ABCs of Sound Dampening/blocking

[talcite]

heres a guide that didn't quite take as well at AT as my wiring one. At the suggestion of a member, i've decided to post it here. Its not quite complete, i've had alot of other things going on at the same time.

here it is:

------

Part 1 is http://forums.anandtech.com/messageview ... erthread=y if you're interested. It really has more to do with the basic stuff about noise reduction. Part 2 is more abstract and really more useful if you've already finished doing Part 1. I wrote this section because i've noticed there's almost zilch in SPCR or here or anywhere about sound absorbtion.

Acoustic Management Guide Part II: Acoustic Absorption

This section is going to come together slowly. Anyways, its all about absorbing sound and preventing it from reverberating. One thing should really be clarified, the goal is to absorb the sound, not to block or reflect it and prevent it from going through.

In the article, i will discuss different techniques for absorbing and also blocking sound, most of these techniques are borrowed from home building and are used on a large scale. I will also discuss different ratings systems and how they work. This article should take a few weeks to finish =p

Ok so here we go...

First, sound absorbing material is rated and classified in many different standards, but the most important to us is the NRC and the STC.

NRC

The NRC or the noise reduction coefficient is a value between 0 and 1. 0 being perfect acoustic reflection, and 1 being perfect acoustic absorption. Its is the average of 4 values, the coefficients at 250Hz, 500Hz, 1000Hz, and 2000Hz. While these 4 frequencies span most of the noise we normally hear, there are a few exceptions. Human hearing spans from 20Hz to 20kHz. Some comprehensive sound absorbing foam will actually have ratings for up to 4000Hz. I have also seen NRC's of over 1, but how that's possible is totally beyond me.

Most computer noise is high pitched. You rarely hear bass coming from a HSF =p. Therefore, you should find materials with good NRCs around the higher pitched ranges.

STC

STC or sound transmission class is generally used in house building, it is a measurement of how well a wall or celing absorbs sound. STC is based on sound loss in dBs at 16 frequencies between 125 and 4000 Hz. The values are then fitted to a transmission loss curve.

STC is measured in dBs and are harder to use in a computer's context because computers rarely generate the same frequencies and intensity of sound generated by a family of 4 in a townhouse. However, they may be used as a qualitative measurement, as to which methods of soundproofing are more effective. This is because acoustic absorption techniques are rarely measured in NRC, but instead in STC.


Alright... lets move on to techniques for reducing sound reflection.

It'd be ideal if you had a sound meter for measuring frequency and intensity, but since most of us don't have professional grade equipment that costs several thousands... well we just try to estimate the sound range.

Foam
Foam is generally a very popular way to reduce sound reflection. They are measured with NRC, and can range in price from several dollars to several hundred. Foam normally comes in large quantities, but it is possible to buy from a reseller for a smaller roll. Most cases will use foam as a preliminary acoustic absorber. Foam can come in many different shapes and sizes. Different sizes and shapes are designed for absorbing different frequencies of sound. The egg carton shape is very popular, as is the wedge. They both have their own respective frequency specialties.

The shape has a huge effect on absorption, but material also has a very large effect aswell. Open-celled foam is almost ALWAYS required for sound absorption. Generally, the more flexible the foam, the more it can absorb. I am not sure about the actual density of the foam, as if you move to really loose foam, it will leak sound, while really dense foam may reflect too much of it. this is where NRC ratings come in handy.

mass-loaded vinyl

Let me start of by saying mass-loaded vinyl is NOT the same as vinyl flooring. MLV is doped up with large amounts of heavy metals such as barrium (health hazard? 0.o) The principal behind MLV, as it is with all sound absorption, is to convert sound energy and vibrations into heat and then disperse. MLV's effectiveness is questionable, but should work nicely as a backing for foam (more on that later.)

fiberglass

Fiberglass is ITCHY. It will get into your skin and make everything very itchy. No joke, you must have professional experience with fiberglass before you try using it. Ontop of that, you must trap the fiberglass fibers in some kind of epoxy resin, which coincidentally releases organic fumes that will probably bond your contact lenses to your corneas or something equally traumatizing. The effectiveness of fiberglass is questionable, although it is employed in some high end speakers as insulation. I really know not much about fiberglass. I don't suggest working with it unless you're really adventurous. Use at own risk!! A safer alternative would be to use something with very fine fibers. Sort of like what the mutemat does, just you can make your own or something.

complicated pathways

Lets say you want to make a maze for your fan airflow, it'll probably completely kill the sound, provided you make the proper angled turns and line the edges with foam. Even just slightly angleing two walls together can create a dramatic effect in reducing sound. This is really more of an experimental kinda thing.

