Take a look at our
ThinkPads.com HOME PAGE
For those who might want to contribute to the blog, start here: Editors Alley Topic
Then contact Bill with a Private Message

Shimodax fan control tool : sharing values

Forum for scripts, utilities like TPFanControl, IBM-ECW, 2-finger scrolling, etc.
Message
Author
nrj45
Sophomore Member
Posts: 212
Joined: Mon Apr 25, 2005 2:33 pm
Location: switzerland

Shimodax fan control tool : sharing values

#1 Post by nrj45 » Mon Nov 28, 2005 1:35 am

The goal of this thread is to compare values from Shimodax fan control tool (see this thread : http://forum.thinkpads.com/viewtopic.php?t=17715 ).

I'm curious to see if we could isolate some temperatures and by examinating the difference between them, we could perhaps understand to which sensor it belongs...

My configuration is in signature.
Desktop use : 800Mhz @ 0.700Vcore, graphics to 104MHz/120Mhz (according to the powerplay minperf setting).
After 30 minutes of desktop use in a 20°C room i get these values :

Fan : 00 / Temp.: Highest 51°C ( 49 47 40 51 33 n/a 21 n/a 43 51 44 n/a )
After 3 minutes of prime95 (1866MHz @ 1.212Vcore) with fan blocked to 0 :
Fan : 00 / Temp.: Highest 71°C ( 71 53 43 61 35 n/a 21 n/a 52 54 51 n/a )
After some minutes of desktop use (prime95 stopped and fan always blocked to 0 :-P ) :
Fan : 00 / Temp.: Highest 55°C ( 54 51 45 55 36 n/a 27 n/a 47 54 48 n/a )
Perhaps some of you remember of my "homemade cooling system" ? It consists in a copper sheet that bridges the 4 chips (cpu, gpu, north and south bridge).

As you see the 3rd temperature will continue to increase even after i stopped prime95 (this value will make 43°C -> 45°C as the CPU will make 71°C -> 54°C). Please confirm but i'm pretty sure that this value monitores one of the chips that my cooling system is bridging. It would be logical : the cpu gives its heat to the copper sheet, but as the fan is off the copper sheet will just share this heat with the other chips it is linked to.

So i guess that the 3rd value is the temperature of the gpu or northbridge or southbridge (the 3 other chips that are cooled by the copper sheet).

EDIT : BTW this 3rd value will also make a significant drop when i run the fan to level 7 :
Fan : 07 / Temp.: Highest 50°C ( 47 45 36 49 35 n/a 27 n/a 44 50 45 n/a )
2nd EDIT : After activating max perf for powerplay, the 4th value made a jump of 8°C (from 47°C to 55°C) in about 3 minutes. The 3rd value made a jump of 4°C and others remained stable. So it confirms more or less that the 3rd value belongs to the temp of one of the chipsets (north or south ?) since it is bridged to the gpu through the copper sheet. And the 4th is likely the gpu temperature.

Summary : after inspecting values changes in some situations i guess that :
- After gaming it seems that the 2nd and 3rd values are not those of chipsets... it seems that chipsets are monitored in the 3 last values
- 4th value is for the gpu
17.06.05:
t43p (2668G4G), PM750, 2Go dual, 1032GAX (100gb/5k/16mb) 2010 error msg, SXGA+ 14.1", V3200, DVD-RW, GBeth, Intel abg, bt, 9 cells, XPPro/Ubuntu, Fingerprint,
800MHz-0.7Vcore, LCD min -> 13Watts

Aristotle11
Sophomore Member
Posts: 134
Joined: Wed Jun 01, 2005 1:56 am

#2 Post by Aristotle11 » Mon Nov 28, 2005 3:31 am

Tried it on my T42 and it works great. I modified the fancontrol.ini file to...

Level#0= 52°C -> Fan 0
Level#1= 54°C -> Fan 3
Level#2= 56°C -> Fan 4
Level#3= 60°C -> Fan 5
Level#4= 70°C -> Fan 7

Currently Fan : 00 / Temp.: Highest 52°C ( 52 42 31 44 29 n/a 23 n/a 0 0 0 0 )

I'm thinking of knocking all those values up 2 degrees. I have CHC running the 1.7 GHz processor at 1.004volts.

