KT7A, Video card, and power....

Discussion in 'Abit' started by Robert Usdin, Dec 30, 2006.

  1. Robert  Usdin

    Robert Usdin Guest

    Hi folks,

    I have a KT7A with a T-bird 1.4 athlon. Have a 300 Watt Power supply,
    USB 2.0 PCI card and a wireless card in the PCI slots. Also have an
    old Soundblaster AWE64 (t's in the last slot - is that ISA - I can't
    remember right now). Have one hard drive in there (though I plan on
    addking another in a couple of days) Anyway - Was using an old 16 MB
    ATI All in Wonder AGP, got tired of not being able to run at least some
    games, so I bought a BFG 3D Fuzion GeForce MX4000 - 128MB. Installed
    yesterday without a hitch, worked great all night. This morning go to
    boot up the PC - it won't boot. 3 beeps from the PC -
    beeeeeeep..beep-beep. Power off, and on again, same thing. Power off
    and on, won't power up.

    Pull new video card, put back old one, boots up like before.

    Questions: What does the beeping mean? Do I need a bigger power
    supply?

    I have replaced the power supply once before - I originally had a 250
    Watt one in there.

    I can go up to a 350 if that will help. ANyone? Or did I just luck
    out briefly with the card working and the card is not going to cut it?

    --*Rob
     
    Robert Usdin, Dec 30, 2006
    #1
    1. Advertising

  2. Robert Usdin wrote:
    |
    | Hi folks,
    |
    | I have a KT7A with a T-bird 1.4 athlon. Have a 300 Watt Power
    | supply, USB 2.0 PCI card and a wireless card in the PCI slots.
    | Also have an old Soundblaster AWE64 (t's in the last slot - is
    | that ISA - I can't remember right now). Have one hard drive in
    | there (though I plan on addking another in a couple of days)
    | Anyway - Was using an old 16 MB ATI All in Wonder AGP, got
    | tired of not being able to run at least some games, so I bought
    | a BFG 3D Fuzion GeForce MX4000 - 128MB. Installed
    | yesterday without a hitch, worked great all night. This morning
    | go to boot up the PC - it won't boot. 3 beeps from the PC -
    | beeeeeeep..beep-beep. Power off, and on again, same thing.
    | Power off and on, won't power up.
    |
    | Pull new video card, put back old one, boots up like before.
    |
    | Questions: What does the beeping mean? Do I need a bigger
    | power supply?
    |
    | I have replaced the power supply once before - I originally had a
    | 250 Watt one in there.
    |
    | I can go up to a 350 if that will help. ANyone? Or did I just luck
    | out briefly with the card working and the card is not going to cut it?
    |
    | *Rob
    |

    Hi Rob -

    You may have tipped your PSU over the edge. The computer was probably
    "warm" when you installed the new card (having been run for some time just
    before shutting down and installing the new card) and now it's cold first
    thing in the morning. "Cold" generally makes heavier demands on the PSU to
    spin/warm everybody up to POST the system.

    Fist .. "The List"

    Basic Beep Codes for ABIT AWARD BIOS Motherboards

    No beep at all - this means your motherboard is dead, either due to a
    defective or underpowered power supply, poorly seated CPU or RAM, or a
    dead-on-arrival board

    1 short (Beep) System booting is normally

    1 long beep then machine shuts down - faulty, improperly installed or
    missing CPU

    2 short (Beep) CMOS setting error

    1 long - 1 short (Beep) DRAM ERROR

    1 long - 2 short (Beep) Display card or monitor connected error

    1 long - 3 short (Beep) Keyboard Error

    1 long - 9 short (Beep) ROM Error

    Long (Beep) continuous - DRAM isn't inserted correctly

    Short (Beep) continuous - POWER supply has a problem

    A two-tone siren - generally caused by overheating or out of specification
    voltages

    Four beeps then machine shuts down - this is because this version of the
    BIOS will shut down your machine if no fan tachometer signal is detected on
    the fan header. Make sure you attach a fan to the CPUFAN header or clear
    CMOS to reset to default (no checking).

    ____

    One long and two short beeps generally means you have no display card
    installed or no monitor could be detected.

    Could be the PSU. Test with your 350 watt PSU and see if all is okie-dokie.
    The fact that it comes back to life with the original card whispers this to
    me .. so I'd start here.