That's about all i can think of interms of sound absorption, now onto the sound blocking.

vacuum

Sound needs a medium to travel in. Remove medium, remove sound. 100% effective if you can pull off an absolute vacuum 0.o Not sure what effect a partial vacuum would have. I assume its a better way to stop sound waves?

Changing densities

Basically sound waves lose energy as they travel through different types of material. If you manage to sandwich many different density materials together, then sound will be diminished.
One way is actually to put sound absorpative material ontop of sound reflective material. Theoretically, it should force the sound waves to travel through the material twice, hence more absorption. I'm not sure about this though, never seen it really implemented anywhere.

Sound reflective material

Certain materials with very low NRC ratings will reflect sound waves. That's basically the point of a double paned window in a sound studio. Glass is a horrible medium to absorb sound with. It's really a better bet to reflect it back and let the other material handle it instead.

active noise cancellation

I've heard alot of people talk about this, none of them try it. Personally i'm very skeptical about the entire thing, plus not very enthusiastic about the piggy-bank breaking cost. If any of you brave souls out there wanna give it a try, document it for us.

I think that's all about sound blocking techniques.

Many building techniques use a combination of multiple walls, soft mounted walls, staggered walls, and insulation. look it up, some of them look promising but impossibly hard to do.

Theres more that i thought of on the 12 hour bus rides, but i can't remember any of it right now. I'll add if it comes to me.

In the next section i'll be actually much more step-oriented, and it'll really be more of a "guide". I'm going to construct my optical drive cage. This should be interesting. It should also take a long time since my exams are coming in a month heh. If it gets a bit too technical, give me a shout and i'll try to explain better. Unfortunately, i might start to use more complicated tools, since i have 2 machineshops at my disposal. Don't worry, i'll suggest suitable alternate tools for you guys.

Before i forget! i also need a suggestion for a cheap way to measure the dBs my computer is making, and also if possible the frequency range. That way i'll know which foam to purchase. Thanks guys.

----

ok then... i've actually drawn out the optical case silencer in ProDesktop already. CNC is waiting to be programmed and cut. I'll put a picture up of the renders. It will be build with 3/8" art rubber, like Leo's HDD case. There will be a metal bracket, which isn't drawn, securing the Optical drive, since they don't like vibration. The CNC cut version is going to be done in plastic, and pieces used as a template. I need a sound meter pretty badly. If i'm going to make it a guide format, i need some quantitative measurements, meaning real dB readings. I'm actually not even out of high school yet, so money's a huge issue. I'm already spending way too much for this case. its only a hobby 0.o I still need to line my case walls with foam and mlv, build a double layered soundproof plexiglass window, AND design + construct a waterblock for my new system using CNC and fluid mechanic software. But... one step at a time. Theres also going to be acoustic foam inserts in the two little gaps for the cables in the optical case.

[talcite]

Heres the jpeg captures of the assembly. Theres another sound blocking insert intended. i removed it to show the ribs. Everything's gotta be glued with exception to the purple lid. That's where the optical drive opens. When it closes, it should more or less seal. Acoustic foam intended to go between the two inserts and the back wall. Metal brackets for securing drive not shown. Could be added in on request, but it'd take awhile. i have to measure my drive holes then. The drawing is to scale. based on the largest optical drive i could find, a Lg 16x dvd rom. Feedback/comments greatly appreciated.






ProDesktop files avail on request.

[efcoins]

Making a sound proof enclosure is not hard to do. What is hard is making a soundproof enclosure which lets out the heat. It looks to me that anything you put into your enclosure will quickly overheat. Tell us how the cooling works with your design.

[talcite]

the drive will be mounted with 2 wide strips of metal, which will hopefully transfer the heat. I plan to use steel, but i can also use aluminum or copper alloys. thermal tape is also an option, but i'll resort to that if theres no other choice.

