1084 Transactions of the American Institute. 



should be determined. This may be done by blocking the engine 

 upon either half stroke, and letting on a full head of steam. By then 

 examining the exhaust, the leakage, if any, ^rill be detected, and may 

 be measured by causing it to be condensed in cold water, weighing 

 the water both before and after a given time ; or by notinjf the rise 

 in temperature of a given weight of water. 



But the question of re-eyaporation of condensed water within the 

 cylinder while the exhaust-valve is open, is not so easily determined. 

 It may be, and has been measured, by using a surface condenser 

 weighing the water discharged therefrom, and comparing it with the 

 weight of steam exhausted, as determined by the capacity of tlie cyl- 

 inder and passages, and the specific weight of steam at the terminal 

 pressure, and has been found to vary from nine to fifty per cent of 

 the whole amount of steam used, according as means for its preven- 

 tion have been used or neglected. 



This loss, if measured and accounted for in the theoretical diagram, 

 would reduce the apparent rate of the engine ; but as it is nearly the 

 same in all cases, having the same amount of protection to the cylin- 

 der, and is, in a measure, independent of the mechanism of the engiae, 

 we may ignore it in the foregoing calculation, without seriously 

 affecting the value of the latter as a means of comparing the effi- 

 ciency of one engine with another. But we should bear in mind in 

 such comparison, that the effect of this practically unavoidable loss is, 

 in all cases, to belittle the performance of an engine having a 

 good degree of expansion within the cylinder when compared with 

 one having little or none ; because, in the former, a portion of the 

 reevaporation occurs during expansion, and enters into our calcula, 

 tion. We should also remember that when a steam-jacket is employed, 

 its usefulness is not fully apparent in this mode of comparison, for 

 the principal benefit of the steam-jacket being to lessen the loss by 

 condensation, and to increase the proportionate re-evaporation during 

 expansion, the saving in condensation is not represented, while the 

 increase of re-evaporation during expansion, raises the terminal pres- 

 sure, and thus lowers the percentage of the indicator diagram. 



It will be seen that in this mode of comparing the performance 

 of engines, the questions of the relative economy of expansion or 

 non-expansion, of high or low pressure, condensing or non-conden- 

 sing, are not mooted ; the comparison merely being as to the relative 

 value of the mechanical arrangement for utilizing the steam used at 

 the particular pressure, and degree of expansion, and under the cir- 



