High-Density Active Silicon Carbide Power Electronics: Enabling Responsive Power Conversion

In pursuit of Carbon net-zero, it is imperative to develop technologies that enhance the efficiency and reliability of energy conversion, e.g. in drivetrain and rapid chargers of electric vehicles (EVs). To put this into context, the larger battery size (i.e. 350 kWh at 800 V & 440 A for higher consumption) and long-range driving nature of heavy-duty EVs mandate ubiquitous access to extremely fast chargers at 350 kW for financially justifiable charging delays. These are proposed to directly connect to 11 kV feeders by high-frequency solid-state-transformers (SST), needing energy-dense fast power modules. Literature indicates that the emergence of wide-bandgap semiconductor devices, especially Silicon Carbide devices, enables us to deliver ultra-efficient reliable converters that deliver the next leap.

Wide-bandgap power electronics is, however, currently being slowed down due to issues such as high dV/dt, common-mode interference and degradations. This means the full potential of wide-bandgap devices is still far from being obtained. The IEEE International Technology Roadmap for Wide-Bandgap Power Semiconductors (ITRW) has indicated that to unlock this potential, these limitations must be broken-through by 2028. As the UK is leading toward automotive electrification with a ban on the sale of new petrol & diesel engines by 2030, the UK needs to develop this technology locally, and earlier than this, to remain a global competitor in 'driving the electric revolution'.

Research on SiC devices has shown that they are prone to progressive degradations, with a 'memory' effect that leads to a drift of electrothermal parameters away from the datasheet values. This can lead to failures in long-term operations. Nevertheless, it is demonstrated that under certain conditions the devices can recover to close to the initial state, if the devices are subjected to specific electrical and thermal conditions. This proposal, in a nutshell, aims to take advantage of these findings to explore ways of controlling and reversing degradation in devices using non-contact sensors which feed information to smart, active gate drivers, which, in turn, control the recovery of the power devices.

To this end, this New Investigator Award project aims to make the power electronic core of these power converters responsive to operating conditions and functional degradations. This will be achieved by closing the loop between detection of change in SiC devices and how devices are controlled via their gates. This would permit SiC devices to be operated safely at higher switching speeds and thus efficiencies, than current datasheet limits allow. This is because datasheet nominal values are conservative in order to take every situation into account, whereas new situational awareness will allow these limits to be safely exceeded when appropriate. This is so important, particularly in the case of SiC power conversion, because whilst it is successfully taking over from silicon, it is also known that the potential performance of SiC is over an order higher than today's systems. Being able to safely break through these nominal limitations will reduce converter volume in cars and aircraft 2x or more, and bring a similar reduction in power loss in wind and solar power generation. Perhaps most importantly, it will reduce operational risk, by changing to safer driving modes as devices age or overheat. For example, this will reduce the cost of offshore wind power generation by generating more power at a lower risk of damage, and allow maintenance to be pre-empted. In the future, responsive power conversion with awareness of operating conditions and degradation could allow electric vehicles to detect the onset of drive failure, and activate a safe mode to get people home.