PC won't power on

Discussion in 'Computer Support' started by alice, Mar 13, 2007.

  1. alice

    Leythos Guest

    On Sat, 17 Mar 2007 09:45:27 -0700, alice wrote:

    > On Mar 16, 6:24 am, Leythos <> wrote:
    >> On Thu, 15 Mar 2007 22:15:14 -0700, w_tom wrote:
    >>
    >> > Get the meter. Discover what is and is not functional before
    >> > changing anything more. Do not shotgun.

    >>
    >> By the time they find/buy a meter they could have swapped out the PSU and
    >> been up and running without the cost of a Meter they may never use again.
    >>

    >
    > I already have swapped the PSU (ie shotgunning), and that didn't work,
    > and it does not prove or disprove what is or isn't working.
    > The above posts are right, only a multimeter can tell me anything at
    > this point.


    I never suggested that the idea of purchasing a multimeter was wrong, only
    that by the time you get one, test the power supply, find out it's bad
    (most cases), then go buy a power supply, you've wasted hours.

    In most cases, if the system won't power up it's one of two causes (1 bad
    PSU, 2 Bad motherboard) - in most cases.
     
    Leythos, Mar 17, 2007
    #41
    1. Advertising

  2. alice

    w_tom Guest

    On Mar 17, 1:47 pm, Leythos <> wrote:
    > I never suggested that the idea of purchasing a multimeter was wrong, only
    > that by the time you get one, test the power supply, find out it's bad
    > (most cases), then go buy a power supply, you've wasted hours.
    >
    > In most cases, if the system won't power up it's one of two causes (1 bad
    > PSU, 2 Bad motherboard) - in most cases.


    Even if a new power supply 'appears' to work, we still don't know
    until that new power supply is verified by the meter. It is a power
    supply 'system'. Power supply is only one 'system' component. Meter
    reports on the entire 'system'.

    For those who have motherboard monitors: 3.5 digit multimeter is
    also necessary to calibrate that motherboard voltage monitor.

    Meanwhile, time taken to swap a power supply is many times longer
    than what the meter would report without disconnecting anything.
    Furthermore, numbers from the meter that might, at first, appear OK
    can also elicit other useful facts from the newsgroup. Those replies
    will only be as good as data in the original post. Another reason why
    the tool is as necessary as a screwdriver.

    In another thread, the OP not only discovered how to fix the
    problem. OP also discovered what had failed and therefore what was
    probably the reason for that failure. Just another reason to identify
    before replacing. Using a meter, the OP also knows heat, power
    cycling, or surges did not cause his failure. More useful information
    because the problem was identified before parts were replaced: "Dead
    motherboard?" in the newsgroup
    microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    2007.
     
    w_tom, Mar 18, 2007
    #42
    1. Advertising

  3. alice

    Leythos Guest

    On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:

    > Meanwhile, time taken to swap a power supply is many times longer
    > than what the meter would report without disconnecting anything.
    > Furthermore, numbers from the meter that might, at first, appear OK
    > can also elicit other useful facts from the newsgroup. Those replies
    > will only be as good as data in the original post. Another reason why
    > the tool is as necessary as a screwdriver.
    >
    > In another thread, the OP not only discovered how to fix the
    > problem. OP also discovered what had failed and therefore what was
    > probably the reason for that failure. Just another reason to identify
    > before replacing. Using a meter, the OP also knows heat, power
    > cycling, or surges did not cause his failure. More useful information
    > because the problem was identified before parts were replaced: "Dead
    > motherboard?" in the newsgroup
    > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    > 2007.


    And for 90% of all cases the OP won't have a multimeter and most likely
    the place the OP buys a PSU from won't have a multimeter for sale.

    In 90% of all cases of no-power, it's the power supply, or at least
    disconnecting the drives will indicate it's down to the motherboard. You
    can even remove the RAM/Video Card and fans, and if it still doesn't power
    up, without a multi-meter you're down to PSU or Motherboard.

    So, you can spend an hour finding a store that sells you a multimeter -
    about $50 on the low end, or you can get that PSU after doing the above 5
    minute test (disconnecting parts) and see if that fixes it - if not, then
    it's the motherboard.

    Oh, lets not forget that most people don't have a clue as to how to use a
    multimeter let alone know what the tolerance is on the voltage levels on
    the PSU, don't have a clue about current draws, don't know what colors are
    suppose to be what voltages...

    By the time they do all needed test with a multimeter, in 99% of the
    cases, they could have spent less time by replacing the power supply.

    --
    Want to know what PCBUTTS1 is really about?
    *** WARNING - these links contain foul/pornographic content of an
    abusive nature created by PCBUTTS1 and still hosted on his public
    website ***
    http://www.pcbutts1.com/rlk/rlk.htm ,
    http://www.pcbutts1.com/license.htm ,
    http://www.pcbutts1.com/downloads/max.htm ,
    http://www.pcbutts1.com/downloads/mpv.htm ,
    http://www.pcbutts1.com/downloads/wtcpcb.htm ,
    http://www.pcbutts1.com/cracks.htm ,
    http://www.pcbutts1.com/Loutheasshole.htm
    All while spamming his company website at: http://www.seedsv.com
     
    Leythos, Mar 18, 2007
    #43
  4. alice

    alice Guest

    On Mar 18, 6:04 am, Leythos <> wrote:
    > On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:
    > > Meanwhile, time taken to swap a power supply is many times longer
    > > than what the meter would report without disconnecting anything.
    > > Furthermore, numbers from the meter that might, at first, appear OK
    > > can also elicit other useful facts from the newsgroup. Those replies
    > > will only be as good as data in the original post. Another reason why
    > > the tool is as necessary as a screwdriver.

