Quantitative Model-Based False Turn-on Evaluation and Suppression for Cascode GaN Devices in Half-Bridge Applications

Owing to the high operation frequency and fast switching speed, gallium nitride (GaN) devices are prone to the false turn-on phenomenon, causing high switching loss, shoot through, and even sustained oscillation. However, most of research on false turn-on aims at the enhancement-mode GaN devices, whereas there is still little published research on GaN devices in cascode configuration due to the comparatively complicated structure and various parasitic components. This paper conducts a comprehensive and in-depth study for the problem of false turn-on in cascode GaN devices based half-bridge circuits, which has a great guiding significance in the device selection, printed circuit board (PCB) design, and debugging of cascode GaN converters. In this paper, a novel analytical model of the device suffering false turn-on is first developed, with all the parasitic parameters being fully considered. Based on the model, the quantitative and accurate expression of the induced gate-to-source voltage is derived, which can serve as a precise and significant reference for judging and evaluating the false turn-on problems. Then, the influences of device and circuit parameters on the peak value of induced voltage are first investigated in detail, providing a reliable guidance for the usage and PCB design of cascode GaN devices based half-bridge converters. What's more, guidelines to effectively suppress the false turn-on phenomenon are also given, which can be adopted both in the design and debugging process of cascode GaN applications. Finally, all the theoretical calculations and analysis are verified by experiments with satisfying results and performances.