PC freezing randomly

By Kroenen1897
Dec 20, 2009
Topic Status:
Not open for further replies.
  1. Well guys this one officially has me stumped, after almost 2 months of having problems and replacing almost every major part of hardware on the PC I have come to the conclusion that its got to be something besides RAM, HDD, VC or CPU and I just dont seem to be able to put my finger on it. I have read through several other posts on this forum and found alot of people having similar problems but nothing that directly relates to the one I am having. With all that being said I will see if I can discribe the issue to the best of my ability, ok so what actually is happening is at random times the PC will freeze with no paticular program running, I have had it freeze while running Internet Explorer 7, Call Of Duty 4, CorelDRAW X2, rFactor, Ghostbusters, Batman: Arkham Asylum, Warhammer 40K Dawn Of War II, Windows Media Player, Nero 7 Ultimate, Adobe Photoshop CS2 and in general Windows XP Pro SP3 while on the desktop, I have tried running Memtest for 10 hours with 0 errors, I have also run the stress tests with PassMark Performance Test, Prime95, 3DMark06, FurMark and BurnIN all which run full tests and give accurate readings for the specs of the PC and no errors. Now thats not the part that truely has me stumped, what gets me is when running Speedfan my readings for the -12V and -5V are in the positive and the same can be said for Everest as well with the -12V reading at 1.79V to 1.82V and the -5V reading at 2.36V to 2.56V which I know they should be reading well below that but even with all of the fans and non-essential drives plugged in it is still reading like that, these readings are coming off of a OCZ Stealth 500W PSU and prior to this freezing the stats were reading alot lower for those two pipes, now I know this clearly indicates that the PSU is bad but when I check the voltages with a multimeter they are within normal operating levels which is the part that really floors me, I mean I doubt that the motherboard could have that much draw but damn thats alot of power being sucked out of the PSU with no reason for it...anyone else have any ideas on what it could be?

    Thanks

    KRK
  2. raybay

    raybay TechSpot Evangelist Posts: 10,716   +6

    We will need to know more about your system... Brand and model, or motherboard and configuration.

    Strange problems that sometimes work. Replace the CMOS battery CR2032. Removing the heat sink to remove and replace the thermal past with the thinnest possible layer... as too much thermal paste acts as an insulator.
    Others have had problems with the OCZ Stealth in higher speed units. Works well in normal computer installs, so you might want to try a test with a different power supply that is not OCZ and Not Antec... Perhaps Seasonic, FSP Group, Corsair...
  3. Kroenen1897

    Kroenen1897 Newcomer, in training Topic Starter

    raybay, thanks for the quick response, I have tried replacing the thermal paste with Arctic5 and OCZ Freeze but nothing changes at all, the CPU itself never gets above 58 degrees and the CMOS battery was replaced when I got the board since it was second hand (its a habit I have with wanting everything to be in peak shape...lol), as for the PSU I havent tried replacing it yet but I figure it is either a power issue or the northbridge overheating but I cant be 100% sure.

    I was toying with the idea of pulling the heatsink off the northbridge and replacing the thermal paste on it with the OCZ Freeze as well just to see if that clears up the issue since I did have a video card that was blowing extremely hot air directly onto the northbridge and I would assume that it has more than likely dried up the thermal past that was originally on the bridge.

    Now thats pretty much everything I have done with the exception of mentioning that I have replaced all the major parts as of recently, this is the whole issue is I have replaced all the parts that could be an issue and now I am left looking baffled by whats going on, I guess the last course of action would be to replace the PSU and hope for the best if applying new thermal paste to the CPU and northbridge doesnt work.

    Let me know if replacing thermal paste on a northbridge is a good or bad idea please guys, I dont want to risk losing this board this close to christmas or at all for that matter. Thanks

    KRK
  4. Puiu

    Puiu TechSpot Addict Posts: 1,036   +91

    It's the MB for sure and it's most likely nothing you can about it besides replacing it with a another one. To be sure try another PSU (borrow one if you don't have an old one), but only the MB (and from what i read yours is old and second hand) can make the PC freeze, not the PSU.

    PS: this is only valid only if the CPU, video card and HDD work. try a bios update if one is available. if it isn't then buy a new MB (your present for x-mas)
    PS2: don't buy a second hand one unless the one you buy it from is a friend. (just to be on the safe side)
    PS3: unless you overclocked it the northbridge shouldn't need better thermal paste even with some heat from the video card
  5. raybay

    raybay TechSpot Evangelist Posts: 10,716   +6

    Nothing is ever for sure. Without specifics, how can you say, "...it's the MB fpr sure..."
    Motherboards seldom fail, although they do sometimes fail, of course. And the Northbridge over heating is rare, particularly after what you have already done. Best to try all possibilities before tossing out the board.

    Some suggestions:
    Borrow a power supply from a friend to test... anything over 305 watts will be adequate for the test.
    Consider a wipe of the partition, and then a full reinstall and update or drop back one on the BIOS to test. Be sure your system is not overclocked during these reviews.
    Borrow memory modules to test, even though your memtest came out ok...