    >
    > > In another thread, the OP not only discovered how to fix the
    > > problem. OP also discovered what had failed and therefore what was
    > > probably the reason for that failure. Just another reason to identify
    > > before replacing. Using a meter, the OP also knows heat, power
    > > cycling, or surges did not cause his failure. More useful information
    > > because the problem was identified before parts were replaced: "Dead
    > > motherboard?" in the newsgroup
    > > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    > > 2007.

    >
    > And for 90% of all cases the OP won't have a multimeter and most likely
    > the place the OP buys a PSU from won't have a multimeter for sale.
    >
    > In 90% of all cases of no-power, it's the power supply, or at least
    > disconnecting the drives will indicate it's down to the motherboard. You
    > can even remove the RAM/Video Card and fans, and if it still doesn't power
    > up, without a multi-meter you're down to PSU or Motherboard.
    >
    > So, you can spend an hour finding a store that sells you a multimeter -
    > about $50 on the low end, or you can get that PSU after doing the above 5
    > minute test (disconnecting parts) and see if that fixes it - if not, then
    > it's the motherboard.
    >
    > Oh, lets not forget that most people don't have a clue as to how to use a
    > multimeter let alone know what the tolerance is on the voltage levels on
    > the PSU, don't have a clue about current draws, don't know what colors are
    > suppose to be what voltages...
    >
    > By the time they do all needed test with a multimeter, in 99% of the
    > cases, they could have spent less time by replacing the power supply.
    >
    > --
    > Want to know what PCBUTTS1 is really about?
    > *** WARNING - these links contain foul/pornographic content of an
    > abusive nature created by PCBUTTS1 and still hosted on his public
    > website ***http://www.pcbutts1.com/rlk/rlk.htm...htm,http://www.pcbutts1.com/Loutheasshole.htm
    > All while spamming his company website at:http://www.seedsv.com- Hide quoted text -
    >
    > - Show quoted text -


    I see everyone's point here, but I already have swapped out the power
    supply with one that was last known to be working, and the original
    PSU was working just minutes before I removed the
    CPU fan. I haven't used a multimeter since high school, but I should
    probably learn and no doubt will use it again. This is for a very low
    budget company so it's better to spend more time
    and figure out what part(s) need replacing rather than just buy a
    bunch of spare parts that we don't need, and I'd rather learn what is
    wrong, for future reference. I want to know
    how and why and what is failing exactly. My *guess* is that it is
    indeed either the CPU or the MB, but I can't be certain at this point.
     
    alice, Mar 18, 2007
    #44
  5. alice

    Leythos Guest

    On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:

    > On Mar 18, 6:04 am, Leythos <> wrote:
    >> On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:
    >> > Meanwhile, time taken to swap a power supply is many times longer
    >> > than what the meter would report without disconnecting anything.
    >> > Furthermore, numbers from the meter that might, at first, appear OK
    >> > can also elicit other useful facts from the newsgroup. Those replies
    >> > will only be as good as data in the original post. Another reason why
    >> > the tool is as necessary as a screwdriver.

    >>
    >> > In another thread, the OP not only discovered how to fix the
    >> > problem. OP also discovered what had failed and therefore what was
    >> > probably the reason for that failure. Just another reason to identify
    >> > before replacing. Using a meter, the OP also knows heat, power
    >> > cycling, or surges did not cause his failure. More useful information
    >> > because the problem was identified before parts were replaced: "Dead
    >> > motherboard?" in the newsgroup
    >> > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    >> > 2007.

    >>
    >> And for 90% of all cases the OP won't have a multimeter and most likely
    >> the place the OP buys a PSU from won't have a multimeter for sale.
    >>
    >> In 90% of all cases of no-power, it's the power supply, or at least
    >> disconnecting the drives will indicate it's down to the motherboard. You
    >> can even remove the RAM/Video Card and fans, and if it still doesn't power
    >> up, without a multi-meter you're down to PSU or Motherboard.
    >>
    >> So, you can spend an hour finding a store that sells you a multimeter -
    >> about $50 on the low end, or you can get that PSU after doing the above 5
    >> minute test (disconnecting parts) and see if that fixes it - if not, then
    >> it's the motherboard.
    >>
    >> Oh, lets not forget that most people don't have a clue as to how to use a
    >> multimeter let alone know what the tolerance is on the voltage levels on
    >> the PSU, don't have a clue about current draws, don't know what colors are
    >> suppose to be what voltages...
    >>
    >> By the time they do all needed test with a multimeter, in 99% of the
    >> cases, they could have spent less time by replacing the power supply.
    >>

    >
    > I see everyone's point here, but I already have swapped out the power
    > supply with one that was last known to be working, and the original
    > PSU was working just minutes before I removed the
    > CPU fan. I haven't used a multimeter since high school, but I should
    > probably learn and no doubt will use it again. This is for a very low
    > budget company so it's better to spend more time
    > and figure out what part(s) need replacing rather than just buy a
    > bunch of spare parts that we don't need, and I'd rather learn what is
    > wrong, for future reference. I want to know
    > how and why and what is failing exactly. My *guess* is that it is
    > indeed either the CPU or the MB, but I can't be certain at this point.


    Learning about the power and such IS a great way to learn about hardware,
    but, there are so many things that can be wrong with the power when not
    under load, when under load, when shorted out by certain failed devices,
    and then there are many PSU units that have multiple outputs for different
    taps that don't impact each other if they fail - you could have the loss
    of a 12+ power on one set of power connectors and +12 on the other
    segments could be fine.

    What it gets down to is this:

    If the Computer does not post, there are some simple basics that don't
    require a multimeter:

    1) When you plugin the AC cord, does anything bump (fans)?

    2) When you have the AC power on, do you see any indicator lights (LED's)
    on the motherboard that indicate power is on?

    3) Disconnect all devices except the main power, the power button, and the
    CPU extra power connector - no fans, no drives, no extra power for the
    video card if it has it, no external devices - press the power button,
    does it come on at some level?

