[ltp] Re: [T410s] Fan running constantly

[Daniel Näslund]

On Wed, May 18, 2011 at 2:37 PM, Stefan Monnier
<monnier@iro.umontreal.ca> wrote:
>> When googling, I find several forum posts describing the T410s fan as
>> noisy when running Linux but it appears as  a non-issue for Windows
>> users.  Has anyone found a workaround? I've found [2] but I'm resistan=
t
>
> IIUC the problem is not the fan, but the power management.
> I.e. the fan will turn itself off if the temperature goes down
> sufficiently, but that doesn't happen under GNU/Linux because the power
> management is not as good as under Windows.
>
> I don't have a T410 so maybe I'm talking about something unrelated, but
> at least my T60's power consumption when idle has never gone nearly as
> low under GNU/Linux (lowest ever for me: 12.3W) as under Windows
> (lowest below 10W; only time I booted Windows on this machine).

On a sidenote, I too have experienced higher power consumption when
switching from Windows to Linux as detected by shorter battery life.
But many laptops running Linux do not have a fan running at high
speed!

If I'm to speculate freely about the possible causes of the
fan-always-on problem I'd say that it must be one of these:

1) The BIOS firmware causes the fan to be too aggressive.
2) There's a bug in the thinkpad-acpi kernel module causing the
erroneus behaviour.
3) The acpi trippoints in the kernel are set too low causing the
4) The fan itself can not correctly adjust it's speed - a hardware problem
5) The T410s is built in such a way that it gets too hot - the fan is
too small or there's too little space.

If it's (1) or (3) then there's a reasonable chance of getting the
problem fixed. If it's (2) it's a little more involved but doable if
someone could collaborate with the linux-acpi team. (4) or (5) means
that the only viable solution is buying myself a new laptop in order
to get rid of the fan problem.

An obstacle for me getting further with this is my limited
understanding of how the BIOS and acpi systems interact. AFAICT the
fan is either controlled by the acpi system or it's not and then it's
controlled directly by firmware in the BIOS. If that's a correct
assumption, then the first thing to do would be to determine if the
acpi system in fact is controlling the fan. As I wrote earlier in this
thread, I tested to supply the boot argument "acpi=off" but the fan
still ran on high speed. I'm not sure if that's a proof of the acpi
system not controlling the fan... Just realized that I can probably
alter some behaviour of the fan through the /proc filesystem. If I can
change the speed of the fan through /proc then the fan is indeed
controlled by the acpi system. Will test that tomorrow. Sorry about
thinking out aloud.

Daniel