particular case no gearing. 
 This is the same thing as multiplying EC by the efficiency 
 of conversion. When gearing is employed the loss therein 
 must be accounted for in the efficiency of conversion. 
 The electrical efficiency is easily obtained, since it is the 
 
 Tjl 
 
 ratio ==, and is shown in Fig. 4. The efficiency of conver- 
 sion, as also the commercial efficiency, are more difficult to 
 
22 
 
 ELECTRICAL TRACTION. 
 
 obtain in the absence of experimental data ; an approxima- 
 tion can, however, be made. What we, of course, want is 
 for any load, at any speed, the rate of dissipation of energy 
 due to wind, bearing and brush friction, and magnetic 
 hysteresis and eddy currents. This we assume amounts to 
 1500 watts in the two motors, and for our present purpose 
 we consider it to be constant. 
 
 A series of differential tests have been carried out by 
 the Author in the Siemens Laboratory, King's College, 
 London, on two Siemens machines, each intended to give 
 11 k.w. at 880 revolutions per minute, to test the variation 
 of this quantity with various currents in the armature at 
 different speeds. The figures in Table II. give the results, 
 and refer only to the motor ; they show that this loss is 
 greater the greater the speed ; it is to some extent depen- 
 dent upon the load for a given speed. 
 
 TABLE II. 
 
 
 
 
 Rate of 
 
 
 
 
 dissipation of 
 
 
 
 
 energy by mag- 
 
 Kevs. per min. 
 
 Amperes in 
 armature. 
 
 Volts across 
 brushes. 
 
 netic hysteresis, 
 and eddy currents 
 
 
 
 
 in armature ; and 
 
 
 
 
 bearing, brush, 
 and wind friction. 
 
 
 
 
 Watts. 
 
 1273 
 
 42-1 
 
 107-0 
 
 589 
 
 1216 
 
 97-9 
 
 107-0 
 
 657 
 
 1130 
 
 99-2 106-5 
 
 633 
 
 1110 
 
 42-2 
 
 109-0 
 
 56 L 
 
 1060 
 
 729 
 
 105-0 
 
 499 
 
 1015 
 
 42-6 
 
 106-0 
 
 526 
 
 960 
 
 96-8 
 
 107-0 
 
 591 
 
 700 
 
 94-6 
 
 106-5 
 
 501 
 
 630 
 
 41-5 
 
 104-0 
 
 467 
 
 620 
 
 125-0 
 
 100-0 
 
 553 
 
 618 
 
 69-0 
 
 104-0 
 
 453 
 
 613 
 
 100-0 
 
 102-0 
 
 445 
 
 573 
 
 96-4 
 
 86-0 
 
 398 
 
 350 
 
 51-0 
 
 78-4 
 
 210 
 
ACQUIREMENT OF MAGNETISM. 33 
 
 We have then E - " ^ 5QO and g " 150 as the con- 
 ki(j VO 
 
 version and commercial efficiencies respectively. These are 
 
 plotted in Fig. 4. EC " gives the horse-power 
 
 VO 
 
 delivered to the rail, whilst gives the horse-power 
 
 actually taken from the supply circuit. 
 
 The Acquirement of Magnetism. The current cannot 
 vanish, since the commercial efficiency is zero when the 
 current is 3-| amperes. We have seen that the torque 
 depends solely upon I and for a series-motor, I being a 
 function of 0, as shown by the characteristic curve. "We 
 have assumed that the induction I, corresponding to any 
 value of C, always exists when C exists. This is not 
 necessarily the case. To test this point the Author made 
 some experiments upon a Siemens machine, the section of 
 the magnets being 15 x 4-| square inches ; the magnet 
 winding had 1984 effective turns, having a resistance of 
 6' 6 ohms. An extra non-inductive resistance of 21-5 ohms 
 was placed in the circuit of this winding, and provision 
 made for suddenly switching on 283 volts. The result 
 shows that the current is a maximum in this circuit three 
 seconds after switching on ; but ,the induction in the arma- 
 ture takes eight seconds before being fully established, 
 although most of the induction was induced in four 
 seconds. If it were not for the considerable air-space, the 
 induction would take much longer to attain its maximum 
 value. In the case of a core twelve inches in diameter, the 
 air-space being almost negligibly small, experiment shows 
 that with such a magnetizing force as that just used, the 
 magnetism takes fifty seconds to be completely reversed, in 
 spite of the fact that the current is reversed almost imme- 
 diately.""" "We see, therefore, that the initial torque in series 
 
 * See Journal of Institution of Electrical Engineers, part 116, vol. 
 xxiv. ; also Engineering, February 21, 1896. 
 
24 ELECTRICAL TRACTION. 
 
 motors having magnets of solid metal, is not nearly so great 
 as one would infer from the characteristic curve. There are 
 two remedies for this : first, lamination of the magnets ; 
 second, a sufficient magnetizing force always applied to 
 ensure a large value for I initially. The practice at present 
 in tramway work is to employ multi-polar motors. This 
 will help matters even if the cores be solid, and when the 
 cast steel shell has no great thickness. But, then, again, 
 armatures are used with toothed cores, thereby diminishing 
 the air-space. This will go against rapid acquirement of 
 magnetism. In ordinary tramway work the motors are 
 small, but in the case of electric traction on railways where 
 heavy loads have to be dealt with, the motors become large. 
 There is no reason, in such cases, why the motors should 
 not be compound wound, the currents in the series and shunt 
 windings helping one another, always provided that more 
 than one shoe be supplied to collect the current in order to 
 diminish the risk of breaking the shunt-circuit. The shunt- 
 circuit would always provide an initially large value for I, 
 and the series coils would only be called upon to give a 
 smaller augmentation of the induction, thereby providing 
 for. the drop of potential due to armature reaction and 
 current into resistance, and to prevent an abnormal current 
 in the event of the shunt being broken. In such cases the 
 assumption we made in connection with I and C would be 
 in great measure removed, and greater economy secured. 
 It must be remembered that the effect of temperature upon 
 a shunt winding is to reduce the magnetizing current for 
 the same applied potential difference, since copper alters its 
 resistance about 0*4 per cent, per 1 C. This heating effect is 
 not so important in the series-motor. A rheostat in the shunt- 
 circuit is generally employed to compensate for heating of 
 the coil. It appears that compound wound-motors are to 
 be experimented upon on the Belgian railways (see 
 Electrician, March 19, 1897.) 
 Agreement between Theory and Practice. It is worth 
 
THEORY AND PRACTICE. 25 
 
 while noting the agreement between the curves in Fig. 4, 
 and the results actually obtained in working. If reference 
 be made to Mr. Siemens' paper, read before the British 
 Association, 1892, it will be seen that curves obtained on 
 the locomotive are given. Four complete sets of curves 
 are given for an " up," and four for a " down " journey. If 
 the curves be examined it will be seen that certain portions 
 of a particular journey give the same conditions, or nearly 
 so, and therefore an average for the four can be taken. 
 Take sheets 2, 5, 8 and 11, Oval to Stockwell 
 
 Sheet. Miles per hour. Amperes. Volts. 
 
 2 13 80 400 
 
 5 13 60 420 
 
 8 13 60 420 
 
 11 13 70 420 
 
 Revolutions per minute = 1G2 for 13 miles per hour. The 
 curves in Fig. 4 give for this speed a current of 65 amperes. 
 Take Sheets 1, 4, 7, 10, Stockwell to Oval- 
 
 Sheet. Miles per hour. Amperes. Volts. 
 
 1 18J 30 440 
 
 4 19 30 460 
 
 7 19 30 405 
 
 10 19J 45 440 
 
 Revolutions per minute for 10 miles per hour = 237. The 
 curves for this speed in Fig. 4 give a current of 40 amperes. 
 Take Sheets 1, 4, 7, 10, Elephant and Castle to Borough- 
 sheet. Miles per hour. Amperes. Volts. 
 1 10 100 . 460 
 4 10 100 440 
 7 10 100 440 
 10 10 100 440 
 
 Revolutions per minute for 10 miles per hour = 125. The 
 curves in Fig. 4 give a current of 100 amperes. These 
 results are in fair agreement with calculation. 
 Sheets 1, 4, 7, 10, Borough to City 
 
26 ELECTRICAL TRACTION. 
 
 Sheet. Miles per hour. Amperes. Volts. 
 
 1 22 50 443 
 
 4 25 25 455 
 
 7 20 25 450 
 
 10 20 40 480 
 
 Sheets 2, 5, 8, 11, Elephant and Castle to Kennington 
 
 Sheet. Miles per hour. Amperes. Volts. 
 
 2 5 103 420 
 
 5 5 120 420 
 
 8 5 120 420 
 
 11 5 120 440 
 
 The highest speed given in the curves is between the 
 Borough and City. The curves in Fig. 4 would indicate 
 37 amperes. There is here for 40 and 50 amperes an indi- 
 cation of a shunt across the magnet windings ; but with 
 25 amperes without a shunt the calculated speed would be 
 greater than that actually observed. We shall see, at p. 27, 
 that, instead of shunting the field, the motors can be placed 
 in parallel for high speeds. Take a low speed, five miles per 
 hour ; the curves in Fig. 4 give 132 amperes for 62 revolu- 
 tions per minute. This shows that the actual acceleration 
 was not so great as shown in Fig. 4, and is, of course, 
 controlled by the motor-man. It would appear that under 
 these circumstances, from about 40 to 100 amperes, the 
 locomotive was running without extra resistance, R, and 
 without a shunt across the magnet windings. The effect 
 of this latter is shown by the dotted extension of the speed 
 curve, from 40 to 10 amperes. Since 140 amperes are 
 necessary to start this locomotive and train, we can at once 
 get the initial pull in pounds when the current is 140 
 amperes. The locomotive weighs 13'5 tons, and the com- 
 plete train weighs, say, 40 tons with passengers. The pull in 
 pounds per ton of locomotive = 216 ; and the pounds per 
 ton of train = 73. 
 
 Series - Parallel Operation of Motors. In ordinary 
 tvarnway work it is usual to operate the motors in an 
 
SERIES-PARALLEL CONTROL. 27 
 
 extended manner to that just mentioned in connection 
 with the locomotives (Fig. 3). The motors are almost 
 universally series wound, but are not permanently con- 
 nected in series with one another. The control usually 
 adopted is, first, to have the motors in series with one 
 another and extra resistance. When the extra resistance 
 has been cut out, simply the motors themselves are in series 
 between the conductor and the rail. The next step is to 
 place the motors in parallel, but this is not effected at once ; 
 for instance, the usual method is to first re-insert the extra 
 resistance, place the motors in parallel, and finally cut out 
 the extra resistance. This is known as the series-parallel 
 control, and was patented by Dr. J. Hopkinson in 1881. 
 For high speeds the magnet windings are shunted. We 
 see from Fig. 4, that, as the speed of the motors increases, 
 the horse-power they deliver to the rail decreases, and this 
 with decreasing efficiency after about 20 miles per hour. 
 But the motors work at high efficiency within the range 
 of speeds experienced, omitting, of course, the initial stages. 
 Suppose the speed be 526 revolutions per minute, that is, 
 40 miles per hour. With the motors in series and without 
 extra resistance, K, the commercial efficiency is 80 per cent., 
 and the current 20 amperes; that is, 7,200 watts are 
 delivered to the line. Now place the motors in parallel. 
 The total ohmic resistance is then 0*66 ohms, for one 
 motor. If the current be 10 amperes in each motor, E 
 = 450 - 6'6 = 443 volts. Revolutions per minute = 1063. 
 YO = 9000 motors in parallel. EC = 443 X 20 = 8860 ; 
 
 8860 - 1500 = 7360. 5? = 81'8%. 
 
 Under these conditions as to tractive resistance, the 
 motors in parallel do work at the rate of 81*8 per cent. 
 of YO at 1063 revolutions per minute, whereas when in 
 series they do work at the rate of 80 per cent, of VC, at 
 526 revolutions per minute, YC being the same in each 
 case. Suppose the current is 20 amperes per motor, and 
 
28 
 
 ELECTRICAL TRACTION. 
 
 they are in parallel. VO in this case = 450 X 40 = 18,000, 
 as against 9,000 when in series. E = 450 13 = 437. 
 Revolutions per minute = 500. EC = 437 X 40 for the 
 two motors. EC - 1,500 = 15,980. Commercial efficiency 
 = 88'8 per cent. Under these conditions as to tractive 
 resistance the motors in parallel do work at the rate of 
 15,980 watts, at 89 per cent, efficiency ; whereas, at practi- 
 cally the same speed, when in series, they do work at the 
 rate of 7200 watts, at 80 per cent, efficiency. In series, 
 the motors would do work at the rate of 15,500 watts, at 
 86 per cent, efficiency ; but the speed would be 244 revolu- 
 tions per minute. We see, therefore, that for high speeds 
 the motors can with advantage be placed in parallel, and 
 it is usual in traction work to carry this into effect ; that 
 is to say, to change over from series to parallel for the higher 
 speeds. Suppose the current be 80 amperes, that is, 40 
 amperes per motor in parallel. Then VC = 450 X 80 
 = 36,000 watts. E = 424. Revolutions per minute = 291. 
 EC= 33,920. EC - 1,500 = 32,420. Commercial efficiency 
 
 = ^ = 90*1 per cent. From the watts delivered to 
 
 the rails we can at once deduce the horse-power and pounds 
 pull at radius --5 feet. Table III. gives these figures in 
 better form for comparison. 
 
 
 TABLE III. 
 
 
 
 SERIES. 
 
 
 Revs, per 
 minute. 
 
 VC. 
 
 \Vatts delivered 
 to Kails. 
 
 Commercial 
 efficiency 
 per cent. 
 
 H.P. 
 
 delivered 
 to Rails. 
 
 Force in Ibs. 
 at radius 
 ft feet. 
 
 52G 
 244 
 138 
 123 
 
 9,000 
 18,000 
 36,000 
 45,000 
 
 7,200 
 15,500 
 30,500 
 3G,500 
 
 80 
 8G 
 85 
 81 
 
 9-65 
 
 20-8 
 40-9 
 48-9 
 
 85-7 
 398-0 
 1380-0 
 1890-0 
 
SERIES-PARALLEL CONTROL 
 
 I'ARALLEL. 
 
 1063 
 
 9,000 
 
 7,360 
 
 82-0 
 
 9-87 
 
 43-3 
 
 500 
 
 18,000 
 
 15,980 
 
 89-0 
 
 21'4 
 
 200-0 
 
 290 
 
 36,000 
 
 32,420 
 
 90-0 
 
 43-5 
 
 700-0 
 
 271 
 
 45,000 
 
 40,200 
 
 893 
 
 53-9 
 
 929-0 
 
 It might be found with motors in parallel that one is 
 doing more work than the other. If nothing wrong with 
 the connections can be found, the motor which is doing 
 least work can be made to do more by giving it a slightly 
 longer air-space ; for instance, by placing a thin sheet iron 
 packing strip between the junction of the halves of the 
 magnetic circuit. This will weaken the field of this motor, 
 and make it tend to run at a greater speed before gene- 
 rating a given electromotive force, E. 
 
 The curves in Fig. 5 are taken from Mr. Baylor's paper, 
 recently read before the Institution of Electrical Engineers 
 (vol. xxvi. of the journal), and illustrate the saving of 
 energy when series-parallel control is substituted for control 
 simply by resistance, the motors being in parallel. The 
 average amperes for the first 18 seconds are 43, and 62 in 
 the two cases. 
 
 Within the limits of Table III. the motors in parallel deliver 
 from 2 to 10 per cent, more power to the rails at from 102 to 
 120 per cent, greater speed. The question really turns on the 
 horse-power delivered at a certain speed and efficiency ; if, 
 for instance, 54 H.P. is required at 270 revolutions per 
 minute, clearly the motors should be in parallel. In series 
 at 244 revolutions per minute, the motors deliver only 21 
 H.P. to the rails. It would appear, then, that the require- 
 ments of the service on the South London Railway are 
 such that the motors, as constructed, can be connected 
 permanently in series, with the magnet windings shunted 
 for high speeds if necessary. At the same time series- 
 parallel control might have been employed with advantage. 
 
ELECTRICAL TRACTION. 
 
 Hamburg may be cited as an interesting example, since 
 out of some 300 cars only about 50 are worked by two 
 motors and series-parallel control, the rest are single motor- 
 cars. This is not universal, and it is considered better 
 
 wv 
 
 96 
 90 
 65 
 60 
 75 
 70 
 65 
 60 
 
 CO ** 
 
 a 50 
 
 < 40 
 35 
 30 
 25 
 20 
 15 
 10 
 
 5 
 
 
 
 / 
 
 \ 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 / 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 X 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 / 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 / 
 
 \ 
 
 
 7 
 
 X. 
 
 s^ 
 
 
 
 
 
 
 
 V 
 
 
 
 
 
 
 , 
 
 ^s 
 
 
 
 f 
 
 ^~ 
 
 
 
 
 
 
 
 
 * - 
 
 ^i^. 
 
 ^ *. 
 
 -^^ 
 
 
 \ 
 
 / 
 
 \ 
 
 
 
 
 / 
 
 
 \ 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 / 
 
 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 T 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ave 
 
 a* 
 
 Af 
 
 npc. 
 
 res 
 
 for 
 
 I8s 
 
 e c 
 
 s- 
 
 3 . 4 
 
 2.9 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ava 
 
 ^3 
 
 > Af 
 
 npt 
 
 reg 
 
 for 
 
 (8a 
 
 ec 
 
 Rhc 
 
 06 
 
 2.4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ft 2 3456789 tO II 12 13 14 15 16 17 18 19 20 2) 22 23242 
 SECONDS 
 
 FIG. 5. 
 
 practice to employ the series-parallel control in tramway 
 work. The higher the range of speed the more valuable 
 does this method of control become. 
 
 Effect of Variation of Applied Potential Difference on 
 Efficiency. It is interesting to see the effect of variation 
 of potential difference applied to the locomotive upon its 
 efficiency, speed, etc. Imagine V has droppsd to 300 volts, 
 and that we take 18,000 watts from supply, the motors 
 
SERIES-PARALLEL CONTROLLER. 31 
 
 being in parallel. Current is 60 amperes ; E = 280 ; revo- 
 lutions per minute = 420; EC - 1500 = 15,300; com- 
 mercial efficiency = 85 per cent. Comparing this with VC 
 = 18,000, and V = 450, we see that the speed is reduced 
 from 500 to 420, and the efficiency from 89 to 85. To 
 obtain the speed of 500, and do work at the same rate, 
 a greater current would be demanded ; and this is what 
 would happen, since the motor-man must maintain his 
 speed at the expense of current. In this particular case 
 he would have to shunt the fields, since there is no extra 
 resistance to be cut out. We see, therefore, what an 
 important bearing the constancy of potential difference 
 over the whole system has upon efficiency. 
 
 The Controller and Extra Resistance. The apparatus 
 used in practice for effecting the successful operation of 
 motors on the series-parallel control system is of great 
 importance, and is the outcome of much experience and 
 trial. 
 
 One of the early difficulties encountered in the operation 
 of motors on tram-cars was the sparking at contacts of 
 the controlling apparatus. The action of a magnet on the 
 electric arc is well known, and its adoption in tramway 
 apparatus of all kinds where arcs may occur has, no doubt, 
 had a great influence upon the success of electrical traction. 
 This idea is due to Professor Elihu Thomson, and the 
 patents for magnetic blow-out apparatus are owned by the 
 British Thomson-Houston Company. 
 
 The Thomson-Houston " K2 " Controller, in which this 
 magnetic blow-out is employed, is shown in Fig. 6, in which 
 the magnet with one of its poles swung back on a hinge 
 is clearly shown. This pole piece is so arranged that all 
 the contacts on the vertical spindle are, when the controller 
 is shut up, in the magnetic field produced by such magnet ; 
 and, therefore, whenever a contact is broken the field 
 immediately blows out the arc. Many hundreds of these 
 controllers are in use at the present time, and they are 
 
32 ELECTRICAL TRACTION. 
 
 adopted on the Leeds, Bristol, Dublin, and other lines in 
 the United Kingdom. 
 
 In Fig. 6 the handle at the top of the controller serves 
 to operate the vertical spindle and its series of contacts, 
 
 FIG. 6. 
 
 and interlocked with this is a subsidiary spindle shown to 
 the right which operates the reversing switches, the arrange- 
 ment being such that the reversal of current cannot take 
 place until the main switch has broken contact with supply. 
 
SERIES-PARALLEL CONTROLLER. 33 
 
 The K2 controller is so arranged that when the handle 
 is placed on the first notch the motors are in series with 
 one another and the full extra resistance. Notches 2 and 
 3 reduce this resistance to one-third and one-twelfth 
 respectively. Notch 4 places the motors in series across 
 the mains without extra resistance. Notch 5 prepares for 
 placing the motors in parallel by first shunting the fields, 
 and then placing only one motor across the mains with 
 one-third the extra resistance in its circuit. Notch 6 places 
 the second motor in parallel with the first, the one-third 
 extra resistance still in series. Notch 7 reduces this extra 
 resistance to one-twelfth. Notch 8 places the motors in 
 parallel across the mains without extra resistance. Notch 
 9 shunts the fields of the motors in parallel for higher 
 speeds. Notches 1, 2, 3 are merely for the purpose of 
 cutting out resistance, and the handle should pass over 
 them with fair rapidity ; 4, 5, 6 are working positions, 
 giving speeds suitable for level stretches ; 8 and 9 are 
 positions for steep gradients and high speeds, and a pause 
 should be made on each notch to enable the car to obtain 
 the speed due to that position of the switch-handle. The 
 addition of an extra reversing switch enables the con- 
 troller to deal with four motors instead of only two. 
 
 The Westinghouse controller, No. 28 A (Fig. 7), deals 
 with the series-parallel control of two motors of from 25 to 
 30 H.P. A vertical steel shaft, terminating with the 
 handle shown in the figure, is fitted with iron lugs, which 
 are faced with copper, such lugs being insulated from the 
 shaft by means of mica. These lugs make contact with a 
 series of projections on the side of the controller frame, 
 such contacts depending upon the position of the operating 
 handle. A series of fibre rings are placed between the lugs 
 on the shaft, for the purpose of preventing any inter- 
 sparking which might occur, such rings turning with the 
 shaft. This controller has a series of seven definite posi- 
 tions or notches at which the controller handle can be 
 
34 
 
 ELECTRICAL TRACTION. 
 
 FIG. 7. 
 
SERIES-PARALLEL CONTROLLER. 35 
 
 placed, and which are very easily found by the conductor 
 of the car. If the handle be placed at the first notch, the 
 circuit is closed on the two motors in series with extra 
 resistance, which is cut out in steps until notch number 
 four is reached. The fifth position places the motors in 
 parallel with extra resistance in the ' circuit which is partly 
 cut out on the sixth position. Notch seven places the 
 motors in parallel directly on to the line without extra 
 resistance. 
 
 The small handle to the right is for reversing the 
 direction of current in the field coils, so as to reverse the 
 direction of rotation. The two handles are interlocked 
 with each other in such manner that the reversing switch 
 cannot be operated unless the circuit be broken by the 
 main handle. Further than this, if it is necessary to cut 
 either motor out of circuit by means of a special short 
 circuiting plug and this might be necessary if one motor 
 became disabled the car starts on the first notch, and is full 
 on at the fourth, beyond which the handle cannot be turned. 
 
 The construction of a suitable rheostat or resistance for 
 motor work is one of the most important considerations in 
 the success of electric traction. Coils of wire, such as 
 German silver, stretched between insulators in the manner 
 usually adopted in the test-room, are not a success on 
 a car or locomotive. The vibration and extension due 
 to heating give great trouble. In some cases such spirals 
 are supported in plaster-of-Paris, or sand, in iron tubes. 
 Fig. 8 gives the Westinghouse method of dealing with this 
 problem. Iron or nickel steel band is wound on a former 
 with mica insulation between, and free access is given to 
 the air to come in contact with such coil internally and 
 externally. In the figure, one such coil is shown, and 
 combinations of such coils can be made by securing them 
 to a cast-iron frame, the number depending upon the 
 requirements of the rheostat. 
 
 The Tramway Motor. In traction work it has become 
 
36 ELECTRICAL TRACTION. 
 
 general to use multi-polar motors with at least four 
 poles. In a four-pole motor two coils are used on poles 
 
 FIG. 8. 
 
THE TRAMWAY MOTOR. 37 
 
 on opposite sides of the armature ; the other two poles, at 
 right angles to these, being simply short projections on the 
 cast steel shell which protects the motor from injury and 
 seals it up so that water cannot get into the working 
 parts. The armature of these motors is, perhaps, the 
 most important part, and deserves special attention. The 
 practice now is to use the two-circuit four-pole drum arma- 
 ture, the conductors being placed in slots on the surface. 
 This gives a very strong construction, and the strain upon 
 the copper conductors is thereby diminished. 
 
 As an example of the best practice, we may mention the 
 motors made by the Westinghouse Company. Their present 
 standards are what are termed "No. 12A," 25 and 30 H.P. ; 
 and " No. 38," 50 H.P. In these armatures the number of 
 slots and the number of commutator bars are each prime 
 to the number of poles ; for instance, in the 12A type, 
 there are 47 slots and 93 commutator bars. Each slot 
 contains wires belonging to four separate coils, two of 
 which are attached together before insertion. Since there 
 are 47 slots and 4 ends per slot, there are 188 ends. But 
 the commutator has only 93 parts, and therefore two ends 
 have to be discarded. These are the ends of any one coil 
 on the armature. The diagram in Fig. 9 shows a four- 
 pole armature, in which there are 19 slots and 37 com- 
 mutator parts, each number being prime to the number 
 of poles ; and shows the " dead coil," as it is termed, 
 this being placed in position to make the mechanical 
 balance perfect. The long, radial lines are the upper 
 conductors, and the short ones the lower conductors in 
 a slot. It will be seen that the upper conductors in one 
 slot are connected to the lower conductors in a slot three in 
 advance, and this is continued all round the armature. 
 Although only one conductor is shown for a particular coil 
 in the diagram, it is obvious that each coil can enclose the 
 teeth round which it passes as many times as is required 
 for the purpose in view. This number is necessarily 
 
38 ELECTRICAL TRACTION. 
 
 dependent upon the speed, volts, and induction in the arma- 
 ture. The pitch, as it is sometimes called, or the arc of 
 circle embraced by the coil, is a variable. For instance, 
 
 FIG. 9. 
 
 in the armature referred to, the coil embraces an arc 
 of a circle given by placing it in slots 1 and 11 ; but 
 in another instance, the coils are placed in slots 1 
 and 12, although the total number of slots is the same 
 viz. 47.* On the other hand, the "Westinghouse, No. 38, 
 50 H.P. armature has 45 slots, each containing three 
 
 * This variation of pitch is well set forth in " Armature Windings," 
 by Messrs. Parshall and Hobart. 
 
THE TRAMWAY MOTOR. 3$ 
 
 separate coils, making six ends per coil. The commutator 
 has 135 bars, two leads being connected to each. There is, 
 therefore, no " dead " coil in this armature, and yet 135 
 
 FIG. .10. 
 
 and 45 are each prime to the number of poles. It is only 
 necessary to place the brushes on the neutral line at 90 
 apart, in this type of armature. The brashes are in all 
 cases made of carbon, the carbon being held in a small 
 
 FIG. 11. 
 
 holder, and lightly pressed upon the commutator by means 
 of a spring. The brushes are shown in position in the 
 diagram. The coils are machine wound, and carefully 
 insulated before being placed in position. Fig. 10 shows 
 
40 ELECTRICAL TRACTION. 
 
 No. 12A armature partially wound. Fig. 11 shows a com- 
 pletely wound armature, with half its connections made to 
 the commutator. 
 
 If I be the total induction in the armature due to one 
 pole, the average E.M.F. induced in the turn on the 
 armature during one quarter of its revolution is 
 
 21 _8I_ 
 
 time of \ rev. in sees. ~~ time of 1 rev. ' 
 
 and since there are -^ effective turns between the brushes 
 
 ! 
 
 FIG. 12. 
 
 where m = total number of convolutions on the armature, 
 or 2m = the total number of wires counting all round, it 
 follows that, if m be large enough, the machine would give 
 as a generator a practically steady uni-directional electro- 
 revs. ~ e, 41 w revs, 
 motive force of 4Im x C.G.S. units ; or -^- x 
 
THE TRAMWAY MOTOR. 
 
 X 10 ~ 8 volts. This is our familiar equation connecting 
 these quantities (see p. 16), only we now have four poles 
 instead of two. 
 
 Fig. 12 gives a cross sectional elevation of the motor in 
 detail, the two poles shown having field coils, the two 
 others at right angles to these are without coils. The 
 motor is what is generally called a " consequent pole " 
 machine. The armature core is provided with internal 
 passages for the purpose of ventilation a method largely 
 used at the present time in generators as well as motors. 
 The centrifugal force, acting upon the air in these radial 
 passages consequent upon the rotation, sets up circulation 
 which keeps the armature cool. The sketch shows in 
 detail the bearings and method of lubrication, and the 
 steel pinions at the end of the motor shaft. Fig. 13 
 shows one of the armature 
 discs, the internal passages 
 for air, and the armature 
 conductors in position in a 
 slot. In these motors the 
 brushes can be fixed rigidly 
 on the neutral position be- 
 tween two poles. It is 
 fortunate that this is the 
 case, for adjustment of the 
 brushes to suit various loads 
 would be costly and trouble- 
 some. It is well known 
 that in a motor the arma- 
 ture reaction is not so 
 
 severe as in a generator. The eddy currents and magne- 
 tizing force due to hysteresis in the armature core are the 
 same in a motor as in a generator ; but in the latter they 
 help the current turns on the armature, whilst in the 
 former they oppose, thereby causing less reaction. 
 
 The General Electric Company's Motors. Figs. 14 and 
 
4 2 
 
 ELECTRICAL TRACTION. 
 
 15 give the speed and torque curves of a G.E, 800 motor, 
 with 3 and 4 turns per armature slot respectively, given 
 
 1 
 
 10 
 
 2 14 
 
 3 18 
 
 | MILES P 
 
 2O 22 
 
 ER HOUR 
 
 24 26 
 
 ON 33" 
 23 30 
 
 WHEELS. 
 
 32 34 
 
 3-6 s'a 
 
 40 4 ' 2 
 
 44 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 A 
 
 I 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 \ 
 
 
 ^ 
 
 ^ 
 
 FO 
 
 G. E. 800 
 
 RM B. RAILWAY MOTOR 
 3 TURN ARMATURE. 
 
 
 > 
 
 <^ 
 
 
 
 SPEED & TORQUE. 
 
 SPEED REDUCTION 4-78. 
 
 X 
 
 
 Pour 
 
 "" ^ 
 
 DS HORI 
 
 "*" 
 rONTAL E 
 
 
 
 FORT ON 
 
 33 "WHEELS. 
 
 
 
 200 400 600 800 IOOO 1200 I4OO I60O I8OO 
 
 R.P.NJ. AXLE 
 
 *0 60 80 100 120 '40 I6O ISO 2OO 22O 24O 26O 28O 300 32O 34O 360 38O 
 
 G.E.8OO FORM B. RAILWAY MOTOR 
 4-TORN ARMATURE 
 
 SPEED & TORQUE 
 
 SPEED REDUCTION 4 T 78 
 
 FIGS. 14 AKD 15. 
 
 to the Author by the British Thomson-Houston Company. 
 The speed reduction is 4' 78 to 1. Data taken from these 
 curves are as follows : 
 
THE TESTING OF MOTORS. 
 
 43 
 
 A.*,,. Force in Ibs. at rim ' \ mtlM ~ Q Revolutions per 
 
 Armature. of 33 in. wheel. Amperes. minute of motor. 
 
 3-Turu 
 4-Turn 
 
 800 
 1060 
 
 50 
 50 
 
 121 
 93 
 
 It is interesting to compare the same motor with the two 
 windings. For the same current, say 50 amperes, the field 
 will be the same ; we should therefore expect the torque to 
 vary as the turns, 1060 X f = 795 as against 800. Further, 
 the total mechanical energy developed by a motor in a given 
 time varies as IwC x revolutions per minute. We have 
 800 X 124 = 992, also 10GO X 93 = 986. 
 
 Another set of curves, relating to a G.E. 1200 motor, 
 give with a ratio of reduction of 3*53 to 1 : 
 
 Armature. 
 
 Force in Ibs. at rim 
 of 33 in. wheel. 
 
 Amperes. 
 
 Revolutions per 
 minute of motor. 
 
 2-Turn 
 
 1200 
 
 96 
 
 160 
 
 3-Turn 
 
 1820 
 
 96 
 
 110 
 
 In this case 1820 X | = 1210 as against 1200 ; also 
 1200 X 1GO = 192 X 10 3 ; whereas 1820 X HO = 200 X 10 :5 . 
 
 The Testing of Motors. In addition to the test of the 
 motor or motors on the car itself (see p. 88), it is im- 
 portant to see that they will fulfil certain conditions 
 before they leave the testing-shop. An accurate method 
 for such testing is mentioned on p. 202 ; but, in any case, 
 the motor should be run at the average speed it will ex- 
 perience in practice, say the speed corresponding to seven miles 
 per hour of the car, for six hours, when it is developing the 
 requisite torque. For instance, in the G.E. 800 motor 
 (p. 42) the force at the rim of a 33-inch wheel through 
 
44 ELECTRICAL TRACTION. 
 
 single reduction of 4*78 to 1, is 800 Ibs. in the three- 
 turn armature. When inclosed, the armature and field 
 under such working conditions should not exceed a rise in 
 temperature on that of the surrounding atmosphere of the 
 number of degrees to be stated in the specification. Such 
 motors should be able to stand safely 50 per cent, over- 
 loading for short times. 
 
 The Author is indebted to the "VVestinghouse Electric and 
 Manufacturing Company for the following particulars of a 
 workshop test of their No. 4G 500-Volt Eailway Motor. 
 The ratio of reduction between armature and car-axle is 
 4'86 to 1 ; and the duration of test one hour. During 
 this time the motor developed a force of 1000 Ibs. at the 
 rim of a 30-inch car-wheel at a speed corresponding to 
 about 8 miles per hour of the car. This is equivalent to 
 doing work on the rails at the rate of 21 H.P. at this speed. 
 At the end of the hour the armature had risen in tempera- 
 ture 49 C., the field G4 0., and the commutator 30 C. 
 The current in the armature was 38 amperes, and the 
 applied potential, 500 volts. The net efficiency cold was 
 80 per cent., and with an increased resistance of 20 per 
 cent, this fell to 78 per cent. This test, the Author is 
 informed, is much more severe than those which the motor 
 would be subjected to under normal running conditions. 
 
 The Magnetic Properties of Cast Steel. The shell of 
 these inclosed motors is now made of cast steel, and this 
 material has played an important part in the development 
 of the tramway motor, in addition to the multi-polar 
 generator. It is instructive to compare the magnetic 
 properties of cast steel with good wrought iron. 
 
 The figures in Table IV. relate in part to a sample of 
 almost pure iron tested by Messrs. Lydall & Pocklington,* 
 which can be taken as a standard with which to compare 
 the other samples given in the Table. This table gives 
 
 * See " Magnetic Properties of Pure Iron," Proc. Roy. Soc., vol. 52, 
 p. 228. 
 
CAST-STEEL. 
 
 45 
 
 the magnetic induction per square centimetre, B, in terms 
 of magnetizing force, H, both in O.G-.S. units. Sample I. 
 is the pure iron just referred to. Sample II. is an un- 
 forged steel casting. Sample III. is a steel forging furnished 
 by Mr. R. Jenkins as a sample of forged ingot metal for 
 dynamo-magnets. Sample IV. is of Low Moor bar forged 
 into a ring, annealed and turned. These figures have been 
 taken from a paper on the " Magnetic Testing of Iron and 
 Steel," by Professor Ewing.'* 
 
 TABLE IV. 
 
 Magnetizing force H 
 
 Magnetic induction B. 
 
 
 I. 
 
 ir. 
 
 III. 
 
 IV. 
 
 5 
 
 12,700 
 
 10,900 
 
 12,300 
 
 10,900 
 
 10 
 
 14,980 
 
 10,320 
 
 14,920 
 
 13,120 
 
 15 
 
 15,800 
 
 14,350 
 
 15,800 
 
 14,010 
 
 20 
 
 16,300 
 
 14,950 
 
 16,280 
 
 14,580 
 
 30 
 
 16,950 
 
 15,660 
 
 16,810 
 
 15,280 
 
 40 
 
 17,350 
 
 16,150 
 
 17,190 
 
 15,760 
 
 50 
 
 
 
 16,480 
 
 17,500 
 
 16,060 
 
 GO 
 
 
 
 16,780 
 
 17,750 
 
 16,340 
 
 70 
 
 
 
 17,000 
 
 17,970 
 
 16,580 
 
 80 
 
 
 
 17,200 
 
 18,180 
 
 16,800 
 
 90 
 
 
 
 17,400 
 
 18,390 
 
 17,000 
 
 100 
 
 
 
 17,600 
 
 18,600 
 
 17,200 
 
 Take magnetizing force 40. The pure iron, sample L, 
 gives for B, 17,350 as against 16,150 for the steel casting, 
 whilst the steel forging, sample III., gives 17,190. Both 
 samples II. and III. are but little inferior to the pure iron, 
 and are superior to the annealed wrought iron, which gives 
 15,7 GO for this magnetizing force. It will be seen, therefore, 
 that steel castings can be produced which are equal to the 
 best wrought iron, and therefore the castings for multi-polar 
 
 See Proceedings Institution of Civil Engineers, vol. cxxvi. p. 185. 
 
46 ELECTRICAL TRACTION. 
 
 dynamos can be made much lighter than was formerly the 
 case when cast-iron was used, since, with the cast-steel, 
 the sectional area of the frame effective for magnetism need 
 not be greater than the actual section of the magnet cores 
 themselves, whereas for the cast-iron it was usual to make 
 it about twice as great. 
 
( 47 ) 
 
 CHAPTER III. 
 
 OVERHEAD AND OVERGROUND CONDUCTOR SYSTEMS. 
 
 General Remarks. Financially speaking, the most suc- 
 cessful means of transmitting energy in the form of 
 electricity to motor-cars on a tramway is, undoubtedly, the 
 overhead trolley system. A copper conductor, connected to 
 the source of supply, is fixed to and insulated from brackets, 
 which are supported by poles placed at intervals along the 
 route. Contact is made between this conductor and the 
 one pole of a motor, or motors, on the car by means of 
 a small trolley wheel, which is carried at the end of a 
 flexible pole attached to the car roof. In another arrange- 
 ment, contact is made with the overhead conductor by 
 means of a transverse horizontal rod suspended by flexible 
 end poles. The other pole of the motor, or motors, is con- 
 nected almost universally to the wheels of the car, and 
 hence to the rails, which form the return conductor. There 
 are instances in which a second overhead conductor is pro- 
 vided, in which case a second trolley wheel is used to connect 
 the other pole of the motor with the return, not using the 
 rails as a return conductor at all. This latter type is called 
 the "double trolley" system, and has advantages and 
 disadvantages. Its advantages with direct currents are 
 that little or no interference is occasioned to telephones 
 or telegraphs, and there can be, in this system, no earth 
 currents. Its disadvantages are its higher first cost, and the 
 difficulty of suitably arranging the conductors at crossings. 
 A modern and interesting example of this double trolley 
 
48 ELECTRICAL TRACTION. 
 
 system, in which the rail is also used as a return, is at 
 Lugano in Switzerland, where the line, which is about three 
 miles in length, is operated on the three-phase alternate 
 current system. Another instance is that of Cincinnati, 
 where the two overhead conductors transmit energy to the 
 cars, the rails not being used as a conductor. There are 
 not many instances of this system, but there are some 
 hundreds of single trolley systems in which the rails are 
 used as the return conductor. At the end of 1896 there 
 appears to have been close upon 13,000 miles of tramway in 
 the United States operated on the overhead trolley system, 
 as against, at the most, 1000 miles for all classes in Europe. 
 
 An allied system to the overhead trolley is that in which 
 a continuous conductor is carried on insulators above or 
 near the surface of the ground. This system was probably 
 the earliest adopted for the electrical transmission of energy 
 to cars. It is, of course, out of the question for roadways, 
 but it is used for elevated railways, underground subways, 
 and, in fact, anywhere where the track is not used for any 
 other purpose than the railway. We have instances of 
 this system in the City and South London Railway, and 
 the Liverpool Overhead Railway. Such conductors are also 
 used on the Chicago Elevated Railway. 
 
 The transmission of energy to motors in a tramway is quite 
 a different problem to the simple transmission of energy 
 from, say, a generator to a motor at a fixed distance ; here a 
 certain drop of potential takes place, and is allowed for. In 
 an electric tramway the conditions, as regards potential 
 difference, more nearly approximate to the supply of electrical 
 energy from central stations for lighting purposes. The 
 maximum potential difference, which the Board of Trade 
 sanction, and which is almost universally used, is 500 volts. 
 This is a potential difference which can be handled with 
 tolerable ease in motor and apparatus, and is not fatal to 
 human beings, but may be fatal to horses. The car is con- 
 structed to work properly at 500 volts, say, and, of course, 
 
SPEED SERVICE, ETC. 49 
 
 it must have this applied potential, or nearly it, at any part 
 of the system. The importance of constancy of potential 
 difference is pointed out at p. 30. This consideration 
 makes the problem more difficult to handle, and we shall 
 have to consider the various remedies adopted in practice. 
 
 Before commencing the design of any scheme, certain 
 information must be obtained as to general matters. For 
 instance, the service or frequency at which the cars must 
 run, the average speed and the length of the line give the 
 number of cars to be operated ; and provision must be made 
 for spare cars to meet greater demands upon the system, 
 such, for instance, as at holiday times, or in summer, if the 
 place is a pleasure resort, and also to replace breakdowns. 
 In Great Britain the average speed of cars, including stops, 
 may be taken to be about seven miles per hour. If v be 
 this velocity and I the length of the journey in miles, then 
 
 - is the time taken for a car to complete the round trip. 
 
 v 
 
 If the cars have to run at intervals of m minutes, then 
 
 gives the number of cars. Take, as an example, the 
 vm 
 
 Leeds Tramways (see p. 105) ; here the round trip is about 
 fourteen miles. Assume that the average speed is seven 
 miles per hour, then the car completes the trip in two hours. 
 If the service is seven and a half minutes, then the number 
 of cars is sixteen. The problem is, of course, influenced by 
 consideration as to single or double track, and also in branch 
 services, where the demands of the different branches vary. 
 
 The power taken to propel a tram-car is, as we have seen, 
 an important item, and on a given system is greatly 
 influenced by the nature of the track, as to whether 
 there are steep gradients, sharp curves, etc. An ordinary 
 tram-car, complete with motors, etc., to seat about 44 
 passengers, weighs, with passengers, about ten tons. At an 
 average velocity of seven miles per hour, such a car can 
 be propelled one mile, on ordinary lines, by the expenditure 
 
So ELECTRICAL TRACTION. 
 
 of about one Board of Trade unit ; that is, 1000 watt 
 hours. A Board of Trade unit is equivalent to about 
 2,650,000 foot-pounds, or 1*34 H.P., acting or exerted 
 for one hour. We have seen that the power taken to 
 get a train under weigh is about three times the average 
 power taken to keep it going when once accelerated. Take 
 the average speed at seven miles per hour. The time for 
 one mile is one-seventh of an hour. One thousand watt 
 hours equals the watts multiplied by one-seventh hours. 
 Therefore the watts = 7000 ; H.P. = 9'4 ; and, since the 
 volts are assumed to be 500, the amperes = 14. We see 
 that, whereas the average rate at which work must be done 
 on the car is from 9 to 10 H.P., the maximum H.P. may 
 be 33. According to Cunningham,* the power taken at 
 Montreal equals about 260 watt-hours per ton mile. The 
 average speed is 7 '5 miles per hour. The cars are stated to 
 weigh six tons, and on this basis we get 1*56 Board of Trade 
 units per mile, per car, which, with passengers, might weigh 
 ten tons. It must be remembered that Montreal is on a 
 slope, and the gradients at right angles to the river are very 
 severe. They are stated to be 1 in 10 and 1 in 8. In any 
 case, contractors guarantee the propulsion of a tramcar, 
 weighing about ten tons, one mile, at about seven miles per 
 hour, by means of the expenditure of one Board of Trade 
 unit. If the cars are many, say twenty or thirty or more, the 
 average power differs less from the maximum power than if 
 the cars are few. Suppose, in a certain district, there are 
 two cars only ; then they may demand 70 H.P. if they start 
 together, which is very likely ; whereas, if there are many 
 cars, it is, one may say, almost infinitely improbable that 
 they will all start at the same moment, and therefore the 
 average power differs less from the maximum power. 
 
 In a simple system, where no precaution is taken to 
 prevent drop of potential due to current in the line, the 
 
 * See Journal Institution Civil Engineers, vol. cxxiv., 1895 96, 
 p. 287. 
 
FEEDERS. 51 
 
 potential difference will fluctuate very much in the outlying 
 districts, whereas, in the vicinity of the station, the fluctua- 
 tions will be smaller. Although the Board of Trade issue 
 no limitation as to the drop of potential in the working- 
 conductor itself, it is of importance for the successful 
 operation of the line to limit it (see p. 30). This brings us 
 to the important question of Feeders. 
 
 Feeders. By the term "feeder" the Author wishes to 
 imply any system by means of which energy can be given 
 to the cars at distant points, which energy is transmitted 
 from the power-house proper, by means other than the 
 ordinary line and return. Suppose A is the power-house, 
 then, in Fig. 1C, we have a simple system, the cross lines 
 
 being intended to represent motors placed between - the 
 conductor and the rail. Such a system as we have seen 
 is open to the objection that the fall of potential difference, 
 as one proceeds to the distant points, becomes greater, and 
 the fluctuations severe. The system might, of course, be 
 tolerated on a short line. 
 
 Fig. 17 illustrates a simple system, to which feeders have 
 been added. These consist of heavy conductors, which 
 transmit energy directly from the station at A to the 
 feeding-points ; that is, the point where such conductors 
 are attached to the working conductor and return respec- 
 tively. Such a system has long been used in connection 
 with electric lighting, and is well known. The South 
 London Railway may be quoted as an instance.* In large 
 
 * See Dr E. Hopkinson, Proc. List. Civil Eng., vol. cxii. p. 209. 
 
52 ELECTRICAL TRACTION. 
 
 installations the first cost of the copper is a great con- 
 sideration in the design of feeders. 
 
 Fig. 18 illustrates another system of feeders. This system 
 receives energy from the station at a high potential, thereby 
 requiring a small conductor, since its size is in part deter- 
 mined by the current required to be transmitted. At the 
 point where the energy is to be delivered to the system a 
 motor-generator is placed. This consists essentially of two 
 machines coupled together, the one a motor, which receives 
 the energy from the station at high potential, the other 
 a generator, which delivers this energy to the system at 
 the lower required potential. In such a system of feeders 
 the consideration is not altogether the first cost of the 
 copper, which may be small, but also the first cost of 
 the motor-generators, their efficiency, the housing and 
 the attention they require. The tramway system at 
 Dublin gives an illustration of this kind of feeder. Here 
 energy is transmitted by means of three-phase alternate 
 currents at 3000 volts to a synchronous motor, directly 
 coupled to a generator, giving energy, in the form of a 
 direct current, at 500 volts to the line. We note that in 
 this case there is no storage except from the momentum 
 of the moving parts. 
 
 Fig. 19 shows another system of feeders, which consists 
 of storage-cells capable of delivering energy to the line 
 and receiving energy from the station at A. This system 
 has the advantage that the batteries can be steadily charged 
 almost continuously, and when called upon can suddenly 
 discharge energy at a great rate. This aspect of storage is 
 important. An illustration of this system is at Leeds, 
 where the simple feeder (Fig. 17) is also used in combina- 
 tion. As shown in Fig. 19, the current from the cells to 
 the station takes path along the rails. A drop of potential 
 will, therefore, occur on account of this independently of 
 the cars. 
 
 The whole subject of feeders is one of great importance, 
 
FEEDERS. 
 
 53 
 
 and is influenced by many considerations. An examination 
 of the figures just given shows that one effect of a feeder 
 is to seriously disturb the distribution of current in the 
 rail return. Take Fig. 16, which represents a simple 
 system of supply. Between A and one end of the rails 
 the potential difference is that due to the gradually 
 
 
 
 .1 
 
 
 
 
 L 
 
 
 
 
 
 
 " 
 
 
 A 
 
 FIG. 17. 
 
 JL 
 
 
 5> 
 
 A 
 
 Via. IS. 
 
 A 
 
 FIG. 19. 
 
 diminishing current into the resistance of the rail. A very 
 stringent provision made by the Board of Trade, with 
 regard to drop of potential in the rails, is that this shall 
 not exceed seven volts (see Regulation 7, Appendix, p. 
 227), the object being to secure that the earth shall not 
 be compelled to carry more current than necessary, on 
 account of the injurious effects due to electrolytic action 
 
54 ELECTRICAL TRACTION. 
 
 on underground pipes. A little consideration shows that 
 in Figs. 17, 18, 19, the drop of potential due to the 
 current in the rails between the point A and where the 
 feeder is attached, may be very small, although just as 
 many cars are being operated as before. The fall of 
 potential difference, then, between A and the feeding-points 
 is no criterion that there are, or are not, earth currents. 
 A safer condition would be to state that the maximum 
 potential difference between any two points of the rails 
 shall not exceed such a potential difference as it can be 
 demonstrated will not be injurious. This matter will 
 receive attention when we come to deal with the track 
 itself, for which there are other regulations with regard to 
 leakage (see p. 82). 
 
