Is my PSU dying?

Hello :)
I have a Corsair TX950 PSU that is about two years old and in that period I have had no end of troubles with my PC. My other components change but I'll start with the original build:

XFX nForce 680i motherboard
Intel Pentium Dual Core E2180 @ 2.00Ghz
GeForce GTX 295

The original problem occurred one day when playing WoW; the mouse started to die. Thinking it was the wire, I replugged it in and it started working. Then the keyboard and mouse died, and soon the PC turned off. I was worried that the CPU could have been running too hot so I left it a while. When I turned it back on, it would last maybe 5 minutes then turn off. Soon the fans would spin a little then it would die. I could swap everything out except the PSU; I changed the motherboard to one with onboard graphics (microATX nForce 610i) and success! It booted. I assumed either the motherboard or the GPU had failed.

About 3 months later, my boyfriends graphics card started to fail in his PC. I suggested he try mine out, and miraculously the 295 worked. The motherboard is still sat up in its box in the loft.

A while later I moved into my own place and wanted to do some gaming now I had a cosy hub to do it in. I decided to try his card in my system (an EVGA GeForce 8800 Ultra). That worked well, too. I overclocked my CPU to 2.2Ghz and for a week there seemed to be no problems - i was monitoring my GPU this whole time, worried that it might overheat from the OC but the temps stayed within normal boundaries even on 100% load. This morning I loaded up SWTOR and went to do my hair, when I came back the PC was off. I tried turning it back on and I got the same pathetic fan spin then nothing. For a few hours I've been thinking I need to replace that graphics card, but now I'm wondering if it could be the PSU. I noticed that the GTX 295 needs way more power than the 8800 to run, so could it be that my PSU is the culprit?

I'm gonna take the graphics card to my boyfriends to try it out in his PC tonight. If it works okay, should I start hassling Corsair for an RMA? :P Is it safe to try the 295 in my PC to test the socket?
6 answers Last reply
More about dying
  1. Tried to edit but apparently I cant :P I meant to say, when I took the 8800 out and let it run onboard graphics, it boots fine
  2. It could be the PSU, but it would be unusual for a Corsair 950 TX to die after 2 yrs of service.

    You seem to be following the correct method of testing, swapping out components to see if one fails. As long as you use the proper ansi-static protocols it should be safe to try the 295 in your rig. If it wasn't killed before by the Corsair PSU, it should not be when it is restored to the original mobo. But I predict that you will see it fail.

    You could do an analysis of both PSUs with 3D Mark Vantage and see if it indicates failure. Look at this article to see how it indicates PSU failure:
  3. Could be the TX but would be unusual for it to die so quickly, more likely the 680i or 610i is suffering heat stroke, as an experiment, take the side off and point a cooling fan at the northbridge chipset.

    Had a 580i that had similar symptons and bolted a small fan to its northbridge and cured its issues.
  4. wow talk about a mismatched system... who ever convinced you to buy a gtx 295 and put it in a system with what is a very budget dual core, doesnt know what there talking about...

    as for your problem... its such a miss match of parts it could be anything. my bet would be an unstable oc... reset the bios via the motherboards jumpers and try rebooting. listen for any clicks or whistling from the cpu area. listen to the rest of the system and check all the wires.. turning the system on and moving the wires may show a brake in a cable which can happen if your constantly swapping out parts.
  5. Heh I kinda upgraded as I could afford to. I'm saving for an i5 but that'll be set back if I need a new graphics card :( Also I hear these CPUs are nifty for OCing. I originally reset the bios as soon as the problem started today but it didn't change anything. My system was running very cool, everything except gpu ~35-45c idle according to HWMonitor.
    I would also suspect the mobo but it is fine running onboard graphics. Unfortunately I can't get access to another PSU or card until tomorrow to test things. I have a multimeter though, let's hope I don't blow myself up. I'll also try using 3D mark vantage, thanks for that.
  6. 35-45 is very cool for an air cooled gpu so thats unlikely to be the issue.
    your psu is off a good brand and will still be under warranty so if it is blown or the rails are dying then just email em and RMA it...
    also put the 295 aside for the moment and just run the system with the 8800 that way 1s you do get it working you wont have such a huge bottleneck choking your system and it will run a lot smoother.
    in the world of pc's there is such a thing as to much power...

    but hey 12 hours isnt a long time just be patient and good luck.
    while your waiting run the paperclip test on yer psu at least that will tell you if its supplying current.
Ask a new question

Read More

Power Supplies Components