    If the above doesn't work, could be you have a bad video card. Got another
    computer where you can test? Install the card and try to boot into Safe
    Mode, just to confirm that the card makes video. I suggest Safe Mode so you
    don't have to muck around with drivers in the other computer .. start
    tapping F8 just before you see the Windows banner page to get the boot menu.
    Safe Mode with your arrow keys.

    Another possibility: The KT7A is of the vintage that suffered from the "Bad
    Caps" problem. A bad electrolyte formula was prevalent during this period
    that affected most manufacturers (there was a class-action settlement agreed
    to by Abit, that has now closed from claims, where boards were made good ..
    repaired/replaced free of charge). Look at the capacitors on the
    motherboard, especially those around the CPU. If any are domed on the top
    or are leaking a brownish/reddish crud on the motherboard, you have been bit
    by "Bad Caps". In order to continue using the board you will need to have
    it repaired. If you're in the United States or Canada, Abit will repair
    your board for US$25.00 (see the RMA page at http://www.abit-usa.com).

    There are other resources for repair as well. Homie®, who frequented this
    group for many years (but has gone missing from here for some time) runs
    http://www.motherboardrepair.com/. Also see http://www.badcaps.net.

    ____

    To answer your first question: yes, your sound card is installed in one of
    the two ISA slots on your motherboard.

    Good luck!

    Jef
     
    Bird Janitor®, Dec 30, 2006
    #2
    1. Advertising

  3. Robert  Usdin

    Exray Guest

    "Bird Janitor®" <> wrote in
    message

    <Huge Snip>

    Bird, your post was informative (I trust) for the original poster, also for
    me.

    Turns out that a no beep condition doesn't necessarily mean death, it just
    means that one little bitty vitally critical thing is not right.

    My KT7AR currently won't POST at all...no beep...until I clear the CMOS,
    fail on the cold boot, and then hit the reset button. (thank goodness for
    reset buttons). I was mucking with a pretty good (and extremely dusty) XP
    1900+ board, and now all I can get out of it is 6 x 100. The 2200 uFd
    capacitors look great, some of the smaller ones, the 10 u at 10 V look a
    little bit wavy on top. (I have an old KT7 on which two of the 2200s have
    domes on them, but the smaller caps all look perfect).

    While I was putting the machine back together, before I realized I did
    something weird to the mobo during the annual removal of dust, I eventually
    noticed that I wasn't seating the AGP card well. The card and one of the
    handles on the memory slot closest to the CPU were interfering unless I
    pushed that handle to the "loaded" position, although that slot is empty.

    It's certainly possible that the OPs problem might be solved with a guttier
    power supply than the one he's using, but if he goes out and buys a larger
    supply, he should verify the store's return policy first before handing over
    his payment.

    By the way...only one ISA slot on my version 1.0.
     
    Exray, Dec 30, 2006
    #3
  4. Exray wrote:
    |
    | <Huge Snip>
    |
    | Bird, your post was informative (I trust) for the original poster, also
    | for me.
    |
    | Turns out that a no beep condition doesn't necessarily mean death,
    | it just means that one little bitty vitally critical thing is not right.
    |
    | My KT7AR currently won't POST at all...no beep...until I clear the
    | CMOS, fail on the cold boot, and then hit the reset button. (thank
    | goodness for reset buttons). I was mucking with a pretty good (and
    | extremely dusty) XP 1900+ board, and now all I can get out of it is 6
    | x 100. The 2200 uFd capacitors look great, some of the smaller ones,
    | the 10 u at 10 V look a little bit wavy on top. (I have an old KT7 on
    | which two of the 2200s have domes on them, but the smaller caps all
    | look perfect).
    |
    | While I was putting the machine back together, before I realized I
    | did something weird to the mobo during the annual removal of dust, I
    | eventually noticed that I wasn't seating the AGP card well. The card
    | and one of the handles on the memory slot closest to the CPU were
    | interfering unless I pushed that handle to the "loaded" position, although
    | that slot is empty.
    |
    | It's certainly possible that the OPs problem might be solved with a
    | guttier power supply than the one he's using, but if he goes out and
    | buys a larger supply, he should verify the store's return policy first
    | before handing over his payment.
    |
    | By the way...only one ISA slot on my version 1.0.
    |

    Let's be clear first .. I wasn't suggesting that Rob go out and buy a new
    PSU. He said he had a 350 watt unit to try on it and I agreed that this
    would be the best first step. I'd hate to be responsible for needless
    spending (unless it was on me!).

