«

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

> On Mar 16, 6:24 am, Leythos <v...@nowhere.lan> 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.

On Mar 17, 1:47 pm, Leythos <v...@nowhere.lan> 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.

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

On Mar 18, 6:04 am, Leythos <v...@nowhere.lan> 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,...theasshole.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.

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

> On Mar 18, 6:04 am, Leythos <v...@nowhere.lan> 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

On Mar 18, 3:19 pm, Leythos <v...@nowhere.lan> wrote:
> On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
> > On Mar 18, 6:04 am, Leythos <v...@nowhere.lan> 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,...theasshole.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.

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

> On Mar 18, 3:19 pm, Leythos <v...@nowhere.lan> wrote:
>> On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
>> > On Mar 18, 6:04 am, Leythos <v...@nowhere.lan> 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

On Mar 18, 3:50 pm, Leythos <v...@nowhere.lan> wrote:
> On Sun, 18 Mar 2007 15:42:28 -0700, alice wrote:
> > On Mar 18, 3:19 pm, Leythos <v...@nowhere.lan> wrote:
> >> On Sun, 18 Mar 2007 11:42:35 -0700, alice wrote:
> >> > On Mar 18, 6:04 am, Leythos <v...@nowhere.lan> 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,...theasshole.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?

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

On Mar 18, 6:42 pm, "alice" <a...@fearofdolls.com> 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.