Controlled experiments on vaccum breakdown

An experiment is described which was designed to study the factors influencing high voltage breakdown in vacuum, with particular application to problems encountered in the development of high power vacuum tubes. The first part was devoted to developing procedures for preparation and voltage conditioning of electrodes in vacuum with minimal sparking. The next objective was to determine if there was a generally reliable and consistent criterion which could be used to predict electrical breakdown of vacuum gaps without damaging the electrode surfaces. In pursuance of this, instrumentation and monitoring techniques were developed in order to describe the gap as fully as possible during the pre-breakdown and breakdown stages. Thus, during stepwise application of voltage, the following were monitored: micro-discharge activity, magnitude and waveshape of gap current, light output, partial pressure of hydrogen or nitrogen and X-radiation. For the second phase of the experiment, the voltage range was extended to 300 kv using a stainless steel bakeable chamber. Also, a statistical design was chosen because it provides a powerful tool for the analysis of the results, and yields information on the effects of both selected individual factors and two-factor interactions with a minimum number of experiments. The first such experiment involving seven factors has been completed and the results analyzed both on a statistical and a physical basis. Some significant trends have emerged from the statistical analysis and, encouragingly, these have been found to be consistent with theories of the mechanism of breakdown developed from the physical analysis.