1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Strange Problems, 2800+ now dead it seems...

Discussion in 'Hardware' started by Tim S, 2 Feb 2004.

  1. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    Strange Problems

    Right this is a confusing one... I've been having problems with my computer ever since I installed everything into my testbed... I can't work out what it is.

    I have a problem getting the computer to POST on a regular basis.

    Here are my findings so far.

    Memory swapped out: no problems with memory
    No memory: Gives correct error message
    No video card: Gives correct error message
    Changed CPUs: No problems with processors
    Changed Video Cards: No problems with video cards
    2 Monitors connected: refuses to POST
    Sony Monitor Connected: POSTs
    Other monitor connected: No POST (ok here is the first problem... this has now been removed from the equation)

    I wake up today and my screen is blank... much like the problems that mrplow was having a while back, his turned out to be power. I have an enermax PSU at the moment, and have done for a good while... I'm starting to think that this is the problem.

    My lines aren't too bad though
    vCore: 1.94 - 1.97v idle 1.90 - 1.93v load (bios set to 1.975)
    vDIMM: 2.92 - 2.98v
    +3.3v: 3.26 - 3.33v (I have noticed that this has dropped slightly in recent times, it used to sit at 3.33 all day long)
    +5v: 5.00 - 5.03v (used to sit at 5.05v all day long)
    +12v: 11.73 - 11.98v (used to be much steadier in it's fluctuation, fluctuates a lot in recent times)

    Anyone got any ideas at all? :confused:

    This has baffled me for around a week now...
     
  2. bsodmike

    bsodmike Banned

    Joined:
    11 Oct 2002
    Posts:
    763
    Likes Received:
    0
    I'm gonna suspect your GFX or your monitor. What is the GFX card you have? Is this your replacement to the 9600 that died? Hope you can get it sorted m8!
     
  3. yodasarmpit

    yodasarmpit Modder

    Joined:
    27 May 2002
    Posts:
    11,354
    Likes Received:
    205
    You already know the question coming, but have you got another PSU to test with?

    The fluxuations on your lines are very minimal, unless the whole system is on the edge, you are puting a fair bit into the vCore &vDIMM.
     
  4. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    I've had 3 different graphics cards in there... one of them being 3 days old, so I doubt it's the graphics card some how...
     
  5. yodasarmpit

    yodasarmpit Modder

    Joined:
    27 May 2002
    Posts:
    11,354
    Likes Received:
    205
    Forgot youve already fried one GPU on this setup, what voltage are you putting through the AGP?
     
  6. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    I'm gonna try and get a PSU off mrplow later today to stick in, before now my lines have been VERY stable though... no fluctuations, or minimal at best, lines have dropped and the fluctuations have increased, to me that means that the power supply is possibly deciding to end it's life.

    I remember when mrplow had his problems with the enermax after he changed the fans, his lines started buggering about and he got the black screen problem that I woke up to this morning.
     
  7. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    That GPU came back to life... randomly

    my VDDQ is 1.5v, default, any more is bad for your GPU memory :)
     
  8. Will

    Will Beware the judderman...

    Joined:
    16 Jun 2001
    Posts:
    3,057
    Likes Received:
    2
    I'd suggest testing a new PSU as well, just to rule it out. I've seen some funny things with PSUs in my time :duh: such as, almost at random, not letting a system post, so I'd test that as well.

    Have a look at the area of the motherboard around all the voltage regulators and MOSFETs as well (including the back of the board), I do recall someone on the OcUK forums slowly killing the board over time through using too much vcore (1.9v+ through Bartons), basically it was pulling too much power through the board and it slowly died. When he had a look at the board, some of the tracks were discoloured on the back of the board :eeek:. I'm not sure if he had the same symptons as you, but he had no mosfet cooling or anything as was (like you) using pretty high vcores.....
     
  9. mushky

    mushky gimme snails

    Joined:
    24 Mar 2003
    Posts:
    5,755
    Likes Received:
    3
    I'm tempted to say it sounds a bit like there might be a short, so i will...
    "there might be a short".

    there.

    the one variable that gives constant results is.. testbed vs case.
     
  10. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    It's a possibility Will, but since I've moved my board out of a case, I've had an 80mm fan over the mosfets/vreg and clockgen. I'm in the motion of creating some heatsinks for these parts as well as the 80mm fan. I'll whip the board out when I next crash and have a look at the traces. :)

    Edit: Monkeyspank, surely a short would mean a no-POST? but I'm posting from the system here at the moment, so it runs but it doesn't all of the time.
     
  11. bsodmike

    bsodmike Banned

    Joined:
    11 Oct 2002
    Posts:
    763
    Likes Received:
    0
    Enermax are the crap. I've had 2 before *those* days. A 350W which went, bang....and scared the living jesus out of me....and the replacement 450W which almost killed me and did the same bang, with alot more light, sparks and smoke....

    Antec baby :D
     
  12. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    I've had this enermax for erm, I forget how long... from the day I bought a Tbird 1400, which was around October/November 2001.

    It's been great to me over the time, but recently I've noticed it's lines becoming unstable, and I remember when I was installing my Barton 2800+, I was getting even more than normal undervolting on my vCore especially, I was close to testing another PSU then, but withheld for some reason.

    1.65v (bios) was reading 1.61v idle and 1.58v load
     
  13. mushky

    mushky gimme snails

    Joined:
    24 Mar 2003
    Posts:
    5,755
    Likes Received:
    3
    i was thinking there might be an intermittent short. you know, like something flapping against two contacts..
    just a thought. If it was me i would put everything back in the case with baited breath, just to see if that solved it.

