Constraint Layer Damping(options)

[MidnightSensi]

How do you guys adhere the aluminum side of peel-n-seal to the exterior wood?

Someone else might have something cheaper, but I use West System G/Flex for like everything haha

http://westsystem.com/ss/g-flex-epoxy/

It's like my duct tape of adhesives.

 

[highfigh]

Okay, officer, I don't have a license for the accelerometer, I was just transporting it for a friend.

Actually, I want to measure the panel vibration directly and minimize the ambient room noise and sound from the driver.

I can also specify the exact point on the panel I want to perform the measurement.

Was that what you were looking for?

I was just curious because it's not something most people know about, never mind having one. If someone were to ask where the accelerometer is, I expect that it would be similar to what a former neighbor heard when she asked for some government agency in Okechobee, Fl- "Honey, I cain't tell you where it is, I cain't even tell you what it is".

Have you thought about using roll-on truck bed liner? What about a box inside a box with a layer of sand around the inner one?

 

[Alex2507]

How do you guys adhere the aluminum side of peel-n-seal to the exterior wood?

Clean the aluminum with denatured alcohol and use Liquid Nails. Some say to use an adhesive called PL 4000 (?) but I think liquid nails is good enough. It's all Home Depot stuff.

 

[lsiberian]

Clean the aluminum with denatured alcohol and use Liquid Nails. Some say to use an adhesive called PL 4000 (?) but I think liquid nails is good enough. It's all Home Depot stuff.

I found the post where Chris posted the adhesive he uses. We all know I'm copying him in many ways. Though his work is much more clean and precise.

It's a PL Polyurethane construction adhesive. I'd cheat on any other project, but these speakers are supposed to be my reference ones.

 

[Loren42]

I was just curious because it's not something most people know about, never mind having one. If someone were to ask where the accelerometer is, I expect that it would be similar to what a former neighbor heard when she asked for some government agency in Okechobee, Fl- "Honey, I cain't tell you where it is, I cain't even tell you what it is".

Have you thought about using roll-on truck bed liner? What about a box inside a box with a layer of sand around the inner one?

The accelerometer is just a tool to tell me how well I am doing at absorbing acoustic energy.

I have not considered the bed liner material. I really don't know what acoustic properties it might have, but I am sure the manufacture had acoustics pretty low on the requirements list.

The sand filled walls is a good solution, but perhaps the most difficult to execute.

The accelerometer will allow me to test a number of different concepts and quantitatively tell me what works best for the effort.

The best solution to dampen vibrations may not be the optimum solution if it costs 5 times the price of the closest competition, but only yields a small improvement.

Everything is a trade-off and quantitative measurement is the best way to find how those alternatives rank, for me.

 

[Ted White]

I would be curious what you find out from your trials.

To break it down:

• Damping is the conversion of energy over time and distance.
  
• We are relying on the panel flexing when hit with a sound wave. This includes wall panel movement as a wave (bending wave) is traveling laterally in the wall through the drywall (whatever).
  
• This flexing of the two panels (with VE material in between) will generate a shear force on the VE material. The force will stretch the VE material, and the stretching converts the kinetic (acoustic) energy to heat.
  
• So to create the shear force in the first place, the panels need to flex. Stiffer = less flex. Things that start to inhibit the flex will reduce the shear force and thereby reduce damping. Stiff adhesives, close wall framing (16” vs. 24”), etc. limit flex.

Back in 2003 when we were running damping trials with an accelerometer, we found that damping pads and mats didn’t damp as well as products that started out as a paste. The conclusion was that the pad damping layers were too thick and not intimately bonded well enough to the substrate.

With respect to damping layer thickness, we know that a semi-liquid paste VE damping compound if applied too thick (over 0.75 mm or so) will have less damping capacity than a layer of the same material at 0.5mm. The mats and pads, even those with pressure sensitive adhesive on both sides, were too thick to generate sufficient heat during the shear force. Thinner layers really converted much more energy to heat. The accelerometer studies demonstrated this quite clearly.

The conversion to heat was more efficient if there was intimate bond between the VE and the substrate. A semi-liquid VE damping material seeps into the pores of the plywood, MDF or drywall. This really anchors the material, allowing full stretch during shear. The pressure sensitive adhesive mats often “rolled” across the top of the substrate slightly, making it less susceptible to the shear force generated.

Lastly, you want those panels to flex, and the epoxies, adhesives, etc mentioned here will stiffen the boards, limiting flex and shear, thereby producing a much less damped system.

