THEORY OF THE STEAM ENGINE. CXXXVII. 
(A) Quality of steam at admission. 
(j) Several special expedients arbitrarily applied, ¢.g., steam 
jacketing, superheating, introduction of air into the 
cylinder, and so forth. 
(a) Increasing the speed of an engine lessens in a very marked 
degree the condensation waste, but authorities differ widely as to 
the precise mode of variation. In several proposed formule, 
based on experimental results, the loss is stated as varying 
inversely as the square root! of the number of revolutions per 
minute, in others inversely as the first power’ of the number of 
revolutions, while yet in at least one other as the 2 power.? It 
should be remembered that although increasing the speed 
diminishes the loss by initial condensation, it increases the fric- 
tional loss and may be the cause of incurring other expenses to 
an extent that will discount very considerably the primary gain. 
(8) Variations in the admission steam pressure have an impor- 
tant influence on the magnitude of the thermal loss, whereas the 
effect of altering the back pressure appears to be almost insensible, 
With any particular engine the total number of heat units given 
up to the cylinder walls per revolution increases with the pressure* 
and may be expressed by an equation of the form 
H=-KP+C 
where H is the heat loss, P the initial pressure, and K and Care 
constants depending on the other working conditions. The weight 
of steam condensed does not however increase as fast as the total 
weight of steam used, so that the percentage condensation 
diminishes with increase of initial pressure. This fact is of great 
importance in that it distinctly contravenes the assumption very 
commonly made that percentage condensation is directly propor- 
tional to the temperature range from admission to exhaust, which 
is equivalent, if the back pressure be kept constant, to assuming 
ea Gee 
1 Cotterill— The Steam Engine,” p. 339. ‘Thurston—“ Manual of the 
Steam Engine,” p. 510. 
2 Journ. Frank. Inst., Feb. 1886. 3 Industries, Oct. 1890. 
* Proc. Inst. C.E., Vol. cxx., p. 327. 