[jaganath]

I have also seen NRC's of over 1, but how that's possible is totally beyond me.

An explanation of NRC's >1.

Measuring Sound Absorption

Basically it's a flaw in the measuring methodology. No material can absorb more than 100% of the sound striking its surface.

[Lupo]

Hello!

First post on this forum. A big thanks to all for making this magnificent site!

A little background info: Am running a smallish sound studio where the main selling point is great monitoring and acoustics. There's literally hundreds of kilos of sound absorption in the room! Acoustics seems to be an area that is mostly unknown to the computer crowd. Big thanks to Talcite for bringing up the subject!

Some general thoughts regarding acoustics and absorption: the foam and fluffy fibre glass/mineral wool works by friction on the passing sound waves. This is dependent on the velocity of the waves. Whenever a wave approach a hard surface the velocity slows down and the pressure rises. At the surface itself, the velocity is zero and the pressure is at maximum. Once the wave bounces off the surface it starts to accelerate again. Play some music and sneak along the walls and corners to confirm this - the pressure is much louder there than in the middle of the room.

As most absorbents are really thin and needs velocity to do their job, they have to be placed some distance from the wall to be most effective. The general rule is to space it out from the wall with the same distance as the thickness of the material. So for one inch of damping, once inch of air gap behind it will improve things quite dramatically. If you can fill the gap completely, so much the better, but this is often not economically practical.

Now the stuff that's sold for computers are generally not even an inch thick - more like 1/8 inch. To make matters worse, this is glued directly to the wall. The effectiveness can't be very high. Audio waves are huge things. At 20hz, the wavelength is 17 meters. At 200 hz - 1.7 meters, at 2kHz - 17 centimeters and at 20kHz, the upper limit, it's still 1.7 centimeters. To work as well as possible, the absorbent needs to be of the same size as 1/4 of the wavelength. This means that a 1/8" thick absorbent will have some effect at 20kHz and less below that. To be effective down to about 500Hz, where the airborne noise from the computer starts, the absorbent needs to be about 6 inch thick.. Less thickness will still do something - but not as good.

Most of the sound sheets sold for computer use is of the heavy type that damp vibrations in the case. This might be very well in a flimsy case, but I guess most people now use the P180 or similar cases with great damping built into the system. What these systems may need is not vibration damping - but acoustic damping of reflections. The sound escaping through the fans is inevitable, but this can be reduced a bit by applying generous amounts of THICK absorption sheets inside the computer. The thicker, the better. Sheets with mass barrier damping and a tiny bit of foam on top is probably not the best choice. Go for foam all the way. The echoes that bounces off the bare walls inside the case will be dampened, which will help a bit. Not much, just a little bit, but that's what all the SPCR tweaks are about, right?

Applying everything to one wall is less effective than to spread it out over several surfaces. Cramming it in wherever possible will have better effect than to apply all to, say, the side door.

I personally use 2" thick foam made from a reputable studio acoustic supplier. It's works fairly well, at least a lot better than the stuff sold for computer use, and is very easy to handle. The 'easy to handle' bit is the only reason I've not used high density mineral wool - which is vastly more effective. One inch sheets should be easy to fit in most computers.

This is all nice but really doesn't make a startling difference. What really helps though is dampen the reflections around the computer case itself. Some absorption placed behind the computer, on the underside of the desk if it's below one of those and generally everywhere the sound bounces before it reaches your ears will help tremendously. Damping these reflections and blocking any direct airways between the 'puter and yourself is highly effective. The foam type damping is still easier to handle, but micro fibers are so much more effective that it's easily worth the bother. More on this below!

Foam
.. The shape has a huge effect on absorption, but material also has a very large effect as well. Open-celled foam is almost ALWAYS required for sound absorption. Generally, the more flexible the foam, the more it can absorb. I am not sure about the actual density of the foam, as if you move to really loose foam, it will leak sound, while really dense foam may reflect too much of it. this is where NRC ratings come in handy.

The point about open cell foam is very important. This is what makes it work. Normal foam have closed cell structures, like the cheap bed mattresses. These have near zero acoustical effect. The open cell types have a zillion tiny little open ducts that lets the sound waves propagate through the material. Every time the audio waves hit the walls of the open cells some energy is lost through friction.

