Measurement techniques to characterize SiC, GaN, and Si power MOSFETs for transportation applications

Abstract

Wide Bandgap (WBG) semiconductors have the potential to cross the barriers that limit the widespread adoption of power electronics (PE) circuits in various energy conversion applications. The WBG devices improve power density and allow PE circuits to reach operating points and efficiency beyond what standard Silicon (Si) can offer. Currently, Silicon Carbide (SiC) and Gallium Nitride (GaN) are the most viable WBG semiconductor candidates to replace Si-based devices. For now, the state-of-the-art technologies still bring confusion when it comes to choosing between SiCs and GaNs to get the most benefits of them for power converter applications. To tilt the balance in the face of this debate, this thesis presents a comprehensive comparative computation and experimental analysis among Si, SiC, and GaN MOSFETs in energy conversion applications. A presentation of simulation equations based on intrinsic parameters of MOSFET’s equivalent mathematical model highlights the assumptions, limitations, and opportunities for improvement of the computational model. Then, simulation of multiple experimental conditions reveals favorable switching results towards SiCs while highlighting the marked advantages of this technology at high temperature. Although the simulation results are promising, the experimental validation remains necessary to make the concluding remarks. Analysis of experimental data extraction methodologies highlights the perturbation from parasitic inductances, parasitic capacitances, mutual coupling, and electromagnetic interference (EMI) on the switching process and measurements. Based on the detailed assessment of parasitic perturbations, this work proposes a voltage and current extraction methodology to remarkably minimize disturbances on measurements and the switching process. Finally, interpretation of experimental results through statistical analysis of relative comparison presents major trends from each technology while providing an interpretation of observations based on intrinsic parameters.

Date	15 Sept 2021
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Handy Fortin Blanchette (Supervisor) & Amin Ghazanfari (Co-supervisor)