Development of high frequency (>1MHz) power converters

Details

High density power conversion is critical to a range of fields: from the electrification of vehicles, in small on-board chargers, DC-DC converters, and rapid DC chargers; in renewable energy, for solar and wind applications in the pursuit of net-zero; and in AI datacentres, where new 800V architectures require several stages of conversion down to 1V at power levels of 10s–100s of kW. Across these applications there is demand for smaller, lighter and higher power capable converters.

A recent key innovation in this field is the widespread availability of wide-bandgap semiconductor devices based on SiC and GaN. These devices provide a step-change in capability, enabling higher frequencies, higher voltages, and more efficient operation, all of which support smaller, high power density converters. However, realising this potential in practice requires technological advances across the whole converter, passive components, topology, and control, alongside semiconductor gains. In particular, passive components are increasingly a limiting factor, with losses scaling with frequency and eroding the efficiency gains WBG devices otherwise enable.

Despite this potential, converter design has not kept pace with the frequencies now possible with WBG devices. Conventional magnetic materials suffer increasing core and winding losses above 1 MHz, and much of magnetics design still relies on empirical rules-of-thumb and iterative prototyping rather than predictive, physics-based design tools. Topologies and control strategies suited to high-frequency, high-density operation are similarly underdeveloped. This project targets the critical boundary where WBG-enabled operating frequencies have outrun the passive components, topologies, and design methods needed to exploit them.

This PhD will explore high-frequency (>1 MHz) power converter design, with a focus on achieving improvements in power density. This includes developing novel magnetics, utilising finite element analysis and modern manufacturing techniques, and exploring the use of AI tools and state-of-the-art materials. Whilst magnetics are almost exclusively used in power conversion, this research will also explore the use of alternative technologies, such as piezoelectric devices, as a direct alternative to magnetic materials for high-frequency operation. Piezoelectric resonators are electrically equivalent to high-Q resonant circuits and can be used to create high-density converters with efficient high-frequency operation.

Depending on your background and research interests the project offers scope to explore topology development, characterisation of high-frequency passives, PCB layouts, control methodologies, and automated design tools. Crucially, this research will be grounded in real-world application, as you will design, build, and test experimental high-frequency converters to validate your theoretical models in the lab.