 With regard to the simple feeder system (Fig. 17), several 
 conditions have to be decided before determining the size 
 and number of feeders for a given line. If the feeder is 
 to keep the line potential where it makes contact, the same 
 or nearly the same as the potential difference of supply at 
 A, a fair value of the current in such a feeder can be 
 obtained by a consideration of its point of contact with 
 the line and the number of cars between A and the end 
 of the line. Clearly under the assumption of its main- 
 taining the line potential difference the same as at A, and 
 assuming the line and rail resistance, per unit length, 
 constant, the feeder would supply current to one-half the 
 cars between it and A if equally spaced out, and to all 
 the cars beyond it to the end of the line. From this the 
 current can be roughly judged, for if the cars be many, 
 the average current per car could be taken. The drop of 
 potential in the feeder itself has to be considered ; if there 
 are many feeders, and one portion of the plant is devoted 
 to them, its potential must suit them all. The drop of 
 potential difference in a feeder for a given current depends 
 upon its resistance, and this, in turn, for a given length, 
 depends upon the material of the feeder and its cross- 
 
ECONOMY OF ELECTRICAL CONDUCTORS. 55 
 
 sectional area, In any case, if the potential difference 
 between the ends of the feeder and the current in it be 
 given, the rate of dissipation of energy in such feeder is 
 also given. This, then, brings us to the continuing cost 
 of the conductor. 
 
 Continuing Cost. Lord Kelvin, in 1881, read a paper 
 before the British Association on the " Economy of Con- 
 ductors of Electricity," and there gave his law with regard 
 to the same. It is this : That the interest on the first 
 cost of the conductor and allowance for depreciation should 
 be equal to the cost of energy actually dissipated in the 
 conductor in a given time, say, per annum. The first cost 
 of a conductor for an installation of a given size and 
 type is easy to arrive at, by obtaining quotations from 
 cable manufacturers, which prices, per mile say, will vary 
 with the quality, the amount ordered, and the company sup- 
 plying it. So that price lists can only be taken as a rough 
 guide in estimating the cost. Then, again, there is the 
 varying price of material. If a conductor be laid in a 
 trench specially got out for it alone, it would seem only 
 fair to charge such cost to the cable. But this does not 
 always happen. If, for example, the cable be laid in earthen- 
 ware ducts, there are probably other cables in the same 
 duct, in which case the duct is really put down for a 
 number of cables, and perhaps provision made for drawing 
 in more. It is only experience and an examination of the 
 given conditions which can enable an engineer to properly 
 judge as to the first cost of his feeder. But, looking at the 
 feeder alone, it is only for large sizes that one pays in 
 almost direct proportion to the copper used. The curve #, 
 in Fig. 20, gives the relation between the cost in shillings 
 per yard per square inch cross-section of a lead-cased iron- 
 sheathed cable, and the cross-section in square inches. 
 Beyond about 0'2 square inch the curve is almost straight 
 and gradually sloping towards the horizontal ; that is to 
 say, so far as the cost of this cable is concerned, it is more 
 
56 ELECTRICAL TRACTION. 
 
 nearly proportional to the volume of copper beyond 0-2 
 square inch than up to 0'2 square inch. This shows that 
 it would be more economical, from this point of view, 
 to put down this cable at about 0*2 or 0*3 square inch 
 sectional area or greater. The curved portion, up to about 
 0*2, shows that the covering material is an important factor. 
 The curve &, Fig. 20, gives the relation between the cost 
 in pounds sterling per statute mile and square inches cross- 
 
 2000 
 
 1000 
 
 100 
 
 6 
 
 50 
 
 1-0 
 
 1-5 
 
 FIG. 20. 
 
 section, and this, although curving down towards the origin, 
 would not pass through it. 
 
 It seems to be pretty well established that in traction 
 work with efficient machinery a Board of Trade unit can 
 be produced for 0*5 pence, or a little over. . Taking this as 
 a basis, and assuming the system as working 5000 hours 
 per annum ; one horse-power = 746 X 5000 = 3,730,000 
 watt-hrs. = 3730 Board of Trade units ; therefore, the 
 cost for one horse-power per annum is 7' 6. Let us, how- 
 ever, assume that it actually costs 10 per annum. A 
 feeder, 0'25 square inch cross-sectional area, weighs, say, 
 5000 pounds per statute mile. Suppose it is three miles 
 long, then it weighs 6'7 tons. Taking the cost of the 
 
ECONOMY OF ELECTRICAL CONDUCTORS. 57 
 
 cable at 550 per mile, if sheathed, the total cost equals 
 1650. The cost per ton is 246. Allow on this 10 per 
 cent, for interest and depreciation, that is, 25 per annum. 
 We ought, therefore, to waste one horse-power for every 
 |4 ton, that is, in this particular case we ought to waste 
 17 H.P. Take the resistance of the cable as 0'177 ohm 
 per statute mile ; the total resistance is 0*53 ohm. Then 
 
 C 2 X 0*53 
 should equal 17 ; whence C equals 150 amperes. 
 
 The current density, then, is 600 amperes per square inch, 
 and the drop of potential between the ends of this feeder 
 is 150 x 0*53 = 79 volts. We see from this that the 
 heating effect will not be injurious to the cable, and the 
 drop of potential difference should suit the particular 
 requirements in hand. The engineer should at least see 
 that this law of economy is fairly fulfilled, always having 
 due regard to heating and fall of potential. In laying 
 down feeders in connection with the rail return, in order to 
 comply with the Board of Trade limit of 7 volts drop in 
 the rail, this law of continuing cost may have to be deviated 
 from. For instance, if these feeders are connected together 
 at the negative terminal of the machines, the long feeders 
 may have more copper than would be dictated by Kelvin's 
 Law, owing to the requisite diminished fall of potential 
 difference, the short feeders complying with it. A way 
 out of the difficulty would be to connect the long- 
 feeders together, and supply through them at greater 
 potential difference by means of a separate machine, the 
 short feeders being treated similarly. This leads to sub- 
 division of plant. A further remedy would be the use of 
 storage-cells, or another dynamo of small potential difference. 
 Such cells or dynamo are sometimes known as " Boosters." 
 When it is considered that the conductors might cost as 
 much as 30 per cent, of the cost of the plant, one can 
 realize how important it is to lay down copper to the 
 greatest advantage. The actual laying of cables is a subject 
 
58 ELECTRICAL TRACTION. 
 
 beyond the scope of this work ; but it may be mentioned 
 that nowadays the best practice seems to be to run insulated 
 cables either through porcelain or bitumen ducts, or through 
 iron pipes. 
 
 These remarks on feeders refer to electric traction in 
 general. The Author does not intend to enter into a full 
 discussion as to the relative continuing and first cost of 
 the other forms of feeders given in Figs. 18 and 19. The 
 problem must be attacked in the same manner, but great 
 experience and care are required to form a good judgment 
 as to which is the best to adopt in a particular case. 
 
 Board of Trade Regulations with regard to Underground 
 Feeders. It is usual in tramway work to place feeders under 
 the ground, and to continuously insulate them, although in 
 some cases feeders are carried overhead and insulated on poles. 
 In the case of railways, bare copper conductors, or ordinary 
 rail sections can be used for feeders (see p. 78). An important 
 regulation with regard to underground feeders is as follows : 
 "The insulation resistance of all continuously insulated cables 
 used for lines, for insulated returns, for feeders, or for other 
 purposes, and laid below the surface of the ground, shall 
 not be permitted to fall below the equivalent of 10 megohms 
 for a length of one mile. A test of the insulation of all 
 such cables shall be made at least once in each month."'"' 
 This is a very modest insulation resistance for a good class 
 of cable, when one considers that as much as 2000 megohms 
 per statute mile, after one minute's electrification, is a 
 guaranteed insulation resistance when the cable is new. 
 But it must be remembered that at any time during the 
 life of a cable it must have 10 megohms per statute mile 
 resistance, which is a very different matter. Of course 
 this should be covered by the agreements which contractors 
 enter into with regard to the maintenance of the cables. 
 
 Overhead Conductor. It will be well to state here the 
 
 * See "Blackpool Corporation Tramways," February 11, 1897, 
 Regulation XI., Appendix, p. 228. 
 
THE OVERHEAD CONDUCTOR. 59 
 
 principal regulations of the Board of Trade with regard 
 to overhead insulated conductors. They are as follows : 
 " The insulation of the line, and of the return when insu- 
 lated, and of all feeders and other conductors shall be so 
 maintained that the leakage current shall not exceed one 
 hundredth of an ampere per mile of tramway. The leakage 
 current shall be ascertained daily, before or after the hours 
 of running, when the line is fully charged. If at any time 
 it be found that the leakage current exceeds one-half an 
 ampere per mile of tramway, the leak shall be localized and 
 removed as soon as practicable, and the running of the 
 cars shall be stopped, unless the leak is localized and 
 removed within 24 hours ; provided that where both line 
 and return are placed within a conduit this regulation shall 
 not apply." ' 
 
 " The electrical pressure or difference of potential between 
 any suspended conductors used in connection with the work- 
 ing of the tramways by electrical power and the earth, or 
 between any two such suspended conductors, shall in no 
 case exceed 500 volts continuous pressure." f 
 
 " The suspended conductors used in connection with the 
 working of the tramways by electrical power shall be 
 in no part at a less height from the surface of the street 
 than 17 feet, and shall be securely attached to supports 
 at intervals not exceeding 120 feet." \ 
 
 " The line wire shall be divided up into sections not exceed- 
 ing (except with the special approval of the Board of Trade) 
 one quarter of a mile in length, between every two of 
 which shall be inserted an emergency switch and a safety 
 fuze or cut-out, constructed to act with a current exceeding 
 the maximum working current by 50 per cent., which 
 
 * See "Blackpool Tramways," February 11, 1897, Kegulation X., 
 Appendix, p. 228. 
 
 t See Dublin Southern District Tramways, August 17, 189G, 
 Kegulation X., Appendix, p. 233. 
 
 \ Ibid , Eegulation XI., Appendix, p. 233. 
 
60 ELECTRICAL TRACTION. 
 
 apparatus shall be so enclosed as to be inaccessible to 
 pedestrians." * 
 
 This is in conflict with Regulation No. 9 of the Board 
 of Trade, dated July 24, 1896, and February 11, 1897, 
 Appendix, p. 228, in which this distance is stated to be one- 
 half mile, which is quite short enough. 
 
 We see from the regulations just quoted, that the insula- 
 tion resistance of the line shall be so maintained that it 
 shall not fall below 50,000 ohms per mile. This is on the 
 assumption that the potential is 500 volts, and that if the 
 resistance gets down to 1000 ohms per mile the fault is to 
 be at once localized and removed. When one compares 
 this with the 10 megohms for underground insulated con- 
 ductors (see p. 58), it is evident that, if possible, the over- 
 * head conductor should be the only conductor. But this 
 is not always possible, although there is a tendency towards 
 heavier conductors. The question of sufficiency of copper 
 depends upon the distance between the feeder-points and 
 the number of cars to be supplied with energy between 
 such points, for the drop of potential difference should not 
 be excessive (see p. 30). A very common size of hard- 
 drawn copper for overhead trolley lines is 0'324 inch 
 diameter, or 0*0829 square inch cross-sectional area ; but 
 at Leeds Dr. Hopkinson has employed a copper conductor 
 0-4 of an inch diameter, having a cross-sectional area of 
 0'12G square inch, and 98 per cent, conductivity. One 
 must bear in mind that a great increase in the size of con- 
 ductor means also an increase in strength and cost of the 
 poles and insulators. If the drop of potential between any 
 two feeding-points is too severe with a given overhead 
 conductor, the remedies are to increase its diameter, or to 
 lay an underground insulated conductor placed in contact 
 with the overhead conductor at intervals, so as to increase 
 the effective cross-sectional area. 
 
 * See Dublin, etc., August 17, 1896, Regulation XII, Appendix, 
 p, 233, 
 
THE OVERHEAD 'CONDUCTOR. 61 
 
 Suppose that we have two points in a conductor, a 
 and (3, and that the potential differences between each of 
 these and the earth are maintained constant and equal. 
 Let the resistance between a and /? be 2R, let the distance 
 a /3 be divided into 2n equal parts, and let there be 2ti - I 
 trains, making simultaneous contact with the conductor at 
 each of the intermediate points, and each demanding the 
 same current x. Then the whole current X entering this 
 section at a is equal to x(n -|) ; also R = nr, where r 
 is the resistance of the conductor between two consecutive 
 
 trains. Then the maximum drop in volts equals xi 1 ^- \ 
 
 - , since r = ~. Of course these conditions are never 
 V n 
 
 realized, nor is it possible to get at the exact conditions 
 in working. All one can do is to get a fair idea of what 
 the maximum drop might be in a given case. As an 
 example, let us examine the system at Dublin (see Fig. 21, 
 and p. 190 ; also Engineering, June 5 and 12, 1896). In 
 Fig. 21, A is the Ballsbridge Power House, B and C are 
 sub-stations at Blackrock and Dalkey. Take the length 
 of this line to be 7f miles ; then the round trip equals 
 15-| miles. Assume there are 20 cars and trailers on 
 the line at the same time, and that the average velocity 
 is 7f miles per hour ; then the service would, under 
 these conditions, equal six minutes. The average distance 
 between the trains equals about 1360 yards. There are two 
 trolley wires, each 0'0829 square inch section, and between 
 the sub-stations at Blackrock and Dalkey an under- 
 ground insulated conductor of 0*1435 square inches section 
 connected to the trolley wire at eight intermediate points. 
 Assume, then, we have a continuous copper conductor 
 between these two points, B and C, Fig. 21, of 0*309 square 
 inch section. Take the distance to be 4-| miles. The 
 resistance of such a length is, say, 0'664 ohm. Assume 
 that there are 10 cars, each with trailers on the two lines 
 
62 ELECTRICAL TRACTION. 
 
 between these points, equally spaced out, and that each car 
 is taking 20 amperes ; 20 amperes at 500 volts corre- 
 sponds to about 13 H.P., and is taken to be the average 
 
 Two TROLLEY WIRES EACH -324 IN. DIAMETER -- 
 
 
 ^FEEDER 
 
 0754 5Q. IN. 
 
 
 
 ^FEEDER '1435 SQ. IN. 
 
 
 i 
 in 
 
 ^C DOUBLE 
 
 11 
 
 TV 
 
 TRACK 76 LBS. PER YARD * 
 \ 
 
 \ 
 
 TRIPLE CONCENTRIC CABLE 
 FIG. 21. 
 
 power per car. At an average speed of seven miles per hour 
 the work required per mile is about 1^ Board of Trade 
 
 units per car and trailer. Then the drop in volts = 
 2QX 6 ^ 0-332 
 
 V"D 
 
 ficiently numerous, the drop in volts reduces to^ , 
 
 2 
 
 which is then quite near enough, since the exact conditions 
 are not known.] This is the drop of potential in the report 
 alluded to. Take the maximum drop in rail. The resist- 
 ance neglecting joints is about 0*00984 ohm per mile,'"' 
 since the rail has 76 Ibs. weight per yard ; 4 '5 miles have 
 
 resistance 0'0444 ohm. Then -^ R '= 1-33 volts. 
 
 Fig. 22 gives a diagram of the system of conductors put 
 down by Dr. Hopkinson at Leeds. The total length of 
 the line is about 12,300 yards. There is no underground 
 conductor in parallel with the trolley wire ; therefore we 
 have only the drop in the wire itself to consider. Take 
 the points B' and C' in Fig. 22, distant about 3250 yards. 
 Assume that there are three cars between these points, and 
 that the potential difference at these points is maintained 
 
 * See page 76. 
 
THE OVERHEAD CONDUCTOR. 63 
 
 constant at 500 volts. The trolley wire has a diameter 
 of 0-4 of an inch and a cross-section of 0'126 square inch ; 
 3250 yards have a resistance of 0-668 ohm. In this case 
 
 KIRKSTALL WELLINGTON ROAD BOAR LANE SHEEPSCAR 
 
 ROUNDHAY 
 
 STATION 
 FIG. 22. 
 
 n = 2, and assume x = 20 since the cars are few ; this is 
 greater than the average and less than the maximum 
 current. The drop is 6'7 volts. In actual work the 
 potential difference at the feeding-points is not, of course, 
 absolutely constant as has been assumed. It must vary 
 to some extent. But these two instances are given simply 
 to show the sort of figure one might obtain in actual 
 working. 
 
 Appliances used in Erection of Conductor. The methods 
 and materials used in the erection of overhead wires is a 
 subject of great practical importance, and about which 
 much has been written. There are a variety of ways of 
 suspending the overhead conductor ; for instance, in narrow 
 streets it can be supported by an insulator fixed to a span 
 wire bridging across the opposite walls, or the span wire 
 may be supported by poles on each side of the track. This 
 latter method is adopted in many places where plenty of 
 width is available. Again, the conductor or conductors can 
 be supported on insulators by means of poles, which can be 
 placed either in the middle or at the side of the roadway. 
 When placed in the middle of the road, the track being 
 double, such poles can be fixed in an island of pavement 
 with a pillar at each end to protect it from injury. Such 
 poles can also be used for the erection of arc lamps, and 
 so serve a double purpose ; this is done at Dover and 
 
64 ELECTRICAL TRACTION. 
 
 also at Leeds (see Fig. 29). At the present time steel 
 tubular poles are universally adopted, these being securely 
 fixed in concrete foundations. A great deal of time and 
 money has been spent upon obtaining efficient means for 
 insulating the conductor from the bracket arm or span 
 wire, and the devices used are somewhat numerous. The 
 great point in such apparatus is the insulating material ; 
 this should be hard, durable, and protected as much as 
 possible from moisture. In all such devices this material 
 is in compression, for insulation materials can be made 
 which will stand great compressive stresses. Well-known 
 insulation materials are the 2Etna and Medbury. One of 
 these insulators is shown in Fig. 23. The clamp to which the 
 
 FIG. 23. 
 
 conductor is soldered is screwed on to a bolt covered with 
 hard insulating material. The bolt again is supported by 
 a casting in such manner that the material round the 
 underneath side of the bolt head is in compression. This 
 is the general method adopted in suspending a conductor. 
 A. part of the same casting consists of a hood, which protects 
 the insulation material from rain and moisture. 
 
 Various methods are adopted for supporting the overhead 
 
ERECTION OF OVERHEAD CONDUCTORS. 
 
 wire. This is almost always circular, and is soldered into 
 a support as shown in Fig. 24. This method, although 
 mechanically good, would seem to disturb the smooth 
 running of the trolley wheel at high speeds. Another 
 
 FIG. 24. 
 
 section for the wire, as shown in Fig. 25, has been tried, 
 the support being clamped to the upper portion. This 
 gives smooth running for the trolley-wheel, but an objection 
 to this section is that the wire is liable to kink. The 
 circular pattern can be supported by means of a clamp, 
 which grips the wire by two grooves, milled out as shown 
 
 FIG. 25. 
 
 in Fig. 2G. The circular conductor can also be gripped as 
 shown in Fig. 27. 
 
 A V' sna P e( l casting, usually called a " Frog," is used for 
 branch circuits, and guides the trolley-wheel in the desired 
 direction. 
 
 The Board of Trade regulation (see ante, p. 59 ; and 
 Appendix, p. 228) states that the line wire shall be divided 
 
 F 
 
66 
 
 ELECTRICAL TRACTION. 
 
 up into sections. This is generally carried out at a pole 
 by means of a section insulator. The wire at this point 
 is broken, and the space between the two ends, of about 
 one foot, filled up with a piece of hard wood or fibre in 
 such manner that the trolley runs along the same. The two 
 ends are carried by conductors down the pole to a street box 
 
 FIGS. 26 AND 27. 
 
 containing fuzes, and, if necessary, the feeder is connected 
 to the line at one of these points. In this manner, by 
 withdrawing the fuze at any two adjacent boxes, the line 
 in between them can be isolated and submitted to test. 
 Such a section insulator is shown in Fig. 28. With a single 
 
 FIG. 28. 
 
 trolley on the car it is obvious that the circuit must be 
 broken at these points. The motorman knows when this 
 would occur, and breaks the circuit just before reaching 
 such point, making it immediately after. In this manner 
 the flash does not occur at the insulator. This breaking 
 
LIGHTNING ARRESTERS. 67 
 
 of contact between the trolley wheel and the line is con- 
 sidered by some to be objectionable ; in such cases, storage- 
 cells are usually employed on the car for lighting purposes 
 (see p. 215). It is usual to erect guard- wires where neces- 
 sary to prevent telephone and telegraph wires from falling 
 on the over-head conductor. 
 
 Lightning Arresters. The subject of lightning arresters 
 is of importance since the line wires are liable to be struck. 
 There are several good arresters for use in traction work. 
 The " Ajax," the British Thomson-Houston, and the West- 
 inghouse or Wurts arresters are largely used, and are placed 
 on each car, on the posts and in the central station. 
 
 The " Ajax " arrester consists of a series of specially 
 constructed fuzes, arranged in such manner that one after 
 the other is automatically brought into operative position 
 by the action of the discharge current as it consumes each 
 fuze in turn. The fuze itself consists of two insulated 
 wires placed with their ends overlapping, but not touching, 
 the distance between the parallel portions of the ends being 
 secured by sealing such ends into a plug of insulating 
 material at each end of a small glass tube. When the 
 lightning strikes, the current leaps across the gap and 
 burns out the wire. This allows a small metal ball to fall 
 into contact with the end of the next fuze lower down, 
 thereby leaving it in position for the next discharge. The 
 general arrangement is such that a fresh set of fuzes can be 
 quickly inserted. 
 
 The well-known magnetic blow-out principle is used in 
 the Thomson-Houston lightning arrester. It provides a 
 ready path for the lightning across a small air-gap, and 
 the line current immediately tends to flow to earth ; but 
 this is prevented by making it traverse a coil which 
 energises an electro-magnet, the field of which immedi- 
 ately extinguishes the arc formed. It is obvious that such 
 an arrester provides very great protection, since a quick 
 succession of lightning flashes can be as quickly dealt with. 
 
68 ELECTRICAL TRACTION. 
 
 The Wurts non- Arcing lightning arrester is very simple 
 and effective. The discharge is caused to pass between 
 two brass electrodes, separated by half an inch, over 
 narrow grooves burned in a block of lignum vitae. This 
 differs from the old telegraph arrester, which consists of 
 a series of opposing points on two metal blocks with a 
 short air-space between them, in that the discharge leaps 
 from block to block over the charred surface, and the 
 conducting vapours, which are necessary to the formation 
 and maintenance of a dynamo arc, are suppressed by the 
 cover which fits tightly over the metal electrodes. The 
 well-known Wurts arrester, with metal cylinders, is used 
 in the case of alternate current systems. 
 
 Sag in Wires. "We see from the Board of Trade Eegu- 
 lations (see Appendix, p. 233 ; Regulation XL) that the 
 distance between the ground and any point of the con- 
 ductor shall be at least 17 feet, and that the distance 
 between the supports is not to exceed 120 feet. This 
 naturally introduces the subject of sag in wires, for the 
 minimum distance between the earth and the line will be 
 midway between the supports, if these be horizontal, at 
 the highest temperature. The question of sag in wires 
 is an important one, and has been worked out from theo- 
 retical considerations, which have been proved to be all that 
 is required in practice. The true form of the curve is the 
 catenary, but it is usual in calculating tensions for small sag, 
 to suppose it a parabola without introducing serious error. 
 
 The overhead line has of course to encounter very varying 
 atmospheric conditions, such as strong winds, temperature 
 variations, and the deposit of snow and ice. Wind-pressure 
 brings great stresses to bear upon the insulators and poles. 
 Variation of temperature produces changes in the tensile 
 stress of the conductor itself, thereby varying the sag on 
 account of expansion and contraction, whilst snow and ice 
 add to the weight per unit length, and therefore to the 
 stress on insulators, etc. It is important for a given safe 
 
SAG IN OVERHEAD CONDUCTORS. 69 
 
 stress in the conductor at insulator, to see that the required 
 dip or sag has a given value at a certain temperature. 
 For instance, if the line be put up in summer when the 
 temperature is high, the sag must be such that in winter, 
 or at the lowest temperature, the conductor will not be 
 overstrained. From this we see that the overground distance 
 does not give the actual length to the conductor. Allowance 
 must be made for sag. If S be the actual length of wire 
 between the two supports in feet, T the stress in the con- 
 ductor at the insulator in pounds per square inch, BS the 
 breaking stress in pounds per square inch, D the dip or 
 sag in feet, that is, it is the difference between the heights 
 of the conductor at the insulator and the lowest point, 
 A the distance between the poles in feet, and iv the weight 
 in pounds per foot of the conductor, then we have the 
 
 A?ic 8D 2 
 
 following equations : D = -^ ? s = ^ + ^ an d, as ~ 
 
 T>Q 
 
 suming a factor of safety of 4, T = r . 
 
 In a given case assume that A = 120 feet, D = 1^ feet 
 at its maximum, BS = 60,000 Ibs. per square inch, which 
 is a very fair figure for hard drawn copper. Take 0'4 inch 
 as the diameter of the trolley wire, that is 0*126 square 
 inch section. Since a cubic inch of copper weighs 0-32 lb., 
 the weight of 12 inches of wire = 3 '84 times the cross- 
 sectional area in square inches ; hence the weight per 
 foot = 3'84 x 0-126 = 0'484 lb. ; then T = 581 ; this 
 corresponds to 4,600 Ibs. per square inch. 
 
 We have still to take into account wind pressure, for 
 when this is severe the resultant force will be that due 
 to gravity and such wind pressure. A very high wind 
 exerts a pressure of perhaps 30 or 40 Ibs. per square foot ; 
 so that the force due to wind, say per foot length of the 
 conductor, will be some function of this pressure and the 
 diameter of the wire. The surface of the wire is cylindrical 
 instead of flat, and, therefore, a correction is due for this. 
 
70 ELECTRICAL TRACTION. 
 
 Take it that the pressure on the wire is 0'6 of what it 
 would be if the surface exposed to the wind were flat. 
 Then the maximum force due to wind pressure per foot 
 
 length = O'G X X the maximum pressure per square foot 
 
 due to wind pressure ; where y is the diameter of the wire 
 in inches. This is equal to 0'8 lb. in the case we are 
 considering, as against 0*484 lb. due to gravity. The 
 resultant due to these two causes combined is equal to 
 \/0*G4 4- 0*23 = 0*94 lb. Inserting this in equation 
 
 T = ^, for w, instead of 0*484 we have double the 
 8JJ 
 
 tension in the wire at the insulator, or, say, 1200 Ibs., 
 which gives a factor of safety of about G. 
 
 We have next to consider the contraction of the wire 
 when at its lowesb temperature. This will bring greater 
 tensile stress to bear upon it, since it diminishes the dip D. 
 Take the co-efficient of expansion of copper at 0*00001 per 
 degree Fahrenheit, and allow for a fall of 30 Fahr., say on 
 50 Fahr., the temperature at which the wire is to be 
 erected with 1^ foot dip. The contraction on 120 feet is 
 0*030 foot. The length of wire at 50 Fahr. is 120*05 feet ; 
 then 120*014 is the length of the wire at 20 Fahr., the lowest 
 assumed temperature. D = / / ?>A(S - A) = (y ^ foofc> 
 
 T = 2000 Ibs. if we assume w = 0*94 ; the factor of safety 
 = 3'G. Similarly one could find the dip and stress in a 
 conductor at any other temperature. 
 
 The Supporting Pole for Overhead Conductor. The 
 pole has to support two wires on a double track, and has 
 also to withstand wind pressure on its own surface. The 
 test which sample poles, chosen from a batch by the 
 engineer, have to pass is severe, and proves whether the pole 
 will telescope or loosen at the joints when let fall end fore- 
 most three times from a distance of six feet, and whether 
 
POLES FOR OVERHEAD CONDUCTORS, 71 
 
 it will not take a permanent set when fixed horizontally at 
 one end, and a weight is applied at or near the other end. 
 
 Two grades of poles, manufactured by Messrs. Russell, are 
 used at Leeds ; each has an overall length of 31 feet made 
 up as follows. The smaller grade is made up of two 
 sections, the bottom section consisting of 7-inch pipe 17 
 feet long and \ inch thick, and the top section consisting 
 of 6-inch pipe tapering to 4^ inches, 14 feet long and f inch 
 thick, the weight of the pole being about 020 Ibs. The 
 larger grade is also made up of two sections, the bottom 
 section being 8-inch pipe 17 feet long and f inch thick, 
 and the top section tapering from 7 inches to 4f inches, 
 14 feet long and f inch thick ; its weight being about 
 1300 Ibs. 
 
 The sections of each pole are shrunk together hot under 
 pressure so as to make perfect joints, with 18-inch insertion 
 for each joint, which is then covered by a cast-iron ring. 
 
 The test for these poles provides that the pole shall be 
 fixed horizontally, one support being afe the butt end and 
 the other six feet therefrom. The small pole is to be 
 loaded with 700 Ibs. at a point 18 inches from the top 
 end, and shall not be deflected more than six inches at the 
 extreme end. A weight of 1200 Ibs. shall then be applied 
 at the same position, and the permanent deflection shall not 
 exceed half an inch. The large size pole is to be similarly 
 tested, the weights being 1000 and 1700 Ibs. respectively, 
 the temporary and permanent deflections not to exceed six 
 inches and half an inch respectively. 
 
 Fig. 29 shows one of these poles, which is also arranged 
 for carrying a Brockie-Pell 10-arnpere arc lamp. There are 
 40 of these arc lamps attached to the poles at Leeds ; they 
 are arranged in four circuits of 10 in series, and are 
 supplied from the tramway feeder at a potential of from 
 480 to 500 volts, one end of each circuit being connected 
 to the rail return. These arc lamps are in the centre of 
 the town, and the circuits are so arranged that, in the event 
 
FIG. 29. 
 
THE OVERGROUND CONDUCTOR. 73 
 
 of one giving out, -alternate lamps will still be left burning. 
 A single trolley wire as described, with insulators, would 
 cost roughly 200 per mile, and the poles for two wires 
 would cost roughly 1200 per mile. In the case of span 
 wires the supports must be such as to allow for their sag, 
 since the conductor at the insulator must be at the required 
 height above the ground in order that its lowest point is 
 not less than 17 feet from the ground. 
 
 Overground Conductor. Probably the first method em- 
 ployed to convey energy electrically to cars was a steel 
 conductor near the surface of the ground, supported at 
 intervals on insulators of glass or porcelain. Sir William 
 Siemens read a paper on " Electrical Transmission and 
 Storage of Power " before the Institution of Civil Engineers 
 in 1883, in which he mentions such a method being used 
 in Berlin by his brother in 1879, the line being 900 yards 
 long and the gauge 2 feet. He also described the Port- 
 rush line which he was then constructing, in which a 
 similar device was used. In this case it consists of a T 
 girder placed at the side of the rails, against which rub the 
 conductor or conductors fixed to the car and attached to 
 one pole of the motor on the car, the other pole being con- 
 nected to the rail which forms the return. Table V. on next 
 page gives particulars of the early lines in Great Britain, and 
 is taken from " Electric Railways," by Dr. E. Hopkinson,* 
 It shows the early use of this form of conductor. 
 
 Notable instances of the use of a channel steel conductor 
 are the City and South London Railway and the Liverpool 
 Overhead Railway. A requirement of such a conductor is that 
 it must be strong in addition to its conducting properties. 
 Table VI. is taken from a paper by Dr. J. Hopkinson on the 
 " Magnetization of Iron," f for the purpose of showing- 
 how the electrical conductivity of iron and steel varies 
 according to the composition. 
 
 * See Proceedings Institution of Civil Engineers, 1893, vol. cxii. p. 230. 
 
 t See Transactions Royal Society, 1885. 
 
ELECTRICAL TRACTION. 
 
 TABLE V. 
 
 Date of 
 opening. 
 
 j 
 
 * \ 
 System. 
 
 
 1883 
 
 (Portrush & Giant's! ( 
 \ Causeway .../! 
 
 3 Side conductor ... 
 
 Water power 
 
 1883 
 
 Brighton Beach ... 1 ; Rails Gas engine 
 
 1885 
 
 ( Bessbrook & Newry j o 
 \ Tramway ...j 
 
 J Central conductor 
 
 Water power 
 
 1886 
 
 RydePier 
 
 \ i Side conductor ... 
 
 Gas engine 
 
 1886 
 
 Blackpool ... i ' 
 
 2 Conduit ' Steam 
 
 1889 
 
 Carstairs 1 
 
 J Side conductors ... Water power 
 
 1890 
 
 Birmingham ... I 
 
 3 Accumulator 
 
 
 
 1891 
 
 Southend Pier . . . i J 
 
 I Central conductor 
 
 Steam 
 
 1891 
 
 Guernsey 2 
 
 $ Overhead conductor 
 
 1891 
 
 Leeds 3} 
 
 1893 
 
 South Staffordshire : 7i 
 
 TABLE VI. 
 
 
 a 
 
 
 
 
 
 
 
 G 
 
 
 
 1 
 
 
 
 
 g 
 
 
 si 
 
 Material. Temper. 
 
 
 
 - 
 
 
 
 O G ^3 
 
 3 
 
 f 
 
 .- 
 
 
 
 "5 '| 
 
 
 
 H 
 
 
 CO | GO 
 
 
 
 
 Cit ^ o 
 
 Wrought iron 
 
 Annealed 
 
 
 
 - i - 
 
 
 
 13-78 
 
 Bessemer mild steel 
 
 
 
 0-045 
 
 0-2 
 
 03 
 
 none 
 
 0-040 
 
 _ 
 
 10-5 
 
 Whitworth mild steel 
 
 Annealed 
 
 0-090 
 
 0-153 
 
 0-016 
 
 none 
 
 0-042 
 
 
 
 10-8 
 
 M 
 
 M 
 
 0-890 
 
 0-165 
 
 0-005 0-031 
 
 0-019 
 
 
 
 15-59 
 
 J >! 
 
 Oil hardened 0-890 
 
 165 0-005 0-081 
 
 0-019 
 
 
 
 16-95 
 
 Maneanese steel 
 
 As forged 0-674 
 
 4-73 i 0-023 0-608 
 
 0-078 
 
 
 
 53-68 
 
 i 
 
 Annealed i 0-674 
 
 
 
 
 
 . 
 
 39-28 
 
 
 Oil hardened 
 
 0-674 
 
 
 
 ( 
 
 
 
 
 
 55-56 
 
 Chrome steel 
 
 As forged 
 
 0-687 
 
 0-028 0-020 0-134 
 
 0-043 
 
 j Chromium 1 
 1 1-195 } 
 
 17-91 
 
 ,, ,, Annealed 
 
 0-687 
 
 ._ j 
 
 
 
 18-49 
 
 Oil hardened 
 
 0-687 
 
 ._ . . 
 
 
 
 30-35 
 
 Tungsten steel... 
 
 As forged 
 
 1-357 
 
 0-036 
 
 none 0-043 
 
 0-047 
 
 / Tungsten \ 
 \ 4-649 } 
 
 22-49 
 
 ( 
 
 Annealed 
 
 1-357 
 
 
 
 
 _ 
 
 22-50 
 
 , 
 
 /Hardened imi * 
 ( cold water } l 357 
 
 
 
 
 
 
 
 
 
 22-74 
 
 Grey cast-iron 
 
 345 
 
 0-173 
 
 0-042 
 
 2-04 
 
 0-151 
 
 (Graphitic ) 
 < carbon J 
 
 114-0 
 
 
 
 
 
 
 
 \ 2-06 
 
 
 Spiegel eisen 
 
 4-51 
 
 7-97 
 
 trace 
 
 0-502 0-128 
 
 
 105-2 
 
THE OVERGROUND CONDUCTOR. 75 
 
 The great variation of the conductivity shows how impor- 
 tant it is to have regard to the composition of the material 
 to be employed, but first cost must also be borne in mind. 
 The weight per yard is, on the assumption of equal specific 
 gravity, the quantity which determines the sectional area, 
 and also the cost for a given form or shape of section. 
 Other things being equal it is of course best to choose that 
 material which has the highest conductivity. 
 
 The specific resistance of copper is about re X 10 ~ 6 ohm. 
 Compare this with those in Table VI. Bessemer and 
 Whitworth mild steel, the latter annealed, are the best, 
 with a specific resistance of 10'5 and 10'8 X 10 ~ 6 ohms 
 respectively. We see that increasing the carbon from 0*09 
 to 0*89 in the Whitworth steel increases the resistance 50 
 per cent. ; whereas oil-hardening the latter specimen does 
 not make so much difference an increase, in fact, from 15*6 
 to 16*9. Of the other steels given chrome is the best, with 
 17*9 as forged and 18 '5 annealed. Wrought iron is better 
 than any of these, with 13'8. The worst of the irons or 
 steels given is undoubtedly manganese steel, with 4'73 per 
 cent, of manganese ; for even annealed its specific resistance 
 is 39*3. This subject has been discussed by Dr. E. Hop- 
 kinson before the Institution of Civil Engineers, 1887, and 
 the steel used by him on the Bessbrook and Newry Tram- 
 way was manufactured by the Darlington Steel and Iron 
 Co., and is stated to have the following composition : 
 carbon 0'09, silicon 0'02, manganese 0'63 per cent., and 
 the specific resistance is 12'1 X 10 ~ 6 ohms. The weight 
 per foDt of the conductor is 4*33 Ibs., and the cross- 
 sectional area T367 square inch. The cost delivered at 
 the wharf at Newry was 7 10-9. per ton, whilst copper at 
 the time was 84 per ton. This shows in favour of steel, 
 when taking cost and conductivity into account for a given 
 
 weight, in the ratio of ^ x 1? = |. 
 
 i Jij JL u 
 
 Let w = the weight in pounds per cubic inch of the 
 
76 ELECTRICAL TRACTION. 
 
 steel employed = 0*28, say ; let A = the cross-sectional 
 area in square inches ; and let p = the specific resistance in 
 ohms. Then 30 x A X w = the weight in pounds per 
 
 1 rm "HPT* Vf~LT*(j 
 
 yard. Therefore A = --' F J . Then the resistance per 
 
 O U /\ / ' 
 
 P X 91-44 X 1760 X 36 X w 
 mile of such a conductor = - lbs . per yard x 6 . 45 
 
 or, taking w at 0'28, and p at 12 x 10 ~ 6 , which is an 
 average value, the resistance per mile = rr ; -r ohms. 
 
 This is, perhaps, a useful way of expressing the relation ; 
 and of course, when possible, one should always get the 
 specific resistance of the material to be employed. We 
 have taken herein no account of joints,* which, as we shall 
 see, are of very great importance in traction work. Further, 
 the two conductors in a double track are probably cross- 
 bonded, so that, in using the above formula, one must divide 
 again by 2 to get the resistance per mile of the two con- 
 ductors in parallel ; or, in using this formula for tramway 
 rails, one should divide by 2 for single and 4 for double track. 
 
 By eliminating silicon and slightly diminishing the 
 carbon in the mild steel used on the City and South London 
 Eailway, although the manganese was somewhat greater, 
 a specific resistance of 10'5 X 10 ~ 6 ohms was obtained. t 
 The conductor (Fig. 2, p. 18) is supported on glass insulators, 
 and weighs 10 lbs. per lineal yard. The shoe for collecting 
 energy is shown in Figs. 1 and 2. The leakage of the 
 electric system, consisting of dynamos, feeders, and working 
 conductor, is stated to be generally about one-half of an 
 ampere, and rarely exceeds one ampere when tested at 500 
 volts. If we assume this to be all in the conductor and 
 equally distributed along its length, since the track has a 
 length of say 3'1 miles, the resistance per mile is 3120 ohms. 
 
 Fig. 30 gives a view of the insulator and conductor 
 
 * For Electrical Resistance of Joints, see p. :84. 
 
 t See Journal of the Institution of Civil Engineers, vol. cxii. p. 215. 
 
THE OVERGROUND CONDUCTOR. 
 
 77 
 
 used on the Liverpool Overhead Railway, and has been 
 taken from List. C.E., vol. cxvii. pt. iii. The conductor is 
 of steel, rolled in lengths of 32-| feet, has a cross-sectional 
 area of 4 square inches, and weighs 40 Ibs. per lineal 
 yard. It is supported every 1\ feet (except at the joints, 
 where the insulators are 2-| feet apart) on porcelain in- 
 sulators designed for oil ; but this was not used, inasmuch 
 as the leakage was considered low enough without the oil. 
 
 FIG. 30. 
 
 There are no feeders, and the return circuit is through the 
 rails. These latter weigh 56 Ibs. per yard. 
 
 This figure also shows how the conductor is arranged at a 
 crossing. It is broken and bent parallel to the main rails, 
 and since it is three-quarters of an inch above these, the 
 iron shoe attached to the car can bridge across, and so clear 
 the rail return. Neither the composition nor the specific 
 resistance of this steel conductor is given. If it is mild 
 steel, it might have a resistance per mile, neglecting joints, 
 
ELECTRICAL TRACTION. 
 
 of 0-075 ohm, or one-half of this 
 for the two conductors in parallel. 
 The leakage current is stated to be 
 two or three amperes. Take it at 
 2'5 ; take the length of the track 
 at six miles ; then the resistance per 
 mile = 1200 ohms for the double 
 track. This and the 3120 ohms per 
 mile of the City and South London 
 Railway are very low as compared 
 with the 50,000 ohms per mile laid 
 down by the Board of Trade for 
 tramways, and the minimum of 1000 
 ohms per mile, for which the cars 
 may be stopped running (see Regu- 
 lation X., Appendix, p. 228). But 
 it must be remembered that there is 
 no traffic over these lines other than 
 the trains. 
 
 On the Chicago elevated railway, 
 where there are 12 miles in opera- 
 tion, the girder rail is used for the 
 conductor, having 48 Ibs. per yard, 
 and a conductivity equal to a copper 
 rod one inch in diameter. The area 
 of this conductor might be 4'76 
 square inches ; that is, the specific 
 resistance - 9' 7 X 10 ~ 6 ohms, which 
 would indicate a very good quality of 
 steel. Steel rails are used for feeders 
 on this railway, and Fig. 31 gives a 
 section of the roadway, showing the 
 rails, the insulated overground con- 
 ductor, and the steel feeders, which 
 are placed in the centre of the way 
 and protected as shown. 
 
COLLECTORS FOR OVERGROUND CONDUCTOR. 79 
 
 Fig. 32 illustrates the shoe used to collect energy by 
 rubbing against the insulated rail conductor. It will be 
 
 FIG. 32. From Engineering. 
 
 seen from the figure that the shoe is free to move up or 
 down, and overhangs the car. 
 
8o ELECTRICAL TRACTION. 
 
 CHAPTER IV. 
 
 THE TltACK AND ROLLING-STOCK. 
 
 General Remarks. The history and development of the 
 construction of rails for tramways is well set forth in 
 D. K. Clark's " Tramways," to which the reader is referred 
 for this and other valuable information. At the present 
 time, for electrical tramway work the girder rail is almost 
 universally used, its weight per yard depending, in the first 
 instance, upon the strength required and other mechanical 
 considerations ; and, in the second place, upon its electrical 
 properties, since in electrical tramways the rail is also used 
 as a conductor. The engineer has, therefore, to look at this 
 matter from another point of view. Fig. 33 gives a re- 
 presentative section of roadway now met with in tramway 
 
 
 FIG. 33. 
 
 work. The rails for each track are connected by tie-bars, 
 and generally laid on concrete, granite setts being laid in 
 between and on each side of the track, the rest of the 
 surface being the ordinary roadway material. The spaces 
 between the setts and the rails are generally filled up with 
 asphalte. Wood blocks are also used instead of granite. 
 Irrespective of any electrical considerations, the price per 
 mile of an ordinary tramway line is of interest. According 
 
THE TRACK, 81 
 
 to Clark, the following are examples of the cost per mile 
 of single track, without the bonding necessary when the rails 
 are used as electrical conductors : 
 
 
 Vignoles rails, 42 Ibs. per yard ... ... 1823 
 
 Paying for same ... ... ... ... 2275 
 
 Total ... ... 4098 
 
 Kincaird girder rails (Fig. 33), 78 Ibs. per yard 1831 
 Paving for same ... ... ... ... 1641 
 
 Total 3472 
 
 At Liverpool in 1892, where there were 47-| miles of 
 tramway on the Lever System, the cost of maintenance of 
 the ten years up to 1892 is under one per cent, per annum, 
 on a basis of 6000 per mile, which price is for a first- 
 class permanent way on concrete foundations, into which 
 the rails are secured. The wear for 100,000 tons of traffic 
 per annum is stated as being about 0'015 inch, or 0*0086 
 of the original weight, which was 40 Ibs. per yard. 
 
 Gauge. The gauge on tramways varies very much. 
 Where plenty of width of roadway is available, it is gene- 
 rally 4 feet 8|- inches, but varies from 2^- feet to 5 feet. At 
 Dublin the gauge is 5 feet 2^- inches, concrete bed, four 
 tie-bars per rail. At Leeds the gauge is 4 feet 8-| inches 
 concrete bed, with granite setts and asphalte. At Walsall 
 the gauge is 3 feet 6 inches. On the Continent the metre 
 gauge is very popular. The subject of gauge has especial 
 interest at the present time on account of the activity in 
 the construction of light railways consequent upon the 
 passing of the Light Railways Act, 1896 (see p. 5). The 
 standard gauge of 4. feet 8 inches should be adopted where 
 practicable, and when a " light railway " is intended to feed 
 into a main line. At the same time a narrower gauge 
 would seem imperative in the case of narrow roads, and to 
 save expense when taking the line across fields. Few roads 
 
82 ELECTRICAL TRACTION. 
 
 can spare the width of way necessary to take the standard 
 gauge, and the first cost of a tramway capable of carrying 
 loaded cars as on a railway would often be prohibitive. 
 It is, nevertheless, important that some uniform narrow 
 gauge should be adopted, such, for instance, as 2 feet 
 6 inches. The weight of rail per yard varies within wide 
 limits from 40 or 50 Ibs. per yard to over 100. The 
 tendency now is towards heavy sections. 
 
 Gradients. Gradients are in many instances severe ; 
 for example, at Montreal 1 in 8 and 1 in 10 are common. 
 These have an influence upon the power supplied, unless 
 means be provided for returning the energy to the system 
 from the car when descending (see p. 136). Some of the 
 gradients in Bristol are 1 in 15, whilst in Dublin 1 in 20 is 
 met with. Then, again, the curves are often sharp ; and 
 this introduces a greater tractive resistance. 
 
 The Rail as a Return Conductor. In Great Britain the 
 engineer, when considering the line as a return conductor, 
 must comply with certain regulations of the Board of Trade. 
 See Appendix, p. 226, regulations 3 to 8 inclusive, under 
 date February 11, 1897. 
 
 Eegulation 7 settles to some extent the size of rail to 
 be put down, since it gives the limit as to drop of potential 
 difference. Assume, for instance, that the rail weighs 
 100 Ibs. per yard, which is a comparatively heavy section. 
 Then it might, for a double track, when acting as an 
 electrical conductor, and if properly jointed, have a resist- 
 ance of 0'0075 ohm per mile. Suppose the line were 10 
 miles long, and the average current 200 amperes, then 
 clearly the Board of Trade Regulation would be exceeded. 
 In order to comply with this regulation, it is clear that, for a 
 given system and each one must be treated individually 
 the cross-sectional area, or weight in pounds per yard for a 
 given quality, is an important quantity ; for if this be too 
 small, another conductor must be laid in parallel with the 
 track, and connected with it at intervals not exceeding 100 
 
THE CHICAGO RAIL BOND. 
 
 feefc if such conductor be uninsulated. See Regulation 
 4, Appendix, p. 226. For resistance of rails, see p. 7G. 
 
 We have already discussed the influence of feeders on the 
 distribution of currents in the rails (pp. 53, 54). If the 
 line is used as a return, it must be preserved as a good con- 
 tinuous conductor. The ordinary tramway track, although 
 perfectly sound at the joints mechanically, becomes at 
 
 these points a bad conductor in time, owing to rust. Many 
 devices for getting over this difficulty have been adopted. 
 One of these, and perhaps the oldest, is to bridge across the 
 joints with a copper conductor. At the present time the 
 Chicago rail bond: (Fig. 34) is very largely used. It con- 
 sists of a copper conductor C which is bent at each end, the 
 ends A being expanded into freshly drilled holes in the web 
 
84 ELECTRICAL TRACTION. 
 
 a of each rail by means of a driffc-pin B driven into a hole 
 in the end of the bond. Such a device has been well tried, 
 and is very much used at the present time.* As many as 
 two, or even three, such bonds are used per joint, according 
 to the size of the rail. This is a costly operation, and forms 
 a fairly large item in the cost of a tramway line. It may 
 cost 300 per mile of double track with two bonds per joint. 
 The great cause of difficulty is, no doubt, the up and down 
 motion of the rail joints when traversed by cars. This 
 works the bonds loose, and bad connections may ensue. 
 There are various other forms of rail bond besides that just 
 mentioned. The Edison- Brown plastic rail bond has 
 much to recommend it, if after long trial it proves efficient. 
 This device (Fig. 85) makes use of the fish-plate itself as 
 the intermediate conductor. Between the web of each of 
 the two rails to be jointed and the fish-plate is placed an 
 amalgam, the contact surfaces being thoroughly cleaned 
 and treated with a solid alloy, which silvers them and keeps 
 them free from rust. An elastic cork, with a hole through 
 it, is placed against the web of the rail, and supports the 
 amalgam before placing and fixing the fish plate in 
 position. The current then passes from one rail through 
 the amalgam, through the fish-plate, through the other 
 piece of amalgam to the next rail. It is claimed that 
 in the case of a 9-inch rail, the bond will carry 1500 
 amperes with a drop of potential of 0'03 volt. Another 
 method of applying the amalgam is to drill a hole through 
 the top of the rail into the fish-plate, or through the bottom 
 of the fish-plate into the bottom of the rail, in either case 
 filling with amalgam and sealing up with a metal plug. 
 There is another method which, if it gives a perfectly good 
 result, electrically speaking, will be of very great value. It 
 is to cast- weld the joints, thus making a continuous rail. 
 Ordinary welded joints, butt to butt, have been tried on 
 
 * The electrical resistance of a newly made joint is stated to be from 
 0-000017 to 0-000034 ohm. C. Billberg, Electrician, Sept. 24, 1897. 
 
