Reducing fan noise? Heatsink/pipes?

Toggle sidebar Toggle sidebar
Mephistopheles

Mephistopheles

Distinguished
Feb 10, 2003
2,444
0
19,780
I currently have (as some of you might already know... sorry :wink: ) a dual P3 933Mhz...

I'm considering giving it to my brother in exchange of a small help in my new purchase (a new rig!) soon.

However, he has complained that the damned thing is too loud, and I have to agree with him. (but I couldn't be bothered when it was my computer... I don't care enough about the noise) He's a musician, which means he'll probably and eventually record stuff on this computer. And it doesn't help if there are two 933Mhz p3 stock fans and a GF4 Ti4200 fan in the background!

Is there anything I can do? Something cheap? I was considering letting go of one CPU for some spare cash (I wonder if I can sell it? for a few bucks?) and put a CPU terminator; that could decrease overall noise.

Might be a dumb question, but it never hurts to ask... Is there any other way? At all? A cheap one? I mean, each processor is only responsible for 27W or so...

<i>Edit: This kind of lets me frustrated... I mean, a highly sophisticated quantum physics-based apparatus cooled by... an aluminum or copper block with a fan? Isn't that rudimentary? There must be another cost-effective way. Isn't there some sort of system that can remove heat more efficiently without the need for water? Something better... How low must power dissipation be in order for passive cooling to be possible? I wonder...</i>

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 02:02 AM.</EM></FONT></P>
 
Resistors? You mean, thereby reducing rpm?...

I was currently reading on how heat pipes work. They're quite impressive; I wonder why heat pipes aren't more widely used in CPU cooling. Are these more expensive?

I was under the impression that, having access to a materials workshop that handles copper, I <i>might</i> be able to fashion an efficient passive cooling solution myself, <i>without</i> the much added cost of importing something. Of course, it never hurts to have ambitions... :smile:

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 01:17 AM.</EM></FONT></P>
 
I found these highly interesting readings... on heatsinks!

First... An <A HREF="http://faq.arstechnica.com/link.php?i=1194" target="_new">introduction</A> to heat dissipation and the physics involved (quite introductory indeed, but informative nonetheless)...

Second... a <A HREF="http://faq.arstechnica.com/link.php?i=1192" target="_new">discussion</A> on using diamond, the best known thermal conductor.

This reminds me of one of my laboratory classes... We deposited a thin aluminum film on a piece of paper. Looked metallic and nice. We also deposited aluminum on a small piece of glass used on microscopes (damned, I forgot the name for that)... And it looked like a perfect mirror. This is done in vacuum. Two metal pieces can also be joined perfectly in vacuum, with optimal surface contact!

And before you think that diamond heatsinks is a stupid idea, consider instead that nowadays it is possible to use these in-vacuum thin film techniques to deposit a small diamond-like carbon (DLC) layer on top of any surface. This layer would, of course, have the thermal conductivity of pure diamond, which is absurdly greater than, say, copper.

So, if you have a copper heatsink, you can coat it with DLC, as thick as you can, and it'll probably conduct better...

Of course, I don't know that much about cooling... just researching now...

One more thing: It seems heat pipes are truly stellar conductors. Check <A HREF="http://www.deltatronic.de/bilder/pipe_cpu_klein.jpg" target="_new">this out</A>!

<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 02:10 AM.</EM></FONT></P>
 
Oh my, just look at that!

What's that? A plane? Superman? No, it's a <A HREF="http://www.aavidthermalloy.com/products/microp/037704.shtml" target="_new">passive heatsink for a <b>Nocona Xeon!</b></A> Up to 135W processors. Wow.

Gotta hand it to these guys. They even have <A HREF="http://www.aavidthermalloy.com/products/microp/037863_2u.shtml" target="_new">multiple passive cooling solutions</A>...


<i><font color=red>You never change the existing reality by fighting it. Instead, create a new model that makes the old one obsolete</font color=red> - Buckminster Fuller </i>
 
yeah..reduce rpm
I've found out that most of the fans are rotating too fast, therefore making too much noises, the fans don't necessary have to rotate at such high speed

I bought the TT's Silent Boost, the fan is supposed to be 2500rpm, which is quite silent compare to some generic fan, but still too loud for me, so I add a resistor to it, the fan is become extremely silent, run at about 1900rpm, and I can hardly notice any temperature raising!

