Abstract: High-power electric drive devices typically employ insulated-gate bipolar transistors (IGBTs) as power semiconductor switching devices. During system development, comprehensive and in-depth testing of the IGBTs and their drive circuits is necessary to ensure system reliability. This paper describes the main testing contents and methods for IGBTs and their drive circuits in high-power electric drive devices. Verified on various high-power electric drive devices, the methods demonstrate strong practicality.
1 Introduction
In electrical drive systems with power ratings reaching hundreds of kilowatts or even megawatts (such as high-power frequency converters and wind power converters), insulated-gate bipolar transistors (IGBTs) are typically selected as power semiconductor switching devices. The selection of IGBTs and the design of their drive circuits have a significant impact on the operational reliability of the system.
In the development of electric drive devices, comprehensive and in-depth testing of IGBTs and their drive circuits is necessary to ensure the reliability of the system. This paper describes the main test contents and methods for IGBTs and their drive circuits in high-power electric drive devices. These test contents and methods have been verified through various high-power electric drive devices and have strong practicality.
2. Basic Test Units and Driver Modes
To ensure consistency between test conditions and results and actual applications, power modules actually used in electrical drive systems are generally selected as test units. This article uses an IGBT half-bridge module as an example.
In the actual operation of electrical drive devices, IGBT modules are usually in a high-frequency switching state. Therefore, the testing of the test unit should also fully reflect this characteristic, such as using the same switching frequency.
Commonly used test modes include the dual-pulse drive test method and the multi-pulse drive test method. The dual-pulse drive test method, as shown in Figure 1, requires testing the lower IGBT and the upper IGBT in the test unit separately. Here, attention should be paid to the wiring method of the load reactance and the blocking of the drive of the non-test IGBT.
The multi-pulse drive test method can use multiple pulse series with fixed frequency and duty cycle to drive the test. The test results are usually shown in Figure 2.
3. Specific test content and methods
3.1 Current spikes at IGBT turn-on
Typically, when an IGBT is turned on, a reverse recovery current flows through its freewheeling diode, causing a current spike in the IGBT. In the design, it is essential to ensure that the maximum turn-on current of the IGBT under normal operating conditions remains within its reverse bias safe operating area (RBSOA).
In this test, the DC bus voltage can be selected from the rated operating voltage of the DC bus, because the bus voltage has little effect on the reverse recovery current of the diode.
3.2 VCE spike at IGBT turn-off time
Typically, when an IGBT is turned off, its VCE will generate a spike due to the parasitic inductance of the circuit. In the design, it is necessary to ensure that when the IGBT turns off its maximum operating current at the highest possible bus operating voltage, the peak value of its VCE remains within the IGBT's RBSOA.
In Figure 4, the blue waveform represents the VCE of the IGBT. During the test, the peak value that appears during the turn-off process needs to be recorded.
3.3 Enable/Disable Delay Test
A certain interval (usually called dead time) is typically inserted between the turn-off of one IGBT transistor and the turn-on of the other to prevent shoot-through. The actual dead time presented on the IGBT module is affected by the drive signal transmission link. If the difference between the turn-on and turn-off delays is large, it will affect the actual dead time and operational reliability of the IGBT. As shown in Figure 5, the turn-on delay and turn-off delay are measured by comparing the drive signal issued by the controller and the VGE drive signal at the IGBT port. It is generally necessary to ensure that the turn-on delay and turn-off delay are consistent; for example, the difference in delay time should not exceed 10% of the dead time.
3.4 Straight-through protection test for upper and lower pipes
The IGBT shoot-through protection test is an important test for bridge circuits to evaluate the system's immunity to driving interference.
IGBT shoot-through protection testing typically involves two types of tests. One test method, as shown in Figure 6(a), involves keeping the upper IGBT always on and inputting a 12µs drive pulse to the lower IGBT. Under normal conditions, the protection circuit should provide protection within 10µs; even if protection fails, 12µs usually won't damage the IGBT under test. The other test method, as shown in Figure 6(b), involves simultaneously inputting a 12µs drive pulse to both the upper and lower IGBTs to check the effectiveness of the protection circuit. In both test methods, the drive circuit should provide effective protection within 10µs.
