No abstract

No abstract

No abstract

This paper reviews and contrasts two complementary device modeling approaches based on data readily obtainable from a nonlinear vector network analyzer (NVNA) [1]. The first approach extends the application of waveform data to improve the characterization, parameter extraction, and validation methodologies for "compact" transistor models. NVNA data is used to train artificial neural network -based constitutive relations depending on multiple coupled dynamic variables, including temperature and trap states for an advanced compact model suitable for GaAs and GaN transistors. The second approach is based on loaddependent X-parameters* [2], [3], [5], [6], measured using an output tuner working with the NVNA. It is demonstrated that Xparameters measured versus load at the fundamental frequency predict well the independent effects of harmonic load tuning on a 10W GaN packaged transistor without having to independently control harmonic loads during characterization. A comparison of the respective merits of the two approaches is presented.

IV knee walk-out in AlGaN/GaN high electron mobility transistors (HEMTs) on a Sapphire substrate is analyzed using dynamic radio frequency (RF) load-lines acquired with a large signal network analyzer (LSNA) for both continuous-wave (CW) and pulsed-IV/RF excitations. When thermal effects and traps are bypassed using pulsed-IV biasing and pulsed-RF excitations, the IV knee walk-out observed in CW load-lines is found to be effectively suppressed and the device delivers the maximum output power expected for class A operation. It is also demonstrated using pulsed-IV/RF measurements at various substrate temperatures that the IV knee walk-out primarily arises from thermal effects at high bias rather than trapping in the on-wafer devices characterized.Index Terms-GaN, high electron mobility transistor (HEMT), large signal network analyzer (LSNA), load-line, traps.