An almost quiet laptop [I drilled into]

Show off your quiet rig.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
impaire
Posts: 7
Joined: Wed Sep 01, 2010 11:57 am
Location: France

An almost quiet laptop [I drilled into]

Post by impaire » Wed Sep 01, 2010 1:01 pm

Hi,

Some weeks ago, I decided to patch a brand new laptop, I'll share the pictures, they may help.

Before patching it like I did, the fan of this laptop, an U350 with a 1,3GHz ULV, never switched off for approx 20°C room temperature. The CPU was remaining at 42°C for a zero system load, with the fan active (noisy). I opened this U350 and noticed that this laptop had no fresh air intakes, none close to the fan.

The fan had several speeds, the fastest (like the first speed) being really loud for my needs (and ears). Some people do not complain, I think they never heard an almost quiet professional laptop. But several other do complain about "excessive fan noise"...

A few days later, I decided to drill inside, to add a fresh air intake to this case. From that point, drilling more and more into it, the CPU temp decreased by approx 10°C for highest system loads. I then decided to slow down also the fan speed using a common transistor/resistor circuit, to decrease the fan supply voltage.

This extra transistor itself heats a little bit up. For it's own cooling, I've mounted it pushed against the laptop's keyboard frame which is metallic. This provides an acceptable heatsink for this extra transistor.

I've lost the warranty. But I've now something quiet under my hands, even by highest system loads and room temperatures. This laptop survived over the summer, which is a good sign... I assume so.


Initially, there are no holes close to the fan. They are like forseen, but none in the box:

Image

As internal parts and connectors are really sensitive, instead of unmounting it again, once, I just returned this laptop and started to drill more and more into it (taking care of the fan, just behind, where I drilled). This is an intermediate result, with a sticker I used to close again those holes I drilled, to make comparisons. I drilled finally more than this:

Image

Just below it, you'll notice a Thinkpad. I made comparisons, this U350 is now almost as quiet as a r61. This is the transistor (plus a small SMD resistor) I could house under the U350 keyboard, to slow down the fan, which is now almost quiet, even at highest speeds:

Image


If you are interested in some temperature measurements I made, you could have a look here, from page 6 (initial measurements) to page 10 (tests with those two patches, holes plus a transistor):

http://forums.lenovo.com/t5/IdeaPad-Y-U ... 811#M23311

Oops... I couldn't post links, neither pictures. You must have 3 posts before you can post URL's/Links. I came back and posted some pictures.

Regards

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Fri Sep 03, 2010 10:00 am

Amazing that such a simple thermal solution was not implemented by the manufacturer. You should submit your mod and info about why it was necessary and how it has improved things.... and demand $$ for your expert consulting! :lol:

impaire
Posts: 7
Joined: Wed Sep 01, 2010 11:57 am
Location: France

Post by impaire » Fri Sep 03, 2010 12:05 pm

MikeC wrote:Amazing that such a simple thermal solution was not implemented by the manufacturer.
I am sometime also wondering why those holes below the fan weren't forseen, drilled or molded throught, by the factory. I assume that Lenovo has excellent reasons to deliver the U350 as this. But I couldn't guess or figure out which reasons those are.

First, I thought that the fan was also sucking fresh air from holes managed over the RAM and over the Wifi board. But this wasn't the case, I checked it, by fully removing the rear panel. I first drilled some holes only, to check how it would more globally behave, thinking that other parts could heat more or even overheat. But this never happened; I assume it would have crashed over the summer, with room temperatures far above 30°C.

In my opinion, noise and over heating issues are real nightmares for both manufacturers and end users. I've read several news about those, since last june. The last one concerns Toshiba, here just some excerpts:

30 Jun 2010 ... Potential overheating problems on Sony's Vaio laptops prompt the company to announce a recall of 535000 units...

3 Sep 2010 ... Toshiba is the latest to recall laptops that pose a fire hazard or risk of injury--highlighting the growing concern of excessive heat ...

The Sony recall was immediately following an HP recall of laptop batteries in May. HP recalled over 100,000 laptop batteries in all as a result of mounting reports that the batteries were overheating, rupturing, and posing a risk of injury to users.

While quality control during the engineering and manufacturing process at Toshiba, Sony, and HP is apparently lacking, the issue goes beyond the individual manufacturers and exemplifies problems associated with the demands placed on notebook computers.


You can read more here: http://www.pcworld.com/businesscenter/a ... ptops.html

impaire
Posts: 7
Joined: Wed Sep 01, 2010 11:57 am
Location: France

Post by impaire » Fri Sep 03, 2010 12:57 pm

MikeC wrote:Amazing that such a simple thermal solution was not implemented by the manufacturer. You should submit your mod and info about why it was necessary and how it has improved things....
I was looking for other articles about cooling issues, for such extra flat laptops, they are known by manufacturers. But I now found following, it may explain why this laptop had no holes close to the fan:

Intel on Notebook Cooling Jet Engine Style – Sort of
http://www.google.fr/search?q=laminar+j ... ing+laptop

"Sort of", like I've read, via Google. But from the jet, I had the noise, this is why I patched it.

Like I shortly explained earlier, I've also taken care about the heat of the bottom of this computer. I'm often using it on my lap. The main heat or hot area, there, was and is still over the Wifi board. I'm still trying to decrease the temps around there, around the RAM also. Yesterday, I decided to remove all the stickers in this area, from the SODIMM and the Wifi board, to see how this helps.


Edit: I made further tests and comparisons, following those last readings.

Removing internal stickers makes no noticeable changes. But as I've already lost the warranty, I prefer having those components or chips in direct contact with internal air, instead of being partially recovered by paper or stickers.

On the U350, the back remains much more cooler when the radios are turned off. But the bottom heat is still acceptable, even with Wifi active.

I used my r61 on my lap, on heavy load, for several hours. According to me, its bottom gets hotter than this patched U350 bottom, and the r61 heat is better spread over all its bottom. But for me, this r61 hot bottom is still acceptable, I won't complain about it. Other people and manufacturers would anyway say that this r61 isn't foreseen for laptop usage, considering that the r61 is a notebook, more to be used sited on desks, not on laps.

With the Wifi active on the U350, its bottom heat remains mainly concentrated around the Wifi modem. This is acceptable, this heat is more in the middle of the PC, not on a leg. I'm now trying to find a solution to spread the heat currently concentrated over the Wifi modem. But this has no effect on CPU heat or fan speed.

Playing around, I finally noticed that this always active fan may consume a fulll hour of power from the battery. I still didn't find a way to slow down the fan using softwares or time for more advanced solutions, like PWM. My current transistor patch to slow down the fan is wasting power. But with this ULV, I've an 8 cells battery and more than 9 hours live with activities, including Wifi.

Post Reply