PC Abruptly switches off

Mad_Matt01

Reputable
Aug 5, 2014
13
0
4,510
I've had my PC for a little over 2 years now, and in the last 2 weeks I've had a problem where my PC would shut off after normal use, without warning and as if it was a power cut. By normal use I mean gaming for an hour and idle at 3 hours. I've looked at temps for CPU, GPU etc and none of them are even close to 70 Degrees when the PC goes off. I'm led to believe it is somehow my PSU overheating as I can let the PC rest for 20 minutes and it will work just as fine, as long as before turning it on I drain the system of power and turn it back on.

With that meaty paragraph outta the way here are my specs:

PSU: EVGA Supernova NEX 650G
CPU: AMD FX-8320 (Not overclocked from default; what a waste)
CPU Cooler: CM Hyper 212 Evo
GPU: ASUS GTX 970 STRIX
MOBO: Gigabyte GA-990FXA-UD3

As of writing this, I've just finished a 40 minute Prime95 stress test and the CPU had a max of 55 degrees C.
I am incredibly confused as to what the problem is here and any insights will be much appreciated.
 
Solution
Temp issues aside (and please read the links aquielisunari posted, they're important), the fact that you have to muck about with draining power points to a bad PSU to me. The G1 series were not especially well built PSUs. Go through the links aquielisunari posted, monitor your CPU correctly and see if you have a heating issue. If not, then I'd seriously look at replacing out the PSU to a better version. GQ, GS, G2 and G3 are all solid gold class PSUs from EVGA
40 minute Prime95 stress test and the CPU had a max of 55 degrees C

stands out as odd.

I would download http://www.amd.com/en-us/innovations/software-technologies/technologies-gaming/over-drive and try that test again. Whatever monitoring program you're using is giving a false reading.

http://www.tomshardware.com/faq/id-2122665/understanding-temperature-amd-cpus-apus.html can help if you don't understand how AMD Overdrive works. You are looking at margins instead of an actual temperature.
 
Temp issues aside (and please read the links aquielisunari posted, they're important), the fact that you have to muck about with draining power points to a bad PSU to me. The G1 series were not especially well built PSUs. Go through the links aquielisunari posted, monitor your CPU correctly and see if you have a heating issue. If not, then I'd seriously look at replacing out the PSU to a better version. GQ, GS, G2 and G3 are all solid gold class PSUs from EVGA
 
Solution

Mad_Matt01

Reputable
Aug 5, 2014
13
0
4,510
Well after 2 hours of idling and then 1 hour of pure Prime95 Stress testing my PC finally shut off, again. CPU temps seemed normal, AMD overdrive gave 23 Degress C thermal margin before the PC shut off. PSU was cold to the touch, and I'm now leaning towards a failing component of the PSU or as ko888 said a "protection circuit in the PSU is being triggered." While looking around in my PC I noticed the Northbridge (At least what I assume is the Northbridge heatsink) was incredibly hot to the touch, 65+ C I'd say. Is it possible that this is causing the shut offs? Just thought I'd mention it.

I'll look further into replacing my PSU but it seems I cannot get anything from the 5 Year Warranty I have on the PSU because I don't have the original packaging; go figure. So I'll most likely go with a full new purchase of a PSU unless anyone else proposes an idea in this thread.

Thanks for your thoughts so far guys!