578 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1953 



to have a heavy layer of tin constituent at the contact surface. From the 

 test, it can be concluded that wrapped connections are sufficiently tight 

 to allow solid state diffusion and are therefore good electrical contacts. 



On the other hand the contacts are not welded contacts such as occur 

 when two pieces of aluminum are cold pressed together with strains in 

 excess of 75 per cent. This is shown experimentally by the simple process 

 of imwinding the wire from the terminal which takes place with no excess 

 force when the wire is removed from a terminal corner. Since the strains 

 at the points of contact do not exceed 30 to 40 per cent, one would not 

 expect cold welding. It is possible, however, that in tinned terminals, 

 some long time diffusion takes place at room temperature in the manner 

 demonstrated at higher temperatures by the data of Fig. 19. This would 

 occur very slowly and cannot be relied upon solely to maintain the 

 contact. 



Hence, it appears that long life in the connection depends on main- 

 taining sufficient hoop stress in the wrapping wire to keep the elements 

 of the connection sufficiently tightly pressed together so that no corro- 

 sion can occur in the connection in such a manner as to interrupt the 

 electrical continuity. This is a problem in stress relaxation rather than 

 creep. Stress relaxation is intrinsically a simpler phenomenon since the 

 major fraction of stress that can be relaxed will be relieved through 

 viscous flow in previously formed slip bands or along grain boundaries, 

 and no generation of new slip bands is required. However, under ordinary 

 creep conditions, an increase in stress is presumably attended by the 

 generation of new slip bands. It appears likely then that stress relaxa- 

 tion phenomena even at quite high stresses should more nearly follow 

 the conditions that have been established for low stresses than would be 

 the case for creep phenomena. 



A good deal of work has been done on stress relaxation at low stresses, 

 particularly by Zener'^ and his coworkers, and this will be briefly re- 

 viewed. According to these studies, stress relaxation can be caused by 

 several mechanisms including stress induced migration of impurities in 

 the metal, viscous behavior of slip band material and the viscous be- 

 havior of grain boundaries. At- the common junction between the two 

 metal grains, there is an amorphous layer of material which acts as a 

 viscous medium, i.e., if there is a shearing stress applied across it, the 

 two grains will move with respect to each other with a velocity 



* Zener, C, Elasticity and Anelasticity of Metals. University of Chicago Press, 

 1948. T'ing-Su K6, Experimental Evidence of the Viscous Behavior of Grain Bound- 

 aries in Metals. Phys. Rev., 71, No. 8, pp. 533-546, April 15, 1947. T'ing-Su Kd, 

 Anelastic Properties of Iron. Tech. Publication No, 3370, Metals Technology, 

 Jqne, 1948, 