The only thing I don't like is that it makes a loud beep every time the fan goes on or off. And I get a lot of "Can't read status" messages, but that may be CHC's temp monitor's being on (I don't seem to get the messages when I turn CHC temp monitoring off). But that is minor, since this is the greatest program since sliced bread. Now if I can only get this program to run in systray as a service

:wink:

Thanks Shimodax!
Aris

vpn-user
Sophomore Member
Posts: 168
Joined: Thu May 20, 2004 5:56 am
Contact:

#3 Post by vpn-user » Mon Nov 28, 2005 4:05 am

I think the 3rd value is the HDD temp! At least the value is nearly the same as shown in NHC or MobileMeter.
X300 (6478-15G) with 3GB of RAM and builtin 3G/UMTS, running Vista Business x86-32

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

Known sensors

#4 Post by Thinkerer » Mon Nov 28, 2005 6:36 am

Here are the known sensors, from the new Linux fan control script (by the time you read this the script may have migrated to the article page).

Code: Select all

#  Sensor     ThinkPad model
#             R51      T43
#  ---------- -------  -------
#  EC 0x78    CPU      CPU
#  EC 0x79    miniPCI  HDAPS (through EC)
#  EC 0x7A    HDD      ???
#  EC 0x7B    GPU      GPU
#  EC 0x7C    BAT      BAT (SouthWest)
#  EC 0x7D    n/a      n/a
#  EC 0x7E    BAT      BAT (NorthEast)
#  EC 0x7F    n/a      n/a
#  EC 0xC0    ???      ???
#  EC 0xC1    ???      mini-PCI?
#  EC 0xC2    ???      RAM?
Also, to repeat the questions from that Wiki talk page:

Any idea what are the sensors at EC offsets 0x79, 0x7A, 0xC0, 0xC1, 0xC2? On the T43, 0x79 seems to be the same as the HDAPS sensor (it never deviates by more than one degree from what the HDAPS sensor tells directly), but I don't know where it's located. Sensor 0x7A is uncorrelated with disk temperature and activity, so it can't be HDD like reported for the R52. Sensor 0xC1 seems to be under the palm-rest (see discussion in http://thinkwiki.org/wiki/Talk:Problem_with_fan_noise).

vpn-user
Sophomore Member
Posts: 168
Joined: Thu May 20, 2004 5:56 am
Contact:

#5 Post by vpn-user » Mon Nov 28, 2005 7:23 am

WTF is "HDAPS"?
X300 (6478-15G) with 3GB of RAM and builtin 3G/UMTS, running Vista Business x86-32

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#6 Post by Thinkerer » Mon Nov 28, 2005 7:40 am

vpn-user wrote:WTF is "HDAPS"?
In the table, "HDAPS" is the temperature reported by the accelerometer chip of the Hard Disk Active Protection System. No idea where it's located.

thommyf
Posts: 4
Joined: Tue Nov 01, 2005 5:44 am
Contact:

the chip is shown in the hardware maintenance manual!

#7 Post by thommyf » Mon Nov 28, 2005 10:06 am

the location of the acceleration chip is shown in the hardware maintenance manual !

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

Re: the chip is shown in the hardware maintenance manual!

#8 Post by Thinkerer » Mon Nov 28, 2005 11:13 am

thommyf wrote:the location of the acceleration chip is shown in the hardware maintenance manual !
Right! Page 140 here.

So on the T43, the HDAPS sensor (2nd value) is on the sytem board under the middle of the left edge of the mini-PCI card (close to the top left edge of the touchpad).