    You say you had a 300 watt V-Tech in the machine in your other post (not the
    prettiest crayon in the box when it comes to good quality PSU's). Who makes
    the 480 watt unit in there now? There are a lot of PSU's out there,
    regardless of maximum rated wattage, that serve better to stop the door than
    they do to power a computer.

    For some reason you seem to be defaulting to failsafe defaults when you
    finally do get the machine to POST with the reset switch. Are you able to
    go into the BIOS and make changes (Load Optimized Defaults) then save and
    exit, and have the changes still present when you reboot? Obviously this
    ends up a round-trip on the next cold boot.

    I say all the former because you may have your CCMOS1 jumper set on pins 2
    and 3 (clear CMOS) instead of 1 and 2 (retain CMOS with backup power or
    motherboard battery when PSU is switched off).

    I see you've replaced your motherboard battery. Have you metered the
    replacement? Could be dead out of the package.

    Another possibility is that you have a short between your board and the
    case. If you're into tear-down testing, try the board out of the case on a
    non-conductive surface. Connect up only PSU, video card, RAM, CPU, HSF,
    monitor and keyboard. Start the board by turning on the PSU then
    momentarily shorting the pins for the power switch with a small blade
    screwdriver. If it reliably starts, saves the correct FSB and multiplier
    for your CPU and can be repeated on a cold boot (turn off PSU, walk away for
    a while, then return and repeat) then you can be reasonably certain you had
    a short to the case (in which case .. check for bent or errant studs on your
    motherboard pan). If you lose your settings and you still haven't metered
    the motherboard battery, might be worth a trip to the local Radio Shack if
    you don't have a meter.

    As to the ISA slots on the KT7/KT7A .. it's been a while (quite a few years)
    since I worked on my old KT7-RAID. It had two ISA slots (the inner slot
    sharing the bus with the last PCI slot .. if I remember correctly). Also ..
    if you read back through the group, on December 5th I was accused by TomG of
    doddering over the edge.

    Jef
     
    Bird Janitor®, Dec 30, 2006
    #4
  5. Robert  Usdin

    Exray Guest

    Well, you are not doddering over any edges as far as I'm concerned.

    I agree, you weren't after the OP to go buy a new supply. I don't recall
    who was the maker of the 480 W one that I tried, but since behavior was
    identical with either supply, 300 or 480, it would be unusual for the issue
    to be primarily associated with the supply. The local computer store here
    even emphasized that they'd take it back if it didn't do the job. It
    didn't...they did.

    I've developed real expertise in moving the CMOS jumper from where it
    normally lives to the clear position. It succeeds at clearing the calendar
    to January 2002 midnight, and then I put it back after a minute or so.

    Good question about the 2032 battery. When they die, they seem to go to
    between 2.5 and 2.7 volts. The new one was reading 3.17, (isn't the
    precision of digital meters great? I don't know the real voltage, but it's
    about a half volt higher than the old battery).

    The case I have has the Mobo mounted on the back chassis. When I took the
    machine down, I didn't remove the board from that plate, I just pulled the
    whole plate and laid it down sort of half in and half out of the case. I
    can see that there's nothing even close between board and plate, and I did
    all kinds of finger poking and prodding to verify that there's not much that
    can move, nothing that is loose.

    When the machine does POST (after clearing the CMOS), I get a 600 meg
    default, and a CMOS checksum error and a message that the CPU is unworkable
    or has been changed...and an invitation to press delete to fix the settings
    or hit F1 to just press on. The interesting thing about trying to do any
    settings (and I have perhaps been too aggressive -- it's defaulting to
    6x100, with a PCI bus at 133. I keep on trying to get it to something x
    133, where it always worked well before I improved it), and when I say
    save -- it's into a cold boot mode and it's not gonna do anything useful for
    me.

    It won't cold boot with an ISA video card and one stick of RAM and nothing
    else, 'cause it wants to default to 133. Definitely a sick machine.

