ia39.] 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



135 



ilistingiiisli the piecise value of cacli particular unit of resistance, than to de- 

 teriuine the relative sura of resistance and the relative expendititrc of power 

 at all velocities and under all circumstances. Now the term duty may be ap- 

 jdied in tlie strictest sense of the term to the work done by a locomotive 

 engine ; for whether the engine drag a load whose resistance is 8 Ihs. per 

 ton, or whether a weight of 8 lbs. for each ton of matter moved descending 

 over a pidley and attached to the load, be considered as the moving force, 

 the result is the same. If, then, the tractive force, or resistance per ton of 

 matter in motion, which is the real load on the engine, be ascertained, the 

 whole effect is found by multiplying tliis sum by the space jiassed over in 

 feet ; and the consumption of water as steam and of eo!;o being known, we 

 have all tlio elements requisite for determining the duty [lerfiirmed Iiy the 

 steam or coke. The ju-cssure against the pistons may he decbiced from the 

 sum of tlie resistances first calculated on the assumed resistance overcome .it 

 the velocity of the engine in each experiment ; and the pressure on the pis- 

 tons may also be deduced from tlic ratio of the volumes of the steam and 

 water consumed. The results which may be olitained on these principles are 

 tabulated, for the experiments of M. de Panibour, Robert Stephenson, and 

 Dr. Lariluer. In .mother table the author has recorded the reduction of 

 each of these exiieriments to terms of horses' power, and has exhibited under 

 that denomination the absolute power resulting from the steam used — that 

 reipiired to overcome the assigned resistance — their differences — and the 

 power whicli balances the gross and useful duty. The construction of these 

 most elaliorate tables is described in great detail, and the consetpiences which 

 follow from tlie tests thus obtained are fully stated, and the author comes to 

 the conclusion, th.at results inconsistent with the capabilities of tlie locomo- 

 tive are perceptible in almost every one of the expei'iments. A condensing 

 engine ]daced on wheels, with water of condensation transported for its sup- 

 ply, and m.ide to drag a train along a railway, woul.l require the same ex- 

 penditure of water as steam, to produce a given effect, as if fixed ; a non-con- 

 densing engine also is one and the same machine, whether fixed or locomotive, 

 excepting that the latter must consume more power than the former, to do 

 etpial work, at like pressures, by the amount of the additional resistance aris- 

 ing from the contraction of its eduction pipes, in order to produce a fierce 

 blast of steam through the chimney. Troni these and other causes the fixed 

 non-condensing engine must be the more economical of the two ; hut if the 

 results derived from M. de Pamlionr's data be correct, we must acknowledge 

 the fixed non-condensing engine, with its simple atmospheric resistance, to lie 

 fjir inferior in economy of steam to the locomotive, with its ]dus .atmospheric 

 resistance. The experiments by Dr. Lardner were made for the purpose of 

 determining the resistance opposed to progressive motion on railways. They 

 cojisisted in dismissing trains at various speeds from the summit of inclined 

 planes, and in observing their velocity when it became uniform, the resistance 

 at such velocity lieing equal to the accelerating force of gravity down the in- 

 clined plane. The results of these .are tabid.ited in tlie same manner as the 

 preceding, and the most singular discrepancies present themselves. For in- 

 stance, it would appear that in one particular case a duty of double the amount 

 of that effected by the condensing engine was performed by an equal expen- 

 diture of jiower; that compared with a fixed non-condensing engine at equal 

 pressure, the locomotive, tliongh labouring against the heavy counter-pressure 

 of the blast from which the other is free, is assumed to have jierformed equal 

 work with less than one-half the cxjicnditure of power. That if the resistance 

 assigned by Dr. Lardner as opposed to the progressive motion of the train be 

 correct, tlic efticiency of the steam in the locomotive is more than double 

 that obtained by the best condensing engines; more than treble that derived 

 from stationary non-condensing engines, and c(pial to the performance of a 

 Cornish expansive engine, doing a ,^iO million duty with a bushel of coals. 