    4) If nothing happened in #3, remove memory, remove video card - does it
    come on now? You won't see much, but you should hear the PSU fans start if
    they are going to run.

    5) Replace PSU with one from a spare computer or one that you've already
    bought - does it come on?

    6) If nothing working after #5, then it's down to the CPU/Motherboard -
    your call, and nothing from a Multimeter will give you any hope here, it
    won't tell you a dang thing.

    So, without a multimeter, we've spent 10 minutes to test that you have a
    bad CPU/Motherboard.

    If at any point above you saw signs of life, well, you start doing the
    reverse one device at a time until it stops - another 10 minutes. In less
    than 30 minutes you can strip the PC down to what is failed and all
    without any other parts, except the PSU that you can often get anywhere in
    a reasonably size city and you don't need to know how to use a PSU.

    --
    Want to know what PCBUTTS1 is really about?
    *** WARNING - these links contain foul/pornographic content of an
    abusive nature created by PCBUTTS1 and still hosted on his public
    website ***
    http://www.pcbutts1.com/rlk/rlk.htm ,
    http://www.pcbutts1.com/license.htm ,
    http://www.pcbutts1.com/downloads/max.htm ,
    http://www.pcbutts1.com/downloads/mpv.htm ,
    http://www.pcbutts1.com/downloads/wtcpcb.htm ,
    http://www.pcbutts1.com/cracks.htm ,
    http://www.pcbutts1.com/Loutheasshole.htm
    All while spamming his company website at: http://www.seedsv.com
     
    Leythos, Mar 18, 2007
    #45
  6. alice

    alice Guest

    On Mar 18, 3:19 pm, Leythos <> wrote:
    > On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
    > > On Mar 18, 6:04 am, Leythos <> wrote:
    > >> On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:
    > >> > Meanwhile, time taken to swap a power supply is many times longer
    > >> > than what the meter would report without disconnecting anything.
    > >> > Furthermore, numbers from the meter that might, at first, appear OK
    > >> > can also elicit other useful facts from the newsgroup. Those replies
    > >> > will only be as good as data in the original post. Another reason why
    > >> > the tool is as necessary as a screwdriver.

    >
    > >> > In another thread, the OP not only discovered how to fix the
    > >> > problem. OP also discovered what had failed and therefore what was
    > >> > probably the reason for that failure. Just another reason to identify
    > >> > before replacing. Using a meter, the OP also knows heat, power
    > >> > cycling, or surges did not cause his failure. More useful information
    > >> > because the problem was identified before parts were replaced: "Dead
    > >> > motherboard?" in the newsgroup
    > >> > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    > >> > 2007.

    >
    > >> And for 90% of all cases the OP won't have a multimeter and most likely
    > >> the place the OP buys a PSU from won't have a multimeter for sale.

    >
    > >> In 90% of all cases of no-power, it's the power supply, or at least
    > >> disconnecting the drives will indicate it's down to the motherboard. You
    > >> can even remove the RAM/Video Card and fans, and if it still doesn't power
    > >> up, without a multi-meter you're down to PSU or Motherboard.

    >
    > >> So, you can spend an hour finding a store that sells you a multimeter -
    > >> about $50 on the low end, or you can get that PSU after doing the above 5
    > >> minute test (disconnecting parts) and see if that fixes it - if not, then
    > >> it's the motherboard.

    >
    > >> Oh, lets not forget that most people don't have a clue as to how to use a
    > >> multimeter let alone know what the tolerance is on the voltage levels on
    > >> the PSU, don't have a clue about current draws, don't know what colors are
    > >> suppose to be what voltages...

    >
    > >> By the time they do all needed test with a multimeter, in 99% of the
    > >> cases, they could have spent less time by replacing the power supply.

    >
    > > I see everyone's point here, but I already have swapped out the power
    > > supply with one that was last known to be working, and the original
    > > PSU was working just minutes before I removed the
    > > CPU fan. I haven't used a multimeter since high school, but I should
    > > probably learn and no doubt will use it again. This is for a very low
    > > budget company so it's better to spend more time
    > > and figure out what part(s) need replacing rather than just buy a
    > > bunch of spare parts that we don't need, and I'd rather learn what is
    > > wrong, for future reference. I want to know
    > > how and why and what is failing exactly. My *guess* is that it is
    > > indeed either the CPU or the MB, but I can't be certain at this point.

    >
    > Learning about the power and such IS a great way to learn about hardware,
    > but, there are so many things that can be wrong with the power when not
    > under load, when under load, when shorted out by certain failed devices,
    > and then there are many PSU units that have multiple outputs for different
    > taps that don't impact each other if they fail - you could have the loss
    > of a 12+ power on one set of power connectors and +12 on the other
    > segments could be fine.
    >
    > What it gets down to is this:
    >
    > If the Computer does not post, there are some simple basics that don't
    > require a multimeter:
    >
    > 1) When you plugin the AC cord, does anything bump (fans)?
    >
    > 2) When you have the AC power on, do you see any indicator lights (LED's)
    > on the motherboard that indicate power is on?
    >
    > 3) Disconnect all devices except the main power, the power button, and the
    > CPU extra power connector - no fans, no drives, no extra power for the
    > video card if it has it, no external devices - press the power button,
    > does it come on at some level?
    >
    > 4) If nothing happened in #3, remove memory, remove video card - does it
    > come on now? You won't see much, but you should hear the PSU fans start if
    > they are going to run.
    >
    > 5) Replace PSU with one from a spare computer or one that you've already
    > bought - does it come on?
    >
    > 6) If nothing working after #5, then it's down to the CPU/Motherboard -
    > your call, and nothing from a Multimeter will give you any hope here, it
    > won't tell you a dang thing.
    >
    > So, without a multimeter, we've spent 10 minutes to test that you have a
    > bad CPU/Motherboard.
    >
    > If at any point above you saw signs of life, well, you start doing the
    > reverse one device at a time until it stops - another 10 minutes. In less
    > than 30 minutes you can strip the PC down to what is failed and all
    > without any other parts, except the PSU that you can often get anywhere in
    > a reasonably size city and you don't need to know how to use a PSU.
    >
    > --
    > Want to know what PCBUTTS1 is really about?
    > *** WARNING - these links contain foul/pornographic content of an
    > abusive nature created by PCBUTTS1 and still hosted on his public
    > website ***http://www.pcbutts1.com/rlk/rlk.htm, http://www.pcbutts1.com/license.htm...htm,http://www.pcbutts1.com/Loutheasshole.htm
    > All while spamming his company website at:http://www.seedsv.com- Hide quoted text -
    >
    > - Show quoted text -


