To those who have "sound proof'd" their cars...

[StereoInstaller1]

Luke, do you happen to have the spectrograms from your measurements?

No, it was in Car Audio and Electronics magazine, not me.

I was hoping YOU had done some analysis since you kick ass on such things!

BTW, I wanna ship you my 3 Harman/Kardon CA260s, see what you wanna do. I am finally ready!

[nmlss2006]

I am not so certain that I agree with you on the 80/20 rule in this case: it really depends on what you're trying to accomplish and what angle you choose to accomplish it by. The least questionable as far as I am concerned is separating the air inside from the vibrations in the body (with the possible side benefit of adding mass to the flat panels of the body to dampen said vibrations). If that is the case, then any gap will be very detrimental. The experimental evidence I've seen - which granted, is qualitative, not quantitative - seems to support this theory as valid in practice.

[StereoInstaller1]

Actually, I believe your example perfectly addresses the "80/20" rule. You just hit the 100% mark, as exemplified by the fact that you have 220+lbs of the stuff.

220 lbs of cotton would do the exact same thing. Mass is mass, right? All it does is shift the resonance lower in frequency, which is why that much still only did 3db.

It slows resonances in metal, period. In no way does it "separate the air vibrations", that is the side benefit. Read through the silly marketing BS, look at the physics.

[bmwman91]

No, it was in Car Audio and Electronics magazine, not me.

I was hoping YOU had done some analysis since you kick ass on such things!

BTW, I wanna ship you my 3 Harman/Kardon CA260s, see what you wanna do. I am finally ready!

I can probably do something with them. How about you shoot one over and I dig in to see where some room for improvement is?

I am not so certain that I agree with you on the 80/20 rule in this case: it really depends on what you're trying to accomplish and what angle you choose to accomplish it by. The least questionable as far as I am concerned is separating the air inside from the vibrations in the body (with the possible side benefit of adding mass to the flat panels of the body to dampen said vibrations). If that is the case, then any gap will be very detrimental. The experimental evidence I've seen - which granted, is qualitative, not quantitative - seems to support this theory as valid in practice.

The effectiveness of mass loading is really pretty location-dependent. Here's a simple example to illustrate why. Imagine a circular piece of sheet metal that is in open air, held rigidly around the perimeter. You push on it with a finger in different locations. Intuitively, you know that you get the most movement for the least force right in the middle where the stiffness seems to be least.

I'll stay out of the math of this, but just remember that the middle gives the most movement for the least force. There will also be a resonant frequency of this membrane. That resonant frequency is related to the material's stiffness and the mass of the membrane (size is a factor in both). This is the frequency where, if energy is added at the same frequency, it will accumulate in the membrane. Say one side has a loudspeaker producing white noise (noise with an equal amount of energy in each octave). Most of the frequencies' energy will not pass through the membrane efficiently since they will fail to really excite it, so it won't vibrate air on the other side much. However, the energy in the frequency that matches the resonant frequency will be heard most loudly on the other side since the membrane will be excited and "ring" at that frequency, effectively acting like a loudspeaker by vibrating & displacing air.

Now, add some mass to that membrane right in the middle (something heavy but with little stiffness...like rubber). The resonant frequency of a spring-mass system is equal to the square root of spring stiffness divided by the mass in the system. So, mass on the bottom of the fraction means that more mass = lower frequency. It is actually a bit more complicated than that for a 2D system like a circular plane, and basically the middle of the membrane has a lot more influence on the behavior than the perimeter where it is clamped. Adding a little mass there is equivalent to adding a lot further from the middle. There is a lot more to it than that, but ultimately that is all you really need to know.

Anyway, adding a little mass to the middle is the easiest way to lower the resonant frequency. At the same time, lower frequencies also require larger moving diaphragms to produce the same amount of acoustic energy, so if you lower the resonance enough then the sheet metal panel will actually produce less acoustic noise because it is less efficient as a loudspeaker.

In the end, the sheet metal panels have many different resonant frequencies since they are of complex geometry, and let a wide band of noise pass through (as anyone that has been on a highway can attest). Think of the panels as being low-pass filters in how they transmit noise into the car. Mass-loading the panels will lower the cut-off frequency, and you want that cut-off to be as far below the voice-band as possible. Lower frequencies will be transmitted even less efficiently as their wavelength grows to about 6x the size of the diaphragm (floor pan) as well.