    ,
  6. Puiu

    Puiu TechSpot Addict Posts: 1,036   +91

    Considering that he tested everything else besides the MB and PSU i don't think i'm wrong and he has a second hand MB. I've had a lot of problems with models older than 4 years (computer freezing being one of them). But as you said he should always be sure before deciding to buy something.
    Try a few more tests and then give us a report.
  7. raybay

    raybay TechSpot Evangelist Posts: 10,716   +6

    We disagree. When you suggest a change in motherboard, you might suggest why it is the motherboard.

    He still has to test the power supply, he could try a reformat and reinstall of the hard drive, he could replace the memory, or it may be a defective cable component.

    True, motherboard is a possibility, but to say, "...It is the motherboard for sure..." when you have not said why is simply over reaching...
  8. Puiu

    Puiu TechSpot Addict Posts: 1,036   +91

    "to be sure try testing another PSU"
    It wasn't my intention to start an argument.
    I did say that he should test his PSU first. Also i was under the imprecision that he already tried reinstalling a fresh copy of windows and it was him who said that he replaced almost everything (including the ram) and still nothing. Besides testing the PSU and HDD cable there's nothing he can do. Kroenen1897, can you tell us if they work OK? And try to give us more details.
  9. Kroenen1897

    Kroenen1897 Newcomer, in training Topic Starter

    Puiu, raybay is correct in assuming it could be the PSU, esspecially since in both everest and in speedfan the voltages are all over the place for the -12V and -5V and the power for the cores is bouncing all over as well. I had a really funny feeling that the PSU was burnt out for a while but their is the chance that the northbridge could be the cause but that overheating could also be caused from the PSU surging. However I am going to still test the mainboard, PSU and all of the SATA and IDE cables today and I will see how things go from their.

    KRK
  10. Kroenen1897

    Kroenen1897 Newcomer, in training Topic Starter

    Ok after 5 hours of the system running stable I finally got a freeze when trying to compress a zip file, unfortunately it is looking more and more like the mainboard but then again I am running the system off a 350W PSU that wasnt tested prior to being put into the PC. Hopefully someone will post on or offer me one for a reasonable price in the city I live in and that way I can get back up and running without the freezing.

    KRK
  11. Puiu

    Puiu TechSpot Addict Posts: 1,036   +91

    All that remains is to test the components (including CPU) with another MB. Hopefully you'll fins someone to help you with it.
    At my workplace it took me hours just to test the CPU (without a stress test, just to see if the PC powers up) by placing it into another computer.

    About the 350W PSU, try removing everything (extra fans, cd-rom, etc) and don't do any graphics intensive tests which use the VC card then try it again. Maybe you'll get lucky.
     
  12. duffskeeze

    duffskeeze Newcomer, in training

    Possibly Ram or any number of things, lol

    IF the ram is not new, swap it. Never trust a ram test... ever! But from the details of your post it looks like the only original hardware on your machine is the mobo, power supply, optical drives, case, and various cards. Remove all of the cards from the system. Upgrade the bios on the mobo if you haven't. Since you've tried another power supply but do not know its working condition you may want to try another. Doubtful its anything to do with the optical drives (very very doubtful) but disconnect them anyway. If you've tried all of this and I'm correct in the new/ old parts listing, it's the motherboard. Remember... it's not that hard people just over-tech things :)
  13. Kroenen1897

    Kroenen1897 Newcomer, in training Topic Starter

    I think it has come down to being almost 100% sure that it is the motherboard, I have tested the CPU and it has always come back as being a bit dogish but nothing failing on any of the tests. When running yesterday I only had the MB, VC, DVD and HDD plugged in with the side of the case removed but yet I did get the exact same results still with no errors in event log and no signs of overheating or starving for power so I guess we can almost rule out the PSU as being a possible cause unless something was totally overlooked. I am going to try and run it today without doing anything that will put major strain on the PSU and see how things go, if it locks up like I think it will I am going to be forced to buy a new 939 motherboard if I can even find one with a AGP slot on it and if not I may have to sell this new VC I got which I would rather not do. Wish me luck though guys!

    KRK
  14. Puiu

    Puiu TechSpot Addict Posts: 1,036   +91

    Since it's a 939 MB i think you'll need to buy it SH again. Buy the cheapest working one you can find and don't forget to test it first. There's no need for a brand name MB for something this old. It either works or it doesn't.
  15. Kroenen1897

    Kroenen1897 Newcomer, in training Topic Starter

    I'm actually looking for a AM2+ or 775 board now with AGP on it, I found a few on ebay but until after Christmas things are looking grim, if all else fails I think I might have to go with another ASRock board until I can bank the money to buy a new rig but it will basically only be a bandaid until then. Really crappy thing to have happen prior to Christmas but what can we do...lol
Topic Status:
Not open for further replies.


Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...


Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.