Development of Multistage Pseudospark Switches for High Voltage Applications

Abstract

For the injection/extraction kicker magnet system of the future heavy ion synchrotron accelerator complex SIS 100/300 at GSI the design for pulse forming network (PFN) is under discussion. One of the key elements of the PFN is the high power switch. Different switch types like thyratrons, solid-state devices and ferrite switches have been studied. A promising alternative represents the pseudospark configuration. The pseudospark is a low- pressure, high power switch, similar to the thyratron, but without an hot cathode. Cathode emission is described by the so-called superdense glow model. Therefore it is also indicated as cold-cathode thyratron. This switch combines excellent switching performance with low housekeeping power requirement and a simple setup. Commercially available single gap pseudospark switches have already proved their reliability and performance for hold-off voltages up to 32 kV and peak currents up to 30 kA. The current rise rate can be as high as 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">12</sup> A/s and the overall switching characteristic is comparable to that one of thyratrons. The aim of the development is to extend the one-gap geometry to a multi-gap switch system which can handle hold-off voltages of 100 kV and peak currents of 8 kA for ~6-8 mus pulse length. Another demanding requirement is its extremely low misfiring-rate of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-9</sup> . To overcome the physical limit of 35-40 kV in hold-off voltage for a one-gap system a four-gap device is proposed to guarantee the reliable operation for 100 kV. In order to study basic fundamental problems related to multi-gap systems, actually a two-gap pseudospark is used to test different design parameters and components, e.g. trigger modules, plasma coupling between the gaps and so on, and first results are presented in this paper. The two-stage system is used as prototype for the development of the four-stage pseudospark switch. The preliminary trigger setup leads to delay values of 185 ns and jitter below 15 ns (full width half maximum) without optimizing the working-gas pressure within the switch. At comparable pressure the hold-off voltage of the two-gap switch is about twice as high as in a equivalent single-gap system, but this switch was not yet tested beyond 25 kV due to limitations in high-voltage equipment.