Determining the losses in spark gap switches

The variation in losses in gas plasma switches is difficult to measure directly for most practical switches. The approach usually adopted is to measure the voltage across the switch and the current passed through it. The problem area is measuring the voltage, which fast becomes very small compared to the voltage applied to the system as the current through the switch rises. The measurement inevitably loops some magnetic field, and the B signal becomes a significant part of the measurement. Other extraneous signals can also interfere with the measurement. The method described in this paper is a curve fitting method. It relies on the assumption that the loss mechanisms can be closely approximated by a model with a resistance in series with a constant voltage drop, and that the value of these elements is the same for alternating and unidirectional current flows. Excellent fits were in fact obtained which lends credence to the method.