[SOLVED] My computer keeps switching off without warning

Jun 23, 2020
1
0
10
0
Hi,
My computer keeps switching off at random intervals without warning. Sometimes it will be on for an hour, another time it will be on for only two minutes. Every time it happens, I will have to switch off and then on the power supply before it will allow me to try switch the pc on again. I usually also have to wait for a while, or the pc will quickly turn off again, soon.

I have a Corsair HX850i power supply, intel i5 4440, Gigabyte Z87M D3H motherboard, EVGA 970sc gpu, 2 hdd, 2 ssd (one is the c drive), my cpu is water cooled (Cooler Master Nepton 240M), 8gb ddr3, and a Thermaltake Armor A60 AMD edition tower.

If it makes a difference, it will perform perfectly for most of the year, and then come June, this starts. Every year. Without fail. June is winter in South Africa, but it isn't too cold.
 
Jun 24, 2020
10
0
20
2
It's likely a PSU issue when feeding power to the GPU. If you haven't already, download GPUz and MSI afterburner. GPUz will allow you to monitor current clock speeds, voltages, and temps of your GPU. Also download OCCT and stress test the CPU, Mem, GPU and PSU and let it run for 5 mins each. I predict your system will crash on the GPU and PSU test but see what happens.

Use MSI Afterburner to reduce the core clock speed (DON'T touch voltages) and see if you're still getting a full system crash. In GPUz for me, my GPU (default speed approx. 1600MHz) was running above its set boost speed (approx. 1800MHz) by +100MHz so I just used Afterburner to reduce the clocks by 200-250MHz to get it down to its default and now I've been able to game and work without a problem (so far, haha). Once you've made those changes, just go about how you'd normally use your PC and see if the problems persist.

If you have a spare PSU or know someone who wouldn't mind lending you theirs, it's worth testing your system with that. Personally, 850w is definitely overkill for your current build. A 650w or even 500w PSU would do just fine and be a much cheaper replacement should your PSU be the issue.
 
Jun 24, 2020
10
0
20
2
It's likely a PSU issue when feeding power to the GPU. If you haven't already, download GPUz and MSI afterburner. GPUz will allow you to monitor current clock speeds, voltages, and temps of your GPU. Also download OCCT and stress test the CPU, Mem, GPU and PSU and let it run for 5 mins each. I predict your system will crash on the GPU and PSU test but see what happens.

Use MSI Afterburner to reduce the core clock speed (DON'T touch voltages) and see if you're still getting a full system crash. In GPUz for me, my GPU (default speed approx. 1600MHz) was running above its set boost speed (approx. 1800MHz) by +100MHz so I just used Afterburner to reduce the clocks by 200-250MHz to get it down to its default and now I've been able to game and work without a problem (so far, haha). Once you've made those changes, just go about how you'd normally use your PC and see if the problems persist.

If you have a spare PSU or know someone who wouldn't mind lending you theirs, it's worth testing your system with that. Personally, 850w is definitely overkill for your current build. A 650w or even 500w PSU would do just fine and be a much cheaper replacement should your PSU be the issue.
 
If it makes a difference, it will perform perfectly for most of the year, and then come June, this starts. Every year. Without fail. June is winter in South Africa, but it isn't too cold.
It only does this once a year?!?!?!

I was going to say it sounds like over temperature protection. But you're saying that it's Winter and NOT hot and it works other times of the year.
 

ASK THE COMMUNITY