As of late, if I unplug my PC I can not power back up unless I jump across the
black/green pins on the 20 pin mb connector and them press my start button.
Under mormal shutdown conditions in which I power off it restarts fine.. it's
only whne I unplug it this happens.
Any ideas?
--
--------------------------------- --- -- -
Posted with NewsLeecher v3.8 Final
Web @ http://www.newsleecher.com/?usenet
------------------- ----- ---- -- -
brian@yahoo.com wrote:
> As of late, if I unplug my PC I can not power back up unless I jump across the
> black/green pins on the 20 pin mb connector and them press my start button.
>
> Under mormal shutdown conditions in which I power off it restarts fine.. it's
> only whne I unplug it this happens.
>
> Any ideas?
>
The power supply is divided into two pieces. One tiny power supply, creates
+5VSB. +5VSB is used to power the motherboard logic, including the logic
handling the power switch. +5VSB must be present, in order for the power
supply to be started.
The second part of the power supply, has the 3.3, 5.0, 12V, -12V on it.
A voltage used to make the +5VSB, may also be used to operate the supervisor
for the other rails. If +5VSB is defective or shorted out, you might well
not see operation of the other half.
For reference, this is the only ATX power supply schematic that I've found
that is available for free.
This could also be related to the condition of the CMOS battery.
Use a multimeter, and check the CR2032 coin cell. A good value
is around 3 volts or so. Below 2.4V might cause a problem with
remembering the settings, or keeping the RTC clock running.
Replace the battery, and see if the symptoms change. It could
be, that your battery is completely flat, and that is preventing
the computer from starting normally.
chat00chat@gmail.com wrote:
> I have an idea... stop unplugging it.
>
> It sounds like you might have a faulty power supply... luckily they
> are cheap.
chat00chat@gmail.com wrote:
> I have an idea... stop unplugging it.
>
> It sounds like you might have a faulty power supply... luckily they
> are cheap.
On Jun 13, 10:09*am, brian (br...@yahoo.com) wrote:
> As of late, if I unplug my PC I can not power back up unless I jump across the
> black/green pins on the 20 pin mb connector and them press my start button..
>
> Under mormal shutdown conditions in which I power off it restarts fine.. it's
> only whne I unplug it this happens.
According to what was posted, the power supply is typically not the
problem. Problem is more likely in another section of the power
supply 'system' - the supply controller. Doing what Paul has
suggested will help trace that problem to its source. Another test
uses the multimeter to measure voltages on the orange, red, purple,
and yellow wires both before and when that switch is pressed (after
unplugging). Also useful is to measure voltages on green and gray
wires as power switch is pressed after an unplug. Report both the VDC
numbers and how those voltages responded as power switch is pressed.
Numbers in volts and seconds will make useful replies. Time, cost,
and labor to do what Paul has suggested and to obtain these numbers
will be less than 'swapping a supply only on a whim'.
On Jun 13, 10:09=A0am, brian (br...@yahoo.com) wrote:
>> As of late, if I unplug my PC I can not power back up unless I jump across the
>> black/green pins on the 20 pin mb connector and them press my start button.
>>
>> Under mormal shutdown conditions in which I power off it restarts fine..
>>it's only when I unplug it this happens.
>W_Tom Replied
> According to what was posted, the power supply is typically not the
>problem. Problem is more likely in another section of the power
>supply 'system' - the supply controller. Doing what Paul has
>suggested will help trace that problem to its source. Another test
>uses the multimeter to measure voltages on the orange, red, purple,
>and yellow wires both before and when that switch is pressed (after
>unplugging). Also useful is to measure voltages on green and gray
>wires as power switch is pressed after an unplug. Report both the VDC
>numbers and how those voltages responded as power switch is pressed.
>Numbers in volts and seconds will make useful replies. Time, cost,
>and labor to do what Paul has suggested and to obtain these numbers
>will be less than 'swapping a supply only on a whim'.
>from paul
>The power supply is divided into two pieces. One tiny power supply, creates
>+5VSB. +5VSB is used to power the motherboard logic, including the logic
>handling the power switch. +5VSB must be present, in order for the power
>supply to be started.
>The second part of the power supply, has the 3.3, 5.0, 12V, -12V on it.
>A voltage used to make the +5VSB, may also be used to operate the supervisor
>for the other rails. If +5VSB is defective or shorted out, you might well
>not see operation of the other half.
>For reference, this is the only ATX power supply schematic that I've found
>that is available for free.
>This could also be related to the condition of the CMOS battery.
>Use a multimeter, and check the CR2032 coin cell. A good value
>is around 3 volts or so. Below 2.4V might cause a problem with
>remembering the settings, or keeping the RTC clock running.
>Replace the battery, and see if the symptoms change. It could
>be, that your battery is completely flat, and that is preventing
>the computer from starting normally.
> Paul
Thanks to both of you for your replies. I have not been able to take any voltage measurements yet,
But to follow up"
Is it fair to say that the startup switch itself is not bad, if it were, it would not start
even when grounding the green wire on the 20 pin connection?
