Can a dead CMOS battery stop POST?

Discussion in 'DIY Computers' started by Jon Parker, Jan 2, 2010.

  1. Jon Parker

    Jon Parker Guest

    My system has stopped POSTing.

    Video is fine, it just stops at the BIOS screen where it invites you to
    press DEL to enter setup, and the processor is now reported at 1050MHz
    instead of 2.1GHz it used to.

    I've re-seated memory, no effect. Video card re-seated just for
    avoidance of doubt. No bulging caps visible.

    Does this sound like a dead CMOS battery?

    I don't have any spares so I'm going ot get one tomorrow, but I'd be
    grateful for some other opinions.

    It's a home built machine, board is an Asus A7N8X-E Deluxe, and is about
    5-6 years old. Is that about right for a CMOS battery to last?


    --
    Regards
    Jon
     
    Jon Parker, Jan 2, 2010
    #1
    1. Advertisements

  2. On Sat, 2 Jan 2010 19:25:30 -0000, Jon Parker
    <> wrote:

    >My system has stopped POSTing.
    >
    >Video is fine, it just stops at the BIOS screen where it invites you to
    >press DEL to enter setup, and the processor is now reported at 1050MHz
    >instead of 2.1GHz it used to.
    >
    >I've re-seated memory, no effect. Video card re-seated just for
    >avoidance of doubt. No bulging caps visible.
    >
    >Does this sound like a dead CMOS battery?


    Could be, leading to BIOS settings corruption. Have you tried
    clearing the CMOS/setting the BIOS settings to default? That would fix
    it temporarily if so.

    >I don't have any spares so I'm going ot get one tomorrow, but I'd be
    >grateful for some other opinions.
    >
    >It's a home built machine, board is an Asus A7N8X-E Deluxe, and is about
    >5-6 years old. Is that about right for a CMOS battery to last?


    Could be. I've seen them go dead after 3 years, and keep going for
    over 10.

    Cheers - Jaimie
    --
    "I went to a planet where the dominant lifeform had no bilateral symmetry,
    and all I got was this stupid F-Shirt." -- Eric Pivnik
     
    Jaimie Vandenbergh, Jan 2, 2010
    #2
    1. Advertisements

  3. Jon Parker

    Jon Parker Guest

    In article <>,
    says...
    > Could be, leading to BIOS settings corruption. Have you tried
    > clearing the CMOS/setting the BIOS settings to default? That would fix
    > it temporarily if so.


    Would rather try a replacement first, as I will lose all my overclock
    settings. I can't access setup so would have to do this by the jumpers
    too.

    --
    Regards
    Jon
     
    Jon Parker, Jan 2, 2010
    #3
  4. The message <-september.org>
    from Jon Parker <> contains these words:

    > My system has stopped POSTing.


    > Video is fine, it just stops at the BIOS screen where it invites you to
    > press DEL to enter setup, and the processor is now reported at 1050MHz
    > instead of 2.1GHz it used to.


    That looks supsiciously like you're using an Athlon XP2100 cpu which
    requires an FSB of 133 (266DDR)MHz (real clock speed of 1400MHz) and is
    now being misreported as an Athlon clocked at 1050MHz (100MHz x 10.5).

    > I've re-seated memory, no effect. Video card re-seated just for
    > avoidance of doubt. No bulging caps visible.


    It's always good idea to reseat socketed items (and check the caps) in
    a 6 year old system box but these aren't the root cause of your problem.

    > Does this sound like a dead CMOS battery?


    It most certainly does! However, similar symptoms can arise if the coin
    cell holder contacts become high resistance. If this has happened,
    removing the coin cell and cleaning it with a lint free cloth and
    reinstating it (rotating it half a turn back and forth in its holder)
    usually clears this problem.

    > I don't have any spares so I'm going ot get one tomorrow, but I'd be
    > grateful for some other opinions.


    You, quite obviously, don't possess a voltmeter. This is a pity since
    you could have checked the voltage. New (and even used ones) will read
    about 3.2v +/- 0.1v, getting down to 2.9v or lower when close to
    exhaustion but not yet giving any symptoms.

    If you're in the habit of switching the mains power on and off with use
    (as the fact that you're now seeing the classic symptoms of low cmos
    'battery' suggests), the first hint of trouble is usually manifest by
    the RTC losing time (coin cell voltage less than about 2.5v) followed by
    corruption of the RTC registers (the 'cmos memory' where the bios
    settings are stored) when the voltage dips below the 2v mark.

    > It's a home built machine, board is an Asus A7N8X-E Deluxe, and is about
    > 5-6 years old. Is that about right for a CMOS battery to last?


    It does rather depend on how much current the "RTC Chip" on the MoBo
    draws (as well as whether you shut off the mains power between each time
    you use the machine). With PC Chips MoBos, the coin cells needed to be
    replaced every 18 to 24 months. In the case of most other manufacturers
    (as with my Jetway V600DAP socket A board), they'll last 5 years
    plus[1].

    The usual indication of low cmos battery is a POST time error message
    which says exactly that, typically:

    "CMOS checksum error; cmos battery low. Press F1 to continue or DEL to
    enter cmos configuration"

    The exact wording of the above will vary between different makes,
    depending on the hotkey chosen to initiate a cmos setup (probably F2
    instead of DEL for an Asus board).

    If you're getting this symptom (if not such an explicit error message -
    the fact that the FSB has reverted to 100MHz is enough evidence of
    such), you'll have already lost all your overclocking settings so have
    nothing to lose by removing the coin cell (extremely unlikely to be
    anything other than a CR2032 BTW ;-).

