RAM is Being Physically Damaged Over & Over Again.

Discussion in 'Asus' started by Joe Near Portland, OR. USA, Nov 24, 2006.

  1. My current computer problem is as follows.
    I have a Pentium 4 processor at 2.4 GHz currently running one gig of
    RAM (Patriot).
    I have replaced computer case, power supply and video card.
    I have returned the motherboard five times in the past four months due
    to the fact that every time I get the replacement motherboard and
    install it and to the new computer case and turn it on within the
    first five to six hours of operation it cooks the RAM. or better words
    destroys RAM.
    This has happened six times & nobody can tell me what's wrong or what
    is causing this problem, I Hope someone out there can help? Or at
    least lead me in the right direction as to what is wrong and what I
    need to do to solve this problem.

    The motherboard I am using is an ASUS P4P800SE socket 478.
    & a SoundBlaster live 24 bit sound card
    & a promise ultra ATA / 133 (TX2) IDE Controller so that I can have up
    to eight IDE devices.
    Four on the motherboard and four on the IDE Controller card.


    This is for my Brother, i Have asked him for More Detail's on the
    Hardware & More Detail's on What Happened & When?

    The End result's have Been His RAM is Being Physically Damaged Over &
    Over Again.
     
    Joe Near Portland, OR. USA, Nov 24, 2006
    #1
    1. Advertisements

  2. Joe Near Portland, OR. USA

    Outback Jon Guest

    I'd suspect that the power supply is to blame. Sounds like it is giving
    too much voltage somewhere.


    --
    "Outback" Jon - KC2BNE

    AMD Opteron 146 (@2.8) and 6.1 GHz of other AMD power...
    http://folding.stanford.edu - got folding? Team 48435

    2006 ZG1000A Concours "Blueline" COG# 7385 CDA# 0157
    1980 CB750F SuperSport "CoolerKing"
     
    Outback Jon, Nov 24, 2006
    #2
    1. Advertisements

  3. Joe Near Portland, OR. USA

    Paul Guest

    When you say it cooks the RAM, does the RAM get extremely hot ? Is
    the RAM being verified in another computer, after the failure ?
    Does the motherboard also fail at this point, or can you plug
    more RAM in it and get another six hours from it ?

    Also, are the named PCI cards always present when it happens ?

    It sounds like you've replaced virtually everything, and then
    stuck the same two PCI cards into the system. Perhaps it has
    something to do with the PCI cards. Try your next test, without
    the PCI cards present.

    Also, has the computer case been inspected for its standoffs ?
    A brass standoff should only be installed, where there is a
    matching tin-plated hole in the motherboard. Make sure any
    standoffs that don't line up with a plated hole on the
    motherboard, have been unscrewed and removed. Standoffs have
    been known to short to traces on the bottom of the
    motherboard.

    A minimal test config would be motherboard, CPU, video card,
    RAM, and floppy drive. Connect the monitor, then boot the
    system with a memtest86+ floppy from memtest.org . See how long
    it lasts with that config (and no PCI cards).

    In terms of voltages, the power supply has its output rails,
    but a significant part of the motherboard is powered by
    regulator circuits located on the motherboard. So at least
    some things are isolated from power supply problems. But if
    one rail is bridged to another, there is little some of these
    regulator circuits can do about that. Many of the regulators
    are designed to "push" but not "pull". If a higher voltage
    rail is bridged to a lower voltage rail, the regulator may not
    say a word. Motherboard regulators are not protected against
    overvoltage/undervoltage. In fact, some of the Asus motherboard
    regulators consist of an opamp and a pass transistor, and
    they use components like that, because they are cheaper than
    using a proper regulator chip that will shutdown when there is
    a problem.

    I would look at the edge connector of the two PCI cards,
    and see if any of the pins are burned or discolored. The
    flow of current might not be enough, to leave a trace,
    but you can have a look anyway. You might get lucky.
    Check out the two pictures at the bottom of this page,
    for inspiration.

    http://www.xtremesystems.org/forums/showthread.php?p=1555484

    Paul
     
    Paul, Nov 24, 2006
    #3
  4. Joe Near Portland, OR. USA

    DaveW Guest

    With that many drives in the system, it is VERY possible that if he is not
    using a HIGH QUALITY, HIGH POWER output power supply that the RAM is being
    fried when the power supply tries to supply enough current to drive all
    those harddrives. An inexpensive and/or underpowered PSU will become VERY
    unstable in its voltage and current output if stressed overly and will ruin
    delicate components like RAM.
     
    DaveW, Nov 25, 2006
    #4
  5. Joe Near Portland, OR. USA

    Leythos Guest

    I agree, with 8 drives and all the rest, at least a 480W PSU should be
    installed.

    One other poster suggested the standoffs be checked.

    As RAM power is furnished by the motherboard, there are only a couple
    things that can be the problem:

    1) Standoffs/mechanical shorting/rubbing
    2) Bad Power supply or bad quality power

    Since nothing else can reach the RAM itself, those are the only two
    things that can impact the RAM directly.
     
    Leythos, Nov 25, 2006
    #5
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.