THE EDISON-BROWN RAIL BOND. 
 
86 ELECTRICAL TRACTION. 
 
 \ miles at St. Louis, 5 miles at Cleveland, and 32 miles at 
 Brooklyn, the welding being carried out by electricity. St. 
 Louis had o miles of cast-welded joints ; and in Chicago 
 there are, it is reported, about 12,000 such joints. Molten 
 metal is poured into a mould round the end of the rails, and 
 thus makes a bond, joining them. It seems that sometimes 
 a sliding action sets in on expansion and contraction, which, 
 from an electrical standpoint, is bad. This process is carried 
 out in cold weather preferably, so that when hot weather 
 steps in the rails expand and do not so much tend to sever 
 the joints. Only time and experience will sift out the best 
 from the many types. 
 
 Bonding in Great Britain is of very great moment, on 
 account of the Board of Trade regulations (see Appendix) ; 
 but with proper care there is no reason why these should 
 not be more than complied with. In America the matter 
 is different, since the regulations are not so stringent. At 
 the same time bonding is necessary from the point of view 
 of economy. Since the rules in Great Britain are so severe, 
 it follows that the bonding of rails has been brought more 
 prominently to the front, and has exercised, and will exercise, 
 a great influence upon the question of feeders. 
 
 Rolling Stock. The whole subject of rolling stock is 
 of great importance, and cannot be dealt with exhaustively 
 in a work of this magnitude. We illustrate in Fig. 36 
 a Leeds motor-car, which may be taken to represent the 
 ordinary tramcar with overhead seats. Twenty-five such 
 cars are supplied at Leeds, each capable of seating 50- 
 passengers. The car bodies are mounted upon Peckham 
 standard cantilever extension trucks, and were constructed 
 for the contractors, Messrs. Greenwood & Batley, of Leeds, 
 by Messrs. G. F. Milnes & Co. The length of wheel base 
 is G feet. Each car is supplied with two motors, having 
 single reduction gear, and operated on the series-parallel 
 control system. The cars are supplied with two meters 
 one for recording the ampere-hours, the other for recording 
 
88 ELECTRICAL TRACTION. 
 
 the maximum current. Each car is wired for 10 16 C.P. 
 lamps, connected five in series between the trolley wheel 
 and rail. The trolley is of the swivelling- arm type, fixed 
 to a standard on the car roof, the springs being such that a 
 pressure of about 28 Ibs. is exerted between the trolley 
 wheel and wire. 
 
 Leeds Tramcar Test. The test imposed upon the con- 
 tractors by Dr. J. Hopkinson, in connection with the Leeds 
 tramcars, is of interest. The test to which the motor-cars 
 must be subjected, and which they must pass, is that each 
 one shall stop for five seconds every five minutes, and shall 
 complete the round trip, loaded with three tons, in one hour 
 fifty minutes, with a consumption of not more than 28 
 ampere-hours, and that the current required shall never 
 exceed 50 amperes, the supply being maintained at a 
 potential of not less than 480 volts. 
 
 Taking the line to be about 7 miles long, corresponds 
 to an average velocity of about 1\ miles per hour, and an 
 expenditure of about one Board of Trade unit per mile. 
 
 The Board of Trade regulations with regard to tramcars 
 and trailer cars are dealt with generally on pages 5 and G. 
 They are given in full in the Appendix. 
 
 The discussion as to whether large cars are to be used, or 
 smaller ones with trailers, is a very broad one. But a very 
 important thing to note is the Board of Trade regulations 
 with regard to attendants on cars and trailers. In Regula- 
 tion III., relating to the Dublin Southern District Tramway 
 (see Appendix, p. 231), it is stated that each trailing car 
 must carry a brake-man as well as a conductor. This rule 
 is of great importance apart from the fact that this brake- 
 man may or may not apply his brake in keeping with the 
 motor-man on the leading car. For particulars of the Dublin 
 cars, see p. 189. 
 
 When it is considered that the cost of coal, oil, water, 
 and wages at the generating station, other than wages for 
 repairs, may be less than a halfpenny per car mile run, 
 
ROLLING-STOCK. 89 
 
 whereas the cost of wages of car-driver and conductor may 
 be more than twopence per car mile run, one can under- 
 stand how important it is to have as few conductors and 
 drivers as possible. According to the above regulation, we 
 see that two men are required on the trailer as well as on 
 the motor-car, and this already large expense is doubled 
 if we add the trailer. On the other hand, such an arrange- 
 ment of cars and trailers makes the system much more 
 flexible, since the trailers need only be used when the traffic 
 is great. At the same time the motors must be able to cope 
 with both loads. The long cars take more time to empty, 
 and do not give so much flexibility to the system. We find 
 trailers at Dublin and on many other lines. 
 
 The discussion as to whether locomotives and coaches, as 
 on the South London Railway, or motor-cars, as on the 
 Liverpool Overhead Railway, should be used depends upon 
 the size of the trains and conditions as to head room, 
 and the engineer himself. In subways where room is 
 limited, the locomotive would seem to be the better solution ; 
 on the other hand, with long trains, if each car has its 
 own motors, and all the motors have to be controlled by 
 one conductor, the switching arrangement becomes com- 
 plicated. The Chicago Elevated Railway has motor-cars 
 similar to the Liverpool Overhead Railway. This line 
 resembles the elevated portions of London railways, and was 
 opened in 1895. The cars are 40 feet long, and mounted on 
 steel framing with bogies at each end. There are 55 motor- 
 coaches and 100 trail coaches, the proposed length of line 
 when complete being 18 miles, of which 12 are in opera- 
 tion. The cross-section of the roadway is given in Fig. 29.* 
 
 The under-truck and bogie are structures upon which 
 the car body itself is mounted, and which have received 
 careful attention from the makers. Such structures must 
 be properly braced to resist deformation ; they must be 
 strong and light, and carry the car body with a minimum 
 * See Engineering, January 10, 1896. 
 
90 ELECTRICAL TRACTION. 
 
 of vibration. The axle-boxes must be dust-proof and self- 
 lubricating, and the wheels fitted with an adjustable brake, 
 the shoes for which must be easily got at for removal. 
 The construction of trucks has been carried to great per- 
 fection, and there are several well-known builders. A 
 Peckham truck is illustrated in Fig. 53. Fig. C7 shows a 
 
 FIG. 37. From Engineering. 
 
 single motor mounted on an under-truck for an ordinary 
 tramcar. 
 
 The suspension of the motor on the truck has been 
 well thought out. Fig. 37 gives a plan of a General Electric 
 800 motor. One end of the motor is supported by the 
 car axle itself, which works in bearings formed by extensions 
 of the shell. The other end of the motor is usually carried 
 by a lug cast on the shell of the motor, such lug fitting 
 
THE UtiDER-TRUCK AND &RAKES. 91 
 
 into a hole in a transverse bar, supported from the under- 
 truck by means of springs. Another method is to support 
 this end of the motor from two lugs carried by longitudinal 
 side bars also supported by springs on each side of the 
 motor. The former method is more used, and is well 
 shown in Fig. 37. This plan shows the gear-case, which 
 protects the spur-wheel and pinion from dust and dirt, and 
 gives details of the cast-steel shell of the motor. 
 
 Brakes. The subject of brakes has been touched upon 
 at p. G. It may be mentioned that Mr. Baylor, in his 
 recent paper before the Institution of Electrical Engineers,* 
 describes and illustrates a very ingenious brake, which utilizes 
 the motor when converted into a generator and driven as 
 such by the momentum of the car. It does not bring 
 excessive stress to bear upon the armature, since the arma- 
 ture conductors need only carry their normal current and 
 operate the brake, which may be arranged to be very 
 powerful. By means of an inter- connection between motor- 
 car and trailer, such a brake could be operated by the 
 motor-man on the motor-car, thus dispensing with the 
 brake -man on trailer, and making the operation much 
 more secure, -since the motor-man can see at once what to 
 do, and time is lost if he has to signal to the man behind. 
 
 A usual provision is to short-circuit the motor armature 
 in extreme cases, but this brings very severe stresses to bear 
 upon the armature and gearing. 
 
 Brakes are also worked by compressed air, as on the 
 City and South London Railway. 
 
 The Board of Trade regulations with regard to brakes 
 are given in !., II., Appendix, pp. 231, 23%. 
 
 Tractive Resistance on Tramway Lines. A great many 
 experiments have been made to determine the tractive 
 resistance in, say, pounds per ton of the gross load on the 
 driving-wheels of tramway lines. The results, as one would 
 expect, depend very much upon the state of the rails. In 
 
 * See Proceedings, vol. xxvi. 
 
92 ELECTRICAL TRACTION. 
 
 dirty weather, when the rails get filled up with mud, a 
 greater resistance is introduced than in fine weather, when 
 the grooves may bs clear. Experiments made by Mr. Holt 
 at Leeds show a tractive resistance of about 22 Ibs. per 
 ton of gross load, at an average of 6 miles per hour on 
 a flat groove rail in fair condition. He also found that 
 the initial pull exerted might be from 100 to 1GO Ibs. per 
 ton. On the Nottingham tramways Mr. Perrett found that 
 the tractive force varies very much with the weather ; he 
 obtained on a straight and level track, when clear, 25 Ibs. 
 per ton ; and on a curve 22 feet radius, up a gradient of 
 1 in 130, the force rose to 94 Ibs. per ton. To start a 
 car on the level in dirty weather, he found a pull of 6G Ibs. 
 per ton necessary ; whereas on the curve and gradient 
 mentioned, this rose to 130 Ibs. per ton. 
 
 Reckenzaun states in the Electrician, May, 1800, that the 
 mean tractive force in tramway work may be anything from 
 30 to 50 Ibs. per ton ; it is seldom below 20 on level roads. 
 This, is very different to an ordinary railway, where 15 Ibs. 
 per ton is a fair figure to take. 
 
( 93 ) 
 
 CHAPTER V. 
 
 THE SLOTTED CONDUIT SYSTEM. 
 
 General Remarks. Probably the avoidance of an over- 
 head conductor, and the examples set by cable tramways, 
 led the early workers in electric traction to experiment 
 with what is generally termed the " slotted conduit system." 
 This consists of an underground conduit, generally of an 
 oval section, running the whole length of the line, with a 
 narrow opening on the surface to permit of a plough or 
 metal conductor passing down from the tram-car into the 
 conduit. The conductor or conductors from which electrical 
 energy has to be taken are supported on insulators fixed 
 to the walls of the conduit, and rubbing contact is made 
 between them and the collector. In this manner the car 
 can, as it travels along, receive energy in the form of 
 electricity, with the same facility that energy is received 
 from the overhead-conductor, only that now the live con- 
 ductor is under the surface of the roadway. It can be 
 easily understood that such a system as this is expensive 
 to construct, especially in towns or cities where there are 
 already gas and water pipes and electric-light mains. 
 Further than this, the whole system has to be properly 
 drained, since water would, if deep enough to come in 
 contact with the live conductor, cause very serious trouble ; 
 and provision must be made for cleaning out the conduit. 
 This is generally carried out by having sumps at intervals 
 along the line. The earliest attempts at electric traction 
 in streets where this system was tried, showed how 
 
94 ELECTRICAL TRACTION. 
 
 important it was to have the conduit of ample dimensions 
 and sufficient strength to keep the slot from closing in 
 owing to the severe stresses due to frost and ordinary 
 traffic. Holroyd Smith in England, and Bently-Knight in 
 America, were early in the field with slotted conduits, and 
 at the present time there are several examples in practice. 
 Crosby and Bell, in their second edition (189 G) of " Electric 
 Railways," state that "the Bently-Knight roads in this 
 country" (meaning America) "have now been totally 
 abandoned after long and costly experiments ; and there is 
 to-day no conduit road in operation in America, unless we 
 except an experimental one in Chicago." On the other 
 hand, Dawson, in his " Electric Railways and Tramways," 
 1897, p. 471, states, in connection with the "Love" 
 system, that "it was put into use at Washington on 
 March 3, 1893, and has been continually at work since ; " 
 so that authorities differ. But in any case there are 
 examples in England, America, and on the Continent, of 
 slotted conduits. This system has advantages and disadvan- 
 tages. Once the conduit is constructed, both the positive 
 and negative mains can be supported on either side, and in 
 this way electrolytic action on water and gas-pipes can be 
 avoided, since the rails need not be used as a return circuit. 
 There are examples in which the positive and negative 
 mains are supported on insulators in the conduit, and also 
 in which only the positive main is so carried, the rails 
 themselves serving as the return ; the former, of course, is 
 the better to adopt. Naturally the strength required in 
 the conductor demands a larger cross-sectional area than 
 one meets with in overhead wires, and this is an advantage 
 from an electrical point of view, since greater conductivity 
 is secured. 
 
 Electric Osmosis. The same effects due to electrical 
 osmosis observed in the ordinary electric lighting system 
 of conduits are also observed with this system ; that 
 is to say, if the insulation resistance be measured, the 
 
ELECTRIC OSMOSIS. 95 
 
 positive conductor is found to be more highly insulated 
 than the negative, and on reversing the poles this effect 
 is also reversed, which proves that it is not casual. A 
 simple experiment carried out in ths Siemens Laboratory, 
 King's College, London, under the direction of Dr. J. 
 Hopkinson, by Messrs. Highfield and Adlard, illustrates 
 this well-known effect. An ordinary insulator of porcelain, 
 as used in electric light conduits, with slots for the reception 
 of copper strip conductors, was placed in a tray of water, 
 the water just touching the bottom of the insulator. Two 
 bare copper conductors placed in the slots were connected 
 to the poles of a battery of 56 storage-cells, and the leakage 
 current between the positive and water and the negative 
 and \vater measured, means being provided to reverse the 
 poles of the battery when desired. The following figures 
 show the character of the results : 
 
 Time 10'55 a.m. 12-10 p.m. 1*0 p.m. 
 
 Positive leakage current 51 ... 14 ... 14 
 
 Negative leakage current 40 ... 75 ... 74 
 
 The currents are stated in arbitrary units. On reversing 
 the poles the effect was reversed. This effect may be 
 explained as follows. Suppose two vessels containing water 
 to communicate by a cotton wick, and two electrodes, 
 differing much in potential, to dip into the two vessels ; 
 then it is found that water passes over continuously from 
 the vessel into which the positive electrode dips to that 
 into which the negative electrode dips. The action of bare 
 conductors is the same upon a larger scale. The main 
 source for moisture is from the walls and floor of the 
 culvert ; this moisture will constantly creep over the porce- 
 lain insulators towards the negative conductor, but will 
 creep away from the positive conductor, drying the neigh- 
 bourhood of the latter and improving its insulation. 
 
 It is reported by Dawson* that at Washington the 
 insulation resistance of two miles of conduit varies from 
 * See " Electiic Railways and Tramways," p. 481. 
 
96 ELECTRICAL TRACTION. 
 
 8300 to 27,600 ohms on the positive side, and from 400 
 to 910 on the negative side, for varying states of weather, 
 which shows the same effect. 
 
 For the Board of Trade regulations with regard to 
 conduits, see Appendix, pp. 229, 230. 
 
 Objections. The great objection to the slotted conduit 
 is its cost, which must of course vary with the locality in 
 which it is placed. It is just in crowded thoroughfares, 
 where opposition to the overhead conductor is great, that 
 so many other interests have to be considered ; and this is 
 just the place where the slotted conduit system is likely to 
 be useful, for in the suburbs the overhead trolley can be 
 used without much opposition. Other matters which are 
 said to militate against the use of the slotted conduit are 
 the danger and inconvenience caused by the slot itself, and 
 the serious and prolonged interference with the streets both 
 during construction and repairs. 
 
 Descriptive Matter. The following short descriptions 
 of some of the more important conduit lines at present 
 in use will serve to show the different methods adopted. 
 
 Budapest. One of the most important examples of the 
 slotted conduit system is that of this city, which has a 
 population of 500,000. The streets are generally very 
 wide, and the trams run on either side, leaving the centre 
 free for ordinary traffic. There are a few steep gradients, 
 the steepest being about 1 in 18, but generally the streets are 
 flat. The paving is generally granite, but asphalte is used, 
 and the rails used are of a special section, each one being 
 practically a double girder of about 170 Ibs. per yard. 
 The gauge is throughout 4 feet 8-| inches, and when the 
 conversion of horse lines is finished there will be about 
 44 miles, as follows : 
 
 Length of line Miles of 
 
 in miks. track. 
 
 Electric underground ... 2'3 ... 4-7 
 
 Electric conduit ... 14'5 ... 30'0 
 
 Electric overhead 26*7 ... 54'0 
 
ELECTRICAL TRACTION. 
 
 The Budapest tramways are owned by two companies, 
 the Street Tramways Company and the Electrical Town 
 Line Company founded by Messrs. Siemens and Halske, who 
 
 laid the first electric line 
 in 1887. Storage-cell cars 
 have been tried, but were 
 not considered a success. 
 It is stated that all the 
 horse lines are to be con- 
 verted into electric within 
 the next four years. Figs. 
 38 and 39 give details of 
 the slotted conduit at Ber- 
 lin, and will serve to illus- 
 trate that of Budapest, 
 $ except that at Budapest 
 -j gjj both rails are of the double 
 z o girder type. The under- 
 f=< ground conduit is about 
 5 22 inches deep, and is 
 o under one rail, the slot 
 ~" between the rails varying 
 from l~ to 2 inches at 
 curves. At intervals of 
 66 to 130 yards there are 
 manholes, which give ac- 
 cess for cleaning and re- 
 pairing ; and there are two 
 handholes midway between 
 the manholes. The two con- 
 ductors, one on each side, 
 are carefully insulated, and 
 
 form the going and return circuits. There is, therefore, 
 no return by the rails. A collector attached to the car 
 pcT/sses down through the slots and rubs against the live 
 conductors. Fig. 40 gives the collector in plan and sectional 
 
 M S 
 
BUDAPEST AND BERLIN. 99 
 
 elevation.* It will be seen that the rubbing contacts, 
 , , are mounted on springs, S, S, arranged in such manner 
 that the blocks are pressed outwards, so as to make good and 
 efficient contact with the conductors, which are of L section. 
 The cost per mile varies from 4600 to a maximum of 
 6900, against which the overhead is placed at about 
 3000 when using the rails as a return. The maximum 
 voltage is 500. The cars are arranged to seat 50 persons 
 inside and out ; some are fitted with one and some with 
 two motors. The weights are : for a one-motor-car 6' 7 5 
 tons, and for a two-motor-car 8' 3 tons, each of which has 
 a daily mileage of 186. The fares vary from \\d. to M. 
 The maximum speed allowed is 15 miles per hour, but this 
 
 FIG. 40. 
 
 varies with the street traffic. The service is from 5 to 
 3-| minutes, and the tramway runs for 11 hours per diem. 
 It is reported that on exceptional occasions -the line has been 
 stopped by snow. 
 
 Berlin. Electric cars have been running in this city 
 since May, 1896, and when complete there will be more 
 than 150 miles of overhead conductors alone. The length 
 of overhead is 8 miles, and of conduit 5^ miles. Although 
 the overhead line is allowed in some streets in the city, the 
 police object to it in the centre, and to meet this the 
 company is adopting conduit. Meanwhile it has been 
 decided to run storage cell cars. There is conduit on 
 three routes, mostly constructed by Siemens and Halske, 
 
 * See H. Smith, Journal Northern Society of Electrical Engineers 
 1895. 
 
ioo ELECTRICAL TRACTION. 
 
 the Union Company having laid a short length. The cross 
 section of the roadway is shown in Fig. 38, and a longi- 
 tudinal section through the slot in Fig. 39. The combina- 
 tion of overhead and conduit seems to work well ; the 
 change from one to the other can be made while the car 
 is running and is scarcely perceptible. It is stated that 
 the cost per mile overhead is 2400 as against 6450 for 
 slotted conduit, not including concrete work. The voltage 
 is 500, and the power is supplied by another company at 
 about Id. per unit. There are about 50 motor-cars, each 
 capable of seating 34 passengers, weighing about 7 tons 
 empty, and 9f tons full. The fares vary from Id. for from 
 1^- to 2 miles, and %\d. for about 7 miles ; and the maximum 
 speed is 12-^ miles per hour. The service is from 4 to 10 
 minutes. The objections to the slotted conduit system 
 are its first cost, length of time occupied in construction 
 and repairs, and dirt and moisture in the conduit interfering 
 with the working. 
 
 New York Electric Conduit Tramways. In this system 
 the conduit is laid in the centre of the track, and contains 
 two conductors, one on each side of the conduit. These 
 form the going and return conductors of the system, so 
 that the rails are not used for electrical purposes. Fig. 41 
 gives details of the insulator for supporting the T conductor, 
 the other half of the conduit being the same as shown in 
 the figure. The general features of this conduit are very 
 good, and represent the best practice. The lower part of 
 the framework up to the line MN is one casting. Firmly 
 bolted to the frame is a cast-iron cap, G-, which supports 
 the insulator and working conductor. The insulator, P, is 
 of porcelain, and is cemented into the cap Gr by means of 
 concrete, 0, the distance piece, J, preserving a sufficient 
 space for the cement. The insulators are arranged at 
 intervals of 14 feet along the conduit, and a handhole 
 over each gives immediate access. Manholes are placed 
 every 200 feet along the line, having the dimensions 4 feet 
 
NEW YORK, 
 
 icr 
 
 FIG. 41. 
 
102 ELECTRICAL TRACTION. 
 
 10 inches deep, 4 feet long, and 17 feet 2 inches wide. 
 The working conductors are 6 ins. apart, and each one 
 weighs 21 Ibs. to the yard. The slot rails weigh 57 Ibs. 
 per yard, and the track rails 107. The latter are in 
 30 feet lengths and 9 inches deep. This system has been 
 in operation in New York city on Manhattan Island, 
 and has proved so satisfactory that the company have 
 decided to employ these designs on about 40 miles of 
 track. 
 
 Brussels. The earliest electric line in this city has 
 been in use since May, 1894. The overhead and conduit 
 systems are used, and are being equipped by the Union 
 Company of Berlin, under a guarantee as to the cost of the 
 working of the line. The guarantee was 2^. per car 
 per mile ; and as to the overhead lines the cost came 
 out at less than this, so that the company were relieved 
 of their responsibility before the term of guarantee had 
 expired. 
 
 With regard to the conduit line it is stated that the cost 
 was about 19,000 per mile of double line including every- 
 thing, and at this price it was handed over to the company 
 ready for service. In January, 1897, about seven miles of 
 conduit line were in course of construction. The cars and 
 trailers are arranged to carry from 82 to 34 passengers, but 
 frequently 60 or 70 persons are carried on the car and 
 trailer, with the result that the speed is reduced. On the 
 better class of line the cars are divided into first and second 
 class, the first class being provided with cushioned seats, and 
 in the second only bent wood is used. There are two motors 
 on each car, and these are usually 15 and 25 H.P. The 
 weight of the car is 6^ tons empty, and each car runs about 
 80 miles per day. The cars cost 670 each. At present 
 there are 28 motor-cars ; this number is to be increased to 
 100 during 1897. As is usual, the cars take up passengers 
 at any point on the line ; but definite stopping-places are 
 arranged from 200 to 300 yards apart, and only stop there 
 
BRUSSELS AND BLACKPOOL. 103 
 
 to set down. The speed is nominally about 1\ miles per 
 hour, but this is often exceeded. It is stated that the 
 company's receipts are 70 per cent, more on the electric 
 lines than on those worked by horses. The service is 
 frequent, being every 3-| minutes .on some days, and every 
 7 minutes on others ; but next year, it is stated, there are 
 to be 18 cars per hour. 
 
 Blackpool. The electric traction scheme in Blackpool, 
 which is of the conduit type, is due to Mr. Holroyd Smith, 
 and the conditions of working are necessarily very varied 
 on account of this being a seaside resort. As a matter 
 of fact, the population in winter is perhaps about 30,000, 
 whereas in summer it is said to be as much as 120,000. 
 Special difficulties have to be contended with on account 
 of the effects of sea-water and sand blowing into the 
 conduit. The line runs along the promenade or sea-front, 
 which is generally a good width. The steepest gradient 
 is about 1 in 50, and the sharpest curve has a radius of 
 about 40 feet. The rails are of the girder type, and weigh 
 about 92 to 97 Ibs. per yard. The slot and conduit are in 
 the centre between the rails. The electric tramway is 
 two miles in length ; the gauge is 4 feet 8-| inches. The 
 slot is about -f inch wide, which is much narrower than 
 that already cited at Budapest. The motor-cars are of two 
 sizes. There are four large ones, each with two 8-H.P. 
 motors, and carry 82 persons, and weigh 11 tons 14 cwts. 
 The small cars are ten in number, each has one 4-H.P. 
 motor, and carries 54 passengers ; the weight of the small 
 cars is 4-| tons. The large cars are 36 feet long, and cost 
 about 800, 360 of this being for the car body. The 
 fares on the electric route are 2d. in summer, for the whole 
 journey of about two miles, and Id. in winter. No less 
 fares than those just mentioned are taken. The voltage is 
 250. No accumulators are used either in the station or on 
 the cars. The speed of the cars is eight miles per hour, the 
 Board of Trade limit ; and the service is almost constant 
 
104 ELECTRICAL TRACTION. 
 
 in summer, whereas in winter it is at intervals of ten 
 minutes. 
 
 A distinctive feature of Mr. Holroyd Smith's system is 
 that he employs worm-wheel instead of ordinary spur- 
 gearing on the motor-cars. Some years ago Mr. Reckenzaun 
 found that the efficiency of this form of gearing increases 
 with the speed, and was from 84 to 95 per cent. ; the ratio 
 of reduction was 8 to 1, and speed of motor varied from 
 300 to 900 revolutions per minute. 
 
 The collector is shown in Fig. 42, and the following is 
 Mr. Smith's description of it. It consists of three main 
 parts, one centre-piece, C, and two clearing ploughs, PP, 
 for enabling it to run in either direction. One of_the 
 elements of success is the employment of flexible col- 
 lectors flexibly connected to the car, and automatically 
 detachable therefrom. For this purpose the two ploughs 
 are joined to the centre-piece, either by a tempered steel 
 strip, S, or a hinged wrought-iron plate, H. The steel 
 plates forming the ploughs are held by cast-iron cheeks 
 placed at an angle, and extending downwards a little past 
 the bottom of the steel troughs forming the surface of the 
 channel ; their upper ends terminate in a hook or finger 
 curved backwards. The hooks received the "snap loop" 
 of the hauling ropes, which are attached to the front and 
 rear of the car. These snap links are strong enough for 
 ordinary work, but should an absolute block occur, the 
 break and the collector is left behind, while the car, by its 
 momentum, still travels forward, and the snap link of the 
 trailing rope slips off the hook of the rear plough. The 
 centre-piece consists of a cast-iron clamp holding a plate 
 of strong brass, which is thoroughly insulated and protected 
 by hardened steel guards where it passes the channel slot. 
 The bottom of the plate is bared of insulating material, 
 and has attached to it a flapper or wing, held outward by 
 a spring so as to maintain contact with the conductor-rod. 
 The current is taken from the collector by a short length 
 
io6 ELECTRICAL TRACTION. 
 
 of insulated wire, provided at either end with a dip, one 
 end attached to a terminal under the car, the other to the 
 upper end of the brass plate that passes through the iron 
 clamps. In case of the collector breaking away these clips 
 readily let go. 
 
 The Author is indebted to Mr. Smith for information 
 respecting his new system of underground conduits, which 
 has received the approval of the Demonstration Francaise 
 des Fonts et Chaussees of Paris. The channels are formed 
 under the rails, and, in the case of a double tram-line, lie 
 next to the inter-space. The underground conduit is large 
 enough for a man to get along the whole length of the line, 
 and in that way repairs can be effected without taking up 
 the roadway. Also the great depth of the conduit allows of 
 dirt and water being kept far enough away from the rails. 
 The channels are supported on transverse H girders let into 
 the walls of the conduit, and any dirt which may get in 
 between the slot must fall to the bottom of the conduit, and 
 can easily be swept away. At intervals along the line there 
 are switches by means of which the sections of the conductor 
 can be insulated, and a live conductor is carried on a 
 porcelain insulator inside the conduit, and is attached at 
 different points to the rubbing conductors. The Author has 
 no data with regard to the probable first cost of this system, 
 so that he cannot compare it with the other methods 
 already given. 
 
 Combined Cable and Electric Traction. An allied 
 system has been proposed by Mr. Scott (see Journal of 
 the Inst. of E. E., vol. xxvi.), in which a cable tramway 
 is worked by means of electric motors in order to diminish 
 the friction due to running a cable round corners, and to 
 supply power to cars when ascending very steep gradients, 
 but the Author is not aware of this system having been put 
 into operation. 
 
 Other Slotted Conduit Systems. The Author has not 
 the space to discuss all the proposed slotted conduit systems- 
 
OTHER SLOTTED CONDUIT SYSTEMS. 107 
 
 It is not necessary to remark that many ways of solving 
 this problem are possible. For instance, the Lachmann 
 and " Simplex " systems present novel features (see 
 Electrician, July 2, 1897, and The Railway World, April, 
 1897). 
 
io8 ELECTRICAL TRACTION. 
 
 CHAPTER VI. 
 
 THE SURFACE-CONTACT SYSTEM. 
 
 General. In this system of supplying electrical energy for 
 traction purposes, a series of surface-conductors or buttons 
 are arranged in the line of the track, and project slightly 
 above the ground, perhaps f of an inch. These conductors 
 are charged by automatic apparatus under the ground only 
 when a car is over them, and are intended to become 
 electrically " dead " when the car has passed on its journey. 
 Fixed underneath the car are sliding conductors, which 
 make contact with these buttons and collect the current 
 for propelling the car, and also actuate the apparatus under 
 the ground for connecting the surface-conductors to the 
 supply main. This system has advantages and dis- 
 advantages. 
 
 The advantages are, of course, that no overhead wires 
 are required ; further than this, by charging only a portion 
 of the conductors, the leakage is very much diminished, 
 and in this way economy of power is secured. The dis- 
 advantages are that the automatic apparatus may possibly 
 fail to act, in which case the surface-conductors might be 
 left charged, thereby making it very dangerous to foot 
 passengers, and horses might be killed. This is rendered 
 very improbable by the use of a trailing piece which touches 
 the buttons in turn, and disables the apparatus in the event 
 of one being left alive. For the purpose of classification 
 the Author divides this system into three groups. 
 
 First. The system in which insulated sections are 
 
THE SURFACE-CONTACT SYSTEM. 109 
 
 employed, these being charged or rendered alive by a 
 mechanical contrivance. These appliances can be actuated 
 by other means than the car, and this makes the system 
 dangerous. So far as the Author is aware the first system 
 patented is that of Ayrton and Perry in 1881. In this 
 system the flange of the wheel on the car is made to depress 
 a lever just before entering a section, such depression con- 
 necting that section to the source of electric supply, which 
 can be an insulated feeder running parallel with the track. 
 Having entered the section, the car is propelled by the 
 current passing through this, but the flange immediately 
 presses another lever, which disconnects the section just 
 left, and makes it dead. Thus contact is broken between 
 it and the supply feeder. 
 
 Second. The system in which the sections are made 
 alive by the lifting up of the conductor, which is placed in a 
 closed conduit underneath the ground, such lifting being 
 brought about by a magnet attached to the car. LinefTs 
 1888 patent describes a system in which an insulated 
 flexible conductor, in the form of a strip, is connected to 
 the source of supply, and rests normally on the bottom 
 of a closed underground conduit. Attached to the car 
 is a long electro-magnet, with poles near to the surface, 
 and this, by virtue of its magnetism, sucks up the flexible 
 conductor underneath, and establishes contact with the 
 section to be made alive. This system has been tried on 
 a length of experimental line in London, and reported upon 
 by Mr. Kapp.* 
 
 Diatto. In this system a series of electro-magnets, 
 Mj, M 2 , M 3 , M 4 , M 5 , in Fig. 43, are mounted upon an iron 
 skate, E, which is long enough to touch two consecutive 
 contacts, j, # 2 . These contacts slightly project above the 
 ground level, and serve to transmit electrical current to 
 the motor M, when connected to the insulated supply- 
 feeder C. Underneath each contact, a lt a^ etc., an iron 
 * See Electrician, Sept. 12, 1890. 
 
no ELECTRICAL TRACTION. 
 
 bolt floats in mercury in a vertical cylinder, the mercury 
 being connected through the metal cylinder permanently 
 to the feeder, 0. When the skate, carrying the electro- 
 magnets Mj, M 2 , etc., which are excited by a shunt circuit 
 across the motor, comes over a surface-contact, the magnet 
 nearest it attracts the iron bolt b ly lifting it into contact 
 with <7 15 and electrical connection is thereby established. 
 After the car has passed on its journey, the bolt ^ drops 
 
 r 
 
 FIG. 43. 
 
 and severs contact. This system, according to Merechal, 
 is being tried at Toure and Saint Nazaire. 
 
 Wheless, McLaughlin, and others have patents for systems 
 in which a magnet on the car lifts an armature under the 
 ground, thereby actuating a switch which places a surface- 
 contact in connection with a supply-feeder. 
 
 Third. The system in which sections are made alive 
 solely by electro-magnetic means, thereby necessitating 
 the connection between the car, and suitable contacts only 
 for the passage of the electric current. A battery carried 
 on the car generally supplies current for starting such 
 connections, but in some cases the line current also 
 does this. 
 
 The objects to be attained may be briefly summarized as 
 follows : First, absolute safety as regards disconnecting 
 the surface conductors from the supply-feeder after the car 
 ceases to be over them. Second, the reliability of obtaining 
 a supply of current, especially at high speeds. Third, 
 
THE SURFACE-CONTACT SYSTEM. 
 
 in 
 
 leakage to be guarded against, as this might operate the 
 electro-magnets, rendering the surface-contacts alive, and 
 thereby occasioning great danger to pedestrians and horses. 
 Fourth, facility for getting at the automatic apparatus in 
 case of failure. Fifth, means of propelling the car over 
 short distances independently of the switch apparatus. 
 Sixth, to remove complications at points and crossings. 
 
 This system would seem to offer the best solution to the 
 problem. The manner in which different inventors have 
 dealt with this matter will be best understood by describing 
 their individual systems. The following will serve to 
 illustrate several different methods of attacking the same 
 problem, and there are no doubt others. 
 
 Hopkinson, 1882. Dr. J. Hopkinson was, so far as the 
 
 Mi; 
 
 S 3 
 
 FIG. 44. 
 
 Author is aware, the first to patent a system in which 
 surface-contacts or conductors are provided, with which a 
 conductor attached to the car makes contact successively as 
 it proceeds upon its journey. In Fig. 44, A, B are the 
 rails upon which the motor-car or train is running. is 
 an insulated supply-feeder laid parallel with the track. The 
 car is provided with a battery, I), a motor or motors, M, and 
 a switch, S ; and underneath the car a sliding-conductor or 
 shoe, E, is attached, which makes contact with the surface- 
 conductors, #!, #25 etc., placed between the rails. The shoe 
 E is long enough to make contact with two contiguous 
 
112 ELECTRICAL TRACTION. 
 
 surface-contacts, and the car is long enough to cover these con- 
 tacts, which are, when it is over them, and then only, connected 
 to the feeder C. The length of the car or train then limits 
 the distance between the surface-contacts a, a^ etc. There 
 are preferably under the ground as many electro-magnetic 
 switches, S 1? S 2 , as there are surface-contacts. Each switch 
 is provided with a magnet, M 19 M 2 , etc., wound with two 
 coils, WH w<n w 39 WH etc., and the armature of each electro- 
 magnet, when in its normal position, makes contact with two 
 studs &u #2 in switch No. 1, # 5 , b & in switch No. 2, and so 
 on ; S 3 shows the normal position of armature. But when the 
 magnet is excited the armature is lifted, and contacts 3 , # 4 , 
 switch one, are made, and b 19 b 2 broken ; but, since the lower 
 contact-bridge piece is free to move on the shaft, there is a 
 time during the operation when all four contacts are made. 
 Suppose a car be in the position shown in the figure, and 
 that the shoe E is in contact with contacts a l9 a z ; now 
 make switch S. This allows the battery D to send an 
 electric current round the coils ?r 2 , ?r 4 , which are of fine wire, 
 in parallel, and so to the return conductor B. The arma- 
 tures are lifted, and sever contact with b l9 # 2 > ^ ^>> 
 respectively, making contact with # 3 , & 4 , # 7 , & 8 , thereby 
 placing the shoe E in contact with the feeder ; that is, the 
 motor M can take current from the supply system through 
 the thick coils tv 19 W B . When, finally, the shoe E breaks 
 contact with a^ 9 and the motor-car moves from left to 
 right, the armature of S x drops and closes the fine 
 wire coil w 2 ready for the next car which enters the 
 section. 
 
 It is stated that, in this system, by having supply simul- 
 taneously between two magnet coils in parallel, they may, 
 owing to their different resistance and each circuit has 
 three contacts so subdivide the current as to cause it to 
 almost wholly traverse one magnet winding and not the 
 other, thereby rendering the latter inactive. 
 
 Wheless. The patent rights of this system are owned by 
 
HOPKINSON AND WHELESS. 
 
 the Westing-house Company, who have thoroughly tested it. 
 It has been successfully at work in Washington for over a 
 year on a branch line nearly a mile long, where a five- 
 minutes' service was maintained, or twelve return trips per 
 hour for twelve hours a day during 1895, one car running 
 over 40,000 miles. The potential difference is that usually 
 employed in tramway work, namely, 500 volts, and during 
 the time it was at work no accident of any kind occurred. 
 This is not the only example of this system having been 
 tested. The Westinghouse Company have had it at work in 
 their factory at Pittsburg, and it has proved of great con- 
 venience, since the shunting work in the shops is carried out 
 without the smoke associated with the ordinary steam 
 
 =111, 
 
 aft, 
 
 wlLEl 
 
 s, 
 
 FIG. 45. 
 
 locomotive. Fig. 45 gives a diagrammatic sketch of the 
 system. As in the Hopkinson system, A, B are the rails, 
 C the insulated feeder, b lt & 2 , t> 3 a row of surface-contacts 
 which transmit energy from the supply-circuit to the 
 motors when the switches, S^ S 2 , S 3 , etc., are operated. The 
 two rows of conductors are arranged side by side between 
 the rails, as shown, and the car is provided with two sliding 
 conductors, E : , E 2 , making contact respectively with the two 
 rows of surface-contacts. The electro-magnets, M 1? M 2 , etc., 
 as in the Hopkinson system, are wound with two coils each ; 
 
ELECTRICAL TRACTION. 
 
 
 for instance, in switch S 2 , u\ 
 is a thick wire coil, and u\ 2 a 
 fine wire coil ; the thick wire 
 coil is connected to the surface- 
 contact # 2 > and the contact 1 6 
 of the switch S 2 , contact 1> 5 
 being connected to the feeder 
 C. The fine wire coil u\ 2 is 
 connected to the surface-con- 
 tact 2 and the return con- 
 ductor B. Suppose a car to 
 be placed as shown. The 
 battery D can be made to send 
 a current through E : to <7 2 , 
 thence through coil ic z to the 
 return. This current operates 
 switch S 2 , and the conductor 
 > E 2 is connected to the feeder 
 - C through contacts & 5 , 5 6 , and 
 the coil u\ in switch S 2 . 
 Energy can now be delivered 
 to the motors to operate the 
 motor M on the car. Fig. 46 
 gives a cross-section of the 
 roadway showing the rails and 
 a pair of surface-contacts or 
 buttons in position. These 
 buttons are about 10 or 12 
 centimetres in diameter, and 
 stand about one centimetre 
 above the ground level. The 
 casting supports them firmly, 
 and also provides means for 
 leading in the conductors from 
 the electro-magnetic switch. 
 This latter has been very 
 
WHELESS. 
 
 carefully thought out with a view to accessibility, and re- 
 placing any part quickly. Fig. 47 gives sectional side and 
 front elevations of the switch and electro-magnet, together 
 
 with the cast-iron box which contains these. The box is 
 placed at the side of the rails and flush with the ground 
 level. It also shows the electro-magnet supported by 
 
n5 ELECTRICAL TRACTION. 
 
 means of a bell-shaped cast-iron box with the switch inside. 
 The cables are brought in through water-tight joints in the 
 
 FIG. 48. 
 
WHELESS AND JOHN&ON-LUNDELL. 117 
 
 outer box, and led up through a circular casting to vertical 
 pillars fixed to the same. When the bell-shaped top of the 
 electro- magnet is placed in position, connection is automati- 
 cally made with this vertical contact-rod ; and the bottom 
 of the cover fits into a V-shaped circular groove, in which 
 oil or other substance can be placed to render it air and 
 water tight. The contacts #. 5 , # 6 , in Fig. 45, are made of 
 carbon, as shown in Fig. 47, and fixed to the core of the fine 
 and thick wire solenoids. There is nothing, therefore, to 
 prevent this core from dropping immediately the current in 
 the solenoid is interrupted. Fig. 48 gives a view of the per- 
 manent way as installed at Washington. One sees nothing 
 different from an ordinary horse tramway except these 
 surface-buttons, which are automatically rendered dead 
 when the car is not over them. The sliding conductors 
 under the car are supported by springs from insulators on 
 a cross-bar, and are slightly curved upwards at each end, 
 so as to make sure of passing over the surface-contact. 
 
 Johnson-Lundell. This system has been in operation at 
 New York, and has demonstrated that surface-contact rail- 
 ways can be made a practical success. Like that of Wheless, 
 this system has two rows of contacts on the surface, %, a z , 
 etc., &!, 6 2 , etc., but these are not abreast as in the Wheless 
 system ; they are as shown in Fig 49. A distinctive feature 
 of this system is that there is no coil of wire on the electro- 
 magnet between the feeder C and the contact series a l9 a^, 
 etc. This, of course, renders the system very secure against 
 leakage. Another feature of this invention is that two 
 electro-magnetic switches are operating at the same time 
 and by the same current, as against one in the Hopkinson 
 or Wheless. This will be best understood from the descrip- 
 tion. The car is provided with a battery, D, a switch, S, 
 and a motor or motors, M. It has fixed to it, as in the Wheless 
 system, two sliding-conductors, E x , E 2 , which are also 
 long enough to project over two contiguous contacts in 
 each series ; M 15 M 2 , etc., are, as before, electro-magnets ; 
 
J 
 
 D 
 
 Ch 
 
 VWiWt > 
 
 Q- 
 
JOHNSON-LUNDELL. 1 1 9 
 
 Si, S 2 , etc., are switches ; and each electro-magnet is wound 
 with two coils, connected in the following manner : Surface- 
 contact 6 4 is connected by a conductor to coil w% on magnet 
 M 4 , and is continued to coil w 2 on magnet M 3 , and thence 
 to the return conductor B. Similarly, b s is connected to the 
 other coil u\ on M 3 , and continued through a coil on M. 2 
 to the return B. The surface-contacts, t , a^ etc., are con- 
 nected to one contact of each of the switches, S 15 S 2 , etc., 
 the other contact being connected to the feeder C. Suppose 
 all the switches, S 1? S 2 , be down, and the switch S on the car 
 be made. The battery D causes a current to flow through 
 the motor or motors M to E 2 , from thence through & 3 , round 
 the upper coil u\ on magnet M 3 to the lower coil on 
 magnet M. 2 , and thence to the return B. This makes 
 switches S 2 , S 3 ; that is, # 2 > s become charged, since they 
 are connected to the feeder through the switches S 2 , 
 S 3 . The distance between # 2 , a 4 must be covered by the 
 car. The motor can now receive energy from the supply 
 circuit through tf 3 , such supply-current now passing through 
 the magnet windings on the magnets M 2 , M 3 , which we 
 saw just now were traversed by the battery current. The 
 advantage of this method of connection is that all the coils 
 on the magnets are connected to the return B, and therefore 
 they are kept at earth potential. By another arrangement of 
 Johnson and Lundell, the contacts b l9 b. 2 , 6 3 , etc., are placed 
 outside the rails, and therefore any surface leakage must get 
 over or under the rails, if it proceeds from a lt 2 3, etc., 
 before traversing the magnet windings, thereby making the 
 system dangerous. In this latter case the surface-contacts 
 &i &2> &3> etc., are in the form of longitudinal rails, thereby 
 making the conductor E 2 much shorter ; but it must be 
 long enough to project over the space between the longi- 
 tudinal rails. 
 
 The Lundell magnet is shown in Fig. 50. A A are blocks 
 of carbon clamped to insulated supports, BB, and the object 
 of the switch is to bridge across between AA when a current 
 
FIG. 50. 
 
THE SURFACE-CONTACT SYSTEM. 121 
 
 passes through the magnet windings. These windings, w l9 
 iv^ are shown in section, and surround a vertical bolt of soft 
 iron, free to move vertically within them. An extension of 
 the iron core carries six metal cross-arms, provided with con- 
 tact-conductors, CO, of such shape that when the solenoid is 
 energized by an electric current, the conductors make good 
 contact with the carbon pieces AA. The whole is rendered 
 waterproof by means of a cast cover which rests in a groove 
 on the base as in the Wheless construction. We see that 
 the Wheless and Lundell magnets consist of a solenoid and 
 iron core, which is different to the form of electro-magnet 
 generally met with. It is well known that if a potential 
 difference be suddenly applied to the ends of the coil of an 
 electro-magnet, the magnet does not immediately operate, 
 owing to the reaction of the magnet upon the circuit. 
 Quickness of action is of importance in tramway working, 
 especially if the speed is great. A speed of ten miles an 
 hour corresponds to about five yards per second, and this 
 might be the length of the skate or collector under the car, 
 that is, at this speed contact might be made with a button 
 for one second. We see, therefore, how important it is to 
 have an electro-magnet which acts at once, so that the 
 contact between feeder and button may be as long as 
 possible. 
 
 Taking the different systems in order, and as shown in 
 Figs. 44, 45, 49, we see that in Hopkinson's, when the 
 shoe touches two buttons, the switch holds solely by the 
 current passing through the thick coil, and therefore 
 unequal division of current in the two parallel coils may 
 result in the failure of one magnet to act properly. The 
 same remark might apply to Wheless', so far as the thick coil 
 is concerned ; but the current through the thin wire coil is 
 not automatically interrupted when the armature lifts, and 
 action is therefore certain. In the Johnson-Lundell no 
 thin wire coil is employed ; but, since there are always two 
 coils in series, the coils being on contiguous magnets, it 
 
122 ELECTRICAL TRACTION. 
 
 follows that, for a given current in the motor, if the turns 
 be the same on each coil of each magnet, the one magnet 
 must have the same net ampere-turns in spite of unequal 
 division of current, and action is certain. 
 
 With regard to leakage, taking the diagrams as they 
 stand, we see that in. Wheless the earthed coil is the coil 
 excited by the battery, and therefore the more nearly the 
 potential difference of this battery is equal to the maximum 
 potential difference of the system, the more secure is the 
 system against operation of the magnet by leakage. The 
 castings which support the buttons, if put directly to the rail 
 return, would help considerably to shield such buttons from 
 leakage. In the Johnson-Lundell, all the coils are to earth, 
 but they have low resistance ; it is therefore necessary to 
 shield the buttons connected to such coils as much as possible 
 from leakage, and for this reason in practice the rail is 
 placed in between the two series of buttons shown in the 
 diagram. 
 
 The Author inclines to the opinion that, since power is lost 
 by departing from the nearly closed magnetic circuit, the 
 best way would be to operate such magnets at the maximum 
 potential difference of the system, interposing in the circuit 
 of the magnet a considerable resistance as non-inductive as 
 may be. In this way, during the initial stages, when the 
 current is small, nearly the whole potential difference would 
 be applied to the magnet, thereby making it act quicker, and 
 leakage would not operate such magnet since the maximum 
 potential directly applied is necessary to do this. 
 
 Esmond's British Patent, No. 22519 of 1895. This 
 patent describes a system in some respects resembling those 
 just described. Esmond arranges his surface-conductors in 
 sets of three, and excites the magnets of advanced sections, 
 thereby attaining, to some extent, the same results as with 
 the two magnets in the Johnson-Lundell system. 
 
 Wynne. This is an allied system to those just men- 
 tioned, and has been tried on an experimental line at 
 
CLARE T> WIULLE UMIER. 
 
 123 
 
 Wolverhampton. Mr. Thomas Parker has given a favour- 
 able report on this system. 
 
