!)2 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



[March, 



ON THE GENERAL THEORY OF THE STEAM ENGINE. 



Notwithstaiuling the number of years Avhich have elapsed since 

 the invention of the Steam Engine, and the immense extent to which 

 it is now employed, as well as the great importance at present 

 attached (particularly in some of its applications) to the perfection 

 and economy of its working : yet our knowledge of the general theory 

 of its action is still but very limited. We were led to investigate 

 this subject more fully on the jiernsal of a pamphlet wliicli appeared 

 in the course of last 'year, entitled, " A New Theory of the Steam 

 Engine," by the Chevaler G. de Pambour, and purporting to be an 

 analysis of a memoir by the same author, which was read at the 

 Institute Royale of France, during the year 1837. The author, after 

 shewing the inaccuracy of the ortlinary mode of calculation used to 

 determine the effects and proportions of steam-engines, exposes what 

 he calls his new theory, by means of which he undertakes to solve all 

 problems relating to those effects and proportions. M. de Pambour's 

 theorj- consists essentially in the following laws : — 



1. That the pressure in the cylinder is strictly regulated by the 

 resistance on the piston and by nothuig else. 



2. That the velocity of an engine is determined by the quantity of 

 water which can be evaporated m the boiler in a given time, and 



3. That the pressm-c in the boiler is indifferent, provided it be at 

 least equal to the pressure in the cylinder. 



The last of these laws is of ver%' little consequence, since it can have 

 no effect on any of the calculations ; but the two former serve to 

 solve all the problems relating to steam-engines ; for, having as- 

 certained the quantity of steam generated in the boiler, and trans- 

 mitted to the cylinder, as well as the velocity of the piston, which 

 gives the volume occupied by that steam, we find its density and 

 elastic force, and consequently the resistance on the piston. Inversely, 

 either of the two other quantities might be determined, the rest being 

 given. The principal difference between M. de Pambour's and the 

 ordinary mode of calculation is that, according to the former, the 

 effect of an engine is measured by the quantity of steam generated in 

 the boiler ; and according to the latter, by the quantity used in the 

 cylinder : both which quantities must be equal, if correctly measured ; 

 unless, of course, thei-e be any discharge through the safety valve, in 

 which case M. de Pambour's methoci would fail, if he had not the 

 means of measuring or estimating the quantity so lost; in place of 

 which he assiunes (for locomotive engines) an average loss of one 

 fourth of the whole of the steam generated, and therefore considers 

 the effective evaporating power of a boiler to be three fourtlis of its 

 total evaporating power. Now this clearly cannot be true for all 

 locomotive engines, nor even for one engine on all occasions ; the 

 safety valve will be more or less open, according to the load of the 

 engine and the pressure in the boiler; and the discharge of steam 

 through the valve, when open to the same degree, will depend on the 



Eressure in the boiler. The latter circumstance would, however, 

 ave but an imperceptible influence if the difference of pressure were 

 not very considerable, the velocity of efflux being directly as the 

 effective pressure, and inversely as the density of the steam. 



The mamier in which M. de Pambour arrived at the average loss 

 of steam thi'<jugh the valve, shows that it is seldom, if ever, correct. 

 He first ascertained what rise of the valve was necessary for the dis- 

 charge of all the steam generated, which corresponds to dilferent 

 numbers of degrees of the spring balance for different boilers, and 

 then observed the actual rise in a certain number of experiments 

 made with different engines ; he then compared the sum of all those 

 rises with the sum of all the rises necessary to give egress to all the 

 steam generated in the various boilers, taken once for each experi- 

 ment. In eleven experiments, the sum of actual rises was 12 degrees 

 of the various spring balances, and the sum of rises necessary to 

 give egress to all the steam, in each case 46'5 degrees : namely, five 

 experluicuis, \\ ilh engines requiring 5 degrees, give 2.5 ; tliree expe- 

 riments, with one requiring 4, give 12: two experiments, with one 

 requiring 3, give 6 ; and one experiment, with an engine requiring 

 ,3'5, gives 3-5 ; which numbers, when added together, will be found 

 to make up 46-5. The i-atio of 12 to 46-5 being very near one-fourth, 

 this has been taken as the average rise of the valve, and conse- 

 quently as the average loss of steam. 



Thus we see that we can make no use of the constant coeflicienl 

 075 in determining the effective evaporating pow-er of a boiler; for 

 it might very well happen that only one-eighth should escape 

 through the valve, in which case we" should find the effect one- 

 seventh too little : and if in any case there should be a loss of one- 

 half of the steam, then v,e should arrive at a result 50 per cent, too 

 great. 



