GaN-on-Si: Monolithically Integrated All-GaN Drivers for High-Voltage DC-DC Power Conversion

Chin Hsia*, Deng Fong Lu

*Corresponding author for this work

Research output: Contribution to journalJournal Article peer-review

5 Scopus citations

Abstract

This paper presents a novel integrated half-bridge driver architecture using GaN-on-Si process for high-speed and high-voltage DC-DC converters. The entire circuit includes only enhancement mode (E-mode) and depletion-mode (D-mode) GaN transistors. The high-side driver circuit adopts the E-stacked E/D-mode (EED) architecture, which can directly drive the gate of the high-side transistor with a low-voltage signal without using an additional level shifter, which simplifies the design and reduces propagation delay. In addition, the low-side power transistor is driven by stacking two D/E-mode devices. This architecture separates the high-side pulse from the low-side drive signal to prevent false triggering of the low-side driver. The designed fully integrated GaN driver can output a high-voltage pulse wave with an operating frequency greater than 1 MHz when the input voltage is greater than 200 V. The rise and fall times of the high-voltage pulse wave operating at a peak voltage of 200 V are 54.4 ns and 57.6 ns, respectively. The experimental results show that the circuit can effectively drive the half-bridge circuit and be applied to a buck converter. The designed buck converter can deliver up to 20.5 W of output power, and the maximum efficiency achieves 90.7%.

Original languageEnglish
Article number5109
JournalApplied Sciences (Switzerland)
Volume12
Issue number10
DOIs
StatePublished - 01 05 2022

Bibliographical note

Publisher Copyright:
© 2022 by the authors. Licensee MDPI, Basel, Switzerland.

Keywords

  • DC-DC conversion
  • GaN technology
  • high-side driver
  • integrated power converter

Fingerprint

Dive into the research topics of 'GaN-on-Si: Monolithically Integrated All-GaN Drivers for High-Voltage DC-DC Power Conversion'. Together they form a unique fingerprint.

Cite this