PC randomly switches itself off

[UKTomm]

Hi guys,

 

I've been having some issues with my PC where it will turn itself completely off and then restart without warning. Sometimes when it does this and tries to reboot itself, it will turn itself off again and get into some vicious restarting cycle. It happens at any time such as having just booted the PC, watching a video, playing games. It even did it when I was uninstalling an old version of Java. Am I right to suspect it might be a dying PSU?

 

Just wanted to get your opinions before I go out and buy a new PSU which I may or may not need. 

Thanks

[profesorfrink42]

Is your CPU overclocked?

Is it running too hot?

Try replacing the cable (the one that goes from the PSU to the wall)

[vorticalbox]

Hi guys,

 

I've been having some issues with my PC where it will turn itself completely off and then restart without warning. Sometimes when it does this and tries to reboot itself, it will turn itself off again and get into some vicious restarting cycle. It happens at any time such as having just booted the PC, watching a video, playing games. It even did it when I was uninstalling an old version of Java. Am I right to suspect it might be a dying PSU?

 

Just wanted to get your opinions before I go out and buy a new PSU which I may or may not need. 

Thanks

Might be overheating, what are CPU temps like?

[UKTomm]

Is your CPU overclocked?

Is it running too hot?

Try replacing the cable (the one that goes from the PSU to the wall)

 

I'll replace the cable and see if that helps!

 

Might be overheating, what are CPU temps like?

 

I thought that but I wasn't too sure. I've got a AMD FX-8350 (not OC) and it's running between 10C and 24C at the moment. 

[Elliott Game]

I'll replace the cable and see if that helps!

 

 

I thought that but I wasn't too sure. I've got a AMD FX-8350 (not OC) and it's running between 10C and 24C at the moment. 

10 Degrees Celcius! thats very cold, your air con can only get down to 18 Degrees

[UKTomm]

10 Degrees Celcius! thats very cold, your air con can only get down to 18 Degrees

 

It is! The room is a bit chilly at the moment though so I'm sure it'll warm up as time goes on.

[StareX]

10 Degrees Celcius! thats very cold, your air con can only get down to 18 Degrees

Nothing wrong there but an excellent cooling!!

Also, try Furmark. And Aida64, tell us what happens there.

Furmark is for the GPU and aida64 is for the CpU.

If one fails, it's probably either the one or your PSU not outputting enough OMFFF to one of them.

[Elliott Game]

It is! The room is a bit chilly at the moment though so I'm sure it'll warm up as time goes on.

oh yea, sorry my mistake. It does currently snow in the UK, I'm from Australia.

[vorticalbox]

AMD temps idle aren't always the best. better stress it with OCCT, prime or something to check unload

[UKTomm]

Nothing wrong there but an excellent cooling!!

Also, try Furmark. And Aida64, tell us what happens there.

Furmark is for the GPU and aida64 is for the CpU.

If one fails, it's probably either the one or your PSU not outputting enough OMFFF to one of them.

 

Running Furmark at the mo. How long should I run these for?

 

Edit: Well scratch that, PC just shut itself off while running Furmark. So it's something to do with the GPU then?

[StareX]

What model is the GPU? Brand? how old is it..

PSU too.. Model.. Etc.

Also, when it happened, did it crash at WHAT temprature exactly???

[Sauron]

Please provide full system specs before anything

[UKTomm]

Please provide full system specs before anything

 

 

What model is the GPU? Brand? how old is it..

PSU too.. Model.. Etc.

Also, when it happened, did it crash at WHAT temprature exactly???

 

AMD FX8350 4.0Ghz (not OC)

Asus Radeon HD7850 (not OC)

Asus M5A99X EVO R2.0

Crucial SSD

Few standard HDDs

CiT 750UB 750W PSU

 

Didn't spot the temps exactly when it crashed but it was in the high 70s. Graphics card isn't that old that it should be causing problems. Probably about 18months old.

[Sauron]

Didn't spot the temps exactly when it crashed but it was in the high 70s. Graphics card isn't that old that it should be causing problems. Probably about 18months old.

 

Got bad news for you: amd cpus start thermal throttling at 64c and risk permanent damage when they hit the 72c mark. You definitely have an overheating problem. Are you using the stock cooler? Make sure it's clean and not caked in dust, if necessary remove it and reapply thermal paste.

[UKTomm]

Got bad news for you: amd cpus start thermal throttling at 64c and risk permanent damage when they hit the 72c mark. You definitely have an overheating problem. Are you using the stock cooler? Make sure it's clean and not caked in dust, if necessary remove it and reapply thermal paste.

 

That was the GPU, not the CPU. I was running Furmark at the time

[Sauron]

That was the GPU, not the CPU. I was running Furmark at the time

 

oh, ok. What were cpu temps though?

[StareX]

ASUS Radeon... what? DirectCU II? what.. and if it has 2 fans on it.. it means it's pre overclocked.

 

Run MSI Afterburner, test your GPU on the lowest core clock you can put it on (like if it's on 800.. try to decrease it to 700 just for the test...) and your voltage.. try to maximize it.. (if it crashes then... decrease the voltage.. like if it's 1.15V, decrease it to 1.00V (1000mV) or 1300mV as the max I believe.. or something. play with it.)

 

And then run Furmark and test it .. until it goes beyond 70C.. if it does and it crashes, it's most likely your GPU. if it freezes, it's your GPU.. if it restarts or doesn't boot.. it's your PSU.. but lets not jump to any conclusion here.. first test it and get back to us.

[UKTomm]

ASUS Radeon... what? DirectCU II? what.. and if it has 2 fans on it.. it means it's pre overclocked.

 

Run MSI Afterburner, test your GPU on the lowest core clock you can put it on (like if it's on 800.. try to decrease it to 700 just for the test...) and your voltage.. try to maximize it.. (if it crashes then... decrease the voltage.. like if it's 1.15V, decrease it to 1.00V (1000mV) or 1300mV as the max I believe.. or something. play with it.)

 

And then run Furmark and test it .. until it goes beyond 70C.. if it does and it crashes, it's most likely your GPU. if it freezes, it's your GPU.. if it restarts or doesn't boot.. it's your PSU.. but lets not jump to any conclusion here.. first test it and get back to us.

My GPU is the one seen here

 

Edit: I decreased my core clock from 860MHZ to 760MHz. No changes made to voltages. Ran it for 12 minutes with Furmark and it hit 80C with no crashes or freezing. So, I bumped the core clock  back up to 860MHz with no changes to voltage and ran it for 12 minutes. It hit 80C but didn't crash or freeze. These were the exact settings that it crashed with before so unless that it was just coincidence that it crashed while running Furmark the first time? 

[StareX]

Weird.. your computer is OK. Nothing wrong there. Reinstall drivers.. entirely.. and install them back.

 

To be fully honest with you? It's not your GPU nor your CPU.. nor your RAM. as I can see, on what you tell me.. you're fine 90%. the other 5% might be the PSU.. But I doubt it.. you could try to go to a local computer store and try another PSU, ask them to just test it... if they have one laying around. and the other 5? Probably drivers.. or just voltage issues.

 

Moreover.. try to run a CPU temp test on some games.. launch something and run a temp test .. use HWinfo64 .. along with MSI Afterburner ..