Output details

Exploration of Power Device Reliability Using Compact Device Models and Fast Electrothermal Simulation

The work reported here describes new techniques that allow temperature cycling to be predicted over extended period (>30mins) without the requirement of excessive computational resources. It has given us the basis to carry out a number of collaborative projects with companies in the UK such as Jaguar Landrover, Prodrive (Pete Tibbles, ptibbles@prodrive.com) and Ricardo (John Reeve, john.reeve@ricardo.com). The predictive capability of these models is state of the art. We are continuing to develop this capability in the automotive applications area. Because of the accuracy and the computational speed of the methodology, it is in high demand by industry.