Page 1 of 1

CMOS Checksum Error

Posted: Mon Apr 22, 2002 10:21 pm
by Griffinok
I've got an ABIT KA7 with an Athlon 850 Slot A. It was my old gaming machine converted into a server. Its been running for 2 years fine. A week ago it started to do some wierd things.

The first thing I noticed was that it would randomly reboot my machine. It would be running Win 2000 server and then reboot itself.

A couple of days ago, the CMOS Checksum error appeared, and would stop after reporting the memory and HD info. It would pause like that for 1-10 secs, then reboot. After the reboot, I get nothing at all. I've unhooked all components except for Mobo, CPU, memory, video, and floppy. I've tried switching memory, replacing the CMOS battery. Everything results in the same outcome.

I'm now down to trying to figure out whats left that could be causing the problem? CPU dying? Mobo dying? or CMOS/BIOS is junk now? The components are so old now, I'm going to have a hell of a time finding a replacement part. But thats another problem I'll deal with later. How can I pinpoint now what is going on? I don't have a similar board or slot A CPU lying around that I can test to further figure out the problem.

Anyone have any ideas?

Thanks
-Griff

Posted: Tue Apr 23, 2002 3:04 am
by ajzchips
You didn't mention checking the CPU fan and the PSU.

CMOS Checksum Error; ABIT SH6 motherboard.

Posted: Mon Jan 20, 2003 2:26 pm
by dave clark
Hi...
I have a similar problem with an ABIT SH6 motherboard.
The system will only start after "clear CMOS", and only then on third or fourth attempt.
At that point it reports "CMOS Checksum error" and loads 'defaults'.
I am then able to get into BIOS setup, and configure for correct CPU speed, however on "save and exit" the system fails to restart and hangs before video/bleep are enabled. CPU fan is OK, Hard Disk spins up.

I have had the BIOS Chip reprogrammed with the latest (-7u) code for this motherboard, but still get the same result.

I am now suspecting that the CMOS chip is damaged.

For what it's worth, when I received the motherboard, it had the "Clear CMOS" jumper set to "clear..." instead of "normal". If the board had been powered up in this state would this damage the CMOS chip?

By much research I am almost certain that the CMOS chip is chip U24 (ATMEL ATF16V8C, EE-PLD) adjacent to the BIOS EEPROM (U19).

Apart from the obvious hazards and difficulties of changing chips of this nature, has anybody got any comments on my strategy so far?

Dave Clark

Posted: Mon Jan 20, 2003 3:28 pm
by NickS
That Atmel device is a Programmable Logic Device, not a CMOS/Clock chip.

Posted: Mon Jan 20, 2003 4:59 pm
by dave clark
The ATMEL ATF 16V8C is an Electrically Eraseable Programmable Logic Device (EE-PLD) which is the 'CMOS' referred to in the "save to CMOS" routines of the BIOS setup.

The BIOS Chip is an EEPROM (Electrically Programmable Read Only Memory), which contains the BIOS defaults.

It is the default BIOS settings which are modified during BIOS setup, and saved to the CMOS EE-PLD chip, and retained so long as the battery is OK.

The system then uses the settings from the CMOS EE-PLD chip to start and configure the system each time it boots.


The following was extracted from 'Tomshardwarepage'

The FWH - 82802

Behind the name 'FWH' = 'Firmware Hub' you'll find a chip that's not much else than a 4 Mbit EEPROM plus a tiny bit of active silicon. The EEPROM contains the motherboard BIOS and the active silicon is a random number generator. This random number generator, otherwise known as the 'RNG', can supposedly be used by software to provide extra security during computer related transactions.

Posted: Mon Jan 20, 2003 8:03 pm
by NickS
dave clark wrote:The ATMEL ATF 16V8C is an Electrically Eraseable Programmable Logic Device (EE-PLD) which is the 'CMOS' referred to in the "save to CMOS" routines of the BIOS setup.
'Fraid not. In the Intel 815E the 256 bytes of CMOS RAM and the RTC are in the ICH2. Check out http://www.intel.com/design/chipsets/de ... 823401.pdf

Posted: Mon Jan 20, 2003 8:57 pm
by Rainbow
CMOS settings are NOT stored in EEPROM memory. EEPROM memory does not need power to keep data - so why the battery is there? CMOS settings are saved in CMOS. It's part of chipset since later Socket 7 boards, it was separate chip before (CMOS+RTC [Real Time Clock] in one chip).
Have you replaced CMOS backup battery?