Gas- Engine Indicator -Diagram, 

 Table VII. 



67 



A. 



£ 



Observed. 

 p. 



Calculated. 

 p. 



Expansion- 

 curve 

 calculated. 



p. 







•889 



45 



45 



173 



•061 



•950 



86 



83 



157 



•111 



1-000 



108 



108 



146 



•170 



1-059 



123 



124 



134 



•211 



1-100 



126 



125 



126 



•311 



1-200 



111 



111 



111 



•511 



1-400 



88 



88 



89 



•711 



1-600 



72 



73 



73 



•911 



1-800 



60 



61 



61 



1-111 



2-000 



52 



52 



52 



The differences between the observed and calculated values 

 of p are within the limits of errors of observation therefore. 

 It may be that there is a discontinuity in the ignition-curve 

 at \=*061 for a mixture of gas and air; but no discontinuity 

 shows itself in the indicator-diagram in a sufficiently marked 

 manner to be distinguished from vibrations due to the spring 

 of the indicator. 



As we have already stated, the recentness of the last explo- 

 sion affects the temperature of the mixture of gas and air, 

 and therefore the mass of the mixture. Hence we often find 

 that the expansion parts of successive indicator-diagrams do 

 not coincide. The law 



where k/, k, and n are constants, satisfies all the diagrams. 



n is a constant whose value depends on the point in the 



stroke at which the maximum pressure occurs ; and this, 



for a given speed of engine, depends principally on the 



proportion of gas to air in the mixture ; 



M is a constant which depends on the recentness of the last 



explosion. 

 7. Simple Formula? for the Ignition and Expansion parts of 

 the Diagram. — The complete formula (3) assumes no want of 

 continuity in proceeding from the ignition to the expansion 

 part of the diagram. For the sake of ease in practical calcu- 

 lations we may, however, assume that the ignition- and expan- 

 sion-curves are quite discontinuous, employing some method 

 of indicating continuity when results are plotted on squared 

 paper. In this case we assume that the lines of fig. 3 are 



F2 