Also it seems to me I can rule out the CMOS battery or I would be losing date time and other info in my bios
if I lost power and the CMOS battery were dead.
I forgot to mention, this is an HP PC with a mb with no off/om rocker panel on th PSU. Also when this first
started happening simply unplugging the 20 pin connector and reconnecting it resloved the issue.
Luckily for me I don't lose power too often.
Thanks to both of you foe your replies.
Brian
--
--------------------------------- --- -- -
Posted with NewsLeecher v3.8 Final
Web @ http://www.newsleecher.com/?usenet
------------------- ----- ---- -- -
brian@yahoo.com wrote:
> On Jun 13, 10:09=A0am, brian (br...@yahoo.com) wrote:
>>> As of late, if I unplug my PC I can not power back up unless I jump across the
>>> black/green pins on the 20 pin mb connector and them press my start button.
>>>
>>> Under mormal shutdown conditions in which I power off it restarts fine..
>>> it's only when I unplug it this happens.
>
>> W_Tom Replied
>> According to what was posted, the power supply is typically not the
>> problem. Problem is more likely in another section of the power
>> supply 'system' - the supply controller. Doing what Paul has
>> suggested will help trace that problem to its source. Another test
>> uses the multimeter to measure voltages on the orange, red, purple,
>> and yellow wires both before and when that switch is pressed (after
>> unplugging). Also useful is to measure voltages on green and gray
>> wires as power switch is pressed after an unplug. Report both the VDC
>> numbers and how those voltages responded as power switch is pressed.
>> Numbers in volts and seconds will make useful replies. Time, cost,
>> and labor to do what Paul has suggested and to obtain these numbers
>> will be less than 'swapping a supply only on a whim'.
>
>
>>from paul
>> The power supply is divided into two pieces. One tiny power supply, creates
>> +5VSB. +5VSB is used to power the motherboard logic, including the logic
>> handling the power switch. +5VSB must be present, in order for the power
>> supply to be started.
>
>> The second part of the power supply, has the 3.3, 5.0, 12V, -12V on it.
>> A voltage used to make the +5VSB, may also be used to operate the supervisor
>> for the other rails. If +5VSB is defective or shorted out, you might well
>> not see operation of the other half.
>
>> For reference, this is the only ATX power supply schematic that I've found
>> that is available for free.
>
>> http://www.pavouk.org/hw/en_atxps.html
>
>> This could also be related to the condition of the CMOS battery.
>> Use a multimeter, and check the CR2032 coin cell. A good value
>> is around 3 volts or so. Below 2.4V might cause a problem with
>> remembering the settings, or keeping the RTC clock running.
>> Replace the battery, and see if the symptoms change. It could
>> be, that your battery is completely flat, and that is preventing
>> the computer from starting normally.
>
>> Paul
>
> Thanks to both of you for your replies. I have not been able to take any voltage measurements yet,
> But to follow up"
>
> Is it fair to say that the startup switch itself is not bad, if it were, it would not start
> even when grounding the green wire on the 20 pin connection?
>
> Also it seems to me I can rule out the CMOS battery or I would be losing date time and other info in my bios
> if I lost power and the CMOS battery were dead.
>
> I forgot to mention, this is an HP PC with a mb with no off/om rocker panel on th PSU. Also when this first
> started happening simply unplugging the 20 pin connector and reconnecting it resloved the issue.
>
> Luckily for me I don't lose power too often.
>
> Thanks to both of you foe your replies.
>
> Brian
>
This is a diagram of the power switch thing.
PS_ON#
Power -------- Motherboard ------------------ ATX Power
Switch Logic green wire supply
(momentary) (latched (logic 0 ground
state) to turn on)
The power switch does not directly drive the PS_ON# signal. The
power switch is conditioned by motherboard logic, the momentary
contact is latched and stored. That gives a steady level, as
required, on the PS_ON# signal. Whenever PS_ON# is grounded
at zero volts, the power supply fan spins, and the main rails
(3.3V/5V/12V/-12V) come on.
If the CMOS contents and RTC appear good, then the motherboard is
a candidate. Either the motherboard cannot drive PS_ON# to ground
properly, or there is some other problem with it. You could use
a multimeter to investigate what appears on the main power cable.
And see if PS_ON# gets a good level. Say, for example, you
press the front power button, and PS_ON# goes from reading 5.0
volts (keeps ATX PSU turned off), but only drops to 2.0 volts.
That is not low enough to make the power supply turn on. A
reason for this, can be a damaged driver chip on the motherboard,
or alternately, the power supply is using too much pullup current
on the PS_ON# signal. You should also verify the voltage on the
+5VSB pin, and see if while all of this is happening, the
+5VSB rail stays at 5V and doesn't wink out.
Power supply specs (to get wiring harness signal names) - oldest to newest
If the computer case has both a power switch and a reset switch,
you can swap the power switch two pin connector for the reset
switch connector. Then use the reset switch to turn the power
on and off. That is one way to verify the switch works, without
using a multimeter.
Power up problem 
Linear Mode
