Mathematical and Physical Papers. 243 



paper or stretching of the cord may cause this. The pencil should 

 be well sharpened and held lightly against the drum, and fire wire 

 substituted for long lengths of cord to remedy this. A good braided 

 cord should be used, as they do not stretch as much as others. 



The deviations of the cord from the perfect card explained so 

 far have been due to faults in the mechanism for taking the card. 

 We will now take up some departures that indicate faults in the 

 engine itself, beginning with the admission line and going around 

 the card in the order in which the events occur. 



The admission line for a properly designed and regulated engine 

 should start from a point one-third to two-thirds the distance up 

 from the bottom or back-pressure line of the card and rise perpen- 

 dicularly to nearly the boiler-pressure line: It is probably better 

 with some types of engines if this line slopes out a little. Should 

 the line slope too much outwardly, it indicates that this event is' 

 too early, and steam is therefore being admitted too much before 

 the end of the stroke has been reached, and will be working against 

 the steam in the other end of the cylinder. If the" line, on the other 

 hand, slopes inward from the perpendicular the valve is not open- 

 ing soon enough, and the piston has started on its return stroke 

 before the steam is admitted. If the admission line rises to the 

 proper height and falls below a horizontal position before reaching 

 cut-off, the passages leading from the boiler to the engine are in- 

 sufficient at some point. Either the pipe-line or the engine-ports 

 or steam passages do not admit enough steam. A diagram taken 

 from the steam-chest will show whether the trouble lies in the 

 steam line or the engine. 



If in the steam line, the steam-chest pressure will drop as the 

 piston travels outwardly on its stroke. At cut-off the line should 

 round with a short bend into the expansion line; failure to do this 

 shows too slow a closing of the valve at cut-off or a leaky valve. 

 If a leaky valve is the cause, it can be told by the expansion curve. 



In an unjacketed cylinder, we would expect the expansion line 

 of the steam to be adiabatic. In J;he actual engine, however 

 (jacketed usually with some non-contracting material), where 

 the cylinder walls furnish some heat during the expansion and en- 

 trained and condensed water is evaporated during the latter part 

 of expansion, the line is found to approach very near to the iso- 

 thermal expansion curve. By drawing this curve in, then, we can 

 determine whether the expansion took place as it should. An ex- 

 pansion curve which is too high shows that steam entered the 

 cylinder after cut-off, and that the valve must leak. A leaky valve 