I suspect that the 9th sensor (0xC1), which is giving a lot of trouble, is further right under the mini-PCI card, near the soubhbridge and closer to the GPU.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#9 Post by Shimodax » Mon Nov 28, 2005 11:53 am

Btw, I had the thinkpad sitting on my desk for a few hours now (office temperature about 20°C) with AC but max. battery life scheme. Temperature readings are:

CPU 41°C (0x78)
HDD 38°C (0x79)
??? 31°C (0x7A)
GPU 40°C (0x7B)
BAT 32°C (0x7C)
BAT 30°C (0x7E)
??? 36°C (0xC0)
??? 43°C (0xC1)
??? 41°C (0xC2)


For those who did not follow the discussion on ww..thinkwiki.com note the C1 sensor. With tests yesterday I observed that the fan kicks in with the C! reaching 43°C or the CPU heating up to about 50°C. It goes off again if both fall below 40°C (probably 38°C).

In my office environment the C1 fluctuates around 41°C to 44°C so with just 2°C more on the BIOS contoller it would probably be off most of the time.

Also, yesterday evening I was using the machine while sitting on the sofa with a blanket and the laptop in my lap. It grew a lot warmer (around 40°C to 55°C) with light use (web broswing, compiling the fanctonrol tool).

Thinker observed on thinkwiki that the bottom area is used for cooling. Also, there are some air vents at the front and on the bottom near the front. These were sure blocked by the blanket, probably rendering the fan ineffective for this area even when it was on.


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#10 Post by Shimodax » Mon Nov 28, 2005 12:31 pm

Btw, I have something else that I would like others to test.

When I start XP (shutdown, turn off, turn on) and then run Fan Control it comes up with a value of 7. This means that XP/BIOS starts with the fan locked to it's fastest pace.

If I close FC, hibbernate, wakeup, start FC it comes up with a 0x80 (BIOS Auto) setting.

Does anybody else experience this?


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#11 Post by Thinkerer » Mon Nov 28, 2005 12:34 pm

Shimodax wrote:Also, there are some air vents at the front and on the bottom near the front.
Those vents are mostly for speakers or decoration. If you look at the maintenance manual page 108 and peek in the holes, you'll find out the following. The front of the T43 looks like this:

Code: Select all

+---------------------------------------------+
|     :::#:::::::::  ::#:::::::::::::    # #::|
+---------------------------------------------|
      AAA BBBBBBBBBBBBB CCCCDDDDDDDDD       EE
The holes in areas A and C are blocked by the speakers. Area B is blocked by the speakers assembly. Areas D and E are obstructed by the HDD. There is probably some air getting through D and E, but not very much -- blocking the whole front with sticky tape causes no change in idle temperatures.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#12 Post by Shimodax » Mon Nov 28, 2005 12:46 pm

Maybe it was caused by blocking the holes in the middle of the bottom. Or of course by insolating the whole bottom area.



Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Aristotle11
Sophomore Member
Posts: 134
Joined: Wed Jun 01, 2005 1:56 am

#13 Post by Aristotle11 » Mon Nov 28, 2005 3:17 pm

How are these numbers? I consistently get the highest temps for the HDD and GPU. Do you?

CPU 41°C (0x78)
HDD 49°C (0x79)
??? 35°C (0x7A)
GPU 50°C (0x7B)
BAT 33°C (0x7C)
BAT 27°C (0x7E)
??? 0°C (0xC0)
??? 0°C (0xC1)
??? 0°C (0xC2)
N/A 0°C (0xC3)

Right now I have the smart settings set to trun the fan on at 52 degrees, and if I run NHC in adaptive mode (600 @ 0.7v, 1.7 GHz @ 1.004v)), the fan never turns on.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#14 Post by Shimodax » Mon Nov 28, 2005 4:15 pm

The second one (HDD) is cooler on my T43 but this may be a different model and the sensor may have a different location or meaning (besides the fact that on the T43 it's not the harddisk anyway, I just misnamed that).

Personally I consider everything under 55°C quite safe ... but that just a humble opinion, I don't know where most of the sensors are and I don't know all the hardware specs.

What model is that (I see that it does not have the C0-C4 sensors)?



Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Aristotle11
Sophomore Member
Posts: 134
Joined: Wed Jun 01, 2005 1:56 am

#15 Post by Aristotle11 » Mon Nov 28, 2005 4:47 pm

Hi Marcus,

My T42 is a 2378-EUU. The HDD is slightly hot since it is a 60GB 7200RPM Hitachi 7K60. Here are the specks...

