Welcome to our peer-to-peer forums, where owners help owners. Need help now? Visit eSupport here.

English Community

ThinkPad NotebooksThinkPad: T400 / T500 and newer T series Laptops
All Forum Topics
Options

62 Posts

06-08-2011

USA

74 Signins

339 Page Views

  • Posts: 62
  • Registered: ‎06-08-2011
  • Location: USA
  • Views: 339
  • Message 831 of 1217

Re: T420s Fan noise Issue

2012-01-19, 0:20 AM

So I did a little test. When my t420s is idle after boot using battery power the temp is between 32c and 38c with no fan running. When I boot up using AC power connected to my lenovo dock the core temp when idle is between 46c and 52c (it jumps up to 70c with web browsing).  Fan is always at 4k rpm at any temps above 48c.

 

I have the Nvidia GPU in mine which I understand is always on when the laptop is connected to a dock. This seems to be the real source of the high temp/ max fan speed for me. 

 

Has anyone found a way to disable the Nvidia GPU when using the dock?

Reply
Options

87 Posts

01-25-2010

Canada

388 Signins

3631 Page Views

  • Posts: 87
  • Registered: ‎01-25-2010
  • Location: Canada
  • Views: 3631
  • Message 832 of 1217

Re: T420s Fan noise Issue

2012-01-19, 0:59 AM

Anyone have any idea what is the room temp that they are testing it in?  I'm in an air conditioned room at 20 degrees C with the same issues.

_______________________________________________________
T420s, i5-2520M @2.5, nVidia 4200m, 8GB DDR3, OCZ Vertex 3 Max IOPS 240GB SSD, Hitachi 320GB @7200rpm ulrabay, 8Pro x64
T61, Centrino Duo @1.8, GM965, 4GB PC2-5300 DDR2 @667MHz, OCZ Agility 120GB SSD, 7Pro x64
Reply
Options

5 Posts

01-19-2012

Poland

5 Signins

95 Page Views

  • Posts: 5
  • Registered: ‎01-19-2012
  • Location: Poland
  • Views: 95
  • Message 833 of 1217

Re: T420s Fan noise Issue

2012-01-19, 11:00 AM

@David44,

 

Also new to this board, although I have also been anxiously following this thread for a long time before I bough my t420s. I have been using it for about a week now, and what you describe corresponds  to what I see: the stuck fan at 4000rpm happens to me after 1-2h of work when the fan is Bios-controlled and only when the PC was plugged in at boot time. Changing the AC vs battery settings in the bios did not change this behavior.

 

@all,

 

I start to think that the stuck fan problem is actually very widespread, and possibly affects all units. Case in point, I recently noted that the problem was actually already decribed in NotebookReview's lengthy article on the t420s.

However, for many (a majority?) of people, software control (Lenovo Power Manager, Thinkfan, etc) appears to work. This is I guess why many people (myself included) don't notice it with Windows where Lenovo Power Manager fixes the problem (don't recall if LPM was enabled by default, but I guess most users would anyway quickly enable it to get a quieter life).

 

Although I did think that the fan was still a little loud in Win7 with LPM, I never got a stuck fan until I installed Linux where the fan was BIOS-managed. As written above, this would only happen when the machine is  AC powered. Rebooting would solve the problem, and IIRC suspend works too.

Then I did a test: after the fan had been stuck for an hour even though the CPU was all the time at 38-39C (!), I installed Thinkfan and started it while the fan was stuck and ... miracle, silent fan! I have now been using Thinkfan for a couple of days and never got again the stuck fan problem, even though my Thinkfan configuration is such that it lets the BIOS take over fan management above some temperature treshold.

 

Bottom line: as quite a few people have already pointed out, the problem (which I start to believe is extremely widespread) has nothing to do with i7 (I have i5), with discrete graphics (I have integrated) or with the OS (I had this problem on Linux). I guess that for most users, however, software fan management works, whereas according to this thread for others it doesn't.

