What happened? Can anyone explain? (Power problems)

[deadman_uk]

First of all I don't know if this is an issue with my PSU or not so I will explain what happened and ask for advice. My computer has been working perfectly for ages when suddenly 5 days ago the whole computer powered itself off. I saw no "Windows is now shutting down" screen, it just powered off as if someone flipped the power switch at the back of the PSU (which they didn't).

I was shocked to see this, and immeditely turned the computer back on, went into the BIOS and checked the CPU temperature. It was a massive 80C!!!. I turned the pc off and gave the inside of my PC a good clean. I then powered my PC back on and monitored the temperature of the CPU for 20 minutes on idle.. didn't pass 45C. Afterwards, I ran video editing software (I was encoding a video) which would take 2 hours to complete. While this was going on, I was monitoring the CPU temperature.. didn't pass 56C. An hour into this, the PC decided to power itself off. I turned the PC back on, checked the temperatures.. all fine.

I continued to use the PC and it powered itself off once more a few hours later. I decided to run MemTest which I have installed on a Floppy Disk. I ran it for 5 hours and it found no errors. I then ran a torture test with Prime95 for 6 hours.. it passed all tests. After this, I left chkdsk running overnight (I told it to attempt to scan and fix bad sectors too). I am unsure of the results but when I woke up the next morning, I was greeted with my desktop so I know the PC didn't power itself off while I was sleeping.

It's been 5 days now and I have had no problems at all. What could cause this? My PC and all computer equipment (speakers, monitor, printer etc) are plugged into a 6 way surge protector.

 

[dandragonrage]

That wouldn't happen if someone flipped the switch at the back of the PSU, FYI. When it turned itself off the second+ time, did it do Windows is shutting down, or did it just turn immediately off?

 

[michael100]

It happened only ones and 5 days ago?
I don't think that's a big problem as long as it don't happen again.

 

[deadman_uk]

That wouldn't happen if someone flipped the switch at the back of the PSU, FYI. When it turned itself off the second+ time, did it do Windows is shutting down, or did it just turn immediately off?

If I flip the switch at the back of my PSU, my whole PC loses power instantly, which is exactly what has happened 5 days ago multiple times.

No, it never said Windows is shutting down, it just powered off immediately

It happened only ones and 5 days ago?
I don't think that's a big problem as long as it don't happen again.

It happened around 4 times in one day, then it never happened since. Why did it do this? Why did my CPU temp reach 80C?

 

[dandragonrage]

Oh, I misread. Thought you said the first time you saw that. What's your BIOS have set for shutdown temperature?

 

[engiNURD]

It sounds like a heat issue. Answer Dan's question above. Run Prime95 on smallFFTs for 24hrs and see if its stable. What CPU HSF are you using? What are your current idle and load temps? It could also be your PSU. How old is that Antec NeoHE?

 

[deadman_uk]

It sounds like a heat issue. Answer Dan's question above. Run Prime95 on smallFFTs for 24hrs and see if its stable. What CPU HSF are you using? What are your current idle and load temps? It could also be your PSU. How old is that Antec NeoHE?

Hi, my Antec NeoHE PSU is just over 2 years old. I am using the stock HSF with my Opteron 185. My current idle CPU temperature is in the low to mid 40C's. My load CPU temperature is around the mid 50C's - low 60C's. My motherboard temperature idle is 25-30C.

I have looked in every menu in my BIOS and can find no safety features relating to the shutdown of my system. However, I did find a CPU setting relating to temperature. If my CPU reaches a certain temperature, the fan will kick in at 100%. I currently have this set to 57C.

Must I run the smallFFT test for 24 hours? I have been running games like Crysis and Prostreet for the last 5 days fine. I have also been doing alot of video encoding.... surely this is good to test the stability of my system?

Am I still unsure what has caused this though.

 

[engiNURD]

Games are fine tests for usability, but I doubt they stress the CPU enough to be used as CPU stability tests. The smallFFT test would just be to double check your system's stability. Another way to do this, if you suspect it may be GPU related, is to loop 3DMark for a long time.

Your temps look fine, btw.

 

[paulmr]

whats your temps normally idle and while gaming..... I'd freak if my hit 80C...

 

[deadman_uk]

Games are fine tests for usability, but I doubt they stress the CPU enough to be used as CPU stability tests. The smallFFT test would just be to double check your system's stability. Another way to do this, if you suspect it may be GPU related, is to loop 3DMark for a long time.

Your temps look fine, btw.

I do not believe it's GPU related because my GPU is only a couple of months old and I don't think it would cause the whole PC to turn off. If I have time, I will try out the smallFFT test.

whats your temps normally idle and while gaming..... I'd freak if my hit 80C...

I already said above and I did freak out when I saw my temperature at 80C.

 

[paulmr]

oh yeah whoops sorry.. still odd that it jumped up that high and never has before and hasn't again since....

were you able to see your cpu fan moving just prior to it shutting off?.. maybe for some reason it just wasn't turning.. I've seen that happen before

personally if I had done all the tests you have done and it has not happened again since I would stop worrying about it.

Might be worth doing some preventive things to protect you CPU in case it happens again not sure what temps it would take to seriously damage it but maybe set the temp shutoff a little lower. Also you could use a monitoring program like speedfan http://www.almico.com/speedfan.php probably a better progeam out there for this but I think it can log your temps and fan speed.. I use it to monitor my temps but never have tried the log feature.... but if it happens again you see if the fan was the cause.

 

[deadman_uk]

oh yeah whoops sorry.. still odd that it jumped up that high and never has before and hasn't again since....

were you able to see your cpu fan moving just prior to it shutting off?.. maybe for some reason it just wasn't turning.. I've seen that happen before

personally if I had done all the tests you have done and it has not happened again since I would stop worrying about it.

Might be worth doing some preventive things to protect you CPU in case it happens again not sure what temps it would take to seriously damage it but maybe set the temp shutoff a little lower. Also you could use a monitoring program like speedfan http://www.almico.com/speedfan.php probably a better progeam out there for this but I think it can log your temps and fan speed.. I use it to monitor my temps but never have tried the log feature.... but if it happens again you see if the fan was the cause.

When it first powered off, I didn't know it would do this, I was caught by suprise so no, I wasn't watching the CPU fan at all. My case was shut and I only open it when I have a problem. When I went to the BIOS and saw the 80C temperature, that's the next thing I did... check to see if the CPU fan was spinning (it was).

I will forget this has happened but if I get any more problems, I will try out speedfan. Thanks all.