An Analytical Switching Loss Model for SiC MOSFET Considering Temperature-Dependent Reverse Recovery Over an Extremely Wide High-Temperature Range

Accurate switching loss prediction is crucial for studying the failure mechanisms of power modules at extremely high temperatures. However, the temperature range of existing loss models for SiC MOSFETs is below 175 degrees C, which cannot fulfill the requirements of high-temperature applications. This study proposes an analytical switching loss model for SiC MOSFETs over an extremely wide temperature range of 25 degrees C to 475 degrees C. First, the proposed device-level model improves accuracy by considering the temperature-dependent channel transfer and output characteristics, body diode forward characteristic, and focusing on the two main charge storage effects of the reverse recovery process. Importantly, new phenomena in device characteristics at high temperatures are considered, especially the sudden increase phenomenon in reverse recovery, thereby broadening the applicable temperature range of the model. In addition, the proposed circuit-level model is based on the commonly used four-lead Kelvin source connection case in conventional high-temperature packages and includes almost all parasitic inductances and capacitances. In this study, a low-inductance double-side embedded package structure with a temperature resistance of 550 degrees C is used to develop a high-temperature double-pulse test platform to verify the accuracy of the model. Experimental results show that the average estimation error of the turn-ON energy loss is less than 8% and the maximum estimation error of the total energy loss is less than 14%. The proposed model holds the potential to be a powerful tool for studying extremely high-temperature failure mechanisms of power modules and guides the design of high-temperature converters.