 \\"Mi such results before us, the resistances assigned as opposed to and over- 

 come by the locomotive at different velocities, must lie reganled as utterly 

 iMconsistcnt with reality, and as resting on no solid foundation. 



The preceding results show also that errors have crept in by the adoption 

 of the theoretical method of reducing undnlatoi'y surfaces to a level. M. de 

 I'anibonr extends the length of the road as a compensation for the acilivities 

 or for the help afforded by the hank engines, and Dr. Lardner diminishes the 

 time of the trip to that which he assumes would be occupied in ]ierformingit 

 on a dead level. If the jirinciples on which these corrections for the acclivi- 

 ties and declivities are made be correct, other facts than we are at present 

 acquainted with must be taken into account before it can he demonstrated 

 that a given power will convey a given load at some certain increased velocity 

 along a level compared with the actual velocity along any given undulating 

 line. The resistances w^iich enter into the comjiosition of the sum of the 

 forces are ever varying to such an extent, tliat it may be doubted whetlicr the 

 theoretical level be not a pure fiction with reference to the jiractical results 

 of the experiment. 



The cficctive jiower of a locomotive engine, or the excess of po\rer after 

 overcoming its proper friction and tlie resistance from the blast, is solely ex- 

 pended in the generation of momentum. This which is the product of the 

 mass and the velocity represents the useful jnecbanical effort exerted by the 

 steam, and may always be ascertained under all the practical circumstances 

 of railway traffic. Tlie consumiition of power as ^v.atcr, in the shape of steam, 

 is a third quantity which may also be readily ascertained. The application 

 which may be made of the ,ibove data is comprehended in the following pro- 

 positions. First, that equal momenta would result at all velocities from an 

 equal amount of jiower expended in equal times by tlie same engine, if the 

 forces opposed to progressive inolion and to the effective use of steam in the 



engines were uniform at all velocities. Secondly, the difference between the 

 momenta generated bv a unit of power in a given time at various velocities, 

 measures the difference in the sum of the resistances opposed to the power at 

 those velocities. Having ascertained the gross weight of an engine tender 

 and train— their mean velocity— and the expenditure of water as steam dur- 

 ing the trip, simple computations will inform ns of 



1. The mechanical effect realized by a given power at all velocities. 



2. The totiil increase or decrease of resistance at all velocities. 



3. The ratios which the increase or decrease of resistance at difterent velo- 

 cities bear to the ratios of those velocities. 



Two other results also follow from the above, and which may he terined 

 tlrc commercial results, viz. the amount of gross and useful tractive etVect 

 realized bv an e(|ual expenditure of power at all velocities. The difference 

 between these is a useless quantity in a practical sense, being the costly waste 

 of power incident to the locomotive functions of the engine and tender over 

 and above the waste arising from the iinaseertained and ineffective portion of 

 the whole power required for the blast. The reductions and computations 

 necessary for the exhibition and development of these views are contained ill 

 two tables. Tliev relate to forty-nine experimeuts, being those already re- 

 f.'ned to. and those by Mr. \. Wood, on the Great Western, and T^ondon and 

 Birmingham liailvvay, and sonic others. One of these tables contains the 

 velocity of the engines, the con,suniption of water as steam, the loads, the 

 absolute momenta per second ; the momenta generated by equal power in 

 equal times, viz. hv 1 lb. of water as steam per second ; the weights of the 

 gross and useful loads moved by equal powers, viz. by one cubic foot ofw.ater 

 as steam, at the velocity of each experiment, with various other elements. 