    What happens is that I press the power button, the front power light
    on the tower comes on, the PSU fan and the CPU fan start to spin, then
    a few seconds later, it all turns off.
    I'm not at it right now, so I can't try it with all of your above
    suggestions, however, I have already tried it with another PSU (that
    worked the last time it was in a PC), tried
    unplugging the hard drives (one at a time), and tried it without the
    CPU fan, as well as a different CPU fan, and the results have been
    exactly the same each time.
     
    alice, Mar 18, 2007
    #46
  7. alice

    Leythos Guest

    On Sun, 18 Mar 2007 15:42:28 -0700, alice wrote:

    > On Mar 18, 3:19 pm, Leythos <> wrote:
    >> On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
    >> > On Mar 18, 6:04 am, Leythos <> wrote:
    >> >> On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:
    >> >> > Meanwhile, time taken to swap a power supply is many times longer
    >> >> > than what the meter would report without disconnecting anything.
    >> >> > Furthermore, numbers from the meter that might, at first, appear OK
    >> >> > can also elicit other useful facts from the newsgroup. Those replies
    >> >> > will only be as good as data in the original post. Another reason why
    >> >> > the tool is as necessary as a screwdriver.

    >>
    >> >> > In another thread, the OP not only discovered how to fix the
    >> >> > problem. OP also discovered what had failed and therefore what was
    >> >> > probably the reason for that failure. Just another reason to identify
    >> >> > before replacing. Using a meter, the OP also knows heat, power
    >> >> > cycling, or surges did not cause his failure. More useful information
    >> >> > because the problem was identified before parts were replaced: "Dead
    >> >> > motherboard?" in the newsgroup
    >> >> > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    >> >> > 2007.

    >>
    >> >> And for 90% of all cases the OP won't have a multimeter and most likely
    >> >> the place the OP buys a PSU from won't have a multimeter for sale.

    >>
    >> >> In 90% of all cases of no-power, it's the power supply, or at least
    >> >> disconnecting the drives will indicate it's down to the motherboard. You
    >> >> can even remove the RAM/Video Card and fans, and if it still doesn't power
    >> >> up, without a multi-meter you're down to PSU or Motherboard.

    >>
    >> >> So, you can spend an hour finding a store that sells you a multimeter -
    >> >> about $50 on the low end, or you can get that PSU after doing the above 5
    >> >> minute test (disconnecting parts) and see if that fixes it - if not, then
    >> >> it's the motherboard.

    >>
    >> >> Oh, lets not forget that most people don't have a clue as to how to use a
    >> >> multimeter let alone know what the tolerance is on the voltage levels on
    >> >> the PSU, don't have a clue about current draws, don't know what colors are
    >> >> suppose to be what voltages...

    >>
    >> >> By the time they do all needed test with a multimeter, in 99% of the
    >> >> cases, they could have spent less time by replacing the power supply.

    >>
    >> > I see everyone's point here, but I already have swapped out the power
    >> > supply with one that was last known to be working, and the original
    >> > PSU was working just minutes before I removed the
    >> > CPU fan. I haven't used a multimeter since high school, but I should
    >> > probably learn and no doubt will use it again. This is for a very low
    >> > budget company so it's better to spend more time
    >> > and figure out what part(s) need replacing rather than just buy a
    >> > bunch of spare parts that we don't need, and I'd rather learn what is
    >> > wrong, for future reference. I want to know
    >> > how and why and what is failing exactly. My *guess* is that it is
    >> > indeed either the CPU or the MB, but I can't be certain at this point.

    >>
    >> Learning about the power and such IS a great way to learn about hardware,
    >> but, there are so many things that can be wrong with the power when not
    >> under load, when under load, when shorted out by certain failed devices,
    >> and then there are many PSU units that have multiple outputs for different
    >> taps that don't impact each other if they fail - you could have the loss
    >> of a 12+ power on one set of power connectors and +12 on the other
    >> segments could be fine.
    >>
    >> What it gets down to is this:
    >>
    >> If the Computer does not post, there are some simple basics that don't
    >> require a multimeter:
    >>
    >> 1) When you plugin the AC cord, does anything bump (fans)?
    >>
    >> 2) When you have the AC power on, do you see any indicator lights (LED's)
    >> on the motherboard that indicate power is on?
    >>
    >> 3) Disconnect all devices except the main power, the power button, and the
    >> CPU extra power connector - no fans, no drives, no extra power for the
    >> video card if it has it, no external devices - press the power button,
    >> does it come on at some level?
    >>
    >> 4) If nothing happened in #3, remove memory, remove video card - does it
    >> come on now? You won't see much, but you should hear the PSU fans start if
    >> they are going to run.
    >>
    >> 5) Replace PSU with one from a spare computer or one that you've already
    >> bought - does it come on?
    >>
    >> 6) If nothing working after #5, then it's down to the CPU/Motherboard -
    >> your call, and nothing from a Multimeter will give you any hope here, it
    >> won't tell you a dang thing.
    >>
    >> So, without a multimeter, we've spent 10 minutes to test that you have a
    >> bad CPU/Motherboard.
    >>
    >> If at any point above you saw signs of life, well, you start doing the
    >> reverse one device at a time until it stops - another 10 minutes. In less
    >> than 30 minutes you can strip the PC down to what is failed and all
    >> without any other parts, except the PSU that you can often get anywhere in
    >> a reasonably size city and you don't need to know how to use a PSU.
    >>

    >
    > What happens is that I press the power button, the front power light
    > on the tower comes on, the PSU fan and the CPU fan start to spin, then
    > a few seconds later, it all turns off.
    > I'm not at it right now, so I can't try it with all of your above
    > suggestions, however, I have already tried it with another PSU (that
    > worked the last time it was in a PC), tried
    > unplugging the hard drives (one at a time), and tried it without the
    > CPU fan, as well as a different CPU fan, and the results have been
    > exactly the same each time.