    IF its not the psu ;)
     
  14. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    I've had a look inbetween the motherboard tray and the motherboard, and the only thing that has contact with the tray (apart from the standoffs) is the grounded back plate from the graphics card, so shorting can't be a problem... having said that, I've got to remove the motherboard to check the traces to the vregs/mosfets so I'll do that in the same breath. :)
     
  15. mrplow

    mrplow obey the fist!!

    Joined:
    5 Sep 2001
    Posts:
    2,563
    Likes Received:
    23
    Mine was about the same age, and it just kinda stopped working properly..

    Enermax = poop.
     
  16. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    right, I crashed again with the black screen problem about 30 mins ago, maybe an hour... took the motherboard out, and photographed the mosfet area, top and bottom of the board...

    [​IMG]

    [​IMG]

    From what I can see there is no damage whatsoever there. My high vCore seems to have been ok as I've cooled them fairly adequately.

    I've also insulated inbetween the motherboard standoffs and the motherboard, so that can no longer be a problem. :)

    Now it's time to try another PSU and see if that proves any better. :worried:
     
  17. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    right after more investigations, and a long chat with my good friend Isaac, I think the problem has been worked out.

    Code:
    [18:56:26] (IS2): have you used that psu before?
    [18:56:35] (IS2): with that sort of load?
    [18:56:56] (bigZ\brb): yeah for 2 years
    [18:57:08] (IS2): ok, and now you have stability problems?
    [18:57:20] (bigZ\brb): lines are all over the place
    [18:57:51] (bigZ\brb): I posted what I've done here: [url]http://forums.bit-tech.net/showthread.php?t=50856[/url]
    [18:58:10] (bigZ\brb): there's quite a bit of stuff Ive done, diagnostic wise
    [18:58:28] (IS2): I think I know what the problem is
    [18:58:57] (bigZ\brb): what's that?
    [18:59:19] (IS2): rfi
    [18:59:31] (IS2): it's not in a case, you've got no grounded plate under the mobo
    [18:59:46] (IS2): you've got large RFI sources right beneath the mobo
    [18:59:50] (bigZ\brb): right
    [19:00:03] (IS2): the switch mode, hard disks, etc
    [19:00:06] (bigZ\brb): so I need to ground the motherboard tray?
    [19:00:30] (IS2): first of all just try moving the psu and drives out from under the mobo to out to the side
    [19:01:25] (bigZ\brb): the motherboard tray is attached to 2 of the struts
    [19:01:40] (IS2): it's perspex though isn't it?
    [19:01:57] (bigZ\brb): nah, I've got a motherboard tray in there on top of the plexy
    [19:02:32] (IS2): ok
    [19:02:57] (IS2): still try moving the stuff out from underneath to away to the side
    [19:03:14] (IS2): it might sort it... the mobo tray is probably just floating about
    [19:03:15] (bigZ\brb): ok, totally out from under the case?
    [19:03:28] (bigZ\brb): from under the mobo tray even
    [19:03:29] (IS2): what are hte struts attached to electrically?
    [19:03:34] (IS2): yes
    [19:03:34] (bigZ\brb): nothing
    [19:03:37] (IS2): ok
    [19:03:42] (IS2): connect them to the psu casing
    [19:03:48] (IS2): that will ground the mobo tray
    [19:03:51] (bigZ\brb): kk with some wire?
    [19:04:05] (IS2): yes.... in a normal case it's done with the screws that hold it in
    [19:04:16] (IS2): some people had stability probs with the plexy cases for these reasons
    [19:04:50] (bigZ\brb): if I solder some wire to the fan guards on the PSU to the motherboard tray, that would eliminate rfi?
    [19:05:05] (IS2): it will ground the mobo tray, so that will act as a shield, yes
    [19:05:26] (IS2): you are still vulnerable to RFI from the top side of the board, but you've got major rfi sources under the tray
    [19:05:35] (IS2): switch mode psus, motors, clocks
    [19:06:35] (bigZ\brb): righto, I'll try that before I try another psu, as you're pretty good with electronics, being an electronic engineer and all that :)
    [19:07:17] (bigZ\brb): will monitors affect RFI?
    [19:07:36] (IS2): yes
    [19:07:43] (IS2): they ge
    [19:07:48] (bigZ\brb): righto, I'll move them away too
    [19:07:50] (IS2): generate large fields
    [19:07:59] (IS2): but the fields are relatively static
    [19:08:07] (bigZ\brb): kk
    [19:09:04] (bigZ\brb): I'll still move them away
    Could be the problem... time to move the case away from the monitors and time to ground the Motherboard tray to the PSU :D
     
  18. yodasarmpit

    yodasarmpit Modder

    Joined:
    27 May 2002
    Posts:
    11,354
    Likes Received:
    205
    Interference, wouldnt have even occured, not seen Isaac for ages say hi.

    Oh and was that some home made spiral wrap I seen in the above pic.
     
  19. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    yep, that's some home made spiral wrap :)

    I crashed again earlier, I've now grounded the PSU, so we shall see about whether this will help or not overnight... Will post some findings tomorrow morning and then try the other PSU if this doesn't fix the problem.
     
  20. Tim S

    Tim S OG

    Joined:
    8 Nov 2001
    Posts:
    18,881
    Likes Received:
    78
    The grounding the PSU didn't work, and nor did the change of PSU... back to square one, I can't boot at all at the moment and I'm getting rather ****ed off with it all :grr:

    Going to change the design of the case a bit and change the plexy out for some steel, hopefully that should solve the problem of Interferance totally.
     

Share This Page