Temporal development of spark breakdown in nitrogen and air

Abstract

Measurements have been made of the formative time lag of spark breakdown in nitrogen, dry air and humid air, for overvoltages of 0.05–5%. The effects of variation of gap length and pressure have been investigated for gap lengths of 1–3cm and pressures in the range 200–750mm Hg, and the influence of humidity on the formative time in air has been studied for water-vapour partial pressures up to 10mm Hg. A comparison of results obtained using ultraviolet irradiation with those obtained using cobalt 60 irradiation has shown that the former is to be preferred for accurate formative-time-lag measurements. The influence of pressure, overvoltage and electrode conditioning on the shape of the breakdown oscillogram has been studied, and the significance of the form of the voltage collapse is discussed. The measured time lags in nitrogen are in good agreement with those calculated on the basis of Davidson's theory with 100% photoelectric secondary process, and the correlation is supported by comparison of calculated values of current-growth constant with those measured from oscillographic observations using a photomultiplier technique. The results for dry air are discussed using a simplified theory, and are found to be consistent with a Townsend process having a photoelectric secondary mechanism. It is suggested that the statistical fluctuations in the formative time, observed in humid air, are due principally to the effects of nonuniformity of the cathode surface.