    Don't try things ONE at a time, disconnect device 1, then test, then
    device 2 (not reconnecting device 1), and keep going - as you could have
    several devices that are fried.

    If the new PSU didn't fix it, and you've removed the drives/fans, and
    you've removed the memory and video - then it's the motherboard/CPU and a
    multimeter won't tell you which one.



    --
    Want to know what PCBUTTS1 is really about?
    *** WARNING - these links contain foul/pornographic content of an
    abusive nature created by PCBUTTS1 and still hosted on his public
    website ***
    http://www.pcbutts1.com/rlk/rlk.htm ,
    http://www.pcbutts1.com/license.htm ,
    http://www.pcbutts1.com/downloads/max.htm ,
    http://www.pcbutts1.com/downloads/mpv.htm ,
    http://www.pcbutts1.com/downloads/wtcpcb.htm ,
    http://www.pcbutts1.com/cracks.htm ,
    http://www.pcbutts1.com/Loutheasshole.htm
    All while spamming his company website at: http://www.seedsv.com
     
    Leythos, Mar 18, 2007
    #47
  8. alice

    alice Guest

    On Mar 18, 3:50 pm, Leythos <> wrote:
    > On Sun, 18 Mar 2007 15:42:28 -0700, alice wrote:
    > > On Mar 18, 3:19 pm, Leythos <> wrote:
    > >> On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
    > >> > On Mar 18, 6:04 am, Leythos <> wrote:
    > >> >> On Sat, 17 Mar 2007 20:58:01 -0700, w_tom wrote:
    > >> >> > Meanwhile, time taken to swap a power supply is many times longer
    > >> >> > than what the meter would report without disconnecting anything.
    > >> >> > Furthermore, numbers from the meter that might, at first, appear OK
    > >> >> > can also elicit other useful facts from the newsgroup. Those replies
    > >> >> > will only be as good as data in the original post. Another reason why
    > >> >> > the tool is as necessary as a screwdriver.

    >
    > >> >> > In another thread, the OP not only discovered how to fix the
    > >> >> > problem. OP also discovered what had failed and therefore what was
    > >> >> > probably the reason for that failure. Just another reason to identify
    > >> >> > before replacing. Using a meter, the OP also knows heat, power
    > >> >> > cycling, or surges did not cause his failure. More useful information
    > >> >> > because the problem was identified before parts were replaced: "Dead
    > >> >> > motherboard?" in the newsgroup
    > >> >> > microsoft.public.windowsxp.perform_maintain between 10 and 14 Mar
    > >> >> > 2007.

    >
    > >> >> And for 90% of all cases the OP won't have a multimeter and most likely
    > >> >> the place the OP buys a PSU from won't have a multimeter for sale.

    >
    > >> >> In 90% of all cases of no-power, it's the power supply, or at least
    > >> >> disconnecting the drives will indicate it's down to the motherboard. You
    > >> >> can even remove the RAM/Video Card and fans, and if it still doesn't power
    > >> >> up, without a multi-meter you're down to PSU or Motherboard.

    >
    > >> >> So, you can spend an hour finding a store that sells you a multimeter -
    > >> >> about $50 on the low end, or you can get that PSU after doing the above 5
    > >> >> minute test (disconnecting parts) and see if that fixes it - if not, then
    > >> >> it's the motherboard.

    >
    > >> >> Oh, lets not forget that most people don't have a clue as to how to use a
    > >> >> multimeter let alone know what the tolerance is on the voltage levels on
    > >> >> the PSU, don't have a clue about current draws, don't know what colors are
    > >> >> suppose to be what voltages...

    >
    > >> >> By the time they do all needed test with a multimeter, in 99% of the
    > >> >> cases, they could have spent less time by replacing the power supply.

    >
    > >> > I see everyone's point here, but I already have swapped out the power
    > >> > supply with one that was last known to be working, and the original
    > >> > PSU was working just minutes before I removed the
    > >> > CPU fan. I haven't used a multimeter since high school, but I should
    > >> > probably learn and no doubt will use it again. This is for a very low
    > >> > budget company so it's better to spend more time
    > >> > and figure out what part(s) need replacing rather than just buy a
    > >> > bunch of spare parts that we don't need, and I'd rather learn what is
    > >> > wrong, for future reference. I want to know
    > >> > how and why and what is failing exactly. My *guess* is that it is
    > >> > indeed either the CPU or the MB, but I can't be certain at this point.

    >
    > >> Learning about the power and such IS a great way to learn about hardware,
    > >> but, there are so many things that can be wrong with the power when not
    > >> under load, when under load, when shorted out by certain failed devices,
    > >> and then there are many PSU units that have multiple outputs for different
    > >> taps that don't impact each other if they fail - you could have the loss
    > >> of a 12+ power on one set of power connectors and +12 on the other
    > >> segments could be fine.

    >
    > >> What it gets down to is this:

    >
    > >> If the Computer does not post, there are some simple basics that don't
    > >> require a multimeter:

    >
    > >> 1) When you plugin the AC cord, does anything bump (fans)?