Foam have a bad reputation in studio circles. Most of the products that are sold have very little effect. If you want the real stuff, go for the big names like Auralex. Most foam damping only affect a small frequency range and it does so with far less effect than what could be have with proper foam or other means of absorption.

fiberglass
Fiberglass is ITCHY. It will get into your skin and make everything very itchy. No joke, you must have professional experience with fiberglass before you try using it. Ontop of that, you must trap the fiberglass fibers in some kind of epoxy resin, which coincidentally releases organic fumes that will probably bond your contact lenses to your corneas or something equally traumatizing. The effectiveness of fiberglass is questionable, although it is employed in some high end speakers as insulation. I really know not much about fiberglass. I don't suggest working with it unless you're really adventurous. Use at own risk!! A safer alternative would be to use something with very fine fibers. Sort of like what the mutemat does, just you can make your own or something.

Fiberglass IS itchy. To handle it, suit up in a raincoat and don the breath mask and safety glasses. The effectiveness is NOT questionable. It's both the cheapest and most effective way of absorption. The foam stuff is easier to handle, but you wont find that in any real sound studio. Mineral wool or fiberglass all the way. The absorption of foam is so little compared to fiberglass/mineral wool that it's practically unusable for the purpose. The Owens Corning 703 is the most popular brand in the states. This have twice the density of the normal fluffy fiberglass. IIRC, the density is 4.5 pounds per square foot, which is perfect for sound absorption. 705 is even denser, but this can be too much, making the high frequencies bounce off the material if the waves does not hit perpendicular to the surface.

Mineral wool is much easier to work with and not as itchy, most construction workers don't bother to even put on gloves when handling it. Here in Norway, the leading supplier is Rockwool, who have a product called 'acoustic sheets' that works very well.

In either case, the stuff needs to be wrapped in something to avoid getting micro fibres everywhere. Burlap works well, preferably with a flame retardant treating. This is no biggie, just wrap it up and use it! Am not quite sure how much use it is inside a computer though, as the minimum thickness is one inch, with the wrapping, it might be a tight fit inside any computer case except for the door and free drive bays. As for absorption around the computer case - this is the stuff to use! 1 inch sheets works but two inches are better at absorbing frequencies from about 500 hz and up, the range we're concerned with here. The Owens Corning and Rockwool sheets can be supplied with paper backing that keeps the front/back surfaces sealed, leaving only the sides to be dealt with. A bit of diluted paint seals this up nicely, or some duct tape, or a bit of cloth - whatever works.

Placing this absorbent behind the 'puter is the most important step, to block reflections from the main offender - the rear outlets. Cramming in as much as possible in the general area, without blocking airflow, helps a lot.

My puter is hidden behind drapes of thick clothing and sheets of rockwool. The only path the sound can take is by bouncing off this stuff several times before reaching my ears. This have been good enough in itself to help me live with a regular noisy machine for years. The faint whirr and racket eventually got me at last. Am currently building a more silent system based on the Antec P180, which is why I ended up here at SPCR! =) Along with the extensive treatment of the general area the computer resides, I hope this will good enough to drown in the aircon noise.


The energy in sound waves is, btw, diminutive. Think I saw a concern about this mentioned somewhere. At the threshold of pain, 120dB, the pressure is 1 watt on one square meter. At the levels inside a PC case, it's at least hundreds of thousand times less power. Whatever heat is generated through the friction of the absorbents is totally negligible.

Hope this might help someone!


Cheers,

Andreas Nordenstam


PS: The reference book on the subject is Everest's Master Handbook of Acoustics. Much recommended!
PPS: about mineral wool/fibreglass and health: http://recording.org/ftopict-21013.html

[Bluefront]

Good as it is at sound dampening, I won't touch fiber-glass for computer use. Terrible stuff....breaks up easily, and airflow scatters it around. Not worth it to me. It even has health concerns when breathed in.

Thick dampening is unrealistic to most computer users.....about 1" thick is the best you can do in most cases. And covering all possible hard surfaces inside the case is your best bet.

Start off with quiet components, then add case dampening. This still will give you the quietest computer.

