Abstract :
It is pointed out that although quite a number of nonlinear descriptions are available for modeling microwave GaAs MESFETs (GaAs FETs), their accuracy is much debated. Like their silicon bipolar counterparts, the GaAs FET models are based on experimental curve tracer measurements and a limited set of scattering parameters (S-parameters) measured at convenient bias points. Hence, it is impossible to be confident of their validity under conditions for which they have not been tested. This situation is complicated by a trend toward proprietary models, developed by software companies who are unwilling to share the details of their work with customers for fear of giving secrets away to potential competitors. Most of the GaAs FET models available today as part of a commercial design packages are characterized by a hyperbolic tangent or polynomial fit to the device current-voltage (I-V) curve, which is determined by actual measurement, not by studying the underlying physics. Thus, the different mathematical manipulations employed by the various developers yield significantly different results when applied to a nonlinear circuit analysis.<>
Keywords :
III-V semiconductors; Schottky gate field effect transistors; gallium arsenide; nonlinear network analysis; semiconductor device models; solid-state microwave devices; GaAs field effect transistors; IV characteristics; bias points; current-voltage curve; curve tracer measurements; microwave MESFETs; nonlinear circuit analysis; scattering parameters; semiconductor; Current measurement; Gallium arsenide; MESFETs; Microwave FETs; Packaging; Physics; Polynomials; Scattering parameters; Silicon; Testing;
