On covering bass traps - Page 2

Give this man a ceegar.

Since it takes longer for sound to pass through the absorber, it's the same as having a room that's slightly longer in that dimension.

--Ethan

Ethan, my apologies, I was unclear. I was commenting on Nyal's statement that "isolation clips lower the resonance frequency." The boundary conditions I reference involve panel installation - how the panel is mounted and the compliance of those mounts - not room boundaries.

I can see isolation clips reducing transmitted energy, and perhaps increasing the panel's ability to absorb energy, but not altering panel resonance frequency. Uniformly supported resonant structures have the same modes as an unsupported panel. The big change in resonant modes occurs when support is non-uniform.

Think of a straight, inelastic string of length L in free space. The lowest frequency mode supported would be wavelength 2L, but it would support two modes, (center flex/end stationary) and (center stationary/end flex). The availability of 2 modes enables greater energy absorption and storage.

Now tie the ends down - add the boundary condition of no movement at the ends. The lowest frequency mode remains wavelength 2L, but there's only one mode as the ends can no longer move.

What am I missing?

HAve fun,
Frank

A week or two ago, I included the speed of sound thru absorbers change, in an explanation here. As it's a lengthy post, but toward the bottom in the page beginning "Acoustics can be ...", is where I mention it.


Anyway, I've read and read, measured, experimented, etc., I've seen references claiming the speed delta may be hugely significant, like 200%-300%.

Thoughts?



Thanks

The short answer is that speed of sound delta is different per flow resistance, hence different absorption coefficients.
The long answer is, well..
http://perso.ensait.fr/vromanp/documents/publications/Delany-1970.pdf
http://www.cs.takushoku-u.ac.jp/~ymiki/AcoustProp.pdf
http://perso.ensait.fr/vromanp/documents/publications/Komatsu-2007.pdf


Yes, agreed, but aren't we looking at the difference between a non-uniformly supported structure (drywall on studs) vs. a much more uniformly supported structure on isolation clips?
Edited by GIK Acoustics - 3/9/13 at 10:58am

In all honesty, I'm not enough of a physics geek to comment on this knowledgeably.

--Ethan

I've never noticed frequencies shift more than a few percent, but I never did extensive testing. It's tough to imagine any absorber that could shift even 20 percent let alone 300 percent!

--Ethan

The larger difference here is the speed of sound inside the absorber instead of outside. Edit: Also FOH was mentioning speed, not frequency.

If you think about light passing in and out of water:

It travels at the speed of light through air
Hits the water, slows down, bends
Leaves the water at the speed of light again, bends back, amplitude has been decreased due to absorption and reflection.

However, the frequency of the light wave is the same inside and outside of the water, so my previous description actually does not explain the shift in frequency of peaks.

As per the law of conservation of energy, light (or sound in this case) is slowed down in speed and in frequency, but only inside the water (or porous absorber).
The only thing that really changes is the phase and the the position.
Edited by GIK Acoustics - 3/9/13 at 11:56am

I'm not well versed enough to say anything here with confidence, but I wanted to interject that there may be a couple other phenomena at work in a few of these ideas.

I would attribute any increases in absorption associated with isolation clips to the freer movement of the wall, allowing more energy to be stored in the wall structure (during its flexing), therefore made available for absorption in the constrained layer; consider this analogous to damping by a free limp mass.

As a consequence of the motion of the wall, the time between sound wave incidence and reflection is extended, dropping the resonant frequency of the vessel proportionally.

Similarly, a change in speed of sound within a porous absorber would be necessarily quite large in order to impact the resonant frequency of a room, since the porous absorber cannot occupy a large percentage of the room. (This is the same math by which average speed for travelling in a car is only changed slightly by a short period of driving at lower speed limits - a 500 mile drive isn't changed much by twenty minutes of sitting in traffic)

Yes, this part is confusing. fbov has countered by saying iso clips installed correctly provide uniform support, so the modal distribution would be the same regardless. But the second it does start flexing, the modal distribution isn't the same.

This is a very clear example.

I don't think so. Both are constrained at the edge. An example of non-uniform support would be a free, unsupported end, like a tuning fork, which supports an L/4 mode unavailable to edge supported panels.

Consider how isolation clips are installed (scroll down).
http://www.soundproofingcompany.com/soundproofing-products/soundproofing-clips/

In normal construction, the drywall is screwed into 2x4's at the edge of each panel - uniform edge support.

With isolation clips, the drywall is screwed into U-channel at the edge of each panel - uniform edge support. That the U-channel can move in the plane of the panel has no impact on modes normal to the panel.

Now, a lot of other aspects of sound proofing can alter resonance frequencies, but isolation clips aren't one of them....

HAve fun,
Frank
Edited by fbov - 3/11/13 at 1:48pm

Ah, I think I understand you now. Thanks for the clear explanation Frank.

Std construction practice is to attach drywall in the field, not just at the edges, and the stud separation is every 16" on center. With isolation clips the drywall is supported every 24" on center. So we're not just talking about a single panel only supported at the edges. Thinking about it there are sound isolation benefits to using 24" on center spacing as opposed to 16" for wooden studs. I guess the flex of the isolation clips absorbs energy in the low bass that would otherwise be transmitted though the stud structure with a normal wall. I don't think the energy is reflected back into the room.

Excuse me for quoting you again but now that I've re-read this I have a few different questions:

Yes, there is better damping of the resonant modes of the drywall with CLD (and therefore the resonant modes of the wall partition). But this doesn't mean you're absorbing the room's resonant modes (as Arny suggested).

The data you have from Kinetics is referring to absorption due to isolation clips or adding Green Glue? And in interior room absorption coefficients or exterior TL? I have no place to argue about isolation clips, I am only curious.

Thanks!

Edit: Nyal your explanation above is clear. If you have any other responses to the above feel free to respond here or e-mail me. I don't mean to put you on the spot.
Edited by GIK Acoustics - 3/12/13 at 5:58pm

One interesting point about isolation is that, having prevented transmission, the only options left are absorption/damping and reflection (or re-radiation to be more accurate). The CLD sandwich should be more effective the longer the energy stays in the panel, and I can see isolation helping in this regard, making the system a more effective absorber. I can't see this not taking energy from room modes, especially with a larger active area (to Nyal's point about 16" vs 24" mounting centers.)

Speaking of which, the stud down the middle ("attached in the field") does the same thing that the edge mounting did - limit vibration modes available to the panel. I would expect that a wider mounting center spacing, 16 vs 24", would affect absorption efficiency at the lowest frequencies, but not materially shift the absorption spectrum, per se, although that's what it will look like.

Think in terms of speaker bracing. A brace connecting two parallel sides of a box, center to center, greatly reduces the ampliltude of the fundamental resonance, leading one to believe that the fundamenal has shifted upward. Certainly the spectral energy content has shifted upward, but that fundamental mode is still present, just damped into invisibility.

At least that's my understanding of vibratory motion...

Have fun,
Frnak