Electrode Effects in the Measurement of Power 



Factor and Dielectric Constant of Sheet 



Insulating Materials 



By E. T. HOCH 



Synopsis: It is shown that, aside from the guard ring type of electrode 

 which can only be used with certain special types of measuring circuit, 

 the most accurate results can probably be obtained by the use of equal foil 

 electrodes and making proper allowance for the edge effects. From the 

 standpoint of convenience, mercury electrodes and foil electrodes of un- 

 equal size have certain advantages. It is believed that the results ob- 

 tained with these two types are also sufficiently accurate for most purposes 

 when the corresponding corrections are applied. 



Introduction 



WHEN it is desired to determine the dielectric constant and 

 power factor of an insulating material, the first problem that 

 presents itself is that of finding a suitable means of applying a potential 

 to the material in question. In order to accomplish this, a sample of 

 the material must, in general, be placed between two conductors or 

 electrodes to which the desired potential is applied. The size, shape 

 and manner of application of these electrodes afifect directly the 

 quantities to be measured from which the dielectric constant and 

 power factor are derived. Therefore, unless proper allowance can be 

 made for them, these features of the electrodes will affect the values 

 obtained for the properties of the insulation. 



It is the purpose of this paper to discuss only the part played by the 

 electrodes in the measurement of power factor and dielectric constant 

 and not to discuss complete methods for measuring these quantities. 

 Experimental data will be presented to show the magnitude of the 

 various effects discussed. 



If any form of test is to be of general use for determining the prop- 

 erties of any material, there are certain fundamental requirements 

 which must be fulfilled. F'irst, the method should lead to exact 

 reproducibility of results. That is, a test on a given sample of ma- 

 terial should lead to the same result regardless of when, where or by 

 whom the test is made. Second, the result obtained should be the 

 correct result, that is, the absolute accuracy should be high. Third, 

 the method should be as convenient as possible to use. If it is not, 

 the method loses its practical value to a large extent since the tendency 

 will be for most people to use a more convenient method even at the 

 expense of accuracy and reproducibility. 



555 



