Surface-Potential-Based RF large signal model for Gallium Nitride HEMTs

S. Khandelwal, S. Ghosh, Y. S. Chauhan, B. Iniguez, T. A. Fjeldly

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

29 Citations (Scopus)

Abstract

We present a physics based large signal RF compact model for Gallium Nitride HEMTs (GaN HEMTs). This surface-potential-based model is called Advance SPICE Model for GaN HEMT or ASM-GaN-HEMT model. Surface-potential (SP) in the triangular quantum well of GaN HEMTs is derived by solving Schrodinger's and Poisson's equations consistently and analytically. Core analytical drain-current model is derived using the developed SP model and drift-diffusion transport. The core model is enhanced with models for key real device effects to represent a real GaN HEMT device. A consistent intrinsic charge model is also derived from SP. The developed model is implemented in Verilog-A. Excellent model agreement with DC, S-parameters and large signal RF power sweep measurements are shown for a GaN HEMT device with width W = 40 μm and number of fingers NF = 8.

Original languageEnglish
Title of host publicationCSICS 2015
Subtitle of host publicationProceedings of the 2015 IEEE Compound Semiconductor Integrated Circuit Symposium
Place of PublicationPiscataway, NJ
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages1-4
Number of pages4
ISBN (Electronic)9781479984947
DOIs
Publication statusPublished - 30 Oct 2015
Externally publishedYes
EventIEEE Compound Semiconductor Integrated Circuit Symposium (37th : 2015) - New Orleans, United States
Duration: 11 Oct 201514 Oct 2015

Conference

ConferenceIEEE Compound Semiconductor Integrated Circuit Symposium (37th : 2015)
Country/TerritoryUnited States
CityNew Orleans
Period11/10/1514/10/15

Keywords

  • AlGaN/GaN HEMTs
  • Compact models
  • Intermodulation distortion

Fingerprint

Dive into the research topics of 'Surface-Potential-Based RF large signal model for Gallium Nitride HEMTs'. Together they form a unique fingerprint.

Cite this