Frequently Asked Questions

eGaN Reliability

Is there reliability data (MTBF, FITS) available for the EPC eGaN® transistors?

We have MTTF (Mean Time to Fail) Published in Reliability Report Phase 5,

Over 1.7 million accumulated device hours of reliability testing validate the readiness of eGaN FETs to support most commercial power switching applications. This report and all of our reliability testing reports can be found at

What is the expected HBM (Human Body Model) ESD rating of the parts? Is there special handling required?

ESD tests were performed per JEDEC and EIA standards for HBM and Machine Model (MM) respectively. On HBM these devices meet Class 3A or higher and for MM Class C. HBM ESD level can be found in the following document: “EPC GaN Transistor Application Readiness: Phase Two Testing

How do I predict failure rates for the EPC eGaN FETs?

Life testing is ongoing for the eGaN Devices and based on this, EPC has published several phases of reliability reports.

After 30 years of study, silicon MOSFET degradation mechanisms are well understood and therefore the failure rates can be estimated at various operating conditions. The structure of EPC's eGaN® FETs is significantly different from MOSFETs, and therefore the degradation mechanisms are quite different. While to date we have run over 1,800 devices through more than 1,700,000 hours of maximum stress testing, we have not yet generated enough information to quantify the key degradation mechanisms in eGaN FETs. Since eGaN technology employs a new type of semiconductor material, acceleration factors used for silicon may not be applicable We are continuing to run thousands of parts through various reliability tests with the goal being to give the customer information they need to project failure rates under actual operating conditions. In future work, temperature and voltage acceleration factors will be investigated.

EPC has been publishing reliability reports on a regular basis that give the user more and more detail, and at higher statistical confidence levels ( We will be using this publication to present data on failure mechanisms as they emerge from our testing.