Putting mass in the middle of the panels makes the most difference, which is why the seemingly small amounts of BMW deadening are all in the middle of panels. Covering everything will still help, but you have to understand that it is a lot of money and work for smaller & smaller improvements. If that doesn't concern you, then go for it. I did.

Check out Acoustics by Leo L. Beranek if you want to start getting at the fundamentals of acoustic engineering. It is an engineering text though, so if you aren't hot on differential equations / partials, math in the complex domain & analog electronics it might not be the most useful. Still, there are enough diagrams and verbiage that you could probably skip the math and at least get the ideas (understanding the math just helps).

Hopefully it is more clear now that I am not just pulling things out of my ass :p.

Actually, I believe your example perfectly addresses the "80/20" rule. You just hit the 100% mark, as exemplified by the fact that you have 220+lbs of the stuff.

220 lbs of cotton would do the exact same thing. Mass is mass, right? All it does is shift the resonance lower in frequency, which is why that much still only did 3db.

It slows resonances in metal, period. In no way does it "separate the air vibrations", that is the side benefit. Read through the silly marketing BS, look at the physics.

Mass is mass, so long as it does not add stiffness to the panel. Mass lowers resonant frequencies, but stiffness increases them. My statement above that gluing a steel bar to the middle would help is actually wrong. It would add mass, but it would also add stiffness, so it would sort of cancel itself out. It would change the 2D vibration modes of the panel which might make things better, or worse.

This is why heavy, compliant stuff like tar and rubber are used for deadening. The aluminum foil on there is probably just there to keep everything else from sticking to the material. I'd actually say that thick rubber with minimal aluminum foil thickness is ideal.

[StereoInstaller1]

My understanding is the ceramic absorbs resonant energy better than anything.

BRB, gonna go see if I can find a kiln for my car.

[AndrewBird]

I've used Peel & Seal in my Jetta on both the rear firewall behind the seat and on the backside of the door panels. The only thing I was expecting it to do is lessen and/or remove vibration, which it did very well. The same amount of road noise still come through, but the panels don't rattle anymore.

For $15 a roll, I am very happy with the results, considering I only used 2 rolls for all of it.

[StereoInstaller1]

You spent $30 on a Jetta?

[AndrewBird]

Tell me about it...

[nmlss2006]

Luke, bmwman
unless I am very much mistaken - physics exams were sometime in 1998 and I stopped playing seriously three years earlier - coating the inside of an echo chamber (like a guitar or a car) with something anechoic or even just not rigid changes the sound inside the chamber pretty significantly - indeed, any interface changes the behaviour of sound waves, unless I am high on something. I am not sure I would discount the effect. If all the effect were had by mass loading large flat areas of the car, you could glue basketballs to the interior and have an identical effect, which I am not sure is the case. There is no doubt that there is a material and well perceivable difference to having no gaps in the coverage, which is something that as far as I am concerned needs an explanation :).
I will check the book, however, I fooled a few professors into believing I knew something about differential equations and analogue electronics.

[bmwman91]

Covering the remaining 80% of stuff will make some quantifiable difference. I think that the discussion is more about whether or not it is really "worth" the time and money to do so. Since coverings are sold by area / square feet, covering the last 80% costs 80% more, and it is logical to assume it'll add more labor, although not 80% more (since the interior has to be stripped either way).

For the person concerned with money, they will reap some benefit from covering the worst-offending areas. For everyone else, cover it all up. I did, and I doubled or tripled-up on the central parts of various metal panels. If I'd had more time and patience, I would have driven around with my measurement mic and Fast Track Pro to measure the change.

[Martin C.]

Interesting discussion. My e46 came with some sort of mat at the door panels, and at the floor. Is very thin but is factory applied.

My e46 is quiter than my previou e36, and far quieter than my e30s. I don't know how it works, but the mat seems to work (don't have hard figures to share though). At the doors I know that the road noise from the neighboring cars travel upward likely onto the lower portion of the door. Here it appears the mat is more effective. Am I right or should I quit drinking tequila?