    >
    > >> 2) When you have the AC power on, do you see any indicator lights (LED's)
    > >> on the motherboard that indicate power is on?

    >
    > >> 3) Disconnect all devices except the main power, the power button, and the
    > >> CPU extra power connector - no fans, no drives, no extra power for the
    > >> video card if it has it, no external devices - press the power button,
    > >> does it come on at some level?

    >
    > >> 4) If nothing happened in #3, remove memory, remove video card - does it
    > >> come on now? You won't see much, but you should hear the PSU fans start if
    > >> they are going to run.

    >
    > >> 5) Replace PSU with one from a spare computer or one that you've already
    > >> bought - does it come on?

    >
    > >> 6) If nothing working after #5, then it's down to the CPU/Motherboard -
    > >> your call, and nothing from a Multimeter will give you any hope here, it
    > >> won't tell you a dang thing.

    >
    > >> So, without a multimeter, we've spent 10 minutes to test that you have a
    > >> bad CPU/Motherboard.

    >
    > >> If at any point above you saw signs of life, well, you start doing the
    > >> reverse one device at a time until it stops - another 10 minutes. In less
    > >> than 30 minutes you can strip the PC down to what is failed and all
    > >> without any other parts, except the PSU that you can often get anywhere in
    > >> a reasonably size city and you don't need to know how to use a PSU.

    >
    > > What happens is that I press the power button, the front power light
    > > on the tower comes on, the PSU fan and the CPU fan start to spin, then
    > > a few seconds later, it all turns off.
    > > I'm not at it right now, so I can't try it with all of your above
    > > suggestions, however, I have already tried it with another PSU (that
    > > worked the last time it was in a PC), tried
    > > unplugging the hard drives (one at a time), and tried it without the
    > > CPU fan, as well as a different CPU fan, and the results have been
    > > exactly the same each time.

    >
    > Don't try things ONE at a time, disconnect device 1, then test, then
    > device 2 (not reconnecting device 1), and keep going - as you could have
    > several devices that are fried.
    >
    > If the new PSU didn't fix it, and you've removed the drives/fans, and
    > you've removed the memory and video - then it's the motherboard/CPU and a
    > multimeter won't tell you which one.
    >
    > --
    > Want to know what PCBUTTS1 is really about?
    > *** WARNING - these links contain foul/pornographic content of an
    > abusive nature created by PCBUTTS1 and still hosted on his public
    > website ***http://www.pcbutts1.com/rlk/rlk.htm, http://www.pcbutts1.com/license.htm...htm,http://www.pcbutts1.com/Loutheasshole.htm
    > All while spamming his company website at:http://www.seedsv.com- Hide quoted text -
    >
    > - Show quoted text -


    And then the only way to tell if it's the MB or the CPU would be to
    replace the CPU first and see if it works, and if not, then the MB?
     
    alice, Mar 18, 2007
    #48
  9. alice

    Leythos Guest

    On Sun, 18 Mar 2007 15:56:02 -0700, alice wrote:
    >
    > And then the only way to tell if it's the MB or the CPU would be to
    > replace the CPU first and see if it works, and if not, then the MB?


    That's a toss-up, but, I've seen more bad motherboards than I've seen
    CPU's. In almost 30 years of working on, designing, building computers
    I've seen 2 dead CPU's (since I don't use Cryix or AMD) and in that same
    time I've seen hundreds of dead motherboards.

    Have you looked at the larger capacitors on your computer?

    http://home.earthlink.net/~doniteli/index27.htm

    That link will shows you what a BAD (expanded) cap would look like (as
    well as other failures) - if you have one of those, even slightly bulging,
    it's a motherboard 99% chance.

    I would start with the motherboard, but I don't know what brand/model you
    have or what CPU either.

    --
    Want to know what PCBUTTS1 is really about?
    *** WARNING - these links contain foul/pornographic content of an
    abusive nature created by PCBUTTS1 and still hosted on his public
    website ***
    http://www.pcbutts1.com/rlk/rlk.htm ,
    http://www.pcbutts1.com/license.htm ,
    http://www.pcbutts1.com/downloads/max.htm ,
    http://www.pcbutts1.com/downloads/mpv.htm ,
    http://www.pcbutts1.com/downloads/wtcpcb.htm ,
    http://www.pcbutts1.com/cracks.htm ,
    http://www.pcbutts1.com/Loutheasshole.htm
    All while spamming his company website at: http://www.seedsv.com
     
    Leythos, Mar 18, 2007
    #49
  10. alice

    w_tom Guest

    On Mar 18, 6:42 pm, "alice" <> wrote:
    > What happens is that I press the power button, the front power light
    > on the tower comes on, the PSU fan and the CPU fan start to spin, then
    > a few seconds later, it all turns off.


    Leythos suggests shotgunning. Shotgunning can further complicate
    the problem as you have already seen. We sometimes have to teach
    people to stop doing this. Sometimes they learn after making more
    problems. But then Leythos need only charge the customer for new
    parts - blaming his damage on failed parts. Just another reason why
    we teach techs to not shotgun - AND why auto mechanics that shotgun
    have a poor employment history.

    What do voltages do when system is powered on? Learn and post
    those numbers before knowing what to do next. Is problem the
    motherboard and CPU? That will become apparent by steppng through the
    problem.

    But until the power supply 'system' is confirmed with the
    multimeter, even shotguning may be time wasted and more created
    complications.
     
    w_tom, Mar 19, 2007
    #50
  11. alice

    alice Guest

    On Mar 18, 5:43 pm, "w_tom" <> wrote:
    > On Mar 18, 6:42 pm, "alice" <> wrote:
    >
    > > What happens is that I press the power button, the front power light
    > > on the tower comes on, the PSU fan and the CPU fan start to spin, then
    > > a few seconds later, it all turns off.