 

 

----------------------------------------------------------

T420s, i5-2520M integrated, intel 310 80GB, BIOS 1.26, OS: Debian Wheezy + Win 7

 

Reply
Options

23 Posts

12-17-2011

Germany

257 Signins

377 Page Views

  • Posts: 23
  • Registered: ‎12-17-2011
  • Location: Germany
  • Views: 377
  • Message 834 of 1217

Re: T420s Fan noise Issue

2012-01-22, 14:20 PM
Hello, I made another observation: If you set the fan control of Power Manager to "Reduce fan speed dynamically" and start DPC Latency-Checker every 5-10 seconds a noticeable peak of around 1000µs pops out. I don't know what this could mean for our problem, just wanted to let you guys know.
Reply
Options

107 Posts

06-28-2011

Italia

2191 Signins

3793 Page Views

  • Posts: 107
  • Registered: ‎06-28-2011
  • Location: Italia
  • Views: 3793
  • Message 835 of 1217

Re: T420s Fan noise Issue

2012-01-22, 19:49 PM

Hi everybody,

 

My T420s is back from techy land. They have changed my fan and it does work better now, even under Linux. To everybody who has his fan permanently stuck at 4000 RPM, you might be lucky with a new fan.

 

Now, even though I am happier with my new fan, I see that the fan sometimes gets stuck at 4000 RPM (before, it was always spinning fast). As previously reported by others, as soon as I put the laptop to sleep and resume, the fan slows back at 2000 RPM. The fan getting stuck at 4000 RPM seems to happen randomly for now, I haven't identified a reliable pattern. However, it can guarantee that it happens, so please Lenovo, keep the testing unit for longer than an hour to address this!!

 

Secondly, I believe that the switching threshlods are a little too aggressive. Could you consider raising the temperature limits (and maybe whatever else is used for regulating the fan speed) a bit, to make things quiter. Just a bit would suffice and make things much better!

 

Cheers,

Alessandro

Reply
Options

5 Posts

01-19-2012

Poland

5 Signins

95 Page Views

  • Posts: 5
  • Registered: ‎01-19-2012
  • Location: Poland
  • Views: 95
  • Message 836 of 1217

Re: T420s Fan noise Issue

2012-01-23, 22:05 PM

Alessandro, what is your power consumption under linux? Depending on the kernel and boot options, I noted that one may get anything between 7W to 16W when idle in GUI mode (gnome/kde/xfce/whatever), thanks largely to the infamous kernel power regression. My current understanding is that the "stuck" fan happens mainly when the laptop's body temperature becomes too high to allow the CPU to be cooled down to ~37C, and since the temperature of your laptop after an hour or so is strongly correlated to the average power, reducing the power consumption really helps a lot (especially if you are reluctant to use Thinkfan).

 

I will perhaps write someday a longer post on this topic, but in the meantime I strongly recommend the new kernel serie (3.2) together with boot options i915.i915_enable_rc6=1 and i915.i915_enable_fbc=1 (but without pcie_aspm=force). With this and a few tweaks the laptop runs at very lower power and cooler (extra bonus: longer battery life than even Win 7!).

 

Reply
Options

1 Posts

01-23-2012

France

5 Signins

21 Page Views

  • Posts: 1
  • Registered: ‎01-23-2012
  • Location: France
  • Views: 21
  • Message 837 of 1217

Re: T420s Fan noise Issue

2012-01-23, 22:21 PM

Any solution for T410 ? using tpfan control but i busy from it.

 

i want a regular computer no fuc*** softwares lol

 

Fan running around 3500 RPM.  Actually studing very importante the the silence !!!!!