 Claret- Wiulleumier. This system has been at work in 
 Paris for some considerable time, on a line 4f miles long, of 
 which 2J miles are in the city. The maximum gradient 
 is 1 in 217, and there are numerous curves, some having a 
 radius of 97 feet. The generating station contains three 
 200-H.P. horizontal type condensing engines, each driving by 
 belt a 140-kilowatb four-pole over-compounded generator, 
 giving 500 volts potential difference. The surface con- 
 ductors in this system consist of metallic buttons embedded 
 in bitumen eight feet apart, connected in pairs by an under- 
 ground cable having a cross-sectional area of 0*012 square 
 
 FIG. 51. 
 
 inch to the supply-feeder by means of automatic switches 
 called " distributors." A conductor, ten feet long, fixed 
 underneath the car, makes contact with these surface-con- 
 ductors, and consists of an iron band pressed down upon the 
 contacts by springs. There are two supply-feeders, one 3f 
 miles long and 0*46 square inch section, the other 1^ 
 miles long and 0'35 square inch section. Fig. 51 gives a 
 diagram of connections, from which it will be seen that the 
 surface-contacts are divided into groups. In Paris there are 
 18 per group, each group being connected to an under- 
 ground distributor. In the sketch only eight contacts are 
 shown with each distributor, D x , D 2 . This number, of 
 
124 ELECTRICAL TRACTION. 
 
 course, depends upon the distance between the cars and the 
 distance between two consecutive contacts on the line. 
 Initially, the distributor is as shown in D 2 . There are two 
 brushes bearing on contiguous contacts, each brush being 
 connected to metallic rings, on which a single fixed brush 
 bears. The distributor is driven by means of a weight, or 
 motor, in such a manner that it carries round the two rings 
 with their brushes so as to make contact with the blocks 1, 2, 
 3, 4, etc., successively. Suppose a motor-car is in the position 
 shown, that is, the conductor E is making contact with No. 
 5 button. The current can pass from the feeder C to the 
 inner ring of the distributor, thence to contacts 5, 5, on 
 distributor and surface, through the motor M, to the return 
 A. If the car moves in the direction of the arrow, it will 
 bring the collector E into contact with No. G. The current 
 can now pass from this contact to No. G in the distributor, 
 through the coil on the electro-magnet M b to the return B. 
 This magnet, when operated, releases the mechanism which 
 turns the distributor to the next position, thereby placing 
 the brushes on G and 7 respectively. The motor M now 
 takes its current from contact G, through the distributor, 
 instead of contact 5. Surface-contact 8 is connected 
 by means of a cable to No. 1 on the next distributor ; 
 and so the car, when arriving there, begins to operate 
 distributor D 2 , and so on, leaving distributor D x at its initial 
 position, ready for the next car. If the distributor should 
 fail to act and the car goes forward, it must be turned round 
 until it is in the proper position for the then position of 
 the car. This is clearly a defect in the system, as here 
 described, but it may now be remedied. Another defect in 
 this system, as regards tramways, is that two cars cannot 
 be in the same section at the same time, as the distributor 
 would not have worked round, and consequently would 
 not supply current to a second car. This, however, be- 
 comes a decided advantage in the case of railways, con- 
 stituting a sort of block system. Means are provided for 
 
CLA RE T- W1ULLE UM1ER. 125 
 
 reversing the distributor so that the cars can traverse in 
 either direction. The cars are 28 feet long, and take 56 
 passengers, with a compartment for luggage. There are on 
 each car, two single reduction motors. The number of cars 
 running is 15, and 5 are under construction. The service 
 is a five-minutes' one. Street lighting is carried out by 
 means of 10 ampere arc lamps, 9 in series, and 14 such 
 circuits. 
 
126 ELEC7R1CAL TRACTION. 
 
 CHAPTER VII. 
 
 STOKAGE-CELLS. 
 
 General Remarks. Before entering upon the actual 
 application of storage-cells to traction work, it will be well 
 to look into one or two matters in connection with storage- 
 cells generally. Let a fully charged storage-cell be dis- 
 charged, say, by placing a rheostat and current meter 
 between its electrodes, and let the current be kept con- 
 stant by variation of the resistance. Place a voltmeter 
 across the terminals of the cell, and before starting the 
 discharge, let the volts be registered ; as the discharge 
 proceeds let them be observed and plotted in terms of time. 
 The general result is that the potential difference drops 
 immediately on making circuit, the extent of the drop 
 depending upon the internal resistance of the cell and the 
 magnitude of the current ; a slight recuperation occurs, and 
 then the potential is maintained almost constant for a con- 
 siderable time ; finally, the potential begins to fall rapidly, 
 and if the test be continued it drops to zero. The point 
 at which the potential difference begins to rapidly diminish 
 depends upon the rate at which the cell is being discharged, 
 and the density of the acid employed. It is usual not to 
 discharge cells beyond the point at which rapid drop of 
 potential sets in. The usual figure to take for this is 1*7 
 to 1*85 volts, according to the rate of discharge. The area 
 bounded by the current curve between the ordinates at time 
 zero, when the discharge starts, and any other time, and the 
 base-line gives the total quantity discharged in that time, 
 
S TOR A GE- CELLS. 1 2 7 
 
 say in ampere-hours. Let a curve be drawn giving the 
 products of volts and amperes at any moment. Then the 
 area bounded by this curve and the ordinates at time zero, 
 and any other time, and the base-line gives the actual work 
 done in that time on the circuit by the cell in, say, watt- 
 hours. If a series of charges and discharges be carried 
 out, the minimum volts on discharge being, say, 1'8 to 1*7 
 according to current, the ratio of watt-hours and ampere- 
 hours on discharge and charge give respectively the work 
 and quantity efficiency of the cell ; the criterion with 
 regard to full charge being the abundant liberation of gas 
 at the plates. But due notice must be taken of the rate 
 of charge. If the rate be great, gas will be liberated at the 
 plate before the cell is properly charged, and the test should 
 be continued for a longer time in order to properly charge 
 the plates. A fair figure for lead batteries is about 90 per 
 cent, for quantity and 80 per cent, for work efficiencies 
 respectively, but the latter may drop below 70 per cent, 
 (see p. 140) for high rates of discharge. For instance, it 
 is stated in connection with the Tudor stationary cell, that 
 the work efficiency is 80 per cent, for the 3 to 10 hour 
 rates, and 75 per cent, for the 1 to 3 hour rates. Let a 
 series of discharges be made each with constant current, 
 but let the currents be gradually increased with the suc- 
 cessive discharges, and let us take as a criterion either that 
 the discharge be stopped when the volts are 1*75, 'say, or 
 that it be stopped when the net drop in volts between their 
 initial and final values with the current passing is, say, 0'2. 
 It is found that the total discharge in ampere-hours becomes 
 smaller as the current is increased ; or you can put it this 
 way if the rate at which you discharge a cell be halved, 
 the test is continued for more than double the time before 
 the limiting conditions are reached, and consequently the 
 ampere-hours are greater. This is well shown by the curves 
 in Fig. 52, which give the total quantity in ampere-hours, 
 the rate of discharge in amperes in terms of the time 
 
128 
 
 ELECTRICAL TRACTION. 
 
 during which the test is continued for one Electrical 
 Power Storage Company's K type plate, measuring about 
 9f by 9 inches and the same Company's Faure-King type 
 central-station cell-plate, the data being taken from the 
 1897 list. Take the K type. We see that if we spread 
 
 20 
 
 NOTE: CURVES MARKED X SHOW RELATION BETWEEN TIME AND AMPS. PER POS. PLATE 
 
 ,, n ., ,. ., ,, ,, AMP. HOURS PER OOS.PLATE 
 
 FIG. 52. 
 
 the discharge over seven hours, the amperes being 5' 6, 
 the quantity in ampere-hours is 39 '3, whereas, for one 
 hour, if the amperes are 17*5, the quantity is only 17*5 
 ampere-hours. The Faure-King type shows the same cha- 
 racter of result, and the intermediate points at three and 
 
STORAGE-CELLS. 129 
 
 five hours have been calculated by the Author on the same 
 basis as in the K type. 
 
 The data given in the Electrical Power Storage Company's 
 1807 Price List show that for a given time of discharge, 
 before the limiting conditions are reached, the amperes at 
 which the cell can be discharged are proportional both to list 
 price and total weight of the complete cell. The data 
 given in connection with the Tudor cell give the same 
 character of result. 
 
 We see, therefore, that when specifying the capacity of 
 storage-cells, care must be taken to give the limits with 
 regard to time of total discharge, final volts with current 
 passing, and the rate in amperes. 
 
 Economy. The question may be asked for a given cell 
 At which rate is it best to charge or discharge ? In 
 answering this the actual requirements must be carefully 
 studied. Economy of working must be viewed from more 
 standpoints than actual efficiency ; that is to say, high 
 efficiency does not necessarily mean best economy. For 
 instance, it might be more economical to work a small battery 
 at high rates of charge and discharge with diminished life, 
 if by so doing a great saving can be effected by doing away 
 with the removal of cells from the car. If positive plates 
 could be made very cheaply, the actual life of the cell would 
 become of less importance, and in any case the maintenance 
 of storage-cells is in this connection a most important 
 item. Central station-cells are maintained at from 3-^ 
 to 10 per cent, of the first cost, according to the conditions 
 of working ;* but in traction work, with the cells on the 
 car, the maintenance would probably be higher if high rates 
 of charge and discharge are employed, and it is usual to 
 quote a rate for maintenance per car mile. For instance, 
 the Birmingham Tramways paid Id. per car mile for main- 
 tenance (see Electrician, August 26, 1892). At Hanover 
 the cost of maintenance per car mile is 0'15rf. (see p. 146). 
 * See p. 197 for maintenance cf storage-cells at Leeds. 
 
 K 
 
ELECTRICAL TRACTION. 
 
 With regard to the life of a cell for given working 
 conditions, only experience over a considerable time will 
 enable one to obtain definite information. The life of 
 ordinary station-cells is well known ; it depends upon the 
 type of cell, but four years is a fair average figure, if the 
 cells receive proper care and attention. 
 
 Density of Acid. The acid, too, is a very important 
 element in a lead storage-cell. Apart from purity, the 
 density has much to do with the life and working of the 
 cell. Table VII. is taken from Everett's "Tables of 
 Physical Constants," p. 180, edition 1891, and refers to the 
 specific resistance of dilute sulphuric acid. 
 
 TABLE VII. 
 
 
 Specific resistance. 
 
 Density 
 
 
 Al C. 
 
 At 8 C. 
 
 At 16 C. 
 
 At 24 C. 
 
 1-10 
 
 1-37 x 1(P 1-04 X 10 10 
 
 0-845 X 10 10 
 
 0-737 x 10 10 
 
 1-20 
 
 1-33 
 
 0-926 
 
 0-666 ., 
 
 0-486 , 
 
 1-25 
 
 1-31 
 
 0-896 
 
 0-624 
 
 0-434 , 
 
 
 1-30 
 
 1-36 
 
 0-94 
 
 
 0-662 
 
 0-472 , 
 
 
 140 
 
 1-69 
 
 1-30 
 
 
 1-05 
 
 0-896 , 
 
 
 1-50 
 
 2-74 
 
 2-13 
 
 
 T72 
 
 l-o2 
 
 
 1-60 
 
 4-82 
 
 3-62 
 
 
 2-75 
 
 2-21 
 
 
 1-70 
 
 941 
 
 
 6-25 
 
 
 4-23 
 
 3-07 
 
 
 We see that the specific resistance is a minimum for any 
 temperature quoted at a density of 1'25 ; but the variation 
 of resistance on either side of this density, within small 
 limits, is not great. 1'25 would seem, therefore, from con- 
 ductivity to be the best ; but a density not so high as this 
 is generally adopted in actual practice. The Electrical 
 Power Storage Company state that the specific gravity, 
 when the cell is fully charged, should be from 1*2 to 1*21, 
 and the discharge should never be continued after the 
 specific gravity has fallen to 1'170. These densities are 
 
METHODS OF WORKING CELLS. 131 
 
 very near those of maximum conductivity given in the 
 above table. The Chloride Electrical Storage Syndicate 
 state 1*215 as the specific gravity for the fully charged cell. 
 A high density has the effect of keeping up the volts on 
 discharge, but it probably shortens the life of the plates. 
 
 The specific gravity of the acid and the colour of the 
 plates are fair criterions as to the state of the cell ; but the 
 potential difference with current passing is the best guide 
 when discharging. 
 
 Methods of Working Cells. We find storage-cells used 
 ut the present time in traction work in the following ways : 
 
 (1) In the central station or power-house proper. 
 
 (2) On the car itself ; (a) where the whole power taken 
 to propel the car comes from the cells ; (/>) where 
 the cells are used in conjunction with an overhead 
 or underground conductor, or with a surface^contact 
 system. 
 
 (:-]) In feeder substations. 
 
 (4) On the car itself for lighting purposes only. 
 
 The Use of Storage-Cells in the Power-house. As 
 regards this division, we have seen that the fluctuations of 
 power delivered to the line by the power-house is con- 
 siderable, and to meet a sudden call for heavy discharge 
 the storage-cell is no doubt well adapted, and especially is 
 this the case on small systems. There are several instances 
 of batteries being used in parallel with the generating 
 dynamos, just as in an ordinary electric lighting station. 
 In the combined lighting and traction ' station at Rome, 
 for instance, a battery of 300 Tudor storage-cells of about 
 1000 ampere-hours capacity is placed in parallel with the 
 station machines, and the regulating cells are varied in 
 steps of three at a time for the purpose of keeping the 
 potential constant. At Hamburg, which is another notable 
 instance of combined lighting and traction station, a large 
 battery of storage-cells is employed in the power-house ; 
 and the sub-stations are provided with motor-generators 
 
132 ELECTRICAL TRACTION. 
 
 and storage-cells. Zurich is another instance of the 
 employment of storage-cells in central stations for traction 
 purposes. 
 
 The Use of Storage-Cells on the Car. () A self-con- 
 tained car equipped with storage-cells is no doubt, in some 
 respects, the ideal motor-car, the success of which at the 
 present time depends upon the storage-cell and its manage- 
 ment. The storage-cell is used on cars as the only pro- 
 pelling power, and to make it as successful financially as 
 the overhead trolley, a low first cost and maintenance are 
 the great things to be aimed at. On a tramcar, for 
 instance, since the whole weight may be on the driving 
 wheels, it follows that lightness is a clear gain, not only on 
 account of the actual weight to be carried, but also on 
 account of the permanent way. A battery of storage-cells 
 on an ordinary tramcar, which would weigh about 10 tons, 
 with passengers but without storage-cells, may weigh from 
 \\ to 2J tons. Two enclosed geared motors weigh about 
 1-J- tons for such a car, so that the battery is an important 
 extra so far as the weight is concerned, and of course this 
 is an objection. This aspect would not be so bad if the 
 battery were used on a locomotive drawing a train of 
 carriages. We find the storage-cells placed under the seats 
 of cars in many places ; but no doubt the best arrangement 
 is to make the undertruck to carry the cells as well as the 
 motors. The Peckharn storage-cell truck is illustrated in 
 Fig. 53, from which it will bs seen that the cells can be 
 placed in between the car axles and the motors overhang, 
 having their usual place occupied by the cells ; this arrange- 
 ment has the advantage that great strength can bs secured 
 for their support whilst they are removed from the wood- 
 work of the car, and are not so liable to be troublesome to 
 passengers. The cells can be very conveniently lowered 
 into a pit under the car, and this method is adopted at 
 New York. The storage-cell used on cars is, as a rule, 
 enclosed in teak lead-lined boxes, although ebonite is 
 
134 ELECTRICAL TRACTION. 
 
 sometimes used. The number of elements and their capacity 
 vary with the requirements of the car and the method of 
 working. We are at present considering the storage-cell 
 when worked at rates commonly used with cells, and there- 
 fore the battery must be removed from the car, properly 
 charged over a considerable time, and its place filled by 
 another one. The handling of storage-cells in this system 
 is an important item, to set against which one must con- 
 sider the longer life of the cell. 
 
 Birmingham. During the last seven years storage-cells 
 have been persevered with at Birmingham. The line is 
 about three miles in length, and equipped with twelve motor- 
 cars. It is stated that for the year 1803 the number of 
 car miles run was over 140,000, and the total cost per 
 car mile was lid. This figure is of course very high, the 
 most important item included therein being 5rf. for repairs 
 and maintenance. 
 
 The Author is indebted to the Chloride Electrical Storage 
 Syndicate for the following particulars with regard to the 
 working of their cells on this line. For upwards of three 
 years to June, 1897, this company have been miming the 
 cars at Birmingham, during w r hich time the aggregate 
 mileage run has been 383,000 miles. The battery in eaoh 
 car consists of 72 cells, some with 11 and some with 13 
 plates of the chloride TR type, the positive of which 
 measures 10 x 7 X -f^ inches, and weighs 6-| Ibs. when dry. 
 The negative plate measures 10 x 7 X yV inches, and 
 weighs dry 3 Ibs. 3 ozs. The cells are placed under the 
 seats of the carriages ; the weight of the battery in each 
 car is 50 cwts., and the cars themselves weigh about 10 
 tons. The average mileage per car per diem during the 
 three years has been C5. 
 
 New York. The cars on the Maddison Avenue, New 
 York, line are equipped with storage-cells of the Philadelphia 
 Electric Storage Battery Co. The trucks are constructed as 
 above described, and each is supplied with two General 
 
STORAGE-CELLS AT BIRMINGHAM AND NEW YORK. 135 
 
 1 
 
 Electric motors. There are 60 cells per truck, arranged in 
 two sets of 30 each, and a very convenient method of lower- 
 ing the complete battery from the truck into a pit underneath 
 is adopted, connection being automatically made with the 
 poles of the cells when in position. The proper handling of 
 cells is a point the importance of 
 which cannot be over-estimated, 
 and this method seems to give 
 special facilities. The superficial 
 area occupied by this battery is 
 about 24 square feet, and the 
 controller is so arranged that 
 initially the two sets of 30 cells 
 each are placed in parallel, with 
 the motors in series ; whereas, 
 finally, the motors are in paral- 
 lel, with their windings shunted 
 for high speeds, the GO cells 
 being in series, and the potential 
 difference about 120 volts. The 
 different stages are shown in 
 Fig. 54, and are carried out by 
 means of a controller similar 
 to that described in Chapter II. 
 The capacity of these cells is 400 ainpere-hours. 
 
 Paris. The Paris tramways furnish a very good example 
 of the employment of storage-cells in traction work. Three 
 lengths of line, as under, are equipped with two distinct 
 types of cars 
 
 Miles. 
 
 5-7 
 
 5-7 
 
 2-73 
 
 6 
 
 FIG. 54. 
 
 Madeliene Saint Denis 
 Opera Saint Denis ... 
 Neuilly Saint Denis 
 
 Total 
 
 14-13 
 
 The first type contains 108 Laurent-Cely type storage-cells, 
 having a capacity of 175 ampere-hours at a mean current 
 
136 ELECTRICAL TRACTION. 
 
 of about 35 amperes, placed under the seats. There are 
 nine cells in a box, and six such boxes on each side of the 
 car, which carries 52 passengers. The weights, etc., are 
 stated by Marechal* to be as follows : 
 
 Car, 17,000 Ibs. ; battery and accessories, GG10 Ibs. 
 = 2*95 tons ; passengers, 7720 Ibs. ; total, 14 '3 tons. 
 
 A fully charged battery takes the car 31 miles. The 
 motors have double reduction gear, and are worked on the 
 series-parallel control system. There are three steam- 
 engines in the power-house, each driving by a belt a shunt 
 machine giving about 2G<> volts. The working expenses 
 per car mile in 1893 are stated to be a little over 8d. t of 
 which 2%d. goes towards maintenance and handling of cells. 
 
 The second class of car has the cells placed on the imder- 
 truck between the axles, as is the case at Maddison Avenue 
 (see p. 135), and there are 5G cells instead of 108. The 
 advantages gained by this arrangement are reduced weight 
 and labour in charging operations. The weights are as 
 follows : 
 
 Car, 14,300 Ibs.; battery, 3750 Ibs. = T7G tons; pas- 
 sengers, 7720 Ibs. ; total, 11*5 tons. 
 
 A feature of these cars is the arrangement whereby 
 the energy due to gradients (see p. 10) on descending is 
 returned to the cells by converting the motors into gene- 
 rators. Allowing for efficiency on two parts of the track, 
 namely, Rue de Rome and Avenue de Saint Ouen, 27*3 
 and 21 '8 per cent., respectively, of the total energy is given 
 back to the cells by the motors on the cars. 
 
 The Paris tramways also give an illustration of rapid 
 rates of charge and discharge, whereby the cells are never 
 taken from the truck, but at the end of the journey the 
 charge is effected as the car stands, in, say, a quarter of an 
 hour. There are four lengths of line having a total length 
 of about 15 miles as follows : 
 
 * Sec Les Tramways Elcdriqucs. 
 
STOK AGE- CELLS AT PARIS. 137 
 
 Miles. 
 Madeliene Courbevoie ... ... ... 4'18 
 
 Madeliene Bineua Courbevoie ... ... 4'10 
 
 Madeliene Levallois ... ... ... 2'93 
 
 Neuilly Avenue du Konle ... ... ... 3'55 
 
 The cells are charged at a centre common to the four 
 lines by means of feeders, each transmitting energy from 
 a distant station. The cells are of the Tudor type, and 
 each car contains 200 5-plate cells, the plates having the 
 dimensions of 21 cms. high and 21 cms. broad. The 
 weight of a cell is 33 Ibs., and the entire battery 8-45 
 tons. The minimum guaranteed capacity is 32*5 ampere- 
 hours for a going and return journey at a maximum speed 
 of 10 miles per hour out of Paris, and 7*5 miles per hour 
 in Paris. If we assume the length of journey to be eight 
 miles, this gives about 1-| Board of Trade units per mile 
 (see p. f)0). The cells are charged in series at 540 volts, 
 the current being 120 amperes, which corresponds to about 
 8 amperes per kilogram of plate. There are 35 cars on 
 the four lines, each of which with passengers weighs about 
 14 tons. This novel way of dealing with the problem 
 will be watched with interest. It remains to be shown 
 whether this method, which does away with the cost of 
 handling the cells, is on the whole cheaper. We see that 
 for the weight the capacity is small, and this is what one 
 would expect with such high rates of discharge. 
 
 Under this system of rapid charge there is no reason 
 why provision should not be made at certain points along 
 the route for automatically making contact with a charging 
 feeder, so that the cells could be charged as the car stands. 
 For instance, an ordinary pole could support a contact 
 making-piece, which would be automatically connected as 
 the car runs underneath. 
 
 The copper-zinc storage-cell is deserving of attention, 
 as being a rival to the lead storage-cell for traction work. 
 In the Desmazures cell the positives are composed of fine 
 copper powder compressed upon copper gauze or net, under 
 
138 ELECTRICAL TRACTION. 
 
 a pressure of from 800 to 1200 kgm. per square centi- 
 metre, and have a total thickness of about seven mms. over 
 the paper cells which surround them. The negatives are 
 made of steel-wire netting, about eight to nine mms. thick ; 
 the plates being placed in an alkaline-zincate * solution. 
 Although the potential difference of these cells is low, about 
 0*8 volt, their weight efficiency is stated to be high. In 
 America the Waddell-Entz cell has attracted attention, and 
 differs only from the Desmazures in constructive details. 
 The copper-zinc cell is being used at Vienna, and appears to 
 give satisfactory results ; but the Author has no exact 
 figures with regard to maintenance and general expenses. 
 
 The Author has recently carried out, in conjunction with 
 Mr. H. H. Hodd, a series of experiments upon two totally 
 different types of traction-cells, and he wishes to thank 
 this gentleman for his assistance. The experiments were 
 made in the Siemens Laboratory, King's College, London. 
 
 The first type is a Faure-King five-plate traction cell, 
 supplied by the Electrical Power Storage Company. It 
 is a pasted cell, the active material of which is secured to 
 the surface of an almost non-oxidizable conductor ; the 
 plate is then inserted in a perforated envelope, composed of 
 a material capable of resisting destructive action on the 
 part of the oxides, the gases, and the electrolyte, such 
 envelope being secured to the plate by means of a series of 
 bolts and washers of this material. 
 
 The two positive plates measure each 7^ x S inches, and 
 have a thickness of \ inch, not including the envelope. 
 They weigh, with lug, when just taken out of the acid, 
 7 Ibs. 7 oz. The total weight of the cell in ebonite box 
 and acid is 21 Ibs. 2 ozs. The specific gravity of the acid 
 in the fully charged cell is 1*275. 
 
 The second type is that of the Chloride Electrical Storage 
 Syndicate, and known as their P.R. type 11-plate cell. 
 Two of these cells were supplied to the Author for the 
 * A solution of zinc hydroxide in excess of alkali. 
 
EXPERIMENTS ON STORAGE-CELLS. 139 
 
 purpose of these tests, by Mr. G-rindle, the manager of the 
 Syndicate. 
 
 Each cell has five positive plates, each measuring 5-^X7 
 inches, and \ inch thick. The five plates with lug, weigh, 
 when just taken out of the acid, 16*5 Ibs. The total weight 
 of the cell in ebonite box and acid is 3G Ibs. The specific 
 gravity of the acid of the fully charged cell is 1/200. 
 
 The positive plates consist of spirals of lead strip inserted 
 in a skeleton of harder material, and therefore represent, so 
 far as absence of paste is concerned, the original Plaute 
 cell. The negatives are the ordinary chloride type, with 
 hexagonal pastilles of spongy lead, round which is cast the 
 sustaining grid. 
 
 It must be clearly understood that these tests have been 
 made upon new cells, and therefore no conclusions can be 
 drawn as to the life of such plates ; and, further than 
 this, tests upon three isolated cells like these may not 
 represent the average results obtained from many. 
 
 The potential differences were taken all through by 
 balancing against a Clark's standard cell by PoggendorfF s 
 method. The currents were measured by a Siemens 
 electro-dynamometer, which could be checked at any time 
 by measuring the potential difference between the ends of 
 a known standard resistance in the circuit by means of 
 Clark's cell and Poggendorff's method. The ampere-hours 
 and watt-hours on discharge are very reliable ; but this is 
 not the case on charge for the high rates, owing to the 
 uncertainty when a cell is properly charged. We have 
 already stated that gas freely libsrated at each plate is a 
 good criterion that the cell is charged ; but at high rates 
 gas is certainly liberated before the cell may be said to be 
 fully charged, owing, no doubt, to the time required for 
 proper penetration. There is no doubt but that, in the 
 working of cells at high rates, a long steady charge (say at 
 night) at a low current density, would prolong the life of 
 the cell. 
 
J 4 
 
 ELECTRICAL TRACTION. 
 
 In the tests here described, the criterion as to when the 
 cell is sufficiently discharged has been to allow a 0'2 volt 
 drop on tho potential difference of the cell when this is 
 steady, two or three niinutos after switching on. The 
 results are given in Tables VIII. and IX. (pp. 141, 1-42, 
 and 143). 
 
 In the case of the Faure-King cell, the one-hour rate 
 final volts are 1-7 as against 1*85 at the nine-hour rate. 
 The points marked X in Fig. 55, give the ampere-hours 
 and watt-hours in terms of the total time of discharge in 
 
 TOTAL TIME OF DISCHARGE IN HOURS 
 
 FIG. 55. 
 
 hours before the limiting conditions are reached. The 
 quantity efficiency remains high all through, but the work 
 efficiency from the figures, drops from about 73 per cent, at 
 the five-hour rate to about 62 at the one-hour rate. In 
 the figure, the points marked are taken from the E.P.S. 
 1897 List ; the agreement is good. It will be noticed that 
 the charge, after any discharge is accomplished at a higher 
 current density per plate than was the case in the discharge. 
 The points Q in Fig. 55 give, for the Chloride cell, the 
 
EXPERIMENTS ON STORAGE-CELLS. 
 
 141 
 
 1 
 
 
 OS 
 
 
 
 . 
 
 
 
 
 
 co 
 
 1 
 
 s 
 
 1 
 
 in 
 
 S rH 
 
 in 
 
 
 ,_, 
 
 IN 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 00 
 
 in 
 
 
 
 
 
 * 
 
 
 o 
 
 
 
 
 OJ 
 
 OS 
 IN 
 
 1 
 
 
 
 ^ 
 
 CO 
 
 a* 
 
 * 
 
 o 
 o 
 
 7; 
 
 00 
 
 
 
 
 
 -* 
 
 ja ""* 
 
 m 
 
 M 
 
 
 
 00 
 
 ; s 
 
 1 
 
 i 
 
 3 
 
 ^ 
 
 a 
 
 a. 
 
 o. 
 
 ^ 
 
 9 
 
 
 
 1-1 
 
 " 
 
 s 
 
 -= ~ i 
 
 S 
 
 1-1 
 
 ' CO 
 
 5 
 
 
 3 
 
 to 
 
 OS 
 
 S. 
 
 OS 
 
 g^ 
 
 as 
 
 ^ 
 
 * 
 
 1 m 
 
 
 
 
 
 
 
 C5 
 
 
 
 b 
 
 
 " 
 
 
 
 N 
 
 C3 
 
 m 
 
 
 t- 
 
 
 
 
 
 
 
 
 , 
 
 
 
 
 
 s 
 
 1 
 
 
 ^ 
 
 Cl 
 
 II 
 
 CO 
 
 co 
 
 s 
 
 
 
 9 
 
 35 
 
 _ 
 
 1 
 
 s 
 
 {2 
 
 
 g'25 
 
 
 
 OS 
 
 CO 
 
 
 
 
 H 
 
 Ai 
 
 s 
 
 .dri 
 
 e* 
 
 1-1 
 
 
 Ci 
 
 
 
 to 
 
 in 
 
 
 
 
 
 
 
 
 . 
 
 CJS 
 
 3 
 
 ^ 
 
 S ~"i 
 
 op 
 
 c* 
 
 OS 
 
 1^ 
 
 1-1 
 
 1 
 
 ^ 
 
 * 
 
 (H 
 
 ffl M 
 
 CO 
 
 s 
 
 00 
 CO 
 
 So 
 
 *. 
 
 1 
 
 oo 
 
 oo 
 
 O 
 
 S- 
 
 CO 
 
 o 
 
 9 
 
 i 9 
 
 
 
 - 1 
 
 -' 
 
 00 
 
 .a'* 
 
 s 
 
 s 
 
 t- 
 
 
 2 
 
 OJ 
 
 
 
 00 
 
 7< 
 
 at 
 
 
 ^ 
 
 00 
 
 9 
 
 
 w 
 
 
 
 2 
 
 a* 
 
 
 
 s 
 
 s 
 
 CO 
 
 00 
 
 2 
 
 O5 
 
 
 
 
 
 00 
 
 a 
 
 s 
 
 2 
 
 
 
 g 
 
 OS 
 
 
 m 
 
 i 
 
 op 
 
 
 32 
 
 
 
 
 
 
 01 
 
 (^1 
 
 ^ 
 
 00 
 
 - ; 
 
 in 
 
 ^4 
 
 in 
 
 ^ 
 
 
 
 
 
 
 
 
 1-1 
 
 CO 
 
 00 
 
 cs 
 
 
 
 1 
 
 oc 
 
 _, 
 
 as 
 
 os 
 
 
 co 
 
 CO 
 
 
 
 
 
 00 
 
 
 CO 
 
 
 in 
 
 
 
 
 
 
 
 
 
 1-1 
 
 
 OS 
 
 
 
 9 
 
 in 
 
 oo 
 
 OS 
 
 a' 
 
 <N 
 
 m 
 
 M 
 
 t- 
 
 
 
 
 
 
 ^2 
 
 00 
 
 1-1 
 
 CO 
 
 SB 
 
 
 1 
 
 I 
 
 2 
 
 s 
 
 c S 
 
 o 
 b 
 
 
 00 
 
 9 
 
 
 
 
 
 - 1 
 
 43 in 
 
 00 
 
 S 
 
 CO 
 
 oo 
 
 
 
 m 
 
 m 
 
 o 
 
 
 o 
 
 in 
 
 
 
 CO 
 
 
 IN 
 
 ,1, 
 
 
 
 ,4 * 
 
 OS 
 
 "< 
 
 
 
 
 1 
 
 9 
 
 S 
 
 p 
 
 
 
 
 in 
 
 
 1 
 
 
 
 IN 
 
 1-1 
 
 
 
 43 
 
 
 
 
 
 
 
 .s~s^T? 
 
 TS7 
 
 ~"^" 
 
 1 
 
 &s> 
 
 
 ' 
 
 
 i 
 
 1 
 
 3 
 
 jS 
 
 Volts when current 
 zero; i.e. open circ 
 just before start! 
 test 
 
 c> 
 
 Final volts at end 
 test ; current passi 
 
 Amperes constant a 
 
 Total time of dischai 
 in hours and minul 
 
 Arnpere-hours ... 
 
 Watt-hours 
 
 Work efficiency 
 
 Quantity efficiency 
 
142 
 
 ELECTRICAL TRACTION. 
 
 11 
 
 
 1 
 
 IN 
 
 
 
 m 
 
 g w 
 
 
 
 CO' 
 
 j 
 
 1 
 
 M 
 
 
 <N 
 
 IN 
 
 
 
 ^j-- 1 
 
 
 
 
 
 O 
 
 1 
 
 1 
 
 
 
 o 
 
 g = 
 
 
 
 
 
 | 
 
 I 
 
 10 
 
 
 IN 
 
 M 
 
 S 
 
 fl^ 
 
 
 
 r " ( 
 
 
 
 ?! 
 
 1 
 
 
 
 i 
 
 9 
 
 |1 
 
 CO 
 
 s 
 
 1 
 
 1 
 
 
 1 
 
 C; 
 
 3 
 
 N 
 
 g'O 
 
 00 
 
 g 
 
 | 
 
 I 
 
 IN 
 
 
 
 
 
 
 
 
 
 
 
 
 M 
 
 
 IN 
 
 ' ' 
 
 
 
 
 
 
 1 
 
 
 P 
 
 eo 
 
 a's 
 
 * 
 
 CO 
 
 | 
 
 | 
 
 C( 
 
 ,- 
 
 
 S 
 
 
 S 
 
 & 
 
 CO 
 
 1-1 
 
 
 
 ^^ C 1 ! 
 
 1 
 
 S 
 
 p 
 
 -* 
 
 gO 
 
 CO 
 
 co 
 
 1 
 
 | 
 
 C^ 
 
 
 IN 
 
 IN 
 
 CM 
 
 ^3 w 
 
 
 
 
 
 y 
 
 
 
 7-1 
 
 
 
 
 
 
 
 3 s 
 
 1 
 
 S 
 
 5 
 
 w 
 
 2 
 
 a'2 
 
 O5 
 
 GO 
 
 1 
 
 1 
 
 1 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 >? S 
 
 1 
 
 g 
 
 IN 
 
 8 
 
 2 
 
 a'S 
 
 2 
 
 s 
 
 1 
 
 1 
 
 H 
 
 
 <N 
 
 
 IN 
 
 w 
 
 
 
 
 
 2 
 
 1 
 
 N 
 
 3 
 
 O 
 
 GO 
 
 go 
 
 9 
 
 
 
 m 
 
 1 
 
 1 
 
 "o 
 
 
 (N 
 
 c* 
 
 
 r^H in 
 
 C5 
 
 
 
 
 g 
 
 
 M 
 
 IN 
 
 
 m 
 
 ^ 
 
 ^ 
 
 
 
 H S) 3 
 
 1 
 
 GO 
 
 9 
 
 S 
 
 a -*, 
 
 M 
 
 S 
 
 1 
 
 1 
 
 1 I 
 
 
 C-l 
 
 
 IN 
 
 * 
 
 C5 
 
 
 
 
 S 2 
 
 1 
 
 CO 
 
 9 
 
 
 
 a 1 ^ 
 
 S 
 
 1 
 
 1 
 
 I 
 
 i 
 
 
 IN 
 
 
 (N 
 
 ^= ^ 
 
 
 
 
 
 2 
 
 1 
 
 GO 
 
 w 
 
 
 
 EO 
 
 
 
 O 
 
 1 
 
 I 
 
 bd 
 
 
 IN 
 
 IN 
 
 C1 
 
 A *' 
 
 CO 
 
 rH 
 
 
 
 i 
 
 1 
 
 in 
 
 9 
 
 00 
 
 S'S 
 
 IN 
 
 S 
 
 1 
 
 1 
 
 ci 
 
 
 7^ 
 
 N 
 
 rt 
 
 ^^ 
 
 5 
 
 N 
 
 
 
 CO 
 
 1 
 
 S 
 
 CO 
 
 00 
 
 a 
 
 9 
 
 GO 
 
 1 
 
 1 
 
 O 
 
 
 (N 
 
 O* 
 
 1-1 
 
 .s 10 
 
 05 
 
 ^ 
 
 
 
 en 
 1 
 
 -1 
 
 || ! 
 
 r 
 
 
 o 
 
 = 
 
 : 
 
 : 
 
 : 
 
 
 t>3 O 
 
 .21 
 
 i : 
 
 i" 
 
 
 
 
 
 
 
 1 
 
 "S 
 
 fa 
 
 when current 
 i circuit just be 
 
 when they are 
 ost immediately 
 on current ... 
 
 volts at end of 1 
 ling 
 
 res constant at 
 
 1 ' 
 
 2 
 B.S 
 
 -n s 
 
 1 
 1 
 
 
 
 efficiency 
 
 1 
 
 <u 
 
 
 38.1 
 
 *J 
 
 11.3 
 
 ii 
 
 a 
 
 I 1 
 
 f 
 
 1 
 
 o 
 
 C 
 
 03 
 
 O 4 
 
EXPERIMENTS ON STORAGE-CELLS. 
 
 '43 
 
 
 
 
 ,0 
 
 
 
 
 i 
 
 
 
 (0 
 
 1 
 
 1 
 
 1 
 
 IN 
 
 i 
 
 S^ 
 
 S 
 
 3 
 
 1 ! 
 
 
 
 * 
 
 1 
 
 1 
 
 S 
 
 9 
 
 o 
 
 aS 
 
 S 
 
 ta 
 
 1 
 
 1 
 
 
 
 
 
 
 IN 
 
 IU 
 
 ^3 "" 
 
 * 
 
 
 
 
 
 
 2 
 
 1 
 
 1 
 
 
 <N 
 
 p 
 
 35 
 
 te 
 
 
 
 1 
 
 1 
 
 3 i o 
 
 3 * J 
 
 1 
 
 1 
 
 10 
 
 BO 
 
 3 ^ 
 
 _rj IN 
 
 te 
 
 9t 
 
 
 1 
 
 <u i 5 
 
 
 
 
 
 
 i 
 
 
 j 
 
 3 
 
 1 
 
 1 
 
 S? 
 
 M 
 
 e2 
 
 M 
 
 M 
 
 
 1 
 
 Pn , 
 
 
 
 IN 
 
 ' 
 
 ^3 ^ 
 
 * 
 
 
 
 
 T 1 
 
 1 
 
 W 
 
 | 
 
 | 
 
 5 
 
 2 
 
 g2 
 
 g 
 
 ifl 
 
 | 
 
 I 
 
 PH 
 
 
 
 
 (N 
 
 71 
 
 J3 M 
 
 88 
 
 H 
 
 
 
 EH i 
 
 * 
 
 | 
 
 | 
 
 ip 
 
 M 
 
 go 
 
 P 
 
 S 
 
 I 
 
 | 
 
 r. 
 
 
 
 
 
 N 
 
 n 
 
 ^3"* 
 
 ^ 
 
 ^~* 
 
 
 
 
 
 
 
 
 co 
 
 
 . 
 
 
 
 
 
 PH 
 
 
 IN 
 
 1 
 
 I 
 
 ^? 
 
 
 a 
 
 
 3 
 
 
 1 
 
 
 
 
 
 
 IN 
 
 
 
 
 
 
 
 | 
 
 
 in 
 
 (0 
 
 g 
 
 S 
 
 ^- 
 
 ia 
 
 a ^ 
 
 m 
 
 S 
 
 <N 
 
 9 
 
 CO 
 
 
 
 IN 
 
 M 
 
 
 * 
 
 J3 ^ 
 
 
 
 
 OS | 
 
 1 
 
 M 
 
 10 
 IN 
 
 IN 
 
 S 
 
 n 
 
 12 
 
 T' 
 
 S 
 
 H 
 
 t- 
 
 9 
 
 00 
 00 
 
 0} 
 
 1 
 
 1 
 
 J5 
 
 te 
 
 ag 
 
 - 
 
 
 
 
 
 p 
 
 g 
 
 IN 
 
 - 
 
 H 
 
 C-I 
 
 * 
 
 ia 
 
 
 *" 
 
 
 QQ 
 
 
 10 
 
 | 
 
 ^ 
 
 o 
 
 . 
 
 -* 
 
 
 
 00 
 
 lO 
 
 
 
 
 9 
 
 
 - 
 
 
 
 ^^ 
 
 S 
 
 S 
 
 CO 
 
 00 
 
 'S 
 
 
 
 
 <* 
 
 t- 
 
 H 
 
 us 
 
 
 
 s k 
 
 o 
 
 1 
 
 
 "" 
 
 1 
 
 9 s 
 
 2 
 
 5 
 
 1^ 
 
 3 
 
 
 03 
 
 at 
 
 W 
 
 
 
 
 | 
 
 S 
 
 10 
 
 -r 
 
 a^ 
 
 
 
 
 CO 
 
 ? 
 
 S 
 
 i 
 
 
 
 -< 
 
 ^ 
 
 74 
 
 ^5 M 
 
 u 
 
 l ~ l 
 
 00 
 
 at 
 
 o & 
 
 
 
 CO 
 
 j 
 
 * 
 
 ia 
 
 H 
 
 la M 
 
 CO 
 
 s s 
 
 co 
 
 I 
 
 OT 
 
 
 ^ 
 
 a ^i 
 
 M 
 
 -f 
 
 i 
 
 10 
 
 o i 
 
 
 
 H 
 
 ^ 
 
 N 
 
 ^3 w 
 
 ** 
 
 
 *" 
 
 
 03 
 
 
 
 
 OJ 
 
 CO 
 
 
 a* N 
 
 to 
 
 -f 
 
 (N 
 
 <N 
 
 
 
 
 " 
 
 i 
 
 
 
 ^ 
 
 A^ 
 
 (~ 
 
 
 CO 
 
 01 
 
 
 
 1 
 "S 
 
 I 
 
 Volts when current is) 
 zero ; i.e. open circuit > 
 just before starting ... ) 
 
 Volts when fairly steady ) 
 almost immediately ( 
 after switching on f 
 current ) 
 
 Final volts at end of) 
 test ; current passing 5 
 
 Amperes constant at ... 
 
 Total time of discharge ) 
 in hours and minutes \ 
 
 Ampere-hours 
 
 Watt-hours 
 
 Work efficiency 
 
 Quantity efficiency 
 
144 ELECTRICAL TRACTION. 
 
 ampere-hours and watt-hours in terms of total time of test 
 before the limiting conditions are reached. These results, 
 in each type of cell, can be expressed in terms of ampere- 
 hours and watt-hours per pound of positive plate, or per 
 square foot of superficial area. 
 
 It is interesting to note the relation between the weight 
 of a battery of cells and the distance such a battery would 
 propel a tramcar when discharged at a given rate. Take, 
 as a basis, that one kilo-watt-hour is required to drive a 
 tramcar one mile at the average rate of seven miles an hour. 
 Take the case in which a cell discharges 110 watt-hours 
 
 in 2 -25 hours, the average rate of discharge is o-2 r x 1000 
 
 kilo-watt-hours per hour. In 2*25 hours the whole battery 
 must discharge 16 Board of Trade units, the car travelling 
 16 miles ; therefore, if the weight per cell be 30 Ibs., the 
 weight in tons of the battery would be 1'95. The above is 
 only approximate, since the rate of discharge at starting 
 may be three times what it is when the car is running at 
 its maximum speed, and the discharge in watt-hours does 
 not vary as the total time. The number of such cells would 
 be 145, and the volume they would occupy at 0*16 cubic 
 foot per cell, would be about 23 cubic feet. 
 
 If reference be made to the Locomotives on Highways 
 Act, 1806 (see Appendix, p. 240), it will be seen that a 
 " light locomotive " means a vehicle propelled by mechanical 
 power which is under three tons in weight unladen, that is, 
 without storage cells, if these be used for propulsion. If 
 the weight unladen is 1~ ton, and does not exceed two 
 tons, the locomotive shall not be driven at a greater speed 
 than eight miles an hour ; or, if the weight exceeds two 
 tons, five miles an hour. 
 
 Take a cell weighing about 30 Ibs., and capable of dis- 
 charging 100 watt-hours in \\ hour, that is, at a rate of 
 
 y-^7 watt-hours per hour. Assume that to drive a " light 
 
STORAGE-CELLS. 145 
 
 locomotive," a battery of storage cells must discharge at the 
 average rate of 3 H.P. Then the number of cells in 
 the battery would be 28, and the weight 840 Ibs., or 0'375 
 ton. Thess cells would occupy a floor space of about six 
 square feet, if placed in one layer. 
 
 The Use of Storage-Cells on the Car. (ft) One of the 
 most interesting applications of storage-cells to traction 
 purposes is the combination of cells on the car with over- 
 head trolley conductor, as used at Hanover, or with the 
 slotted conduit as at Dresden. In crowded thoroughfares 
 there is an objection to the overhead conductor, and of 
 course one can combine the trolley-wire, which can be used 
 outside the town, with a surface-contact system, slotted 
 conduit system, or storage-cells. The change over from 
 the trolley-wire to surface-contact is very convenient ; not 
 so the slotted conduit and trolley-wire, since the " plough," 
 as it is termed, or the instrument used to collect the energy 
 from the live underground conductor, has to be placed in 
 and taken out when entering and leaving the sections ; 
 although with proper management this can be carried out 
 in about a quarter of a minute. The storage-cell is also 
 convenient, since it can be made of sufficient capacity to 
 take the car between termini and trolley-wires and gets 
 automatically charged from the trolley-wire, as in the case 
 of Hanover. 
 
 Hanover. The tramway system has a length of about 
 32 miles on which electric traction is used, and 17 on which 
 horse traction is used. The standard gauge is employed, 
 and the maximum gradient is 3*5 per cent., which is favour- 
 able for storage-cell traction. There are 92 cars, of which 
 29 are motor-cars for overhead trolley-wire alone, and G3 
 for either trolley-wire or storage-cells. The company has 
 its own station, which has an output of GOO kilo-watts, 
 including reserve. The kilo-watts per mile of track are 
 10'G, and per motor-car G'5. According to a report 
 recently issued by the directors of this system the storage- 
 
146 ELECTRICAL TRACTION. 
 
 cell is held in great favour. The cost of maintenance has 
 been carefully determined, and comes out on the average 
 at 0*1 5rf. per car mile for the year 189C. But allowing 
 for future renewal of plates this figure may be exceeded by 
 50 per cent. 
 
 On the basis of a storage-cell car mileage of 34,000 to 
 37,000 miles per annum, it is asserted that storage-cell 
 traction costs only 0'4^. more than any other system. 
 Sixteen and a half miles of overhead line has been saved 
 by the use of cells, representing about 32,500 : taking in 
 addition the maintenance of the overhead line at from 
 0'037rf. to ! 063r/. per car mile, this mixed system only 
 costs 0'2rf. more per car mile than simple overhead. 
 
 There are 1 10 storage-cell cars being built, and the battery 
 system is to replace the overhead system on all lines. 
 During the first half of 1896 the generating costs averaged 
 0*66tf. per kilo-watt hour, and during the second half 0'59^. 
 per kilo-watt hour. During November and December the 
 average was 0*54^., the improvement being attributed to 
 the increased use of storage-cells. The entire working 
 cost amounted to G8 per cent, of the year's receipts, giving 
 a net profit of 15,000, after allowing 14,000 for deprecia- 
 tion, from which a dividend of 5 per cent, was declared. 
 
 The Use of Storage-Cells in Sub-stations. The Ham- 
 burg lighting and traction system employs batteries in 
 sub-stations for the purpose of reducing the pressure of 
 500 volts, at which electrical energy is transmitted to them, 
 to the pressure of about 125 volts on each side of the 
 three-wire system of distribution for lighting purposes. 
 Each sub-station contains two motor-generator sets, and a 
 set of storage-cells, giving a potential difference of about 
 250 volts. The motors are each wound for 1 25 volts ; one 
 motor has one of its brushes connected to the 500-volt 
 negative, and the other motor has one of its brushes 
 connected to the 500-volt positive main. The other brush 
 on each motor can be connected to the outer of the 
 
STORAGE-CELLS IN SUB-STATIONS. \M 
 
 three-wire system, the cells for charging, or one pole of the 
 generator to which it is coupled. The other two poles 
 of the generators are connected together, and to the centre 
 of the battery and the centre of the three-wire system. 
 Two instances in which storage-cells are used in sub- 
 stations in a pure traction scheme are those of Douglas 
 and Laxey, and Leeds. 
 
 Leeds. The two sub-stations in this system are placed 
 respectively near each end of the line (see Fig. 22), and 
 contain the Chloride Electrical Storage Syndicate's cells. 
 The Roundhay battery, which is on a level considerably 
 above the power-house, contains 250 Chloride R. 15 type 
 cells, having 160 ampere-hours capacity when discharged 
 to an average of 1-8 volts per cell, at the rate of 1GO 
 amperes. The Kirkstall sub-station contains the same 
 number of cells of the Chloride S. R. 11 type; but their 
 capacity is smaller, namely, 110 ampere-hours. The charging 
 and discharging of cells is carried out automatically by 
 Dr. J. Hopkin son's switches, which we describe. In each 
 station there are two automatic switches, one for charging 
 cells and automatically cutting out the regulating cells as 
 they become fully charged ; the other for keeping the 
 potential difference between the overhead conductor and 
 the rail constant at 500 volts. A front view of the dis- 
 charge switch is given on the right-hand side of Fig. 50 ; 
 a back view is given in Fig. 57, which differs a little from 
 the actual switch. Fig. 58 gives a complete diagram of 
 connections for discharge. 
 
