1893.] 



on Economies in the Utilisation of Energy. 



87 



70- 



60 



as great as 2000° F. Hot air occupies an intermediate position. 

 With ether or ammonia the temperatures are all very much lower. 

 From onr present point of view it is necessary to remember that all 

 these divers working fluids work under the same physical laws and 

 limitations. How, then, do they stand in reference to this problem 

 of economy ? 



First of all, the mere nature of no one gives it any thermo-dynamic 

 advantage over the others. The fact that steam can be liquefied 

 in a condenser is a great convenience to us, but does not make it, 

 per se, one whit worse or 



better than air, which can- F IG< 3, 



not be liquefied — at least % 

 in a condenser. The man- 10< > 

 ner in which the choice of 

 fluid affects the value of the 90 

 maximum possible thermo- 

 dynamic efficiency is indi- 

 cated in Fig. 3. With gas, 80- 

 the highest temperature is 

 the temperature of com- 

 bustion ; the theoretical 

 maximum efficiency is on 

 this account already ex- 

 tremely high. With air, 

 the temperature is only in- 

 directly derived from the 

 temperature of combustion, 

 and the maximum efficiency 40 

 is smaller, although it is 

 still high. With steam 

 (apart from super-heating) 30 

 the temperature is depen- 

 dent upon, and limited by, 

 considerations of pressure, 

 and consequently of safety. 

 With ammonia it is simi- 

 larly limited ; but the en- 

 gine lies much further 

 down the scale, both as to 

 T x and as to T 2 . 



From Fig. 3 we may gather that with gas engines the theoretical 

 efficiency is already so high that we need hardly trouble ourselves 

 about attempting to raise it. With gas, in fact, and to a smaller 

 extent with air, the possibility of improvement lies in bringing the 

 actual up to the theoretical process, and not in attempting to raise 

 the efficiency of the latter. With steam, however, it is different. We 

 want much to raise the theoretical limit of efficiency. But here we 



50- 



20 



10- 



