198 BELL SYSTEM TECHNICAL JOURNAL 



The resistance of a granular conductor has been shown experi- 

 mentally to He almost entirely within very small volume elements in the 

 regions of the contact areas.* It is our hypothesis that there exist 

 minute fluctuations of resistance in the region of contact, and when 

 such an element carries direct current a potential fluctuation between 

 the terminals can be observed. Accordingly we propose for this 

 phenomenon the term "contact noise." 



In a study of the electrical disturbances in a carbon transmitter 

 Kawamoto 2 found that in addition to "carbon burning," which is a 

 sharp crackling noise sometimes present in the carbon transmitter 

 when the voltage across individual contacts is of the order of 0.5 volt or 

 greater,^ there is a continuous rushing sound which is always present 

 no matter how well the transmitter is shielded from external dis- 

 turbances. Kawamoto applied the term "carbon roar" to this 

 phenomenon. Frederick * in discussing the disturbances in the carbon 

 transmitter states that the noise power is proportional to the square of 

 the direct current passing through the transmitter. More recently 

 Otto ^ — ^who has been working on this subject contemporaneously with 

 ourselves — has reported the results of an extended investigation of this 

 phenomenon. The present report parallels to some extent the study of 

 Otto but in addition new aspects of the phenomenon have been in- 

 vestigated, more accurate data have been obtained, and the conclusions 

 drawn from these experimental results are fundamentally different 

 from those of Otto. 



Electrical disturbance in grid leaks, which becomes evident with 

 the passage of current, was first reported by Hull and Williams ^ who 

 observed the phenomenon in resistances formed by an India ink line. 

 Preliminary reports have since been published concerning such noise in 

 thin metallic films on glass.'' The observations of Otto ^ were also 

 extended to fine carbon wires and copper-oxide resistances. More 

 recently Meyer and Thiede ^ have investigated the noise in resistances 

 consisting of thin films of carbon on a refractory base. 



We have performed noise measurements on each of the types of 

 resistance elements mentioned above and the experimental results 



» F. S. Goucher, Jour. Frmtklin Inst. 217, 407 (1934); Bell Sys. Tech. Jour. 13, 163 

 (1934). 



* T. S. Kawamoto, Unpublished Report, Engineering Division, Western Electric 

 Company, April, 1919. 



'This disturbance undoubtedly has its origin in the heat generated at the carbon 

 contact by the passage of current. 



* H. A. Frederick, Bell Telephone Quarterly 10, 164, July, 1931. 



^ R. Otto, Hochfrequenztechnik und Elektroakuslik 45. 1S7 (1935). J 



8 A. W. Hull and N. H. Williams, Phys. Rev. 25, 173 (1925). J 



^ G. W. Barnes, Jour. Franklin Inst. 219. 100 (1935). J 



» Erwin Meyer and Heinz Thiede, E.N.T. 12, 237 (1935). ■ 



