T43 noise

Message

JimL · #1 Post by **JimL** » Mon Nov 27, 2006 11:49 am

I've seen quit a few posts about the T43 making a lot of air noise. It occurred to me to wonder if there has been a related history of failures of some kind or if this is merely a poor "wind tunnel" or fan blade design problem or something like that.

Maybe I should just ask. Why are they noisy?

Senser · #2 Post by **Senser** » Fri Dec 01, 2006 6:39 am

I'm not sure, having the same problems with my T43. Some say it's a BIOS issue, others say it's by design.

I've contacted IBM (or Lenovo) on this and they claim fan noise is no known issue.

agarza · #3 Post by **agarza** » Fri Dec 01, 2006 10:18 pm

After reading a post here on the forums, I've found that the hyperactive fan is because the Northbridge/Southbridge chips produce too much heat, even when the CPU is undervolted.

There's a special fan assembly for the T43 models that spreads the heat on the Northbridge, resulting on lower temps.

dmar · #4 Post by **dmar** » Sat Dec 02, 2006 10:36 pm

Can you detail a bit further what that special fan is, and if it's user replaceable?

gunston · #5 Post by **gunston** » Sun Dec 03, 2006 8:56 am

use TP fan control and show your temperature reading here.

agarza · #6 Post by **agarza** » Sun Dec 03, 2006 10:53 am

dmar wrote:Can you detail a bit further what that special fan is, and if it's user replaceable?

It's the FRU 26R8197, there's one available on eBay:
IBM ThinkPad T43 Laptop HeatSink FAN 26R8197

You can replace the fan by yourself, if your machine then develops a problem that needs replacement or something, most of the time IBM technicians won't notice you opened the machine.

Amigaman · #7 Post by **Amigaman** » Wed Dec 06, 2006 1:08 pm

According to this forum topic:

http://forum.thinkpads.com/viewtopic.ph ... ht=26r8197

You CANNOT use the 1878-XXX "special" Fan in the T43 machines with external GPU (2686-XXX, etc). It says that it is the wrong size.

So I don't think that your suggestion will work, unless you have done it, or someone can say that they did it?

Gustavo · #8 Post by **Gustavo** » Wed Dec 06, 2006 1:47 pm

Amigaman ahhh Amiga that was one fine gaming computer in the old days. I just loved those white little devils.

Fan control does the trick for the T43 noise. But compared to my T42 the T43 is like an hairdryer. But with TP fancontrol it becomes much better but the main reason I love the T43 is its performance which is alot better than off my T42.
But some might disagree with me on that. Great machines both off them thou. But I miss my old X31 portability which was great.

Senser · #9 Post by **Senser** » Wed Dec 06, 2006 1:49 pm

FanControl does indeed solve the general fan noise, however in my case I receive a buzzing noise in return.

Amigaman · #10 Post by **Amigaman** » Wed Dec 06, 2006 9:04 pm

@Gustavo

Yes, Amiga's rule

Does the fan on your T43 pulsate/buzz every ~4.8s in Windows, especially if the fan runs at low-moderate speed?

@Senser

Yes, even though we can control our fan speed, there is a typo/bug in the T43 EC Firmware!!

They fixed the buzz in the T40-T42, but mis-typed the fix for the T43/R52. I can't believe there has not been more of an outcry over something like that, it would probably improve the issue of the pulse/buzz sound for many T43 owners.. A similar fix for the pulsing sound is used in Linux, but I know of no Windows program that fixes the pulse/buzz.

USSS · #11 Post by **USSS** » Thu Dec 07, 2006 12:47 am

Does the fan on your T43 pulsate/buzz every ~4.8s in Windows, especially if the fan runs at low-moderate speed?

Yes, even though we can control our fan speed, there is a typo/bug in the T43 EC Firmware!!

I've had the same pulsing "chime" problem on my T43 (2687) since I purchased it in October 2005. Yet, this is the first post I've seen that actually measures it at every 4.8 seconds (I thought it was about 5 seconds!).

It's a firmware bug in the Embedded Controller? Jeez...where's my hex editor?

Senser · #12 Post by **Senser** » Thu Dec 07, 2006 1:59 am

Amigaman wrote: Yes, even though we can control our fan speed, there is a typo/bug in the T43 EC Firmware!!

They fixed the buzz in the T40-T42, but mis-typed the fix for the T43/R52. I can't believe there has not been more of an outcry over something like that, it would probably improve the issue of the pulse/buzz sound for many T43 owners.. A similar fix for the pulsing sound is used in Linux, but I know of no Windows program that fixes the pulse/buzz.

I'll contact IBM again about this issue. I've contacted them a few weeks ago and they claim fan noise is no known issue. Also the buzzing noise was not something the person I spoke to knew about.

