A fully functional table-based nonlinear model of the heterojunction bipolar transistor (HBT) is presented which includes explicit thermal feedback. The model uses four table-based nonlinear functions: Ic, Qc, Vbe, and Qb, all defined versus Ib and Vce by using a nonuniform bias grid. Thermal modeling (self-biasing and environment temperature dependence, Ta) is done by linearly mapping the table-based current functions versus Ta coupled with explicit thermal feedback. Four table-based nonlinear coefficients are required to accurately predict the device behavior versus temperature. Excellent results have been obtained under dc, small, and large signal excitations for InGaP/GaAs HBTs in the range 10degC to 110degC.

A nonlinear table-based thermal model of the static HBT behavior is presented for the first time and validated under DC operation. The generated model uses table-based nonlinear functions I c and V be in the output and input ports, respectively, and both are defined vs. I b and V ce by using a non-uniform bias grid. Thermal modeling (self-biasing and environment temperature dependence) is done by means of two analytical functions (one for each device port) involving table-based nonlinear coefficients. Five tables are the minimum required to accurately predict the device dc behavior vs. temperature. Model extraction is direct and does not require optimization. Excellent results have been obtained for different InGaAs/GaAs HBTs in the range 10 to 100 degC