PC shutdown issue, PSU -12v at -9.5V ?

[westom]

What is strange is that during both today and yesterdays tests the PC didn't once shutdown, and that's makes this issues so frustrating

One other (rare) event that I forgot to mention. Sometimes a fan gets stuck due to reasons explained later if relevant. Once the fan starts spinning, it keeps spinning. But you should be able to stop the fan. And it restarts every time and quickly when released. A fan that only slowly returned to full speed may not always start - especially when cold.

If the fan does not spin and if that unique power supply design contains a temperature monitor (not required but some have that feature), then a too warm supply will declare a not 'Power Good' and shutdown.

That would explain your ‘I cannot get it to fail’ scenario.

 

[Dutchman]

Cheers again for the comments and suggestions.
Plec, makes sense what you are saying and once finished here I'm gonna give that plug a wiggle and see what happens.

Weird_Dave, If I knew when it was shutting down I could run the test, but it's very random, and each time I have deliberately tried to get it to shutdown by running load and such nothing happens.

Westom, good point indeed, I've not even thought of checking PSU fan, most notably because my PSU is mounted fan down in an Antec p182, but may unscrew it and see what the fan is doing.

 

[weird_dave]

I didn't literaly mean test it before it shuts down, but I realise I missed the bit where you say you've actually done it anyway. I blame it on the man cold....
My money is on the PSU fan or a dodgy connection somewhere, which may now be less dodgy or fixed due to fiddling around.

 

[westom]

Plec, makes sense what you are saying and once finished here I'm gonna give that plug a wiggle and see what happens.

A defective connector would appear in the numbers. Simply measure voltage after the connector. If connector is intermittent, then that defect also appears on the meter as a voltage reduction. Observe a defective plug using the meter; even when connector is not causing failures today. Either voltage will always be lower, or the connector is not causing intermittents.

No reason to remove that fan. A tool, ears, and your hand should be sufficient to feel the fan working. A stick can be stuck into an 80 mm fan to stop it. Then see (hear) how quickly it restarts when released. A fan with sufficient torque will start quickly. A fan with reduced torque (defective) sometimes will not start or will start slowly.

 

[Dutchman]

A defective connector would appear in the numbers. Simply measure voltage after the connector. If connector is intermittent, then that defect also appears on the meter as a voltage reduction. Observe a defective plug using the meter; even when connector is not causing failures today. Either voltage will always be lower, or the connector is not causing intermittents.

No reason to remove that fan. A tool, ears, and your hand should be sufficient to feel the fan working. A stick can be stuck into an 80 mm fan to stop it. Then see (hear) how quickly it restarts when released. A fan with sufficient torque will start quickly. A fan with reduced torque (defective) sometimes will not start or will start slowly.

Very useful as always, thanks Westom. I actually gave the plug a wiggle and nothing, also measured all yellows, orange and red cables again, still getting similar results to before.
I actually didn't mean I would remove the fan, just meant that my psu is mounted fan down and is at the bottom of the case, so I can't stick something in to stop the fan and see if it resumes spinning okay. I can feel a very faint bit of airflow coming out of the back, may need to see if this picks up under load.
I'm out tonight and working tomorrow but reckon Sunday I'll have a look at the PSU fan.

 

[weird_dave]

Dutch,
Your PSU has variable fan speed, so it may not spin as fast as you would normaly expect.
If the fan is spinning fast, don't shove a stick in to stop it as you can damage it, better to power on with the fan stopped by the stick.
Is there a gap between the fan and the bottom of the case?

 

[james.miller]

You know, this happened to me when i first bought my ATi 4830. constantly shutting down mostly with a blue screen, sometimes a straight power down with no error at all. it wasnt the ram, cpu... i realised it wasnt even the card when i tried another 4830 and it did the same thing.

It actually turned out to be a software problem, more specifically something to do with the bios on the motherboard. between numerous bios reset's and a re-flash, it never did it again and that was a good 10-11 months ago i think.

Don't discount anything, you may be looking in the wrong place. It could weel be the gtx that's shot. I;ve only skimmed this thread and the other, but i dont think the pc shut down on you with the other card in it?

 

[Dutchman]

Dutch,
Your PSU has variable fan speed, so it may not spin as fast as you would normaly expect.
If the fan is spinning fast, don't shove a stick in to stop it as you can damage it, better to power on with the fan stopped by the stick.
Is there a gap between the fan and the bottom of the case?

Thanks, I'll do it that way rather than shoving in something after it's already running
There is a gap between the case and the PSU, haven't taken a photo myself but this one here is similar to my setup and shows the gap:
http://ups.imagup.com/04/1240549330_P1010018.jpg

I'm gonna test my PC intensively today( I'm gonna play games ! ) and do more troubleshooting tomorrow

James, thanks for the suggestion, if I don't get anywhere in next couple of days I will slum it and put the ATI 1950 for a few days