«

On Mar 18, 5:43 pm, "w_tom" <w_t...@usa.net> wrote:
> 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.

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.

On Mar 20, 2:38 pm, "alice" <a...@fearofdolls.com> 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.

On Mar 20, 2:38 pm, "alice" <a...@fearofdolls.com> 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.

On Mar 20, 10:56 pm, "w_tom" <w_t...@usa.net> wrote:
> On Mar 20, 2:38 pm, "alice" <a...@fearofdolls.com> 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?

On Mar 26, 10:32 am, "alice" <a...@fearofdolls.com> wrote:
> On Mar 20, 10:56 pm, "w_tom" <w_t...@usa.net> wrote:
>
> > On Mar 20, 2:38 pm, "alice" <a...@fearofdolls.com> 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" <> wrote in message
news: oups.com...
> On Mar 26, 10:32 am, "alice" <a...@fearofdolls.com> wrote:
>> On Mar 20, 10:56 pm, "w_tom" <w_t...@usa.net> wrote:
>>
>> > On Mar 20, 2:38 pm, "alice" <a...@fearofdolls.com> 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...e+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...G=Search&meta=

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

Best
Paul.

On Mar 26, 1:32 pm, "alice" <a...@fearofdolls.com> 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.

On Mar 26, 3:27 pm, "alice" <a...@fearofdolls.com> 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 .

On Mar 26, 3:49 pm, "w_tom" <w_t...@usa.net> wrote:
> On Mar 26, 3:27 pm, "alice" <a...@fearofdolls.com> 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.

On Mar 26, 8:11 pm, "alice" <a...@fearofdolls.com> 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?