ok to disengage fan control on W520 temporarily?

[ianb527]

There are a few things that I use my W520 for that make the CPU kindof warm - eg, kernel builds, bitbake, etc...
The fan control seems to stop well short of the fan speed when the control is disengaged and I have noticed that by disengaging the fan control temporarily while doing these things that cause all cores to go to 100%, it lowers the temperatures significantly (for the most part, they stay below 80C). I couldn't care less about the noise.
My question is this: what is the risk of harming the fan motor if the speed control is disengaged temporarily (usually less than 1/2 hour)?
Also, what is the best way to control the fan and what usable speed range should I be able to get under control (right now, disengaged is usually a little less than 5000 rpm, highest under control is around 3800, which puzzles me a bit).

[brchan]

There is TPFanControl for Windows and a program for linux that you can use to control the fan speeds:
http://www.thinkwiki.org/wiki/How_to_control_fan_speed

[jaspen-meyer]

My question is this: what is the risk of harming the fan motor if the speed control is disengaged temporarily (usually less than 1/2 hour)?

It is a brushless motor.

Here are some fan 'live expectancy' numbers:
http://www.sepa-europe.com/en/technisch ... tancy-mtbf

[tarvoke]

hell with tpfancontrol, what you want for any sandy/ivy -bridge is P-state + thermald.
conferre:
http://forum.thinkpads.com/viewtopic.php?t=117552

in theory, mint / ubuntu / debian are now supposed to prefer pstate over acpi, but in my experience it doesn't happen - so, explicitly pass 'intel_pstate=enable' to grub.

the default / automagic config for thermald has been fine by me, but if you don't want the fan to come on, you may need to mess with the settings.