A Glance at Device Stability in Load-Pull Measurement Setups

Source-pull and load-pull techniques are widely used to characterize and test RF transistors. The measurement system varies source and load impedances to evaluate the performance of DUT. However, RF transistors are not unconditionally stable due to their physical structures. Therefore, they tend to oscillate during source-pull and load-pull measurements. A load-pull setup usually consists of couplers, tuners, bias-tees, circulators, and driver amplifiers. These components might cause resonances and trigger oscillations during measurements. In this study, the frequency response of a passive load-pull setup is analyzed in a wide range to understand the effect of measurement setup equipment on DUT's stability. The measured results show a considerable impedance variation outside of the operation band. Some solutions are recommended, and the necessity of new equipment is discussed.

I. Introduction

High-power RF transistors operate in the nonlinear region to achieve high efficiency and output power. It is possible to get high efficiency or power from a transistor by forming the voltage and current wave shapes at the current source plain of the drain or collector [1] with the suitable source and load impedances. Although the load-line theory could provide optimum load impedance at the current source plane, nonlinear intrinsic components and other passives, such as pads, bonding wires, and package parasitics, cause a frequency-dependent and complex load impedance requirement to achieve high efficiency or output power [2]. In most cases, it is not easy to model or estimate these effects to calculate optimized source and load impedances. Therefore, measuring a DUT's electrical parameters, such as input impedance, delivered input power, output power on a specific load, and efficiency, etc., at particular source and load impedance values is more practical and beneficial for understanding the behavior of the DUT.