nathanieltolbert wrote on 2025-09-18, 15:37:

ChrisK wrote on 2025-09-18, 15:03:

Full datasheet

Thank you. Is there a specific place I should go to look for datasheets online? I did a search and even though I had two links that said they were data sheets, they took me to a site that wanted me to pay money to access datasheets at all and they didn't even have the one I was looking for.

FYI, I guess you missed my previous post?
anyway.
https://theretroweb.com/motherboards/s/asus-k7m
click on chips. click on the chipname, click on documentation.

Right there in the section under PCI to ISA bridge. - Flash EPROM 4mb EPROM and combined BIOS support. So is that possibly the NVRAM at startup that the BIOS says NVRAM not found? I don't see anything else that could possibly line up with NVRAM in the description of the chip.

weedeewee wrote on 2025-09-18, 16:08:

FYI, I guess you missed my previous post? anyway. https://theretroweb.com/motherboards/s/asus-k7m click on chips. click on the […]
Show full quote

nathanieltolbert wrote on 2025-09-18, 15:37:

ChrisK wrote on 2025-09-18, 15:03:

Full datasheet

Thank you. Is there a specific place I should go to look for datasheets online? I did a search and even though I had two links that said they were data sheets, they took me to a site that wanted me to pay money to access datasheets at all and they didn't even have the one I was looking for.

FYI, I guess you missed my previous post?
anyway.
https://theretroweb.com/motherboards/s/asus-k7m
click on chips. click on the chipname, click on documentation.

I did indeed. I did not know that The Retro Web worked that way. I learned something new today. Thank you!

nathanieltolbert wrote on 2025-09-18, 16:58:

Right there in the section under PCI to ISA bridge. - Flash EPROM 4mb EPROM and combined BIOS support. So is that possibly the NVRAM at startup that the BIOS says NVRAM not found? I don't see anything else that could possibly line up with NVRAM in the description of the chip.

Guess I'll have to repeat my first comment on your thread.

from
https://theretroweb.com/chip/documentation/68 … 6e776850263.pdf

VT82C686A "Super South" South Bridge

d) Real Time Clock with 256 byte extended CMOS. In addition to the standard ISA RTC functionality, the integrated RTC also
includes the date alarm, century field, and other enhancements for compatibility with the ACPI standard

weedeewee wrote on 2025-09-18, 17:13:

Guess I'll have to repeat my first comment on your thread. […]
Show full quote

nathanieltolbert wrote on 2025-09-18, 16:58:

Right there in the section under PCI to ISA bridge. - Flash EPROM 4mb EPROM and combined BIOS support. So is that possibly the NVRAM at startup that the BIOS says NVRAM not found? I don't see anything else that could possibly line up with NVRAM in the description of the chip.

Guess I'll have to repeat my first comment on your thread.

from
https://theretroweb.com/chip/documentation/68 … 6e776850263.pdf

VT82C686A "Super South" South Bridge

d) Real Time Clock with 256 byte extended CMOS. In addition to the standard ISA RTC functionality, the integrated RTC also
includes the date alarm, century field, and other enhancements for compatibility with the ACPI standard

Yes I didn't see that at the time I was trying to read posts because I was trying to read the datasheet while reading the comments. So the question becomes the 4Mb EPROM with combined BIOS support, is this separate? 4 Megabit should be 512 Kilobyte if I remember my conversion right 8 bits to 1 byte, so 256 byte extended CMOS with RTC sounds like a different or separate function. Unless specifically the 4Mb EPROM is broken down into multiple sections for various uses. I haven't read datasheets before so I am trying my best. Looking further down and trying to read the information I come across the CMOS/RTC Registers. I see labeled Port 70 through 75, with 71, 73, and 75 being CMOS Data (256 bytes). While port 72 and 74 are listed as CMOS Address (256 bytes) Everything here indicates settings for an internal RTC but nothing that I can easily read that indicates the 4MB EPROM indicated way earlier in the PCI to ISA features Section at the beginning of the Datasheet. I am working my way through the Function 0 Registers - PCI to ISA Bridge and I don't see the 4MB EPROM info. I must have missed it somewhere. When I try to search the PDF for EPROM it just shows the two in that listing at the start under PCI to ISA Bridge. The way the manual for the K7M reads regarding NVRAM is that it is stored on the same chip as the BIOS. And that is the SST39SF020, which could be compatible with 39SF020A chip? I can't help but wonder if the NOS chip I got on eBay wasn't. I just wasn't certain if it was safe to order the 020A from Mouser. I can only think that the EPROM has something to do with storing information specific to the PCI to ISA Bridge that works in conjunction with the BIOS. I really wish I had taken those Electronic Engineering courses when I went back to school.

