Insulated Gate Bipolar Transistor (IGBT) is a composite power switching device that combines power MOSFET and bipolar power transistor. It not only has the advantages of fast switching speed of MOSFET, high input impedance, low driving power and simple driving circuit, but also the advantages of large capacity and high blocking voltage of high-power bipolar transistors. It is currently the most widely used in high-power UPS A kind of high frequency SPWM switching device.
1. Basic structure and working principle of IGBT
The basic structure of ICBT, device symbols, equivalent circuit and low-voltage zone output characteristic diagram are shown in Figure 1. Figure (a) is the basic structure of the IGBT, Figure (b) is the device symbol, Figure (c) is the equivalent circuit, and Figure (d) is the schematic diagram of the output characteristics of the low-voltage area. Compared with MOSFET in structure, IGBT adds a P+ area under the drain of MOSFET, and adds a PN junction J1. It can be seen from the equivalent circuit that it is a composite of MOSFET and bipolar power transistor. Therefore, it has the following characteristics.
①The input stage of IGBT is MOSFET. When driving voltage is applied between G and E, MOSFET enters the on state (or off state). Therefore, IGBT is a voltage control device
②In the IGBT, since the switching speed of the MOSFET is very fast, the switching speed of the IGBT depends on the switching speed of the equivalent transistor. In the IGBT, the N+ region thickness is optimized to suppress the injection of excess carriers, and the lifetime suppression mechanism is introduced to reduce the dissipation time of the stored carriers, thereby shortening the switching time of the equivalent transistor, which can increase The switching speed of ICBT makes it much faster than bipolar transistors.
③When a negative voltage is applied between the collector-emitter of the IGBT, since the P+N junction J1 is in a reverse bias state, it is impossible for current to flow between the collector and emitter. And because IGBT has one more J1 junction than MOSFET, IGBT has a higher withstand voltage than MOSFET.
Due to the existence of the P+ region of the IGBT, when the IGBT is in the on state, positive carriers will be injected from the P+ region and accumulate in the N region, so that the IGBT presents a low resistance state when it is turned on, so the current capacity of the IGBT is also It is larger than MOSFET.
④ When the IGBT is in the on state, the size of UCE can reflect the overcurrent situation. Therefore, the over-current situation can be identified by measuring UCE. Once UCE is higher than a certain value, it indicates that an over-current situation has occurred. At this time, the IGBT can be quickly turned off by controlling the gate voltage to quickly change to zero or negative voltage. Realize the overcurrent protection of IGBT.
It can be seen from the above analysis that the IGBT has the characteristics of high forward and reverse blocking voltage and large on-state current, which can be controlled by voltage to turn on or off. At the same time, due to the MOS gate, the power consumption of the control circuit is small, and the static power loss during turn-on and turn-off is also very small, and there is only a certain dynamic loss during the state transition. For this kind of dynamic loss, it can be minimized by ZVS or ZCS soft switching technology. Because ICBT has these characteristics, it is widely used as SPWM power switching device in UPS conversion circuit.
2. Use gate resistance to control IGBT switch
The setting of the switching characteristics of each IGBT is affected by the gate resistance RG. The input capacitance of the IGBT changes during the switching period and must be charged and discharged. The gate resistance determines the charging and discharging time by limiting the gate current IG during turn-on and turn-off, and the amplitude of the pulse, as shown in Figure 2. Due to the increase of the gate peak current, the turn-on and turn-off time Will be shortened, and the switching loss will also be reduced. Reducing the resistance of RG(on) and RG((off) will increase the peak gate current. When reducing the resistance of the gate resistance, what needs to be considered is the current generated when a large current is switched too quickly Time-varying characteristics di/dt. The presence of stray inductance in the circuit will produce a large voltage spike on the ICBT. This effect can be observed from the waveform diagram of the IGBT when it is turned off as shown in Figure 3. The shaded part shows the relative value of the turn-off loss. The instantaneous voltage spike on the collector-emitter voltage UCE may damage the ICBT, especially in the case of short-circuit turn-off operation. At this time, you can increase the value of the gate resistance To reduce the Ustray to eliminate the risk of IGBT damage due to overvoltage. However, rapid turn-on and turn-off will cause higher du/dt and di/dt, which will generate more electromagnetic interference (EMI) , Resulting in circuit failure.
The switching characteristics of the freewheeling diode that protects the IGBT is also affected by the gate resistance, and limits the minimum value of the gate impedance. This means that the on-switching speed of the IGBT can only be increased to a level compatible with the reverse recovery characteristics of the freewheeling diode used. The reduction of the gate resistance not only increases the 1GBT overvoltage stress, but also increases the overvoltage limit of the freewheeling diode due to the increase of di/dt in the IGBT module. Through the use of a special design and optimized soft recovery function The CAL (Controllable Axial Life) diode can make the reverse peak current small, thereby making the ICBT conduction current in the circuit small. The driver output stage of the gate drive circuit is a typical design, using two MOSFETs configured in the form of totem poles, as shown in Figure 4. The gates of the two MOS FETs in the figure are driven by the same signal. When the signal is high, NiiMOSFETi; when the signal is low, the P-channel MOSFET is turned on, resulting in a configuration of two transistor push-pull outputs. The output stage of the MOSFET can have one or two outputs. Depending on whether the output stage is one or two outputs, one or two gate resistances (on, off) can be realized for symmetrical or asymmetrical gate control. plan.
Generally, in applications, IGBT modules with a large rated current should be driven with a small gate resistance, and IGBT modules with a small rated current should be driven with a large gate resistance. In other words, the resistance values given in the ICBT data sheet must be optimized for each design. Generally, the optimal gate resistance value is generally between the value listed in the manual and twice its value. In addition, the value specified in the manual is the minimum value. Under the specified conditions, a current value twice the rated current can be safely shut off. In practical applications, due to the difference of test and application parameters, the gate resistance in the data sheet may not be the best value. It can be used as the starting point for optimization, and the best value can be finally optimized by the test method. It is necessary to keep the turn-off overvoltage of the IGBT within the specified range of the data sheet, especially in the case of a short circuit. The gate resistance determines the gate peak current IGM. Increasing the gate peak current will reduce the turn-on and turn-off time and switching loss. The maximum value of the gate peak current and the minimum value of the gate resistance are determined by the performance of the output pole of the driver.