Test and measurement challenges with WBG devices

The use of wide bandgap (WBG) semiconductors like SiC and GaN to improve data center, renewable energy, and electric vehicle (EV) powertrain efficiencies and speed up charging times. This requires designers to use new validation testing approaches to better understand device and system performance.

This article begins by reviewing standard double-pulse testing (DPT) of WBGs. It then examines dynamic gate stress (DGS) testing to simulate application conditions and concludes by considering how to test high-power WBG-based converters efficiently.

DPT is used to measure device switching losses and evaluate energy losses during turn-on, turn-off, and reverse recovery performance. It is implemented using two (usually identical) devices.

The inductor (L) simulates circuit conditions expected in the converter design. A power supply is needed as a voltage source for the test. An arbitrary function generator (AFG) produces the triggering pulses for the device gates, and an oscilloscope captures and measures the results (Figure 1).

**Figure 1.** DPT testing setup. (Image: Tektronix)

IEC 60747-9 describes the test setup and measurement results for DPT. WGB DPT testing software suites that comply with IEC 60747-9 are available and can speed up the process.

Some considerations include:

Turn-on and turn-off performance is measured at the falling edge of the first pulse and the rising edge of the second pulse. A proper layout to minimize parasitics improves accuracy and repeatability.
Reverse recovery current is measured during the second pulse’s turn-off and is used to calculate energy losses, which are an important factor in converter efficiency.
For WBG applications, it’s often necessary to use the actual waveforms expected in the application instead of those defined in IEC 60747-9 to get the best results from DPT.

Simulating application conditions

DPT can be “fine-tuned” and improved using anticipated waveforms in the application circuit. However, WBGs are subject to failure mechanisms that traditional testing cannot identify. That’s where DGS comes in.

DGS was originally developed for the automotive industry and is defined in the European Center for Power Electronics (ECPE) Guideline AQG 324, “Qualification of Power Modules for Use in Power Electronics Converter Units in Motor Vehicles.” It defines a procedure for power module characterization and environmental and lifetime testing.

DGS is implemented with automated equipment and uses fast voltage shifts to simulate fault mechanisms at the gate terminals. The amplitude of the square wave signal is based on the maximum value in the device specifications.

The dV/dt rise time must be at least 50 kHz and have a minimum duty cycle of 20%. Guideline AQG 324 requires the stress duration to be at least 10¹¹ cycles. Therefore, higher-frequency testing can reduce the time needed to implement the test.

High-efficiency testing

Optimal testing of high-power WBG-based converters involves using the fewest possible resources, including minimizing the cost and energy consumption of testing. For example, it’s been estimated that the energy consumption needed to test a 100-kW power converter under full output voltage and current conditions for an hour could power a typical household for 3 days.

In response to the potential power consumption from testing high-power converters, a new methodology for measuring WBG device switching characteristics and assessing the thermal management systems in the converters has been proposed. The methodology can test a converter’s full voltage, current, and thermal characteristics and estimate its efficiency while minimizing energy consumption, load, and other test components.

The methodology was used on a 75 kW back-to-back (BTB) converter. A BTB converter connects different voltage and frequency networks, such as 480 and 280 Vac and 60 or 50 Hz, allowing equipment from different countries to be used. The BTB platform was implemented with a six-phase SiC IGBT power module, where three legs form the rectifier, and three form the inverter (Figure 2).

**Figure 2.** A 75-kW BTB converter basic schematic (top) and hardware (bottom). (Image: 2023 IEEE Energy Conversion Congress and Exposition)

Testing begins with a DPT to characterize the SiC device switching performance. Next, a multi-cycle test (MCT) is implemented with the converter connected to a resistive load, which runs at full voltage and current to estimate the efficiency. The operating temperature is close to room temperature, providing an upper bound for efficiency. In an actual installation, efficiency will drop at higher operating temperatures.

Following the MCT, the converter’s thermal resistance and thermal capacitance are measured to evaluate the thermal management system’s performance before a high-power continuous test is implemented. The high-power continuous circulating test is used to fine-tune the efficiency estimates and accelerate device degradation, so that the converter’s lifetime can be estimated.

Summary

Testing WBG-based high-power converters requires blending existing testing methodologies like DPT with new approaches like DGS. In addition, new methodologies are being developed to reduce energy consumption and the cost of testing.

References

Application-Oriented Testing Of SiC Power Semiconductors, Semiconductor Engineering
How to Test Wide-Bandgap Semiconductor Power Modules, Keysight
Identify New Failure Effects with DGS Tests, NI
Testing Methodology for Wide Bandgap High Power Converter with Limited Lab Resources, 2023 IEEE Energy Conversion Congress and Exposition
Validating Wide Bandgap Semiconductor Devices for Power Conversion Systems, Tektronix