    I downloaded and installed Speed Fan today. I didn't realize that one can
    invoke the softmenu from within Windows using that software. Well, a normal
    person with a working computer could do that....when I tell it to go from
    the 100 meg FSB to 133, the machine freezes; and a warm boot won't work
    (since it thinks it's supposed to fire up at 133).

    This is almost an academic issue with me. The machine "works", it is gonna
    be replaced next week, it's totally backed up....but I want to know what
    I've missed, and I'm wondering if replacing some capacitors might be
    helpful. It's quite stable, it warm boots when I do a restart in Windows
    (it thinks it's at 100 mHz, at that point). It'll be interesting to see if
    it deteriorates over time by itself. Perhaps when the CPU fan gets its dust
    back, it'll start working better.

    I thank you for your response...and I apologize to the OP for hijacking his
    thread. I hope he got his video card working right.




    "Bird Janitor®" <> wrote in
    message news:7nAlh.7888$...
    > Exray wrote:
    > |
    > | <Huge Snip>
    > |
    > | Bird, your post was informative (I trust) for the original poster, also
    > | for me.
    > |
    > | Turns out that a no beep condition doesn't necessarily mean death,
    > | it just means that one little bitty vitally critical thing is not right.
    > |
    > | My KT7AR currently won't POST at all...no beep...until I clear the
    > | CMOS, fail on the cold boot, and then hit the reset button. (thank
    > | goodness for reset buttons). I was mucking with a pretty good (and
    > | extremely dusty) XP 1900+ board, and now all I can get out of it is 6
    > | x 100. The 2200 uFd capacitors look great, some of the smaller ones,
    > | the 10 u at 10 V look a little bit wavy on top. (I have an old KT7 on
    > | which two of the 2200s have domes on them, but the smaller caps all
    > | look perfect).
    > |
    > | While I was putting the machine back together, before I realized I
    > | did something weird to the mobo during the annual removal of dust, I
    > | eventually noticed that I wasn't seating the AGP card well. The card
    > | and one of the handles on the memory slot closest to the CPU were
    > | interfering unless I pushed that handle to the "loaded" position,
    > although
    > | that slot is empty.
    > |
    > | It's certainly possible that the OPs problem might be solved with a
    > | guttier power supply than the one he's using, but if he goes out and
    > | buys a larger supply, he should verify the store's return policy first
    > | before handing over his payment.
    > |
    > | By the way...only one ISA slot on my version 1.0.
    > |
    >
    > Let's be clear first .. I wasn't suggesting that Rob go out and buy a new
    > PSU. He said he had a 350 watt unit to try on it and I agreed that this
    > would be the best first step. I'd hate to be responsible for needless
    > spending (unless it was on me!).
    >
    > You say you had a 300 watt V-Tech in the machine in your other post (not
    > the
    > prettiest crayon in the box when it comes to good quality PSU's). Who
    > makes
    > the 480 watt unit in there now? There are a lot of PSU's out there,
    > regardless of maximum rated wattage, that serve better to stop the door
    > than
    > they do to power a computer.
    >
    > For some reason you seem to be defaulting to failsafe defaults when you
    > finally do get the machine to POST with the reset switch. Are you able to
    > go into the BIOS and make changes (Load Optimized Defaults) then save and
    > exit, and have the changes still present when you reboot? Obviously this
    > ends up a round-trip on the next cold boot.
    >
    > I say all the former because you may have your CCMOS1 jumper set on pins 2
    > and 3 (clear CMOS) instead of 1 and 2 (retain CMOS with backup power or
    > motherboard battery when PSU is switched off).
    >
    > I see you've replaced your motherboard battery. Have you metered the
    > replacement? Could be dead out of the package.
    >
    > Another possibility is that you have a short between your board and the
    > case. If you're into tear-down testing, try the board out of the case on
    > a
    > non-conductive surface. Connect up only PSU, video card, RAM, CPU, HSF,
    > monitor and keyboard. Start the board by turning on the PSU then
    > momentarily shorting the pins for the power switch with a small blade
    > screwdriver. If it reliably starts, saves the correct FSB and multiplier
    > for your CPU and can be repeated on a cold boot (turn off PSU, walk away
    > for
    > a while, then return and repeat) then you can be reasonably certain you
    > had
    > a short to the case (in which case .. check for bent or errant studs on
    > your
    > motherboard pan). If you lose your settings and you still haven't metered
    > the motherboard battery, might be worth a trip to the local Radio Shack if
    > you don't have a meter.
    >
    > As to the ISA slots on the KT7/KT7A .. it's been a while (quite a few
    > years)
    > since I worked on my old KT7-RAID. It had two ISA slots (the inner slot
    > sharing the bus with the last PCI slot .. if I remember correctly). Also
    > ..
    > if you read back through the group, on December 5th I was accused by TomG
    > of
    > doddering over the edge.
    >
    > Jef
    >
    >
     
