A Method for Deriving Large-Signal Input Impedance of Microwave Devices Using a Modified Source and Load Pull Technique

Microwave power-amplifier technology is no longer dominated by class-A FET designs. For example, GaAs HBTs offer high-efficiency microwave performance in class-B and class-C operation. When operated at lower conduction angles, device input impedance is often a strong function of input-signal amplitude. Knowing Zin for the expected bias and drive conditions is essential to first-pass success of input- and interstage-matching circuits. One widely-used, yet incorrect, method of estimating large-signal input impedance optimizes the input tuner for maximum power transfer. Assuming that this produces a complex-conjugate impedance match, the designer assigns the device under test an input impedance equal to the complex conjugate of the tuner's s22. This procedure has triggered numerous unnecessary design iterations to correct the frequency responses of high-efficiency amplifiers. The authors present a technique for accurately calculating Zin of a device under test from tuner data obtained in a modified source and load-pull configuration. The new technique inherently improves the robustness of tuner-optimization algorithms, and also allows an input tuner to function for higher device reflection coefficients than previously possible.