306 



SCIENCE. 



[N. 



Vol. VII. No. 166. 



have been somewliat more marked, as the 

 engine of novel proportions has, nominally 

 at least, 65 per cent, higher ratio than its 

 rivals ; but, as a considerable part of this ap- 

 parent expansion-ratio measures free ex- 

 pansion without performance of work, the 

 comparison on this basis would not be 

 strictly correct. ISTo correction is attempted 

 for diiferences in speeds of piston or of revo- 

 lution, on which score the intermediate 

 type of engine would apparently have a 

 very marked advantage ; for, as was long 

 ago pointed out by the writer, where jacket- 

 ing is adopted successfully, variation of 

 piston-speed seems to have little effect on 

 economy. 



The three experimental engines being 

 compared, as proposed, and as indicated 

 in the introductory paragraph, the novel 

 proportions are found, in this case also, to 

 give an efficiency intermediate between the 

 standard compound engine and the standard 

 triple-expansion under similar steam-pres- 

 sures. 



The following table shows the conditions 

 during the most efficient periods of test of 

 the three systems : 



test of the triple and the -most economical 

 test of the 7-to-l compound there is a dif- 

 ference of an inch and a half of mercury 

 in the vacuum, and of four pounds in the 

 boiler pressure. The column showing the 

 three tests reduced to a common back-pres- 

 sure was obtained by increasing or dimin- 

 ishing the mean effective pressure in the 

 low-pressure cylinder of each by the amount 

 each-varied in back-pressure from that of 

 the required mean. In this case the mean 

 was taken as the back-pressure in the 

 triple-expansion test. This correction brings 

 the triple and 7-to-l compound nearer to- 

 gether, but we shall still have a difference 

 in steam consumption of 1.48 pounds of 

 steam per horse-power per hour between the 

 triple and the 7-to-l compound, and a dif- 

 ference of 2.1 pounds between the latter 

 and the compound with the 3-to-l ratio of 

 cylinder volumes. 



The performance, absolute and relative, 

 of these three engines with varying power 

 is illustrated in the accompanying diagram, 

 which well exhibits the curious variation of 

 relation, in this respect, produced by chang- 

 ing conditions of operation. 



COMPAEISON OF THE MOST ECONOMICAL TRIALS. 



Triple. 



7-to-l Com- 3-to-l Com- 

 pound, pound. 



Boiler gauge 



Revolutions per minute 



Vacuum in inches of mercury 



Condensed steam in pounds 



Total jacket- water 



Total steam used 



Total I. H. P 



Distribution of work between cylinders, 



H. P. = 1 



Mechanical efficiency 



Steam per I. H. P. per hour 



Number of expansions 



Steam per I. H. P. corrected to a vacuum of 24. 3 inches mercury. 



119.1 

 84.95 

 24.3 

 1,205 



335.4 

 1,540.3 

 112.65 

 I. C. = 1 

 L. P. C. = 1 

 84.1 

 13.68 

 22 

 13.68 



115 



87.65 



22.84 



1,753.7 



316.7 



2,070.4 



129.97 



1.29 



86.6 

 15.8 

 18.89 

 17.1 



117.5 

 85.52 

 22.7 

 1,030 



190.97 

 1,221.2 

 67.7 

 .635 



90 



18.03 

 15.45 

 17.3 



It will be noticed that in the triple-ex- 

 pansion tests both the vacuum and the 

 boiler pressure are better than in either of 

 the others. Between the most economical 



The curves in this figure leave no room 

 for doubt in regard to the relative economy 

 of the three engines. At about 37 horse- 

 power the steam consumption in each case 



