When talking about solid state in terms of pulsed power, it refers to semiconductors such as silicon, silicon carbide and gallium nitride. If any one material can define the start of the modern age, it could be the solid-state transistor in 1947. Today, solid-state components are ubiquitous, located in everything from cell phones, refrigerators, cars, televisions, computers and so on.
For pulsed power, solid-state components are typically the replacement for switches such as trigger spark gaps, vacuum gaps or mechanical relays. In many applications these older, tried and true, technologies are suitable, if you can find one. However, they suffer from one major drawback which is that they have a short lifetime. Lifetimes might be measured in tens of thousands of pulses or months of operation in a repetitive pulsed power environment.
Solid-state components are not just used for longer lifetime. The solid-state switch can have a shorter turn-on time with less jitter. A solid-state switch can be used to turn-off current, providing pulse width control down to nanosecond time scales. A solid-state switch can also reduce the EMI noise emitted by a circuit due to the their softer turn-on. The addition of a solid-state diode can improve the efficiency of a circuit design. There are always trade-offs that have to be considered but there are many solid-state options to choose from.
There are three basic switching device types for solid-state pulsed power applications. Those are MOSFET, IGBT and thyristors. Each of those types has many variants. MOSFET devices are useful for when precise turn-off control is required. IGBT devices typically have a longer turn-off time than MOSFET but they typically have lower on-state resistance. Thyristors typically have very long recovery times and cannot be used for turn-off applications as easily as MOSFET or IGBT devices. However, Thyristors have very low on-state resistance and are, therefore, the most efficient for typical capacitor discharge applications.
Many of Sanders Pulsed Power’s designs use thyristors for the low on-state resistance. Though, we do use IGBT and MOSFET devices where they are the most applicable, such as using IGBT devices in the Electrostatic Quadrupole Pulse Generators at the Muon g-2 experiment at Fermilab or using MOSFET devices in the inverter stage for a capacitor charging power supply. It is important to use the right device for the application.
Thyristor devices can be used to replace even high-power thyratrons. Examples of this can be seen at Argonne National Laboratory and Oak Ridge National Laboratory. These conversions from thyratrons to thyristor based solid-state switches will result in hundreds of thousands of dollars in cost savings just for Oak Ridge National Laboratory.
On a even higher power commercial scale, fusion power will need to operate at megajoules per pulse with design lifetimes of billions of pulses. This will only be achieved with solid-state devices and thyristor based devices will result in the highest efficiency. Integrated parallel and series thyristor devices will result in the most cost-effective approach.