nathanieltolbert wrote on 2025-09-19, 03:23:

weedeewee wrote on 2025-09-18, 17:13:

Guess I'll have to repeat my first comment on your thread. […]
Show full quote

nathanieltolbert wrote on 2025-09-18, 16:58:

Right there in the section under PCI to ISA bridge. - Flash EPROM 4mb EPROM and combined BIOS support. So is that possibly the NVRAM at startup that the BIOS says NVRAM not found? I don't see anything else that could possibly line up with NVRAM in the description of the chip.

Guess I'll have to repeat my first comment on your thread.

from
https://theretroweb.com/chip/documentation/68 … 6e776850263.pdf

VT82C686A "Super South" South Bridge

d) Real Time Clock with 256 byte extended CMOS. In addition to the standard ISA RTC functionality, the integrated RTC also
includes the date alarm, century field, and other enhancements for compatibility with the ACPI standard

Yes I didn't see that at the time I was trying to read posts because I was trying to read the datasheet while reading the comments. So the question becomes the 4Mb EPROM with combined BIOS support, is this separate? 4 Megabit should be 512 Kilobyte if I remember my conversion right 8 bits to 1 byte, so 256 byte extended CMOS with RTC sounds like a different or separate function. Unless specifically the 4Mb EPROM is broken down into multiple sections for various uses. I haven't read datasheets before so I am trying my best. Looking further down and trying to read the information I come across the CMOS/RTC Registers. I see labeled Port 70 through 75, with 71, 73, and 75 being CMOS Data (256 bytes). While port 72 and 74 are listed as CMOS Address (256 bytes) Everything here indicates settings for an internal RTC but nothing that I can easily read that indicates the 4MB EPROM indicated way earlier in the PCI to ISA features Section at the beginning of the Datasheet. I am working my way through the Function 0 Registers - PCI to ISA Bridge and I don't see the 4MB EPROM info. I must have missed it somewhere. When I try to search the PDF for EPROM it just shows the two in that listing at the start under PCI to ISA Bridge. The way the manual for the K7M reads regarding NVRAM is that it is stored on the same chip as the BIOS. And that is the SST39SF020, which could be compatible with 39SF020A chip? I can't help but wonder if the NOS chip I got on eBay wasn't. I just wasn't certain if it was safe to order the 020A from Mouser. I can only think that the EPROM has something to do with storing information specific to the PCI to ISA Bridge that works in conjunction with the BIOS. I really wish I had taken those Electronic Engineering courses when I went back to school.

Yes, that is a seperate chip. The SST39SF020 512K(4M) contains the bios, not the nvram. The nvram is located in the VIA southbridge chip.

nathanieltolbert wrote on 2025-09-19, 03:23:

I really wish I had taken those Electronic Engineering courses when I went back to school.

I'm not sure if this topic here is something that is taught in Electronic Engineering courses 😉
Anyway, it would be a huge coincidence to me if the newly bought EEPROM would have the exact same failure than the one you had.
IIRC you could program the "new" EEPROM and the board works as before, right? That tells me the problem must be elsewhere.
Still the question stands if the NVRAM within the southbridge really holds the CMOS data, i. e. the settings you make in the BIOS. I think this can only be answered by an ASUS BIOS engineer or someone you could have a look into the BIOS routines and understand what's done there. Things change over time and ways to do things too. Like the RTC, that was for long time an external component, getting part of the chipset the CMOS memory could also have moved into the chipset for whatever reason.
But what does that tell us? If the no-NVRAM failure message relates to the NVRAM within the southbridge then there must be a problem anywhere around there. Since all NVRAM accesses are allegedly totally internal to the southbridge there can't be any external components or buses involved which rules out any passive parts defect imho. Looking at the board there also aren't many components around the BIOS chip itself that could be damaged and if so the symptoms might be worse.
I don't know... I don't like pointing towards something without beeing at least half way sure. But what let's me wonder are the other things you described (slow system, drive problems, no boot with an ISA card installed). These all could have to do with the southbridge or anything around it... I guess you already inspected all components in this area as well as looked for board/trace damages of any kind? The battery is inserted and working correctly? Did you try any other ISA card and have you inspected the ISA slot thoroughly (contacts and solder joints)?
Very tricky case...