Reply
Options

107 Posts

06-28-2011

Italia

2191 Signins

3793 Page Views

  • Posts: 107
  • Registered: ‎06-28-2011
  • Location: Italia
  • Views: 3793
  • Message 838 of 1217

Re: T420s Fan noise Issue

2012-01-23, 23:05 PM

wrote:

Alessandro, what is your power consumption under linux? Depending on the kernel and boot options, I noted that one may get anything between 7W to 16W when idle in GUI mode (gnome/kde/xfce/whatever), thanks largely to the infamous kernel power regression. My current understanding is that the "stuck" fan happens mainly when the laptop's body temperature becomes too high to allow the CPU to be cooled down to ~37C, and since the temperature of your laptop after an hour or so is strongly correlated to the average power, reducing the power consumption really helps a lot (especially if you are reluctant to use Thinkfan).

 

I will perhaps write someday a longer post on this topic, but in the meantime I strongly recommend the new kernel serie (3.2) together with boot options i915.i915_enable_rc6=1 and i915.i915_enable_fbc=1 (but without pcie_aspm=force). With this and a few tweaks the laptop runs at very lower power and cooler (extra bonus: longer battery life than even Win 7!).

 


Well, I am already using all the intel settings you mention :) Why do you recommend avoiding pcie_aspm=force?

 

Secondly, I have enabled a good number of powersave options from the ones listed, for example, at https://bbs.archlinux.org/viewtopic.php?id=134109

 

That means that my laptop should be quite well configured for low power consuption.

 

Despite all this, I have to agree with you that the fan stuck problem mainly happens after some time running programs that require a high computational cost (e.g. C++ simulations, archive compression and such things). These might dramatically raise the laptop temperature, requiring a longer time to cool things down.

 

As a bottom line I still think that the fan management might be improved. As a first suggestion, as many other have proposed, I think that lowering the middle speed of 3500 RPM to 3000 would help. I am strongly sure that's how it worked in early BIOS releases.

 

For better clarity, what I suggest is having a low fan speed of 2000 RPM when the laptop is running light tasks. Then raise the fan speed to an intermediate value of 3000 RPM when the workload increases. Finally, when the temperature goes even higher, sprint the fan to 4000 RPM. The point is that now the 3500 RPM value is very close to 4000 RPM, and the thresholds for the two mode seem to be very close, since the fan only runs at 3500 RPM for short time period. Better separating these two modes, by lowering the intermediate speed and raising the threshold of the 4000 RPM would make a striking improvement in noise management (that is my opinion, of course).

 

Cheers,

Alessandro

Reply
Options

5 Posts

01-19-2012

Poland

5 Signins

95 Page Views

  • Posts: 5
  • Registered: ‎01-19-2012
  • Location: Poland
  • Views: 95
  • Message 839 of 1217

Re: T420s Fan noise Issue

2012-01-24, 0:22 AM

OK Alessandro, I see we have been on a similar Watt-saving hunt ;-)

Note that the laptop-mode-tools package does many of the optimizations from the Arch thread automaticaly.

Option pcie_aspm=force was for kernel 3.1 and earlier. Looks like this option actually hurts on 3.2 and that 3.2 is yet better than 3.1 with, power-wise (at least on the X220):

http://techpatterns.com/forums/about1963.html

I would need to make more investigations to reliably compare 3.1 (with aspm) vs 3.2 (w/o aspm), but I have the feeling that there is >1W of difference when idle.

 

In any case, I agree with you and other people in this thread: the BIOS fan speeds and temperature treshold seem way too conservative. Another thing: it looks like one gains nearly 1.5W by taking the fan from 4000RPM down to 2000RPM! Another very good reason to change the BIOS settings...

Reply
Options

143 Posts

11-15-2009

Stockholm

822 Signins

2288 Page Views

  • Posts: 143
  • Registered: ‎11-15-2009
  • Location: Stockholm
  • Views: 2288
  • Message 840 of 1217

Re: T420s Fan noise Issue

2012-01-30, 21:31 PM
I'm guessing this issue is no longer getting any love and attention from the Lenovo engineers. Does anyone have any updates they would like to share with the rest of us?
Reply
Forum Home

Community Guidelines

Please review our Guidelines before posting.

Learn More

Check out current deals!

Go Shop
X

Save

X

Delete

X

No, I don’t want to share ideas Yes, I agree to these terms