    >
    > Leythos suggests shotgunning. Shotgunning can further complicate
    > the problem as you have already seen. We sometimes have to teach
    > people to stop doing this. Sometimes they learn after making more
    > problems. But then Leythos need only charge the customer for new
    > parts - blaming his damage on failed parts. Just another reason why
    > we teach techs to not shotgun - AND why auto mechanics that shotgun
    > have a poor employment history.
    >
    > What do voltages do when system is powered on? Learn and post
    > those numbers before knowing what to do next. Is problem the
    > motherboard and CPU? That will become apparent by steppng through the
    > problem.
    >
    > But until the power supply 'system' is confirmed with the
    > multimeter, even shotguning may be time wasted and more created
    > complications.


    OK, here are the numbers, assuming I'm using the meter correctly
    (black into COM and red into V)
    Purple wire - 5.08
    Green - 0 / 2.57 when on
    Gray - 0
    Yellow - 0 / .06 when on
    there are several orange and red wires, so I'm not sure which ones to
    measure.
     
    alice, Mar 20, 2007
    #51
  12. alice

    w_tom Guest

    On Mar 20, 2:38 pm, "alice" <> wrote:
    > OK, here are the numbers, assuming I'm using the meter correctly
    > (black into COM and red into V)
    > Purple wire - 5.08
    > Green - 0 / 2.57 when on
    > Gray - 0
    > Yellow - 0 / .06 when on
    > there are several orange and red wires, so I'm not sure which ones to
    > measure.


    Measure any orange wire in that connector - wires are all same.
    Measure any red wire - same.

    Purple wire means power supply controller has sufficient voltage to
    make appropriate decisions. (This voltage is why nothing can be
    removed or disconnected until power cord is removed from AC
    receptacle.)

    Power supply controller is telling power supply to power on. That
    voltage is sufficient - but also raises suspicion (not typically high
    enough but enough to tell power supply to turn on). This we know.
    Power supply controller is working properly.

    Gray wire says a power supply voltage monitor declares a problem.
    Well, computer sees this signal and therefore will not even start.
    Forget problems with CPU, peripherals, memory, etc. Something is
    happening on red, orange, and yellow wires. Problem is severely
    isoated down to a few items.

    Yellow wire implies - only implies - a problem on 12 volt devices.
    Other numbers on red and orange are required to say more. That is 0.6
    volts always? Or does yellow wire voltage first go higher - monitored
    as power switch is pressed? How it goes from 0 to 0.6 is relevant.

    We don't yet know whether this is power supply or other - missing
    numbers. But we now have specific facts to trace problem down to but
    a few items. With red and orange wire numbers, then we can move on.
     
    w_tom, Mar 21, 2007
    #52
  13. alice

    w_tom Guest

    On Mar 20, 2:38 pm, "alice" <> wrote:
    > OK, here are the numbers, assuming I'm using the meter correctly
    > (black into COM and red into V)
    > ...
    > Green - 0 / 2.57 when on


    Actually I think you have this backwards. 2.57 when off and
    something below zero when power switch is pressed. Actually, the
    number should have been something like 0.2 - something just above zero
    and below 0.7 volts. Not a significant concern if as noted - if your
    post was only a typo error.
     
    w_tom, Mar 21, 2007
    #53
  14. alice

    alice Guest

    On Mar 20, 10:56 pm, "w_tom" <> wrote:
    > On Mar 20, 2:38 pm, "alice" <> wrote:
    >
    > > OK, here are the numbers, assuming I'm using the meter correctly
    > > (black into COM and red into V)
    > > ...
    > > Green - 0 / 2.57 when on

    >
    > Actually I think you have this backwards. 2.57 when off and
    > something below zero when power switch is pressed. Actually, the
    > number should have been something like 0.2 - something just above zero
    > and below 0.7 volts. Not a significant concern if as noted - if your
    > post was only a typo error.


    No, that's no typo, that's what it measured.
    I assume you mean the switch on the PSU itself, and I assume all the
    other plugs from the PSU are to not be plugged in when I do this,
    right?
     
    alice, Mar 26, 2007
    #54
  15. alice

    alice Guest

    On Mar 26, 10:32 am, "alice" <> wrote:
    > On Mar 20, 10:56 pm, "w_tom" <> wrote:
    >
    > > On Mar 20, 2:38 pm, "alice" <> wrote:

    >
    > > > OK, here are the numbers, assuming I'm using the meter correctly
    > > > (black into COM and red into V)
    > > > ...
    > > > Green - 0 / 2.57 when on

    >
    > > Actually I think you have this backwards. 2.57 when off and
    > > something below zero when power switch is pressed. Actually, the
    > > number should have been something like 0.2 - something just above zero
    > > and below 0.7 volts. Not a significant concern if as noted - if your
    > > post was only a typo error.

    >
    > No, that's no typo, that's what it measured.
    > I assume you mean the switch on the PSU itself, and I assume all the
    > other plugs from the PSU are to not be plugged in when I do this,
    > right?


    So the red and orange wires both jump to .02, then down to 0.
     
    alice, Mar 26, 2007
    #55
  16. alice

    PeeCee Guest

    "alice" <> wrote in message
    news:...
    > On Mar 26, 10:32 am, "alice" <> wrote:
    >> On Mar 20, 10:56 pm, "w_tom" <> wrote:
    >>
    >> > On Mar 20, 2:38 pm, "alice" <> wrote:

    >>
    >> > > OK, here are the numbers, assuming I'm using the meter correctly
    >> > > (black into COM and red into V)
    >> > > ...
    >> > > Green - 0 / 2.57 when on

    >>
    >> > Actually I think you have this backwards. 2.57 when off and
    >> > something below zero when power switch is pressed. Actually, the
    >> > number should have been something like 0.2 - something just above zero
    >> > and below 0.7 volts. Not a significant concern if as noted - if your
    >> > post was only a typo error.

