Damping and the whoosh
Page 1 of 1

Author:  Puffi [ Sun Nov 04, 2007 8:43 am ]
Post subject:  Damping and the whoosh

I'm building a case out of wood with an air duct... So big, heavily damped and soundproofed that the mechanical sound of the fans is quite unlikely to be an issue. The sound of air moving (the whoosh) probably will. I'm talking about very high CFM fans here. So...

What's the best damping material for the duct and the case walls? I've heard good things about fiberglass and foam?

Would it be a good idea to use different damping materials on the duct and on the walls of the case, that's actually attached to a hard surface?

Would foam that has an uneven surface create turbulence inside the duct that'd create more noise? Is it better to opt for a completely flat surface there?

Will bottlenecking the sound at the end of its journey help reduce the sound of the whoosh? Does this mean the intake should be bigger in contrast?

What about putting something inside the duct to affect the sound of air moving?

If you've got any other relevant info, feel free to share.

Author:  Bluefront [ Sun Nov 04, 2007 9:44 am ]
Post subject: 

I think the "woosh" noise from a duct is dependent on how much CFM you are trying to blow through a particular diameter/size duct. Try to blow 70CFM through a 92mm duct, for instance, and it will make noise.

That being said....in the automotive industry, you'll find air ducts lined with dampening foam (open cell), somewhat like paxmate. You'll find this in higher-price vehicles. The cheap models just use a plastic duct.

On smaller ducts I leave them bare.....larger ducts get 4mm open-cell foam. This is something you'll need to try out. I've made neat-looking setups that amplified the "woosh"......


Author:  Puffi [ Sun Nov 04, 2007 3:59 pm ]
Post subject: 

Yes the CFM is exactly my concern here. I'll be putting three 120mm 104 CFM fans on top of each other...

What in your experience amplified the whoosh? Also, if something amplified it, then there must've been something that was better than the amplified version?

In my initial plan I was thinking of making the duct of about 20mm thick damping material, without any sound blocking, and putting 30mm damping on the walls of the case.

Author:  Puffi [ Fri Nov 09, 2007 3:29 pm ]
Post subject: 

Oh come on, I have to bump this?

There has to be a lot of people here who've built a duct, or dampened something... And I know there are a bunch of engineers and physics enthusiasts here, so step up! One of u has to know something relevant about aerodynamics at least...

Author:  MikeC [ Fri Nov 09, 2007 4:22 pm ]
Post subject: 

bluefront's comments about this being very dependent on details is dead on. The diameter and length of the duct/s themselves as well as the volume and dimensions of the box are part of the equation. So is the noise signature of each fan.

Yeah, you could work out scientifically the most likely problem frequencies and the most effective ways to damp them, but you'd need one hell of a sophisticated software. Trial and error is probably your most realistic approach.

The issue is not really the wall materials themselves, but the AIR in the box and in the duct. They both have resonances, and the noise of the fans WILL be accentuated by these resonances.

My general recommendation is to AVOID the ducts if at all possible. Tubes are acoustic resonators, plain and simple. Add some noise makers in them, and the noise gets accentuated. If you cant avoid them, try making them from open cell foam sheets. Don't add any restrictions in the ducts. Damping materials elsewhere will do very little to reduce the noise.

Author:  Puffi [ Sat Nov 10, 2007 8:57 am ]
Post subject: 

How much louder a whoosh will I get if I squeeze 104CFM through a 90mm hole, in stead of 120mm? Are there some other variables than the CFM and size of the hole that considerably affect the whoosh? If not, then there should be some kind of a simple equation for this!

That's assuming the hole is perfectly round. What if it isn't? Can you reduce the whoosh that way, or is round always the best?

Trial and error is always a good approach, but what surprises me is that no-one has tried this stuff before??

U were talking about open cell foam? So that's better than fiberglass in your opinion?

Author:  Bluefront [ Sat Nov 10, 2007 9:13 am ]
Post subject: 

Well...100cfm through either size hole will be noisy. I don't think the shape of the duct matters much......it's the total area.