I've found out that for 80mm fans, the effective and silent speed would be about 1800rpm, for a 40mm, it would be 4000rpm (my 8500 fan), a 60mm fan would be about 3000rpm (my other pc Duron1GHz's fan)
and another thing is that, at these lower rpm, the fans won't get dust that easy, i think that's a + :lol:

The Heatpipe is quite effective, my sister's miniPC has it inside with a Pentium4 2.4CGHz, the max temperature I've seen is about 44C, usually is around 37C, the fan is at 2000rpm..

u still remember ur lab? I forgot all of them...
diamond? i dun think you'll have money for that? btw, isn't diamond = Carbon? but carbon isn't the best thermal conductor? all i know from diamond is that it's the most rigid element we could find
 
The resistors are a good ideal if you have standard case fans, this could greatly reduce fan RPMs and noise. If you have some thermal fans normally cutting off the sensor will make them run at low speed all the time.

<A HREF="http://www.coolerguys.com/840556014034.html" target="_new">http://www.coolerguys.com/840556014034.html</A>

Those fans running low speed only produce 19 db, which inside the case is pretty much silent.

As far as heat-pipes, they are getting fairly common in CPU heatsinks, generally starting about $30 heatsinks have heat pipes built in.

You might be able to fab up some sort of passive cooling, but is probably more trouble than it is worth, I believe there are a couple pre-made solutions for this.

Personally I think you should thinkg about swapping fans out for quieter ones, then maybe look at noise dampening material for the case.

Noise Dampening Pads:
<A HREF="http://www.coolerguys.com/840556014003.html" target="_new">http://www.coolerguys.com/840556014003.html</A>

My Desktop: <A HREF="http://Mr5oh.tripod.com/pc.html" target="_new">http://Mr5oh.tripod.com/pc.html</A>
Overclocking Results: <A HREF="http://Mr5oh.tripod.com/pc2.html" target="_new">http://Mr5oh.tripod.com/pc2.html</A>
 
One thing I found out when searching the net is that it's not impossible at all to <A HREF="http://www.thermaltake.com/coolers/cl-p0019Fanless103/cl-p0019fanless103.htm" target="_new">cool a prescott</A> passively, with absolutely no noise. It might indeed be a bit cumbersome (like in <A HREF="http://www.thermaltake.com/images/shared/enlargedView.gif" target="_new">this image!</A>), but it's not impossible and that's not bad at all!

I agree, trying to get a passive solution (homemade - or, to be more precise, made with the tools I could get my hands on in the physics department - including workshops, vacuum, etc, etc) to work would be hard... But interesting nonetheless. Oh well...

As for quieter fans, or different fans, I don't really know an easy place for getting those, sadly. Oh boy, buying computers here in this country is really disappointing... People don't know jack about computers. Even the ones who sell them are morons.

<i>Edit: wow, my two CPUs have two fans at 4100rpm and <b>5300</b>rpm, respectively. That's why it's so damned frickin' loud here.</i><P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 08/11/04 02:47 AM.</EM></FONT></P>
 
I use a zalman "fan mate 1", and am running my p4 heatsink fan at 2200 idle. It increases a little as the heat rises, but not much. They cost about $5, and come standard with some zalman heatsinks. Also check into the fanless power supplies. Coolmax and thermaltake make them.
 
> No, it's a passive heatsink for a Nocona Xeon! Up to 135W
>processors. Wow.

Ahem,.. no they are not. First they are not for Nocona since they are for socket 423, secondly, they are to be used in 1U and 2U server chassis that use ducting. rackmount servers almost never have a fan right on the cpu, but they tunnel the airflow from front to back, over the cpu (heatsink) instead. Much less turbulence, better airflow, but still very much using powerfull and noisy fans. There is no way you can cool 135W cpu('s) passively in a 1 or 2U rack.

= The views stated herein are my personal views, and not necessarily the views of my wife. =
 
Just grab a couple potentiometers and series them up with your current fans. That little bit of control is always nice. If you dont have good enough heat transfer, add a layer of copper to the bottum of the heatsink.
 
To answer your real question: you could try voltmodding your current CPU fans (give them 7v) if your temps are low enough to allow this or replace the fans with silent one's from Papst. That is, if it are indeed the cpu fans that cause the noise, could be PSU as well. I doubt the GF4 would make a lot of noise (mine is virtually silent), but if it does, again, buy a better (more silent) fan for it.

