I have an ASUS P4P800-E motherboard with a Pentium 4 processor, running
WinXP. I have been using standby on a regular basis every time I'm away
from the computer. I've also got the power-on via realtime clock set
for 5:30 AM everyday, so the system can do it's backup. This has been
working flawlessly for over a year. Recently standby quit working
spontaneously: Windows says "preparing to standby" then the system
powers down completely. Also the power-on via realtime clock doesn't
work anymore. Everything else about the system works just fine. So far
1. Check all Windows and BIOS settings to make sure standby is still
2. Replaced the CMOS battery
3. Reset the CMOS via jumper on the motherboard
4. re-flashed the BIOS with the latest version.
It still doesn't work. Actually standby worked ONCE after the BIOS
flash, but then quit again, and didn't work after a second BIOS flash.
I contacted ASUS tech support and they said it sounds like a motherboard
problem and suggested I send it in for a test. I'm not interested in
paying to have them fix it, as the motherboard is probably not worth the
cost of shipping and repair (it's long out of warranty). But I would
like to fix it if possible. Anyone have any suggestions on where to
start? Is there any detailed hardware diagnostic software for this
motherboard? I'm no stranger to the soldering iron, so I'm ready to have
at it if I knew where to start.
A soldering iron ? I think this job needs a hot air rework station...
(That's how you remove and replace the Southbridge.)
You can get a PDF schematic from Intel, for a board of this generation.
The problem with that schematic, is the Port Angeles might not be a
"real" chip we can get a datasheet for. So it's hard to say what
some of the logic signals from page 34 actually do, when they connect
to Port Angeles.
When the machine goes to standby, the PS_ON# signal goes to the deasserted
state, meaning 3.3V, 5V, 12V main rails are switched off. All that is left
is +5VSB. The +5VSB is regulated down, to a voltage suited to running the
Northbridge and RAM, so that the standby contents can be maintained.
When the computer comes back up later, there must be some means by which the
BIOS knows the system state. So there have to be some ICH5 status bits that
say "I was put to sleep from S3 at last power off". It might not be external
signals that you can check. You'll have to pull a copy of the ICH5/ICH5R
datasheet, and look at the docs for some of the status signals you can
see identified on page 34 of that schematic.
On page 68, you can see the PS_ON# signal coming out of Port Angeles. And
some input to Port Angeles, would be carrying a signal triggered by the
RTC timer, amongst other things.
So the problem is either ICH5 (Southbridge, BGA, hot air to remove), or
the equivalent of the Port Angeles (SuperI/O chip of some sort). On page
67, you can see the other half of Port Angeles, and it has things like
the Floppy interface, so that makes the chip type SuperI/O.
So that's roughly where you start, and I've left all the hard
work to you :-)