 The switch proper consists of a circular series of solid 
 gun-metal blocks numbered to 51, between each of which 
 is placed another series insulated from the first, the whole 
 being mounted upon a slate base. Contact is made between 
 any one of these blocks, and the gun-metal ring, R, which 
 forms the terminal of the switch, by means of two metal 
 arms carried by a central spindle, and bearing respectively 
 upon the ring and the contact block. A second ring, R l5 
 
148 
 
 ELECTRICAL TRACTION. 
 
 has a subsidiary brush carried by the main brush, but 
 insulated therefrom, bearing upon it and the circular series 
 
 of blocks, but does not touch R. Between R and R x a 
 polarizing resistance is placed, the object of which is to 
 
AUTOMATIC BATTERY SWITCHES AT LEEDS. 149 
 
 carry the discharge current whilst the switch is passing 
 from one block to the next, so that the current shall not 
 be interrupted. This polarizing resistance consists of lead 
 plates in water, and there are two elements in series 
 between the rings. This switch is geared to an electro- 
 motor, M, whose direction of rotation, as well as time of 
 operation, is controlled by a relay, working in conjunction 
 with what is termed a " resetting-drum," D, which is 
 mounted on a spindle of the main switch, and rotates with 
 it. The motor M is a small two-pole machine, with drum 
 
 FIG. 57. 
 
 armature ; its field magnets are permanently excited by the 
 current from one storage-cell preferably the end one or 
 nearly so, since these are not so much discharged as the 
 others of the battery. The motor armature is excited by 
 a positive or negative current, according to the direction 
 of rotation, from say four or five cells, the number depend- 
 ing upon the speed of operation required and the resistance 
 of the connecting wires between the switch and cells. 
 The object of the relay is to so control the switch that the 
 
AUTOMATIC BATTERY SWITCHES AT LEEDS. 151 
 
 potential difference between the trolley-wire and the rails 
 is constant at 500 volts, within about 1-| per cent, on either 
 side. It consists of two small electro-magnets, suspended 
 by springs to prevent vibration (Fig. 59), the protect- 
 ing case not being shown. Each of these magnets has a 
 
 FIG. 59. 
 
 small soft iron armature riveted to an aluminium plate, 
 which rests normally against three pointed screws on the 
 cone, slot, and plane principle, so that once these screws are 
 adjusted, the armature, when resting against them, has a 
 perfectly defined geometrical position with regard to the 
 poles of the magnet. These screws are supported by means 
 of a brass box fixed to the bottom of the magnet ; and 
 in the right-hand relay magnet they are screwed into the 
 bottom of the box, whilst in the left-hand relay they are 
 screwed into the top of the box. The right-hand armature 
 
152 ELECTRICAL TRACTION. 
 
 rests against its supports by gravity, whilst the left-hand 
 armature has to be held against its supports by the mag- 
 netism of the magnet. Since there are three cells between 
 adjacent blocks on the discharge-switch, it follows that three 
 times the maximum potential difference of one cell, say 7*8 
 volts when fully charged, and the charging current passing 
 through it, .is the minimum total variation allowable on 
 both sides of 500 volts, beyond which the switch must act. 
 For instance, if the potential difference between the line 
 and return rises above 504 or falls below 49G volts, the 
 switch must be operated so as to add or subtract cells, 
 until the required 500 volts within the limits mentioned are 
 obtained. 
 
 In Fig. 58, r t , r. 2 are the magnet windings of the relay 
 just mentioned, and have resistance (580 ohms ; r sj r 4 
 are extra manganese resistance coils, of about 11,000 
 ohms each. The junction between i\ and i\ is permanently 
 connected to the rail return, #. The junction between r 2 and 
 r ;! is permanently connected to the trolley-wire, y. Any 
 variation of the potential difference between x and y will 
 produce a corresponding variation in the currents, in the 
 parallel circuits of the relays. The resetting drum, D, is of 
 ebonite, and carries on its surface a copper strip with pro- 
 jections,^!,^, and has bearing upon its surface three brushes, 
 Nos. 1, 2, and 3 respectively. When the switch is central 
 on a block connected to the junction between two cells, 
 the resetting-drum is positioned relatively to the three 
 brushes, as shown in Fig. 58. Brush No. 2 is connected to 
 the junction between r a and r ;? ; brush No. 1 to the resist- 
 ance r B , of about 10,000 ohms, which is connected to the 
 junction of r 2 , r s ; brush No. 3 to the junction of r 29 
 r 4 . If the 'drum D turns in either direction from the 
 position shown, it either short circuits the relay-coil r 2 , 
 or increases the current in r l9 since it places r s , r 5 in parallel 
 with one another. But this is for a short interval only, 
 Pz are not very wide. The armatures, A,, A 2 , are 
 
AUTOMATIC BATTERY SWITCHES AT LEEDS. 153 
 
 shown in their normal positions; that is, the volts being 
 500, the magnetizing currents are such that A! is held up 
 against its supports, ct l9 a. 2 , the third support not being shown 
 for the sake of clearness ; whilst the magnetizing current 
 is not great enough to lift the armature A 2 from its sup- 
 ports, <7 a , 4 , the third support not being shown. Suppose the 
 potential difference between x and y falls below 496 volts, 
 the adjustment is such that the armature A! drops, thereby 
 completing the circuit between contacts b 19 b 2 . The current 
 can now pass from cell No. 1 through a contact-maker, fa, 
 if closed, to # x ; thence through the aluminium plate to # 2 ; 
 thence through the armature of motor M through r 7 to the 
 junction between cells Nos. 3 and 4. This current first 
 operates an electro-magnet, of which r 7 is the winding, and 
 removes the brake-block from a light aluminium disc-wheel, 
 which terminates a train of wheels in gear with the main 
 switch, and therefore with the motor M. It then operates 
 the motor in such a direction that the main switch adds 
 cells to the circuit, thereby increasing its potential dif- 
 ference. 
 
 When the switch has just reached the next contact- 
 block, and has therefore added three cells to the circuit, 
 the slip, p l9 makes contact with brush No. 1, and raises 
 the armature A x against its supports. If now the 
 potential difference be above 49 G volts, this armature will 
 remain against its supports, and the motor M is almost 
 immediately braked, since the magnet r, is not excited, 
 contact being broken at 1> 19 # 2 . After p has come into 
 contact with the brush, the switch has still a little distance 
 to travel before it is directly upon the block. To ensure 
 this the brake is not allowed to grip the aluminium disc- 
 wheel until the switch is in the middle position, and this 
 is effected by having, in addition to the resetting-drum, a 
 gun-metal disc with slots in it. Against this disc bears a 
 set-screw, which is mounted on an extension of the brake- 
 lever, and it is only when this set-screw comes opposite one 
 
154 ELECTRICAL TRACTION. 
 
 of the slots that the brake is applied, although the current 
 in the magnet winding r 7 has ceased, so that what happens 
 is this : When the armature is reset and interrupts the 
 current, the mechanism, in virtue of its momentum, is 
 able to travel forward until the set-screw on the extension 
 of the brake-lever comes opposite a slot in the disc. The 
 brake then presses forward on the aluminium wheel and 
 stops the switch. If this device were not used, there is 
 just the possibility that the switch would either overshoot 
 the mark or the brake would pull up the switch before the 
 little brush had again severed contact with the slip p^ 
 If the switch arrives at contact No. 0, and the potential 
 difference is still too small, the motor cannot act, since 
 #! is automatically broken by the switch itself. Take the 
 case in which a sudden removal of load increases the 
 potential difference between x and y ; the armature A 2 is 
 immediately raised, and completes the contact between b s 
 and ?; 4 . The current can now pass from the junction 
 between cells Nos. 3 and 4 through r r , through the motor- 
 armature in the reverse direction, through r 6 , which is a 
 coil wound on the relay magnet, in such a direction that 
 its ampere-turns augment the magnetizing force already 
 existing, from /> 3 to 1 4 through 7r 2 to the junction between 
 cells Nos. 7 and 8. The motor now rotates in such a direc- 
 tion as to subtract cells from the circuit. When the switch 
 reaches the next block, the relay-coil r. 2 is short circuited ; 
 the armature A 2 falls, and if the potential difference 
 between x and y is below 504 volts, it will remain there ; 
 that is, the switch will come to rest, the brake acting as 
 before. The time taken by the switch in passing from one 
 block to the next in either direction can be varied, and 
 is generally about l~ seconds. In traction work sudden 
 variations of potential difference between x and y have to 
 be dealt with, and consequently the switch must be actuated 
 quickly. The brake is therefore essential with high speeds, 
 in order that it may prevent the switch from overrunning, 
 
AUTOMATIC BATTERY SWITCHES AT LEEDS 155 
 
 owing to the momentum of the moving parts. The coil r 6 
 is inserted in the armature circuit, in order to prevent the 
 possibility of the armatures Aj and A 2 making simultaneous 
 contact with b l9 b. 2 ; # 3 , # 4 , respectively, and this might 
 happen if, when the switch has once started in a given 
 direction, the potential difference between x and y be 
 suddenly varied, so as to tend to reverse the direction of rota- 
 tion of the switch before the next contact block be reached. 
 The advantage of this form of relay is, that when once the 
 armature begins to move from its normal position, the rapid 
 decrease or increase of the air-space allows it to make sure 
 contact with its supports, which are platinum-tipped. But 
 it requires considerable force to bring the armature back 
 to its normal position, and this is controlled by the switch 
 itself. 
 
 The charging switch is similar in construction to the 
 discharging switch, just described, and is shown in Figs. 
 56 and 57. Its object is to prevent the overcharging 
 of the regulating cells, and this it effects by subtracting 
 cells from the circuit as they become fully charged. Fig. 60 
 gives a diagram of connections. M is the electromotor, 
 which operates the switch to which it is geared ; AJL and A. 2 , 
 as before, are the armatures of two relay magnets, of which 
 r.,, r 4 are the magnetizing coils. The armature AI falls 
 from its normal position and connects ^ and & 2 thereby 
 closing the armature circuit. This occurs if by chance the 
 switch gets into such a position that the potential difference 
 between the ring, R, and the rail return falls below 500 
 volts ; the object being to avoid the charging-current 
 passing through the cell or cells in a reverse direction 
 before entering the discharge circuit. The armature Ao 
 is normally upon supports, <7 3 , # 4 , and the magnetizing coil, 
 ? 4, is automatically placed across the terminals of the end 
 three cells which are being charged, by means of two small 
 brushes, P I? c 2 , which rotate with the switch and bear upon 
 a series of smalt blocks, arranged in a circle, and connected 
 
I S 6 
 
 ELECTRICAL TRACTION. 
 
AUTOMATIC BATTERY SWITCHES AT LEEDS. 157 
 
 respectively to every third regulating-cell, as in the case of 
 the blocks to 51. If therefore the three end cells are 
 fully charged, the potential difference having risen to, say, 
 7*8 volts, the armature A 2 is lifted, with the result that the 
 switch moves in such a direction as to cut out three cells. 
 The brushes, c^ c 2 , are so arranged that they break contact 
 with the terminals of the three cells being cut out, just 
 before completing the operation, and then come to rest across 
 the next three cells lower down. At the moment of breaking 
 contact the armature A 2 drops, and if the next three cells 
 are not fully charged, it will rest on its supports until such 
 time as the potential difference has risen to about 7 '8 volts, 
 or whatever potential difference the armature is set to lift 
 at. The resetting drum, D, resets the armature A x by 
 placing resistance, r 5 , in parallel with i\. The magnetizing 
 coil, r 3 , is wound on the left-hand relay magnet, and has 
 for its object the prevention of the two armatures AI and 
 A 2 , making simultaneous contact with the stops, # T b 2 , h, ^4, 
 which would, of course, short-circuit the cells, Nos. to 6. 
 As in the discharge-switch, two contacts, fo, k%, are supplied, 
 so that in no case can the switch overrun the end blocks. 
 The main brush in this switch is the same as shown in 
 Fig. 58, and has been left out for the sake of clearness. 
 
 The Use of Storage-Cells on the Car for Lighting 
 Purposes. An instance of the use of cells on cars for 
 lighting only, is at Bristol. The power-house contains a 
 main set of storage-cells for the cars. These cells are of 
 the Chloride type, the main battery consisting of 55 cells, 
 each capable of discharging about 550 ampere-hours at a 
 six-hour rate, and charged by means of a motor-generator 
 from the 500 volt circuit. The lighting cells for the cars 
 consist per car of two sets of five cells each in boxes. The 
 great advantage of this method of lighting is that the light 
 is steady, and if the trolley wheel becomes detached from 
 the overhead conductor the supply does not fail. 
 
158 ELECTRICAL TRACTION. 
 
 CHAPTER VIII. 
 
 ALTERNATE AND DIRECT CURRENTS. 
 
 IT is well known that the transmission of power by alternate 
 currents can be carried out with great economy when the 
 distance is great. The chief reason for this is the ease with 
 which one can transform from one potential difference to 
 another by aid of the alternate current transformer. It is 
 possible, therefore, to generate currents in the station at 
 low potential with safety and to raise this potential many 
 fold for the line, which can consequently have a small cross- 
 sectional area, thereby reducing its first cost. At the 
 receiving end of the line, another transformer can be placed 
 to reduce the pressure again within safe limits. With 
 direct currents it might be said that the same thing can be 
 done, and with some truth ; but direct current dynamos 
 have not been built, and would be costly to build, if they 
 had to work at such potential differences as, say, 20,000 
 volts. The absence of commutators on the alternate current 
 generator is a great point in favour of alternate currents, 
 since with direct currents and high potentials this becomes 
 a large item in the first cost of the machine. 
 i In large tramway schemes, with the supply from one 
 power-house, the first cost of copper in the feeders becomes 
 great, and one may be justified in reducing this by means 
 of sub-stations containing transforming apparatus. No hard 
 and fast rules can be laid down, as each case must be treated 
 on its own merits, and the justification rests in part with 
 the engineer. In most tramway schemes in Great Britain 
 
ALTERNATE CURRENTS. 159 
 
 where the radius from power-house is of the order of three 
 or four miles, one may say that direct currents need not be 
 deviated from. On the other hand, when the distance gets 
 to eight or ten miles, the alternate current becomes of value. 
 The Central London Railway and Dublin Southern District 
 Tramway are good illustrations of the latter condition, and 
 here alternate currents are employed (see pp. 175, 188). 
 
 It will be well to divide this subject into two parts : first, 
 the use of alternate currents in feeder systems, transforma- 
 tion taking place at sub-stations, such transformation pro- 
 ducing direct currents for the line ; second, the use of 
 alternate currents in feeder systems, the transformation 
 producing alternate currents of smaller potential for the 
 line. 
 
 Alternate Currents for Feeder and Direct Currents for 
 Line. In practice one meets with one, two, and three- 
 phase alternate current machinery, and it will be well to 
 discuss briefly their relative advantages and disadvantages. 
 We have to consider the generation of the current or 
 currents, the transmission of energy by the aid of such 
 currents, and the motor which receives the energy in the 
 sub-station. With regard to generation it can be fairly 
 said that the two or three-phase generator for the size of 
 machine is capable of delivering a greater output than a 
 single-phase generator when what are termed " mixed phases " 
 are employed ; but in any case, the choice of one-, two-, or 
 three-phase is not influenced so much by considerations of 
 the generator as by other parts of the system. In either 
 case if the potentials be of the order of 5000 or 6000 volts, 
 stationary transformers would probably be employed ; this 
 is the case at Portland, Oregon, U.S.A. But at 3000 volts 
 between the conductors, the three-phase transmission at 
 Dublin (see p. 188) is effected without such transformers ; 
 that is to say, the three-phase currents are transmitted 
 directly from the generator to the motor at 3000 volts. 
 
 With regard to the line itself, it comes to laying either 
 
160 ELECTRICAL TRACTION. 
 
 a double or triple concentric cable or two or three overhead 
 conductors according as the system is single, or two-, or 
 three-phase. If the system be three-phase the volume of 
 copper required is three-fourths the volume on the single- 
 phase system, but of course there is the extra insulation of 
 one conductor. The advantage of the two- or three-phase 
 over the single-phase really lies in the satisfactory working 
 of large rotatory transformers, which receive such currents 
 and transform them into direct currents. Consider the 
 rotatory transformer ; this is a motor-generator, the armature 
 of which receives the one-, two-, or three-phase currents, and 
 also gives from an ordinary commutator a direct current at 
 the required potential. In such a machine, if the alternate 
 currents be single-phase, it is stated * that Professor Men- 
 garini's experience at Rome is all against such machines for 
 large sizes, since they are reported to spark so violently on 
 the direct current commutator, this being due to the vary- 
 ing rate of delivery of energy by the single-phase current, 
 whilst the rate of delivery by direct current is constant. 
 With the two- or three-phase transmission the case is different, 
 as is evidenced by the satisfactory working of the GOO 
 kilo-watt rotatory converters at Niagara and Portland. The 
 General Electric three-phase rotatory converters at Portland 
 are illustrated in Fig. 61. It has been pointed out by Mr. 
 Parshall that reaction of the armature is so small, that a 
 three-phase 500 kilo-watt rotatory converter will carry double 
 and treble this load without the least sparking at the brushes. 
 This is attributed to the more steady delivery of energy by 
 three-phase currents. It is well known that the rate of 
 delivery of work by a single-phase generator, and conversely 
 the receipt of work by a single-phase motor, fluctuates 
 continuously ; an inductive load, giving phase-difference 
 between potential and current, returns energy to the system 
 at each half period. With a two- or three-phase transmission 
 
 * See the Journal of the Institution of Electrical Engineers, vol. 
 xxvi. p. 408. 
 
ALTERNATE AND DIRECT CURRENTS. 
 
 161 
 
 the rate of delivery of work is constant if perfect symmetry 
 obtains. Mr. Parshall also quotes the case where, without 
 excitation, a GOO kilo-watt rotatory converter maintains 
 synchronism and perfect collection up to 900 kilo-watts. 
 According to the tests at Niagara, the efficiency of a well- 
 
 FIG. 61. 
 
 designed 600 kilo-watt converter is 96 per cent, on full, and 
 94 per cent, on half load. 
 
 Fig. 62 illustrates a British Thomson-Houston three- 
 phase synchronous motor or generator. In outward appear- 
 ance it differs little from an ordinary multi-polar direct 
 current generator or motor, except that the usual commutator 
 
1 62 
 
 ELECTRICAL TRACTION. 
 
ALTERNATE CURRENTS. 163 
 
 is supplanted by three plain metal rings on the shaft. The 
 radial poles in motor and generator are excited by direct 
 currents, and the generator armature consists of three 
 windings placed in slots in such manner that the angular 
 distance between them is one-third of the distance between 
 two poles. The inter-connection of these three coils is 
 perhaps best described by con- 
 sidering the methods of group- 
 ing three-phase conductors. If 
 we imagine a ring wound as in 
 Fig. 63, which is generally 
 termed the "star grouping," 
 with six coils, 1, 2, 3, 4, 5, (>, 
 equally spaced out and con- 
 nected, as shown in the dia- 
 gram, and revolving between 
 two poles of an electro-mag- 
 net, then, assuming uniform 
 rotation and uniform field distribution, we may say 
 
 that E.M.F. induced in coils Nos. 1 and 4 is A sin ~, 
 
 where A is the maximum ordinate, T is the periodic time 
 or the time of a complete revolution, and t is the time 
 corresponding to any position of the armature. If the 
 poles be placed right and left, obviously t = when the 
 coils 1 and 4 are top and bottom. The E.M.F. in 2 and 5 
 
 is A sin ( ^ + GO ) , and the E.M.F. in 3 and 6 is 
 
 A sin ( -^ 4. 120 J . The terminals a, I, and c are joined 
 
 together, and in some cases connected to the body of the 
 armature, as in the Lauffen experiments,* an,d the terminals 
 d, e,f are connected to three insulated metal rings on the 
 shaft, the brushes bearing on these forming the terminals 
 of the three-phase machine. "We see that the currents in 
 * See Thompson's " Polyphase Electric Currents," p. 107. 
 
164 ELECTRICAL TRACTION. 
 
 the line-wires proceeding from d, e, f are the same at any 
 moment as the currents in the respective coils 2 and 5, 
 o and 6, and 4 and 1, whilst the potential difference between 
 two lines equals \/3 of the potential difference of one pair 
 of conductors, the phase being altered by 30. If the 
 currents be in phase with the potentials that is, the load 
 be non-inductive, or equivalent to non-inductive the total 
 power equals twice the potential difference between any two 
 lines into the current in one line, provided everything is 
 symmetrical. If phase difference is introduced, the power 
 factor or ratio of true to apparent watts must be considered. 
 The inter-action of single-phase generators and motors is 
 
 explained in the Transactions of 
 the Royal Society, vol. clxxxvii. 
 pp. 229-252. 
 
 In the "mesh grouping," 
 shown in Fig. 04, the coils are 
 cross connected as before ; but 
 uf, M 9 ce are joined together 
 respectively. In this case the 
 current in the line is equal to 
 VI of the current in the arm- 
 ature conductors. The three 
 line-wires being connected to 
 the junctions af, M, ce, the potential difference between 
 any two lines is the same as the potential difference of the 
 armature conductor between such lines. The power reduces 
 as before, if current be in phase with potential, to twice the 
 potential difference between any two conductors into the 
 current in one line, provided everything is symmetrical. 
 The power factor must be considered when phase-difference 
 between curreat and potential exists. In actual practice 
 both these methods of grouping conductors are used, and 
 sometimes combinations of the two. One would expect a 
 smaller armature reaction for the mesh grouping for a given 
 line current, but the internal resistance of the armature 
 
ALTERNATE CURRENTS. 165 
 
 for the same volume of copper would be greater. The 
 treatment of two-phase currents is similar to the above. 
 The winding of one-, two-, and three-phase armatures is well 
 set forth in Messrs. Parshall and Hobart's Armature Windings. 
 Alternate Currents for Feeder and Line. The question 
 as to whether an alternate current received at the feeding- 
 point at a high potential should be transformed into another 
 alternate current at a lower potential, say 500 volts, and as 
 such to be transferred directly to the line and thence to the 
 motors on the cars, naturally depends upon the suitability 
 of such motors for traction work. We have in this case the 
 choice of using either a constantly running synchronous 
 motor on the car, in which case a suitable friction clutch 
 must be used to apply the power to the wheels, or a non- 
 synchronous motor. In the first case, the motor would no 
 doubt work well, and a set of storage-cells on the car would 
 serve for exciting the dynamo, as well as for lighting, but the 
 gear is the difficulty. The second case is the one actually 
 adopted in practice, and then the two- or three-phase system 
 is undoubtedly better. The great drawback of the alternating 
 current in the line itself, if this be single-phase, is the want 
 of a single-phase motor, which has all the good points we 
 find in the direct current, and two- or three-phase motors. 
 If the two- or three-phase system be employed, the motor is 
 satisfactory ; but two overhead wires are a necessity, and 
 this introduces complications at crossings. With the slotted 
 conduit system the two conductors and the rail could be 
 used as the three wires in a two- or three-phase transmission, 
 and since all the conductors are near together and parallel, 
 there should not be so much disturbance as is experienced 
 Avhen the overhead conductors are used, since these are 
 near to telegraph and telephone lines. The alternate cur- 
 rent has the advantage of comparatively small action on 
 metal underground, which, as we have seen, has been the 
 means of such stringent rules on the part of the Board of 
 Trade in the case of direct currents and rail returns (p. 53). 
 
1 66 ELECTRICAL TRACTION. 
 
 The Author is indebted to Messrs. Brown, Boveri & Co. for 
 the following particulars of their system at Lugano, Switzer- 
 land. This tramway is an instance of three-phase transmis- 
 sion for about 1\ miles from the power-house at Maroggia, 
 where the power is derived from a 300 H.P. turbine directly 
 coupled to two 150 H.P. Brown three-phase generators. 
 The speed is GOO revolutions per minute, and the potential 
 difference is 5500 volts, the frequency being 40 periods per 
 second. These generators are well known, and are of the 
 inductor type. Fig. Go shows one of these generators with 
 the top half removed. It will be seen that the magnets 
 which rotate consist of four poles on one side of the magnet 
 winding, with four poles on the other, such poles being 
 connected together near the shaft by a concentric circular 
 yoke, on which the exciting coil is wound. The poles on 
 either side have an angular distance between them of 45, 
 and since there are four like poles on one side there are four 
 complete periods per revolution ; that is, since the revolu- 
 tions per minute are 600, the frequency is x = 40 
 
 60 
 
 complete periods per second, as stated. The stationary part 
 of this generator consists of soft iron sheet stampings in the 
 form of rings, which have an internal diameter large 
 enough to clear the poles of the exciting magnets. The 
 conductors are threaded through a series of holes punched 
 out in a circle near the inner edge of the rings ; the three 
 phases are produced by dividing the distance between two 
 poles, that is, an angle of 45, into groups having an angle 
 of 15 between them. The electro-motive forces induced 
 in such groups of conductors must therefore have phase 
 differences of 60 ; that is to say, similar electric events, if 
 there be symmetry, happen in the three groups of con- 
 ductors, but at times differing by -gToth second. These 
 generators are each excited by a small dynamo direct 
 coupled to the shaft. 
 
 There are several motors connected to the mains 
 
ALTERNATE CURRENTS AT LUGANO. 
 
 167 
 
 along the line. A transformer station is placed about a 
 quarter of a mile from Lugano, and reduces the pressure to 
 400 volts, the secondary coils of the transformer being 
 
 connected to the two overhead trolley-wires and the rails at 
 this point. A bare copper wire, 7 mm. diameter, is laid 
 in the earth to the junction-point at Lugano, to reduce 
 
1 68 ELECTRICAL TRACTION. 
 
 the drop of potential difference in the rails. The aggregate 
 length of this line is about three miles. The gauge is one 
 metre ; the steepest gradient is 6 per cent., and the 
 sharpest curve has a radius of 50 feet. The trolley-wires are 
 6 mm. diameter. They are 10 inches apart and 18 feet 
 above the rail level. It is stated that no difficulty has 
 been experienced with the working of the line. There are 
 four cars running, each capable of seating 24 persons, and 
 provided with one 20 H.P. three-phase 12 pole motor, 
 connected through single reduction gear to the axle, the 
 ratio being 1 : 4. The maximum speed of the motor is 
 400 revolutions per minute, corresponding to about 
 miles per hour of the car. The two trolley arms on eacli 
 car are placed one behind the other at a distance of one 
 yard. The motor is of the inclosed type, with an opening 
 at top for the purpose of examination of the collector 
 rings on the armature and brushes. The stationary portion 
 of this motor receives the three-phase currents from the 
 two overhead trolley-wires and the rail, and for ordinary 
 speeds and starting such currents produce at any moment 
 12 poles on the inner surface of this part, which, as in the 
 generator, consists of a number of soft iron ring- 
 shaped stampings, with holes punched out near the inner 
 diameter. The conductors are parallel with the axis, and 
 are threaded through such holes. It follows that since 
 there are 12 poles, there are six complete periods per 
 revolution ; and since the frequency is 40, the revolutions 
 per minute, if perfect synchronism were obtained, would be 
 400. 
 
 The subject of difference of frequency between armature 
 and rotatory magnetic field is full of interest, and such 
 difference depends upon the load on the motor, the magnetic 
 field, and the resistance of the armature conductors. In 
 working these motors, and starting with the armature run- 
 ning light when it is very nearly in synchronism with the 
 * See Electrician, August 28, 1896. 
 
ALTERNATE CURRENTS. 169 
 
 rotating field, the torque gradually rises as the load is 
 applied until it reaches a maximum with a given difference 
 between the frequencies of the armature and field ; any 
 further load brings about a diminished torque, and insta- 
 bility sets in. We have already seen that the maximum 
 power the motor has to develop may be three times the 
 average, and it is, therefore, important to see that the 
 average power is well within the maximum, and of course 
 the motor should work with high efficiency at its average 
 rate. It has been pointed out by Mr. Brown * that under 
 normal working conditions the field of these motors is com- 
 paratively weak, and the torque can be considerably in- 
 creased, since it depends upon the current in the armature 
 and the field by changing the connections from star to 
 mesh-grouping, and without altering the applied potential 
 difference. The armature in these motors consists of a 
 number of soft iron discs threaded on the shaft, and secured 
 to it by a key and end plates. These discs have a series 
 of holes punched out in a circle near the periphery, through 
 which the conductors are threaded. In motors of this size 
 (20 H.P.) such armatures require extra resistance in the 
 circuit of their conductors for starting purposes ; but in 
 small sizes, as is well known, the conductors are simply 
 connected together at each end by soldering them into solid 
 copper rings, such conductors and rings resembling a 
 squirrel's cage. In the larger motors, however, the armature 
 conductors, if wound according to the star grouping (Fig. 
 C3), have their three respective ends connected to three 
 insulated metal rings on the shaft. By means of brushes 
 these rings are connected with an external rheostat, which 
 is under the control of the motor-man ; such rheostat is 
 shown in Fig. GG, and consists of resistances which can be 
 varied by turning round the vertical spindle in a similar 
 manner to an ordinary controller (Figs. G and 7). The 
 truck complete with motor and rheostat is shown in Fig. G7. 
 * See Journal Institution of Electrical Engineers^ vol. xxvi. p. 425, 
 
1 70 ELECTRICAL TRACTION. 
 
 The motor-man has two switches to handle : one of these 
 makes the circuit and regulates the resistance of the arma- 
 ture circuit for starting and stopping ; the other serves for 
 totally breaking the supply circuit, or reversing the direction 
 of current when the direction of the motor requires 
 reversal. For higher speeds outside the town the number 
 of poles is reduced on the stationary part of the motor, 
 thereby increasing the speed ; also, for increased torque, 
 
 FIG. 6G. 
 
 the connections are changed from star grouping to mesh 
 grouping. 
 
 This system has been in operation at Lugano since 
 December, 1895, with satisfactory results, and the novelty 
 of it lies in the use of the induction motor on the cars. 
 We have seen that the three-phase transmission is used in 
 the Dublin tramway feeder system, and there are several 
 instances of long distance transmission for traction work by 
 three-phase currents, notably that at Portland, Oregon, 
 
ALTERNATE CURRENTS AT LUGANO. 171 
 
172 ELECTRICAL TRACTION 1 . 
 
 U.S.A., but in these direct currents are used for motors on 
 the cars. 
 
 In all cases of high tension transmission it is necessary 
 to transform down to a voltage of about 500, in compliance 
 with the Board of Trade regulations and for the public 
 safety. In the transformation, running machinery must be 
 employed if direct currents are to be produced, as we have 
 seen at Dublin ; whereas, if alternate currents be employed 
 on the car itself, the transformation can be effected by a 
 stationary transformer with no moving parts, and requiring 
 little or no attention. 
 
 Messrs. Brown, Bovri & Co. inform the Author that they 
 are now (June, 1897) occupied with the equipment of the 
 following lines with three-phase traction : 
 
 1. The Gorner-Grat railway near Zerrnatt. This is a 
 rack railway with a maximum gradient of 20 per cent. 
 The length is about six miles, and the locomotives will be 
 equipped with two motors of 90 H.P. each. Each train is 
 to accommodate 110 passengers. 
 
 2. Engleberg railway. This is a line 14 miles long, 
 having but slight gradients for the greater part of its 
 length, and here worked by adhesion. For about two miles 
 there are gradients of 25 per cent., and here a special electric 
 locomotive is attached, working on a rack. 
 
 These lines will be at work in the course of about a year. 
 
 This firm has just concluded the contract for the equip- 
 ment of the line from Burgdorf to Thun. This is a 
 standard gauge line, and will carry traffic at ordinary rail- 
 way speeds ; it is to be running in about two years. We 
 see, therefore, that three-phase currents, as applied to the 
 motors on the cars, are making considerable headway. 
 
 The Central London Railway. The electric traction 
 system shortly to be brought into operation on the Central 
 London Railway is at the present time attracting much 
 attention by all persons interested in this work. The 
 contract has been secured by the British Thomson-Houston 
 
THE CENTRAL LONDON RAILWAY. 
 
 173 
 
 Company. The engineers to this railway did not settle 
 the scheme to be adopted, but left this to the competing 
 firms to propose for themselves. The following specification 
 was issued, giving the requirements : 
 
 " CENTRAL LONDON RAILWAY. 
 
 " Electrical Traction Plant. Specification of General 
 Requirements. 
 
 " 1. The work intended to be included in the specification 
 includes the finding of all materials, tools, and labour for 
 the complete manufacture and installation, and, except as 
 regards fair wear and tear, the maintenance for a period of 
 twelve months after the opening of the railway, of all the 
 electrical plant and appliances necessary for working a 
 2-|-:mmutes' service in each direction, of trains of seven 
 carriages, having a seating capacity of 336 persons, and 
 weighing 105 tons loaded, exclusive of locomotive, over 
 the Central London Railway, at an average speed of four- 
 teen miles an hour, including stoppages at stations. 
 
 " 2. The said railway consists mainly of two tubular 
 tunnels of 11 feet 6 inches internal diameter, lined with 
 cast-iron, and of enlarged tunnels of similar construction 
 at fourteen stations, and at certain cross-over roads. 
 
 " 3. The position of the stations and the distances apart, 
 measured from centre to centre of the platforms, are 
 proximately as follows : 
 
 Shepherd's Bush to Holland Park 
 
 Holland Park to Notting-hill Gate 
 
 Notting-hill Gate to Queen's Koad 
 
 Queen's Koad to Westbourne Park 
 
 Westbourne Park to Marble Arch 
 
 Marble Arch to Davies Street 
 
 Davies Street to Oxford Circle 
 
 Oxford Circle to Tottenham Court Road 
 
 Tottenham Court Road to British Museum 
 
 British Museum to Chancery Lane 
 
 Chancery Lane to Post Office 
 
 Post Office to Bank ... 
 
 Bank to Liverpool Street . ; . . . 
 
 Yards. 
 
 1,012 
 683 
 768 
 986 
 
 1,288 
 642 
 699 
 666 
 682 
 746 
 
 1,163 
 828 
 841 
 
174 ELECTRICAL TRACTION. 
 
 " 4. In addition to the preceding length of double line 
 of railway, there is a further length of 1036 yards at 
 Shepherd's Bush from the centre of the platform to the 
 end of the sidings in the depot, and at Liverpool Street 
 a length of 236 yards from the centre of the platforms 
 to the end of the sidings. The total length of continuous 
 railway over which electrical traction must be provided is 
 12,276 yards or 7*6 miles, exclusive of cross-over roads at 
 stations and sidings. 
 
 "5. The gradients of the line and the general details of 
 the railway, so far as they affect the subject-matter of this 
 specification, are shown on the drawings herewith. [These 
 drawings are not reproduced in this book.] 
 
 " 6. The contractor shall be deemed to have full know- 
 ledge of the Central London Railway Acts, and shall 
 perform all the obligations imposed on the railway com- 
 pany thereby. 
 
 " 7. All work and materials shall be of the highest class, 
 to the entire satisfaction of the engineers. 
 
 "8. The whole of the work shall be completed within 
 eighteen months of the acceptance of tender. 
 
 "9. For working the above service of trains, 32 electric 
 locomotives shall be provided of the strongest, simplest, 
 and most efficient type, without gearing or pitch chains, 
 and capable of taking the specified trains over the line afc 
 the required speed, and into the depot at Shepherd's Bush. 
 
 " 10. The generating plant shall be of ample capacity to 
 furnish power for the working of the railway with only 
 three-fourths of the same in operation, and the remaining 
 one-fourth in reserve. The boilers, high-pressure engines, 
 economizers, dynamos, main and return conductors, insu- 
 lators, and all other necessary appliances, shall be such as 
 will ensure the highest attainable economy and facility in 
 working and maintenance, and be least liable to break- 
 downs, so as to secure the safe and uninterrupted working 
 of the traffic at all times with comfort to the passengers, 
 
THE CENTRAL LONDON RAILWAY. 175 
 
 and freedom from injurious interference with any public 
 or private interests. 
 
 "11. The contract will include the construction of all 
 necessary connections with the coal sidings and mechanical 
 stokers in the boiler-house. It will also include the com- 
 plete fitting up of repairing shops, and the supply of all 
 tools necessary for keeping the whole of the work included 
 in this specification in efficient condition and repair. The 
 buildings, engine foundations, and other masonry works 
 form no part of this contract, but, as far as may be neces- 
 sary, will be constructed in accordance with the plans and 
 specifications furnished by the contractor. 
 
 " 12. The locomotive must be provided with compressed 
 air-brakes, but the air-compressed plant is not included in 
 the present contract. 
 
 " 13. An alternative tender must be submitted for two- 
 minutes' service." 
 
 The following is a reported description of the plant to be 
 employed on this railway, which will be installed by the 
 British Thomson-Houston Company. 
 
 Track. This consists of 100 Ib. steel rails on heavy 
 sleepers well ballasted : 85 per cent, of the power necessary 
 to accelerate the trains is saved by means of a 3 per cent, 
 rise and fall at each station (see p. 10). The working con- 
 ductor is in the form of a third rail supplied with ample 
 copper feeders, and means are provided at short intervals to 
 enable the operators in the signal-stations to cut out any 
 section desired. The schedule speed, including stops of 20 
 seconds at each of the fourteen stations, will be 14^ miles 
 per hour. 
 
 The Generating Station. This is above ground at 
 Shepherd's Bush, and will contain six 1300 H.P. horizontal 
 Corliss engines by E. P. Allis and Co., each directly con- 
 nected to a General Electric three-phase 5000 volt generator. 
 These are of the stationary armature type, with rotating 
 field, and are intended to give 850 kilo-watts at 5000 volts, 
 
176 ELECTRICAL TRACTION. 
 
 at 94 revolutions per minute, or a frequency of 25 periods 
 per second. The 5100 kilo- watts are to be transmitted 
 from the generators by heavy conductors to a switchboard, 
 from which feeders convey the three-phase current to the 
 sub-stations along the line. In these sub-stations static 
 transformers will drop the potential difference from 5000 
 to 330 volts, for which the alternating current side of the 
 rotatory converters are wound. The working potential 
 difference of the line is 500 volts. Of the four sub-stations 
 two will contain seven 300 kilo-watt rotatory converters, 
 similar in design to those at Niagara. Only six of these 
 transformers are to be in use at any one time, the seventh 
 being held in reserve. The remaining sub-stations will 
 contain one 900 kilo-watt converter and four 300 kilo-watt 
 transformers. All the transformers are to be of the " air- 
 blast " type ; the air, which circulates freely between the core 
 and windings, being driven through by means of a blower. 
 
 Locomotives. The number of locomotives to be con- 
 structed is 35. The motors, which are four in number 
 and directly connected to each axle, are of the G.E. 5G 
 type. They are to be worked on the series-parallel control 
 system, and are capable of exerting a draw-bar pull of 
 14,000 Ibs. Each locomotive weighs 35 tons. 
 
 Three-phase versus Three-wire for Electric Traction. 
 The recent discussion of Mr. Baylor's paper before the 
 Institution of Electrical Engineers,* brought forward some 
 very useful information with regard to the suitability of 
 the three-wire system for traction work, and a comparison 
 of it with that of the high potential three-phase feeder 
 systems just described. It appears that two large English 
 firms proposed the three-wire system for this scheme, and 
 the " Americans," as it was put (meaning, it is supposed, 
 the British Thomson-Houston and the Westinghouse Com- 
 panies), proposed the two- or three-phase systems for feeding 
 at high potential with sub-stations, similar to the Dublin 
 
 * See Journal of Institution of Electrical Engineers, vol. xxvi, 
 
THREE-PHASE AND THREE-WIRE SYSTEMS. 177 
 
 tramway. So far as the author is aware, there is no English 
 firm which has put down any traction scheme on the three- 
 wire system, and if this be so the system has not been 
 properly tested in the United Kingdom. Without actual 
 experience, it does seem from some points of view that 
 the three-wire system would be a good one to adopt. Pro- 
 fessor A. B. TV. Kennedy, before a select committee of the 
 House of Commons, in March, 1897, on the Brompton 
 and Piccadilly Circus Railway Bill, which only deals with 
 about 1200 H.P. as against five or six times this amount 
 in the Central London Railway, is reported * to have pro- 
 posed the three-wire system as being the most advantageous, 
 employing balancing transformers at at least two places on 
 the line. A balancing transformer in a three-wire direct 
 current system consists of two direct current machines, 
 either rigidly coupled together, or the two armature windings 
 are wound on the same core, one being placed between the 
 middle wire and one outer, and the other between the middle 
 wire and the other outer. If, then, the potential differences 
 between the middle wire and the outers are not equal, it 
 follows that, since the machines are the same size, the one 
 runs as a motor, driving the other as a generator, thereby 
 bringing about an adjustment. The process is reversible, 
 and, since the machines are shunt-wound, no reversal of 
 direction of rotation ensues when the change takes place. 
 The proposed Brompton and Piccadilly scheme has a length 
 of about 3476 yards, with five stations ; the railway would 
 be placed 50 or GO feet below the roadway, the service to be 
 a three-minutes' one throughout the day. The generating 
 station is to be placed at Chelsea Creek, which although 
 some distance from the terminii, is an otherwise eligible 
 site, giving as it does direct communication with the river. 
 
 Again, Mr. Snell, after Mr. Parshall's condemnation of 
 the three- wire system in the discussion before alluded to, 
 is still of opinion that the two- or three-phase system has 
 * See Electrician, 26th March, 1897. 
 
 N ' 
 
i/8 ELECTRICAL TRACTION. 
 
 not any practical advantage over the three-wire system. 
 Turning to" the American side of the question, we find 
 nothing but failure reported on the part of the three-wire 
 system for traction purposes. Mr. Parshall points out, 
 which is perfectly true, that the motor-man must keep his 
 proper time with the other locomotives, and therefore he 
 puts over the controller switch until this is attained, heedless 
 of the current he is consuming. If, therefore, for some 
 reason the potential difference applied to his locomotive 
 is small, he requires more current, and this, again, lowers 
 the potential difference. The effect, therefore, is cumu- 
 lative, and, naturally, more aggravated the fewer the 
 points of supply on either side of the system. The variation 
 of efficiency with applied potential difference is dealt with at 
 p. 30. In the case of electric lighting on the three-wire system 
 the conditions are different. A diminution of potential 
 difference in the mains has the effect of increasing the resis- 
 tance of the lamp filaments, which automatically cuts down 
 the current, so that to some extent the remedy is automatic. 
 Two notable instances of failure on the part of the three- 
 wire system are those of Portland (Oregon), and St. Louis, 
 although others are mentioned in the discussion above 
 referred to. At Portland, where the three-wire system was 
 at first used, the three-phase high potential system has been 
 substituted. The company supplies power for lighting and 
 traction ; the power is obtained from the Willamette Falls, 
 about fourteen miles from Portland. In the power-house 
 (Fig. 68) there are three 450 kilo-watt three-phase gene- 
 rators mounted on vertical shafts of Victor turbine wheels, 
 each supplying current at GOOO volts direct to the trans- 
 mission lines. In the sub-station are placed air-blast 
 transformers, which reduce the potential difference for 
 feeding the low tension lighting system at the city, and 
 also supplying current to two 400 kilo-watt rotatory con- 
 verters (Fig. 61) used to operate the street railway. This 
 scheme has been in operation since May, 1895, and has 
 
i8o ELECTRICAL TRACTION. 
 
 demonstrated the suitability of three-phase transmission 
 for railway work. At St. Louis, where there were several 
 hundred cars, the three-wire system, after a thorough trial, 
 Mr. Parshall states, was abandoned, " since it was found 
 that the extra cost and trouble in compensating from hour 
 to hour was of vastly more consequence than the extra 
 cost of copper." "With so many cars running on a network 
 of conductors, such as at St. Louis, one would have thought 
 that the conditions were favourable. Referring to the 
 Central London Railway, the same authority states that 
 there are to be operated between twenty and thirty trains, 
 each weighing about 130 tons, and to maintain the average 
 of fifteen miles per hour these trains must have at times a 
 speed of about thirty miles per hour, and must be accelerated 
 at least one foot per second per second. Mr. Parshall esti- 
 mates that each train may absorb 1000 amperes at times, 
 such figure being based upon actual test, the average output 
 per line not exceeding 3000 amperes. One train, therefore, 
 may absorb about one-third of the total output per line ; 
 and this introduces the use of such large balancing trans- 
 formers, that the three-wire system cannot compete with 
 the three-phase and rotatory feeder system. He estimates 
 that at least three balancing transformers of a capacity of 
 300 kilo-watts would be necessary, consistent with the Board 
 of Trade regulations as to drop of potential difference in 
 the line. 
 
 The Thames Ironworks Company proposed a balancing 
 transformer at each alternate station along the line ; that is, 
 six in all, each of a capacity of about one-fifth or one-sixth of 
 the whole output of the station that is, about 900 or 1000 
 kilo-watts each. At St. Louis it was found that the neutral 
 conductor had to have as large a cross-sectional area as 
 each of the outers, and under this condition the loss in the 
 neutral conductor and balancing transformers, under favour- 
 able conditions, was about equal to the loss of transforma- 
 tion in a well-designed multi-phase sub-station. According, 
 
THE CENTRAL LONDON RAILWAY. 181 
 
 then, to Mr. Parshall, it comes to a comparison of trans- 
 mitting energy at 5000 volts and 1000 ; and if this be so, 
 the high tension three-phase transmission would appear 
 to be the better. He goes on to point out that if the 
 potential difference be suddenly increased, the balancing 
 transformers must at once receive energy to accelerate them, 
 which is a bad condition, since the removal of load from the 
 locomotive should not entail a temporary transference of 
 the load to the balancing transformers. The remarks 
 already made in this book, at page 10, on the efficiency of 
 motors, show how such efficiency depends upon the potential 
 difference of supply, and consequently any system which 
 has a greatly fluctuating potential difference available at 
 the locomotive terminals must necessarily suffer on the 
 whole. We cannot do better than quote Mr. Parshall's own 
 words, as forming the summary of his remarks on three-phase 
 versus three-wire systems for electric traction. They are as 
 follows : " I would point out that the three-wire system for 
 traction work is purely in an experimental condition ; that it 
 has been tried in practice and proved a failure ; that it neces- 
 sitates the use of cominutating machines and sub-stations, 
 the loss in which, and in the neutral conductor, is comparable 
 in magnitude with that in the sub-stations jn the multi-phase 
 system ; that the advantages of high voltage transmission 
 are lost ; that it introduces additional losses in the locomo- 
 tive, owing to the fluctuation of voltage in the consuming 
 circuit. (I am aware that this variation of voltage may be 
 compensated for by feeders and boosters. Such an arrange- 
 ment, however, is at the expense of efficiency, and I will not 
 complicate my discussion by taking it into consideration.)" 
 It is to be hoped, however, in view of this difference of 
 opinion between such eminent experts, that the English 
 engineers will hold their own, by bringing to a successful 
 issue a large three-wire traction scheme. 
 
 The Use of Cells in a Three-wire System. The use of 
 storage-cells in sub-stations to be used on a three-wire system 
 
i82 ELECTRICAL TRACTION. 
 
 was not discussed at the above meeting. Such 'cells, if 
 automatically regulated, would remove some of the objec- 
 tions given. The recuperation, so far as resistance of the 
 cell is concerned, is instantaneous, and the automatic switches 
 at Leeds can operate at the rate of eight volts per second, 
 or more if required, when adding or subtracting cells to 
 or from the circuit. See p. 154. 
 
 Alternate and Direct Currents and Storage-Cells. 
 According to a system proposed by Deri,* each car con- 
 tains an alternate-current motor, a direct-current motor, 
 and a set of storage-cells. The idea is to keep the load 
 on the station as constant as possible by supplying energy 
 at constant, or n early "constant, rate, by means of alternate 
 currents'. The direct-current dynamo is a generator when 
 the car requires little energy to propel it, such generator 
 charging the storage-cells. On the other hand, when the 
 car requires considerable energy for propulsion, the storage- 
 cells drive the direct- current machine as a motor. 
 
 * See Electrician, July 2, 1897. 
 
( 183 ) 
 
 CHAPTER IX. 
 
 EFFICIENCY. 
 
 General Remarks. The efficiency of an electric traction 
 scheme involves many points of importance to the engineer, 
 and finally resolves itself into the question of how, from 
 the best available source of energy, to carry out the scheme 
 with maximum efficiency, having due regard to expendi- 
 ture of money. In a given time, a certain amount of 
 work is performed by the source, and the engineer, if he is 
 to use this to the best advantage, must consider the losses, if 
 we may so call them, in the different parts of the system, 
 in order to deliver as much energy as possible for the object 
 in view, with the least all-round expenditure. We may 
 conveniently divide our system into the following parts. 
 
 First, the conversion of energy delivered in the form of 
 electricity by the power-station or stations to the working 
 conductors. This includes the efficiency of the electric 
 generators themselves. 
 
 Second, the distribution of such energy by the system of 
 conductors to the cars or locomotives. 
 
 Third, the conversion of such energy by the motors and 
 necessary controlling apparatus into the mechanical energy 
 delivered to the rails. 
 
 A full discussion of the first division involves important 
 questions of average efficiency,* division of plant, spare 
 plant, and the proportion of total power used for purposes 
 other than that delivered in the form of electricity to the 
 conductors of the system. One can readily understand 
 
 * Sec Dr. E. Hopkinson, Proc. Inst. Civil Eng., vol. xci. p. 207. 
 
