When it comes to high-power switching devices (tens of amps at hundreds of volts) its hard to go past the IGBT. As its name suggests, the Insulated-Gate Bipolar Transistor combines the low on-state voltage of bipolar transistors with the straight-forward drive characteristics of the power MOSFET. The schematic symbol for the IGBT (Figure 1) illustrates this nicely.
The schematic symbol for an IGBT gives us a pretty good idea of how this device behaves. It has a MOSFET-like insulated gate and so is easy to drive and a BJT-like low saturation voltage when on.
Internally, IGBTs are 4-layer (PNPN) devices with a structure similar to a power MOSFET but with an additional P+ layer at the collector as shown in Figure 2. When no gate voltage is applied the PN junction between the buffer/drift layer and the body layer is reverse biased preventing current flowing between collector and emitter.
Internally, the IGBT is constructed like a MOSFET but with an extra P+ layer next to the collector terminal. When the gate voltage is positive, electrons are attracted into the body layer creating a conductive channel between the emitter and drift/buffer region. The device can then conduct current through this channel and the PN junction adjacent to the collector.
Note that unlike a MOSFET the IGBT cannot conduct in reverse. This is because the PN junction between the injection layer and buffer/drift layer will be reverse biased. Many IGBTs therefore include a separate antiparallel diode within the package.
A positive voltage applied between the gate and emitter causes electrons to be attracted towards the gate within the body region, forming a conductive channel across the body region, just as for a MOSFET. Current can therefore flow between the collector and emitter via the PN junction between the injection layer and the conductive channel Thus, when conducting, the IGBT looks a bit like a PN junction and channel resistance in series.
You might think from this explanation that the on-state forward voltage of an IGBT would be higher than that of an equivalent MOSFET. While it is true that the forward voltage of an IGBT is always at least one diode drop, an IGBT can have a much lower on-state voltage than an equivalent MOSFET at the same temperature. Unlike a MOSFET which is majority-carrier device (only electrons flow in the case of an N-channel device), both electrons and holes flow in the channel the channel and the on-resistance is significantly reduced (modulated) in the IGBT.
This low on-voltage (and therefore power loss) for high-voltage devices is the IGBT’s big advantage over the MOSFET – but it comes with a downside. The IGBT is slower to turn off than a MOSFET since it takes a finite time for the electrons and holes in the channel to recombine once the electric field induced by the gate disappears.
— ADVERTISMENT—
—Advertise Here—
So, when should we use IGBTs and when should we use MOSFETs? For low voltage devices (say under 300 volts), a MOSFET will almost certainly have lower conduction losses. Above this, assuming you don’t need to switch at frequencies above about 50-100 kHz, IGBTs will likely be in contention. As your voltage requirements increase, IGBTs become more and more attractive.
The simplified equivalent circuit of an IGBT is a Sziklai pair comprising a MOSFET and a PNP transistor. The resistance represents the channel conductivity. Note that this is much lower than for an equivalent MOSFET since both electrons and holes are mobile in the IGBT.
Additionally, the on-state losses and switching speed of IGBTs are almost unaffected by temperature unlike MOSFETs where almost every parameter gets worse as temperature rises. IGBTs are relatively easy to use in parallel, especially if all devices are on the same heatsink as they will naturally tend to share the load current evenly. IGBTs are also very robust devices with many models capable of withstanding short circuit conditions in high voltage converters for 10µs or more.
Table 1 shows the comparison of a couple of similar devices rated at 1200V from the same manufacturer. The forward voltage drop at 10A of the MOSFET is almost 3 times higher than that for the IGBT with an unrealistically low junction temperature of 25°C. At a more realistic 125 °C the MOSFET forward drop is more than 6 times higher. While the turn-on times are roughly comparable, the MOSFET wins in terms of turn-off time as we would expect. The gate charge figures suggest that the IGBT is also a little easier to drive Note that the IGBT is significantly cheaper.
This table compares selected characteristics of an IGBT with a similarly specified MOSFET from the same supplier. The IGBT clearly has superior forward voltage performance at the cost of slower turn-off. The IGBT is also considerably cheaper than the comparable MOSFET.
IGBTs are available with voltage ratings up to 4,500V and 1,200A or more and in a myriad of multi-device modules designed for power converters. Many of these modules include all the necessary gate drivers and protection circuits making converter design that much easier. Unless you need very high frequency switching, choosing an IGBT for high voltage applications is a no-brainer.
References
“What Is an IGBT? | Toshiba Electronic Devices & Storage Corporation | Asia-English.” Accessed June 20, 2023. https://toshiba.semicon-storage.com/ap-en/semiconductor/knowledge/faq/mosfet_igbt/igbt-001.html.
“IGBT Tutorial: Part 1 – Selection – EE Times.” Accessed June 20, 2023. https://www.eetimes.com/igbt-tutorial-part-1-selection/.
“IXA12IF1200HB – Littelfuse.” Accessed June 22, 2023. https://www.littelfuse.com/products/power-semiconductors/discrete-igbts/xpt/planar/ixa12if1200hb.aspx?desktop=true.
“IXFX20N120P – Littelfuse.” Accessed June 22, 2023. https://www.littelfuse.com/products/power-semiconductors/discrete-mosfets/n-channel-hiperfets/polar_hiperfet/ixfx20n120p.aspx.
Sponsor this Article
Andrew Levido (andrew.levido@gmail.com) earned a bachelor’s degree in Electrical Engineering in Sydney, Australia, in 1986. He worked for several years in R&D for power electronics and telecommunication companies before moving into management roles. Andrew has maintained a hands-on interest in electronics, particularly embedded systems, power electronics, and control theory in his free time. Over the years he has written a number of articles for various electronics publications and occasionally provides consulting services as time allows.