But, setting; aside the error we should ronimit by taking the 

 average rise of the valve, how arc the following facts to he accounted 

 for? In the table of experiments at j'age 220 of M. dc Pambour's 



'• Treatise on Locomotive Engines," we see that the Finy ascended 

 the Sutton inclined plane, with a load equivalent to 183 tons on a 

 level, at a speed of 13%33 miles an hour, the rise of the valve being 

 5 degrees of the spring balance, which is sufficient (see the table at 

 page 175) to allow of the escape of all the steam generated in tlie 

 boiler. We also find (page 2.32) that the Vesta ascended the same 

 inclined plane, with a load equivalent to the former, at a velocity of 

 3-25 miles an hour, when the rise of the vulve was equal to 3-5 

 degrees of the spring balance, or sufficient, according to the above- 

 mentioned table, to give issue to all the steam. Also, (see page 

 234), the Vulcan is stated to have ascended the same inclined plane, 

 with a load equivalent to 188 tons on a level, at a velocity of 1I'42 

 miles an hour, the safety-valve being sufficiently open" to allow 

 of the escape of all the steam ; and, in the same page, the same 

 engine is stated to have ascended the Winston inclined jilane on 

 another occasion, witli a load equivalent to 186 tons on a level, at a 

 speed of 18-75 miles an hour, the safety-valve being open to the 

 same degree as in the preceding case. 



Supposing the observations to have been correctly made, we can 

 only account for these apparent anomalies by supposing the evapo- 

 ration to have been more rajjid in the cases quoted than during the 

 experiments which had been made, with the view of determining the 

 rise of the valve necessary for the discharge of all the steam gene- 

 rated; for it is the absolute quantity discharged, and not the pro- 

 portion, which is determined by the size of the aperture. It is, 

 therefore, exceedingly difficult to deduce the power developed by an 

 engine from the evaporating power of the boiler, whenever there is 

 any escape llu'ough the safety-valve. 



Before leaving the subject of these experiments, we must observe 

 that, from their nature, they were not susceptible of that precision 

 which is necessary to allow of their results being made the basis of 

 accurate calculation : for, on account of the irregularities of the 

 road, the circumstances were continually varying, and the momentum 

 of the trains rendered the effect of those variations less perceptible 

 than it ought to have been. 



It is an essential part of the Chevalier de Pambour's theory, that 

 the pressure in the cylinder is independent of the pressure in the 

 boiler, and depends solely on the resistance to be overcome. This is, 

 to a certain extent, true ; but we cannot allow that the pressiu'e in 

 the boiler is altogether independent of the pressure in the cylinder, 

 for it cannot be denied that the law of the flowing of elastic fluids 

 must obtain in this, as well as in all other cases. The pressure of 

 the steam in the cylinder being, therefore, equal to the resistance on 

 the opposite side of the piston, and the velocity of the piston being 

 determined by the effective evaporaliiirj power of the boiler, which we 

 readily allow, the pressure ui the boiler will nccessainly be such as to 

 cause a correspondiirg efflux of steam from the boiler into the steam 

 pipe, and through tliat into the cylinder. If at any moment this 

 were not the case, the pressure would unmediately begin to adjust 

 itself, and would finally remain fixed as soon as it had arrived at that 

 point: suppose, for example, that the pressure in the boiler is too 

 low to cause an efflux at a sufficient velocity to supply the cylinder; 

 the pressure will instantly rise until it be sufficient to cause an 

 efflux at the velocity which will then be required, which is less than 

 before, as the density is greater. This cu-cinnstance has been 

 entirely overlooked, or rather neglected, by M. de Pambour, as well as 

 the effect of velocity on the resistance of the air to the trains, which 

 must considerably affect the results of his experiments. 



Before entering upon the general discussion of the theory, wlien 

 we shall have occasion to revert to M. de Pambour's works, we shall 

 briefly advert to a paper on the application of steam as a moving 

 power, which was published in the second volume of the " Trans- 

 actions of tlie Institution of Civil Engineers." We should not have 

 stopped to notice this paper, but for the medium through which it 

 has been brought before the public, which naturally gives a certain 

 degree of importance to everything therein published, as it must first 

 obtain the sanction of a body of men, who, from their profession, 

 and the eminence which some of its members have attained in that 

 profession, have necessarily considerable influence over the opinions 

 of those who are personally imacquainted with mechanics. 



The author of the paper in question appears to have had less in 

 view the advancement of science, as the title seems to promise, than 

 to create a doubt in the public mind as to the correctness of the 

 official reports of the duty performed by the Coniish engines ; and 

 he appears to have persuaded himself that he has demonstrated, on 

 scientific principles, that those engines could not have performed 

 anything like what they arc reported to have clone, and, consequently, 

 that the reports are erroneous ; but wc hope to be able to show, 

 satisfactorily, that he is labom'ing under a delusion. 



We will, for the sake of argument, allow that the combustion ot 

 71bs. of coal is required to (.-oayert one cubic foot of water into steam, 



