TRANSACTIONS OF SECTION G, 789 



varying from one minute to forty-eight hours. It was found that their diameters 

 were reduced by this treatment, the greater part of the permanent set having 

 taken phice during the tirst fifteen minutes, and that its amount increa?ed with 

 the ' follow ' originally given to the springs. It was also found that a measurable 

 amount of set was producible by merely squeezing the rings between the hands. 



After the rings had been compressed to a definite size for two days or more they 

 acted as reliable springs so long as the initial amount of compression was not exceeded. 



Some slight inaccuracies in published formuhe applicable to rings formed of 

 materials such as steel, not liable to permanent set from such light loads as piston 

 rings are subject to, were briefly alluded to and a correct formula given ; lout it 

 was pointed out that that usually published was simpler and sufficiently accurate 

 in use, though not applicable to cast iron. Some elaborate formulie had been 

 published by K. V. (Uark on the 'Theory of Cast-iron Beams,' ' and could perhaps 

 be raodilifd to suit the special case luider consideration; but they were not con- 

 venient, and the proportions found by direct experiment appeared preferable ; 

 moreover the amount of set, and consequently the actual (as compared with the 

 initial) follow, could only be found by trial. 



The amount of 'spring' in a cast-iron ring with loads not exceeding that 

 which had produced permanent set was much greater than appeared to be generally 

 realised. A spring ring about 5^ inches diameter, 0'25 inch thick radially and 

 with its ends about ]^ inch apart (after deducting the permanent set), could be 

 bent until they met, and when released they sprang back again to 1^ inch apart. 



4. A farther Note on Gas-engine Explosions. By H. E. Wimperis. 



In the autumn of 1901 the Institution of Mechanical Engineers published the 

 Second lieport of the Gas-engine Research Committee drawn up by Professor 

 F. W. Burstall, and in this manner there was presented to the scientific and 

 engineering world a very large mass of experimental results containing most 

 valuable information regarding the internal economy of gas engines. Professor 

 Burstall c<mies to several conclusions, prominent among which is his contention 

 that the results of his experiments cannot be reconciled without the use of a 

 variable specific heat. 



The writer has already published in the engineering press an account of 

 certain consequences which follow from this hypothesis, but before the matter 

 could be considered upon a satisfactory basis it was necessary that the classical 

 experiments of Mr. D. Clerk and Mr. Qrover should be brousrht info line. In a 

 paper contributed to the British Association last year, the wiiter analysed Mr D. 

 Clerk's results and showed that they were in complete harmony with the variable 

 specific heat hypothesis. In the present paper the writer deals with the experi- 

 ments of Mr. Grover. and shows not only that there is no inconsistency between 

 this hypothesis and Mr. Grover's results, but that by its adoption the explanation 

 of certain diiiicnlties in these results ia the more easily given. 



Thus in the present paper and in the one submitted last year the writer claims 

 to have achieved the object with which he set out — namely to ascertain whether 

 there was anything in past records of experimental work with gas engines which 

 would prove to be incompatible with the adoption of a variable specific heat in 

 all future gas-engine calculations. The conclusion is that there is nothing in the 

 classical investigations of Mr. D. Clerk or Mr. Grover which should afl'ord any 

 around for hesitation in the matter. 



5. Preliminary Experiments on Air Friction. By Wm. Odell, A.R.C.Sc. 



These experiments were begun with the object of finding a convenient method 

 of determining the power wasted by the windage of flywheels and dynamo 

 urmatures. The experiments described at length were made with paper discs, which 

 ■^y^re mounted on the shaft of an electric motor. 



> Minuses (\f Proceedings f>(s(itution qf Cjiil Enghiecrs, vol. cxlix. p. 313. 