Fiberglass shreds too easily, plus it is usually much thicker than 4mm, like the open-cell foam. That dampening foam won't solve the "too much CFM through too small a duct" problem......

Reconsider that high-speed fan you intend to use. There's not a whole lot you can do to quiet the thing...

Author:  jaganath [ Sat Nov 10, 2007 9:13 am ]
Post subject: 

How much louder a whoosh will I get if I squeeze 104CFM through a 90mm hole, in stead of 120mm?

I strongly suspect the smaller the hole, the louder and higher frequency of the noise, because the same amount of air has to travel faster to get through a smaller area in the same unit time. At least in airplanes the noise of the jet is very strongly correlated to the average velocity.

I'll be putting three 120mm 104 CFM fans on top of each other...

how do you mean, on top of each other? are you proposing to run 3 high-cfm fans in series? why do you need so much static pressure?

Will bottlenecking the sound at the end of its journey help reduce the sound of the whoosh?

as mentioned above, I think it is very likely to make it worse.

At the end of the day, you are trying to push a large amount of air through a small, enclosed space, this is very similar to how musical instruments work, you may want to look at ways you can reduce your airflow requirements.

Author:  djkest [ Sat Nov 10, 2007 10:17 am ]
Post subject: 

Holy moly, what do you need that much airflow for? Seriously, I'd like to know.

You could also check out the 120mmx38mm fans, I think scythe makes one that moves 100cfm for um... less noise than you might think.

Author:  Puffi [ Sat Nov 10, 2007 4:11 pm ]
Post subject: 

I need three fans for a Thermochill PA120.3 radiator. I'll probably get NMB-MAT 38mm fans. They're 104CFM each and show good promise. I just want to see how loud they are. If they're insane, I'll replace them or lower the voltage. I think they were rated 41,5db (quite accurately so from what I remember), and I'm quite confident they're the quietest fans at that CFM level.

So it's more like a test to see how much sound I can damp and block, and how much noise I can bear on a day to day basis. I'm not really that bothered by a gentle whoosh sound, and if I damp this thing really well the whoosh might be quite low-pitched, and not so disturbing...(?)

Obviously the smaller the hole the more noise u'll hear, BUT is there some sort of golden middle road here? I still have to fit the duct inside the case, so I dont have unlimited space.

And why I need all this airflow in the first place? Well, for overclocking a quad core (TDP 95W). And these new graphics cards push a lot of heat too...

Author:  CyberDog [ Sun Nov 11, 2007 8:34 am ]
Post subject: 

How big your tube/box for fans and radiator can be? Because the best way is to put the rad in canal which has sharp variations in sizes and foam on walls. Maybe even turns. And fans are not in contact with rad. Atleast 6cm away on it. And in middle of canal. Rad can be on any place, but best way is put it before fans. The longer and complex the duct/canal/box is the better result you'll have on cost of flow.

Author:  Puffi [ Sun Nov 11, 2007 9:56 am ]
Post subject: 

Well yeah it's not exactly a "tube" - I just used that to illustrate the problem... But my design as of right now will have 6 90-degree turns altogether, the rad being in the middle of it. I've read u get better performance for the rad if the fans are attached directly to it. Also that's just more simple to build... So why should I put it 6 cms away from it? I have never heard of this.

The size of the duct is about 9 x 40cms. The size of the whole box is supposed to be 60x60x60cms. The duct is completely separate of the actual computer, which has its own ducts.

Currently I have planned no variations in size of the duct. How big variations are u talking about? And is it worth it if it'll make the duct smaller than 9x40?

Author:  CyberDog [ Sun Nov 11, 2007 10:37 am ]
Post subject: 

Distance between fan and rad is based on that the radial fan needs about half to full diameter distance to "solid" wall to operate at full potential. And it also reduce turbulent noise.

Variations may be something like double the diameter as the channel. Like chambers. And remember to put lots of damping material.

Page 1 of 1 All times are UTC - 8 hours
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group