Flashing the BIOS
 

Hi,

I'd like to thank the activists here for all the help. Last year, I
tried to upgrade the my BIOS. Unbeknownst to me at the time, the 3V CMOS
battery was dead. I ran into all sorts of problems, like BIOS checksum
errors. I was happy to downgrade, so I could continue using my 3 drive
RAID 0 storage array, that needs to be setup in the CMOS BIOS program.
This year with a good battery, flashing an upgrade to the BIOS went
smoothly, no errors. In a few days when my heart settles down, I will
replace the dual core with a (presumably good) used quad core CPU.

Who would have imagined that a tiny mercury battery would be important?

X

Flashing the BIOS
 

Norm X wrote:
Hi,

I'd like to thank the activists here for all the help. Last year, I
tried to upgrade the my BIOS. Unbeknownst to me at the time, the 3V CMOS
battery was dead. I ran into all sorts of problems, like BIOS checksum
errors. I was happy to downgrade, so I could continue using my 3 drive
RAID 0 storage array, that needs to be setup in the CMOS BIOS program.
This year with a good battery, flashing an upgrade to the BIOS went
smoothly, no errors. In a few days when my heart settles down, I will
replace the dual core with a (presumably good) used quad core CPU.

Who would have imagined that a tiny mercury battery would be important?

X

A dead battery stops some, but not all, boards from booting.

And the reason remains unexplained. Nothing in the circuit
design says that it has to stop dead like that. 3VSB is
diode ORed power from two sources, and if the battery
is flat, the +5VSB provides the power. You lose your
clock setting, but the computer should still work.

Some SuperIO chips have a VBAT signal as an input.
In theory, that only goes to the voltage measurement
mux, and gets measured like 3.3V, 5V, and 12V get measured
by the ADC (analog digital converter). Now, if the
SuperI/O also derived a logic signal from that input,
it could stop things dead. The question would be,
why would they do that ? And, is that documented ?
I've never found any proof of that possibility.
A motherboard should be perfectly capable of
working without the tiny battery as a
source of power.

You shouldn't get BIOS checksum errors from
a dead battery. You can get BIOS checksum errors
(happened to me here), from overclocking. The
33MHz clock on some clocked BIOS chips, is derived
from your overclocked system clock. Once that
runs around 40MHz or so, you might see the checksum
step fail. Don't panic when that happens, and just
reset the BIOS settings. Some motherboards will
recommend "Why don't you reflash the BIOS", and
if you attempt to reflash the BIOS under those
condition, the motherboard ends up bricked! So
don't do that. Don't panic. Reset the settings
with the CMOS jumper (if the system lacks any
other method). Just don't pay attention to
any recommendations to reflash an unstable
computer! That's exactly the wrong time to
flash upgrade, when it's sick.

Paul

Flashing the BIOS
 

On 11/11/2015 10:30 AM, Norm X wrote:
Hi,

I'd like to thank the activists here for all the help. Last year, I tried
to upgrade the my BIOS. Unbeknownst to me at the time, the 3V CMOS battery
was dead. I ran into all sorts of problems, like BIOS checksum errors. I
was happy to downgrade, so I could continue using my 3 drive RAID 0 storage
array, that needs to be setup in the CMOS BIOS program. This year with a
good battery, flashing an upgrade to the BIOS went smoothly, no errors. In
a few days when my heart settles down, I will replace the dual core with a
(presumably good) used quad core CPU.

Who would have imagined that a tiny mercury battery would be important?

X

I hope that it wasn't really a mercury battery -- they've been banned for a
long time and I don't recall any PCs that actually used them. More likely a
silver-oxide cell although I had at least one machine that actually used a
battery pack with four AA cells.