TBC

This project aims to create a step-change in the electrical, thermal, and magnetic performance of power dense power electronics by developing new fabrication techniques for power electronics circuitry.
The goal being to increase the power density and reliability of the circuits. This project will be run in collaboration with Ricardo UK, an engineering consultancy specialising in the automotive and electrical energy sectors. Both these sectors are reliant on power dense electronics with an added emphasis on the emerging electric vehicle and aerospace market.

The use of Silicon Carbide (SiC) and Gallium Nitride (GaN) semiconductor devices in power electronics has encouraged a shift towards increased power density of power electronics module designs. Both semiconducting materials have superior thermal stability properties allowing higher operating temperatures. Currently circuit designers typically gravitate towards printed circuit boards or point to point connection methods, such as cabling or bus bars to build power electronic circuits. The simplicity of the design process and low costs of both design methods make them preferable for engineers. However, for the application of dense power electronics both have significant disadvantages. Given the glass transition temperature of the FR4 (the substrate used for printed circuit boards) is approximately 110 C the thermal design of a PCB is heavily constrained, with the stability of the soldering material at high temperatures being another issue. The current alternative to a PCB is point to point methods, involving wiring or bus bars, using these it is possible to have conductors with a far larger cross-sectional areas reducing the effects of Joule heating. The low connection density of bus bars and wiring proves to be the major issue with point to point connections. This leads to far lower design flexibility, with other factors such ease of assembly and reliability also being concerns. This leads to circuit designers using PCB's wherever possible despite the significant drawbacks.

The purpose of this project would be to develop novel circuit fabrication methods that combine the high connection density and multi-layer construction of printed circuit boards with the high current carrying capacity of conductors with large cross-sectional area. A number of the different techniques would be explored, utilising recent advances in additive manufacturing techniques. Because of these new techniques a shift in the design philosophy of power electronics will be possible. Traditionally the electrical design was constrained to a two-dimensional planar printed circuit board. By designing the circuit layout in three-dimensional space, additional gains in the thermal and magnetic properties of the power converter may be possible which were not previously when using a traditional 2D design.

The first stage of this project would be to determine the optimal layout of power electronics circuit in three-dimensional space once the limitations of standard planar production methods are removed. A thermal analysis using finite element or finite volume analysis would be a sensible approach in order to experiment with alternative configurations of material and components. Once the key design variables are known a second stage could be to apply topology optimisation, to the thermal management problem. In this project I would look to apply topology optimisation to circuit layouts, constraining the optimisation with constraints that define the electrical connections and the component positions, with scope to increase the complexity of the optimisation to include multiple materials and the addition of magnetic components which are automatically generated in the circuit layout. The effectiveness of this method may be limited by the computational complexity of the resulting optimisation algorithm.

This project falls within the EPSRC Engineering, Electrical Motors and Drive research area