[highlandsun]

Interesting. As an almost complete non-sequitur, I recently had to replace a dead bathroom ventilation fan in my townhouse. Rather, I ordered a new fan motor, the bearing was shot on the old one and the thing was making all kinds of racket when it spun up. I noticed that there were pieces of tar/paper (basically like Dynamat) under each of the fan housing's three mounting screw locations. A couple of those pieces dropped and were lost, and while the new fan motor is nice and quiet, I now hear a lot of sound resonating across the whole bathroom ceiling. Gonna have to go find a sheet of Dynamat and remount the thing. Anyway, the reason for this post is because seeing that stuff reminded me of the things I've read in this forum. (As to why the Dynamat bits got lost - the fan died a few months ago, I pulled it down, and then a lot of time passed before I found a reasonably priced replacement motor, and time passed before I decided to take the time to remount it.)

[BrianE]

Now the stuff that's sold for computers are generally not even an inch thick - more like 1/8 inch. To make matters worse, this is glued directly to the wall. The effectiveness can't be very high. Audio waves are huge things. At 20hz, the wavelength is 17 meters. At 200 hz - 1.7 meters, at 2kHz - 17 centimeters and at 20kHz, the upper limit, it's still 1.7 centimeters. To work as well as possible, the absorbent needs to be of the same size as 1/4 of the wavelength. This means that a 1/8" thick absorbent will have some effect at 20kHz and less below that. To be effective down to about 500Hz, where the airborne noise from the computer starts, the absorbent needs to be about 6 inch thick.. Less thickness will still do something - but not as good.

Hello and welcome to SPCR.

I remember reading about this concept some time last year, thanks for the (huge ) write-up! Very sobering to think that most foam sheets placed inside computer cases aren't thick (or dense) enough to absorb much audible sound... unless it happens to hit the foam at oblique angles or something.

Sometimes I wonder if it would be beneficial to have a large block (or two) of dense, open-cell foam sitting somewhere in a computer case where it won't obstruct airflow. It would be there just to "catch" the reflected sound waves bouncing around the inside of the case. Haven't even come close to testing this idea out though...

[Lupo]

Hi!

Hello and welcome to SPCR. :)

Thanks!

Sometimes I wonder if it would be beneficial to have a large block (or two) of dense, open-cell foam sitting somewhere in a computer case where it won't obstruct airflow. It would be there just to "catch" the reflected sound waves bouncing around the inside of the case. Haven't even come close to testing this idea out though...

It will probably have some effect, but not that much, unless you try to pack the foam into every available crevice and surface.

What you hear is what escapes the case. Most of this comes directly from the components straight out the rear hole. The reflections that bounce around in there does add to the noise output though. If some of the reflections are dampened, the noise output will be less. Having a block of foam, say, in one 5,25" bay, will probably have little effect. Stuffing all available drive bays will do more - as the sound from the CPU and exhaust fan that bounce on the front wall and back out again wil be dampened. But really, getting as much as possible inside there is the best solution.

As seen in another thread, my case is jam packed with acoustic damping foam. It sure made quite a difference!

Andreas

[dano]

Lupo, that is a very great rightup, I learned a lot about some of the more obscure (for me ) types of sound absorbtion. In just a few minutes of looking around the internet for mineral or rock wool (they are the same thing right?), i have found it comes in a number of different densities. www.mcmaster.com - has it at 8lb/ft^3 or about 128 kg/m^3. What would you suggest as a good density for the frequencies found in the typical computer?

[Lupo]

Hi!

45 kg/m^3 / 3 pound cubic feet is the typical recommended density in the states, since rigid fiberglass/mineral fiber/rockwool comes in this weight. This is the owens corning 703 type. Over here, the closest match is rockwool at 60 kg/m^3. At 6 pound cubic feet things start to get too dense, potentially reflecting the sound waves instead of absorbing them.

link

>mineral or rock wool (they are the same thing right?)

Yes! Rockwool is mineral fiber. Much easier to handle than glass fiber.


Have to add one thing to the above long post; some of the stuff are very theoretical. In reality, most waves will not hit perpendicular to the surface of the absorber. The angled waves will have higher absorption and in general most absorbers show higher effeciency than the theoretical predictions.

[cmthomson]

Have to add one thing to the above long post; some of the stuff are very theoretical. In reality, most waves will not hit perpendicular to the surface of the absorber. The angled waves will have higher absorption and in general most absorbers show higher effeciency than the theoretical predictions.

