NF7-S & 3000/400 Barton Random crashes

Discussion in 'Abit' started by John, Nov 14, 2003.

  1. John

    John Guest

    I have some symptoms and am looking for opinions.

    I am running a Abit NF7-7 wit the AMD 3000/400mhz Barton. Everything in the
    system is 3 months old. Has run flawlessly until the other day.

    First thing it did was did not want to boot when it was cold. I let it warm
    up for 5 minutes and it ran fine. I let it run 24/7 for a week and thne I
    start to get a few random reboots which occured regardless of what was

    Today it did the random reboot but would not start back up. It just crashed
    ( no screen) and then would not respond. When attempting to boot up it would
    not beep/ chirp or give any signals. All drives would spin up and the CDROM
    lights would lite like it was going throught the boot process.
    After letting it sit for a while I tried again. It started up fine and ran
    for a minute and the crashed as before only now it would try to reboot and
    fail and then just continue the loop of boot failures on its own.

    I reseated the cards and checked the memory etc and restarted it. It
    started to boot and stopped with the message that "the CPU was unusable or
    had changed- check the CMOS". I did and it was as before. I notice on the
    next atempt that it was idetifing as an AMD 050Mhz when I swear befroe it
    said 3000.

    I turned of the power and reset theCMOS jumper. But this did not change

    I let it sit without power for an hour and retried it. It seem to be running
    fine until I tried a CPU benchmark program on it. It crashed immediately.

    Temps are normal.

    Any opinions?

    Looks like the CPU to me but was wondering what others thought.

    John, Nov 14, 2003
    1. Advertisements

  2. John

    - HAL9000 Guest

    The first thing I would verify is the power supply. The second thing
    I would check is the mobo.

    It is unlikely to be either the cpu or ram as they normally work best
    when cold - which is the opposite characteristic you described.

    Devices that use CMOS transistors, such as the cpu and ram, work best
    when cold. Devices that use Bipolar transistors, such as those
    contained in your power supply, work best when warm.


    Motherboard Help By HAL web site:
    - HAL9000, Nov 14, 2003
    1. Advertisements

  3. John

    Wblane Guest

    This depends on how the BJT is biased. Any circuit appreciably affected by
    temperature would be a bad design and unstable since the q-point would vary
    w/temp. In an emitter-biased BJT circuit the emitter current is essentially
    independent of Bdc and Vbe as long as Re >> Rb/Bdc so temp variations are

    -Bill (remove "botizer" to reply via email)
    Wblane, Nov 22, 2003
  4. John

    - HAL9000 Guest


    Your assuming that the transistor(s) are fully functional. When
    transistors have partially failed the gain or Beta of a transistor
    will often very greatly with temperature. Much more so then when it
    is fully functional or new.


    Motherboard Help By HAL web site:
    - HAL9000, Nov 22, 2003
  5. John

    Wblane Guest

    Obviously, but a properly biased BJT circuit is no more sensitive to
    temperature variation than a CMOS circuit so it doesn't work "better" when
    warm. If a BJT is driven into saturation, normal room temperature variation
    will not produce any significant increases in gain.

    -Bill (remove "botizer" to reply via email)
    Wblane, Nov 24, 2003
    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.