[boyle] 



TENSILE STRESS ON ELECTRICAL RESISTANCE 



187 



Tihe last four points of the Aluminium cume showed that the slope 

 was increasing, and, probably, if greater extensions could be obtained, the 

 results would agree with thosie obtained for the other materials. 



These irregularities, in the Manganin and the Aluminium curves. 

 were too small to make any appreciable chajige in the calculations for 

 specific resistance. 



The values obtamed in all cases for the specific resistance, working 

 on the assumption that the wire has constant volume, are remarkably 

 constant. 



As m all work done htereiSofore on this subject, the conclusion is that 

 a longitudinal tensile stress produces practically no change in the specific 

 resistance of the material. At least, it produces no immediate change; 

 whether, after a prolonged action of the stress, there would be a change 

 or not, is perhaps open to question. The changes in specific resistance 

 for copper and aluminium, in this work are smaller than those obtained 

 by Tomlinson and by Gray and Henderson in their work. Tomlinson 

 says that Uhe specific "resistance of Aluminium is actually decreased by 

 a stress in the line of flow of the current. This also shows here in the 

 results for Aluminium, but the change is only in the fourth significant 

 figure, and that perhaps is too fine a distinction to depend on. 



As a general conclusion, this work points to the fact that for alloys 

 and metals, while they are under a tensile stress, the increase per 

 cent resistance is about twice the increase per cent stretch, and that the 

 specific resistance of the material is practically unchanged. 



It is interesting lo finish this paper with a contrast between the 

 small effect of a tensile stress (or, for that matter, of any mechanical 

 treatment) on the electrical properties of a metal or alloy, and the great 

 etfect of even a trace of impurity. 



This is well shown by the following table, taken from Vol. 2, p. 299. 

 Mathematical arid Physical Papers, Kelvin. 



Qualitative Analysis, 



Cu, Fe, Ni, As, O 



Cu, Fe, Fi, O 



Cu, Fe, Ni, (doubtful), O... 

 Cu, Fe, Ni, (doubtful), O. . 

 Cu, Fe, 



Percentage 

 Cu. 



Percentage 

 Impurity. 



Relative 

 Conductivity 



98.76 

 99.20 

 99.53 

 99.57 

 99.90 



1.24 

 0.80 

 0.47 

 0.43 

 0.10 



42.0 

 71.3 



84.7 



86.4 



102.0 



It can be seen, that the purity of the material to be used for a 

 conductor is of the fir-st importance, while the mechanical treatment of 

 it is of minor consideration as regards its purely electrical properties. 



I desire to thank Dr. Barnes for his advice in this work. 



