[ltp] T61 temperature limits

[Owen Heisler]

--J2SCkAp4GZ/dPZZf
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

I've implemented _passive_ cooling (as little fan usage as possible) using =
a=20
script and would like to discuss the limits that I am using for the tempera=
ture=20
sensors.  I am using three limits which I call "warm", "hot", and "veryhot"=
=2E =20
First, the processor frequency is limited, second the processors are thrott=
led=20
too, and last the system is suspended (at least, that's how I have it now).=
 =20
Any temperature less than warm is considered 'cool enough' for full=20
performance.

Note that the most important limit is "veryhot" and second "hot"; I would l=
ike=20
to know how hot things can safely be for an extended time and then at what=
=20
temperature to give up and just suspend quickly.

Also, passive cooling seems to be working well enough most of the time (a c=
ool=20
environment helps tremendously); however, one temperature which will almost=
=20
never drop while the system is running is the battery's (only one sensor of=
=20
which seems to work [2]).  So can that temperature go any higher than what =
I'm=20
using now?

I am using the temperatures from=20
"/sys/devices/platform/thinkpad_hwmon/temp*_input" and=20
"/sys/devices/platform/coretemp.*/temp1_input".
(These values are all *1000 so the limits below need to be *1000 too.)

Most of these descriptions and limits I'm using are guesses based on=20
information from thinkwiki.org (mostly the tp-fancontrol bash script).

 sensor		 description			warm	hot	veryhot
temp1		CPU [1]				66	70	72
temp2		Between CPU and PCMCIA slot	46	54	60
temp3		PCMCIA slot			45	53	56
temp4		None
temp5		BAT0, sensor0 (front left) [2]
temp6		BAT1, sensor0			35	38	40
temp7		BAT0, sensor1 (rear right)	35	38	40
temp8		BAT1, sensor1			35	38	40
temp9		Between northbridge and DRAM	45	57	61
temp10		Southbridge, under miniPCI	48	58	63
temp11		Power circuitry, under CDC	54	62	64
temp12		None
temp13		None
temp14		None
temp15		None
temp16		None
coretemp.0	CPU0 [1]			73	76	80
coretemp.1	CPU1 [1]			73	76	80

[1] Maybe temp1 is outside of the processor while coretemp.* are inside it?=
 =20
temp1 is always lower.

[2] temp5 is always exactly 50000.  Is this a battery fault or something el=
se? =20
temp 7 works though.  Also ultrabay batteries: do they have two sensors?


Thanks for your assistance.

--J2SCkAp4GZ/dPZZf
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIBTYh9XC8Sh1vpNARAhzkAKCJrVs/OwcS6BIz0UCK9C1T2Sy+6QCaAuQU
4m1JNynfVkzsJqN6g13ju/4=
=/xjC
-----END PGP SIGNATURE-----

--J2SCkAp4GZ/dPZZf--