ELECTRODE EFFECTS 561 



Conducting Paint Electrodes 



Another form of electrode which might be considered as a varia- 

 tion of the foil electrode consists of a coating of conducting paint 

 on either side of the sample. In general the conditions are the same 

 as with foil electrodes. A possible advantage might be more intimate 

 contact with the sample and a possible disadvantage is that the film 

 may have sufficient resistance to materially afTect the power factor 

 measurement. Many metallic paints are almost entirely non-con- 

 ducting and arc therefore entirely unsuiteible for this purpose. If 

 suitable conductivity is obtained, the discussion of foil electrodes 

 given above may be applied to this type also. 



Fixed Gap Electrodes 



Another t\pe of electrode occasionally referred to ^ consists essen- 

 tially of a parallel plate air condenser the capacity of which is meas- 

 ured first alone, and then with the sample of insulating material 

 inserted in the air-gap but not necessarily filling the gap completely. 

 From these measurements and the known dimensions of the air 

 condenser and sample the dielectric constant and power factor of the 

 sample can be computed. This arrangement is capable of high ac- 

 curacy if the dimensions of the sample and the thickness of the gap 

 are known with sufficient accuracy. The computations are not as 

 simple as for the other electrodes referred to and a slight error in 

 determining the thickness of the sample or gap results in a much 

 larger error in the final results. Therefore, this method does not 

 seem to ofTer any marked advantages over the simpler forms. 



Evaluation of Errors 



Having now discussed in a general w^ay the various types of errors 

 and their probable efTect in connection with \'arious types of elec- 

 trodes, an attempt will be made to determine by experiment the 

 magnitude of the more important of these errors with respect to 

 certain definite types of electrodes. From the discussion already^ 

 given, it appears that some form of the mercury or foil electrode 

 should be the most suitable for general use. Hence this investiga- 

 tion will be confined to these two general types. 



The first question, namely, that of reproducibility, is one which 

 cannot be determined by a single observer except as it applies to his 

 own particular method of manipulation. Since the results obtained 



■'A. Campbell, Proceedings of the Royal Society, Volume 78, Page 196 ami D\e 

 and Hartshorn Local Citation. 



