Dynamic Analytical Switching Loss Model of SiC MOSFET Considering Threshold Voltage Instability

Accurate modeling of switching loss is critical for silicon carbide
mosfet
s as well as power converters. However, previous “static” time-independent models did not consider the impact of gate oxide degradation on switching performance during long-term operation. This article proposes a “dynamic” time-dependent analytical model considering threshold voltage (
V
_TH
) instability caused by gate oxide degradation to predict switching loss. The influence of
V
_TH
instability on the turn-
on
and turn-
off

V
_TH
, as well as on switching loss during continuous operation is revealed first. Moreover, the problems suffered in the existing analytical model are investigated, and an improved switching loss model is presented. A measurement-based method to obtain the
V
_TH
instability parameters for modeling is provided. Furthermore, a buck converter is built and operated under different conditions. Comparisons of switching waveforms and switching losses between experiments and the proposed model are given to validate the model. The results indicate that the proposed analytical model can effectively evaluate the switching loss, with an error within 7% under different continuous operating conditions. Finally, the universality of the proposed model for devices with different structures is verified, and a predication application of the model in operation is demonstrated.