Inquiry about PC experiencing crashes without BSOD. Seeking Guidance.
Inquiry about PC experiencing crashes without BSOD. Seeking Guidance.
I previously shared my problems which remained unresolved. I initially doubted it was a hardware issue due to its specific characteristics, but I now believe I've identified the problem. I can't say this with complete certainty, but I suspect that the GPU making contact with my Motherboard at a specific worn area depicted in the second image is the culprit. At least, I hope this is the case so I can finally resolve this frustrating issue. My PC has been crashing unexpectedly without any warning or BSOD, just shutting down and not powering back on, although I can restart it right away. I've tried almost everything possible short of replacing components. Here’s my current situation and my theory on the matter.
I think the GPU's contact with the motherboard at the worn spot is the main problem. I recently rebuilt the PC, reseating everything except for the CPU and CPU cooler. I own an MSI Gaming Trio 4090 and an ASUS Maximus Hero z790. After this maintenance, the issue seemed to vanish. I noticed the worn area on the motherboard's heatsink but dismissed it as a non-issue. However, as a 4090 owner, I wanted to ensure the power cable was connected securely, so I disconnected and reconnected it. The following day, my issues resurfaced. I suspect my vigorous handling of the power cable has exerted pressure on the card, pressing it against the motherboard's heatsink, potentially causing a short or other problems that I cannot identify. After the issue returned, I examined the setup closely and saw that my GPU was indeed touching the motherboard. Normally, I wouldn't think this would cause a problem, but it seems to for me. I gently pulled on the end of the card to relieve some pressure when plugging in the power cable, creating a tiny gap between the GPU and motherboard, and so far, no crashes have occurred. I could still be mistaken, but extensive trial and error has brought me to this point and a certain level of success. I would appreciate everyone’s insights and opinions on this matter. The system is functioning well otherwise, and I don’t believe any damage has occurred since my benchmarks continue to improve over time. I’m not planning to act on this immediately, but I’m considering placing a small piece of silicone or another heat-resistant non-conductive material over the specific wear area to eliminate any chance of this issue permanently. Does anyone have suggestions or recommendations regarding the material I could use?
It seems that the pressure you applied while connecting the PSU cable may have caused a short, but I doubt it since the protective casing on the cards usually mitigates this. It sounds more like you might have misaligned the pins, preventing proper contact with the PCIe port and resulting in data loss, which led to the system crashing hard. You shouldn't need excessive pressure to connect the GPU; that approach is unwise and damaging. The proper method is to grasp the edge of the card gently and connect the cable with care, pushing just enough to hear the click when it seats. If you have to use force, there’s likely an underlying issue, possibly a faulty cable, and less likely, damaged pins on the GPU's power receiver.
I apologize for slightly exaggerating the force I applied. It was firm but not excessive (This PC cost more than my vehicle, and I handle it as carefully as possible). You are correct; it may indeed be that I misaligned the pins in the PCIe slot rather than the wear spot being the problem. However, I assure you, I wasn't being overly forceful, and it saddens me that this issue is so sensitive; one slight bump against my desk might trigger it.
This is why it is recommended to use a support stand for heavier cards to maintain proper alignment.
Every 4090 comes with its own anti-sag bracket; I utilized their bracket along with an additional one.