Solid State Pulsed Power Systems

A Pulsed Power System is characterized by its energy storage system, the energy can be released in the form of a high power pulse to the load, by means of a switching device. The energy storage is usually either of an inductive or a capacitive nature. The limiting device in a Pulsed Power System is often the switch, which limits the pulse peak power and the repetition rate. The conventional approach in Pulsed Power designs is to use a gas switch such as a thyratron, ignitron or spark gap. However these devices have limited lifetime, high cost, low repetition rate and high losses. Thanks to the continued improvement of the high power semiconductors in switching speed, voltage and current rating, solid-state semiconductors have become the device of choice for pulsed power systems. In the last decade Physique & industrie has successfully developed a variety of Pulse Power Systems based on solid-state switches to cover the vast range of application requiring high voltage pulses. This paper describes the architecture of several pulse power systems and their typical characteristics. I HIGH POWER SOLID-STATE SWITCHES The first semi-conductors used as high power switches were the Thyristors and GTOs, they can handle several kilovolts and kilo-amps, however their low switching speed is a serious limitation. Today, the semiconductor devices of choice5 for high power switching are the MOSFETs (Metal Oxide Field Effect Transistor) and the IGBTs (Insulated Gate Bipolar Transistor). The MOSFETs are substantially faster than IGBTs (typical switching time: 200ns for an IGBT, 20ns for a MOSFET), but IGBTs are more efficient at high voltage (less losses), cheaper per kW switched, and are being manufactured at higher voltage ratings (up to 6500V), where the MOSFETs are limited to 1200V. So generally, MOSFETs are used only when a high switching speed is required. However sometimes MOSFETs are preferred even for lower speed switching because they are more easy to stack in a series/parallel assembly than IGBTs. Another type of solid-state high power switch is the saturable core inductor. This switch can perform nanosecond switching times of almost unlimited power. It is by nature extremely robust, yet it is not a gated switch. It is used in conjunction with a semiconductor switch to obtain faster switching time. The switching time of a saturable core inductor is defined by the time it takes to saturate. In the close state the impedance of the switch is the one of the saturated inductor (the relative permeability drops to 1, i.e. as if there was no core). In the open state the impedance of the switch is the one of the unsaturated inductor, which is typically about 10000 times the saturated impedance. Saturable inductor switches are usually used in series with a semiconductor switch to sharpen the pulse shape, or in magnetic compression designs, to obtain pulse power in the hundreds of GW range. The use of saturable core switches leads to generators with fixed pulse width. The choice of the modulator switch is closely tied to the one of the modulator topology, to achieve the requirements II PULSED POWER SYSTEM TOPOLOGIES 1. Capacitor Discharge This section describes different topologies of pulse generators with direct transfer from a capacitive energy storage to the load. The major benefit of these kinds of systems is that they deliver square output pulses synchronous with the switch gate control. However, for the pulse to be quasi-square the energy stored in the capacitor has to be one or two order of magnitude higher than the energy delivered to the load. The most basic Power Pulse Generator requires a simple energy storage bank (capacitor or line) charged with a DC high-voltage power supply, in series with a switch. When the switch is closed, the voltage is directly applied to the load.