Corona losses between wires at extra high voltage

F. W. Peek, Jr.: Prof. Harding has given us some curves up to 500,000 volts on conductors good for operation at 220 kv. or less. His loss measurements under some conditions show a close agreement with the quadratic law over a reasonable range of voltage for the conductors. The difference on several of the conductors is, I believe, due to the fact that sphere gaps were used for voltage measurement. Regarding the divergence at the upper part of the curve where the arc-over voltage is approached, we have sometimes noticed a somewhat similar tendency on measurements that we have made up to one million volts. This has occurred on large conductors at spacing relatively small compared to diameter. It seems to be due to the distortion in the dielectric field that takes place because of the great amount of corona. The ratio of spacing to effective radius becomes small. The flexible corona conductor is distorted and no longer a cylinder. At the voltage when the divergence starts the corona begins to separate into huge cart wheels. These have a shielding effect preventing or lowering the loss between wheels. This does not mean that the quadratic law would not apply for calculation of the loss for million-volt conductors. Larger conductors and greater spacing would be used and the law would hold in the usual way. The tendency to diverge thus seems to apply only to extreme conditions or when spark-over is approached and distortion results.