FL UCTUA TIONS IN MICROPHONES A ND OTHER RESISTA NCES 207 



The single contacts studied were mounted in the cantilever bar 

 device as described in the preceding section. We found it important to 

 wait after the mounting of a contact long enough for the whole bar to 

 come to thermal equilibrium before a measurement was attempted, 

 otherwise very erratic results were obtained. Figure 8 is a typical 

 curve obtained when the mean square noise voltage is plotted in db 

 against the contact resistance on a logarithmic scale. For this 



(O 16 

 O 



z 



_C II 



y^ 



/l» 



1000 5000 10,000 



CONTACT RESISTANCE IN OHMS 



50,000 100,000 



Fig. 8 — The mean square contact noise voltage in a single carbon contact as a 

 function of the contact resistance. The resistance was varied by changing the 

 contact displacement while the applied voltage was held at 0.1 volt. 



measurement the d.-c. voltage on the contact was 0.1 volt. The 

 experimental relationship between noise and resistance is given by 



F 2 = Const. R^. 



(2) 



The data plotted in Fig. 8 give the value 13 = 1.25, which is the average 

 value found for all the contacts studied. For individual contacts j8 

 varied between the extremes of 1.1 and 1.42. The studies on other 

 properties of single contacts also exhibit a rather wide variability from 

 contact to contact, hence the above result is not surprising. 



