At its simplest level, an insulated gate bipolar transistor (IGBT) is a switch used to permit power flow in when it is on and to stop power flow when it is off. An IGBT is a solid-state device, which means it has no moving parts. Instead of opening and closing a physical connection, it is operated by applying voltage to a semiconductor component, called the base, which changes its properties to create or block an electrical path.
The most obvious advantage to this technology is that there are no moving parts to wear out. Solid-state technology is not perfect, though. There are still issues with electrical resistance, power requirements, and even the time required for the switch to operate.
An insulated gate bipolar transistor is an improved type of transistor engineered to minimize some of the disadvantages of a conventional solid-state transistor. It offers the low resistance and fast speed when switching on found in a power metal–oxide–semiconductor field-effect transistor (MOSFET), though it is slightly slower to turn off. It also does not require a constant source of voltage the way other types of transistors do.
When an IGBT is turned on, voltage is applied to the gate. This forms the channel for the electrical current. The base current is then supplied and flows through the channel. This is essentially identical to how a MOSFET operates. The exception to this is that construction of the insulated gate bipolar transistor affects how the circuit turns off.
An insulated gate bipolar transistor has a different substrate, or base material, than a MOSFET. The substrate provides the path to electrical ground. A MOSFET has an N+ substrate, while an IGBT’s substrate is P+ with an N+ buffer on top.
This design affects the way the switch turns off in an IGBT, by allowing it to occur in two stages. First, current drops very quickly. Second, an effect called recombination occurs, during which the N+ buffer on top of the substrate eliminates the stored electrical charge. With the off switch happening in two steps, it takes slightly longer than with a MOSFET.
Their properties allow IGBTs to be manufactured to be smaller than conventional MOSFETs. A standard bipolar transistor requires slightly more semiconductor surface area than the IGBT; a MOSFET requires more than twice as much. This significantly reduces the cost to produce IGBTs and allows more of them to be integrated into a single chip. The power requirement for operating an insulated gate bipolar transistor is also lower than with other applications.