 

[lsiberian]

The conversion to heat was more efficient if there was intimate bond between the VE and the substrate. A semi-liquid VE damping material seeps into the pores of the plywood, MDF or drywall. This really anchors the material, allowing full stretch during shear. The pressure sensitive adhesive mats often “rolled” across the top of the substrate slightly, making it less susceptible to the shear force generated.

Lastly, you want those panels to flex, and the epoxies, adhesives, etc mentioned here will stiffen the boards, limiting flex and shear, thereby producing a much less damped system.

Rubber has been demonstrated to be an effective CLD layer by EARSC. The peel-n-seal roofing material is a layer of rubber sandwiched between aluminum and asphalt. It flexes quite well. Poly glue is pretty stiff, but more flexible than some of the other adhesives. I leave the measurements to folks like Chris(Wmax) and Andrew. I have no desire to get into the business of measuring CLD stuff(I live in an apartment and barely get by as it is lol). I'll leave that to you guys with the real estate and budget for such projects. I simply implement the best practices I learn from the speaker builders I know. This speaker is already insanely better than any previous build I've undertaken so I won't be disappointed. Another expert on CLD told me the added mass was more likely to blame for the improved measurements than the CLD itself. I guess I'm covered either way.

I really want to read the articles that say less is more so please link. I'd be glad to add the ply layer if I was certain the best results had been achieved already.

 

[Ted White]

Thanks Isiberian. I very much like your insights.

The studies I referenced were done as part of a product R&D effort, not a research paper. The researcher was Brian Ravnaas, who has an astounding wealth of knowledge and insight regarding sound isolation. You may have read some of his work developing Green Glue.

Anyway, to your good points I think you’re right. For very stiff cabinet construction, mass would bring more to the table than damping. Especially when the different mass layers have differing resonance points.

In a much larger container, say the room dedicated to an IB sub(s), CL damping could figure in much more effectively. Walls flex, while cabinets not nearly as much. Less flex, less damping.

 

[lsiberian]

Thanks Isiberian. I very much like your insights.

The studies I referenced were done as part of a product R&D effort, not a research paper. The researcher was Brian Ravnaas, who has an astounding wealth of knowledge and insight regarding sound isolation. You may have read some of his work developing Green Glue.

Anyway, to your good points I think you’re right. For very stiff cabinet construction, mass would bring more to the table than damping. Especially when the different mass layers have differing resonance points.

In a much larger container, say the room dedicated to an IB sub(s), CL damping could figure in much more effectively. Walls flex, while cabinets not nearly as much. Less flex, less damping.

If it were a wall based system I'd definitely use a green glue like substance. But in cabinet construction layering the stuff just sounds like a recipe for disaster.

 

[Ted White]

Many have used a VE material for cabinet construction for a long time, though I've not seen / heard any side-by-side comparisons. I also can't say how many have done this. At any rate, this is a great thread you have cooking.

 

[highfigh]

Many have used a VE material for cabinet construction for a long time, though I've not seen / heard any side-by-side comparisons. I also can't say how many have done this. At any rate, this is a great thread you have cooking.

Since you tested for damping ability, did you also test for the effect on the frequency response of the finished speaker system? It seems many speaker system manufacturers use the coloration from the cabinet in their end result- I have to wonder if the response with a well-damped cabinet will be more peaky than many built of MDF with stuffing, but no damping of the panels.

 

[Ted White]

HighFigh the tests Brian conducted were for wall, ceiling and floor damping, not specifically speakers.

 

[highfigh]

HighFigh the tests Brian conducted were for wall, ceiling and floor damping, not specifically speakers.

OK, but the spans between framing members in a wall or ceiling are often much larger than anything on a speaker so I would think that, while the principles are the same, the applications will be different. The sand jacket is something I have first ear experience with, since my uncle built a pair of speakers with it in the late '50s. They were about as inert as possible and considering the fact that with a wife & kids (not a lot of disposable income at the time), the components available and his ability to test/verify/alter the response paid off well.

 

[Ted White]

... the spans between framing members in a wall or ceiling are often much larger than anything on a speaker so I would think that, while the principles are the same, the applications will be different.

I agree and that was my point in my earlier post. The great mass, bracing and lack of cabinet panel flex make damping compounds less valuable in cabinets than walls.

