Correlation between Pulse I-V and Human Body Model (HBM) Tests for Drain Electrostatic Discharge (ESD) Robustness Evaluation of GaN Power HEMTs
Jiahui Sun, Zheyang Zheng, Li Zhang, Kevin Jing Chen
Abstract
Open-access reader
Jiahui Sun, Zheyang Zheng, Li Zhang, Kevin Jing Chen
Abstract
Open-access reader
The drain-to-source HBM (human body model) ESD (electrostatic discharge) robustness of GaN HEMTs has been investigated through the TLP (transmission line pulse) I-V tests in most previous works. However, TLP tests and HBM ESD tests are found to be miscorrelated using the conventional method applicable to Si MOSFETs. In this work, it is found that the drain-to-source breakdown mechanism of GaN HEMTs contradicts the basis of the conventional correlation method, which also leads to much longer rise time of the drain-to-source voltage (Vds) in the HBM ESD test than in the TLP test. Therefore, the HBM ESD test of GaN HEMTs is correlated with a pulse I-V (PIV) test with easily adjustable rise time to establish a more suitable method. The HBM ESD sensitivity is estimated from the PIV test with an error of 4.5%.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The drain-to-source HBM (human body model) ESD (electrostatic discharge) robustness of GaN HEMTs has been investigated through the TLP (transmission line pulse) I-V tests in most previous works. However, TLP tests and HBM ESD tests are found to be miscorrelated using the conventional method applicable to Si MOSFETs. In this work, it is found that the drain-to-source breakdown mechanism of GaN HEMTs contradicts the basis of the conventional correlation method, which also leads to much longer rise time of the drain-to-source voltage (Vds) in the HBM ESD test than in the TLP test. Therefore, the HBM ESD test of GaN HEMTs is correlated with a pulse I-V (PIV) test with easily adjustable rise time to establish a more suitable method. The HBM ESD sensitivity is estimated from the PIV test with an error of 4.5%.
Key concepts: Electrostatic discharge, Human-body model, Robustness (evolution), Voltage, Transmission line, Electric power transmission, Materials science, Breakdown voltage