Prospect · #13 Post by **Prospect** » Thu Dec 07, 2006 4:53 am

The firmware was changed for these kind of issues and can now be resolved by updating your Bios and Embedded Controller Chip.

Access the IBM website and search for:

MIGR-59120
MIGR-59097

Both documents will provide the latest updates for Bios and Embedded controller for T43's.

After updating this and rebooting your machine, the problem should be resolved.

Amigaman · #14 Post by **Amigaman** » Thu Dec 07, 2006 9:46 am

@USSS

I didn't personally measure the ~4.8s, I count around 5s too, but someone else calculated it to be that approximately that

I read it on ThinkWiki and on a forum on this web site.

While the underlying cause for the pulse is not necessarily known, a typo/bug in the T43 EC firmware, prevents us from benefiting from a "treatment" which may reduce the pulsing to an acceptable level.

In Linux they use an ACPI script to implement the same "treatment", we could try to implement this in tpfancontrol, but ideally:

Lenovo could *fix* that typo/bug in the EC firmware, thus alleviating us of having to do it via software in Windows.

It is probably too dangerous for us to apply modifications to our EC firmware, in case something goes wrong..

(Though I doubt that the service centres would notice if a T43 were to die this way. When my TP R32 BIOS flash failed a few years ago, they simply wanted to replace the motherboard, they didn't try to identify why it had died).

Amigaman · #15 Post by **Amigaman** » Thu Dec 07, 2006 9:59 am

@Senser

It's important to point out that the "general fan noise" (a potentially subjective issue) is distinctive to the ~4.8s pulse that affects some machines. IBM made a statement on "general fan noise" in the past, and it's not considered an issue.

Eventhough a "treatment" for the pulsing was placed into the EC firmware of all the late model T4X machines, and it was mentioned in the release notes briefly, I fear that the typical IBM/Lenovo support are unfamiliar with the fact that it was intended to treat fan pulsing, or that a typo sneaked into the T43 EC firmware, rendering the treatment completely ineffective.

But here are the release notes of the EC firmware for the T40/41/42:

Symptom corrected by version 3.03 - 1RHT70WW
Note: This version of Embedded Controller Program will only work with BIOS Version 3.06f (or higher).

* (Fix) Reduced Fan noise in some models.

This is where they added that fan pulsing treatment for the T40/41/42. They added it to our T43 EC firmware too, but they broke it with a typo. I wonder if IBM support would believe it...

Amigaman · #16 Post by **Amigaman** » Thu Dec 07, 2006 10:01 am

@prospect

No, the latest version T43 EC firmware does *NOT* fix the problem. There is a*typo* in the T43 EC firmware, the pulsing is not treated on the T43 machines.

It was treated on the T40/41/42 however, but that is of no use to T43 owners.

christopher_wolf · #17 Post by **christopher_wolf** » Thu Dec 07, 2006 1:29 pm

Could you please use the "Edit" button instead of starting new post replies?

Also the bug in the code doesn't preclude proper treatment of the problem. On Linux it is fairly easy to treat the symptoms, but TP FCU may perhaps have to be modified to do the same on Windows. As it stands, however, I have seen no "silent" T43s that get worse from a firmware update and not many loud T43s that benefit from the same firmware, I still have to look more at the hardware though.

Amigaman · #18 Post by **Amigaman** » Thu Dec 07, 2006 1:48 pm

Re: Reply posts, Oops

Yes there is an easy fix in Linux with the ACPI scripts, but I imagine that many affected users of the T43 run Windows only. Even though I run Linux on some machines here, I want Windows on this laptop

I can't see how any "silent" T43s would get worse, since the newest firmware with the broken "treatment/fix" didn't end up doing anything differently from the previous versions, the wrong register got used imediately, if I recall correctly.

No loud T43 can benefit, since once again the firmware "treatment/fix" is broken. What is interesting to know, is whether any loud T43s benefitted from the Linux script (which is similar to what the EC "fix/treatment" should've done).

I just discovered the Lenovo Blogs, and have spent some time reading through it all. Interesting stuff. I wonder if it will provide an opportunity to raise the "typo" with one of the appropriate Lenovo bloggers.. I notice that one person alread did something similar, with regards to "X60 new BIOS shortens battery life"..

Rose · #19 Post by **Rose** » Thu Dec 07, 2006 2:10 pm

i wonder if anyone has the same understanding as I do. Ive had several t4x before my T43 and still have several pre t43 at work, and on all of them it seems that the noise from the fan-motor it self is different and louder on the T43 compared to the others, at about the same rpm. And I dont aim at the airflow-noise, but simply the electrical/mechanical-noise.

My T41p is more silent at full blow compared to my T43p on level 1, not even concidering pulsation.

T43 noise

T43 noise

Who is online