IBM ThinkPad T42 Model 2378-EUU 14.1" XGA
1.73 GHz Gotham (NHC undervolted to 0.7v min, 1.004v max)
60GB 7200RPM Hitachi 7K60
1.25 GB Memory
CDRW-DVD/RW
GigaBit Ethernet, Intel 802.11b/g, Bluetooth
ATI RADEON 7500 32MB (underclocked to 159/159 mem and core)
6-Cell Li-Ion Battery

All the Best,
Aris

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#16 Post by Thinkerer » Tue Nov 29, 2005 11:45 am

Woa! It looks like the new BIOS+EC change the sensor readouts quite a bit on my T43:

My CPU readouts are now 3-4 degrees lower than before (used to be slightly higher than GPU, now slightly lower), and

Sensor 0xC2, which used to be quite cool, now reads a disturbing 51 in idle (everything else is under 46)! It now triggers low fan at 49deg (and it the first to go off when resuming from a long suspend). Even at 52deg, it does not trigger high fan. What's going on here? The BIOS+EC made 0xC2 now look like the hottest part of the system, either because they made it hotter or because they some interpretation of its value; but the EC is also very lax with it. Spooky.

0xC1 looks a bit cooler than before, and much cooler than the new 0xC2 (and is still clearly correlated with WLAN). The other sensors look pretty much like they used to.

Update: after letting the machine idle some more, 0xC2 is dropped a few degrees. It's still the hottest (which never happened with the old BIOS+EC), but not that much. Odd.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#17 Post by Shimodax » Tue Nov 29, 2005 6:22 pm

Thinker,

I just added the following to the secret sensor talk on thinkwiki.

Just now I see the C2 higher than C1 and rest of the system for the first time. Only difference I can think of is the fact that the battery is loading. I hooked it on with 6% left about 30 minutes ago. Usage was mainly web broswing (firefox, maybe a webpage with animated gif ads). C2 triggered the fan at 50°C two times.

CPU 42°C (0x78)
APS 41°C (0x79)
X7A 34°C (0x7A)
GPU 44°C (0x7B)
BAT 40°C (0x7C)
BAT 31°C (0x7E)
XC0 40°C (0xC0)
XC1 46°C (0xC1)
XC2 48°C (0xC2)

----

C2 was already rising again, this second (about 2 minutes after the above) it triggered the fan again:



Just now I see the C2 higher than C1 and rest of the system for the first time. Only difference I can think of is the fact that the battery is loading. I hooked it on with 6% left about 30 minutes ago. Usage was mainly web broswing (firefox, maybe a webpage with animated gif ads). C2 triggered the fan at 50°C two times.

CPU 42°C (0x78)
APS 41°C (0x79)
X7A 34°C (0x7A)
GPU 44°C (0x7B)
BAT 40°C (0x7C)
BAT 31°C (0x7E)
XC0 40°C (0xC0)
XC1 47°C (0xC1)
XC2 50°C (0xC2)

Batter status at 46% loading.


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#18 Post by Shimodax » Tue Nov 29, 2005 7:50 pm

When I booted again after the system was off for about 20 min. while battery charging (now at 95%) the C2 is still the highest temperature.

CPU 33°C (0x78)
APS 33°C (0x79)
X7A 28°C (0x7A)
GPU 34°C (0x7B)
BAT 28°C (0x7C)
BAT 27°C (0x7E)
XC0 31°C (0xC0)
XC1 33°C (0xC1)
XC2 37°C (0xC2)



Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#19 Post by Thinkerer » Wed Nov 30, 2005 3:32 am

Shimodax, 0xC2 indeed seems related to battery charging (I didn't notice that the EC update reset the charge thresholds so the EC started charging, and kept doing so when I booted to Linux...). When not charging, 0xC2 is still slightly higher than other components at idle with low fan, but nowhere close to 51deg.

