Basically, power semiconductors can be broadly divided into two categories: power discrete components and power integrated circuits (Power ICs). Among them, power discrete components include MOSFETs, diodes, and IGBTs, with MOSFETs and IGBTs being the most important.
Are IGBTs better than MOSFETs? I've completed quite a few power supply designs, including switching power supplies and H-bridge motor controllers. Back in the 1970s and 1980s, my designs were done using NPN bipolar power transistors. We rarely use PNP devices because they are generally more expensive and currently have lower ratings than their NPN counterparts (all other things being equal).
MOSFETs and IGBTs are primarily used to convert the chaotic voltage and frequency current generated by power generation equipment into current with specific electrical parameters through a series of conversions and modulations, thus supplying various terminal electronic devices and becoming one of the core components of electronic power conversion devices. In the global power semiconductor market, industrial control accounts for the highest proportion at 34%, followed by the automotive and communications sectors at 23% each, and consumer electronics at 20%.
Depending on their internal structure, MOSFETs can handle varying currents, generally up to kiloamperes (kA). However, MOSFETs do not have the same voltage tolerance as IGBTs. MOSFETs excel in high-frequency applications, operating from hundreds of kHz to tens of MHz in RF products. IGBTs, on the other hand, are near their optimal operating limit at around 100 kHz. Finally, MOSFETs are superior for high-speed switching, as IGBTs integrate BJTs, which have a longer charge storage time, hindering high-speed switching. Therefore, MOSFETs are suitable for portable rechargeable batteries and mobile devices, while IGBTs are ideal for high-voltage, high-power applications such as electric motors and automotive batteries.
We also avoided Darlington transistors. They have very high gain, but their saturation voltage is also very high, which significantly increases power consumption within the device.
Because the collector of the input transistor is connected to the collector of the output transistor, once the output transistor starts conducting, it draws drive current from the input transistor (or strips the voltage source). As a result, the saturation voltage of the composite device is approximately 1 volt. At high collector currents, the power dissipated in the device makes it very warm to the touch during operation.
From the 1980s to the present, the transistors of choice have typically been N-channel MOSFETs. Like bipolar devices, P-channel FETs have lower power ratings and are more expensive. With FETs, the extremely high input impedance makes driving the gate much easier. Gate-source capacitance somewhat offsets this advantage, especially at high switching frequencies.
I've never used insulated-gate bipolar transistors (IGBTs), partly because I've never fully understood them. I initially thought they were Darlington devices with N-channel FETs replacing the input bipolar transistors. This would produce devices with extremely high input impedance and high overall gain, but of course, still with high saturation voltage and correspondingly high power dissipation.
I've been seeing more press releases about IGBT devices lately, so I decided to take a closer look at what these devices really mean.
Aha! It actually uses an N-channel FET as the input device, but the bipolar device is a PNP device. Now it's much more efficient, and the device can have very high breakdown voltage capability. Just a few volts are needed to turn on the FET, and then you can force-turn on the PNP transistor. There's that parasitic NPN transistor; combined with the PNP, it makes the bipolar portion look like an SCR. In fact, early IGBT devices had a latch-up problem: sometimes, once they were turned on, they couldn't be turned off unless you cut the collector current (turned off the mains power). Modern devices have solved this problem.
By the way, you'll see different symbols for IGBT; this is the semi-common one:
Please note that the upper end is called the collector, but it is connected to the PNP emitter. This is just to simplify everyone's understanding of how it works, not what happens inside.
These devices are not solutions for all applications. Their forward voltage drop is lower than that of a typical MOSFET whenever you compare them to comparable high-voltage, high-current devices (ranging from kilovolts to hundreds of amperes). At more moderate current levels, conventional FETs are better. If you require high switching speeds (reaching hundreds of kHz or MHz) for PWM rates, then use a traditional FET again.
