symtoms of a dead bios on a p2b, please.

  • Thread starter Thread starter eric
  • Start date Start date
E

eric

This is about an Asus p2b rev 1.10 which I have on my work table at
the moment. No POST. Using a bare minimum of hardware, I've swapped
out everything: memory sticks, cpu's, video cards and hard drives ad
finitum. All the hardware works fine on other boards. The FSB jumpers
are set for 66Mhz which should boot anything I throw at it. (I have
altered these to 100Mhz just for variety, every now and then)

On this mobo in particular what I get is a whir of fans from the HSF
and the PSU, the hard drive makes noise as if going to boot and then
stops, nothing but a black screen and no further activity. Oh yeah, it
also starts immediately upon being pluged in.

Does this sound like a dead bios? I could hot swap to find out, but
first looking for opinions.

thanks, eric.
 
You need a video card, ram, a PC speaker connected, and a keyboard. If you
receive any kind of beep from the PC speaker, the bios exists.
 
"Lil' said:
You need a video card, ram, a PC speaker connected, and a keyboard. If you
receive any kind of beep from the PC speaker, the bios exists.
Click to expand...

"Oh yeah, it also starts immediately upon being pluged in."

That is not a good sign. On my P2B-S, this happened once when an IDE
cable was only half installed (I bumped it out of place during
assembly). Some voltage is being upset when that happens, but I
doubt I could describe an exact mechanism. In any case, the problem
you have is more primitive than a BIOS problem.

If it wasn't for the "starts immediately" problem, I'd be looking
for one of the onboard regulated supplies failing, like Vcore or
some termination voltage etc. These are all tied together into a
status signal called "Power_Good" or the like. If any supply isn't
within spec, the processor is never given a chance to come out
of reset. Debugging a problem like that, isn't easy, as on some
440BX boards, four regulated supplies come from the same chip,
and the status signals are not separately available for each
voltage. In a properly equipped lab, I would use a four channel
storage scope, to figure out what is going on.

Paul
 
Delete the RAM - if the BIOS is alive, it will complain with long,
repeated beeps.
"Oh yeah, it also starts immediately upon being pluged in."

That is not a good sign. On my P2B-S, this happened once when an IDE
cable was only half installed (I bumped it out of place during
assembly). Some voltage is being upset when that happens, but I
doubt I could describe an exact mechanism. In any case, the problem
you have is more primitive than a BIOS problem.
Click to expand...

IME there are any number of problems which can cause "starts
immediately", including improperly seated CPUs, PCI cards, drive cables
etc. - even P2B systems with no problems have been known to do this on
occasion. However, I agree the problem is almost certainly a hardware
failure on the mainboard, especially given the amount of swapping done
with no change in symptoms. (I am assuming PSU was swapped or known
good, although the OP didn't mention it).
If it wasn't for the "starts immediately" problem, I'd be looking
for one of the onboard regulated supplies failing, like Vcore or
some termination voltage etc. These are all tied together into a
status signal called "Power_Good" or the like. If any supply isn't
within spec, the processor is never given a chance to come out
of reset. Debugging a problem like that, isn't easy, as on some
440BX boards, four regulated supplies come from the same chip,
and the status signals are not separately available for each
voltage. In a properly equipped lab, I would use a four channel
storage scope, to figure out what is going on.
Click to expand...

The fact the CPU fan continues to spin means onboard supplies are
*probably* OK, but a DMM or (preferably) a scope would be needed to
confirm. Electrolytic filter capacitors are rarely a problem on P2Bs,
but they do fail occasionally which is the reason a scope is preferred
for checking voltages - ripple tells the story.

You really need a POST card, DMM, and scope to debug these symptoms.
First look for BIOS activity with the POST card (none expected given the
symptoms, but worth doing anyway). Then check onboard voltages which
don't have corresponding LEDs on the POST card (typically Vcore, Vtt,
and the 2.5v clock chip supply), then FSB and PCI clock signals with the
scope. Finally, verify that Vcc is present at the appropriate pins on
all the major chips - several are protected by zener crowbars close by.

IME ~60% of dead P2Bs pass all the above tests with no problem found, at
which point I conclude the BX northbridge is dead and stick it in the
parts bin - I don't have a storage scope to confirm the diagnosis, but
on the one occasion I had access to a full lab, the storage scope
confirmed it on 6 of 7 boards tested. The other one had a dead hardware
monitoring chip and was fixed, but I'm not sure it was worth the effort :-)

The other ~40% are usually easily repaired - the most common failure
modes are blown FETs (sometimes with colateral damage), tripped
crowbars, and shorted decoupling capacitors.

HTH

P2B
 
I swapped out the bios chip and I get the exact same results.

Put it in a tower and i didn't get any beeps. The techniques descibed
by these replies (most helpful) go beyond my technical abilities and
a lack of the hardware mentioned is also a factor.

I declared the mobo defective and have it stored. I sure wish I was
adept at going deeper because I could turn a couple of defective
boards into one decent one perhaps.

Oh well, life goes on...

eric
 
Back
Top