Wife's Computer Won't Start

[jim evans]

I posted this problem in the pc-homebuilt group day before yesterday
but didn't thing to crosspost to this group, so I'm also posting here.
Understanding has progressed since my original post, but if you're
interested in the history go to that group and look for this Subject.

When we got home Sunday my wife's computer was hung in shutdown and
had been in that state for a day. Holding down the front panel power
switch would not turn it off. She switched off the rocker switch on
the power supply. After she switched it back on the box is completely
dead. No fans, no lights, no sounds, nothing -- a brick.

With the mainboard power connector disconnected, the PS_ON jumpered,
two hard drives, a floppy drive and a CD drive connected the PS fan
comes on and all voltages look normal.

With the power connector plugged in to the mainboard, and all molexs
disconnected, I checked the PS_ON pin. It was 5v, and never changed
when the power button on the front of the machine was pushed. I know
this button works, so for some reason closing the power on switch does
not cause the PSU to get the power on signal/state.

Is this a hopeless motherboard or might there be a fixable cause?

-- jim

 

[jim evans]

More info

I the power on pins it at the PSU>MB connector.

When the PS_ON is shorted to ground the power supply fans start and I
hear something click-on the in the system, but the moment I remove the
short everything stops.

-- jim

 

[Pen]

More info

I the power on pins it at the PSU>MB connector.

When the PS_ON is shorted to ground the power supply fans start and I
hear something click-on the in the system, but the moment I remove the
short everything stops.

-- jim

Either the front panel switch is bad, easily checked by shorting it out
or the power on flip flop is defective(on the mobo) not easily fixed.

 

[Clark]

I had a strange situation once where I had to actually unplug the power
supply--turning it off did not work, if it even had an off switch.

Clark

 

[jaster]

I posted this problem in the pc-homebuilt group day before yesterday but
didn't thing to crosspost to this group, so I'm also posting here.
Understanding has progressed since my original post, but if you're
interested in the history go to that group and look for this Subject.

When we got home Sunday my wife's computer was hung in shutdown and had
been in that state for a day. Holding down the front panel power switch
would not turn it off. She switched off the rocker switch on the power
supply. After she switched it back on the box is completely dead. No
fans, no lights, no sounds, nothing -- a brick.

With the mainboard power connector disconnected, the PS_ON jumpered, two
hard drives, a floppy drive and a CD drive connected the PS fan comes on
and all voltages look normal.

With the power connector plugged in to the mainboard, and all molexs
disconnected, I checked the PS_ON pin. It was 5v, and never changed when
the power button on the front of the machine was pushed. I know this
button works, so for some reason closing the power on switch does not
cause the PSU to get the power on signal/state.

Is this a hopeless motherboard or might there be a fixable cause?

-- jim

Was working then stopped?

I had a similiar problem. Trouble shooting I disconnected everything
except memory, video, cpu fan. When that failed to boot, I swapped
memory then video cards. Turned out the video card had died.

 

[jim evans]

Either the front panel switch is bad, easily checked by shorting it out
or the power on flip flop is defective(on the mobo) not easily fixed.

Shorting the front panel motherboard connector pins has no effect. I
guess that leaves the motherboard. In 25 years of using and
maintaining PC computers this is my first MB failure. If had multiple
failures of every other component, but never a motherboard.

-- jim

 

[Andy]

Did you measure the standby power, 5Vsb?

 

[w_tom]

Better is to make measurements without disconnecting anything. For
example, the green 'power on' wire is above 2 volts disconnected. What
is it when connected - both before power switch is pressed and during a
press. Although this one measurement is not going to report anything
useful, it is how each measurement provides useful facts.

Now what does gray wire do before and in the seconds as power switch
is pressed. Not just 'good' and not just 5 volts. What is the number?
4.87?

And finally what are numbers on one of purple, red, orange, and
yellow wires when and after power switch is pressed?

Is it a motherboard? Maybe. But the point is that in less than 2
minutes, what is failing can be identified without replacing anything.
Furthermore, taking that 5 volt measurement with molex connector
disconnected can look good when the failure still exists. Just another
reason why numbers taken without disconnection anything are so much
more informative.

Once numbers have identified the defective part of the power supply
'system', then a failure might be identified by inspection.
Inspection of an entire motherboard is futile. But once we limit
inspection only to some parts, then inspection why suddenly find that
tiny metal fragment or that slight component bulge.