 The other table contains a summary of the ratios of the velocities and of 

 their s((uarcs, brougbt into juxta-position with the r.atios of the power ex- 

 pended to produce equal momenta, equal gross and equal useful effects, by the 

 comparison of pairs of experiments on the engines given in the preceding 

 table. Tliis t.-ihle also shows the influence of velocity in the expenditure of 

 power to jiroduce equal mechanical and equal commercial effects ; and the 

 amount of loss attributable to the increase of resistance at the higher veloci- 

 ties. The author discusses in great detail the various circumstances of these 

 experiments, and flic inferences and practical conclusions which may be de- 

 duced therefrom : and comes to the conclusion, that the determinatioit of the 

 performance of locomotive engines by the methods here set forth is .as prac- 

 ticalile, exact, and demonstrative of their relative powers and dynamic ex- 

 celleuee, as the determination of duty done bv pumping engines. 



The intensitv of the pressm-e on the opposite side of the piston arising 

 from the blast' has been but imperfectly stated. By some the discharge of 

 the steam has hccu likened to a jet, and considered continuous. But an at- 

 tentive observer can appreciate by his ear that an interval exists between the 

 alternate discharges of steam from the two cylinders. That these juts are 

 ]ieriodic and uot continuous, is also distinctly evidenced by the audible pulsa- 

 tions in the chimney, even .at the very highest velocities of an engine, and 

 their duration may be measured at lower speeds. Upon this intermittent 

 action of the blast "de|iend, in a great measure, the resultant pressure against 

 the jiiston, and the production of a sufiicient current of air through the fire, 

 both which eft'ects would he malerially changed in intensity liy the substitu- 

 tion of a continuous for a periodic current. The precise duration of the jet 

 or of the time of the steam cvticuiiting the cylinder, can only be determined 

 by dii-crl and careful experiments ; hut its period may be ascertained within 

 definite limits; for since a single discharge is completed within the time oc- 

 cupied by the piston in accomplishing a luilf stroke, and the pauses between 

 two successive discharges .are distinctly perceptible, a single blast cannot oc- 

 ciqiy the fourth p.art of the time of the revolution of the crank shaft, .and very 

 lu-obably does not exceed the eighth part, or the period of aqu.arter stroke of 

 the piston, ruder no circumstances, then, can f lie pressure from the blast 

 oppose the jiiston much longer than during one fourth of the stroke. With 

 an active pressure, then, of ;iO lbs. ]ier square inch, the mean resistaiicc from 

 the lilast would not be greater than 7^ lbs., and with a iircssure of 15 lbs. 

 not grci.te- than ;i:f lbs. per square inch, against the pistons. The author 

 then proceeds to cite several observations and experiments nia.le by himself, 

 which arc confirmatory of the preceding argument respecting the lilast, and 

 he was led couclusivelv to the fact, that one fifth of the power of the engine 

 experimented upon, at" working pressures of 20 lbs. and !."> lbs., was .absorbed 

 in blowing the fire ; and that the escape of the steam from the cylinder was 

 four times swifter than the motion of the piston. 



The author lastly treats of the expenditure of power for a given effect by 

 fixed and locomotive non-condensing engines. But few experiments on the 

 expendifiire of steam for a given cfiect by non-condensing tt.atimiary engines 

 have been made. The lel.ative consumption of fixed condensing and noii- 

 condensing engines has been treated of by the late Mr. Charles Sylvester, of 

 Derby, whose knowledge and accurate theoretical an,alysis of the subject are 

 showii by the close accordance of his conclusions with the facts established 

 on two engines of these classes at certain working pressures. His conclusion 

 that the relative economy of these engines will be as the quantities of steam 

 consumed, or .as 1 to 1-7'J, at those pressures, is accurately confimed by the 

 results here recorded. Mr. Sylvester .ilso showed, that by increasing the 

 pressure upon the same nou-eondensing, and by eiil,argi|!ig the area of the 

 condensing engine's cylinder and .air pump, so as to mainfliin the steam in it 

 at a uniform iiressure'iier square inch for all loads, the economy of the former 

 would gradually approach and fimdly eqn.al that of the latter. The results 

 obtained in the preceding part of the paper, furnish nnincrous comparisons 



T 2 