OK, so sensor 0xC2 monitors some power circuitry affected by the ~2A@11V of battery charging. Can you spot any particular area getting warm when charging? Should be easy to spot when the machine is off and charging.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#20 Post by Shimodax » Wed Nov 30, 2005 10:02 am

Thinkerer wrote: OK, so sensor 0xC2 monitors some power circuitry affected by the ~2A@11V of battery charging. Can you spot any particular area getting warm when charging? Should be easy to spot when the machine is off and charging.
I'll try if I can feel something on the next battery cycle.


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#21 Post by Shimodax » Thu Dec 01, 2005 7:24 am

Thinkerer wrote: OK, so sensor 0xC2 monitors some power circuitry affected by the ~2A@11V of battery charging. Can you spot any particular area getting warm when charging? Should be easy to spot when the machine is off and charging.
I have taken the machine from the car today (stone cold) and plugged it in for charging. My guess is that the C2 would be rear/left (more or less under the Esc/F1 key). There are air vents in bottom, it's near the power socket and to the touch it was the warmest place. When turning on the utility showed C2 and BAT as the highest values, so technically it could also be the battery (BAT) sensor, although I'd expect that to be somewhere inside the battery.


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#22 Post by Thinkerer » Thu Dec 01, 2005 8:27 am

Shimodax wrote:My guess is that the C2 would be rear/left [snip]. When turning on the utility showed C2 and BAT as the highest values, so technically it could also be the battery (BAT) sensor, although I'd expect that to be somewhere inside the battery.
Makes sense. I'll check too on my next cycle. BTW, by "BAT", I assume you mean 0x7C rater than 0x7E? Someone who took apart a T4x battery reported that 0x7C is in the front left (near the connector where the in-battery circuitry is), and 0x7E is in the rear right. My experience is that when the battery isn't in use, 0x7E is pretty much an ambient temperature sensor.

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#23 Post by Shimodax » Thu Dec 01, 2005 8:32 am

I don't remember which of the two BATs that was. But from your description it probably was the left/front because the cells themselves (body of the battery) was cold and front/left would be near where I suspect the C2, so the heat from the charger probably crept in there (or it was indeed the battery's own charging heat).


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

duffy
Freshman Member
Posts: 101
Joined: Wed Jun 23, 2004 6:03 pm
Location: D.C.

#24 Post by duffy » Thu Dec 01, 2005 3:52 pm

Everything works great on a new Z60t. Here's my data:

Lenovo Z60t
1.73GHz (soon to be upgraded to a 2.26)
2GB RAM
80GB 7200RPM Hitachi 7K100
CDRW-DVD
WiFi

CPU 45°C (0x78)
APS 51°C (0x79)
X7A 41°C (0x7a)
GPU 0°C (0x7b)
BAT 24°C (0x7c)
BAT 23°C (0x7e)
XC0 40°C (0xc0)
XC1 43°C (0xc1)
XC2 45°C (0xc2)

I just installed v.15 and am ignoring everything except for the CPU. I'm going to lower the threshold to activate fan speed 1. I did find out that APS is the Active Protection Sensor which is located on the bottom side of the mobo next to the port replicator port. I found the location in the hardware manual. I cooled that section of the machine this morning and saw the APS temp drop immediately.

Fan speeds are as follows:
0 = Off
1 = ~1600-1800 RPM (hd is louder)
2 = ~1600-1800 RPM (hd is louder)
3 = ~2800 RPM (audible) This is the speed the fan usually runs at when on BIOS. Once the fan turns on via the BIOS it doesn't turn off.
4 = ~2800 RPM (audible)
5 = ~2800 RPM (audible)
6 = ~3500 RPM (audible)
7 = ~3500 RPM (audible)

christopher_wolf
Special Member
Posts: 5741
Joined: Sat Oct 08, 2005 1:24 pm
Location: UC Berkeley, California
Contact:

#25 Post by christopher_wolf » Thu Dec 01, 2005 5:28 pm

Hmmm, Suprising....

