So, here is how it should look, if You start the board without CPU - full red square. Caps on KT4AV are OST brand, on KT3V Chemicon. As you can see, these boards are oddly unpredictable - now, after a long storage period without a battery in the box, the D-bracket is on even with the KT4AV, where it wasn't lit before...

Ydee wrote on 2021-08-02, 14:21:

So, here is how it should look, if You start the board without CPU - full red square. Caps on KT4AV are OST brand, on KT3V Chemicon. As you can see, these boards are oddly unpredictable - now, after a long storage period without a battery in the box, the D-bracket is on even with the KT4AV, where it wasn't lit before...

Yes, I tried to turn it on in the same way you did and showed in your pictures. But the D-Bracket remaind 1 green, 2, 3 and 4 red...just like when it was tested in the other conditions.
My board is not exctly the same as the one you showed, this ones is nForce2 based and with this "core cell" thing, but I understand your intention was to show how the D-Bracket should behave.

I guess replacing the KZG capacitors is the only thing left to try.

¡Thank you!

I have one. the little red deomon board.
mine has blown mosfets on it when i got it but the chipset still works is why i haven't bin it.
I think i need an engineers part number reference for later model mosfets compatible.
BIOS only corrupts when overclocking de stablises the power rails happens or spikes in the mains. Its very uncommon for a wrong BIOS part number to be flashed. With so many motherboards these events can occur but mainly units not marketed for gaming and or overclocking and always by the newbies whom think pedal to the metal instead of small steps.
Gaming motherboard systems not only support reasonable overclocking but usually the system has better power protection with purpose built gaming Power Supplies as well.
That being said, I find nobody speaks of PSU and i recommend they be tested every 5 years by a technician not just for caps or mosfets but the included surge and spike protection components. Surprisingly, the most complicated parts of motherboards and PSU's, the IC chips rarely die for decades as they only use small amount of power to regulate the mosfets higher outputs (switching ). A capacitor that is used is a capacitor that eventually dries out.

A motherboard on eBay that says works but doesn't post is a motherboard that should be listed as for parts / not working I think. As a used item working is supposed to work for its intended purpose an appliance that can run without malfunction.

Con 2 botones wrote on 2021-08-03, 12:38:

But the D-Bracket remaind 1 green, 2, 3 and 4 red...just like when it was tested in the other conditions.

And that's wrong, because it means that the test routine doesn't happen - CPU detection is the first step of POST. Your board froze in some mode, where the code for initializing the chipset glows randomly, but it couldn't actually happen because the CPU was not detected. I'm afraid replacing the capacitors won't be enough to fix it, but I'd like to be wrong.
I know You have a board with nForce chipset, but the codes for D-bracket are the same for all boards equipped with it.

I've had lots of problems with my K7N2 MS-6570. Oh well, it runs sometimes OK, then suddenly it just won't post or beeps for memory fault. Can't remember LEDs state in bracket.
I've learned that sometimes setting FSB mode jumper to 100MHz (safe mode) helps but not always. Your 6570E seems to be with different layout so I don't know if it has similar "safe mode" jumper.

I have checked one of both KZG and rubycon caps with Fluke LCR meter and those seems to be fine. I fear there might be some dry KZGs nevetheless one was OK. I took the board once again in my hand after reading this topic. I noticed some strong warping on the memory slot area as you can see in the picture. I'll try this board one last time and try to straighten the board with something. If this doesn't help I'll check LED bracket if it reads something similar. These boards are such a waste (of time).

I used to have one of these - at one time it was my boss' home gaming computer, and he gave it to me when I upgraded his system to to Core 2/P5N-E system in 2007. It worked fine for years after that, mostly it sat in an anti static bag. I used it to build an office computer in probably 2012, then replaced it with something less obsolete and it went back into storage. In probably 2013 I tried it out, and it began having issues POSTing. I thought that maybe I had flashed the incorrect BIOS (there are many different revisions of this board) and gave up on it. The last time I fooled with it, I made absolutely sure I had the correct BIOS and I did a hot swap flash; initially all seemed well, but it soon became clear that it still had an issue with only POSTing once out of every so many tries. I tried different memory and CPUs, then finally gave up on it and put it back into storage. It had no visible signs of trouble with the capacitors.

Yeah...I quit.
Any recommendation on a nforce2 based reliable board (with 12V connector and Sata ports)?

Repo Man11 wrote on 2021-08-03, 19:49:

I used to have one of these - at one time it was my boss' home gaming computer, and he gave it to me when I upgraded his system to to Core 2/P5N-E system in 2007. It worked fine for years after that, mostly it sat in an anti static bag. I used it to build an office computer in probably 2012, then replaced it with something less obsolete and it went back into storage. In probably 2013 I tried it out, and it began having issues POSTing. I thought that maybe I had flashed the incorrect BIOS (there are many different revisions of this board) and gave up on it. The last time I fooled with it, I made absolutely sure I had the correct BIOS and I did a hot swap flash; initially all seemed well, but it soon became clear that it still had an issue with only POSTing once out of every so many tries. I tried different memory and CPUs, then finally gave up on it and put it back into storage. It had no visible signs of trouble with the capacitors.

This generation of NVIDIA chipsets (started from NF4 nForce flip-chip packaging) is notoriously known.
Applying very hot air for a 10-30sec to a bad die resurrects the chip but for unpredictable period of time.

Con 2 botones wrote on 2021-08-04, 15:37:

Yeah...I quit.
Any recommendation on a nforce2 based reliable board (with 12V connector and Sata ports)?

Maybe GIGABYTE GA-7N400S . But what is the purpose to have NF2 based system at all? Is it just to run Windows XP only?
There're plenty platforms are capable of that.

Roman555 wrote on 2021-08-05, 16:14:

Con 2 botones wrote on 2021-08-04, 15:37:

Yeah...I quit.
Any recommendation on a nforce2 based reliable board (with 12V connector and Sata ports)?

Maybe GIGABYTE GA-7N400S . But what is the purpose to have NF2 based system at all? Is it just to run Windows XP only?
There're plenty platforms are capable of that.

True, I already own a completely reliable (bought and owned since back in the day) VIA KM400A chipset based MSI board, which is very capable.
It´s just "lavishness", I just wanted the best from the Athlon XP era (to go with my 3200+) and nForce2 is the best. That which you couldn´t afford in its day, you know.
For w98 I do have other systems built.
Thanks for the recommendation!