For the 7v volt mod, found this on google:"You can use the +12 "Yellow" wire as the "+", and use the +5V "Red" wire as the "-", and end up with 7Volts. The Power Supply needs to be able to counteract this neglegable current being dumped into the 5V rail, but a few fans are peanuts compaerd to a 85Watt processor and some power hungry RAM!"

= The views stated herein are my personal views, and not necessarily the views of my wife. =
 
> No, it's a passive heatsink for a Nocona Xeon! Up to 135W
>processors. Wow.

Ahem,.. no they are not.
Click to expand...
Right, you were right... <A HREF="http://www.thermaltake.com/coolers/cl-p0019Fanless103/cl-p0019fanless103.htm" target="_new">This heatsink</A>, however, is adequate for at least up to 3.6Ghz prescott, which should have the same heat output as a xeon, being identical. (sadly, it's obviously still impossible to put <i>that</i> behemoth in a 1U or 2U server chassis, you're right!...)
 
Yeah, but "passive " is a relative term here. It doesn't use its own fan, but it just uses the case fan. if that is passive, than my old Dell Dimension was passively cooled as well, it used a simple plastic duct over the cpu to the case fan, so one fan acted as both cpu and case fan. Simple, effective, cheap, (why doesn't anyone else do this ??) but hardly "passive". If you truly want a noiseless passive cooling, it takes something like this: <A HREF="http://www.zalman.co.kr/eng/product/view.asp?idx=64&code=020" target="_new">http://www.zalman.co.kr/eng/product/view.asp?idx=64&code=020</A> not so cheap though..

= The views stated herein are my personal views, and not necessarily the views of my wife. =
 
I'd install a couple big, bad, Athlon coolers, then I'd put some low speed fans on them.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 
<i>I'd</i> rivot a nice big external radiator to the top of the case and run a cheap-to-medium water cooling system to it. Then I'd see if I needed to add one or two low-RPM 120mm fans to the radiator or not, though probably not. Or maybe I'd just add two 120mm remote-sensored temperature-controlled fans with their sensors placed in a CD bay reservoir and be done with it. :O

Necessity may be the <i>mother</i> of invention, but DNA testing has proved without a doubt that insanity is the father. 😉

<pre><b><font color=red>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</font color=red></b></pre><p>
 
Ah, but water coolers take up so much space, it would be far easier to use huge heatsinks with slow fans.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 
Ah, but water coolers take up so much space, it would be far easier to use huge heatsinks with slow fans.
Click to expand...
I certainly won't argue the far easier. :) Even a simple watercooling setup is still a pain in the arse compared to a simple heatsink.

However I do have to contest the space comment. Mounting the radiator on the top of the case and running all of the tubing/pump/reservoir inside of the case (to the point of even running the pump's power off of a molex) is hardly any worry. Computer cases have so much unused internal space that it'd be right to finally put it to good use, and if it's all inside (and on top) then there's no notable external space difference.

But yes, air cooling is still by far easier.

...Well, so long as you don't have a really bad case anyway. When you have to drill holes and cut front panel plastic to insert just a single intake fan (as I did for my poor old Celeron microATX box) then air cooling can be a pain too.

<pre><b><font color=red>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</font color=red></b></pre><p>
 
Why not just wire the two CPU fans in serial, off one single Mobo header? Wouldn't require much time or effort, and would have the same effect as wiring resistors in to each separately. You could still leave just the (usually yellow) RPM sensor wire plugged into the header which no longer supplies power to a fan, and so you'd still be able to monitor fan speeds separately.

---
Epox 8RDA+ V1.1 w/ Custom NB HS
XP1700+ @200x10 (~2Ghz), 1.4 Vcore
2x256Mb Corsair PC3200LL/1x512Mb Corsair XMS PC4000 2.5-3-3-7
Sapphire 9800Pro @412/740
 
I've had sever problems finding space for the pump and reservior. Also, most radiators use a 120mm fan which precludes many cases.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 
For those of us that dont care about noise, How ahout a huge sink and a super fast fan :) Cools better no?

"If youre paddling upstream in a canoe and a
wheel falls off, how many pancakes fit in a doghouse? None! Icecream doesn't have bones!!!"
 
OH! How about pulling the radiator and electric fan out of a Geo Metro and replacing your case window with it!

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom