.jpg){kind=link}
A few days earlier, Gigabyte released an official statement regarding its potentially "explosive" and hazardous power supply unit (PSU) models, namely the GP-P850GM and GP-P750GM. In that press release, Gigabyte, among other things, laid out a range of serial numbers (S/N) to indicate the units that would be eligible for the company"s Return and Exchange service.
However, a Redditor named Tinefol - who says they own such an eligible unit with the serial number "SN20243G001306" - earlier today posted a screenshot of their alleged conversation with Gigabyte representatives, with the latter refusing to accept a return request.
To be somewhat fair to Gigabyte personnel, the company did mention in its official statement that the "Terms and conditions for the GP-P850GM and GP-P750GM return and exchange service may differ by region." But the response from Gigabyte contains a different explanation altogether stating (translation) that:
This press release is applicable only to the newer batches.
And there was no mention of batches of any sort in the company"s original response.
.jpg){kind=link}
The Redditor also claimed that they are aware of other people too who have faced similar problems.
PSU expert, Dr. Aris Mpitziopoulos, who first noted this problem last year has also taken issue with Gigabyte"s response. Denoting such unfavorable reviews from people like Aris (via TechPowerUp), Gigabyte stated that it "takes reports of this manner extremely seriously".
However, when Dr. Aris reported about his failed review unit to Gigabyte last year, the company allegedly responded that "its engineers tested five units and found no problems." It also didn"t ask for the "bad sample back for failure analysis, which is the typical procedure" and also "didn’t offer a second sample to continue the review" (via Tom"s Hardware).
Now going to the solution of the problem that Gigabyte offered, the company said that the newer PSU units have a revised Over Power Protection (OPP) trigger limit from 120% ~ 150% down to 110% ~ 120% which should resolve the "potential issues raised by third parties, specifically, those discovered during their testing via DC Electronic Load equipment for extended lengths of time repeatedly close to the 120% to 150% OPP trigger point."
However, as per Dr. Aris, the actual problem may still persist. According to him:
This is not just a badly set OPP, since many units died under moderate loads and within short periods. I strongly believe that this is just a bad design, the FETs are not driven correctly, and although in quick pre-checks and normal conditions, the PSU can be ok, there are cases (not only under stress) where they fail. The timing of the FETs is not correct, and this is due to a lousy gate driver or a lousy implementation.
Finally, he has also added that he cannot be sure if the issue still persists without some more samples for testing.
Source and image: Tinefol (Reddit)