Problem with P4C800-Deluxe

[Dave Thoma]

Recently I purchased a P4C800 Deluxe motherboard along with a 3.2 Ghz
Northwood P4 to go along with it. I received the parts, installed
everything properly (so I thought), and turned on the PC. As soon as
pressed the power button, something on the motherboard burned up.
After turning it off and looking at it, it appears to be whatever is
(was) to the right of the IEEE 1394 internal controller at the bottom
of the board. The green light still comes on if I turn the MB on, but
I get an (auditory) error message: "Warning CPU Test Failed." I'm not
sure if this was related to the "burn out" or not. If my assumption
is correct and it has related to the firewire controller, it seems
conceivable to me the MB would still operate. The only error I could
figure out that I made was failing to connect the secondary 12v power
supply at first, but it doesn't seem plausible that that would have
caused this problem. So my questions are:

What do you think could have caused this to happen?
Do you think the warning message I received regarding the CPU was
related, or a separate issue entirely?
If the problem is isolated to the IEEE 1394 controller, is it safe to
use the motherboard provided I don't need this feature?

Thanks in advance. I've put together my fair share of computers but
I'm by no means an electrical or computer engineer. I've never had a
problem like this before, and I'm really unsure of how to proceed.

Regards,
Dave Thoma

 

[Paul]

Recently I purchased a P4C800 Deluxe motherboard along with a 3.2 Ghz
Northwood P4 to go along with it. I received the parts, installed
everything properly (so I thought), and turned on the PC. As soon as
pressed the power button, something on the motherboard burned up.
After turning it off and looking at it, it appears to be whatever is
(was) to the right of the IEEE 1394 internal controller at the bottom
of the board. The green light still comes on if I turn the MB on, but
I get an (auditory) error message: "Warning CPU Test Failed." I'm not
sure if this was related to the "burn out" or not. If my assumption
is correct and it has related to the firewire controller, it seems
conceivable to me the MB would still operate. The only error I could
figure out that I made was failing to connect the secondary 12v power
supply at first, but it doesn't seem plausible that that would have
caused this problem. So my questions are:

What do you think could have caused this to happen?
Do you think the warning message I received regarding the CPU was
related, or a separate issue entirely?
If the problem is isolated to the IEEE 1394 controller, is it safe to
use the motherboard provided I don't need this feature?

Thanks in advance. I've put together my fair share of computers but
I'm by no means an electrical or computer engineer. I've never had a
problem like this before, and I'm really unsure of how to proceed.

Regards,
Dave Thoma

What I see between the VT6307 Firewire controller and the Promise
20378, is a three terminal regulator. There are fixed voltage
versions and adjustable versions, and this one could be adjustable.
The picture in the manual shows an AMD2910 regulator

http:www.advanced-monolithic.com/pdf/ds2910.pdf

If I had to guess what the function might be, perhaps this regulator
uses +12V (running from the same track as the Firewire header), to make
a clean +5V. Maybe the Promise 20378 needs +5V for the IDE connector ?
I don't know if 5V I/O is the standard for IDE or not.

Maybe something shorted on that IDE connector ? Did you have something
connected to the Promise ?

Another possibility, is you wired the Firewire header to the computer
case, connected a powered Firewire device that feeds power into VP
and VG. If the computer case wiring is screwed up, and VP and VG
get reversed, then the power supply on the external Firewire device
and the computer +12V have a fight. A lot of current flows. The +12V
track, which leads from the single +12V pin on the ATX connector, fries
somewhere. The reversed voltage from the external device continues
to flow, and the reverse bias applied across the protection diode (near
the bottom edge of the board) and the regulator proceeds to fry them.
This is a real long short.

I think an RMA is in order here. This could be an internal fault in
the motherboard, but more likely whatever you connected to the
motherboard is what did it.

When the new motherboard shows up, I recommend not using case wiring
for your Firewire connector (at least until you purchase a multimeter,
and use the ohmmeter ranges to check the case wiring is correct). I have
a couple of Antec cases here, different models, and both have cabling
errors. Fortunately, none of the errors involved power supply signals.

The other thing to watch for, is whatever drives you connected to the
Promise controller. Maybe they were at fault.

The three terminal regulator is thermally protected - if the load on
the regular exceeds its capabilities, it just shuts off its output.
But, it is not protected against every calamity. For example, if the
PSU puts out more than 15V (the max input voltage for the AMS2910),
the AMS2910 will simply pop and burn. Also, if the voltage applied
to it is reversed, the devices will similarly disappear in a whiff of
computer smoke. The regulator is only protected against the expected,
and not the unexpected.

RMA the board and move on

Lots of guesses,
Paul