I was not aware that this "Fan Problem with the T43" was so widespread; T42s, R50s, X Series, and now the Z Series :lol:

Once again, thanks for all your hard work and deciding to share this tool with us Shimodax. :D
IBM ThinkPad T43 Model 2668-72U 14.1" SXGA+ 1GB |IBM 701c

~o/
I met someone who looks a lot like you.
She does the things you do.
But she is an IBM.
/~o ---ELO from "Yours Truly 2059"

Paul Unger
Senior Member
Senior Member
Posts: 565
Joined: Mon Jun 14, 2004 8:18 am
Location: Solomon Islands / Canada

#26 Post by Paul Unger » Fri Dec 02, 2005 7:58 pm

I finally got the courage--after reading all the disclaimers!--to give this thing a try. I'm in the Solomon Islands, so controlling temperature is quite a concern! I have a T42 2373-3UU (1.7GHz, 1GB RAM, 40GB 5400RPM HD, SXGA+) and "Fan Control" ran just fine. Here are some numbers:

CPU 41C
APS 46C
X7A 35C
GPU 46C
BAT 33C
BAT 31C
XC0 0C
XC1 0C
XC2 0C

I say it "ran fine" but not without issues . . . First of all, I disabled CHC (running 'dynamic switching' at .700V@6x/1.020@17x). Then I ran "Manual" mode for a while, checking out each setting and comparing fan RPM: 1/2 were close enough to be called the same; ditto 3/4/5; and ditto ditto 6/7. I then switched to "Smart" mode. No "fried eggs" :) But with the 'factory' settings my computer started getting warmer than before running "Fan Control". I made some changes in the .ini file--changed/added some levels (Level=0 35, Level=1 40, Level=2 45, Level=3 50)--and it still stayed warmer than before.

I noted above that I disabled CHC. I tried enabling it to run concurrently with FC, but got consitent (and frequent) warnings about "Can't read status". Just as advertised, I guess . . . Finally, I switched back to "BIOS" mode, shut down FC, left CHC running, and my temps dropped back to "normal".

So in the final analysis, I think I'll continue running CHC until(?) CHC and FC won't fight! But I can't complain about this program! I'm thrilled that you built this thing, Shimodax! Amazed that the Thinkpad community has contributed so much to its development so quickly! Things like this help to renew one's hope in the human race :wink: I attribute I8kfangui--a fan control utility for the Dell Inspiron family--with saving my Dell I8000, so I'll keep watching this thread for further developments. Thanks for all your work,

Paul

christopher_wolf
Special Member
Posts: 5741
Joined: Sat Oct 08, 2005 1:24 pm
Location: UC Berkeley, California
Contact:

#27 Post by christopher_wolf » Fri Dec 02, 2005 8:35 pm

Mmmm, certain parts of the Thinkpad might get much warmer than they otherwise would on Smart control rather than BIOS. After many trials, I created a power manager scheme that would work with a given *.ini file setup. I am starting to think that the fan is simply spooling up more and more until the change in the rate of temperature in various parts of the Thinkpad level off to certain steady state value. I have watched the output logs for awhile now as well as spending 45 minutes looking at it real-time. I noticed that there are certain sensors that would increase way more than others would for any given drop in fan speed.

It would be great to have a real-time, depending of the temperature-check cycle of course, output of these values to a file for further analysis. I am reasonably sure that we could find some distinct heating//cooling trends in certain areas of the Thinkpad and plan out some more efficient Smart mode control routines. I would *love* to get a good schematic of the cooling system and combine that with some thermal response curves. We could get a really good sense of what to ignore and what not to ignore, when to turn the fan on at what level and when to turn it off, etc. :D

On the issue of ignoring sensors; I am well aware that the BAT Sensors are not essential to the proper cooling operation of the Thinkpad. Since the BIOS seems to not ramp up the input to the fan based on the BAT sensors, there is little cause to worry about those and they are not included in the control profile in either BIOS Mode or Smart Mode. The point being that you should not ignore something like APS, that is just plain stupid...Until I actually know what XC1 and XC2 really measure, I will keep a close eye on them.