[e30iwite]

Do not use dynamat as sound or noise dampening. I speak from personal experience. I had used it to cover the ENTIRE (floorboard, ceiling, firewall, you name it) cabin of my toyota tacoma. Here are the reasons why you shouldn't use it:

1. expensive (overpriced)
2. Too difficult to use (the aluminum foil cuts your hands up and it's so sticky that once it is applied, it's nearly impossible to get off)
3. Too heavy (horrible mpg, handling and braking became terrible)
4. does not dampen noise (it's a tar-like material that keeps panels and screws from vibrating, but is NOT a good material for acoustical dampening. Although it does possess some acoustical dampening properties, there are better options.)

If you're wondering why the fuck i applied so much damn dynamat, it was when i was in my crazed audiophile mode. I was running two square L7s and mb quart speakers tied to two hifonic amplifiers 1600 class D mono and class A/B 170x4 channels.

For my e30, I'm currently running 4 coaxial Polk audio speakers (5 1/4) tied to the 4 channel hifonics amplifier I mentioned above. Still contemplating on putting the sub in the trunk. Don't wanna add the extra weight and doesn't need it. The Polk audio speakers are that good. It's so clear and bumps that hard. Of course, they don't bump as hard as the L7s but adequate enough that a friend thought I had installed a small sub in the trunk.

Okay, I've been doing some research on how to noise dampen the e30 and found out that open cell foam is the wall to go. there's also closed cell foam, but open cell dampen sound better. Dynamat actually sells foam pads but at an outrageous price. I found a company online that sells large sheets of sound dampening foam at different thicknesses for cheap. This is how i plan on setting up the e30. their website is foamforyou.com. scroll down on their left panel to find noise control foam. it should bring up a page of all types of foam. Scroll down for the large flat sheets.

I plan on cutting the sheets into smaller pieces and gluing them directly onto the panels beneath all the carpeting. I'm still contemplating on taking off the dash to apply this foam to the firewall. I've seen pictures of somebody's dash removal writeup and noticed that there is actually a thin sheet of some type of sound dampening material behind the firewall. But I think the factory sound dampening material is inadequate. I'm probably gonna go with 1/4 " thickness on the floorboards and 1/8 " thickness for the doors and ceiling. If anybody decides to go this route, make sure to make a post on the quality of the foam and noise dampening property. I'm not going to be doing this project until next spring.

[Threehz]

Good article and very pertinent to what I'd like to do. I've been wanting to try some sound deadening for the firs time on my E30 and I've been wondering exactly how effective it is. The name brands are all over-priced, you can go buy any number of different materials in bulk and achieve similar or better results for a fraction of the price.

I found this and it seems like a cheap, lightweight, and functionally comparable substance to dynamat and others like it. Has anyone tried this stuff out?

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=268-030

I believe it's just parts-express brand.

Anyhow when I have the time and money I'm definitely going to do some sound analysis, I have a pretty high quality portable mic so if nothing else I can do simple sound comparison before and after.

[dwvw]

Wow.
So there are three things you try to do when sound damping. There is isolation, absorption and damping.

Isolation is easy, you are trying to physically block the sound from passing from one area to another with a physical barrier. This type of sound control works well at mid to high frequencies, and is pretty ineffective at controlling bass.
Absorption is taking the energy and converting it to different types of energy, the so called turning sound into heat type argument.
Damping is taking a sound and moving it's frequencies to a range that is less noticeable to human hearing. This is usually shown by knocking on metal, then covering it with damping and there be less "ring".

Do any of these work? Yes, they all work to some degree. I personally think isolation gets you the best bang for your buck. But if you look at the frequencies of different rod noise you will see that eliminating tire noise is going to be hard. And if you look at where wind noise enters a car, you will be hard pressed to control it unless you are willing to put damping material on your windows!

Can you achieve 3dB of reduction at some frequencies, I think yes, I have tested it myself. Is 3dB worth the time, expense and weight penalty, that is up to you? Is 3dB noticeable? Well it's the same as going from 1 300 watt system to a 600 watt system as far as dynamic range and headroom is concerned.

If you look at the pics I posted so far you will see that I have used some damping so far in my car, but I will not be doing things like the roof and the entire floor, as I don't think it's that much of a priority for me.

[dwvw]

Oh and I do believe there are differences in the qualities of brands, I like Cascade products the best, but I get pretty good deals on Dynamat so that's what I am using.