The internal structure of MOSFETs and IGBTs differs significantly, which directly influences their respective application fields. First, due to its structure, a MOSFET can handle large currents—up to kiloamperes—but typically has lower voltage blocking capability compared to an IGBT. Second, IGBTs are capable of handling high power levels with both high current and voltage, making them suitable for applications where the switching frequency is not extremely high. While IGBTs can switch at around 100 kHz, MOSFETs can operate at hundreds of kHz, even up to MHz or higher, which makes them ideal for RF and high-frequency applications. Third, based on their characteristics, MOSFETs are commonly used in high-frequency power supplies such as switching power supplies, ballasts, high-frequency induction heating, and communication power supplies. On the other hand, IGBTs are primarily used in welding machines, inverters, and SMPS systems, where high power handling is required.
The performance of these power supplies largely depends on the choice of power semiconductor devices, such as switching transistors and rectifiers. Although there's no universal solution for choosing between IGBTs and MOSFETs, comparing their performance in specific SMPS applications can help identify the key parameters. This article will discuss several important parameters, including switching losses in hard-switching and soft-switching ZVS (Zero-Voltage Switching) topologies, as well as conduction loss, turn-on loss, and turn-off loss associated with circuit and device characteristics.
In addition, the article will illustrate how the recovery characteristics of the diode are a primary factor in determining the conduction switching loss of MOSFETs or IGBTs, and how this affects hard-switching topologies. While the conduction characteristics of IGBTs and power MOSFETs are similar, IGBTs have a longer voltage drop time due to the minority carrier storage effect in the PNP BJT structure. This delay results in a quasi-saturation effect, preventing the collector-emitter voltage from dropping immediately to its VCE(sat) value.
The Eon energy consumption listed in IGBT product specifications represents the integral of the product of Icollector and VCE over each cycle, measured in joules. It includes losses related to the saturation state. This energy is further divided into two components: Eon1, which does not include losses from the hard-switched diode recovery, and Eon2, which includes conduction energy from the diode recovery. The Eon2 test circuit is shown in Figure 2, where the IGBT measures Eon by switching between two pulses.
In hard-switching scenarios, the gate drive voltage, impedance, and the recovery characteristics of the rectifier diode determine the Eon switching loss. In conventional CCM boost PFC circuits, the boost diode’s recovery characteristics are critical for controlling Eon energy. Choosing a diode with minimal Trr and QRR, along with soft recovery characteristics, helps reduce electrical noise and voltage spikes. Some high-speed diodes may cause high voltage spikes due to fast current drop rates during recovery, leading to EMI and excessive reverse voltages.
In hard-switching circuits like full-bridge and half-bridge topologies, IGBTs are often combined with fast recovery transistors or MOSFET body diodes. The recovery characteristics of these diodes determine the Eon loss, making it essential to choose a MOSFET with fast body diode recovery. However, the parasitic diode recovery of MOSFETs is generally slower than that of discrete diodes. Therefore, body diodes often limit the operating frequency in hard-switching MOSFET applications.
IGBTs are usually packaged with matching diodes, such as slow ultrafast diodes with low forward conduction losses for motor drives, or soft-recovery ultrafast diodes for high-frequency SMPS applications. Designers can also control Eon loss by adjusting the gate drive source impedance. Lowering the drive impedance increases the conduction di/dt and reduces Eon loss, but higher di/dt leads to increased voltage spikes and EMI, requiring a balance.
To select the correct gate drive impedance, internal circuit testing and verification are necessary. The approximate value can be determined from the MOSFET transfer curve. For example, when the FET current rises to 10A, the gate voltage must increase from 5.2V to 6.7V, resulting in an average GFS of 6.7mΩ.
Similarly, IGBTs can undergo similar gate drive resistance calculations. Using the IGBT’s conversion characteristic curve, VGE(avg) and GFS can be determined, and CIES under VGE(avg) is used instead of Ciss. The calculated IGBT turn-on gate drive impedance is 100Ω, which is higher than the 37Ω for MOSFETs, indicating higher GFS and lower CIES for IGBTs.
When comparing conduction losses between 600V-rated devices, IGBTs typically have lower conduction losses than MOSFETs of the same chip size. This comparison should be made at the collector and drain current densities and at worst-case junction temperatures. For instance, both the FGP20N6S2 IGBT and the FCP11N60 SuperFET have an RθJC of 1°C/W. At a junction temperature of 125°C, the MOSFET has higher conduction loss after a DC current of 2.92A.
However, DC conduction losses are not always suitable for most applications. A comparison of conduction loss in CCM boost PFC circuits at 125°C junction temperature shows that the MOSFET-IGBT intersection point is 2.65A RMS. For PFC circuits, the MOSFET has higher conduction loss when the AC input current exceeds 2.65A RMS.
An IEEE article discusses how to incorporate RDS(on) into conduction loss calculations for high-frequency three-phase PWM inverters. RDS(on) changes have little effect on most SMPS topologies, but in high-pulse-current topologies, the characteristics shown in the figures should be considered.
Calculating conduction losses for IGBTs in similar PFC circuits is more complex due to varying IC per switching cycle. IGBTs cannot be represented as an impedance, so they are modeled as an impedance RFCE in series with a fixed VFCE voltage. Conduction loss is then calculated as the product of average collector current and VFCE, plus the square of the RMS collector current multiplied by RFCE.
In practical applications, the conduction loss of IGBTs is small, but most 600V IGBTs are PT (Punch Through) type devices, which have NTC characteristics and cannot be easily paralleled. MOSFETs, on the other hand, have PTC characteristics, providing better current shunting.
Turn-off loss is another critical parameter. Due to the tail current of IGBTs, their turn-off loss is much higher than that of MOSFETs. The Eoff energy dissipation of a MOSFET is a function of its Miller capacitance Crss, gate drive speed, gate drive turn-off source impedance, and parasitic inductance in the source path. Parasitic inductance Lx limits the current drop rate, increasing turn-off loss.
In summary, there is no one-size-fits-all solution when choosing power switching devices. Circuit topology, operating frequency, ambient temperature, and physical size all play a role in selecting the best option. In ZVS and ZCS applications with minimal Eon loss, MOSFETs can operate at higher frequencies due to faster switching speeds and lower turn-off losses. For hard-switching applications, the recovery characteristics of MOSFET parasitic diodes can be a disadvantage. In contrast, IGBTs come with matched diodes that offer excellent soft recovery, making them suitable for high-speed SMPS devices.
In conclusion, there is no essential difference between MOSFETs and IGBTs. The question of whether one is better than the other is often misleading. The decision to use one over the other depends on the specific application, and a dialectical approach is needed to evaluate the trade-offs.
MINI FAKRA Automotive Cable Assembly
Dongguan Zhuoyuexin Automotive Electronics Co.,Ltd , https://www.zyx-fakra.com