IGBT, MOSFET, and Transistor: Application Scenarios and Differences

Some devices use a component that sounds very impressive—IGBT (Insulated Gate Bipolar Transistor). From high-speed trains and electric vehicles to small appliances like air conditioners and induction cookers, it’s commonly found everywhere. You might be wondering, what exactly is IGBT, what is its purpose, and how does it differ from MOSFETs? To understand these questions, let’s start with a common switch.

In the diagram below, a switch is connected to a power source on the left and a load on the right. When the switch is toggled continuously, the voltage waveform on the load is a PWM (Pulse Width Modulation) wave. The circuit looks simple but is extremely important. You could say that most circuit boards contain a version of this circuit.

For example, in electric vehicles, the internal model is essentially a battery connected to a motor. Of course, a motor controller, also known as an ESC (Electronic Speed Controller), is needed in between. The core component of this ESC is actually a switch, which generates a PWM wave through high-speed switching. If you drive faster, the switch stays closed longer, and if you drive slower, the switch stays closed for a shorter time. Similarly, in air conditioners, the inverter that is often advertised works on the principle of converting 220V AC into DC through rectification and filtering, and then using a PWM switch controller to adjust the compressor speed, thus controlling cooling performance. Even in switch-mode power supplies (SMPS), such a constantly switching switch is used to regulate output voltage. Understanding the principles of switching circuits will make it much easier to learn about switch-mode power supplies, inverters, and motor drives.

Limitations of MOSFETs

In reality, we need a microcontroller to control this switch, so we would use a MOSFET to replace it. As discussed in the previous brushless motor article, MOSFETs are widely used for switching. When you apply a high voltage to the gate, the MOSFET between the drain (D) and source (S) behaves like a closed switch. When a low voltage is applied to the gate, the MOSFET behaves like an open switch. Normally, MOSFETs are sufficient for common switching scenarios. However, in some cases, a MOSFET cannot function as the switch.

If you recall the scenarios involving IGBTs mentioned earlier—high-speed trains, electric vehicles, and induction cookers—they share one common feature: high voltage. This is where the problem lies. The biggest issue with MOSFETs is that they cannot handle high voltage. Typically, a MOSFET can only tolerate a maximum voltage of about 400V. Therefore, in situations with high voltage, like high-speed trains, electric vehicles, and induction cookers, using a MOSFET is not suitable, as it can easily get burned. So, what components can withstand high voltages? The answer is bipolar junction transistors (BJTs). BJTs can handle very high collector-emitter (CE) voltages, up to several thousand volts. So, can we use BJTs in electric vehicles? The answer is no.

Disadvantages of BJTs

While BJTs can tolerate high voltages, electric vehicles require large currents. This means that a large current will flow through the switch. If we replace the switch with a BJT, a large current will flow between the collector (C) and emitter (E). We know that the current between the collector and emitter in a BJT is equal to β (the current gain) times the base-emitter (BE) current. Generally, β is around 100. So, if we want 100A of current to flow between the collector and emitter, we would need to supply a 1A current to the base-emitter. Now the problem is, how do we generate such a large base current?

Can we directly connect the microcontroller’s I/O pin to the base of the transistor? No, this won’t work. A microcontroller’s I/O pin cannot output that much current; generally, the I/O pin can only supply around 20mA. Even with a gain of 100, this would only provide 2A of current between the collector and emitter, which is far from the required 100A.

You might think, if one BJT isn’t enough, can we add another MOSFET in front of it? Since MOSFETs are voltage-controlled, it might work. When the microcontroller’s I/O pin is low, the MOSFET turns off, and a current greater than 1A flows through the BJT’s base-emitter, turning the BJT on and allowing a large current to flow between the collector and emitter. When the I/O pin is high, the MOSFET turns on, and the BJT’s base voltage becomes zero, turning the BJT off. For this type of circuit, as long as we apply a PWM signal to the gate of the MOSFET, the BJT will continuously switch.

Summary

BJTs are slow to switch due to the minority carrier storage effect, and they have a slow recovery time after deep saturation. On the other hand, MOSFETs, which rely on charge conduction, have very fast turn-off characteristics. In theory, with proper driving circuitry, MOSFETs can handle very high switching frequencies. Similarly, regular silicon diodes have slow reverse recovery times due to both electrons and holes participating in conduction. However, Schottky diodes, which only involve electrons in conduction, have very fast reverse recovery times. Sometimes, a Schottky diode is placed in parallel with the BJT’s base and collector to limit the saturation depth and significantly shorten the switching time. This is why the “S” in 74LS (Low Power Schottky) logic circuits stands for Schottky diodes.