1915] on Gasaous Explosions ,293 



temperatures obtained in this way. A direct therm ometric determina- 

 tion, however, has long been desired, and experiments such as 

 Hopkinson's ah'eady described endeavoured to make the direct 

 determination by platinum resistance thermometers. Such thermo- 

 meters, however,' are subject to the melting point of platinum, and 

 cannot be used with accuracy when near that point, and of course 

 cannot be used at all for temperatures above the melting point. 

 Prof. Coker has accordingly made experiments with thermo- 

 couples formed from alloys of platinum with rhodium and iridium. 

 These thermo-couples were made of alloy of from yo^oo" to rot o^ of 

 an inch thick, and they have withstood the temperature of explosion 

 near the walls of a gas engine cylinder for hours and even days, pro- 

 vided the engine was not overloaded and the temperature raised 

 seriously above 200<)° C. Prof. Coker determined gas engine 

 flame temperatures by such couples, the actual temperature measure- 

 ment being made by observing the change in the electromotive 

 force produced in the couple by a change in temperature. Devices 

 were arranged whereby small changes in electromotive force could 

 be accurately measured. 



Fig. 22 shows Prof. Coker's arrangement for determining 

 temperature in a gas engine cylinder in a diagrammatic way. The 

 battery B and resi'stances Rj and Pvo are arranged in circuit, so that 

 the fail of potential between the extreme points of a bridge wire BAY 

 can be adjusted to one milli volt. This is tested by the electro- 

 motive force of a cadmium cell C, which can be opposed to the battery 

 electromotive force by means of the upper key K^ and allowance for 

 the known temperature variation of the electromotive force of the 

 standard cell was being made by an adjustable contact-maker D. 

 The thermo-couple H has one lead connected to the lower lead K^, and 

 the other to a set of resistances S in the main circuit, each of which 

 gives a difference of potential of one milli volt when the adjustments 

 are correct. During an observation the battery electromotive force 

 opposes that of the couple, and the readings of the bridge wire and 

 step resistances taken together measure the electromotive force of the 

 couple when the galvanometer G shows a balance. By this method, 

 Prof. Coker determined the temperature of explosion in a National 

 Gas Engine with a weak mixture as 1690° C, but he considered his 

 instrument to have a lag, and he believed the actual temperature 

 attained by the flame at half an inch from the walls of the cylinder to 

 be above 'l850°C. and less than 1950° C. In later experiments he 

 came to the conclusion that with a similar engine using weak mixture 

 and giving an explosion at every cycle at about | load the temperature 

 of 1800° C. was reached at a distance of half an inch from the walls. 

 He also determined the exhaust temperature by a similar instrument ; 

 the maximum temperature in the exhaust pipe after leaving the engine 

 was about 500° C, and the temperature drop of the gases from the 

 moment of opening the exhaust valve to the temperature of the gases 



