ill I 



.>/ 



G 



HKI'OUT -1884, 



Tlic tlicvnindx imniic tlicory of llio stciim-eii^iiio stands, to-day, sub- 

 stantially as it was kfthy (Mansiiis and l{ankinoat tho (rioso of their work 

 in this fichl, in iho dccado iHoO lo 1H(50. Many treatises liavo leeii 

 published, sonit! of flu'in by men of exceptional ability ; but all have 

 followed tho fjfeiieral line; iirst drawn by tlieso masters, aiiil have ordy now 

 and then found sonio minor point to develo{)c. Uankine's ' Steara-eni^iiic 

 and other Prime iMovers,' written a quarter of a century apfo, is still a 

 standard work on thi; tluory of iIk; heat-engines, and is still used as a 

 text-book in enf»ineerinjr schools in this country and Kurope. 



'^I'ho limitations of tlu; tiiermodynamic theory of tho heat-engines, 

 and of its application in the design and operiitimi of such engines, were 

 tirst discovered by James Watt a hundred years ago and more. They 

 were systcmalieally and I'xperimentally investigated by isherwood in 

 1855 to IHG"), were observed and coi-rectly interpreted by Clark in iH";') 

 and. earlier, and were revi'aled again by the experiments of Hirn, an<l by 

 those of Kmeiy nnd many other recent investigators on both sides of the 

 Atlantic. 'I'liese limitations are duo to the fact that losses occur in tlio 

 operation of steani-cngines which arc not taken into account by tlic 

 hitherto iiccepted theory of tho engine, and have no ])lace in the ther- 

 modynamic treatment of tho case. 



It is generally assumcnl, in tho usual theory of the engine, that the 

 expansion of the working iluid takes place in a cylinder having walls 

 impermeable to heat, and in which no losses by conduction or radiation, 

 or by leakage, can occur. Of those losses which actually take place 

 in the real engine, that due to leakage may be prevented, or, if occur- 

 ring, can be checked ; but it is impossible, so far as is now known, to 

 secure a working cylinder of perfectly non-conducting material. Tln' 

 consequence is that, since the; steam or other working fluid enters at a 

 high temperature and is discharged at a comparativel}' low temjicra- 

 ture, the surfaces of cylinder, cylinder heads, and piston are at one in- 

 stant charged with heat of high temperature, and at the next moment, 

 exposed to lower temperatures, are drained of their surplus heat, wliicii 

 heat is then rejected i'rom the cylinder and wasted. Thus, at each stroke, 

 the metal surfaces, exposed to the action of tho expanding substance, 

 alternately absorb heat i'rom it, and surrender that heat to the 'exhaust.' 

 In the case of tho gas-engines, this waste is rendered enoi'mously greater 

 by the action of tho water-jacket, which is there needed to keep tho 

 cylinder down to a safe temperature, and which takes away, in the circu- 

 lating stream of cooling water, an immense amount -usually about one- 

 half — of the heat received from the burning gas. In tho steam-engine, 

 the loss by the method here referred to is raiely less than one-fourth in 

 unjacketed cyilndei's, and is often more than equal to the whole quantity 

 of heat transformed into mechanical ercrgy. Tho amount of this loss 

 increases with wet steam, and is diminished by any expedient, as steam- 

 jacketing or superheating, wliich jirevcnts the introduction or the pro- 

 duction of moisture in tho midst of tho mass of steam in the cylinder. 

 As the range of temperature worked through in the engine increases as 

 the quantity of steam worked per stroke diminishes, and as the finu' 

 allowed for transfer of heat to and from tho sides and ends of the cylinder 

 and tho piston is increased, the magnitude of this loss increases. Hence 

 the use of high steam, of a high ratio of expansion, and of low piston 

 speed tend to increase the amount of this waste ; while low steam, a low 

 ratio of expansion, and high engine speed, tend to diminish it. These 