 

[Loren42]

OK, but the spans between framing members in a wall or ceiling are often much larger than anything on a speaker so I would think that, while the principles are the same, the applications will be different. The sand jacket is something I have first ear experience with, since my uncle built a pair of speakers with it in the late '50s. They were about as inert as possible and considering the fact that with a wife & kids (not a lot of disposable income at the time), the components available and his ability to test/verify/alter the response paid off well.

The problem is the same.

You have, at least in a sealed box, 50% of the driver's energy radiating into a box. That energy must be turned into heat or it will propagate out of the box, either through the box walls or through the back of the driver's cone.

Stiffening a box is one way to help, but you still must convert the internal acoustic energy into heat.

Constrained layering works exactly that way. Internal stuffing also works that way. However, no single method can do it all, so a combination is required.

 

[Ted White]

The parameters are different, though.

In a full scale wall, the panels can flex, causing shear, and CLD can take place. Cut down on the flex, not as much shear, less CLD takes place.

The relative stiffness of the cabinet limit the benefit of CLD. Not saying there's no benefit, but a wall is a different matter altogether.

 

[highfigh]

The problem is the same.

You have, at least in a sealed box, 50% of the driver's energy radiating into a box. That energy must be turned into heat or it will propagate out of the box, either through the box walls or through the back of the driver's cone.

Stiffening a box is one way to help, but you still must convert the internal acoustic energy into heat.

Constrained layering works exactly that way. Internal stuffing also works that way. However, no single method can do it all, so a combination is required.

I understand what damping is and what it does- my point is that, because the cabinet resonances affect the response, has anyone tested the system's frequency response/waterfall, etc with the cabinet un-damped and then damped? However, all the cabinet does is provide a volume of air that the driver uses to augment it's own vibrations. The air resonates sympathetically and unless the cabinet can move in response (which will alter the results), the sound is considered to be 'colored'. Even resonances that fall outside of the specific driver's range will affect the sound of the complete system and IMO, too many people say that it's OK if these higher frequency resonances exist. Either the cabinet should be inert, or it doesn't matter.

I asked if cabinet resonances have been checked because it seems that most speaker companies don't bother. Does anyone have info that supports or refutes this?

As far as "50% of the driver's energy radiating into a box.", a speaker driver causes the internal volume of air to resonate and in a sealed box, the air acts as a spring to dampen the vibrations. Internal damping materials can be used, but I often wonder if much thought is given by the designer to what specific effect they're trying to cause. Is it there to limit cabinet resonances, standing waves or to cheat the cabinet size so a slightly smaller one can be used?

 

[highfigh]

The parameters are different, though.

In a full scale wall, the panels can flex, causing shear, and CLD can take place. Cut down on the flex, not as much shear, less CLD takes place.

The relative stiffness of the cabinet limit the benefit of CLD. Not saying there's no benefit, but a wall is a different matter altogether.

Right- a wall that's flexing sympathetically can cause all kinds of problems, as I posted when I was working on that home theater. The homeowner's construction crew was instructed to build the walls in a certain way and on one of the long walls, they didn't come close to doing it right. Fortunately, it wasn't too hard to peel the backside of that wall to correct the problems and by doing this, it took care of the issues that were screwing up the bass response and distribution of the modes.

 

[Loren42]

The parameters are different, though.

In a full scale wall, the panels can flex, causing shear, and CLD can take place. Cut down on the flex, not as much shear, less CLD takes place.

The relative stiffness of the cabinet limit the benefit of CLD. Not saying there's no benefit, but a wall is a different matter altogether.

Well, it's relative. Large walls require more energy conversion than small panels on a cabinet do.

So the answer is, yes, less damping takes place because the energy to be dampened is proportionally less.

 

[Loren42]

... has anyone tested the system's frequency response/waterfall, etc with the cabinet un-damped and then damped? ...

The answer is yes. The best resource is Vance Dickason's Loudspeaker Design Cookbook, Section 5.40. Also, consider the papers written by:

J. K. Iverson, The Theory of Loudspeaker Cabinet Resonances, JAES, April 1973.

Juha Beckman, Effect of Panel Damping on Loudspeaker Enclosure Vibration, 101st AES Convention, November 1996.

You wrote, "The air resonates sympathetically and unless the cabinet can move in response (which will alter the results), the sound is considered to be 'colored'. I think that you meant to say that if the cabinet walls move or resonate it will color the sound.

I think that it is good practice to reduce cabinet resonances as much as reasonable. There becomes an asymptote where the amount of money and effort applied to squelching resonances does not provide a good return on investment. I suspect that point is pretty variable depending on the manufactures' and DIY goals and budgets.