    [1] I recently had low cmos battery symptoms on this V600DAP system
    (after nearly 6 years use) in the form of lost time between sessions (I
    power the mains off between sessions). To my surprise, after using the
    "Don't switch the mains off" workaround for a fortnight or so, the coin
    cell was still showing 3.12v so, quite patently, wasn't the culprit
    which only left a high resistance contact (battery holder or cmos clear
    jumper contacts) as a possible cause.

    The reason for the protracted 'leave the mains on' workaround was that
    I was also aware of the need to replace some of the MoBo capacitors (or
    else repeat my 5 yearly MoBo/CPU/Memory upgrade cycle) and needed a
    window of opportunity in which to apply the remedial work.

    When I'd finished replacing the caps and reassembled it, I was hit by
    that other bugbear, which seems common to a lot of Jetway boards that
    have socketed PLCC bios rom chips, which stopped the machine from
    booting (and left me in a bit of a panic until I recalled this
    particular peculiarity).

    After regaining my composure and lightly tapping the bios rom chip, the
    machine rebooted with cmos settings intact despite the swift coin cell
    swap out.

    At this point, I recalled having purchased a suitable PLCC extractor
    tool and reseated said chip after cleaning its contacting legs with an
    indiarubber erasor which, hopefully, should permanently cure the
    original ("It isn't going to reboot without powering it up and down a
    few times, or lightly tapping the case to 'dislodge the high resistance
    contact') problem.

    The 'bad contact' issue is pretty common with contacts of the PCI slot
    style (and, seemingly, with the PLCC 'socket' for a chip that was
    primarily designed for surface mount soldering to a board rather than
    for a 'socket'). This issue can be almost entirely eliminated if the
    parts are reseated a few times on initial assembly to establish a more
    reliable contact through the self cleaning rubbing action between the
    contacts.

    You can take the above as a "Heads Up" as to why a system will
    sometimes fail to post (if it has a socketed PLCC BIOS ROM chip ;-). The
    reason as to why the symptom of such a fault only appears at boot time
    is that once the OS has booted, the rom services become redundent (and
    even in the case of msdos, the bios rom is usually copied into ram as a
    'shadow copy' unless this cmos default has been disabled). IOW, the
    intermittent contact only has to remain good for the few seconds it
    takes between power up and the time it gets to copying the bios rom into
    shadow memory.

    HTH

    --
    Regards, John.

    Please remove the "ohggcyht" before replying.
    The address has been munged to reject Spam-bots.
     
    Johnny B Good, Jan 3, 2010
    #4
  5. On Sat, 2 Jan 2010 19:46:17 -0000, Jon Parker
    <> wrote:

    >In article <>,
    > says...
    >> Could be, leading to BIOS settings corruption. Have you tried
    >> clearing the CMOS/setting the BIOS settings to default? That would fix
    >> it temporarily if so.

    >
    >Would rather try a replacement first, as I will lose all my overclock
    >settings.


    Odds are high that they've gone already - the BIOS is not running your
    CPU at the correct settings, let alone your overclocked ones. Did you
    not note down the speed+multiplier+voltages you were using? I used to
    pop them onto a post-it note inside the case, since they're bound to
    get lost sometime - on the next BIOS update, for instance.

    Plus, on a six year old machine, is it really worth gaining 10% speed?
    You could probably get a faster CPU off eBay for a tenner used.

    Cheers - Jaimie
    --
    Do not meddle in the affairs of wizards, even if you are soggy
    and hard to light.
     
    Jaimie Vandenbergh, Jan 3, 2010
    #5
  6. Jon Parker

    Rob Morley Guest

    On Sun, 03 Jan 2010 10:18:54 +0000
    Jaimie Vandenbergh <> wrote:

    > Plus, on a six year old machine, is it really worth gaining 10% speed?
    > You could probably get a faster CPU off eBay for a tenner used.
    >

    Indeed, I just got a XPM2500 for a tenner - remember when that was the
    ultimate overclocking chip back in the day? :)
     
    Rob Morley, Jan 3, 2010
    #6
  7. Jon Parker

    Jon Guest

    In article <O2X%m.31508$2>, ~~@...madmail says...
    > If you keep turning it on you may keep enough power in it to retain some
    > settings, but to save all farting about, just buy a new battery, plop it in,
    > set your bios back up again and see what happens. If its not the battery at
    > least its only cost you ~£2 and not hours of messing about to get to the
    > same point.


    Thanks for everyones replies. Turns out the mobo went pop after all.
    After much faffinf about and transplantation I'm no wusing a different
    board and processor with the old memory and HDDs. So far no probs, apart
    from a re-install.

    Regards
    Jon
     
    Jon, Jan 6, 2010
    #7
    1. Advertisements

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

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Sandoo

    CMOS Battery

    Sandoo, Nov 2, 2003, in forum: Hardware
    Replies:
    1
    Views:
    1,095
    James
    Nov 2, 2003
  2. Jon

    CMOS battery - are they all the same?

    Jon, Dec 23, 2008, in forum: DIY Computers
    Replies:
    1
    Views:
    1,633
    Adrian C
    Dec 23, 2008
  3. Stuart Crow

    CMOS battery in NEC laptop

    Stuart Crow, Jan 15, 2009, in forum: PC Technical Talk
    Replies:
    1
    Views:
    2,323
    Adrian C
    Jan 15, 2009
  4. Data888

    Computer CMOS battery

    Data888, May 5, 2004, in forum: Computing
    Replies:
    3
    Views:
    454
    Bill Ruys
    May 7, 2004
  5. spodosaurus
    Replies:
    6
    Views:
    341
    spodosaurus
    Dec 27, 2008
Loading...

Share This Page