My brother's Windows 11 PC randomly loses its display and powers down with no warning. The fans and lighting keep running briefly, then gradually stop and turn off. There are no useful entries in Event Viewer or Reliability Monitor. The problem happens both while gaming and while completely idle, with anywhere from an hour to two weeks between incidents. The latest shutdown occurred about two weeks after replacing the power supply and its cables.
So far we have replaced the PSU, reinstalled Windows, monitored temperatures with HWiNFO, checked system logs, removed overclocks, and disabled XMP. We are reluctant to replace more components at random and would appreciate ideas for narrowing down the cause.
The system has an ASUS ROG B550-F motherboard, Windows 11 Home, 32GB of Corsair RAM in a 2x16GB kit, an NVIDIA RTX 2080 Ti Founders Edition, a Corsair RM750e PSU, a 1TB SSD, and five fans with a radiator.
4 Answers
Check for a motherboard short caused by an incorrectly placed case standoff. One way to test is to remove the board from the case and run it with only the correct standoffs installed, using the minimum hardware needed. A misplaced standoff can create intermittent faults that are difficult to reproduce.
A failing motherboard is one possibility, especially with completely random timing and no obvious thermal or software errors. I’d also run MemTest86 overnight rather than relying only on Windows’ built-in memory test, since unstable RAM can cause strange shutdown behavior.
I’ll run MemTest86 overnight. The built-in test reported no errors before, but I’d rather use a more thorough test.
There are several possible sources of instability here: RAM, the graphics card, the motherboard’s VRM, or power delivery. Try isolating components methodically: test with one RAM stick at a time, remove the GPU if the CPU has integrated graphics, and consider disabling CPU C-states because some Ryzen systems can become unstable during low-power idle states. Also try a different wall outlet. The B550-F may not expose a dedicated VRM temperature sensor, so keep an eye on general case airflow and motherboard temperatures.
The long gap between failures makes testing difficult, but these suggestions give me a useful order to work through instead of replacing parts blindly.
Fast Startup can make Windows behave more like it is hibernating than fully shutting down, which may affect what gets recorded. Disable Fast Startup and check for hibernation-related events after the next failure. That may provide more information than looking only for a normal shutdown event.
A senior colleague suggested checking the standoffs too, so I’m going to inspect that.