184 ELECTRICAL TRACTION. 
 
 what a, wide subject this is when one considers the many 
 ways in which energy is procured for the purpose of driving 
 the electric generators in the power-house. We have 
 instances of power transmission over long distances in order 
 that water-power may be turned into good account ; and in 
 such a scheme the efficiency of this system becomes of much 
 importance, since it has to be considered whether the first 
 cost of turbines, generators, land-lines, and motors, with their 
 continuing cost, is to be less than the employment of steam- 
 engines in the power-house, involving the cost of coal, 
 water, etc. The whole subject of water turbines is of great 
 importance, and is dealt with in works specially devoted to 
 the subject. It may be mentioned that from 75 to 80 per 
 cent, is an efficiency to be obtained with turbines of fair size 
 and good construction. The governing of such turbines is 
 an important question, for if the load be suddenly removed, 
 and racing occurs, the generators coupled to such turbines 
 get severely strained, and may break down. All these 
 are points to be carefully considered. The requirements of 
 the plant used in electric traction are in some respects 
 different to those in ordinary electric supply systems, on 
 account of the sudden fluctuations in the demand for power. 
 If steam be used, for instance, it is of extreme importance 
 that the engine should have sufficient momentum to main- 
 tain the speed nearly constant for variations of the load, and 
 the engine should be well governed. The class of electric 
 generator, too, is different in some respects to the generator 
 usually employed in electric supply central station work. 
 The fluctuations, which occur very rapidly, render it im- 
 possible to adjust the brushes of the generators for each change 
 of load, and they must be fixed. It is well known that car- 
 bon brushes are well suited to varying loads, but copper 
 is also employed in traction work, when the brush is supplied 
 with what is generally termed a carbon " tip ; " that is, a 
 small plate of carbon which bears on the commutator just 
 in front of the main copper brush, and has the effect of 
 
EFFICIENCY. 185 
 
 suppressing the sparking ; but, for satisfactory working, 
 armature reaction must be comparatively small. Large 
 machines of the ordinary two-pole type have, it is well known, 
 great armature reaction, and this would seem to be the 
 reason why they should be superseded by the multi-polar type ; 
 otherwise the copper on the magnet windings, owing to the 
 necessary great air-space, becomes excessive. Armature 
 reaction can be reduced by splitting the field magnets 
 (see p. 14). There is no doubt that for large units in 
 traction work the multi-polar type has special advantages. 
 It is lighter and cheaper to make for a given output in 
 watts, and a given spaed of rotation, and, on account of the 
 subdivision of the field magnets, the armature reaction is 
 comparatively smaller, thereby giving greater satisfaction on 
 account of the diminished sparking with load variations. 
 The subject of relative weights of bi-polar and multi-polar 
 dynamos has been touched upon in the Journal of the 
 Institution of Electrical Engineers, vol. xxvi. p. 127. There 
 is no doubt that the introduction of steel into dynamo build- 
 ing has had an important effect upon the development of the 
 multi-polar type, for good castings can now be obtained, 
 which have a permeability equal to the best wrought 
 iron. 
 
 The efficiency of the prime mover and dynamos can only 
 be properly dealt with by considering the average effect. 
 Let a curve be drawn which gives the relation between the 
 rate at which energy is being delivered by the source (be it 
 coal, or water in motion, for instance) and time. Let 
 another curve be drawn giving the relation between the rate 
 of delivery of energy in the form of electricity to the system 
 of conductors, and time. Then the areas bounded by each 
 of these curves between the ordinates corresponding to any 
 two times, and the base line, give respectively, for the 
 intermediate time, the total energies delivered to and by 
 the power-house as a whole, and their ratio gives the 
 average efficiency required. The instantaneous efficiency 
 
1 86 ELECTRICAL TRACTION. 
 
 may fluctuate greatly, but the engineer must strive to make 
 the average approximate as nearly as possible to the 
 maximum. In electric supply systems the load curve tells 
 us the rate at which work is done by the station, and one of 
 the chief difficulties is to so arrange the plant that, con- 
 sidering first and continuing costs, the station may be worked 
 with a maximum economy. The load curve for a traction 
 station may have fluctuations (see Fig. 75), but the rate at 
 which energy is delivered is, for, say, 14 hours per diem, on 
 the average more constant, and therefore the average supply 
 may more nearly approximate to the maximum capacity of 
 the station. It is well known that in the case of electric 
 supply for lighting, the bulk of the plant installed is only 
 brought into requisition at those hours of the day when 
 the demand is great, and of course storage will reduce the 
 engine plant required. There is no doubt but that 
 traction schemes lend themselves to greater economy, owing 
 to the greater proportion of the 24 hours during which the 
 full power of the station is required ; that is to say, the 
 average is more equal to the maximum, and greater economy 
 of production consequently results. 
 
 An important question is the combination of electric 
 lighting and traction pLtnts, and it is, perhaps, unfortunate 
 that electric traction has come to the fore in this country 
 after electric lighting ; for, as a general rule, one may say 
 that electric lighting plants may not be suitable for traction, 
 whereas traction plants might be suitable for electric light- 
 ing. It may therefore be desirable to design electric-lighting 
 plants which will be able to deal with traction schemes if 
 required, so that the two can be combined. The lighting and 
 traction schemes at Rome and Hamburg are good examples 
 of such a combination (see pp. 131, 146). 
 
 The subject of power production is a very wide one, and 
 includes water power, steam power obtained from coal and 
 refuse destructors, and gas-engines. 
 
 The second division involves the form of conductors to 
 
EFFICIENCY. 187 
 
 be used, whether they be feeders forming a simple distribu- 
 tion, or a three- wire or three-phase system. 
 
 The third division is dependent, not only on the motor 
 or motors employed, but also upon the variations of applied 
 potential difference, and the controlling apparatus (see 
 Chap. II.). 
 
 The actual methods employed under each of these 
 headings are numerous, and experience only can show the 
 best to be adopted. The Author considers it best to discuss 
 a few representative systems, rather than enter upon a 
 general statement. 
 
 Dublin. The two sections of electric tramway under the 
 control of the Dublin United Tramways Company are those 
 of the Southern and Clontarf Districts. The following short 
 description gives some points of importance in connection 
 with the Dublin Southern District Tramways. The line 
 has a length of 7f miles of girder rail, 76 Ibs. per yard, 
 which, with the exception of two short lengths, is double 
 throughout, and the gauge is 5 feet 2^ inches. The design 
 is due to Mr. H. F. Parshall, consulting engineer to the 
 British Thomson-Houston Company, by whom the whole 
 of the work has been carried out. The power-house 
 contains three Babcock - Willcox boilers of 250 H.P. 
 each, working at a pressure of 140 Ibs. per square inch. 
 Vicars's stokers, driven by an electro-motor, supply the 
 fuel to the boilers. The scrapers in a Green's economizer, 
 as also the three throw-pumps for the feed water, are driven 
 by shunt-wound electro-motors at 500 volts. The feed- 
 pumps are capable of delivering 16,000 Ibs. of water per 
 hour, and mild steel ring main 8-inch steam-pipes are 
 used. The engine-room at present contains four Willans 
 compound condensing two-crank engines, each of 150 B.H.P., 
 working at 380 revolutions per minute with 140 Ibs. per 
 square inch steam pressure, but capable of developing 175 
 horse-power each for short times if necessary. 
 
 The dynamos are coupled to a three feet fly-wheel pulley 
 
1 88 ELECTRICAL TRACTION. 
 
 on each engine by belts. There are two British Thomson- 
 Houston 100 kilo-watt four-pole generators, compound 
 wound to give 500 volts at full load; and two 120-kilo- 
 watt three-phase generators, which, at GOO revolutions 
 per minute, or 30 periods per second, give from 2300 
 to 2500 volts. The power-house is situated about half 
 a mile from the Dublin end of the line. There are 
 two sub-stations, one at Blackrock, about 3^- miles from 
 the power-house, and the other at Dalkey, at the other 
 end of the line, and distant about seven miles from 
 the power-house. Each of these sub-stations contains two 
 synchronous three-phase alternating-current motors, each 
 direct-coupled to a four-pole generator, giving, at full load, 
 120 amperes at 500 volts, and connected to the overhead 
 trolley wires and rail return. A few turns of the con- 
 ductor leading from this generator are wound on the 
 magnets of the synchronous motor to compensate for the 
 reaction of the motor armature upon its field. These 
 motors are run up to speed, and synchronized by means of 
 the 500-volt generators, which are temporarily run as 
 motors from the 500-volt circuit. The commercial effi- 
 ciency of such motor-generators is stated to be 83 per 
 cent, at full load. The Board of Trade regulations' with 
 regard to three-phase high tension transmission on this line 
 are as follows (see also Appendix, p. 233). 
 
 " The electrical pressure between the conductors in any 
 electric line, or between any such conductor and the earth, 
 shall not in any case exceed 3000 volts." 
 
 " All electric lines laid for the purpose of supply to trans- 
 forming stations on the ' three-phase system,' shall have 
 their conductors arranged concentrically, the outer con- 
 ductor being efficiently connected with earth at the gene- 
 rating station, but insulated at all other points ; and the 
 thickness of insulation between the several conductors of 
 any such electric line shall not be less in parts of an inch than 
 the number obtained by dividing the number expressing 
 
THE DUBLIN SOUTHERN DISTRICT TRAMWAYS. 189 
 
 the maximum electrical pressure, in volts, by 20,000. 
 No such electric line shall be brought into use unless the 
 insulation of every part thereof has withstood the continuous 
 application, during one hour, of twice the maximum pressure 
 to which it is intended to be subjected in use." 
 
 At Dublin the car bodies are mounted upon Peckham 
 standard extension trucks. Both the car bodies and frames 
 and the trailer-cars were built by Messrs. G-eorge F. Milnes 
 & Co. of Birkenhead, and the principal weights and 
 dimensions are given below, from which it will be seen that 
 the total weight of the passengers on the cars might be 
 about 8*2 tons, and on the trailers about 5'76, making in 
 all 14 tons of passengers ; that is, for the motor-car and 
 trailer with a full complement of passengers, we have a 
 grand total of about 22 tons. 
 
 MOTOR-CAB. 
 
 ft. ins. 
 
 Outside length of car body . . 1C 8 
 
 Total length over platforms . . 27 2 
 
 Extreme width of car body . 71 
 
 Height of platforms above rails . 22 
 
 Height of top deck above rails in centre . 9 4 
 
 Wheel base ...... . 60 
 
 Diameter of wheels .. . . 26 
 
 Weight of car body and frame . . 6720 Ibs. 
 
 Weight of truck .. . . 4480 
 
 Total weight ... .. . . 5 tons 
 
 Seating capacity in saloon, 24 ; on top deck, 29. 
 TRAILER-CAB. 
 
 Outside length of car body 
 Total length over buffers 
 Extreme width of car body 
 Height of platforms above rails 
 Height of top deck above rails 
 Wheel base ... 
 Diameter of wheels 
 Total weight of car 
 
 Seating capacity in saloon, 24 ; on top deck, 29. 
 
 ins. 
 
 9 
 
 9 
 10 
 
 
 10 
 
 6 
 
 6 
 3 tons 
 
 Each motor has a single reduction spur gearing, 4'78 to 
 
190 ELECTRICAL TRACTION. 
 
 1, between its shaft and the car axle, and is stated to be 
 able to exert a force of 800 Ibs. at the rim of a 33-inch 
 driving-wheel at a speed of eight miles per hour (see Figs. 
 14 and 15). The two motors per car can therefore exert 
 a force of 1,GOO Ibs. at this speed. The weight of each 
 motor complete is 1,455 Ibs. ; of which 715 Ibs. is taken by 
 the axle. The motor is stated to have a maximum com- 
 mercial efficiency of 80 per cent. 
 
 Efficiency of System of Conductors. Fig. 21, p. 62, gives 
 a diagram of the conductors at Dublin. A is the power- 
 house at Balls Bridge, B and C are the sub-stations at 
 Blackrock and Dalkey respectively. Consider the three- 
 phase transmission between A and B. There are two triple 
 concentric cables in parallel, the two outers being to earth 
 at the station. The effective area of one conductor is 
 0'0938 square inches, and the resistance 1*75 ohms. The 
 kilo-watts delivered to this feeder at A are stated to be 
 261 ;* that is, 87 per line. The maximum potential 
 difference between the two wires is 3000 volts, which is the 
 maximum allowed by the Board of Trade (see regulations 
 just quoted, and Appendix, p. 233). The apparent watts 
 are 2 X 3000 X 60, since 60 amperes is the current in each 
 line ; and since the true watts are 261,000, the power factor 
 would appear to be 0*72. The rate of dissipation of energy 
 in the feeders due to ohmic resistance and current - 3 x 
 1-75 X (60) 2 = 18,900 watts, and this is 7'25 per cent, 
 of the maximum, which is 261,000 watts. The fall of 
 potential between A and B is about 100 volts. Of the 
 242,100 watts delivered at B, a portion is transmitted by 
 a feeder to C, and the rest is transformed into direct current 
 at 500 volts for the line at, say, 83 per cent, efficiency. 
 Next consider the transmission of power between B and C. 
 One triple concentric cable is used ; the area of one con- 
 ductor is 0'0469 square inch, and the resistance 4'05 
 ohms. The power delivered to the feeder at B is shown as 
 
 * Sec Engineering, 5th and 12th of June, 1896, Fig. 11, p. 771. 
 
EFFICIENCY OF DUBLIN CONDUCTORS. 191 
 
 3 x 27 -81 kilo-watts. The amperes per line are 20, 
 giving about 80 volts fall of potential between B and C. 
 The rate of dissipation of energy in the feeder due to ohmic 
 resistance and current = 3 x 4 '05 X (20) 2 = 4860 watts. 
 The power given to the feeder at B = 81,000 watts, so that 
 6 per cent, is lost in this part of the transmission. Since 
 the apparent watts are 2 x 2818 x 20 = 113,000, the power 
 
 factor is about - - = 0'72. "We have, therefore, 76,100 
 1 lo 
 
 watts delivered to the synchronous motor at C, and at 83 
 per cent, efficiency, we get 63,200 watts delivered to the 
 line, which, at 500 volts, gives 126 amperes. Return 
 to sub-station B, the total watts delivered by the concentric 
 cable are 242,000 ; and 81,000 of this goes on to C, the 
 difference, 161,000, goes to the synchronous motor to be 
 transformed directly, at 83 per cent. This gives 134,000 
 watts to the line, which, at 500 volts, gives 268 amperes. 
 The whole system as a feeder delivers 197,200 watts to the 
 line in return for 261,000. Therefore, its efficiency as a 
 feeder is equal to about 76 per cent. Take the cost of 
 power at 10 per annum per horse-power. Then, if the 
 rate of supply were constant, we clearly waste in this 
 system, as a feeder, energy at the rate of 86 H.P, ; under 
 this heading the annual charge will be 860. Other in- 
 cidental expenses are wages of attendants and running ex- 
 penses in the sub-stations as a part of the feeder system. 
 The amount for these per annum, when added to the 860, 
 should, in order that the whole expense may be a minimum, 
 be equal to the interest on the first cost of the system as a 
 feeder, plus the depreciation of the same. The weight of 
 the copper in the concentric feeders the Author estimates 
 at about 15 tons. 
 
 We have now to get the efficiency of the whole system as 
 a means of conveying energy from the power-house to the 
 cars. Beyond the power delivered to the line as above, we 
 have that delivered directly from the power-house to the 
 
192 ELECTRICAL TRACTION. 
 
 line. The power-house is assumed near enough to the line 
 to neglect the loss in the connecting mains. We want also 
 the dissipation of energy in the trolley wires and rails. 
 Suppose on each line there are five trains between B and C, 
 three between A and B equally spaced out, and a train each 
 at A, B, and 0. Allow 20 amperes per car. Take the 
 distance B, C. Then considering the half of this, and 
 adopting the same method of procedure as on p. 61, n = 3, 
 r = O'lll ohms, since there are 2n intervals between B and C. 
 The watts dissipated = O'lll X (40) a X {(f) 2 +>(f) 2 + (|) 2 } 
 = 1550. Twice this gives 3100 as the total rate of 
 dissipation of energy between B and C in. the trolley wires 
 and the parallel feeder. Similarly, between A and B, since 
 n = 2, and r = 0-166, the watts dissipated = O'lGG X (40) 2 
 X {(f) 2 + (i) 2 } = 064. Twice this gives 1328, so that 
 the total = 4430 watts. The rail might have a resistance 
 of 0'0099 ohms per mile for the double track. We note 
 that the maximum drop in the rail might be about two volts. 
 Farther, the watts dissipated in the rails may be neglected. 
 We have then, if we assume the power-house supplies 
 120 X 500 = 60,000 watts direct to the line, a total of 
 321 kilo-watts supplied from the power-house. Of this we 
 may say 68 '4 kilo- watts are dissipated in conductors and sub- 
 stations : then 253 kilo-watts are delivered to the cars. The 
 total efficiency, as a system for conveying energy from the 
 power-house to the cars, is 79 per cent., taking the figures 
 as here given. 
 
 The Clontarf section has a length of about three miles 
 with 15 motor-cars, and presents a notable contrast to the 
 Southern District so far as plant is concerned. Fig. 69 
 shows a sectional elevation, and Fig. 70 a plan, of the power- 
 house. The units adopted are three British Thomson- 
 Houston 150-kilo-watt multi-polar generators directly coupled 
 to Mclntosh and Seymour condensing engines fitted with 
 automatic cut-off gear. Provision is made for two more 
 such sets when necessary. No sub-stations are employed, 
 
THE LEEDS ELECTRIC TRAMWAY. 195 
 
 and only direct currents are used, the potential difference 
 being 500 volts. 
 
 Leeds. The following is a short description of the 
 electric tramway scheme at Leeds. The line is about seven 
 miles long, with a double track of girder rail, 100 Ibs. per 
 yard, and overhead conductors, each 0*4 inch diameter of 
 hard-drawn copper, carried on steel tubular poles 120 feet 
 apart. The gauge is 4 feet 8-| inches. This system has 
 been designed by Dr. J. Hopkinson, F.R.S., the consulting 
 engineer, and has been carried out under his direction. 
 Work was started in August, 1896, and cars commenced 
 running on the 2nd day of August, 1897. The power-house 
 is illustrated in plan in Fig. 71, and is situated about 700 
 yards from the centre of the line, and contains at present 
 two 400 H.P. Lancashire boilers by Messrs. Clayton, Son 
 and Company, 30 feet long by 8 feet in diameter, shown 
 in full lines. A Green's Economiser and Bennis mechanical 
 stokers are used. There are, at present, two horizontal 
 compound condensing engines, shown also in full lines on 
 the plan, and made by Messrs. Fowler & Co., of Leeds, 
 each capable of delivering at the dynamo pulley, by rope 
 transmission, 400 H.P., with a steam consumption of 
 not more than 21 Ibs. per brake horse-power. The fly- 
 wheel of each engine has a diameter of 15 feet, and weighs 
 19 tons. Each engine drives a pair of Greenwood & Batley 
 two-pole dynamos, coupled together on one base plate. 
 Each dynamo gives 550 volts 200 amperes compound, 
 and 600 volts 200 amperes shunt, at 450 revolutions per 
 minute ; so that one pair, if working as above, would 
 deliver 230 kilo-watts, or 308 H.P., to the system. A 
 differential test carried out with a pair of these machines 
 shows that at full load they have each a commercial 
 efficiency of 95'2 per cent. Fig. 22 gives a diagram of 
 conductors in the system. There are two battery stations 
 at C and C', situate respectively about 5000 and 4600 
 yards from the power-house. Each battery is charged and 
 
EFFICIENCY OF LEEDS CONDUCTORS. 197 
 
 discharged automatically. The switches, of which we give 
 a description in Chap. VII., maintain the potential differ- 
 ence between the line and the track constant at 500 volts, 
 and cut out cells, when charged ; but provision is made for 
 the charging feeder never to be placed below the discharge 
 feeder on the cells. The Roundhay battery is at a consider- 
 able level above the power-house, and is equipped with 250 
 Chloride Company's cells, each capable of discharging 160 
 ampere-hours at the rate of 160 amperes without undue 
 drop of potential. The Kirkstall battery has a capacity of 
 110 ampere-hours, the rate of discharge being 110 amperes. 
 
 The contractors must maintain the battery for the first 
 year without charge, and thereafter at the rate of 8 per cent, 
 on contract price as long as the Corporation shall require. 
 
 There are 25 cars made for the contractors, Messrs. 
 Greenwood & Batley, by Messrs. Milnes & Co., of Birken- 
 head, and mounted on Peckhain trucks (see Fig. 34). There 
 are also some trailers. The motor-cars are each supplied 
 with two single reduction motors, constructed by Messrs. 
 Greenwood & Batley. 
 
 Efficiency of System of Conductors. In the diagram, 
 Fig. "22, A is the nearest point in the line to the power-house ; 
 B, B' are feeding points ; 0, C' are the battery sub- 
 stations. The feeders to the points A, B, and B' consist of 
 double concentric cables, each conductor having 0*34 square 
 inch cross sectional area. The inner conductor is connected 
 to the trolley line, whilst the outer is connected to the rails. 
 The single feeder to C' has 0*25 square inch sectional area, 
 and the single feeder to C has 0'17 square inch cross 
 sectional area. From the power-house to A, B, and B', the 
 mains are laid in earthenware conduits, having a separate 
 compartment for each feeder, and provision for extension, 
 whilst between B and C, and B' and C', the feeder is laid 
 in an earthenware pipe. Assume that each car takes 
 15 amperes as the average current and we may be justified 
 in doing this with a system of this magnitude. Then the 
 currents might be as follows : 
 
ELECTRICAL TRACTION. 
 
 Feeder. 
 
 Current in 
 amperes. 
 
 Resistance of 
 feeder in ohms. 
 
 Watts dissipated 
 in feeder. 
 
 A 
 
 60 
 
 0-102 
 
 367 
 
 B' 
 
 60 
 
 0-305 
 
 1,100 
 
 0' 
 
 90 
 
 0-65 
 
 5,270 
 
 B 
 
 60 
 
 0-305 1,100 
 
 C 
 
 90 
 
 0-56 4,540 
 
 
 360 
 
 12,377 
 
 Twenty-five cars, at 14 or 15 amperes per car, give about 
 360 amperes, which, at 500 volts, give 180,000 watts to be 
 delivered to the cars. Neglecting for the moment the rate 
 of dissipation of energy in the trolley wire and rails, which 
 we can safely do, we require 180,000 watts to be supplied to 
 the line. The battery stations have, on our assumption, to 
 deliver 45 kilo-watts each, and we assume they do this at 
 80 per cent, efficiency. "We lose in the battery 22,400 watts, 
 which, added to 12,377, gives, as a total, 34*8 kilo-watts- 
 The efficiency, then, of this system of feeders, under the above 
 assumed conditions, when delivering 180 kilo- watts, is 84 per 
 cent. If we allow 10 per horse-power per annum, we have 
 a continuing cost under this head of 460 per annum, to 
 which must be added the cost of supervision and annual 
 running expenses in the sub-stations as a part of the feeder- 
 system, and this whole amount should be equal to the 
 interest on the first cost of mains, batteries, and switches, 
 and depreciation, in order that the whole may be a minimum- 
 According to the assumed conditions, we dissipate energy 
 in the trolley wire at the rate of about 740 watts ; the loss 
 in the track we neglect ; so that the efficiency of the whole 
 of the conductors and batteries between the power-house 
 and the cars, and under the above assumed conditions, 
 practically remains at 84 per cent. 
 
 The Author estimates the weight of copper in the feeders 
 at about 27 tons. 
 
EFFICIENCY. 199 
 
 The Author is not prepared to enter upon the first costs 
 or depreciation of the plant either at Dublin or Leeds. The 
 following points, however, have struck him with regard to 
 these two systems, which are of similar magnitude. At 
 Dublin (on the Southern District), two distinct kinds of 
 apparatus are used, namely, high-tension alternating cur- 
 rents at 3000 volts, combined with direct-current machinery 
 working at 500 volts. At Leeds the direct-current at 500 
 or 600 volts only is used ; but batteries require careful 
 attention. The power-house at Leeds is more favourably 
 situated with regard to the line than at Dublin in its present 
 state. At Leeds the weight of copper in the feeders is 
 about 1*8 times that at Dublin ; but we have seen (p. 55) 
 that the cost of conductors and their laying is not to be 
 judged from the copper alone. At Dublin there is the 
 cost of the sub-stations and rotary transformers, and at 
 Leeds there are the batteries to be considered. Batteries 
 certainly give very great security, for they are not only able 
 to meet at once heavy demands, but, for a time at least, 
 would keep some cars running, and would be able to do the 
 shunting on occasions when perhaps the plant is not 
 running. It must also be remembered that the efficiencies 
 above deduced are so deduced under assumed conditions. 
 Their average values are not discussed. 
 
 Efficiency of Locomotives and Cars. We have seen, in 
 Chap. II., how the efficiency of motors and controlling 
 apparatus can be calculated under given conditions ; but, in 
 actual working, the important question is not what the 
 efficiency is at any moment, but what its average is over a 
 certain time. This, for a given motor, is affected by the 
 maximum speed allowed, the schedule of time, the frequency 
 of stopping and starting, the load, and the resisting force. 
 The definite stopping-places on a railway, no doubt, conduce 
 towards economy of working, as compared with the frequent 
 stoppages on a tramway ; and the further apart the stopping 
 places are, the greater is the economy. Let a curve be 
 
200 ELECTRICAL TRACTION. 
 
 drawn giving the relation between the watts supplied to the 
 locomotive, or car, in terms of time plotted horizontally. 
 The area of this curve between any two limits, with regard 
 to time, gives in suitable units the actual work done upon 
 the car in that time, and can be expressed in Board of Trade 
 units if desirable. Let a second curve be drawn, giving 
 the relation between the watts actually delivered by the car 
 to the rail, having time plotted horizontally. Then the 
 area of this curve between the same two limits, gives the 
 actual work done upon the rail by the car in the same time. 
 The ratio of this last area to the former is the average nett 
 efficiency of the car between the limits of time con- 
 sidered. 
 
 The interesting series of curves given by Mr. Siemens, 
 and alluded to in Chap. II., refer to the locomotives 
 supplied to the City and South London Eailway, and give 
 the volts, amperes, and speed as taken on the locomotive 
 itself at short intervals of time daring the journey. From 
 the product of volts and amperes we have the rate at which 
 work is done on the locomotive at any time. These 
 products are plotted in terms of the time. The revolutions 
 per minute of the armature are given, since we know the 
 speed and diameter of the driving-wheel. We have seen 
 
 ~P ro v i o^ 
 (p. 21") that the revolutions per minute = x ; 
 
 1 4m 
 
 this enables us to find E, since I is known from, the cha- 
 racteristic curve, and we know C. Therefore, the rate at 
 which the motors do work, EC, is known at any moment. 
 But we want the rate at which they do work on the rails 
 themselves. Assume, as before (p. 22), that the rate of 
 dissipation of energy due to wind, bearing, and brush 
 friction, magnetic hysteresis and eddy currents in the 
 armature cores, amounts to 1500 watts in the two motors, 
 and that it is constant. Then subtract from EC, at any 
 moment, 1500 watts, and plot the resulting difference in 
 terms of time. A curve drawn through these points 
 
EFFICIENCY OF LOCOMOTIVES. 201 
 
 gives us a good approximation to the rate at which work is 
 done on the rails themselves. It is necessary. to consider 
 over what lapse of time we want the average efficiency. 
 Do we want it during the time of one-half, or the whole of 
 the journey ? It is of interest to obtain the average 
 efficiency during the time between the successive stopping- 
 places, and also to see what are the variations as well as the 
 nett result. Two sets of curves (Sheets I. and II.) have been 
 treated in this manner, and give 75 per cent, as the average 
 efficiency for the round journey, starting from the City 
 station and returning to the same. The locomotive weighs 
 13| tons, and the carriages, with passengers, 21 tons. The 
 total length of the line is 5550 yards, and since there are 
 six stopping-places, the average distance between each is 
 1110 yards. Table X. gives a summary of the results 
 obtained from Sheets I. and II. 
 
 TABLE X. 
 
 
 Sheet II. 
 
 Sheet I. 
 
 Number of passengers... 
 
 34 
 
 29 
 
 Total weight of train, in tons ... 
 
 36-5 
 
 36-2 
 
 Total Board of Trade units supplied to locomotive 
 
 5-394 
 
 5-378 
 
 Total time of journey in minutes 
 
 15-5 
 
 14-7 
 
 Average speed, including stopping, in miles per hour 
 
 12-2 
 
 12-9 
 
 Board of Trade units per mile 
 
 1-71 
 
 1-71 
 
 Board of Trade units per ton per mile ... 
 
 0-0469 
 
 0-0472 
 
 The whole of the eight sets of curves, making four com- 
 plete round trips, have been treated in the same manner, 
 and the average efficiency of the locomotives comes out at 
 about 70 per cent. 
 
 In the tests (Sheets I. and II.) the average current for 
 the journey is about 50 amperes, at about 420 volts ; but, for 
 the whole series of tests, the volts varied from about 400 
 to 460, which, as we saw in Chap. II., has an important 
 
202 ELECTRICAL TRACTION. 
 
 effect on efficiency. The maximum amperes were from 
 130 to 140. Most of the acceleration is accomplished in 
 half a minute from starting, during which, on account of 
 the low net efficiency of the locomotive, most of the energy 
 is dissipated in the resistance coils on the car. The average 
 maximum speed is 18 miles per hour. We see from the 
 summary of results that the Board of Trade units per mile 
 for an average speed of 12^ miles per hour is T71, whilst 
 the Board of Trade units per ton per mile is 0'0469. Com- 
 pare this with tramway working. Take Montreal * ; here, 
 at an average velocity of 1\ miles per hour on an ordinary 
 tram line, the consumption is 0'26 Board of Trade units 
 per mile per ton, the maximum load being, perhaps, 10 
 tons, as against 36 on the City and South London Railway. 
 The motors on the City and South London Railway 
 have their armatures directly connected to the axle of the 
 driving wheels. In ordinary tramway work one sees almost 
 universally the motor geared to the car axle by spur gearing. 
 It is, of course, advantageous to reduce the weight as much 
 as possible when the whole load is on the driving wheels. 
 Therefore the geared motor has this advantage, that it is 
 lighter, and runs at a higher speed ; but the net efficiency 
 is not necessarily greater, since the gearing has to be con- 
 sidered. At the present time the efficiency of good spur 
 gearing is stated to be over 90 percent., but the Author has 
 not made direct tests to prove this. It could be accurately 
 determined by taking two motors on an undertruck, coupling 
 their axles together, and obtaining a direct measurement by 
 running the system so arranged differentially, preferably 
 making all the measurements electrical, as shown in Engi- 
 neering, March 24, 1893. Of course the dynamo has this 
 advantage over the steam-engine, for instance, that its 
 efficiency does not depend solely upon frictional variation. 
 The electrical dissipation of energy in the motor becomes 
 less with the load, and therefore the efficiency is maintained 
 * See Proceedings Inst. Civil Etigineers, vol. cxxiv. p. 287. 
 
EFFICIENCY. 203 
 
 at a higher figure for the corresponding smaller loads. At 
 the present time only one reduction by gearing is usually 
 used in tramway work, and, in a few instances, like the South 
 London Eailway locomotives, the motors are driving directly 
 on the axle ; two gears seem to have been entirely superseded 
 by the single reduction. The ratio varies in actual practice, 
 but is generally from 1:3 to 1 : 5. Steel is probably the best 
 material for the pinion, although cast-iron and phosphor- 
 bronze give good results. Cast-iron is used for the spur- 
 wheels. As we have seen in the consideration of the City 
 and South London locomotives, Chap. II., the series- 
 parallel control certainly raises the average efficiency, since 
 a greater range of speed can be obtained with high actual 
 efficiency, and this is of extreme importance where motors 
 have to run at greatly varying speeds. 
 
204 ELECTRICAL TRACTION. 
 
 H A P T E K X. 
 
 THE POWER-HOUSE. 
 
 THE sudden fluctuations in load to be dealt with in 
 traction work call for extra strength in engines and 
 dynamos, as well as very good governing with regard to 
 speed under quickly varying demands. Especially is this the 
 case in small schemes, where the average demand differs 
 considerably from the maximum, and where a larger relative 
 plant is consequently required. The conditions of working 
 are such that the plant must be able to run for long periods of 
 time, and, therefore, in the power-house special care must 
 be taken in the choice of engines and dynamos. The 
 indicated horse-power per car running is dependent upon 
 the size of the undertaking, since, if this be large, the 
 average demand at the station terminals does not differ so 
 much from the maximum. In small plants, therefore, we 
 find the indicated horse-power per car high ; then provision 
 has to be made in small plants for extra demands, owing to 
 fetes or any general holiday, such occasions calling for a 
 much greater supply, since the cars are crowded, and 
 perhaps extra ones running whereas, in large schemes, any 
 extra demand becomes of less importance. In a system of 
 about the size of Dublin or Leeds, the brake horse-power of 
 steam-engines is about eighteen to twenty per car, not 
 making any allowance for power used for other purposes in 
 the station, such as stokers, economizers, pumps, etc. This 
 figure must be approximate, and can only be got at by a careful 
 
THE POWER-HOUSE. 205 
 
 consideration of any actual undertaking. In August, 1896, 
 there appears to have been generated on the German Tram- 
 ways an average of 15 kilo- watts per motor car. Allowing 
 80 per cent, for efficiency of engines and dynamos, we get 
 25 I.H.P. per motor-car. In January, 1897, in the United 
 Kingdom, the average kilo-watts per motor-car for tramways 
 comes out at about 19. This would correspond to 32 I.H.P. 
 per motor-car, if 80 per cent, be allowed. Dawson gives, at 
 page 217 of " Electric Railways and Tramways," the 
 variation of indicated horse-power per car in terms of 
 the total number of cars running at 35 to 15, as the 
 total number of cars varies from 1-5 to 25-50. 
 
 The subdivision of plant is a very important question ; 
 that is to say, the size and number of units and reserve 
 machinery for a given scheme. At Leeds the number of 
 engines at present are two, each of 400 brake horse-power, 
 one being spare. At Dublin, on the Southern District, 
 there are four engines of 150 brake horse-power each. 
 At Bristol there are four engines of about 230 H.P. at 
 one-third cut off, which is the most economical load. 
 
 On the City and South London Railway, in February, 
 1893, from 60 to 67 per cent, of the total steam produced 
 was absorbed by the electric generator engines. 
 
 Boilers, Engines, Turbines, etc. The subject of boilers 
 demands a separate work to itself. This problem is to 
 a great extent the same as that which has to be dealt with 
 in electric lighting stations. The water tube and Lanca- 
 shire types are much favoured by engineers. The only 
 mention we can make of boilers will be when describing 
 an existing station ; and the same remark applies to tur- 
 bines. Economisers and mechanical stokers are almost 
 always used nowadays in traction stations. 
 
 A full discussion of the steam-engine is not to be 
 attempted in this book. The whole question as to high 
 and low speed, horizontal and vertical, direct coupled, and 
 belt or rope driving, depends greatly upon the size of the 
 
206 ELECTRICAL TRACTION. 
 
 units, the actual floor space available, and the engineer 
 designing the undertaking. The important points in any 
 engines for traction work are great strength, since the 
 load fluctuations are severe and sudden, great momentum 
 to compensate in some measure for such fluctuations, and 
 a powerful governor, which must be automatic and quick 
 acting. 
 
 The average load is, for fairly large systems, from 0'6 to 
 0*7 of the maximum, and steam engines should therefore 
 work at maximum efficiency at about 0*6 or 0-7 of their 
 maximum power. The engine should also be constructed 
 to stand considerable overloading. The whole subject of 
 speed regulation cannot be too well considered, and espe- 
 cially is this the case in alternate current working, where 
 the generators have to synchronize. 
 
 Governors. The large Mclntosh and Seymour com- 
 pound engines have a governor acting on both cylinders. 
 
 Figs. 7 "2 and 73 represent a centrifugal governor as made 
 by this firm, and the following is the maker's own descrip- 
 tion : 
 
 It consists of centrifugal weights which vary the point of 
 cut-off by revolving the governor eccentric upon the shaft, 
 and is designed so that the centrifugal force of each weight 
 is very great, making the governor exceedingly powerful. 
 These forces are opposed in a direct and frictionless manner 
 by a plate spring through hardened steel pins resting in 
 cups at each end, so that there is no friction or pressure 
 due to these forces upon the pins upon which the weights 
 turn. Dash pots are provided, which give stability to the 
 governor. Great care is taken in the mechanical construc- 
 tion throughout. The pins are made of tool steel, hardened 
 and ground, turning in bushes of hard phosphor bronze, 
 with provision for lubrication. The governor can be ad- 
 justed as to sensitiveness by changing the length of the 
 pins between the centrifugal weights and spring, these pins 
 being arranged to telescope for this purpose. The speed 
 
GOVERNORS. 
 
 207 
 
 is regulated by changing the weight of bushings in the 
 centrifugal weights. This governor can be adjusted to 
 give practically perfect regulation, without any tendency 
 whatever to race under a sudden decrease of load. 
 
 The Willans' centrifugal governor with throttle valve 
 
 FIG. 72. 
 
 and variable expansion gear is used in traction plants. 
 Referring to a Willans direct coupled set at Stockwell, Mr. 
 McMahon, in the discussion of Mr. Eaworth's paper before 
 the Institution of Electrical Engineers,* on the " Generation 
 * See Proceedings, vol. xxvi. p. 483. 
 
208 
 
 ELECTRICAL TRACTION. 
 
 of Electrical Energy for Tramways," states that the governing 
 was about the most perfect he had ever seen in a traction 
 station ; the variation of speed being not more than four or 
 five per cent, at the most. The total output of the set was 
 250 amperes at 500 volts. With the engine running light, 
 
 FIG. 73. 
 
 and full load instantly thrown on, the cut-off in the high- 
 pressure cylinder changed instantly from 0'15 to 0-6, with 
 a speed variation of about four per cent. The description 
 of this governor is too long to give here, but it is well 
 worth studying. Of course fly-wheel action is included in 
 
FLY-WHEELS. 
 
 209 
 
 this effect. Fig. 74 gives two curves relating to a Willans 
 engine, showing variation of speed per cent, and load in 
 electrical horse-power. 
 
 It should be here mentioned that there are other good 
 governors, for instance, the Proel, but space does not 
 allow of a description of these. 
 
 Fly-wheels. The following is an endeavour to ascertain 
 
 A 
 
 \ 
 
 FIG, 74. 
 
 on what basis the fly-wheel for a given load should be 
 constructed according to modern practice in traction 
 work. 
 
 The Author has taken Table XI. from the list of Messrs. 
 Mclntosh and Seymour, which refers to their standard com- 
 pound railway engine. This is a horizontal type engine 
 fitted with piston valves, and a central governor which 
 works directly on the valve rod, and is stated to regulate 
 the speed of the engine within two per cent, for all changes 
 of load from zero to full. 
 
 p 
 
2 to 
 
 ELECTRICAL TRACTION. 
 
 TABLE XI. 
 
 Type A condensing engines. 
 
 Indicated 
 horse- 
 power. 
 
 Range of 
 initial pres- 
 sure in Ibs. 
 
 Revolu- 
 tions per 
 minute. 
 
 Fly-wheel Two wheels per 
 engine. 
 
 Weight 
 of engine 
 in Ibs. 
 
 Diameter 
 in inches. 
 
 Face 
 inches. 
 
 Weight 
 in Ibs. 
 
 90 
 
 90-110 
 
 260 
 
 66 
 
 12-5 
 
 2300 
 
 11,500 
 
 215 
 
 90-110 
 
 210 
 
 82 
 
 18-5 
 
 4500 
 
 29,500 
 
 400 
 
 110-120 
 
 195 
 
 88 
 
 26-0 
 
 7000 
 
 46,750 
 
 500 
 
 110-130 
 
 175 
 
 108 
 
 32-0 
 
 8000 
 
 66,500 
 
 If iv be the angular velocity of a wheel, then any particle 
 of mass m at radius r is revolving at velocity v = rw, and 
 its energy is mv 2 , or, if preferred, mr 2 w 2 . The energy of 
 the whole wheel is therefore ^(\mv z ) = |f0*S(j#r a ) ; or the 
 sum of the second moments of inertia of all the particles 
 multiplied by fyv 2 . 
 
 Taking the table just given, we assume that the radius of 
 gyration for the rim of the wheel is one inch less than the 
 external diameter would give ; further, we assume that 
 the weight of the rim is 85 per cent, of the total 
 weight of the wheel, and that this is acting at the 
 above assumed radius. From this radius, the mass, and 
 the normal speed, we deduce the energy \mv in foot-pounds, 
 VQ being the mean normal velocity in feet per second. 
 Assume now that a five per cent, diminution in the speed 
 has been effected. Then we deduce \mv? where i\ is the 
 mean velocity of the rim at this reduced speed ; \mv* 
 \mv? is the work in foot-pounds delivered by the wheel 
 to the shaft. The time of this delivery in minutes is taken 
 to be the difference of the initial and final revolutions per 
 minute, divided by the mean of the two revolutions ; and 
 
PLY-WHEELS. 
 
 211 
 
 therefore the rate in foot-pounds per minute will be the 
 quotient of these two quantities. It is further assumed 
 that 85 per cent, of the indicated horse-power is delivered 
 at the terminals of the dynamo, which is being driven by 
 the engine. This has been reduced to foot-pounds per 
 minute, and the whole of the figures are given in Table XII. 
 A comparison shows that, under the assumed conditions, each 
 engine fly-wheel is able to deliver energy on the average 
 at the same rate as the dynamo, and the two wheels com- 
 bined would deliver the same average energy for twice the 
 time. Of course this does not mean that if the engine 
 could be instantaneously robbed of steam the fly-wheels 
 would maintain the load on the average for double this 
 time, since there would be the friction in the engine to 
 consider as well as the dynamo efficiency. It seems, there- 
 fore, according to the practice of Messrs. Mclntosh and 
 Seymour, that the fly-wheels are constructed to operate as 
 above described. The variation of speed during one revolu- 
 tion is not here discussed. 
 
 TABLE XIL 
 
 "0 
 
 72-6 
 
 o a 
 
 4wo a 
 ft.-lbs. 
 
 "i 
 
 i 
 
 Imvi"- 
 ft -Ibs. 
 
 iwiv 2 
 - imvi"- 
 ft.-lbs. 
 
 Ft.-lbs. 
 per minute. 
 
 Assumed 
 ft.-lbs. pel' 
 minute at 
 dynamo 
 brushes. 
 
 5270 
 
 159,600 
 
 69-0 
 
 4760 
 
 144,000 
 
 15,600 
 
 304,000 
 
 252,000 
 
 73-3 
 
 5370 
 
 319,000 
 
 69-5 
 
 4830 
 
 287,000 
 
 32,000 
 
 596,000 
 
 604,000 
 
 73-2 
 
 5360 
 
 495,000 
 
 69-4 
 
 4820 
 
 445,000 
 
 50,000 
 
 950,000 
 
 1,120,000 
 
 80-9 
 
 6550 
 
 693,000 
 
 76-7 
 
 5890 
 
 623,000 
 
 70,000 
 
 1,320,000 
 
 1,400,000 
 
 The limiting velocity for the rims of cast-iron wheels 
 varies from 80 to 100 feet per second (see Unwin's "Machine 
 Design "). 
 
 In the engines at Leeds, which have to deliver 400 H.P. 
 to the dynamo pulley, the piston speed is not greater than 
 
212 
 
 ELECTRICAL TRACTION. 
 
 490 feet per minute, with a boiler pressure of not less than 
 115 Ibs. nor more than 120 Ibs. This piston speed corre- 
 sponds to about 100 revolutions per minute. The fly-wheel 
 has a diameter of 15 feet and weighs 19 tons. Comparing 
 this roughly with Mclntosh and Seymour's 400 indicated 
 horse-power engine, we see that the revolutions per minute 
 are 195 as against 98, and the two fly-wheels combined have 
 
 5000 
 
 4000 
 
 FIG. 75. 
 
 a weight of 14,000 Ibs. and a diameter of 88 inches, as 
 against 42,560 Ibs. and a diameter of 180 inches for the 
 Leeds engines. 
 
 Size of Unit. The size of units to be employed in 
 traction plants deserves special attention, since the experi- 
 ence gained in electric lighting does not wholly apply. 
 The mean load-curve of a traction station is of course not 
 constant, as is shown by the curve in Fig. 75, which refers to 
 
SIZE OF UNIT. 
 
 2 13 
 
 the Philadelphia traction power station, and Mr. McMahon's 
 remarks,* with regard to the City and South London Railway, 
 agree with the same statement. At the same time there is no 
 doubt a wide difference between the curve of a traction 
 station and an electric lighting station ; that is to say, in 
 traction, the engineer is warranted in putting in larger and 
 fewer units than in electric supply, on account of the more 
 constant average load and greater proportion of the twenty- 
 four hours during which it is experienced. Dawson gives 
 Table XI II., which refers to the proposed relation between 
 the total indicated horse-power and the number and size of 
 units (see "Electric Railways and Tramways," p. 218). 
 
 TABLE XIII. 
 
 Maximum indicated 
 
 Number of 
 
 Indicated horse- 
 
 horse-power required 
 to work road. 
 
 engines 
 recommended. 
 
 power of each 
 engine. 
 
 200 
 
 2 
 
 200 
 
 400 
 
 3 
 
 200 
 
 GOO 
 
 3 
 
 300 
 
 1,0(10 
 
 3 
 
 500 
 
 1,500 
 
 4 
 
 500 
 
 2,000 4 
 
 750 
 
 5,000 
 
 6 
 
 1000 
 
 10,000 
 
 8 
 
 2000 
 
 As a recent example in actual practice we may look at 
 the Bristol Station (see also Dublin, p. 187 ; Leeds, p. 195). 
 
 Fig. 7G gives a sectional elevation of the Bristol electric 
 tramway station. It exhibits the best practice at the 
 present time, and was designed by Mr. Parshall, and carried 
 out by the British Thomson-Houston Company. It con- 
 tains four Adamson Lancashire boilers, working at about 
 1GO Ibs. pressure ; the length of the boilers is 30 feet, 
 and the diameter inside 7 feet G inches, fed with Vicar's 
 mechanical stokers and Green's economizers, driven by an 
 * See Journal of Institution of Electrical Engineers, vol. xxvi. p. 483. 
 
STEAM-ENGINE TESTS. 215 
 
 electromotor. Two feed-puinps, capable of supplying 
 16,000 Ibs. of water per hour to the boilers, are also driven 
 by electromotors. The engine room contains Mclntosh 
 and Seymour engines of the horizontal type, running at 
 200 revolutions per minute. The total weight of each 
 engine is 46,000 Ibs., and the weight of each fly-wheel is 
 4500 Ibs. There are four of these engines direct coupled 
 to four 150-kilo-watt six-pole generators with toothed 
 armatures, having 95 per cent, efficiency at full load, and 
 88 per cent, at quarter load. The guaranteed speed regula- 
 tion of these engines is two per cent, between no load and 
 full load. The governor we have already illustrated and 
 described. There are 22 motor-cars on this line, each 
 capable of seating 18 persons inside and 26 outside, and 
 these carry storage-cells for lighting purposes. The 
 charging of these cells is carried out by means of a motor- 
 generator, having an output of 135 volts and 230 amperes. 
 
 Direct Coupling. There is no doubt that direct coupling 
 between engines and dynamos has advantages over other 
 modes of driving, and its suitability is amply demonstrated 
 both in America and in Europe. At the same time autho- 
 rities differ. A notable instance in the United Kingdom 
 of conversion from rope to direct coupling is Bristol ; 
 and there is little doubt but that for sizes of generators 
 over 100 kilo-watts output direct coupling is the best. It 
 is certainly more economical to do away with belts and rope, 
 since energy and the cost of such belts are thereby saved, 
 and at the same time the floor space is diminished. 
 
 Steam-engine Tests. The testing of engines and boilers 
 or turbines, if water-power is used, is of importance, since 
 the efficiency is specified, and the engineer must satisfy 
 himself that it is attained. The test imposed upon the 
 contractors by Dr. J. Hopkinson in the case of the Leeds 
 steam-engines is as follows : 
 
 " The engine shall be run with a Buss tachometer attached, 
 and the maximum load of about 400 brake horse-power 
 
216 ELECTRICAL TRACTION. 
 
 shall be thrown on and off as many times as the engineer 
 may consider desirable ; the speed shall at no time be less 
 than five per cent, below the normal speed, or more than 
 five per cent, above the normal speed. If the engineer 
 sees fit, the engines may further be tested as follows : 
 The engine will be run with its dynamo machine fully 
 loaded so as to receive about 300 H.P., and the engine 
 will be indicated from time to time and the mean taken. 
 The engine will be run with its dynamo machines light, 
 and the engines will be indicated. The difference between 
 the two indications shall be taken as the net power of the 
 engine. The steam supplied during four hours or there- 
 abouts to the engine and its jackets shall be determined. 
 If the condenser or circulating pumps are driven by a 
 separate engine, the steam supplied to such engine shall 
 be added thereto. The steam supplied shall not be more 
 than 21 Ibs. per net horse-power." 
 
 The Generator. In America and Europe the generator 
 for traction work has received special attention at the hands 
 of the designers. In America the multi-polar machine is 
 almost universally used at the present time, and for large 
 sizes there is no doubt that this is the best type to adopt. 
 Fig. 77 illustrates one of these large direct coupled gene- 
 rators in motion. 
 
 The subject of slotted or toothed and smooth core 
 armatures in generators is of great interest and importance 
 (see Chapter II.), and although the toothed armature is 
 largely used, it is not universal. The Leeds generators 
 have smooth cores, and the City and South London 
 Railway motors and generators* are other instances ; yet 
 in America the toothed armature is almost universally used, 
 and there seems little doubt but that traction motors at 
 any rate will have toothed armatures for reasons already 
 discussed (p. 15). The subject of armature reaction and 
 brushes has been discussed in Chapters II. and IX. 
 
 The testing of dynamos should receive special attention 
 
218 ELECTRICAL TRACTION. 
 
 from the following points of view first, sparking ; second, 
 maximum rise of temperature ; and third, efficiency. 
 