Great Job, Shimodax, keep up the good work :D
IBM ThinkPad T43 Model 2668-72U 14.1" SXGA+ 1GB |IBM 701c

~o/
I met someone who looks a lot like you.
She does the things you do.
But she is an IBM.
/~o ---ELO from "Yours Truly 2059"

Thinkerer
Posts: 46
Joined: Mon Nov 28, 2005 6:04 am
Contact:

#28 Post by Thinkerer » Fri Dec 02, 2005 9:09 pm

duffy wrote: Lenovo Z60t

CPU 45°C (0x78)
APS 51°C (0x79)
X7A 41°C (0x7a)
GPU 0°C (0x7b)
BAT 24°C (0x7c)
BAT 23°C (0x7e)
XC0 40°C (0xc0)
XC1 43°C (0xc1)
XC2 45°C (0xc2)

I just installed v.15 and am ignoring everything except for the CPU. I'm going to lower the threshold to activate fan speed 1. I did find out that APS is the Active Protection Sensor which is located on the bottom side of the mobo next to the port replicator port. I found the location in the hardware manual. I cooled that section of the machine this morning and saw the APS temp drop immediately.
I don't think anyone checked the sensor mapping on the Z60t before, so the names may be wrong for your machine. You verified the APS, and the BAT look right, but do CPU and GPU also behave as expected? Also, can you see correlation between sensor 0xC1 and WiFi activity (heavy traffic vs. radio off), like on the T43? And between sensor 0xC2 and battery charging?

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#29 Post by Shimodax » Sun Dec 04, 2005 4:15 pm

Just for the record, this was after a round of gaming (20 minutes of Quake4 demo)

CPU 60°C (0x78)
APS 52°C (0x79)
X7A 38°C (0x7a)
GPU 61°C (0x7b)
BAT 36°C (0x7c)
BAT 24°C (0x7e)
XC0 55°C (0xc0)
PCI 53°C (0xc1)
PWR 52°C (0xc2)


7A is still rather cool, while C0 went quite high (highest I've ever seen). Maybe it's the Northbridge, catching some stray heat from the CPU and GPU. Would probably make sense if the C1 is the Southbridge. Maybe 7A is more in the area of the Ultrabay.

But that's all just wild guessing.


This is about 4 minutes later (just after starting the web browser and writing the above) with fan set to 7:

CPU 49°C (0x78)
APS 45°C (0x79)
X7A 35°C (0x7a)
GPU 51°C (0x7b)
BAT 34°C (0x7c)
BAT 24°C (0x7e)
XC0 44°C (0xc0)
PCI 49°C (0xc1)
PWR 46°C (0xc2)

XC0 cools quicker than CPU/GPU.


Markus
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Shimodax
Moderator Emeritus
Moderator Emeritus
Posts: 106
Joined: Fri Nov 25, 2005 6:29 pm

#30 Post by Shimodax » Sun Dec 04, 2005 5:12 pm

Here's an explanation of the North/Southbridge. With that it would not be unlogical that the northbridge heats up during a game, where large bulks of data is exchanged between CPU, GPU and RAM.

I just wish I had the guts to open the baby and play with some cooling spray.


Markus

North/South

The old way of doing chipsets has two components, the northbridge and the southbridge. The northbridge communicates with the CPU over the Front Side Bus (FSB) and acts as the controller for memory, AGP and PCI. The type of FSB, memory and AGP varies from northbridge to northbridge. Some northbridges integrate video as well.

The southbridge takes care of most basic forms of I/O, such as USB, serial ports, audio, IDE and more. What I/O is controlled depends on the specific southbridge. The southbridge sits on the northbridge's PCI bus, which is usually a 32-bit, 33MHz bus capable of providing 133MBps of bandwidth. Intel's 440BX chipset and all current VIA and AMD PC chipsets use this type of northbridge/southbridge architecture.
T43 2668-97G (2GHz, 512MB, 80GB, 15" SXGA)

Annoyed by fan noise? Check out the TP 43 Fan Control thread
http://forum.thinkpads.com/viewtopic.php?t=17715

Post Reply
  • Similar Topics
    Replies
    Views
    Last post

Return to “ThinkPad Utility Work Area”

Who is online

Users browsing this forum: No registered users and 15 guests