Figure 7 shows the test waveforms when the driving method shown in Figure 6(a) is used. It can be seen that the protection circuit turns off the IGBT within 10µs, and there is a soft turn-off time of about 1µs. During the soft turn-off process, the upper transistor VGE drops from 15V to 12.5V within 1µs.
3.5 Overcurrent Protection Test
Although the maximum operating current of an electrical drive system is significantly lower than the overcurrent protection point during actual use, preventing the overcurrent protection from being triggered, the effectiveness of the overcurrent protection function remains crucial for the safe operation of the electrical drive system. The overcurrent protection function can be tested using a multi-pulse drive test method. After the IGBT is turned on multiple times consecutively, the accumulated current may reach the overcurrent protection point, thus triggering the overcurrent protection function.
Figure 8 shows the relevant waveforms under multi-pulse drive test, with each drive pulse having a width of 20µs. The width of the last drive pulse is significantly narrower than the previous ones because this pulse triggers the IGBT's overcurrent protection after the IGBT is turned on, and the drive circuit automatically turns off the IGBT.
3.6 Power-on/Power-off Test of Drive Circuit
The power-on test of the drive circuit is mainly to confirm that the IGBT drive signal will not be erroneously triggered at the moment of power-on. The power-off test is mainly to confirm that the IGBT drive signal must enter a blocked state when the auxiliary power supply suddenly fails under normal operating conditions.
As shown in Figure 9, (a) shows the waveform of the drive circuit output signal VGE at the instant of system power-on and power-off. There is no false triggering signal on the IGBT drive, and a negative voltage can be established after power-on. (b) simulates the sudden disconnection of the power supply to the drive circuit under the premise that the IGBT drive port has been receiving drive pulse signals. It can be seen that the positive voltage in the drive pulse gradually decreases, accompanied by a negative voltage. Until the positive voltage is lower than a certain level, the drive pulse is blocked, and the negative voltage can last for a period of time to ensure the safe shutdown of the system.
3.7 Active clamping branch test
The active clamping circuit shown in Figure 10 is mainly used to clamp the VCE spike during the switching process of the IGBT. It is typically triggered when a large current is turned off. During testing, the main focus should be on whether the VCE spike value is within a safe range, but the losses of the TVS diode must also be considered. By measuring the voltage across resistor RF1 and combining this with its resistance value, the current flowing through the TVS diode can be calculated, allowing for further calculation of its losses. It is essential to assess whether these losses meet the TVS diode's temperature rise requirements within one output current cycle.
3.8 Test the temperature rise of key components
Among all the single-board circuits, the driver circuit is closest to the IGBT module and therefore withstands very high ambient temperatures. If the system's permissible operating ambient temperature reaches 55°C, the ambient temperature around the driver circuit can typically reach 70°C. Therefore, the operating temperature and temperature rise of related components need to be tested.
During testing, the IGBT module can be placed in a temperature chamber, and the temperature rise experiment can be conducted by driving the IGBT at the actual operating switching frequency. Typically, the temperature of key chips and components such as optocouplers and drive transformers needs to be closely monitored, and thermocouple contact point measurements can be used. For other components such as resistors and capacitors, loss theory calculations and thermal imaging scans can be used to confirm whether they are operating within a reasonable temperature range.
4. Conclusion
This article describes the main test contents and methods for IGBTs and their drive circuits in high-power electric drive devices. These test contents and methods can comprehensively and thoroughly simulate and evaluate the working performance of IGBT modules and the effectiveness of related protection functions under various operating conditions, which is of great significance for improving the operational reliability of electric drive devices. These test contents and methods have been developed and applied in various high-power electric drive devices (including the HD2000 series engineering frequency converters) by Hopewind Electric and have strong practicality.
