Reliability of three-electrode spark gaps for synthetic test circuits

Summary form only given. Development of a three-electrode spark gap, intended for use in synthetic circuits, is considered in this paper. Two types of three-electrode gas insulated spark gaps have been tested: one with the third electrode located inside the main electrode, an the other with a separate third electrode. A theoretical model which enables optimal design of the spark gap has been presented. Application of the model to both types of spark gaps enabled their optimal sizing. The following characteristics have been determined experimentally: the influence of insulating gas parameters on spark gap operation, the influence of the polarity of working and trigger voltages on spark gap operation, the influence of the trigger voltage rate of rise on spark gap operation, and the degree of spark gap irreversibility under conditions of exploitation. Three types of insulating gas have been used: SF
₆
, N
₂
, and a mixture of 60% SF
₆
with 40% N
₂
Additionally, three different electrode materials have been implemented: copper, steel and tungsten. Spark gap switching and delay times have been measured as indices of spark gap operation. The statistical analysis of the obtained results is presented in the paper. The testing showed that N? insulated spark gaps have better characteristics than those insulated by SF
₆
, which may be explained by a higher concentration of free electrons in N
₂
compared to the electronegative SF
₆
. It has been found that that both the switching time and its statistical dispersion increase as the trigger pulse rate of rise becomes higher, while the delay time and its statistical dispersion decreased. Behaviour of the switching time random variable indicated the appearance of irreversible changes in the SF
₆
insulated spark gap with copper electrodes. Irreversibility of the spark gap was not affected by either the spark gap type or the triggering mode.