STHKSS SYSTEMS IN TIIK SOLDKKLKSS A\'RAPPKD CONNECTION 1105 



oj 45 



35 



O 30 



O 

 9 25 



Z 20 



1 2 3 4 5 6 7 



HOOP STRESS IN THOUSANDS OF POUNDS PER SQUARE INCH 



Fig. 10 — Activation energy for diffusion as a function of hoop stress. 



that the strength stops decreasing at the end of six months and even- 

 tually increases to 20 per cent over the initial value at the end of forty 

 years. Measurements on connections made one year and ten months 

 ago, taken by V. F. Bohman, confirm this calculation. 



In all probability, the stress causing the diffusion is the shearing stress 

 applied to the tin layer by the hoop stress of the wire acting on the fixed 

 terminal. For a given initial winding stress, this shear stress will decrease 

 in proportion to the relaxed hoop stress as confirmed by the data of 

 Fig. 10. If we start with a lower winding stress, a smaller indentation 

 is made in the wire and the area of contact is smaller about in proportion 

 to the ratio of the hoop stresses. Hence, although the compressional 

 force on the corners is smaller in proportion to the hoop stress, it is 

 supported by a smaller area and hence the stress on the supporting 

 area is approximately independent of the initial value of the hoop stress. 

 This compressional stress produces shear stresses on the layer of tin 

 since the material of the terminal and the wire have to slide with respect 

 to each other in producing the indentation. Hence, we should expect 

 that the diffusion forces would start at a time independent of the value 

 of the initial hoop stress. 



This supposition is confirmed l)y the data of Fig. 11, obtained by 

 winding tin i)lated copper wire on a nickel silver terminal with a hoop 

 force half of the usual winding force of 1,300 grams for a 24 gauge wire. 

 The initial stri]) off force was 00 per cent of that for a 1 ,300-grain winding 



