I recently experienced some strange behavior from my system and I wanted to get your opinions on what it could be. First off, dem specs:

  • ASUS P5Q Pro Mobo
  • PC Power and Cooling S75QB 750W
  • Intel Core 2 Duo E8400 Wolfdale Dual-Core 3.0GHz (OC to 3.9GHz)
  • ARCTIC Freezer 7 Pro (w/Artic Silver 5)
  • 2x ASUS 1GD5 Radeon HD 5850 (Cypress Pro) 1GB 256-Bit GDDR5 (X-fire, Stock speeds)
  • G.SKILL 8GB (4 x 2GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400)
  • Rosewill THOR V2 Full ATX
  • Kingston HyperX 3K 120g SSD (system files, EVE)
  • Some 500mb SATA 7200rpm HDD I had lying around (mostly media storage)

So, a fairly old system I've been upgrading incrementally over the years. The CPU OC has been running the entirety of it's life thus far and is a rock solid champ. I use this system mostly for media playback and gaming (primarily EVE, MOBAs, SC2, Elder Scrolls/Fallout etc) and for the most part it has been extremely solid with admirable performance for it's age/cost.

About a week ago I experienced a series of random hard shut downs- No BSOD, no error messages, just straight to black screen. I was running 2 clients of EVE on separate screens, one on each GPU (I was not running them in x-fire at that point cause EVE doesn't like it that way). I start everything back up and go right back to where I was, figuring it was simply a fluke. Within 5 minutes of dual-screening the same thing happens... I get out the dust off and give my case a good clean, then go back to what I was doing. It happens 3-4 more times, until I move down to a single client on a single GPU. Throughout this ordeal I've set up all my temp monitors and everything seems nice and frosty at all times. The fan on my PSU is working normally as far as I can tell.

I've since replaced my old 5850s (suspect PSU issues? clearly it's time to upgrade your graphics!) with a single MSI gtx 760. I figured it would be less of a draw on my PSU, and in general I prefer a single card situation over sli/crossfire. I haven't experienced the issue since getting the new card, no matter how hard I stress the system with prime95/furmark (though I only ran them on full for ~3 hours). Once again, temps were plenty acceptable, if not even a little chilly, for the workload.

However, I noticed a fairly quiet buzz/clicking emanating from the PSU yesterday, and realized it was something that I've heard in the past but always assumed was my CPU fan...which is currently in something of a 'custom' configuration involving case fans attached to my cooling tower via hair ties which occasionally come into contact with the fan blades. Anyway,with my current sensitivity to potential PSU issues I checked to be sure and indeed the noise it coming from the PSU- it subsided after about 5 minutes.

Thanks for reading- I appreciate any suggestions you might have for me.

  • It really isnt clear what your asking? It does most sound like you had a PSU issue to begin with, you reduced the power draw by lots, and now you might still have a psu problem, which could be just a fan in need of replacement or cleaning? How many years has the psu been in operation (calculating the ON time)? Has it ever been cleaned out? They are not usually the easier things to dust off, but they come apart easy, and will clean up like anything else. (you just watch out for charges on the high voltage caps) – Psycogeek Jun 24 '14 at 4:28
  • I suppose what I'm asking is what caused the cluster of shutdowns I experienced the other day. I understand these problems can be nebulous, so I tried to provide as much context as I could. I bought the PSU in 2008- quite a while ago no doubt- and it has seen many hours. Is there any way to check if a PSU is going bad other than to wait and see? – user338053 Jun 24 '14 at 4:33
  • There are other things you could check while your in there. the high powered mosfet things and the heat sink connection to them. It is not logical how the screw mounting for them comes undone, but it can, the large ammount of power even for a solid state component hard switching, can make noises. Also The big coils can rattle more over time as they get loosened up, and anything else you can spot while your in there. The mosfets not being cooled could even reduce the output ability by large ammounts over time, themal compound and all that stuff. 2008 Capacitors also Age. – Psycogeek Jun 24 '14 at 4:35
  • I'm also starting to wonder if it may have had something to do with my second 5850... it never ran at the factory specified settings well, and always with more heat than the other. Also, the problem stopped as soon as I stopped using that card. – user338053 Jun 24 '14 at 4:36
  • the 5850 wasnt it a brute force :-) kind of GPU, used a lot of power for what it did and blew a lot of heat in the process. – Psycogeek Jun 24 '14 at 4:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.