Close to 4-pin header, check the sop-8.
Maybe it's nothing.

Nexxen wrote on 2024-06-02, 00:41:

Close to 4-pin header, check the sop-8.
Maybe it's nothing.

Do you mean the 4 pin header near the molex connector?

I removed the heatsink over the VRM mosfets.

The high side is showing 12v and the low side is showing 0v (it should be 1.5v). Hmmm

zuldan wrote on 2024-06-02, 01:51:

I removed the heatsink over the VRM mosfets.

The high side is showing 12v and the low side is showing 0v (it should be 1.5v). Hmmm

VRM.jpg

If those (brown) caps are UCC KZG series then possibly one or more 'silent' failures on their part!

zuldan wrote on 2024-06-02, 00:34:

- Motherboard powers on as soon as the board has power from the ATX

This is not good, it means dead chipset most of the time, there's a big inductor close to the chipset, measure its resistance to the ground.
also meassure the resistance of this mosfet to the ground where I have pointed the arrow :

The vcore enable signal comes directly from the chipset and goes through this diode to the pwm IC,
you should have atleast 3v on both side of the diode:

it looks like someone has changed the chipset heatsink, the original one has a fan in it, I wouldn't be surprised if the chipset was cooked to dead.

zuldan wrote on 2024-06-02, 01:04:

Nexxen wrote on 2024-06-02, 00:41:

Close to 4-pin header, check the sop-8.
Maybe it's nothing.

Do you mean the 4 pin header near the molex connector?

Yes. There is a little sop-8 chip. Look what it is and maybe it is connected to mosfets.
Just a guess. Maybe it's nothing at all.

Tha fact it powers on immediately means to me that the circuit responsible is broken.
There should be a driver chip for the mosfets.

Karbist wrote on 2024-06-02, 09:10:

The vcore enable signal comes directly from the chipset and goes through this diode to the pwm IC,
you should have atleast 3v on both side of the diode:
Vcore.EN.JPG

Can you give me the name of the big chip on the left of the arrow? That should be the driver.

boardview for deluxe 1.02, I think its same thing with just couple more components fitted
cold cpu = no Vcore as you later found out

The SC2643VX provides a solution for microprocessor core voltage regulation
https://pdf1.alldatasheet.com/datasheet-pdf/v … 2643VXTSTR.html

8 pin ic:
https://pdf1.alldatasheet.com/datasheet-pdf/v … ECH/SC1211.html

To me it's one of these two.
I can be wrong, unlike rasz_pl I'm not a pro.

The two sot-23 are BAT54C, pretty common.

PC Hoarder Patrol wrote on 2024-06-02, 02:09:

If those (brown) caps are UCC KZG series then possibly one or more 'silent' failures on their part!

They are KZG caps but the ESR meter seems to be happy with them. I'll keep this in mind.

Karbist wrote on 2024-06-02, 09:10:

meassure the resistance of this mosfet to the ground where I have pointed the arrow :

It measures 80 ohms

Karbist wrote on 2024-06-02, 09:10:

The vcore enable signal comes directly from the chipset and goes through this diode to the pwm IC,
you should have atleast 3v on both side of the diode:
Vcore.EN.JPG

Here are the measurements. Did you get this location from the boardview file?

Mistake

Nexxen wrote on 2024-06-02, 10:44:

Can you give me the name of the big chip on the left of the arrow? That should be the driver.

Under the microscope.

rasz_pl wrote on 2024-06-02, 12:41:

boardview for deluxe 1.02, I think its same thing with just couple more components fitted
cold cpu = no Vcore as you later found out

Thanks for that!

Any chance you have a boardview for a A8N32-SLI Deluxe?

Ok so the low side mosfet is connected to pin 7 of the chip. AGND, pin 18 is 35 ohms.

zuldan wrote on 2024-06-03, 06:36:

Here are the measurements. Did you get this location from the boardview file?

diode.jpg

You forgot to measure the resistance of inductor close to the chipset to the ground.
you're missing the vcore enable signal, so either the nforce chipset is dead or has bad connections under it.

Karbist wrote on 2024-06-03, 07:32:

You forgot to measure the resistance of inductor close to the chipset to the ground.
you're missing the vcore enable signal, so either the nforce chipset is dead or has bad connections under it.

Sorry missed that one. It's 80 ohms.

I also removed the 3rd party heatsink and the chip looks toast. The surface surrounding the die has bubbles on it (can't see from the picture). Is there a way to tell for sure it's gone? I'm here to learn. I love doing this stuff 😀

It looks like someone has used extreme heat to reflow the chip, although the chip internal resistance still looks good after that heat abuse, mine has dropped to 50 ohm.
at this point you need either to properly reball the chip or inject lots of flux under the chip and reflow it again and hope for the best.

Karbist wrote on 2024-06-03, 09:36:

It looks like someone has used extreme heat to reflow the chip, although the chip internal resistance still looks good after that heat abuse, mine has dropped to 50 ohm.
at this point you need either to properly reball the chip or inject lots of flux under the chip and reflow it again and hope for the best.

The NB popcorned?