ChrisK wrote on 2025-09-19, 06:50:

I'm not sure if this topic here is something that is taught in Electronic Engineering courses ;) Anyway, it would be a huge coin […]
Show full quote

I'm not sure if this topic here is something that is taught in Electronic Engineering courses 😉
Anyway, it would be a huge coincidence to me if the newly bought EEPROM would have the exact same failure than the one you had.
IIRC you could program the "new" EEPROM and the board works as before, right? That tells me the problem must be elsewhere.
Still the question stands if the NVRAM within the southbridge really holds the CMOS data, i. e. the settings you make in the BIOS. I think this can only be answered by an ASUS BIOS engineer or someone you could have a look into the BIOS routines and understand what's done there. Things change over time and ways to do things too. Like the RTC, that was for long time an external component, getting part of the chipset the CMOS memory could also have moved into the chipset for whatever reason.
But what does that tell us? If the no-NVRAM failure message relates to the NVRAM within the southbridge then there must be a problem anywhere around there. Since all NVRAM accesses are allegedly totally internal to the southbridge there can't be any external components or buses involved which rules out any passive parts defect imho. Looking at the board there also aren't many components around the BIOS chip itself that could be damaged and if so the symptoms might be worse.
I don't know... I don't like pointing towards something without beeing at least half way sure. But what let's me wonder are the other things you described (slow system, drive problems, no boot with an ISA card installed). These all could have to do with the southbridge or anything around it... I guess you already inspected all components in this area as well as looked for board/trace damages of any kind? The battery is inserted and working correctly? Did you try any other ISA card and have you inspected the ISA slot thoroughly (contacts and solder joints)?
Very tricky case...

I did a visual inspection of the board and I don't see any trace damage or spots that look like the traces are corroded under the mask. But I don't have the tools to examine the traces super closely. The VIA 686 chip looks okay? - but again I don't have magnification or the knowledge to check what appears to be a BGA chip. I wonder if I can find a replacement 686A and someone to replace the chip for me. I can't help but think after all of this work that there is something wrong with that chip. But I don't know enough to be able to full ascertain. I mean, I could send the board to someone more knowledgeable and they could test it, but I am already taking up everyone's time as it is here.

Have you tried pressing down the SB?

Nexxen wrote on 2025-09-23, 17:42:

Have you tried pressing down the SB?

I have not. It's currently stored away in the mobo box while I try to figure out my next step. Right now I'm fiddling with a BIOSTAR M7MKE and finding that my Thunderbird 700 works when all of the documentation says it shouldn't?

So after Wednesday when I was handed some crap medical news I couldn't sleep. I pulled this board out of the box and tried pressing on the South Bridge while booting. I was unable to alter the booting process, but I will admit that I didn't press very hard. I will try again here this evening after I get done with all of the medical calls and provide an update to if there is any difference. Thank you again everyone for all of the help you have been providing while I try to figure out what's going on with this board. It really does mean a lot to me.

nathanieltolbert wrote on 2025-09-26, 21:03:

medical news

Be strong and get well soon.
Maybe faster than the motherboard 😉

I got the board back out and I pushed hard on the southbridge chip while it was boot and while rebooting. There is sadly no change in the functionality of what the board does. If the southbridge chip is defective, this board is effectively dead in the water as I cannot replace a BGA chip nor do I even know if I can get a replacement chip at all. Thank you again everyone for all of the help. I do really appreciate it. I hate to leave the board in this state but I have stretched my ability far beyond what I thought I could do just getting to this point. I would like to think I could do it if I tried, but I don't want to permanently destroy the board. It's supposed to be one of the best of the Irongate boards. I want it to be that again, even if I have to have someone else do the final stages of fixing it.