In article <1078861376.811598 DeleteThis @vanessa.tekla.fi>, "Tapio Kohonen"
<tko DeleteThis @tekla.com> wrote:
> Hi
>
> Second try...in case someone could help me...
>
> Problem: When I switch power on, fans start spinning, but PC powers off
> immediately.
>
> I have a new ASUS P4P800-VM and P4 2.8C installed in a new case.
> Video card and RAM also installed.
> No hard drive or CD connected.
> Also, only power on and speaker leads are connected to the mainboard.
>
> No beeps.
> Same behaviour, if I remove Video card or RAM, or without video card
> (there is onboard video), or with different RAM.
> I also tried so that main board was not installed in the case, to avoid
> possible short circuit.
> I have tried also with another PSU.
> In every case same happens, fans start but PC shuts down.
> I have also reinstalled the CPU and cleared the mainboard CMOS.
>
> What could be wrong? Bad CPU or bad mainboard?
> Any way to know without buying second item of both...?
>
> If I remove CPU and CPU fan, again video card fan starts when I put power
> on,
> then PC powers off immediately.
> Well, maybe this does not mean anything, or could the CPU be bad?
> Could boxed Intel P4 really be dead on arrival?
>
> Thanks for any suggestions.
>
> Tapio
1) Are both the ATX 20 pin power connector and the 2x2 +12V square
power connector both connected to the board ? The processor cannot
operate without the 2x2 connected. With no power_good signal, I don't
think the computer will even issue the powerup beep without the 2x2.
Now I'll walk through the power states:
a) Computer unplugged. CMOS runs off battery. AFAIK, a dead battery
should not cause any problems, other than forgotten settings.
b) Plug the computer in and switch it on using the switch on the
back of the computer. This causes the +5VSB standby output on
the PS to be energised. Components should not be added or removed
with +5VSB operating. This is one reason for recommending
unplugging the computer whenever maintenance inside the case
is required.
On some motherboards, there is a BIOS setting for rebooting the
computer after a power failure. When you flip the back power
switch on these computers, you'll hear the computer main power come
on for a second or two. This is the amount of time required
for the processor to start up, read the BIOS setting for whether
or not automatic reboot has been selected - if not, the computer
will turn off the power again and return to its sleeping
state.
c) Now, momentarily press the power button on the front of the case.
This signal is latched by a chip on the motherboard, to become
the PS_ON# signal on the ATX 20 pin power connector. Even though
the power switch was touched for only a moment, the PS_ON# signal
is grounded by the motherboard, for as long as the computer is
to operate.
The motherboard chip also has a timer, for timing how long the
button is pushed. If you later push the button for 4 seconds or
more, that is recognized as a request to change the power state
of the computer, based on whatever you've set as the BIOS
option for the power switch. I'm not sure whether this timer
is used when the power button is initially pushed or not.
d) Now, the computer boots, as when PS_ON# is grounded by the
motherboard, the power supply enables all of its outputs.
There are a number of power conversion circuits on the
motherboard, and when the PS is fully turned on, these
circuits power up. If all of them reach a stable voltage,
then a signal called Power_good is sent to the chipset,
indicating that the processor can start reading the flash
code. The speaker on the computer will beep once when the
Power_good first appears. If you don't hear the initial
beep, maybe this is because one of the power conversion
circuits on the motherboard is shorted or broken.
e) Later, when you select "shutdown" in Windows, one of the
last things an ATX computer does, is deassert the PS_ON#
signal to the power supply. The power supply returns to
just supplying the +5VSB signal. When the computer is
"suspend to RAM", the +5VSB keeps the DRAM powered, so
the contents of RAM won't be lost. Similarly, as long
as +5VSB is running, there is no drain on the CMOS battery.
If you think through that sequence, maybe a few ideas will come
to mind, such as:
1) Stuck power switch (i.e. it isn't momentary like it should
be).
2) Stuck reset switch or signal. This will prevent the processor
from executing its first instruction, and I think it may
also prevent the Power_good beep from the speaker.
3) Bad power supply (if interval from power up to shutdown is
not related to the 4 second timer perhaps).
4) Shorted power conversion circuit on the motherboard.
Something else that can affect operation, is if an IDE cable
is partially seated in its connector - I had one motherboard
that misbehaved at power up, while the cable was askew.
AFAIK, with no components on the motherboard, pushing the power
button should cause the power to come on and stay on.
Also, if you run out of things to try, there is a simple test
for the power supply. If you disconnect the power supply from
the motherboard, then take a wire or paper clip and short
the PS_ON# signal to an adjacent COM pin, you can check to
see whether the power supply will stay running on its own.
As long as PS_ON# is connected to COM, the power supply fan
should keep running. Note that this is not a highly recommended
practice, as the power supply should really have a load on
its outputs. If you have a spare computer, you could also
use the suspect power supply, to power one of your other
computers. If you only have the one computer, then the
PS_ON# test is better than nothing.
You can see the signal pinouts for an ATX power supply here.
(This is for an Antec PP352X supply.)
<a style='text-decoration: underline;' href="http://media-server.amazon.com/media/mole/MANUAL000009643.pdf" target="_blank">http://media-server.amazon.com/media/mole/MANUAL000009643.pdf</a>
HTH,
Paul<!-- ~MESSAGE_AFTER~ -->
FAQ
Profile
Private Messages
Log in
