Stochastic Nature of Electrical Breakdown in Vacuum

Abstract

This paper presents measurement results for the random variable "vacuum breakdown voltage", obtained by the application of pulse, dc, and ac voltage. Parameters varied during the measurements were: 1) vacuum pressure and 2) interelectrode gap width. The measurements were performed under strictly controlled laboratory conditions. Conclusions about the influence that the variable parameters in the experiment exert on the parameters of theoretical statistical distributions have been drawn, based on the statistical analysis of the obtained experimental results. The obtained results have been explained by the mechanisms constituting the process of electrical breakdown in vacuum.

Time enlargement law for gas pulse breakdown

“…In SF 6 insulation, longterm investigations have revealed a time influence on the electric strength, but it is far weaker than the enlargement law would lead us to expect. This demonstrates the existence of dependences related to the movement of microscopic dustparticles in the insulating gas [3].…”

“…an accurate determination of the breakdown voltage and of the probability assigned to it. The phenomenon of gas electric breakdown represents a self-sustaining process, dependent on the relative activity of the elementary processes of gas discharge: ionization, recombination and attachment [1][2][3]. When the time constant specifying the change of the voltage applied to a gas insulation is significantly longer than the time characterizing elementary processes that lead to gas breakdown, the probability assigned to the value of the breakdown voltage is close to 100%, which means that the process is regarded as deterministic (experimental determination of the breakdown voltage in such a case includes only type B measurement uncertainty).…”

Surface-time enlargement law for gas breakdown

“…Consideration of the dependence of breakdown voltage and breakdown time random variables on the geometrical enlargement factor was based on statistical samples composed of 100 values from a single series of measurements, submitted to the Chauvenet's criterion for rejecting spurious measurement results [3]. Conidersation of the dependence of breakdown voltage random variable on the time enlargement factor was based on statistical samples composed of 1000 values from a single series of measurements, submitted to the Chauvenet's criterion.…”

“…If a largescale insulation medium is exposed to sources of high power, electrical discharges in it can occur in parallel with respect to time and space. Random point at which electrical discharge develops most rapidly produces the breakdown of the entire insulation medium [1][2][3].…”