The following paper has been accepted for publication in IEEE Transactions on Semiconductor Manufacturing.
Power MOSFETs play an important role in power converters as high-performance switching devices. Variation-aware circuit simulation is now considered indispensable to improve the reliability of such devices. This paper deals with the modeling of power MOSFETs, which is essential for variation-aware circuit simulation. In particular, it proposes a method to extract the characteristic variation of MOSFETs as a statistical model of the model parameters.
In contrast to existing parameter extraction methods that can only be applied to model parameters that follow a normal distribution, the proposed method can simultaneously consider model parameters that follow different distributions, such as a lognormal distribution. In addition, the proposed method identifies the dominant set of model parameters that contribute significantly to the current characteristic variation of power MOSFETs.
The drain currents of planar and trench SiC power MOSFETs were analyzed, and it was shown that the threshold voltage and the current gain factor are particularly important in representing the variation of the current characteristics for both MOSFET structures.
Hiroki Tsukamoto, Michihiro Shintani and Takashi Sato:
“Statistical Extraction of Normally and Lognormally Distributed Model Parameters for Power MOSFETs,” IEEE Transactions on Semiconductor Manufacturing, (to appear).