[SOLVED] Intermittent restarts (usually overnight)

[pk1966]

I've recently upgraded my PC based upon

MSI MPG Z390 GAMING EDGE AC
Intel i5 9600KF
MSI GeForce GTX 1050 Ti

But I retained my 10 year old Corsair 750W TX PSU.

The PC runs fine and cool most of the time, but I find that overnight it sometime restarts (sometimes repeatedly). There is nothing in the event log when the system shuts down, just some errors where the system restarts to say it was shutdown unexpectedly. It has never shutdown whilst in use, even when trying to stress the system with memory or graphics diagnostic tools.

e.g. "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."

I also have 4 SATA HDD & 1 SATA SSD drives installed.

My suspicion is the PSU, given its age, but don't want to splash out on a new one if that's not the case - any suggestions on how I can diagnose it properly?

Thanks
PK

 

[PC Tailor]

Even if it WASN'T the PSU. They all have a shelf life, and I wouldn't use many of the best quality ones after 10 years. They deteriote over time.

With random unexpected shutdowns, there can be several causes, so unfortunately there is no quick and easy way to find out, but typically it will end up being; PSU > GPU > RAM.

The only true way of finding out what may be causing it would be:

• First attempt to run in safe mode and see if the shutdowns still occur. If they don't you know it could be a third party module (commonly GPU drivers) that are faulting (usually due to a faulty GPU). Or safe mode is not pushing the component enough to cause the shut down.
• Update your BIOS and see if the issue persists.
• Ensure your CMOS battery is functioning (rarely does this cause shutdowns, but I've seen it before)
• Swap the PSU for a known working one and see if the issue persists. You can to a degree - test the PSU with a multimeter, and test that each rail is outputting the relevant voltag with no variation above 5%. But this only mimics low loads so isn't always reflective as often PSUs can fault at higher loads.
• Swap the GPU for a known working one and see if the issue persists.
• Perform memtest, removing all but 1 module, or swap the RAM modules with known working ones and seeing if the issue persists.

 

[pk1966]

• First attempt to run in safe mode and see if the shutdowns still occur. If they don't you know it could be a third party module (commonly GPU drivers) that are faulting (usually due to a faulty GPU). Or safe mode is not pushing the component enough to cause the shut down.
• Update your BIOS and see if the issue persists.
• Ensure your CMOS battery is functioning (rarely does this cause shutdowns, but I've seen it before)
• Swap the PSU for a known working one and see if the issue persists. You can to a degree - test the PSU with a multimeter, and test that each rail is outputting the relevant voltag with no variation above 5%. But this only mimics low loads so isn't always reflective as often PSUs can fault at higher loads.
• Swap the GPU for a known working one and see if the issue persists.
• Perform memtest, removing all but 1 module, or swap the RAM modules with known working ones and seeing if the issue persists.

Thanks for your input.

Updates on a few of your suggestions:

• I have updated the BIOS to the latest version
• The CMOS battery appears to be functioning - all BIOS settings are retained through a complete power off (at the wall)
• Whilst I haven't tried a multimeter, CPUID HWMONITOR is showing fairly stable voltages (this screenshot shows min/max over a 2 hour period whilst the PC spent about 30 mins idle and the rest encoding video)
• I have run MEMTEST86+ several times using the RAM that's installed and it always passes with no errors. I have just run it for about 3 hours with zero errors.

Tonight, I will leave it in safe mode overnight and see if that makes any difference, then will try swapping out the GPU as the next step.

Thanks
PK