 Since the brashes must be fixed, it follows that, to properly 
 test the machines for sparking, the load should be rapidly 
 varied, and this during a run of at least six hours' duration, 
 although in traction work the machines sometimes have to 
 be run for days together. 
 
 The maximum temperature to which a machine will rise 
 if running at full load requires careful attention. The most 
 stringent temperature limit in dynamos is usually that 
 known as the Admiralty temperature specification. This 
 states that the maximum rise of temperature of any part of 
 the dynamo above the atmosphere shall not exceed 70 Fah. 
 after six hours' run at full load. The taking of temperature 
 by means of thermometers is not, of course, the best method, 
 but it is the most practicable. The temperatures specified 
 are to be taken on thermometers in the usual way by placing 
 them on the parts to be examined, and protecting the bulb 
 from air draughts. The average temperatures may be taken 
 by the measurement of resistance.* 
 
 The subject of efficiency of dynamos is best dealt with by 
 the differential test, if two similar machines are available, 
 and this is the case in most power-houses. The differential 
 test was devised by Drs. J. and E. Hopkinson, and given in 
 the Philosophical Transactions of the Royal Society, 1886. 
 It can, however, be carried out in such a manner that all the 
 measurements are electrical. This is advisable when a third 
 dynamo, or storage battery, is available to supply the losses 
 in the system. There are two methods by which this 
 can be carried out (a) by placing a battery or dynamo 
 in series with the two armatures ; (b) by placing it in 
 parallel with the two armatures. This subject is dealt with 
 in Engineering, March 24, 1893. The Greenwood and 
 Batley generators at Leeds have (June, 1897) been tested 
 differentially by means of method (a) above mentioned 
 * See Electrician, August 28, 1896. 
 
GENERATORS AND SWITCHBOARDS. 
 
 219 
 
 (see p. 195). The differential testing of alternators is set 
 forth in the Philosophical Transactions of the Royal Society, 
 vol. 187, 1896, A, pp. 229-252. 
 
 Switchboards. The arrangement of switchboards must 
 be such that extension is easy, and in keeping with the rest of 
 the station. This has given rise to the panel system of 
 switchboards, whereby the various switches, complete for a 
 given purpose, can be mounted on a panel of slate or marble 
 and placed in line with those already installed. This makes 
 the arrangement very flexible ; large firms supply standard 
 panels for dealing with feeders, generators, etc. 
 
 The over-compounded machine is largely used in traction 
 
 FIG. 78. 
 
 stations, as is well known ; that is, at full load it gives, say, 
 550 volts, and on open circuit 500 volts, the difference 
 being due to the extra magnetization due to the series 
 and shunt turns, since the potential difference across the 
 shunt increases with the load, and the current to the external 
 circuit passes through the series turns. It is therefore 
 necessary to study the throwing of compound machines in 
 parallel with others running in the station. In Fig. 78 let 
 A C be the positive and negative omnibus bars of a system 
 of feeders in the station, and 1, 2, 3, compound dynamo, 
 of which No. 1 is running ; it is required to place No. 2 
 
220 ELECTRICAL TRACTION. 
 
 in parallel with No. 1 ; close a. 2 and excite the shunt of 
 No. 2 machine across the mains ; then close the main 
 switches J 2 , c- 2 , thereby placing the series turns of No. 2 in 
 parallel with those of No. 1. The machine must now run 
 at such a speed that the potential difference between its 
 terminals is equal to that between A and C ; the switch c! 2 can 
 then be closed, placing the machine in the circuit. The extra 
 shunt resistance serves to adjust the field, so that at normal 
 speed the proper potential may be obtained before closing rf 2 . 
 Further than this, provision is generally made for running 
 the machines half compounded for the feeders of shorter 
 length, in which case the series coils are placed in parallel, 
 there being two separate windings ; for fully compound, they 
 are generally in series. In the case where batteries are 
 employed, it is necessary to operate the machine as a simple 
 shunt when charging, and since the potential difference 
 must be large, it follows that when working compound the 
 shunt magnetizing force must be relatively small, so that 
 when working shunt it can be increased to give the proper 
 potential difference. If the limits of potential difference 
 between simple shunt on full load and compound on open 
 circuit are too great, the machine will either be working too 
 low on the characteristic curve, in the latter case, giving 
 rise to instability, in the sense that small variations of speed 
 and changes of temperatures will greatly influence the 
 potential ; or, in the former case, the shunt coil will have 
 to supply an unusually large magnetizing force, making the 
 machine expensive to build. The change from simple shunt 
 to compound is easily effected by short-circuiting the series 
 turns. 
 
 There are various devices for breaking the circuit 
 automatically if the current gets too great, and these are 
 preferable to the ordinary fuse, since they act quicker, and 
 are not so troublesome to replace. The breaking of the 
 main switch first and the subsequent separation of carbon 
 cylinders, which are easily replaced and take the spark, is one 
 
PARALLEL RUNNING OF DYNAMOS. 
 
 221 
 
 method ; the magnetic blow out is another. This is due to 
 Professor Elihu Thomson, and consists of a substantial 
 pair of electrodes which take the sparking as just mentioned; 
 but these are placed in a strong magnetic field produced 
 by the current itself, such field immediately blowing out 
 the arc formed. 
 
 Fig. 79 gives a view of the British Thomson-Houston 
 automatic circuit breaker, and Fig. 80 a diagram of con- 
 nections for the same. 
 When the circuit is 
 closed, the current 
 flows through a sole- 
 noid T to a contact 
 block M, through the 
 yoke Y to the block 
 MU and thence to the 
 outside circuit. If the 
 current exceeds a pre- 
 determined value, de- 
 pending upon the 
 tension of the spring 
 S, Fig. 79, the arma- 
 ture A is attracted, 
 lifting the lever L. 
 The yoke Y is imme- 
 diately drawn down 
 from the contact blocks 
 M, M by a powerful 
 spring, and the cur- 
 rent then passes 
 through the coils B,B (Fig. 80), and upper contacts C,C. 
 The current produces a powerful field between the poles of 
 an electro-magnet, of which B,B are the magnetizing coils, 
 the contacts 0,0 being directly under the influence of such 
 field. When the bridge-piece P ultimately severs contact 
 between 0,0, the spark is at once suppressed. 
 
 FIG. 79. 
 
222 
 
 ELECTRICAL TRACTION. 
 
 For the breaking of a shunt-circuit the device above 
 mentioned can be adopted ; but by another method the 
 shunt coil is momentarily placed in parallel with a non- 
 inductive resistance, through which the excessive potential 
 
 Slow-out Contact 
 
 Magnet Coil 
 
 FIG. 
 
 difference can send a current until such potential difference 
 has subsided (see Engineering, December 11, 1892). 
 
 The Leeds switchboard is shown in Fig. 81, and is an 
 example of a modern traction board for direct currents. 
 The drawing of the switchboard should be studied in 
 conjunction with Fig. 22. Provision is made for handling 
 four dynamos, each being so arranged that it can be coupled 
 full compound, half compound, or simple shunt ; but close 
 parallel running, as shown in Fig. 78, is not in the first 
 instance to be arranged for. 
 
 The first four panels are devoted each to one machine. 
 On each are mounted an automatic magnetic cut-out, a 
 double pole switch, a Weston ammeter, a Ferranti ampere- 
 hour meter, and shunt resistance regulator. An arrange- 
 ment is provided whereby all the cut-outs can be simul- 
 taneously put back if more than one should fall at the same 
 time* 
 
LEEDS SWITCHBOARD. 
 
 223 
 
224 ELECTRICAL TRACTION. 
 
 Panel 5 supports the positive and negative poles of the 
 four dynamos, with the screw-plug board for connecting 
 each machine to the feeders, whether these be for 
 charging batteries or simple feeding on to the line, or 
 for station purposes, such as lighting, and motors for cranes, 
 pumps, etc. Two recording voltmeters, reading up to 
 10 volts, are mounted on this panel, these being connected 
 each to the middle and extreme ends of the rail return in 
 accordance with the Board of Trade regulations (see 
 Appendix, p. 227, Regulation 7). 
 
 Panel G supports four circuit breakers and recording 
 ammeters for the four feeders. 
 
 Panel 7 is for dealing with the motors and lighting in the 
 station. 
 
 An automatic arrangement is provided on one machine, in 
 the first instance, for short circuiting the series coils of the 
 dynamos in the event of there being a tendency to reverse 
 the main current. 
 
 Three Kelvin electrostatic voltmeters are placed on the 
 upper central panels, and provision is made for placing these 
 across any of the machine terminals by means of flexible 
 twin wires. 
 
 The subject of switchboards and their connections is so 
 wide as to almost demand a separate work. The Author 
 has not space to deal exhaustively with it. 
 
APPENDIX. 
 
 STATUTORY RULES AND ORDERS, 1897. 
 No. 145. 
 
 TRAMWAYS, ENGLAND. 
 
 R. 363/97. 
 
 Regulations, dated February 11, 1897, made by the Board of Trade 
 under the Provisions of Section 12 of the Blackpool Corpora- 
 tion Tramways Order, 1896, ivhich was confirmed by the 
 Tramways Orders Confirmation (No. 1) Act, 1896, for regu- 
 lating the Employment of Insulated Returns, or of Uninsulated 
 Metallic Returns of Low Resistance ; for preventing Fusion 
 or injurious Electrolytic Action of or on Gas or Water Pipes 
 or other Metallic Pipes, Structures, or Substances ; and for 
 minimizing as far as is reasonaQy practicable Injurious 
 Interference with the Electric Wires, Lines, and Apparatus of 
 Parties other than the Promoters, and the Currents therein 
 whether such Lines do or-dd'npt use the Earth as a Return. 
 
 . , Definitions. 
 
 In the following regulations 
 
 The expression " energy " means electrical energy. 
 The expression " generator " means the dynamo or dynamos or 
 other electrical apparatus used for the* generation of energy. 
 The expression " motor " means any electric motor carried on a 
 
 car and used for the conversion of energy. 
 
 The expression " pipe " means any gas or water pipe or other 
 metallic pipe, structure, or substance. 
 
 Q 
 
226 ELECTRICAL TRACTION. 
 
 The expression " wire " means any wire or apparatus used for 
 
 telegraphic, telephonic, electrical signalling, or other similar 
 
 purposes. 
 The expression ''current" means an electric current exceeding 
 
 one- thousandth part of one ampere. 
 The expression "the Promoters" has the same meaning or 
 
 meanings as in sect. 12 of the Blackpool Corporation 
 
 Tramways Order, 1896. 
 
 Regulations. 
 
 1. Any dynamo used as a generator shall be of such pattern and 
 construction as to be capable of producing a continuous current 
 without appreciable pulsation. 
 
 2. One of the two conductors used for transmitting energy from 
 the generator to the motors shall be in every case insulated from 
 earth, and is hereinafter referred to as the " line ; " the other may 
 be insulated throughout, or may be uninsulated in such parts and 
 to such extent as is provided in the following regulations, and is 
 hereinafter referred to as the " return." 
 
 3. Where any rails on which cars run or any conductors laid 
 between or within three feet of such rails form any part of a return, 
 such part may be uninsulated. All other returns or parts of a 
 return shall be insulated, unless of such sectional area as will reduce 
 the difference of potential between the ends of the uninsulated 
 portion of the return below the limit laid down in Regulation 7. 
 
 4. When any uninsulated conductor laid between or within three 
 feet of the rails forms any part of a return, it shall be electrically 
 connected to the rails at distances apart not exceeding 100 feet by 
 means of copper strips having a sectional area of at least one- 
 sixteenth of a square inch, or by other means of equal conductivity. 
 
 5. When any part of a return is uninsulated it shall be con- 
 nected with the negative terminal of the generator, and in such 
 case the negative terminal of the generator shall also be directly 
 connected, through the current-indicator hereinafter mentioned, to 
 two separate earth connections which shall be placed not less than 
 20 yards apart. 
 
 Provided that in place of such two earth connections the 
 Promoters may make one connection to a main for water supply 
 of not less than three inches internal diameter, with the consent 
 of the owner thereof and of the person supplying the water, and 
 provided that where, from the nature of the soil or for other reasons, 
 the Promoters can show to the satisfaction of an inspecting officer 
 of the Board of Trade that the earth connections herein specified 
 cannot be constructed and maintained without undue expense the 
 provisions of this regulation shall not apply. 
 
APPENDIX. 227 
 
 The earth connections referred to in this regulation shall be con- 
 structed, laid, and maintained so as to secure electrical contact with 
 the general mass of earth, and so that an electro-motive force, not 
 exceeding four volts, shall suffice to produce a current of at least 
 two amperes from one earth connection to the other through the 
 earth, and a test shall be made at least once in every month to 
 ascertain whether this requirement is complied with. 
 
 No portion of either earth connection shall be placed within six 
 feet of any pipe except a main for water supply of not less than three 
 inches internal diameter which is metallically connected to the 
 earth connections with the consents hereinbefore specified. 
 
 6. When the return is partly or entirely uninsulated the Pro- 
 moters shall in the construction and maintenance of the tramway 
 (a) so separate the uninsulated return from the general mass of 
 earth, and from any pipe in the vicinity ; (ft) so connect together 
 the several lengths of the rails ; (c) adopt such means for reducing 
 the difference produced by the current between the potential of the 
 uninsulated return at any one point and the potential of the uninsu- 
 lated return at any other point ; and (d) so maintain the efficiency 
 of the earth connections specified in the preceding regulations as 
 to fulfil the following conditions, viz. : 
 
 (i.) That the current passing from the earth connections through 
 the indicator to the generator shall not at any time exceed 
 either two amperes per mile of single tramway line or five 
 per cent, of the total current output of the station, 
 (ii.) That if at any time and at any place a test be made by con- 
 necting a galvanometer or other current-indicator to the 
 uninsulated return and to any pipe in the vicinity, it shall 
 always be possible to reverse the direction of any current 
 indicated by interposing a battery of three Leclanche cells 
 connected in series if the direction of the current is from 
 the return to the pipe, or by interposing one Leclanche 
 cell if the direction of the current is from the pipe to the 
 return. 
 
 In order to provide a continuous indication that the condition 
 (i.) is complied with, the Promoters shall place in a conspicuous 
 position a suitable, properly connected, and correctly marked 
 current-indicator, and shall keep it connected during the whole 
 time that the line is charged. 
 
 The owner of any such pipe may require the Promoters to permit 
 him at reasonable times and intervals to ascertain by test that the 
 conditions specified in (ii.) are complied with as regards his pipe. 
 
 7. When the return is partly or entirely uninsulated a con- 
 tinuous record shall be kept by the Promoters of the difference of 
 potential during the working of the tramway between the points of 
 
228 ELECTRICAL TRACTION. 
 
 the uninsulated return furthest from and nearest to the generating 
 station. If at any time such difference of potential exceeds the 
 limit of seven volts, the promoters shall take immediate steps to 
 reduce it below that limit. 
 
 8. Every electrical connection with any pipe shall be so arranged 
 as to admit of easy examination, and shall be tested by the Promoters 
 at least once in every three months. 
 
 9. Every line and every insulated return or part of a return 
 except any feeder shall be constructed in sections not exceeding 
 one-half of a mile in length, and means shall be provided for 
 isolating each such section for purposes of testing. 
 
 10. The insulation of the line and of the return when insulated 
 and of all feeders and other conductors, shall be so maintained that 
 the leakage current shall not exceed one-hundredth of an ampere 
 per mile of tramway. The leakage current shall be ascertained 
 daily before or after the hours of running when the line is fully 
 charged. If at any time it should be found that the leakage current 
 exceeds one-half of an ampere per mile of tramway the leak shall 
 be localised and removed as soon as practicable, and the running of 
 the cars shall be stopped unless the leak is localised and removed 
 within 24 hours. Provided that where both line and return are 
 placed within a conduit this regulation shall not apply. 
 
 11. The insulation resistance of all continuously insulated cables 
 used for lines, for insulated returns, for feeders, or for other pur- 
 poses, and laid below the surface of the ground, shall not be per- 
 mitted to fall below the equivalent of 10 megohms for a length of 
 one mile. A test of the insulation resistance of all such cables 
 shall be made at least once in each month. 
 
 12. Where in any case in any part of the tramway the line is 
 erected overhead arid the return is laid on or under the ground, and 
 where any wires have been erected or laid before the construction 
 of the tramway in the same or nearly the same direction as such 
 part of the tramway, the Promoters shall, if required so to do by the 
 owners of such wires or any of them, permit such owners to insert 
 and maintain in the Promoters' line one or more induction coils or 
 other apparatus approved by the Promoters for the purpose of pre- 
 venting disturbance by electric induction. In any case in which 
 the Promoters withhold their approval of any such apparatus the 
 owners may appeal to the Board of Trade, who may, if they think 
 fit, dispense with such approval. 
 
 13. Any insulated return shall be placed parallel to and at a 
 distance not exceeding three feet from the line when the line and 
 return are both erected overhead, or eighteen inches when they are 
 both laid underground. 
 
 14. In the disposition, connections, and working of feeders the 
 
APPENDIX. 229 
 
 Promoters shall take all reasonable precautions to avoid injurious 
 interference with any existing wires. 
 
 15. The Promoters shall so construct and maintain their system 
 as to secure good contact between the motors and the line and 
 return respectively. 
 
 16. The Promoters shall adopt the best means available to pre- 
 vent the occurrence of undue sparking at the rubbing or rolling 
 contacts in any place and in the construction and use of their 
 generator and motors. 
 
 17. In working the cars the current shall be varied as required 
 by means of a rheostat containing at least 20 sections, or by some 
 other equally efficient method of gradually varying resistance. 
 
 18. Where the line or return or both are laid in a conduit the 
 following conditions shall be complied with in the construction and 
 maintenance of such conduit : 
 
 (a) The. conduit shall be so constructed as to admit of easy 
 
 examination of and access to the conductors contained 
 therein and their insulators and supports. 
 
 (b) It shall be so constructed as to be readily cleared of accumu- 
 
 lation of dust or other debris, and no such accumulation shall 
 be permitted to remain. 
 
 (c) It shall be laid to such falls and so connected to sumps or 
 
 other means of drainage, as to automatically clear itself of 
 water without danger of the water reaching the level of the 
 conductors. 
 
 (d) If the conduit is formed of metal, all separate lengths shall 
 be so jointed as to secure efficient metallic continuity for the 
 passage of electric currents. Where the rails are used to 
 form any part of the return they shall be electrically con- 
 nected to the conduit by means of copper strips having a 
 sectional area of at least one-sixteenth of a square inch, or 
 other means of equal conductivity, at distances apart not 
 exceeding 100 feet. Where the return is wholly insulated 
 and contained within the conduit, the latter shall be con- 
 nected to earth at the generating station through a high 
 resistance galvanometer suitable for the indication of any 
 contact or partial contact of either the line or the return 
 with the conduit. 
 
 (e) If the conduit is formed of any non-metallic material not 
 
 being of high insulating quality and impervious to moisture 
 throughout, and is placed within six feet of any pipe, a non- 
 conducting screen shall be interposed between the conduit 
 and the pipe, of such material and dimensions as shall 
 provide that no current can pass between them without 
 traversing at least six feet of earth, or the circuit itself shall 
 
2jO ELECTRICAL TRACTION. 
 
 in such case be lined with bitumen or other non-conducting 
 damp-resisting material in all cases where it is placed within 
 six feet of any pipe. 
 
 (/) The leakage current shall be ascertained daily, before or 
 after the hours of running when the line is fully charged, 
 and if at any time it shall be found to exceed half an ampere 
 per mile of tramway the leak shall be localised and removed 
 as soon as practicable, and the running of the cars shall be 
 stopped unless the leak is localised and removed within 
 24 hours. 
 
 19. The Promoters shall, so far as may be applicable to their 
 system of working, keep records as specified below. These records 
 shall, if and when required, be forwarded for the information of the 
 Board of Trade. 
 
 Daily Records. 
 No. of cars running. 
 Maximum working current. 
 Maximum working pressure. 
 
 Maximum current from the earth connections (vide Regulation 
 6 (i.)). 
 
 Leakage current (vide Regulations 10 and 18 (/)). 
 Fall of potential in return (vide Regulation 7). 
 
 Monthly Records. 
 
 Condition of earth connections (vide Regulation 5). 
 Insulation resistance of insulated cables (vide Regulation 11). 
 
 Quarterly Records. 
 Conductance of joints to pipes (vide Regulation 8). 
 
 Occasional Records. 
 
 Any tests made under provisions of Regulation 6 (ii.). 
 Localisation and removal of leakage, stating time occupied. 
 Particulars of any abnormal occurrence affecting the electric 
 working of the tramway. 
 
 Signed by order of the Board of Trade this llth day of 
 February, 1897. 
 
 Francis J. S. Hop wood, 
 
 Assistant Secretary 
 Board of Trade. 
 
APPENDIX. 231 
 
 STATUTORY RULES AND ORDERS, 1896. 
 
 No. 747. 
 TRAMWAY. 
 
 Regulations, dated August 17, 1896, made by the Board of Trade 
 as regards Electrical Power on the Dublin Southern District 
 Tramways. 
 
 R. 13,471/96. 
 
 The Board of Trade, under and by virtue of the powers conferred 
 upon them in this behalf, do hereby order that the following regula- 
 tions for securing to the public reasonable protection against danger 
 in the exercise of the powers conferred by Parliament with respect 
 to the use of electrical power on all or any of the tramways on 
 which the use of such power has been authorised by the Dublin 
 Southern District Tramways Act, 1893 (hereinafter called '' the 
 tramways "). be substituted for all other regulations in this behalf 
 contained in any Tramway Act or Tramway Order confirmed by 
 Act of Parliament : 
 
 And the Board of Trade do also hereby make the following bye- 
 laws with regard to the use of electrical power on all or any of 
 such tramways. 
 
 Regulations. 
 
 I. Every motor carriage used on the tramways shall comply with 
 the following requirements, that is to say : 
 
 (a) The wheels shall be fitted with brake-blocks, which can be 
 applied by a screw or treadle, or by other means, and there 
 shall be in addition an adequate electric brake. 
 
 (&) It shall be fitted within six months from the date hereof with 
 a governor which cannot be tampered with by the driver, 
 and which shall operate so as to cut off all electric current 
 from the motors whenever the speed exceeds ten miles 
 an hour. 
 
 (c) It shall be numbered inside and outside, and the number shall 
 
 be shown in conspicuous parts thereof. 
 
 (d) It shall be fitted with a suitable fender, which will act 
 
 efficiently as a life protector, and with a special bell or 
 whistle to be sounded as a warning when necessarj 7 . 
 
 (e) It shall be so constructed as to enable the driver to command 
 
 the fullest possible view of the road before him. 
 (/) It shall be free from the clatter of machinery, such as to 
 
232 ELECTRICAL TRACTION. 
 
 constitute any reasonable ground of complaint, either to the 
 passengers or to the public, and any machinery under the 
 carriage shall be concealed from view at all points above 
 four inches from the level of the rails. 
 
 (</) When running between sunset and sunrise, or during fog, it 
 shall carry in front a bright-coloured light. 
 
 II. Every trailing carriage used on the tramways shall comply 
 with the following requirements, that is to say : 
 
 (a.) The wheels shall be fitted with brake-blocks, which can be 
 applied by a screw or treadle or by other means. 
 
 (&.) It shall be numbered inside and outside, and the number 
 shall be shown in conspicuous parts thereof. 
 
 III. Not more than two carriages shall be coupled together, and 
 when two are so running there shall be, in addition to the conductor, 
 a man on the front platform of the second carriage, whose sole duty 
 it shall be to attend to the brake, means being provided by which 
 the driver can signal to this man when he wishes the brake on the 
 rear carriage to be applied. The carriages shall be connected by 
 double couplings, one of which shall be a screw coupling. 
 
 IV. Every carriage used on the tramways shall be so constructed 
 as to provide for the safety of passengers, and for their safe entrance 
 to, exit from, and accommodation in such carriages, and for their 
 protection from the apparatus used for drawing or propelling the 
 carriages. 
 
 V. The Board of Trade and their officers may, from time to 
 time, and shall, on the application of the local authority of any of 
 the districts through which the said tramways pass, inspect the 
 carriages used on the tramways, and the working arrangements 
 generally, and may, whenever they think fit, prohibit the use on 
 the tramways of any of them which, in their opinion, are not safe 
 for use. 
 
 VI. The speed at which the carriages shall be driven or pro- 
 pelled along the tramways shall not exceed the rate of eight miles 
 an hour, and the speed at which the carriages shall pass through 
 facing points, whether fixed or movable, shall not exceed the rate 
 of four miles an hour. 
 
 VII. The speed shall not exceed the rate of four miles an hour 
 in Upper George Street between Mulgrave Street and Mellifont 
 Avenue, or on the road between Merrion Avenue and the boundary 
 between the parishes of Booterstpwn and Monkstown, and not 
 more than one carriage or two carriages coupled together shall be 
 allowed on the first-mentioned portion of the tramways at one and 
 the same time. 
 
 VIII. The passengers shall not have access to any portion of the 
 electric circuit. 
 
APPENDIX. 233 
 
 IX. All electric mains, leads and connections used must be of 
 ample size, and must be thoroughly insulated and protected by- 
 safety fuses or other cut-outs which will operate to break the circuit 
 before the current has risen to an amount which would cause any 
 injurious heating of the conductors, and the length of any safety 
 fuse in the clear shall not be less than two inches. 
 
 X. The electrical pressure or difference of potential between any 
 suspended conductors used in connection with the working of the 
 tramways by electrical power and the earth, or between any two 
 such suspended conductors, shall in no case exceed 500 volts con- 
 tinuous pressure. 
 
 XI. The suspended conductors used in connection with the 
 working of the tramways by electrical power shall be in no part 
 at a less height from the surface of the street than 17 feet, and 
 shall be securely attached to supports at intervals not exceeding 
 120 feet. 
 
 XII. The line-wire shall be divided up into sections not exceeding 
 (except with the special approval of the Board of Trade) one 
 quarter of a mile in length, between every two of which shall be 
 inserted an emergency switch and a safety fuse or cut-out, con- 
 structed to act with a current exceeding the maximum working 
 current by 50 per cent., which apparatus shall be so enclosed as to 
 be inaccessible to pedestrians. 
 
 XIII. The electrical pressure between the conductors in any 
 electric line or between any such conductor and the earth shall not 
 in any case exceed 3000 volts. 
 
 XIV. All electric lines laid for the purpose of supply to trans- 
 forming stations on the " three-phase system " shall have their 
 conductors arranged concentrically, the outer conductor being 
 efficiently connected with earth at the generating station, but insu- 
 lated at all other points ; and the thickness of insulation between 
 the several conductors of any such electric line shall not be less in 
 parts of an inch than the number obtained by dividing the number 
 expressing the maximum electrical pressure in volts by 20,000. 
 No such electric line shall be brought into use unless the insulation 
 of every part thereof has withstood the continuous application during 
 one hour of twice the maximum pressure to which it is intended to 
 be subjected in use. 
 
 XV. The sectional area of the conductor in any electric line laid 
 or erected in any street after the date of these regulations shall not 
 be less than the area of a circle of one-tenth of an inch diameter, 
 and where the conductor is formed of a strand of wires, each 
 separate wire shall be at least as large as No. 20 standard wire 
 gauge : Provided that this regulation shall not apply to any electric 
 line connected to the rails for the purpose of measuring the fall of 
 
234 ELECTRICAL TRACTION. 
 
 potential in the return and not otherwise connected with the 
 electric circuit. 
 
 XVI. No part of any electric line shall be used for the trans- 
 mission of more than 300,000 watts, except with the consent in 
 writing of the Board of Trade, and efficient means shall be provided 
 to prevent this limit being at any time exceeded. 
 
 XVII.* All electrical conductors fixed upon the carriages in con- 
 nection with the " trolley- wheel " shall be formed of flexible cables 
 protected by indiarubber insulation of the highest quality, and 
 additionally protected wherever they are adjacent to any metal so 
 as to avoid risk of the metal becoming charged. 
 
 The insulation resistance between these conductors and the 
 " trolley standard " and the metal fittings on the carriages respec- 
 tively shall be tested daily with the full working electrical pressure, 
 and shall not be permitted to fall below the following amounts 
 respectively, viz. : 
 
 Between conductors and trolley standard - - 10 megohms. 
 
 Between conductors and metal fittings - 1 megohm. 
 
 XVIIL* All metal fittings upon the roofs of the carriages within 
 six feet of the trolley standard shall be carefully covered with 
 insulating material to a thickness of at least 16th inch, and 
 this covering shall be constantly maintained in efficient condi- 
 tion. 
 
 XIX. An emergency cut-off switch shall be provided and fixed 
 so as to be conveniently reached by the driver in case of any failure 
 of action of the controller switch. 
 
 XX. Efficient guard wires shall be erected and maintained at all 
 places where telegraph or telephone wires cross above the electric 
 conductors of the tramways. 
 
 XXI. Where any portion of any electric line or any support for 
 an electric line is exposed in such a position as to be liable to 
 injury from lightning, it shall be efficiently protected against such 
 injury. 
 
 XXII. Where any accident by explosion or fire, or any other 
 accident of such kind as to have caused or to. be likely to have 
 caused loss of life or personal injury has occurred in connection with 
 the electric working of the tramways, immediate notice thereof 
 shall be given to the Board of Trade. 
 
 * The words " and additionally . . . becoming charged," and the 
 words, " and shall not be permitted ... 1 megohm," in this Rule are 
 omitted in the Regulations of November 19, 1896, for the Hartlepool 
 Tramways. See Regulation XVII., p. 239. 
 
 * The words in this Rule have been omitted in the Hartlepool 
 Tramways Regulations, date November 19, 1896, and others sub- 
 stituted. See Regulation XVIIL, p. 239. 
 
APPENDIX. 235 
 
 Penalty. 
 
 NOTE. Any company or person using electrical power on the 
 tramways contrary to any of the above regulations is, for every such 
 offence, subject to a penalty not exceeding 10, and also in the 
 case of a continuing offence, to a further penalty not exceeding 5 
 for every clay after the first during which such offence continues. 
 
 Byelaws. 
 
 I. The special bell or whistle shall be sounded by the driver of 
 the carriage from time to time when it is necessary as a warning. 
 
 II. Whenever it is necessary to avoid impending danger, the 
 carriages shall be brought to a standstill. 
 
 III. The entrance to and exit from the carriages shall be by 
 the hindermost or conductor's platform. 
 
 IV. The carriages shall be brought to a standstill immediately 
 before passing round the spot known as Hynes Corner. 
 
 V. A printed copy of these regulations and byelaws shall be 
 kept in a conspicuous position inside of each carriage in use on 
 the tramways. 
 
 Penalty. 
 
 NOTE. Any person or corporation offending against or com- 
 mitting a breach of any of these byelaws is liable to a penalty 
 not exceeding forty shillings. 
 
 The provisions of the Tramways Act, 1 870, with respect to the 
 recovery of penalties, are applicable to the penalties for the breach 
 of these regulations or byelaws. 
 
 Signed by order of the Board of Trade, this 17th day of 
 August, 1896. 
 
 Francis J. 8. Hopwood, 
 
 An Assistant Secretary to the 
 
 Board of Trade. 
 
236 ELECTRICAL TRACTION. 
 
 STATUTORY RULES AND ORDERS, 1896. 
 
 No. 997. 
 TRAMWAY. 
 
 Regulations, dated November 19, 1896, made by the Board of 
 Trade as regards Electrical Power on the Hartlepool 
 Tramways. 
 
 R. 15,846/96. 
 
 The Board of Trade, under and by virtue of the powers conferred 
 upon them in this behalf, do hereby order that the following regula- 
 tions for securing to the public reasonable protection against danger 
 in the exercise of the powers conferred by Parliament with respect 
 to the use of mechanical power on all or any of the tramways on 
 which the use of such power has been authorized by the Hartlepool 
 Tramways Order, 1883, which was confirmed by the Tramways 
 Orders Confirmation (No. 1) Act, 1883 (hereinafter called "the 
 tramways "), be substituted for all other regulations in this behalf 
 contained in any Tramway Act or Tramway Order confirmed by 
 Act of Parliament : 
 
 And the Board of Trade do also hereby make the following 
 byelaws with regard to the use of mechanical power on all or any 
 of such tramways. 
 
 Regulations. 
 
 I. Every motor carriage used on the tramways shall comply 
 with the following requirements, that is to say : 
 
 (a) The wheels shall be fitted with brake-blocks, which can bo 
 
 applied by a screw or treadle, or by other means, and there 
 shall be in addition an adequate electric brake. 
 
 (b) It shall be fitted within six months from the date hereof, 
 
 or within such further period as the Board of Trade may 
 prescribe, with a governor which cannot be tampered with 
 by the driver, and which shall operate so as to cut oft' all 
 electric current from the motors whenever the speed exceeds 
 ten miles an hour. 
 
 (c) It shall be numbered inside and outside, and the number 
 
 shall be shown in conspicuous parts thereof. 
 
 (d) It shall be fitted with a suitable fender, which will act 
 
 efficiently as a life protector, and with a special bell to 
 be sounded as a warning when necessary. 
 
APPENDIX. 237 
 
 (e) It shall be so constructed as to enable the driver to command 
 the fullest possible view of the road before him. 
 
 (/) It shall be free from the clatter of machinery, such as to 
 constitute any reasonable ground of complaint either to the 
 passengers or to the public, and any machinery under the 
 carriage shall be concealed from view at all points above 
 four inches from the level of the rails. 
 
 (g) When running between sunset and sunrise, or during fog, 
 it shall carry in front a bright-coloured light. 
 
 II. Every trailing carriage used on the tramways shall comply 
 with the following requirements, that is to say : 
 
 (a) The wheels shall be fitted with brake-blocks, which can be 
 applied by a screw or treadle or by other means. 
 
 (6) It shall be numbered inside and outside, and the number 
 shall be shown in conspicuous parts thereof. 
 
 III. Not more than two carriages shall be coupled together, 
 and when two are so running there shall be, in addition to the 
 conductor, a man on the front platform of the second carriage, 
 whose sole duty it shall be to attend to the brake, means being 
 provided by which the driver can signal to this man when he 
 wishes the brake on the rear carriage to be applied. The carriages 
 shall be connected by double couplings, one of which shall be a 
 screw coupling. 
 
 IV. Every carriage used on the tramways shall be so con- 
 structed as to provide for the safety of passengers, and for their 
 safe entrance to, exit from, and accommodation in such carriages, 
 and for their protection from the apparatus used for drawing or 
 propelling the carriages. 
 
 V. The Board of Trade and their officers may, from time to 
 time, and shall, on the application of the local authority of any 
 of the districts through which the said tramways pass, inspect 
 the carriages used on the tramways, and the working arrangements 
 generally, and may, whenever they think n't, prohibit the use on the 
 tramways of any of them which, in their opinion, are not safe for use. 
 
 VI. The speed at which the carriages shall be driven or pro- 
 pelled along the tramways shall not exceed the rate of eight miles 
 an hour, and the speed at which the carriages shall pass through 
 facing points, whether fixed or movable, shall not exceed the rate 
 of four miles an hour. 
 
 VII. The speed shall not exceed the rate of four miles an hour 
 at the following places : 
 
 (a) In Northgate Street, Hartlepool, between Union Street and 
 
 Tweddle Street. 
 (5) On the double line on each side of the bridge carrying the 
 
 North-Eastern Railway over Cleveland Road. 
 
238 ELECTRICAL TRACTION. 
 
 (c) In front of Christ Church Schools, West Hartlepool. 
 
 VIII. No two carriages shall enter any passing-place from 
 opposite directions at one and the same time. 
 
 IX. Notices shall be exhibited on each side of the two railway 
 bridges crossing the tramways, having in characters legible to 
 passengers the words : 
 
 It is dangerous to touch the electric wires. 
 
 X. The passengers shall not have access to any portion of the 
 electric circuit. 
 
 XT. All electric mains, leads and connections used must be of 
 ample size and must be thoroughly insulated and protected by 
 safety fuses or other cut-outs which will operate to break the 
 circuit before the current has risen to an amount which would 
 cause any injurious heating of the conductors, and the length of 
 any safety fuse in the clear shall not be less than two inches. 
 
 XII. The electrical pressure or difference of potential between 
 any suspended conductors used in connection with the working 
 of the tramways by electrical power and the earth, or between 
 any two such suspended conductors, shall in no case exceed 500 
 volts continuous pressure. 
 
 XIII. The suspended conductors used in connection with the 
 working of the tramways by electrical power shall be in no part 
 at a less height from the surface of the street than 17 feet, and 
 shall be securely attached to supports at intervals riot exceeding 
 120 feet. 
 
 XIV. The line-wire shall be divided up into sections not exceed- 
 ing (except with the special approval of the Board of Trade) one 
 quarter of a mile in length, between every two of which shall 
 be inserted an emergency switch and a safety fuze or cut-out con- 
 structed to act with a current exceeding the maximum working 
 current by 50 per cent., which apparatus shall be so enclosed as to 
 be inaccessible to pedestrians. 
 
 XV. The sectional area of the conductor in any electric line 
 laid or erected in any street after the date of these regulations 
 shall not be less than the area of a circle of one-tenth of an inch 
 diameter, and where the conductor is formed of a strand of wires, 
 each separate wire shall be at least as large as No. 20 standard 
 wire gauge : Provided that this regulation shall not apply to any 
 electric line connected to the rails for the purpose of measuring the 
 fall of potential in the return and not otherwise connected with 
 the electric circuit. 
 
 XVI. No part of any electric line shall be used for the trans- 
 mission of more than 300,000 watts, except with the consent in 
 writing of the Board of Trade, and efficient means shall be provided 
 to prevent this limit being at any time exceeded. 
 
APPENDIX. 239 
 
 XVII. All electrical conductors fixed upon the carriages in con- 
 nection with the " trolley-wheel " shall be formed of flexible cables 
 protected by indiarubber insulation of the highest quality. 
 
 The insulation resistance between these conductors and the 
 " trolley standard " and the metal fittings on the carriages 
 respectively shall be tested daily with the full working electrical 
 pressure. 
 
 XVIII. The " trolley standard " shall be maintained in efficient 
 metallic connection with the metal fittings on the roof of the car 
 and with the metal frame of the car. 
 
 XIX. An emergency cut-off switch shall be provided and fixed 
 so as to be conveniently reached by the driver in case of any failure 
 of action of the controller switch. 
 
 XX. Efficient guard wires shall be erected and maintained at all 
 places where telegraph or telephone wires cross above the electric 
 conductors of the tramway. 
 
 XXI. Where any portion of any electric line or any support for 
 an electric line is exposed in such a position as to be liable to 
 injury from lightning, it shall be efficiently protected against such 
 injury. 
 
 XXII. Where any accident by explosion or fire, or any other 
 accident of such kind as to have caused or to be likely to have 
 caused loss of life or personal injury has occurred in connection with 
 the electric working of the tramways, immediate notice thereof shall 
 be given to the Board of Trade. 
 
 Penalty. 
 
 NOTE. Any company or person using electrical power on the 
 tramways contrary to any of the above regulations is, for every 
 such offence, subject to a penalty not exceeding 10, and also in 
 the case of a continuing offence, to a further penalty not exceeding 
 5 for every day after the first during which such offence continues. 
 
 Byelaws. 
 
 I. The special bell shall be sounded by the driver of the carriage 
 from time to time when it is necessary as a warning. 
 
 II. Whenever it is necessary to avoid impending danger, the 
 carriages shall be brought to a standstill. 
 
 III. The entrance to and exit from the carriages shall be by the 
 hindermost or conductor's platform. 
 
 IV. The carriages shall be brought to a standstill at the following 
 points : 
 
 (a) Throston Iron Bridge. 
 
2 4 o ELECTRICAL TRACTION. 
 
 (6) The railway level crossing in Cleveland Road, if and when 
 the railway is used for traffic. 
 
 (c) The junction of Middleton Road and Clarence Road. 
 
 (d) The corner by the police station in Church Street, West 
 
 Hartlepool. 
 
 (e) The corner by the fountain in Church Street. 
 
 (/) The intersection of Whitby Street with Church Street. 
 
 V. A printed copy of these regulations and byelaws shall be kept 
 in a conspicuous position inside of each carriage in use on the 
 tramways. 
 
 Penalty. 
 
 NOTE. Any person or corporation offending against or com- 
 mitting a breach of any of these byelaws is liable to a penalty not 
 exceeding forty shillings. 
 
 The provisions of the Tramways Act, 1870, with respect to the 
 recovery of penalties, are applicable to the penalties for the breach 
 of these regulations or byelaws. 
 
 Signed by order of the Board of Trade, this 19th day 
 of November, 1896. 
 
 Francis J. S. Hop wood, 
 
 An Assistant Secretary to the 
 Board of Trade. 
 
 LOCOMOTIVES ON HIGHWAYS ACT, 1896. 
 REGULATIONS. GENERAL. 
 
 To THE COUNTY COUNCILS of the several ADMINISTRATIVE 
 COUNTIES in ENGLAND and WALES : 
 
 To the Councils of the several County Boroughs in England and 
 
 Wales ; 
 To the Sanitary Authorities of the several Sanitary Districts in 
 
 the Administrative County of London ; 
 To the Urban District Councils of the several Urban Districts in 
 
 England and Wales ; 
 To the Rural District Councils acting as the Highway Authorities 
 
 in Rural Districts in England and Wales ; 
 And to all others whom it may concern, 
 
APPENDIX. 241 
 
 WHEREAS by Section 6 of the Locomotives on Highways Act, 
 1896 (herein-after called " the Act "), it is enacted that 
 
 " (1) The Local Government Board may make regulations 
 " with respect to the use of light locomotives on highways, and 
 " their construction, and the conditions under which they may be 
 " used. 
 
 " (2) All regulations under this section 
 
 " shall have full effect notwithstanding anything in any other Act, 
 " whether general or local, or any byelaws or regulations made 
 " thereunder." 
 And whereas by Section 2 of the Act it is enacted that 
 
 u During the period between one hour after sunset and one 
 " hour before sunrise, the person in charge of a light locomotive 
 " shall carry attached thereto a lamp so constructed and placed as 
 " to exhibit a light in accordance with the regulations to be made 
 "by the Local Government Board." 
 And whereas by Section 7 of the Act it is enacted that 
 
 " A breach of any * regulation made under this 
 
 " Act, may, on summary conviction, be 
 
 " punished by a fine not exceeding ten pounds." 
 
 Now THEREFORE, in pursuance of the powers given to Us by 
 the Act, and by any other Statutes in that behalf, We, the Local 
 Government Board, Do by this Our Order make the following 
 Kegulations with respect to the use of Light Locomotives on High- 
 ways, and their construction, and the conditions under which they 
 may be used, and Direct that the same shall have effect on and 
 after the Fourteenth day of November, One thousand eight hundred 
 and ninety-six : 
 
 ARTICLE I. In this order 
 
 The expression "carriage" includes a waggon, cart, or other 
 vehicle. 
 
 The expression "horse" includes a mule or other beast of 
 draught or burden, and the expression " cattle " includes 
 sheep. 
 
 The expression " Light Locomotive " means a vehicle propelled 
 by mechanical power which is under three tons in weight 
 unladen, and is not used for the purpose of drawing more 
 than one vehicle (such vehicle with its locomotive not 
 exceeding in weight unladen four tons), and is so constructed 
 that no smoke or visible vapour is emitted therefrom except 
 from any temporary or accidental cause. 
 
 In calculating for the purposes of this Order the weight 
 of a vehicle unladen, the weight of any water, fuel, or 
 
 E 
 
242 ELECTRICAL TRACTION. 
 
 accumulators used for the purpose of propulsion shall not 
 be included. 
 
 ARTICLE II. No person shall cause or permit a Light Loco- 
 motive to be used on any highway, or shall drive or have charge of 
 a Light Locomotive when so used, unless the conditions hereinafter 
 set forth shall be satisfied, namely, 
 
 (1) The Light Locomotive, if it exceeds in weight unladen five 
 
 hundredweight, shall be capable of being so worked that it 
 may travel either forwards or backwards. 
 
 (2) The Light Locomotive shall not exceed six and a half feet in 
 
 width, such width to be measured between its extreme pro- 
 jecting points. 
 
 (3) The tyre of each wheel of the Light Locomotive shall be 
 
 smooth, and shall, where the same touches the ground, be 
 flat and of the width following, namely, 
 
 (a) if the weight of the Light Locomotive unladen 
 
 exceeds fifteen hundredweight, but does not 
 
 exceed one ton, not less than two and a half 
 
 inches ; 
 
 (6) if such weight exceeds one ton, but does not exceed 
 
 two tons, not less than three inches ; 
 (c) if such weight exceeds two tons, not less than four 
 
 inches. 
 
 Provided that where a pneumatic tyre or other tyre of a 
 soft and elastic material is used, the tyre may be round or 
 curved, and there may be upon the same projections or 
 bosses rising above the surface of the tyre if such projections 
 or bosses are of the same material as that of the tyre itself 
 or of some other soft and elastic material. The width of the 
 tyre shall, for the purpose of this proviso, mean the extreme 
 width of the soft and elastic material on the rim of the wheel 
 when not subject to pressure. 
 
 (4) The Light Locomotive shall have two independent brakes in 
 
 good working order, and of such efficiency that the appli- 
 cation of either to such Locomotive shall cause two of its 
 wheels on the same axle to be so held that the wheels shall 
 be effectually prevented from revolving, or shall have the 
 same effect in stopping the Light Locomotive as if such 
 wheels were so held. 
 
 Provided that in the case of a bicycle this Regulation 
 shall apply as if, instead of two wheels on the same axle, one 
 wheel was therein referred to. 
 
 (5) The Light Locomotive shall be so constructed as to admit 
 
 of its being at all times under such control as not to cause 
 
APPENDIX. 243 
 
 undue interference with passenger or other traffic on any 
 highway. 
 
 (6) In the case of a Light Locomotive drawing or constructed 
 
 to draw another vehicle, or constructed or used for the 
 carriage of goods, the name of the owner and the place of 
 his abode or business, and in every such case and in the 
 case of every Light Locomotive weighing unladen one ton 
 and a half or upwards, the weight of the Light Locomotive 
 unladen shall be painted in one or more straight lines upon 
 some conspicuous part of the right or off side of the Light 
 Locomotive in large legible letters in white upon black or 
 black upon white, not less than one inch in height. 
 
 (7) The Light Locomotive and all the fittings thereof shall be 
 
 in such a condition as not to cause, or to be likely to cause, 
 danger to any person on the Light Locomotive or on any 
 highway. 
 
 (8) There shall be in charge of the Light Locomotive when 
 
 on any highway a person competent to control and direct 
 its use and movement. 
 
 (9) The lamp to be carried attached to the Light Locomotive in 
 
 pursuance of Section 2 of the Act shall be so constructed 
 and placed as to exhibit, during the period between one 
 hour after sunset and one hour before sunrise, a white 
 light visible within a reasonable distance in the direction 
 towards which the Light Locomotive is proceeding or is 
 intended to proceed, and to exhibit a red light so visible in 
 the reverse direction. The lamp shall be placed on the 
 extreme right or off-side of the Light Locomotive in such 
 a position as to be free from all obstruction to the light. 
 
 Provided that this Regulation shall not extend to any 
 bicycle, tricycle, or other machine to which Section 85 of 
 the Local Government Act, 1888, applies. 
 
 ARTICLE III. No person shall cause or permit a Light Loco- 
 motive to be used on any highway for the purpose of drawing any 
 vehicle, or shall drive or have charge of a Light Locomotive when 
 used for such purpose unless the conditions hereinafter set forth 
 shall be satisfied, namely 
 
 (1) Regulations (2), (3), (5), and (7) of Article II. of this Order 
 
 shall apply as if the vehicle drawn by the Light Locomotive 
 was therein referred to instead of the Light Locomotive 
 itself, and Regulation (6) of the Article shall apply as if 
 such vehicle was a Light Locomotive constructed for the 
 carriage of goods. 
 
 (2) The vehicle drawn by the Light Locomotive, except where 
 
244 ELECTRICAL TRACTION. 
 
 the Light Locomotive travels at a rate not exceeding four 
 miles an hour, shall have a brake in good working order of 
 such efficiency that its application to the vehicle shall cause 
 two of the wheels of the vehicle on the same axle to be so 
 held that the wheels shall be effectually prevented from 
 revolving, or shall have the same effect in stopping the 
 vehicle as if such wheels were so held. 
 
 (3) The vehicle drawn by the Light Locomotive shall, when 
 under the last preceding regulation a brake is required to be 
 attached thereto, carry upon the vehicle a person competent 
 to apply efficiently the brake : Provided that it shall not be 
 necessary to comply with this Regulation if the brakes upon 
 the Light Locomotive by which the vehicle is drawn are 
 HO constructed and arranged that neither of such brakes 
 can be used without bringing into action simultaneously 
 the brake attached to the vehicle drawn, or if the brake of 
 the vehicle drawn can be applied from the Light Loco- 
 motive independently of the brakes of the latter. 
 
 ARTICLE IV". Every person driving or in charge of a Light 
 Locomotive when used on any highway shall comply with the 
 Regulations hereinafter set forth, namely 
 
 (1) He shall not drive the Light Locomotive at any speed 
 greater than is reasonable and proper having regard to the 
 traffic on the highway, or so as to endanger the life or limb 
 of any person, or to the common danger of passengers. 
 
 ('2} He shall not under any circumstances drive the Light Loco- 
 motive at a greater speed than twelve miles an hour. If 
 the weight unladen of the Light Locomotive is one ton and 
 a half and does not exceed two tons, he shall not drive the 
 same at a greater speed than eight miles an hour, or if such 
 weight exceeds two tons at a greater speed than five miles 
 an hour. 
 
