IEEE Workshop on Wide Bandgap Power Devices and Applications in Asia (WiPDA Asia 2018)

285 / 2017-12-07 10:02:06

A Model Parameter Optimization Method of SiC Power MOSFET

Plenary track > Device characterization and modeling

Summary
In order to accurately simulate the switching characteristics of Silicon Carbide (SiC) power metal-oxide-semiconductor field-effect transistors (MOSFET), the model parameters needs to be optimized after being extracted from the data sheet. To improve the transient precision of behavior model for SiC power MOSFET in PSpice, in this study, an optimization method for the model parameter is proposed. By analyzing the influence of the parameters on the switching transition characteristics of SiC power MOSFET, it can be concluded that the sensitivity of the parameters affecting SiC MOSFETs’ switching behavior in sequence is gm, Vth, Cgd, Cgs, Cj, Ld and Ls. Thus, an effective method to optimize the parameters is proposed by adjusting the sensitive parameters in sequence, according to the errors between experiment and simulation results. Finally, the universality and accuracy of the improved model is verified by double pulse test under different load current.
Motivation
SiC power MOSFET is currently regarded as an attractive power device in high-power density applications. In order to accurately estimate the switching characteristics of SiC power MOSFETs, a valid model is necessary. Some research has been conducted on SiC MOSFETs model in previous work1-3. A physical numerical models based on the carrier drift-diffusion motion equations was proposed in [1] . However, it is not easy to be implemented in circuit simulations due to the complexity. Behavior model is an effective way for circuit simulation. Some improved models have been proposed2,3. Due to different manufacture batches and processes, the model parameters for each device is different. They can be extracted from the characteristics curves. However, to measure each device’s characteristics curves under different conditions is a heavy task. Another way is using the characteristics curve of the data sheet provided by manufacturer. But the parameters by using data sheet curve is lack of simulation accuracy for the switching transition. In this study, an optimization method for SiC MOSFET model parameter in PSpice is proposed.
Results
On the basis of the analysis of the influence of the parameters on the characteristics of SiC power MOSFET, it can be concluded that in the SiC MOSFET behavior model(see Fig.1), the sensitivity of the parameters affecting SiC MOSFETs’ switching behavior in sequence is gm, Vth, Cgd, Cgs, Cj, Ld and Ls. Then an effective method to optimize the parameters is proposed by adjusting the sensitive parameters in sequence(see Fig.2). The improved model has been experimentally evaluated by double pulse test under different load current(see Fig.3-Fig.5). It can be found that the simulation results after optimization have much better consistency with the measured ones.