Abstract
To clarify some of the variables associated with the origin and detection of cable corona, two solid dielectric cables were constructed and tested. One, with various types of corona-producing elements introduced into it, was scanned at progressively increasing voltage. Some corona elements gave signals whose intensity increased with voltage. The intensity of others changed very little, presumably because the number of discharging sites within a single element increased with voltage. This report demonstrates that the magnitude of corona signals does not have a direct relationship to the corona current being discharged. The second cable (1000 feet) was made from a number of lengths of an RG8U coaxial cable (50 ohms) connected by matched T connectors at various positions along its length. One end was connected to a spectrum analyzer, the other left open. To simulate corona, frequencies from 0 to 3 MHz were injected in tum (at each connector) through a 10-pF capacitor. The magnitude of the detector response depended on frequency, the location of the T at which the corona-simulating spectrum was injected, and total cable length.

This publication has 6 references indexed in Scilit: