The transition to electrified transport and energy systems is placing new demands on power semiconductor devices. Electric vehicles, charging infrastructure, and renewable energy systems increasingly rely on power electronics that can operate at higher voltages, higher temperatures, and greater power densities than traditional silicon devices. To meet these requirements, manufacturers are turning to wide-bandgap materials such as silicon carbide and gallium nitride.
While some of these high-voltage devices, such as silicon carbide metal-oxide-semiconductor field-effect transistors (SiC MOSFETs, a type of electronic switch that controls the flow of electricity inside almost all modern electronic devices), are already entering commercial use, the methods used to assess their reliability have not kept pace with their operating conditions. Many existing reliability tests were developed for lower-voltage silicon components and typically operate at a few hundred volts. At much higher voltages, new challenges emerge, including electrical breakdown in air, thermal effects, safety constraints, and sensitivity to how voltage is applied during testing.
Without agreed test protocols and supporting measurement data, manufacturers and end users struggle to assess device lifetime and fitness for purpose. This makes it harder to compare devices from different suppliers, slows qualification, and increases the cost and risk of deploying new technologies in safety-critical applications such as electric vehicles, where reliability is fundamental to performance, safety, and consumer trust.
As the UK’s National Metrology Institute, the National Physical Laboratory (NPL) was well positioned to address this gap by developing robust, traceable test capability that could inform future standards. Working in partnership with Swansea University, and with support from industry partners Vishay, a semiconductor manufacturer designing electronic components, and GEN3, an engineering company developing electronic test and measurement systems, NPL set out to understand how high-voltage reliability testing should be carried out on SiC MOSFETs under realistic operating conditions.
Rather than relying on a single test configuration, the project deliberately established three independent high-voltage test systems capable of operating safely at up to 1,500 V and temperatures approaching 200 °C. This triple test approach made it possible to explore how differences in test methodology influence device behaviour, an essential step in moving from laboratory testing to standardised practice.
Developing these capabilities required more than incremental upgrades to existing equipment. The test systems were redesigned to address insulation, spacing, materials selection, and safety systems suitable for sustained high-voltage operation. Two of the systems were developed at NPL, drawing on its long-standing expertise in high-voltage and high-temperature testing, while Swansea University contributed a third system based on its experience in component-level reliability testing. Industrial partners provided both instrumentation and representative devices, ensuring the work reflected real manufacturing and deployment scenarios.
The project successfully established practical, safe, and repeatable test capability for SiC MOSFETs operating at voltages and temperatures beyond the reach of most commercial systems. By testing identical devices across all three platforms, the team generated comparative data that showed how different test conditions influence observed reliability and failure behaviour.
This enabled long-term performance and lifetime characteristics to be assessed in a controlled and meaningful way, providing earlier insight into how devices are likely to behave in service. Just as importantly, the work identified which aspects of the test environment most strongly affect results, highlighting the parameters that must be defined if measurements are to be consistent and comparable between laboratories.
Together, these findings provide the evidence base needed to develop robust, standardised test protocols for high-voltage wide-bandgap devices. This will give the UK greater clout in the international standardisation committees that are defining market requirements for these products.
Reliable high-voltage power devices are essential to the performance, safety, and lifetime of electrified systems. By enabling repeatable and defensible reliability testing, this work reduces uncertainty for manufacturers and system integrators, supporting faster qualification and more informed design choices.
For industry, the availability of credible test methods improves transparency and comparability between devices, helping to balance performance, cost, and lifetime requirements. For end users, it underpins confidence that new technologies will perform as expected over their intended service life.
More broadly, the work strengthens the UK’s position in an area of growing strategic importance. While large-scale manufacturing may take place elsewhere, the ability to define how high-voltage devices are tested and trusted gives the UK influence over how emerging power electronics markets develop. By investing in metrology and standards capability at this stage, NPL is helping to ensure that the UK remains a trusted place to develop, validate, and deploy advanced semiconductor technologies, supporting both the transition to net zero and the competitiveness of UK innovation.
This project tackles a problem the semiconductor sector has been grappling with for some time. Wide-bandgap devices behave very differently to silicon, and being able to recreate years of real-world stress in the lab is a major step forward for UK capability. At Swansea, as part of the UK’s only Compound Semiconductor Cluster based in South Wales, we are proud to work alongside NPL and industry to build testing infrastructure that manufacturers can actually use. It strengthens the whole ecosystem and helps ensure these technologies reach the market faster and with greater confidence.
Professor Mike Jennings - Royal Academy of Engineering Chair, Swansea University