TRANSACTIONS OP SECTION G. 711 



TUESDAY, SEPTEMBER 6. 

 The following Papers were read : — 



1. The Optical Determination of Stress. 1 

 By Professor E. G. Coker, M.A., D.Sc. 



The experimental determination of the state of stress in a body by purely 

 mechanical means and apparatus has the disadvantage that it is necessary for 

 accuracy of measurement that a definite length, area, or volume be maintained in 

 a standard condition, and the stress at a point cannot therefore be accurately 

 determined if the stress is a rapidly varying one. 



The property possessed by glass of becoming doubly refractive under stress 

 has been frequently utilised to determine the state of plane stress at a point in it 

 by the colour fringes produced, but the difficulty of forming any but the simplest 

 objects in glass has prevented its extensive use for experimental work. 



Other substances have been tried, and a preparation of nitro-cellulose in com- 

 mercial use has been found which answers exceedingly well for experimental 

 work. Its properties are very different from glass, and experiments show that 

 the modulus for tension is approximately 300,000 in pounds and inches and the 

 value of ' Poisson's ' ratio 0-37, plate-glass having the corresponding values of 

 10-5 x 10° and 0"227 respectively. For determining stresses a method of matching 

 colours is adopted, in which a uniformly stressed test-bar is loaded until the 

 colour produced by the retardation of a plane or circularly polarised ray corre- 

 sponds to that produced at a point in the object under stress. The relative 

 retardations R of the ordinary and extraordinary rays is assumed to be similar to 

 glass and to follow the law expressed by R = C (X — Y) T, where X, Y are the 

 principal stresses at a point, T is the thickness of the material and C is an optical 

 constant. 



The stresses at the cross-section of an eccentrically loaded tie-bar and at the 

 principal section of a hook are shown to be in fair agreement with theory. 



To determine the lines of principal stress in a body the loci of points at which 

 the directions of the principal stresses are the same are found by using plane 

 polarised light, and from the curves so found the directions of the principal 

 stresses are determined. 



From the curves of principal stress, coupled with a knowledge of the position 

 of the isochromatic lines, the stresses at any point may be determined by the use 

 of Maxwell's method. 



2. On the Direct Measurement of the Rate of Air or Gas Supply to 

 a Gas-Engine by means of an Orifice and JJ-Tube* By Professor 

 W. E. Dalby, M.A., M.Inst.G.E. 



An orifice in conjunction with an anemometer was used to measure the air- 

 supply at the Ashton trials of the Committee of the Institution of Civil Engi- 

 neers, and more recently Professor Ashcroft contributed a paper to the Institu- 

 tion of Civil Engineers describing a method of using an orifice in conjunction 

 with a specially designed indicator to measure the difference of pressure on the 

 two sides of the orifice. In the Ashton trials the air-supply is inferred from the 

 anemometer readings, and in Professor Ashcroft's method the air-supply is 

 inferred from the difference of pressure in conjunction with the orifice, which 

 was made about the same size as the suction-pipe of the engine, in consequence of 

 which the difference of pressure was very small. In each case calibration was 

 effected by driving the engine from the crank-shaft end, and then from indicator 

 diagrams deducing the weight of air passing through the orifice. This deduction 

 cannot be made accurately unless the temperature can be accurately measured at 

 one point on the indicator diagram. Ln neither case could this temperature be 

 measured. The gas-engine used by the author is fitted with apparatus by 

 mean6 of which the temperature corresponding to the pressure and volume 



1 Published in the PhU. Mag., October 1910. 



2 Published in Engineering, September 9, 1910. 



