Check where the posts for the battery holder come through the PCB and measure the voltage -there have been cases where the contacts were slightly corroded but it was almost impossible to see. It doesn't take much resistance when you're dealing with such low voltage. Try setting the board up outside of a case to make sure that nothing can go to ground that isn't supposed to.

The board was not in the case.
I already measured the voltage that goes through battery holder posts to the motherboard on the back of the board and its ok.

I guess either the CMOS memory chip is bad, or it isn't getting power for some reason. That's made more difficult because it isn't readily apparent where the CMOS chip on that board is. With boards such as the PCChips M507 and M520 the diagram makes it crystal clear which chip is the RTC, and it's socketed; so you know where it is, it can easily be checked for voltage, and it can be easily replaced. The only suggestion I have is that you figure out which chip is the RTC, then check the traces between the CMOS battery and that chip.

Repo Man11 wrote on 2022-10-05, 01:25:

I guess either the CMOS memory chip is bad, or it isn't getting power for some reason. That's made more difficult because it isn't readily apparent where the CMOS chip on that board is. With boards such as the PCChips M507 and M520 the diagram makes it crystal clear which chip is the RTC, and it's socketed; so you know where it is, it can easily be checked for voltage, and it can be easily replaced. The only suggestion I have is that you figure out which chip is the RTC, then check the traces between the CMOS battery and that chip.

At a guess its the one on the far left with the white sticker on it, if thats the BIOS then tracing issues with voltages might be a nightmare.

Are you all talking about the same board ? The Intel Zappa (FX82437 and 82371) based boards are different than the PcChips boards....
The CMOS is usually in the south bridge on most Pentium boards unless they have Dallas type time/cmos chip then it is used instead of the one in SB.
AFAIK the Intel Advanced does not have a external cmos chip and does not have a Dallas or other type. I have one (Advanced ATX) and cannot find it. Maybe someone else knows more....
Also: The BIOS chip on any PC motherboard does not have a cmos ram section, it is always in some other IC.

Horun wrote on 2022-10-05, 02:57:

Are you all talking about the same board ? The Intel Zappa (FX82437 and 82371) based boards are different than the PcChips board […]
Show full quote

Are you all talking about the same board ? The Intel Zappa (FX82437 and 82371) based boards are different than the PcChips boards....
The CMOS is usually in the south bridge on most Pentium boards unless they have Dallas type time/cmos chip then it is used instead of the one in SB.
AFAIK the Intel Advanced does not have a external cmos chip and does not have a Dallas or other type. I have one (Advanced ATX) and cannot find it. Maybe someone else knows more....
Also: The BIOS chip on any PC motherboard does not have a cmos ram section, it is always in some other IC.

I knew someone would come along eventually and explain why there is no obvious CMOS chip on the board OP posted, never ran across this myself what an odd setup.

I guess OP should be looking for an Sram chip on the board then.

Horun wrote on 2022-10-05, 02:57:

Are you all talking about the same board ? The Intel Zappa (FX82437 and 82371) based boards are different than the PcChips board […]
Show full quote

Are you all talking about the same board ? The Intel Zappa (FX82437 and 82371) based boards are different than the PcChips boards....
The CMOS is usually in the south bridge on most Pentium boards unless they have Dallas type time/cmos chip then it is used instead of the one in SB.
AFAIK the Intel Advanced does not have a external cmos chip and does not have a Dallas or other type. I have one (Advanced ATX) and cannot find it. Maybe someone else knows more....
Also: The BIOS chip on any PC motherboard does not have a cmos ram section, it is always in some other IC.

According to the 430FX spec sheet I looked up earlier the southbridge doesn't have CMOS memory. Chart one says it uses the PIIX southbridge, chart 2 says no CMOS memory.
https://flylib.com/books/en/4.57.1.45/1/

I found a manual for the board. This excerpt makes me think maybe the CMOS chip is under, or integrated with, the coin battery holder. Especially since they break down what every component is except for the CMOS chip.

Disregard the above, the answer was a little further into the manual. Time to break out a magnifier and look at all of I/O chip's solder connections.

Here's the 83706 datasheet to help you narrow down the search.

Looks like 7 is the magic number.

Thank you, it' s really helpful. I will check it when I get home.
I assume, aside from checking connection itself, there should be voltage on pad.

Good work ! Yes I had a brain fart and called the peripheral controller the southbridge.

87306 is firmy soldered to the board, however there is no voltage on pin 7. Is it time to use bodge wire and bridge it to battery?

I've never done a trace repair, but there are a few threads here that cover the topic. Here's one I found: Minor trace repair on a motherboard

It is no like that. In this motherboard traces from the battery goes inside, between layers. I think direct bodge is the only way. My question is - is it a good idea? Maybe it shouldn't be 3v on pin 7, but less?

According to the 83706 datasheet the operating VBAT voltage range is between 2.4V and Vcc - 0.5, so 3V should be totally fine on that pin.

However, the "Typical Battery Configuration" in page 88 suggests the use of a diode between the positive terminal of the battery and the VBAT pin. If that's how it was implemented in the motherboard then the actual VBAT voltage should be a bit less due to the diode drop, but more importantly battery + would not be connected directly to the chip. So I would look for that diode, if it's blown or not properly connected it could explain the absence of voltage at the VBAT pin.

Keep in mind that the actual implementation of the battery backup circuit may be different than the suggestion in the datasheet, so you could also have a resistor, another diode, and so on.

OK, I managed to track down a diode, which works fine (shows voltage drop 0.53V, shows OL in reverse). Trace goes: battery -> resistor -> diode -> inside motherboad
That probably means that broken trace is between diode and 83706.
I wonder shall I try bridging a diode with pin7 of 83706.
Edit : as the trace to pin7 goes under the chip, I'll try to reflow that area with hot air first.

Update: I managed to do a bodge wire mod as described above, but nothing changed.
I think it may be faulty 83706 controller and further repair attempts won't make any sense.