Which of course is why anechoic chambers are built with zillions of tetrahedrons.

[Bobendren]

Wow, that was uber informative.

I have two questions:

1) Is there an optimum shape for foam? and

2) Would leaving a gap of air between two pieces of arcylic reduce noise? (talking about a case window here)

[Gorsnak]

Fiberglass IS itchy. To handle it, suit up in a raincoat and don the breath mask and safety glasses. The effectiveness is NOT questionable. It's both the cheapest and most effective way of absorption.

I agree pretty much 100% with your excellent post, but I think you're overstating the requirements for handling fibreglass. A raincoat? That will just make you uncomfortable. Unless you're using power saws or something similar to cut the stuff, the fibreglass dust doesn't really get into every nook and cranny, and apart from when cutting it, I find simple long sleeves and gloves to be quite adequate protection. A dust mask is mandatory when cutting it with power tools, and goggles (not just glasses) are a good idea if not absolutely necessary. When cutting with a knife, I don't even bother with a mask. The first time I ever made acoustic absorbing panels I had a hell of a time - horribly, horribly itchy. Since then I've refined my technique in a few ways designed to limit and contain the creation of fibreglass dust. I just got back from a job fabricating and installing ~1000 sq ft of 2" panels in a large band practice room on an air force base. Not once during the whole week did I experience any significant discomfort from the fibreglass, and I used nothing more than leather gloves and common sense. Of course, having had a fair bit of practice certainly does help.

[capoeira]

what about drywall?

[capoeira]

what about drywall?

come on....no coments on this?

as mentioned above, audio professionals don't use foam as most of them have little effect and mostly on higher frequencies (and yea, you got low frequencies from the HDs)
glasswool and mineralwool are dificult to handle and sensible to heat.
drywall on the other hand is easy to work with

[ces]

what about drywall?

come on....no coments on this?

as mentioned above, audio professionals don't use foam as most of them have little effect and mostly on higher frequencies (and yea, you got low frequencies from the HDs)
glasswool and mineralwool are dificult to handle and sensible to heat.
drywall on the other hand is easy to work with

I would expect that the foam works best at high frequencies and the drywall works best at low frequencies (because of its weight).

Glasswool and mineralwool are not sensitive to heat, but I don't think they are very good for sound absorption.

Take a look at:
Foam And Sound Absorbing Materials
viewtopic.php?p=539399#p539399

[capoeira]

what about drywall?

come on....no coments on this?

as mentioned above, audio professionals don't use foam as most of them have little effect and mostly on higher frequencies (and yea, you got low frequencies from the HDs)
glasswool and mineralwool are dificult to handle and sensible to heat.
drywall on the other hand is easy to work with

I would expect that the foam works best at high frequencies and the drywall works best at low frequencies (because of its weight).

Glasswool and mineralwool are not sensitive to heat, but I don't think they are very good for sound absorption.

Take a look at:
Foam And Sound Absorbing Materials
viewtopic.php?p=539399#p539399

difficult to find a good chart for absorption, but glass and mineralwool are indeed good absorbers (couldn't find a list with drywall included):
http://www.sengpielaudio.com/calculator-RT60Coeff.htm
http://www.sae.edu/reference_material/p ... 0Chart.htm
http://www.bobgolds.com/AbsorptionCoefficients.htm

I will research more on this

[dhanson865]

what about drywall?

come on....no comments on this?

Did you realize that you replied to a thread from 5 years ago? It takes time for people still here to realize if it's even worth bothering to reply to a thread this old.

If they do it'll be over days, weeks or months not hours or minutes.

[capoeira]

what about drywall?

come on....no comments on this?

Did you realize that you replied to a thread from 5 years ago? It takes time for people still here to realize if it's even worth bothering to reply to a thread this old.

If they do it'll be over days, weeks or months not hours or minutes.

ok, I use to use faster forums, now I know. excuse me.

I found a list with over 450 materials listed: http://www.wsdg.com/dynamic.asp?id=reso ... absorbtion

so, acording to this list, drywall is a sound blocker.

[ces]

ok, I use to use faster forums,

Don't worry about it. A lot of threads here will lie quiescent for years then pick up again.... just like this one. I think that is a good thing.