    Exray, Dec 30, 2006
    #5
  6. Exray wrote:
    |
    | Well, you are not doddering over any edges as far as I'm concerned.
    |

    There are those who frequent here who would say otherwise... ;-}

    In any event .. major snippage because I don't feel like formatting for a
    nicely quoted post.

    Sure sounds as if you have a very sick board.

    If you stayed with the 100MHz DDR FSB and played with the multiplier, I
    wonder if you'd have better luck getting the 1900+ closer to it's chosen
    speed. Look for posts by Wes in this group .. he's got a link to his web
    space with lots of information on fiddling with unsupported processors in
    these old KT7/KT7A boards. My old KT7-RAID is still running strong some
    1500 miles from here .. may not be the exact same board, since it went back
    to Abit Mothership during the class action settlement days, and came back
    none the worse for the experience .. my dad is still using it as his Windows
    98SE machine.

    And, you're correct .. it's been my experience the CR2032's loose their
    ability to maintain the CMOS somewhere south of 2.7 volts.

    As to the 'out of body' exercise I suggested: I seem to recall you mention
    that your board went ill about the time of its last de-dust-bunny-ing.
    Something could have gotten knocked ajar in the process that you're not even
    seeing. Might be worth taking a few minutes, since you're planning on
    tearing it down anyway, to try to POST it completely outside the case.

    I've never personally experienced the CMOS going south due to bad caps.
    Usually freaky crashes and failure to POST reliably, if at all, are the
    symptoms. Reset-reset-reset-reset-reset-POST is a common method of
    jump-starting when the caps start to go .. works for a while, until the
    issue gets the best of the board. It's been my experience that a bad CR2032
    or a short is generally the cause of the CMOS going back to Failsafe
    Defaults.

    Your board does sound sick, though. You've diagnosed that correct. Got a
    magnifying glass? Look closer under light and see if you've got tell-tale
    electrolyte starting to ooze. In any event, if the caps are indeed bad,
    it's probably worth it to attempt to repair the board if you've got the
    skills. The KT7/KT7A were great boards of their day.

    Jef


    PS .. Sorry Rob. Exray and I seem to like your thread better than the one
    he started. But .. we're on topic as to sick KT7's at least.
     
    Bird Janitor®, Dec 30, 2006
    #6
  7. Robert  Usdin

    Robert Usdin Guest


    > PS .. Sorry Rob. Exray and I seem to like your thread better than the one
    > he started. But .. we're on topic as to sick KT7's at least.


    No problem. So after I posted, I tried it again after switching the
    AGP mode in the BIOS to 4X and it works perfectly. Go figure.

    Now I have another issue I;m trying to figure out. Since installing
    the new video card, my wireless internet connection has gotten flaky.
    I'm wondering if the new video card can cause enough interference to
    cause a wireless connection to go down about...oh...every ten minutes,
    then reconnect almost immediately. I need to test for sure that it is
    the new video card, which I have not had time to do yet. But I didn't
    see the issue before the new video card. Very strange. I also still
    want to test the 350 watt power supply too. It may still be straining
    with the 300 Watt.... Ah fun fun fun...

    My next thing to do with this machine is add a faster processor (it has
    the 1.3 chipset). I may just be playing with fire.....

    --*Rob
     
    Robert Usdin, Dec 31, 2006
    #7
    1. Advertising

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. Enkidu
    Replies:
    9
    Views:
    572
  2. erik©

    updating kt7a to kt7a-raid

    erik©, Nov 27, 2003, in forum: Abit
    Replies:
    1
    Views:
    410
    Homie
    Nov 27, 2003
  3. Replies:
    0
    Views:
    376
  4. Replies:
    7
    Views:
    695
    Wes Newell
    Dec 16, 2006
  5. Replies:
    2
    Views:
    469
Loading...

Share This Page