 Provided that whatever may be the weight of the Light 
 Locomotive, if it is used on any highway to ^ draw any 
 vehicle, he shall not under any circumstances drive it at a 
 greater speed than six miles an hour. 
 
 Provided also that this Regulation shall only have effect 
 during six months from tlie date of this Order, and there- 
 after until We otherwise direct. 
 
 (3) He shall not cause the Light Locomotive to travel back- 
 
 wards for a greater distance or time than may be requisite 
 for purposes of safety. 
 
 (4) He shall not negligently or wilfully cause any hurt or damage 
 
 to any person, carriage, horse, or cattle, or to any goods 
 
APPENDIX. 245 
 
 conveyed in any carriage on any highway, or, when on tho 
 Light Locomotive, be in such a position that he cannot 
 have control over the same, or quit the Light Locomotive 
 without having taken due precautions against its being 
 started in his absence, or allow the Light Locomotive or 
 a vehicle drawn thereby to stand on such highway so as to 
 cause any unnecessary obstruction thereof. 
 
 (5) He shall when meeting any carriage, horse, or cattle keep 
 the Light Locomotive on the left or near side of the road, 
 and when passing any carriage, horse, or cattle proceeding 
 in the same direction keep the Light Locomotive on tho 
 right or off side of the same. 
 
 (G) He shall not negligently or wilfully prevent, hinder, or 
 interrupt the free passage of any person, carriage, horse, or 
 cattle on any highway, and shall keep the Light Locomotive 
 and any vehicle drawn thereby on the left or near side of 
 the road for the purpose of allowing such passage. 
 
 (7) He shall, whenever necessary, by sounding the bell or other 
 
 instrument required by Section 3 of the Act, give audible 
 and sufficient warning of the approach or position of the 
 Light Locomotive. 
 
 (8) He shall on the request of any police constable, or of any 
 
 person having charge of a restive horse, or on any such 
 constable or person putting up his hand as a signal for that 
 purpose, cause the Light Locomotive to stop and to remain 
 stationary so long as may be reasonably necessary. 
 
 ARTICLE V. If the Light Locomotive is one to which Regula- 
 tion (6) of Article II. applies, and the particulars required by that 
 Regulation are not duly painted thereon, or if the Light Loco- 
 motive is one to which that Regulation does not apply, the person 
 driving or in charge thereof shall, on the request of any constable, 
 or on the reasonable request of any other person, truly state his 
 name and place of abode, and the name cf the owner, and the 
 place of his abode or business. 
 
 This order may be cited as " The Light Locomotives on High- 
 ways Order, 1896." 
 
 Given under the Seal of Office of the Local Government Board, 
 this Ninth day of November, in the year One thousand eight 
 hundred and ninety-six. 
 
 HENRY CHAPMAN, President. 
 HUGH OWEN, Secretary. 
 
INDEX. 
 
 Acceleration, 8, 11 
 
 Accumulators. See Storage cells. 
 
 Acquirement of magnetism, 23 
 
 Act, the Light Railways, 1896, 4 
 
 Adhesion, 7 
 
 Admiralty temperature specification, 218 
 
 Advantages of alternate currents, 159 
 
 of electrical traction, 3 
 
 of high tension alternate currents, 159 
 
 of two- and three-phase over single- 
 phase working, 160 
 
 ^tna insulator, 64 
 
 Agents ; cable, electricity, horse, steam, 2 
 
 Air, compressed, for brakes, 91 
 
 Air-blast transformers on Central London 
 Railway, 176 
 
 Ajax lightning arrester, 67 
 
 Allis Corliss engines, 175 
 
 Alternate currents, advantages of, 1 59 
 for feeder and line, 165 
 
 in feeders for traction systems, 
 
 159 
 
 , use of, as influenced by dis- 
 tance, 159 
 
 Appliances for overhead conductor, 63 
 
 Arc lamps on trolley poles at Leeds, 71 
 
 Armature clearance, 13 
 
 conductors, force on, 14 
 , four-pole, winding of, 37 
 
 reaction, 185 
 
 , toothed and smooth-cored, 13 
 
 , Westinghouse, 37 
 
 Arrester, lightning, the Ajax, 67 
 
 , , for alternate currents, 68 
 
 , , Thomson-Houston, 67 
 
 , , AVestinghouse or Wurts, 67, 68 
 
 Automatic accumulator switches at Leeds, 
 
 147 
 
 circuit breaker, 221 
 
 cut-off gear.McIntosh and Seymour's, 
 
 206 
 , Willans and Robinson's, 
 
 207 
 Average and maximum load, relation 
 
 between, 20 1 
 
 Average efficiency, importance of, 185 
 
 of locomotives, 200 
 
 velocity, 49 
 
 Ayrton and Perry. Surface contact 
 systems, 109 
 
 Babcock-Willcox boilers, 187 
 
 Balancing transformers in three-wire 
 
 system, 180 
 require energy to accelerate 
 
 Bar suspension of motor, 91 
 
 Batteries. Six Storage cells. 
 
 Baylor on brakes, 90 
 
 Belgian railways, motors on, 24 
 
 Bentley-Knight roads in America, 94 
 
 Berlin, 99 
 
 Bessbrook and Newry tramway, conductor 
 
 on, 75 
 
 Bessemer mild steel, conductivity of, 74 
 Birmingham, maintenance of storage 
 
 cells at, 129 
 
 , storage cells at, 134 
 Blackpool, 103 
 
 collector at, 104 
 Corporation tramways, Board of 
 
 Trade regulations, 225 
 Blow-out, magnetic, 31, 67, 221 
 Board of Trade recording instruments at 
 
 Leeds, 224 
 
 Regulations, Blackpool tram- 
 ways, 225 
 
 as to brakes, 6 
 
 as to distance between 
 
 supporting poles, 68 
 
 as to feeders, 58 
 
 as to height of overhead 
 
 conductor from ground, 58 
 as to insulation resistance, 
 
 58, 60 
 
 as to life protector, 6 
 
 as to maximum drop of 
 
 potential in rail return, 53, 227 
 as to overhead conductor, 
 
 58 
 
IA T DRX. 
 
 247 
 
 Board of Trade Regulations as to rail as 
 return conductor, 82 
 
 as to speed, 6, and Ap- 
 pendix 
 
 as to tramway cars and 
 
 trailers, 6, 88 
 
 Dublin Southern District 
 
 tramway, 88, 231 
 
 Hartlepool tramways, 236 
 
 unit, cost of, 56 
 
 , definition of, 50 
 
 units, per car mile, 50 
 
 , per mile at Paris, 137 
 
 , on South London Rail- 
 way, 202 
 
 , per ton on City and South 
 
 London Railway, 202 
 , at Montreal, 50 
 
 Boilers at Bristol, 213 
 
 at Dublin, 187 
 
 at Leeds, 195 
 
 Bonding, cost of, 84 
 
 Bonds, butt-welded, 84 
 , cast-welded, 84 
 
 , Chicago, 83 
 
 , Edison-Brown plastic, 84 
 
 , resistance of, 84 
 
 Boosters, 57 
 
 Brakes, electro-magnetic, 91 
 
 worked by compressed air, 91 
 
 Breaking of shunt circuit, 222 
 
 Breaking-stress for copper, 69 
 
 Bristol station, description of, 213 
 
 British Thomson-Houston controller, 31 
 
 lightning arrester, 67 
 
 Brompton and Piccadilly Circus Railway 
 Bill, 177 
 
 Brussels, tramways at, 102 
 
 Budapest, tramways at, 96 
 , collector at, 98 
 
 Burgdorf, three-phrase system at, 172 
 
 Butt-welded rail-joints, 84 
 
 Cable as an agent, 3 
 
 and electric traction combined, 106 
 
 Car bodies, size of, at Dublin, 189 
 Car mile, cost of, 4 
 
 , at Birmingham, 134 
 
 , at Paris, 136 
 
 Car, test of, at Leeds, 88 
 
 weight of ordinary, 49, 189 
 
 Carbon brushes, 39, 184 
 
 Cars and trailers at Chicago, 89 
 
 at Dublin, description of, 189 
 
 , Board of Trade regulations as to, 
 
 88, 231 
 
 , carrying capacity of, at Bristol, 215 
 
 , at Dublin, 189 
 
 , at Leeds, 86 
 
 , electric, energy consumed by, 49, 
 
 88, 137, 144 
 
 , lighting of, at Bristol, 215 
 
 , long and short, discussion on, 89 
 
 Cars, number of, at Bristol, 215 
 
 - y , at Leeds, 197 
 
 , power necessary to start, 50 
 
 , speed of. See Board of Trade 
 
 regulations. 
 
 , storage cell, at Birmingham, 134 
 
 , , at New York, 134 
 
 , -, at Paris, 135-137 
 
 , tractive force of, 92 
 
 , wheel base of, 189 
 
 Cast steel, magnetic properties of, 44. 
 Central London Railway, locomotives 
 for, 176 
 
 , plant for, 175 
 
 , specification of, 173 
 
 , speed on, 173 
 
 , sub-stations on, 176 
 
 , track, 175 
 
 Centrifugal governors, 206, 207 
 Chicago elevated railway, conductor for, 
 78 
 
 , contact shoe, 79 
 
 , cross section roadway, 78 
 
 rail bond, 83 
 
 Circuit breakers, 221 
 City and South London Railway, 7-9 
 , Board of Trade units per mile, 201 
 
 , locomotives on, 17 
 
 Claret-Wiulleumier, surface contact sys- 
 tem, 123 
 
 Clontarf (Dublin) tramway, 192 
 Collector at Blackpool, 104 
 
 for Budapest slotted conduit, 98 
 
 for Chicago elevated railway, 79 
 
 for City and South London Railway, 
 
 76 
 
 for Liverpool overhead railway, 77 
 
 Combined light and traction schemes, 131, 
 146 
 
 storage cell and conduit system at 
 
 Dresden, 145 
 Compound wound motors, 24 
 
 generators, throwing in parallel, 
 
 219 
 
 Compressed air brakes, 91 
 Conductors, system of, at Dublin, 190 
 
 , , at Leeds, 197 
 
 Conductivity, of different compositions, 
 75 
 
 of iron and steel, 74 
 
 Conduit systems Berlin, 99 
 
 Blackpool, 103 
 
 Brussels, 102 
 
 Budapest, 96 
 
 insulation resistance of, 93 
 
 Smith's 106 
 
 Continuing cost, 55 
 
 modified by fall of potential dif- 
 ference, 57 
 
 heating of conductor, 57 
 
 Controller, British Thomson - Houston 
 
 " K2," 31 
 for storage cells, 135 
 
2 4 8 
 
 INDEX 
 
 Controller, series-parallel, 31 
 
 , Westinghouse, 33 
 
 Copper-zinc storage cell, 138 
 Cost of Board of Trade unit, 5G 
 Cost of bonding, 81 
 
 of car mile, 4 
 
 of construction, 73, 81 
 
 of line at Berlin, 100 
 
 at Brussels, 102 
 
 at Budapest, 99 
 
 of poles, 73 
 
 of tramway track, 81 
 
 of trolley wire, 73 
 
 Coupling direct, 215 
 
 Deflection test of poles Leeds, 7 1 
 Density of acid for lead storage cells, 130 
 Depreciation, 55 
 
 Desmazures copper-zinc storage-cell, 137 
 Deviation from Kelvin's law, 57 
 Diatto's surface contact system, 109 
 Differential test of dynamos, 17, 22, 195, 
 
 218 
 
 of power lost in gearing, pro- 
 posed, 202 
 
 Dip of conductor, 68 
 Direct coupling, 215 
 
 current machine, differential test of, 
 
 22, 195, 218 
 
 motor, theory of, 15 
 
 Discharge of storage cells, rate of, 126 
 
 switch, automatic battery, at Leeds, 
 
 147 
 
 Distance affects use of alternate currents, 
 158 
 
 between poles, 68 
 
 Drop of potential in feeder, 54 
 
 in overhead conductor, 61 
 
 in rail, 52 
 
 Dublin cars and trailers, description of, 189 
 
 , Clontarf (Dublin) tramways, 192 
 
 , feeder system at, 52, 190 
 
 Southern District tramways, 187 
 
 , Board of Trade regulations, 231 
 , efficiency, estimated, of con- 
 ductors, etc., 190 
 
 , motors on cars, 189 
 
 Dynamos, acquirement of magnetism in, 
 23 
 
 , armature conductors of, force on, 14 
 
 , reaction in, 185 
 
 , armatures, smooth and toothed cores, 
 
 13 
 
 , efficiency of, at Leeds, 195 
 
 , full and half compound, 220 
 
 , simple shunt, 220 
 
 , test of, 17, 22, 195, 218 
 
 , ventilated armatures for, 15, 41 
 
 Economising energy by means of inclines, 
 
 8, 10, 175 
 
 Economy in working of storage-cells, 129 
 ~ of series-parallel control, 29 
 
 Eddy currents, dissipation of energy by, 22 
 
 , effect of, on acquirement of 
 
 magnetism, 23 
 
 Edison-Brown plastic rail bond, 84 
 Efficiency, affected by series -parallel 
 control, 29 
 
 affected by variation of potential 
 
 difference, 30 
 
 average and instantaneous of locomo- 
 tives, 200 
 
 , commercial, 17 
 
 J differential test of, 17, 22, 95, 202, 218 
 
 , electrical, 17 
 
 of Chloride Electrical Storage Syndi- 
 cate's cells, 143 
 of conversion, 16 
 
 of Dublin system of conductors, es- 
 timated, 190 
 
 of Faure-King traction cells, 141 
 
 of Leeds generators, 195 
 
 Tsystem O f conductors, esti- 
 mated, 197 
 
 ! of spur gearing, 202 
 
 I of worm gearing, 104 
 
 Electric brakes, 91 
 
 locomotives, 1, 7, 17, 19, 22, 25, 144, 
 
 176, 200, 240 
 
 osmosis, 94 
 
 tramways, average velocity on, 49 
 
 ; Electrical traction, growth of, 1 
 
 i welding of joints, 84 
 
 '. Electricity, advantages of, 3 
 
 on light railways, 4 
 
 Electrolysis, 53, 165 
 
 Electromotive force induced in armature, 
 
 15, 40 
 Elevated railway, Chicago, 78 
 
 , Liverpool, 76 
 
 Energy, application of imv", 10 
 
 at starting, 26, 50 
 
 , dissipation of, in conductors, 190 
 
 1 saved by means of inclines, 8, 10, 175 
 
 Engines, fly-wheels for, 209 
 
 , size of, 205, 213 
 
 , steam, governors for, 206 
 
 , , horse power and number of 
 
 units, 213 
 
 , , test of, at Leeds, 215 
 
 Krection of overhead conductors, 63, 
 Esmond's surface contact system, 122 
 Expansion of copper conductor, 70 
 Expenses, ratio of, to receipts, 4, 146 
 
 I , working, 4, 136, 146 
 
 Experiments on acquirement of mag- 
 netism, 23 
 
 on* efficiency, 1 7 
 
 on storage cells, 138 j 
 
 on tractive resistance, 92 
 
 Faure-King cells, 128, 138, 140 
 Feeders, definitions and examples of, 51 
 
 at Dublin, 52, 190 
 
 at Leeds, 52, 197 
 
INDEX. 
 
 249 
 
 Feeders, Board of Trade regulations re- 
 garding underground, 58, 228 
 
 , continuing cost of, 55 
 
 , fall of potential in, 57 
 
 , relation between cost and copper, 56 
 
 Fender for car as life protector, 6 
 
 First cost of feeder, 55 
 
 Fly-wheels according to the practice of 
 Mclntosh and Seymour, 209 
 
 at Leeds, 212 
 
 , relation between, and power of 
 
 engine, 209 
 
 , safe limit of speed for, 211 
 
 Foot-pound and Board of Trade unit, 50 
 
 Foot-pounds, torque in, 16 
 
 Force required to start car, 50 
 
 locomotives, 7, 9, 26 
 
 , tractive, 7, 11, 92 
 
 Frequency of three-phase alternate cur- 
 rents on Central London Railway, 175 
 
 at Dublin, 188 
 
 at Lugano, 166 
 
 Friction, bearing, brush and wind, 16, 22 
 
 Frictional resistance, 8, 9 
 
 Frog for trolley wire branches, 65 
 
 Gauge, 81 
 
 , importance of standard, for light 
 
 railways, 81 
 
 Geared motors, reduction ratio of, 42, 43, | 
 44, 189, 203 
 
 Gearing, spur, efficiency of, 202 
 
 , worm, efficiency of, 104 
 
 General Electric Company's 800 motor, 
 41, 90 
 
 motor, weight of, 190 
 
 Generating station of Central London 
 Railway, 175 
 
 Generator at Lugano, three-phase, 166 
 , compound, 219 
 
 , differential tests of, 218 
 
 , direct coupling of, 215 
 
 , efficiency of, at Leeds, 195 
 
 , multipolar, 216 
 
 , Westinghouse, in motion, 216 
 
 Governor, Mclntosh and Seymour's, 206 
 
 , Willans and Robinson's, 207 
 
 Gradients, 8, 10, 82, 136, 175 
 
 Gravity, work against, 8, 9, 10 
 
 Greathead, papers by, 8, 17 
 
 Greenwood and Batley generators, effi- 
 ciency of, 195 
 
 Growth of electrical traction, 4 
 
 Guard wire, 67 
 
 Hamburg combined traction and lighting 
 station, 131 
 
 , storage cells in power-house at, 131 
 
 sub-stations with storage cells, 146 
 
 Hanover combined storage cell and over- 
 head conductor systems, 145 
 
 , generating costs at, 146 
 
 , maintenance of storage cells at, 146 
 
 Hartlepool tramways, Board of Trade 
 
 regulations for, 236 
 Heating of conductor affects continuing 
 
 cost, 57 
 
 Heilmann locomotive, 1 
 High rate of working storage cells at 
 
 Paris, 136 
 tension currents on Central London 
 
 Railway, 175 
 Hopkinson, Dr. E., on electric railways, 73 
 
 , Dr. J. and Leeds tramways, 195 
 
 , automatic switches at Leeds, 
 
 147 
 
 , motor-car test at Leeds, 88 
 
 , on conductivity of iron and 
 
 steel, 73 
 
 , on electric osmosis, 95 
 
 , series-parallel control, 26 
 
 , surface contact system, 111 
 
 Horse as an agent, 2 
 
 Horse-power consumed by electric cars, 49 
 
 by locomotives, 9 
 
 exerted by motors, 28 
 
 hour and Board of Trade unit, 50 
 
 indicated in Germany, 205 
 
 in United Kingdom, 205 
 
 per car-running, 205 
 
 , relation between, and number 
 
 of units, 213 
 
 required to start car, 50 
 
 Hysteresis, dissipation of energy by, 22 
 
 Importance of uniform gauge, 81 
 
 Inclines used to save energy, 8, 10, 175 
 
 Induction in four-pole armature, 40 
 
 in two-pole armature, 15 
 
 motor and undertruck complete at 
 
 Lugano, 169 
 
 , maximum and average power 
 
 of, 169 
 
 on tramcars, 1 65 
 
 , torque of, 169 
 
 Insulation material for overhead line, 64 
 
 resistance as affected by electric 
 
 osmosis, 95 
 
 , Board of Trade regulations, 60 
 
 of feeders, 58 
 
 of Liverpool overhead railway 
 
 conductors, 78 
 
 of overhead conductors, 58 
 
 of South London Railway con- 
 ductor, 76 
 
 Insulator, section, for overhead lines, 66 
 
 Interest on first cost, 55 
 
 Iron, conductivity of, 73 
 
 , magnetic properties of, 45 
 
 Johnson-Lundell surface contact system, 
 
 117 
 Joints, cast-welded and butt-welded, 84 
 
 in rails, importance of, 86 
 
 , resistance of, 84 
 
 Journal, brush,.and wind friction, 16, 23 
 
250 
 
 INDEX. 
 
 Kelvin's law, 55 
 
 , deviation from, 57 
 
 Kennedy, Prof., evidence of, on Brompton 
 and Piccadilly Circus Railway Bill, 177 
 Kincaird rails, 81 
 Kinetic energy, *mv% 10 
 King type E.P.S. cells, 128 
 
 Lancashire boilers at Leeds, 195 
 Laurent-Cely cells at Paris, 135 
 Leeds, arc lighting in conjunction with 
 
 tramway at, 71 
 , automatic switches for storage cells 
 
 in sub-stations at, 147 
 
 , battery system at, 52 
 
 , boilers at, 195 
 
 , cars at, 86 
 
 , conductor, overhead at, conductivity 
 
 of, 60 
 
 , of feeder system at, 52, 197 
 
 , description of plant at, 195 
 
 , efficiency of system at, estimated, 197 
 
 , flywheel on engines at, 212 
 
 , generators, efficiency of, at, 195 
 
 , maintenance of battery at, 197 
 
 , overhead conductor at, 60 
 
 , poles at, test of, 71 
 
 , service at, 49 
 
 , steam engines at, 195 
 
 , storage cells in sub-stations at, 147 
 
 , switchboard at, 222 
 
 , test of cars at, 88 
 
 , engines at, 215 
 
 , generators at, 195 
 
 Length of line, 49 
 
 of poles at Leeds, 71 
 
 Lightning arrester, Ajax, 67 
 
 for alternate currents, 68 
 
 , Thomson-Houston, 67 
 
 , Westinghouse orWurts, 67, 68 
 
 Light locomotive, distance propelled by 
 
 a given battery, 144 
 
 Railways Act, 4 
 
 , facilities of, 5 
 
 , electricity can be used on, 4, 5 
 
 , gauge of, 81 
 
 , statistics in respect to, 4 
 
 Lighting and traction plant combined 
 
 at Hamburg, 131, 146 
 Lineff s contact system, 109 
 Liverpool overhead railway, conductor 
 
 for, 76 
 
 tramway system, 81 
 
 Load, equalizing, on engines by means of 
 
 accumulators, 182 
 Locomotives, efficiency of, 199 
 , Heilmann, 1 
 
 on Central London Railway, 176 
 
 on Highways Act, 144, 240 
 
 on South London Railway, 8, 9, 17, 19 
 
 , Siemens' curves, 25 
 
 , wheel base of, 19 
 
 Low moor iron, magnetic properties of, 45 
 
 Lugano, double trolley line at, 47 
 
 , induction motor at, 168 
 
 , three-phase generators at, 166 
 
 undertruck, 169, 171 
 
 Lundell, Johnson-, surface contact system, 
 117 
 
 119 
 
 Maddison Avenue, New York, storage 
 
 cells on, 134. 
 Magnetic blow-out circuit breaker, 221 
 
 lightning arrester, 67 
 
 brake, 91 
 
 properties of iron and steel, 44 
 Magnetism, acquirement of, 23 
 Maintenance of storage cells at Birming- 
 ham, 129, 134 
 
 in Central Stations, 129 
 
 Manganese steel, conductivity of, 75 
 Maximum drop of potential in rails, Board 
 
 of Trade regulations as to, 53 
 
 stress in overhead conductor, 69 
 
 velocity, 6 
 
 Mclntosh and Seymour engines at Bristol, 
 
 215 
 
 at Dublin, 192 
 
 Menjarini, Professor, on rotatory con- 
 verters, 160 
 Mesh grouping for three-phase currents, 
 
 164, 169 
 Mile, Board of Trade units per, on South 
 
 London Railway, 201 
 Mileage of storage cells at Birmingham, 
 
 134 
 Montreal, Board of Trade units per car 
 
 mile at, 50 
 Motor-car. See Cars. 
 Motor, agreement between theory and 
 
 practice, 24 
 and gearing, proposed differential 
 
 test of, 202 
 , compound wound, 24 
 
 for tramway work, 35-41 
 
 , test of, 43 
 
 , General Electric Co.'s, 41 
 
 generator in three-wire system, 160, 
 
 176, 180, 188 
 
 rigidly coupled to axle, 17 
 
 , single reduction gear for, 42, 44, 189, 
 
 203 
 , temperature test of, 44 
 
 , testing of, differentially, 22 
 
 , theory of direct current, 15 
 
 , three-phase, 161 
 
 , Westinghouse, 37, 44 i 
 
 Multipolar generator, 185 
 
 New York conduit tramways, 100 
 , method of handling storage cells at, 
 135 
 
 , storage cells on Maddison Avenue 
 
 line, 134 
 
INDEX. 
 
 251 
 
 Niagara, rotatory converters at, 160 
 Non-synchronous alternate current motor 
 
 on cars, 165 
 Number of bars in four-pole armature, 38 
 
 Open conduit. See Conduit systems. 
 Osmosis, electric, 94 
 Over-compounded dynamos, 219 
 Overground conductors, 48, 73 
 
 , Chicago elevated railway, 78 
 
 , composition of, on Bessbrook 
 
 and Newry Railway, 75 
 , , on City and South London 
 
 Eailway, 76 
 
 , conductivity of, 75 
 
 , early use of, 73 
 
 , insulation resistance of, on City 
 
 and South London Railway, 76 
 , , on Livei'pool overhead 
 
 railway, 78 
 , relation between resistance and 
 
 weight of, 76 
 
 , resistance per mile of, 76 
 
 Overhead conductor, 47 
 
 at Dublin, 59 
 
 at Leeds, 60, 62 
 
 , Board of Trade regulations, 58 
 
 , conductivity of, 60 
 
 , cost of, per mile, 73 
 
 , erection of, 63 
 
 , factor of safety for, 69 
 
 , height of, 68 
 
 , maximum drop of potential in, 
 
 61 
 
 , sag of, 68 
 
 , section insulator for, 66 
 
 , temperature effect on, 68 
 
 Panel switchboards, 219 
 
 Parallel running of compound dynamos, 
 
 219 
 Paris, Board of Trade units per car mile 
 
 at, 137 
 
 , Claret-Wiulleumier system, 123 
 
 , storage cell system at, 135 
 
 , weights of cars, batteries, etc., 136 
 
 , working expenses of storage cells 
 
 at, 136 
 Parshall, H. F., on three-phase versus 
 
 three wire system, 180, 181 
 Peckham storage cell undertruck, 132 
 Permanent way. See Track. 
 Perry, Ayrton and, surface contact sys- 
 tem, 109 
 
 Philadelphia, curve of power at, 212 
 Electric Storage Battery Company's 
 
 cells, 134 
 Pinions, steel, cast-iron, phosphor bronze, 
 
 203 
 
 Plastic rail bond, Edison-Brown, 84 
 Plough collector, 104 
 Poles for overhead conductor, 70 
 , test of, at Leeds, 71 
 
 Poles for overhead conductor, weight of, 
 
 at Leeds, 71 
 Portland, failure of three wire system at, 
 
 178 
 , U.S.A., high tension transmission 
 
 by alternate currents, 178 
 Potential difference, drop of, in overhead 
 
 conductor, 61 
 
 , , , at Dublin, 61 
 
 , , , at Leeds, 62 
 
 , , in rail return, 82 
 
 , modifies Kelvin's law, 57 
 
 , maximum, allowed by Board of 
 
 Trade, 48 
 , maximum drop of, Board of Trade 
 
 regulations as to, 53, 227 
 Power consumed by electric cars, 49, 
 
 50 
 
 locomotives, 3, 9, 10, 201 
 
 factor, 41, 48, 164, 190 
 
 , indicated horse-, per car, 205 
 
 , , and number of units, 213 
 
 station at Bristol, 213 
 
 at Dublin (Clontarf District), 
 
 192 
 at Dublin (Southern District^, 
 
 187 
 
 . at Leeds, 195 
 
 at Portland, U.S.A., 178 
 
 transmitted by three-phase currents, 
 
 164 
 Pressure between trolley wheel and wire 
 
 at Leeds, 88 
 
 Quantity efficiency of storage cells, 127 
 
 Rail as return conductor, 82 
 
 bonds, 83, 84 
 
 , cost of per mile, 84 
 
 return, Board of Trade regulations 
 
 as to, 82, 226 
 
 Rails, cost of per mile single track, 81 
 
 , drop of potential in, 54 
 
 , Kincaird, 81 
 
 , maximum drop of potential in, 82, 
 
 192 
 
 , relation between weight and re- 
 sistance of, 76 
 
 , resistance per mile of, 76 
 
 , Vignoles, 81 
 
 , wear of, 81 
 
 , weight per yard of, 81, 82 
 
 Railway, Belgian, motors on, 24 
 
 , gauge of, 81 
 
 , rack, 172 
 
 Railways, Light, Act, 1896, 4 
 
 Reckenzaun on efficiency of worm gear- 
 ing, 104 
 
 on tractive force, 92 
 
 Recording instruments, Board of Trade, 
 224 
 
 Reduction ratio for motors geared to 
 axle, 42, 43, 44, 189, 203 
 
252 
 
 INDEX. 
 
 Regulations. See Board of Trade Regula- 
 tions. 
 
 Requirements of steam engines in traction 
 work, 206 
 
 Resistance, frictional, 8 
 
 insulation, Board of Trade regula- 
 tion, 58, 59, 60 
 
 of joints, 84 
 
 of overground conductor per mile, "76 
 
 of overhead conductor at Dublin, 61 
 
 at Leeds, 62 
 
 of rails per mile, 76 
 
 , relation between, and 
 
 weight, 76 
 
 per mile of rails, 76 
 
 , specific, of iron and steel, 75 
 
 , tractive, 9,92 
 
 Resultant force due to wind and gravity 
 on conductors, 69 
 
 Rheostat for traction purposes, 35 
 
 Rolling stock, Chicago, 89 
 
 , Leeds, 86 
 
 Rome, rotatory converters at, 160 
 , storage cells in power-house at, 131 
 
 Rotatory converters at Dublin, 188 
 
 at Niagara, 160, 161 
 
 at Portland, 160 ; 
 
 at Rome, 160 
 
 on Central London Railway, 176 
 
 transformer, 160, 161, 188 
 
 Rules and regulations. See Board of Trade. 
 
 Russell's poles at Leeds, 71 
 
 Sag in wires, 68 
 
 Section insulator for overhead conductor, 
 
 66 
 Series-parallel operation, 26 
 
 , Dr. J. Hopkinson on, 27 
 
 ' , economy of, 28 
 
 controller, 31, 35 
 
 Service on tramways, frequency of, 49 
 Shoe for collecting energy on the Chicago 
 
 Elevated Railway, 79 
 
 Liverpool overhead railway, 77 
 
 South London Railway, 76 
 
 Shunt circuit, breaking of, 222 
 Siemens, Alexander, curves taken on 
 
 locomotives, 25, 200 
 
 dynamos, efficiency of, 16 
 
 Sir William, on power transmission, 
 
 73 
 Single-phase system compared with two- 
 
 and three-phase, 159 
 Slotted conduit. See Conduit systems. 
 Smooth core armature, 13 
 South London Railway, Board of Trade 
 
 units per mile on, 202 
 conductor, insulation re- 
 sistance of, 76 
 
 , power of motors on, 25, 200 
 
 Span-wire, 63 
 
 Sparking of dynamos, 13 
 
 , prevention of, 31, 221, 222 
 
 Specific resistance of iron and steel, 75 
 
 Speed and torque curves, 21, 42, 43 
 
 on Central London Railway, 173 
 
 on South London Railway, 25, 202 
 
 Spur gearing, efficiency of, 202 
 
 St. Louis, failure of three wire system at, 
 180 
 
 Standard gauge on light railways, 81 
 
 Star grouping for three-phase currents, 163 
 
 Station. See Power station. 
 
 Steam, as an agent, 2 
 
 engine for traction purposes, 206 
 
 , see also Engines. 
 
 . , test of, at Leeds, 215 
 
 on light railways, 4 
 
 Steel, conductivity of, 74 
 
 , magnetic properties of, 44 
 
 pinions, 203 
 
 Storage cells at Birmingham, 129, 134 
 
 at Bristol for lighting cars, 157 
 
 at Leeds, 147 
 
 at Paris, 135 
 
 , automatic regulating switches 
 
 for, 147 
 
 , Chloride P.R. eleven-plate cell, 
 
 test of, 138 
 
 combined with overhead con- 
 ductor at Hanover, 145 
 
 , control of, on Maddison Avenue 
 
 line, 135 
 
 , copper-zinc, 138 
 
 , density of acid for, 130 
 
 , Desmazures, 137 
 
 , efficiency of, 127, 141-143 
 
 , E.P.S., Faure-King type, test 
 
 of, 128, 138, 140 
 
 in sub-stations at Hamburg, 146 
 
 at Leeds, 147 
 
 , maintenance of, at Birming- 
 ham, 129 
 
 , at Hanover, 134 
 
 , at Leeds, 197 
 
 , at Paris, 136 
 
 under truck, 132 
 
 use of, in power-house, 131 
 
 Waddell-Entz, 138 
 
 Stress in line due to temperature varia- 
 tion, 70 
 
 Surface contact systems, 108 
 
 system, Ayrton and Perry, 109 
 
 , Claret- Wiulleumier, 123 
 
 , Diatto, 109 
 
 , Esmond, 122 
 
 , Hopkinson, Dr. J., Ill 
 
 , Johnson-Lundell, 117 
 
 , Lineff, 109 
 
 , McLaughlin, 110 
 
 , Wheless or Westinghouse, 
 
 112 
 
 , Wynne, 122 
 
 Suspension of motor, 90 
 
 Switchboards, 219 
 
 Switchbpards at Leeds, 222 
 
INDEX. 
 
 253 
 
 Temperature, effect of, oil shunt circuit, 24 
 
 tests, 44, 218 
 
 test of Westinghouse motor, 44 
 
 Test of poles for Leeds, 71 
 
 of steam engine at Leeds, 215 
 
 of tramcars at Leeds, 88 
 
 of Westinghouse motor, 44 
 Testing, differential, of dynamos, 17, 22, 
 
 195, 218 
 
 , , of gearing proposed, 202 
 
 of motors, 17, 43 
 
 Theory of direct current motor, 15 
 Three-phase star and mesh grouping, 
 
 163, 164 
 
 transmission at Dublin, 187 
 
 at Portland, U.S.A., 178 
 
 Three wire system, failure of, in tramway 
 
 work, 178 
 Toothed core armature compared with 
 
 smooth, 13 
 Torque, 16 
 
 and speed curves, 21, 42, 43 
 
 Track, 80 
 
 , cost per mile of, 81 
 
 , resistance per mile of, 76 
 
 welding, 84 
 
 Traction, horse, steam, cable, electrical, 2 
 Tractive force, 7 
 
 , on tramways, 91 
 
 Trailer cars at Dublin, 189 
 
 Trailers, Board of Trade regulations as to, 6 
 
 Train-mile, Board of Trade units per, on 
 
 South London Railway, 202 
 Tram-car, test of, at Leeds, 88 
 
 motor, test of, 43, 44 
 
 Tramway motor, the, 35 
 
 , armature for, 37 
 
 Transmission of power by alternate 
 
 currents, advantages of, 158 
 Trolley line. See Overhead conductor. 
 
 systems, single and double, 47, 48 
 
 wheel, 47 
 
 Track at Dublin, 189 
 
 at Leeds, 86 
 
 at Lugano, 169 
 
 , storage cell 132 
 
 Trucks, construction of, 89 
 
 Turbines, Victor, 178 
 
 Two - phase rotatory converters at 
 
 Niagara and Portland, 160 
 
 Uniform gauge, importance of, 81 
 United States, growth of electrical traction 
 in, 4 
 
 Variation of applied potential difference, 
 
 effect of, 30 
 Velocity. See Speed. 
 
 Velocity, energy in virtue of, 10 
 Ventilation of armatures, 15, 41 
 Vicars mechanical stokers, 187, 213 
 Voltage. See Potential difference. 
 
 "\Vaddell-Entz copper-zinc storage cell, 
 
 138 
 Washington, insulation resistance at, 95 
 
 , surface contact system at, 117 
 
 Water-power transmission at Lugano, 47, 
 
 48 
 
 at Portland, 178 
 
 Wear of tramway rails, 81 
 Weight of poles at Leeds, 71 
 
 locomotives on Central London 
 
 Railway, 176 
 
 on City and South London Rail- 
 way, 7, 9 
 
 per yard and resistance of rail, 
 
 relation between, 76 
 
 per yard of tramway rails, 81, 82 
 
 Welded joints in rails, 84 
 Westinghouse controller, 33 
 
 , generator in motion, 216 
 
 , lightning arrester, 67, 68 
 
 , motor, 37, 44 
 
 or Wheless, surface contact system, 
 
 112 
 
 work-shop tests of motor, 44 
 
 AVheel base of locomotives on South 
 
 London Railway, 19 
 
 of tramcar at Dublin, 189 
 
 Wheels, diameter of, on South London 
 
 Railway locomotives, 19 
 
 , , on tramway cars, 43, 44, 190 
 
 Wheless electro-magnetic apparatus, 114 
 
 surface contact system, 112 
 
 Whitworth mild steel, conductivity of, 
 
 74 
 Willans and Robinson's governor, 207 
 
 engines at Dublin, 187 
 
 Wind pressure on overhead conductor, 
 
 69 
 
 Work, application of mv'\ 10 
 Working expenses, ratio of, to receipts, 4, 
 
 146 
 
 with storage cells at Paris, 
 
 136 
 
 Worm gearing, efficiency of, 104 
 Wrought-iron, magnetic properties of, 
 44,45 
 
 , specific resistance of, 75 
 
 Wurts, lightning arrester, 67, 68 
 Wynne's surface contact system, 122 
 
 Zurich, storage cells in power-house at. 
 131 
 
FEINTED BY WILLIAM CLOWES AND SONS, LIMITED, 
 LONDON AND BECCLES. 
 
SCIENCE. 
 
 A First Year's Course in Experimental Chemistry. 
 
 By Dr. E. H. COOK. 
 
 Lectures on Sound, Light, and Heat. 
 
 By RICHARD WORMELL, D.Sc., M.A., Head Master of the Central 
 Foundation Schools of London. New Edition. Each volume, 
 crown 8vo, Is. 
 
 The Standard Course of Elementary Chemistry. 
 
 By E. J. Cox, F.C.S., Head Master of the Technical School, 
 Birmingham. In Five Parts, issued separately, bound in cloth 
 and illustrated. Parts I.-IV., Id. each ; Part V., Is. The com- 
 plete work in one vol., crown 8vo, 2. Qd. 
 
 The several parts correspond to the carefully arranged divisions of 
 the subject prescribed by the Education Department. The work has 
 been purposely issued at an extremely low price, and the apparatus 
 required for conducting the experiments is simple and inexpensive. 
 Adopted by the School Boards for London, Edinburgh, etc. 
 
 Journal of Education." A capital book both for the teacher and practical 
 elementary student." 
 
 Elementary Natural Philosophy. 
 
 By ALFRED EARL, M.A., Assistant-Master at Tonbridge School. 
 
 [In preparation. 
 
 An Elementary Chemistry. 
 
 By W. A. SHENSTONE, M.A., Lecturer in Chemistry at Clifton 
 College. [ In preparation. 
 
 Physical Chemistry. 
 
 By Dr. ALEXANDER SCOTT, Demonstrator to the Jacksonian Pro* 
 fessor of Natural and Experimental Philosophy at the University 
 of Cambridge. [ In preparation. 
 
 A Manual of Physiology. 
 
 By LEONARD HILL, M.D. Specially adapted to the Syllabus of the 
 Science and Art Department. [In preparation 
 
 A Manual of Botany. 
 
 By DAVID HOUSTON, of the County Technical Laboratory, Chelms- 
 ford. Adapted to the requirements of the Science and Art 
 Department, College of Preceptors) and Oxford and Cambridge 
 Local Examinations. [In preparation* 
 
 EDWARD AKNOLD: LONDON AND NEW YORK. 
 
ARNOLD'S PRACTICAL SCIENCE MANUALS. 
 
 General Editor : Professor KAPHAEL MELDOLA, F.K.S., of the Finsbury 
 
 Technical College of the City and Guilds of London Institute. 
 The object of the Manuals in this Series is to present the subject, 
 as far as possible, in a practical form, and to fill the gap which at 
 present is believed to exist between the elementary treatises dealing 
 with the principles of the Sciences in a purely abstract manner, and 
 the special technological works which are too advanced to be profitably 
 used by readers who are not yet familiar with the subject. 
 
 Steam Boilers. 
 
 By GEORGE HALLIDAY, late Demonstrator at the Finsbury Technical 
 College. With numerous Diagrams and Illustrations. Crown 
 8vo. 400 pages. 5s. 
 
 The Engineer." We cannot put down our pen until we have said that we do not 
 know of any book on boilers more likely to be of use to the student than this." 
 
 Electrical Engineer." The best elementary book on boilers we have seen. 
 The more we examine the book, the better we are pleased with it." 
 
 Electrician. "A very excellent little manual." 
 
 Steamship." Mr. Halliday is to be congratulated on having produced a work on 
 steam boilers which is a real contribution to technical literature. If this first volume 
 of Arnold's Practical Science Manuals is a fair sample of those that are to follow, we 
 feel certain that they will meet with great success." 
 
 Electrical Traction. 
 
 By ERNEST WILSON, Wh. Sc., M.I.E.E., Lecturer and Demonstrator 
 in the Siemen's Laboratory, King's College, London. Crown 8vo, 
 cloth, 5s. 
 
 Agricultural Chemistry. 
 
 By T. S. DYMOND, of the County Technical Laboratory, Chelmsford. 
 
 [./w preparation. 
 
 NOW READY. 
 
 THE CALCULUS FOR ENGINEERS. 
 
 By Professor JOHN PERRY, F.R.S. About 400 pages. Crown 8vo, 
 cloth, 7s. Qd. 
 
 Practical Engineer. '\We heartily commend the book to our readers who 
 would like to make the Calculus a real use to themselves and not merely an orna* 
 mental exercise." 
 
 Electrician. ''The author has been successful in retaining all that liveliness and 
 originality of illustration which distinguishes him as a lecturer. The volume abounds 
 with delightfully humorous remarks, but the wit is never at the expense of sound 
 advice and instruction, of which the student who would teach himself stands in need.*' 
 
 EDWARD ARNOLD: LONDON AND NEW YORK, 
 
MATHEMATICAL & SCIENTIFIC WORKS 
 
 PUBLISHED BY 
 
 MR. EDWARD ARNOLD. 
 
 MATHEMATICS. 
 The Elements of Algebra. 
 
 By R. LACHLAN, Sc.D., formerly Fellow of Trinity College, Cambridge. 
 Crown 8vo., cloth, with or without Answers, 2s. 6d. Answers 
 separately, Is. 
 
 A New Arithmetic for Schools. 
 
 By J. KIBKMAN, Assistant-Master at Bedford Grammar School. 
 
 [In preparation. 
 
 The Mercantile Arithmetic. 
 
 A Text-Book of Principles, Practice, and Time-Saving Processes. By 
 RICHARD WORMELL, D.Sc., M.A., late Head Master of the Central 
 Foundation Schools of London. 
 
 Now issued in Parts as well as complete, in the following editions : 
 The Mercantile Arithmetic, Part I. Cloth. 2s. 
 The Mercantile Arithmetic, Part II. Cloth. 2s. 
 The Mercantile Arithmetic, Parts I. and II. together. 3s. 6d. 
 The Mercantile Arithmetic, complete with Answers. 4s. 
 Answers only, Is. 
 
 The Elements of Euclid, Books I. VI. 
 
 By R. LACHLAN, Sc.D., formerly Fellow of Trinity College, Cambridge. 
 With alternative Proofs, Notes, Exercises, all the Standard Theorems, 
 and a large collection of Riders and Problems. Crown 8vo., cloth, 
 4s. 6d. 
 
 Also issued in the following Divisions : 
 
 Euclid, Book I. Cloth, Is. 
 
 Euclid, Books I. and II. Cloth, Is. 6d. 
 
 Euclid, Books I., II. and III. Cloth, 2s. 6d. 
 
 Euclid, Books III. and IV. 2s. 
 
 Euclid, Books I. IV. 3s. 
 
 An Elementary Treatise on Practical Mathematics. 
 
 By JOHN GRAHAM, B.A., Demonstrator of Mechanical Engineering 
 and Applied Mathematics in the Technical College, Finsbury. 
 Crown 8vo., cloth, 3s. 6d. 
 
 An Elementary Text-Book of Mechanics. 
 
 By R. WORMBLL, D.Sc., M.A. With 90 Illustrations. Crown 8vo., 
 
 cloth, 3s. 6d. 
 
 Specially adapted for the London Matriculation, Science and Art 
 Department, College of Preceptors, and other examinations. 
 %* Solutions to Problems for Teachers and Private Students. 3s. 6d. 
 
 LONDON : EDWARD ARNOLD, 37 BEDFORD ST., STRAND, W.C. 
 
MATHEMATICS. 
 The Calculus for Engineers. 
 
 By JOHN PERRY, M.E., D.Sc., F.R.S., Professor of Mechanics and 
 Mathematics in the Royal College of Science, Vice- President of the 
 Physical Society, Vice -President of the Institution of Electrical 
 Engineers, etc. Crown 8vo., cloth, 7s. 6d. 
 
 Physical Calculus. 
 
 By PERCY E. BATEMAN, M.A., Fellow of Jesus College and Assistant- 
 Demonstrator of Physics in the University of Cambridge. 
 
 [In preparation. 
 
 Dynamics for Engineering Students. 
 
 By W. E. DALBY, Professor of Applied Mathematics and Mechanical 
 Engineering at the Technical College, Finsbury. [In preparation. 
 
 ARNOLD'S PRACTICAL SCIENCE MANUALS. 
 
 General Editor : Professor RAPHAEL MELDOLA, F.R.S., of the Finsbury 
 Technical College of the City and Guilds of London Institute. 
 
 The object of the Manuals in this Series is to present the subject, as far 
 as possible, in a practical form, and to fill the gap which at present is 
 believed to exist between the elementary treatises dealing with the 
 principles of the Sciences in a purely abstract manner, and the special 
 technological works which are too advanced to be profitably used by 
 readers who are not yet familiar with the subject. 
 
 Steam Boilers. 
 
 By GEORGE HALLIDAT, late Demonstrator at the Finsbury Technical 
 College. With numerous Diagrams and Illustrations. Crown 8vo. 
 400 pages. 5s. 
 
 Electrical Traction. 
 
 By ERNEST WILSON, Wh. So., M.I.E.E., Professor of Electrical 
 Engineering at King's College, London. With numerous diagrams 
 and illustrations. Crown 8vo., 5s. 
 
 Chemistry for Agricultural Students. 
 
 By T. S. DTMOND, F.I.C,, Lecturer in the County Technical Labora- 
 tories, Chelmsford. With a Preface by Professor MELDOLA, F.R.S. 
 Crown 8vo., cloth, 2s. 6d. 
 
 %* Other volumes in preparation. 
 
 LONDON : EDWARD ARNOLD, 37 BEDFORD ST., STRAND, W.C. 
 
SCIENCE. 
 Magnetism and Electricity. 
 
 By J. PALEY YORKE, of the Northern Polytechnic Institute, Holloway. 
 Crown 8vo., cloth, 3s. 6d. 
 
 A Text-Book of Physical Chemistry. 
 
 By Dr. R. A. LEHFELDT, Professor of Physics at the East London 
 Technical College. Crown 8vo., cloth, 5s. 
 
 A First Year's Course of Experimental Work in 
 Chemistry. 
 
 By E. H. COOK, D.Sc., F.I.C., Principal of the Clifton Laboratory, 
 Bristol. Crown 8vo., cloth, Is. 6d. 
 
 The Standard Course of Elementary Chemistry. 
 
 By E. J. Cox, F.C.S., Head Master of the Technical School, Birming- 
 ham. In Five Parts, issued separately, bound in cloth and illustrated. 
 Parts I. -IV., 7d. each ; Part V., Is. The complete work in one vol., 
 crown 8vo., 3s. 
 
 The several parts correspond to the carefully arranged divisions of the 
 subject prescribed by the Education Department. The apparatus required 
 for conducting the experiments is simple and inexpensive. 
 
 Adopted by the School Boards for London, Edinburgh, etc. 
 Part I. Properties of the Common Part IV. Carbon and Non - metallic Ele- 
 ments. 
 
 V. Metallic Bodies, Combination, 
 Symbols, and Formulse. 
 
 II. The Atmosphere. 
 III. Water. 
 
 Physical Chemistry for Beginners. 
 
 By Dr. CH. M. VAN DEVENTER. With a Preface by J. H. VAN 'T HOFF, 
 Translated by Dr. R. A. LEHFELDT, Professor of Physics at the East 
 London Technical College. 2s. 6d. 
 
 Contents. Chap. I., Definitions. Chap. II., Fundamental Laws of 
 Combination. Chap. III., Behaviour of Gases. Chap. IV., S<>me points 
 of Thermo- Chemistry. Chap. V., Solutions. Chap. VI., Photo-Chemistry. 
 Chap. VII., The Periodic System. 
 
 A Manual of Elementary Chemistry. 
 
 By W. A. SHENSTONE, M.A., Lecturer in Chemistry at Clifton College. 
 
 [In preparation. 
 
 Lectures on Theoretical and Physical Chemistry. 
 
 Part I. Chemical Dynamics. 
 
 By Dr. J. H. VAN 'T HOFF, Professor at the University of Berlin. 
 Translated by Dr. R. A. LEHFELDT, Professor of Physics at the 
 East London Technical College. One vol., 8vo., 12s. net. 
 
 Lectures on Sound, Light, and Heat. 
 
 By RICHARD WORMELL, D.Sc., M.A., late Head Master of the Central 
 Foundation Schools of London. New Edition. Each volume, crown 
 8vo., Is. 
 
 LONDON : EDWARD ARNOLD, 37 BEDFORD ST., STRAND, W.C. 
 
 ( 3 ) 
 
I D I UO77