    >>
    >> No, that's no typo, that's what it measured.
    >> I assume you mean the switch on the PSU itself, and I assume all the
    >> other plugs from the PSU are to not be plugged in when I do this,
    >> right?

    >
    > So the red and orange wires both jump to .02, then down to 0.
    >


    Alice

    May I suggest a 'Google' session to skill up on Power supplies:

    209,000 hits for 'ATX power supply tutorial' :
    http://www.google.co.nz/search?hl=en&q=atx power supply tutorial&btnG=Google Search&meta=

    Pin connections http://www.hardwaresecrets.com/article/181/10

    1,080,000 hits for 'testing atx power supply'
    http://www.google.co.nz/search?hl=en&q=testing atx power supplies&btnG=Search&meta=

    A good trouble shooting chart: http://www.fonerbooks.com/power.htm

    Best
    Paul.
     
    PeeCee, Mar 26, 2007
    #56
  17. alice

    w_tom Guest

    On Mar 26, 1:32 pm, "alice" <> wrote:
    > No, that's no typo, that's what it measured.
    > I assume you mean the switch on the PSU itself, and I assume all the
    > other plugs from the PSU are to not be plugged in when I do this,
    > right?


    When green wire is 2.57, then power supply is told "do not turn
    on". When front panel power switch is pressed, then green wire must
    drop to near zero - and stay there. When green wire drops to near
    zero, only then is power supply told to power on. Only when front
    power switch is pressed and green wire drops to zero - that is when
    voltage on red, orange and yellow wires are measured.

    Never turn off that power switch on supply - if supply has one of
    those power switches. Most supplies do not have that switch because
    that switch is always on.

    If your measurements were during that power switch trip, then
    voltages did exactly as designed - keep computer off. That is not the
    switch you power on the computer (right?). Power switch on front -
    that is what you use?

    When power switch is not pressed (computer is off) and when power
    switch is pressed - both times the purple wire was always near five
    volts. Always at 5 volts as long as power cord connects to AC mains.
    Was that purple wire only near 5 volts when your pressed a power
    switch? If yes, then you are pressing a power switch that 1) really
    should not exist, 2) must always be on, and 3) we still need numbers
    using the front panel power switch.
     
    w_tom, Mar 26, 2007
    #57
  18. alice

    w_tom Guest

    On Mar 26, 3:27 pm, "alice" <> wrote:
    > So the red and orange wires both jump to .02, then down to 0.


    Again, that green wire is above 2.4 volts when computer is off.
    When computer is off, that purple wire must measure more than 4.87
    volts. When computer is on, that purple wire measures 4.87 volts.
    When the power switch is pressed, the green wire drops to some number
    just above zero. All those requirements are correct as defined in the
    procedure in http://tinyurl.com/yvf9vh .
     
    w_tom, Mar 26, 2007
    #58
  19. alice

    alice Guest

    On Mar 26, 3:49 pm, "w_tom" <> wrote:
    > On Mar 26, 3:27 pm, "alice" <> wrote:
    >
    > > So the red and orange wires both jump to .02, then down to 0.

    >
    > Again, that green wire is above 2.4 volts when computer is off.
    > When computer is off, that purple wire must measure more than 4.87
    > volts. When computer is on, that purple wire measures 4.87 volts.
    > When the power switch is pressed, the green wire drops to some number
    > just above zero. All those requirements are correct as defined in the
    > procedure inhttp://tinyurl.com/yvf9vh.


    So then the sqare, 4 pin plug should be plugged into the MB, and
    nothing else? I'm assuming this since it would seem to be the only way
    that the power switch on the front and the PSU would be communicating.
     
    alice, Mar 27, 2007
    #59
  20. alice

    w_tom Guest

    On Mar 26, 8:11 pm, "alice" <> wrote:
    > So then the sqare, 4 pin plug should be plugged into the MB, and
    > nothing else? I'm assuming this since it would seem to be the only way
    > that the power switch on the front and the PSU would be communicating.


    A twenty something plug must connected motherboard. That is the
    connector with purple, green, grey, red ... etc wires. If you have a
    4 pin square plug, that also connects to motherboard somewhere near
    CPU. Square plug is special power for CPU.

    Power switch connects to motherboard. Motherboard connects to power
    supply via the twenty something nylon connector with green, grey,
    orange, red, etc wires.

    All these wires should never be disconnected; must be connected for
    machine to have worked. Only connectors that might not be connected
    are some rectangular ones (with one corner sliced off) that have red,
    black, and yellow wires. Rectangular connectors would be optional
    connectors for future peripherals (ie DVD player). That option
    connector is irrelevant.

    Your concern is only a connector from power supply to motherboard
    with twenty something wires. That is where all measurements are
    made. That is how front panel power switch commands power supply.

    Were you using front panel switch to power on computer during
    measurements?
     
    w_tom, Mar 27, 2007
    #60
    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. Silverstrand

    HEC Ace Power 580UB 580Watt power supply

    Silverstrand, Jul 23, 2005, in forum: Front Page News
    Replies:
    0
    Views:
    868
    Silverstrand
    Jul 23, 2005
  2. Silverstrand

    AOpen Prima Power AO700-12ALN 700W Power Supply Review

    Silverstrand, Aug 27, 2005, in forum: Front Page News
    Replies:
    5
    Views:
    1,507
    unholy
    Aug 29, 2005
  3. Silverstrand
    Replies:
    0
    Views:
    2,130
    Silverstrand
    Feb 21, 2006
  4. that one

    won't power on: power supply or motherboard?

    that one, Apr 19, 2005, in forum: Computer Support
    Replies:
    8
    Views:
    7,465
    Blinky the Shark
    Apr 21, 2005
  5. Southern Kiwi

    I WON, I WON, I WON !!!!!

    Southern Kiwi, Jun 12, 2006, in forum: NZ Computing
    Replies:
    10
    Views:
    671
    Crumb
    Jun 13, 2006
Loading...

Share This Page