Could a dead fan cause loss of BIOS settings in the CMOS?

--
@~@ Might, Courage, Vision, SINCERITY.
/ v \ Simplicity is Beauty! May the Force and Farce be with you!
/( _ )\ (x86_64 Ubuntu 9.10) Linux 2.6.36.2
^ ^ 16:12:01 up 4 days 22:17 2 users load average: 1.07 1.04 1.00
不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

Man-wai Chang wrote:
>
> Could a dead fan cause loss of BIOS settings in the CMOS?
>

I wouldn't think so. Not in a very direct way.

The CMOS is powered by the CMOS battery, wired-OR with a
regulator powered by +5VSB. To cause a "normal" failure, you'd
need to short that rail to ground. That would be an
electrically induced failure.

It is possible to erase the CMOS, under program control. So
a bad software design can do it. The question would be,
whether any software goes nuts, when the fan reads zero RPM.

On an Asus motherboard, the "overclocking failure" detection,
will reset the BIOS (but not change the "clock time"). So
if you're only losing a few settings, and it's an Asus
motherboard, it could be a computer crash that is causing it.
Can a dead fan cause a computer crash ? Yes, if the CPU
overheats (because it abruptly powers off). Many other fans
could die, without side effects. The CPU fan is a bit more
important.

Paul

> I wouldn't think so. Not in a very direct way.

Same here, unless there is a design flaw... Um...

> ... need to short that rail to ground. That would be an
> electrically induced failure.

If the fan failed to turn, would it send a back-emf to that rail and
clear the CMOS? The PC (HP Pavilion m9088hk) was reporting "System Fan
Failure" lately.

> On an Asus motherboard, the "overclocking failure" detection,
> will reset the BIOS (but not change the "clock time"). So
> if you're only losing a few settings, and it's an Asus
> motherboard, it could be a computer crash that is causing it.
> Can a dead fan cause a computer crash ? Yes, if the CPU
> overheats (because it abruptly powers off). Many other fans
> could die, without side effects. The CPU fan is a bit more
> important.

HP m9088hk is a generic Asus motherboard, but it doesn't have any
overclocking feature. The PC didn't hang, so the heat factor could be
ticked out.

--
@~@ Might, Courage, Vision, SINCERITY.
/ v \ Simplicity is Beauty! May the Force and Farce be with you!
/( _ )\ (x86_64 Ubuntu 9.10) Linux 2.6.36.2
^ ^ 18:06:01 up 5 days 11 min 1 user load average: 1.08 1.06 1.01
不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

On Thu, 06 Jan 2011 18:14:51 +0800, Man-wai Chang rearranged some
electrons to say:


>
> If the fan failed to turn, would it send a back-emf to that rail and
> clear the CMOS? The PC (HP Pavilion m9088hk) was reporting "System Fan
> Failure" lately.
>

No. Fix the fan, but it didn't cause your CMOS memory data to be lost.

> No. Fix the fan, but it didn't cause your CMOS memory data to be lost.

I began to think the fan was causing the CMOS data loss...

--
@~@ Might, Courage, Vision, SINCERITY.
/ v \ Simplicity is Beauty! May the Force and Farce be with you!
/( _ )\ (x86_64 Ubuntu 9.10) Linux 2.6.36.2
^ ^ 20:06:01 up 6 days 2:11 1 user load average: 1.02 1.03 1.00
不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

On 07/01/2011 20:16, Man-wai Chang wrote:
>> No. Fix the fan, but it didn't cause your CMOS memory data to be lost.
>
> I began to think the fan was causing the CMOS data loss...
>

If it's not electronics issue, then someone was playing security game
with that PC.

--
@~@ Might, Courage, Vision, SINCERITY.
/ v \ Simplicity is Beauty! May the Force and Farce be with you!
/( _ )\ (x86_64 Ubuntu 9.10) Linux 2.6.36.2
^ ^ 21:04:02 up 7 days 3:09 1 user load average: 1.01 1.02 1.00
不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

> If conditions are just right;
> 1. This dead shorted fan.
> 2. No recharge from battery circuit.
> 3. Some poor design or blown component.
> 4. Weak CMOS battery.
> Now this leaves the CMOS battery with a short across its leads,
> draining it.

Too deep for me as I am not a EE person. Let me investigate further.

--
@~@ Might, Courage, Vision, SINCERITY.
/ v \ Simplicity is Beauty! May the Force and Farce be with you!
/( _ )\ (x86_64 Ubuntu 9.10) Linux 2.6.36.2
^ ^ 18:50:01 up 8 days 55 min 1 user load average: 1.03 1.03 1.02
不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

On Sat, 08 Jan 2011 08:51:16 -0600, Onion rearranged some electrons to
say:

> Actually, it could.
>
> The user would have to power off the computer from a power strip for it
> to happen.
> As said; the fan is powered off the +5 rail as is the CMOS battery
> charger/battery.
> Bad/shorted fan is not drawing enough current to 'catch fire' but when
> computer is powered off with a strip which effectively is like pulling
> the plug, the battery circuit is now a dead short.
>
> If conditions are just right;
> 1. This dead shorted fan.
> 2. No recharge from battery circuit.
> 3. Some poor design or blown component. 4. Weak CMOS battery.
>
> Now this leaves the CMOS battery with a short across its leads, draining
> it.
>
> Have seen much stranger things happen. Onion

There is no "CMOS Batter Charger". The battery is a primary cell. The
CMOS memory is powered from 5V standby. The fans are powered from the
12V rail, though a fan speed controller. One has nothing to do with the
other.

On Mon, 10 Jan 2011 15:03:29 -0600, Onion rearranged some electrons to
say:

>> There is no "CMOS Batter Charger". The battery is a primary cell. The
>>
>> CMOS memory is powered from 5V standby. The fans are powered from the
>>
>> 12V rail, though a fan speed controller. One has nothing to do with
>> the
>> other.
>
> I was being simplistic. The 'charger' is a circuit powered from the
> main power supply.
> All DC voltage creation originates at the main transformer hence they
> are all connected.
>
> Drastically so in the case of a short.
>
> I can explain futher, but I hope you see the point.

I can see that you know nothing of power supply design or construction.
As I am an electrical engineer, I would love to hear your "further
explanation" of how a fan on the 12V rail could affect a battery-backed
circuit with a reverse-biased diode to prevent battery drain when 5VSB is
on.

Oh and while you're at it, explain how a transformer creates DC voltage.

On Mon, 10 Jan 2011 20:22:19 -0600, edfair
<(E-Mail Removed)> wrote:

>
>Suspect that the protect diodes in the CMOS power circuit would
>eliminate any back circuit drain of the battery.
>Has been a long time since I've dug into the circuitry but recall that
>there were 3 of 4 diodes blocking the supplies from affecting each
>other.

I would think there would have to be to avoid CMOS power being drawn
off into the rest of the board. A powered-down board would look
awfully similar to a short as far as the CMOS battery was concerned.