I am wondering if anyone who has used any of the damping sheets (Dynamat, Megamat etc etc) on in their speaker projects and attempted to quantify the "before and after" panel vibrations once their enclosure has been treated?
I am putting together a 3 way design at the moment using damping sheets from an Australian manufacturer that is a heavy polymer with a sound absorbing internally facing material. These guys supply to heavy industry and are involved in fixing issues with heavy transport vehicles so they test their products extensively. The sheet adds mass to the panel and of course reduces reflections in the enclosure though of course I doubt this effect will extend in any way to lower frequencies.
There are a lot of car-sound oriented suppliers of damping material out their that offer little quantitative figures on the performance of their products and perhaps rely on empirical data to design their product - though I am by no means stating that they don't work! The great majority of these products seem to rely on constrained layer damping (CLD) which consists of a dense polymer (some use bitumen but this best avoided) and a sheet of "damp" aluminium. I am interested in working out how effective the CLD products stack up against the D10 damping sheet. In fact I am interested in finding out whether the implementation of CLD in these sheets is really effective as their intended use is on metal panels. How do they stack up when placed on a thicker substrate such as MDF or Plywood for example which has different mechanical properties?
I was thinking about getting a piezo instrument pickup then running it into my Behringer UMC204 and using REW to get an idea of the waterfall plots of a panel before and after the damping material has been applied. I know the pickup with inherently have an uneven sensitivity over the frequency range but I am probably going to use a subtractive process i.e. measure a reference panel with no damping, then re-measure with damping applied and subtract the 2nd measurement to see what dB attenuation has occurred and at what frequency.
In the time domain, REW allows you to overlay 2 waterfall plots to see the difference at a given time from the original signal.
So does anyone have any experience with this?
Did you try different manufacturer's products to compare them?
I am putting together a 3 way design at the moment using damping sheets from an Australian manufacturer that is a heavy polymer with a sound absorbing internally facing material. These guys supply to heavy industry and are involved in fixing issues with heavy transport vehicles so they test their products extensively. The sheet adds mass to the panel and of course reduces reflections in the enclosure though of course I doubt this effect will extend in any way to lower frequencies.
There are a lot of car-sound oriented suppliers of damping material out their that offer little quantitative figures on the performance of their products and perhaps rely on empirical data to design their product - though I am by no means stating that they don't work! The great majority of these products seem to rely on constrained layer damping (CLD) which consists of a dense polymer (some use bitumen but this best avoided) and a sheet of "damp" aluminium. I am interested in working out how effective the CLD products stack up against the D10 damping sheet. In fact I am interested in finding out whether the implementation of CLD in these sheets is really effective as their intended use is on metal panels. How do they stack up when placed on a thicker substrate such as MDF or Plywood for example which has different mechanical properties?
I was thinking about getting a piezo instrument pickup then running it into my Behringer UMC204 and using REW to get an idea of the waterfall plots of a panel before and after the damping material has been applied. I know the pickup with inherently have an uneven sensitivity over the frequency range but I am probably going to use a subtractive process i.e. measure a reference panel with no damping, then re-measure with damping applied and subtract the 2nd measurement to see what dB attenuation has occurred and at what frequency.
In the time domain, REW allows you to overlay 2 waterfall plots to see the difference at a given time from the original signal.
So does anyone have any experience with this?
Did you try different manufacturer's products to compare them?
I know this is not the same as what you are asking about the sheets that can be added. With wood you can use latex caulking layer in between.
In a foamcore tractrix horn, I saw a huge difference in HD with and without CLD. I used latex caulking (heavy layer) sandwiched in between two sheets of foam core.
Here is HD before the CLD:
Here is the HD after CLD:
It also improved the SPL output because less power is wasted on vibrating the horn walls.
I went from -25 to -30dB levels of HD down to -55dB to -60dB HD. A huge improvement - it sounds so clean afterwards. What was neat was that I was playing music through it as I applied the CLD and you could hear it change before your ears.
More info here:
http://www.diyaudio.com/forums/full-range/259293-prv-5mr450-ndy-fast-applications-13.html
In a foamcore tractrix horn, I saw a huge difference in HD with and without CLD. I used latex caulking (heavy layer) sandwiched in between two sheets of foam core.
Here is HD before the CLD:
Here is the HD after CLD:
It also improved the SPL output because less power is wasted on vibrating the horn walls.
I went from -25 to -30dB levels of HD down to -55dB to -60dB HD. A huge improvement - it sounds so clean afterwards. What was neat was that I was playing music through it as I applied the CLD and you could hear it change before your ears.
More info here:
http://www.diyaudio.com/forums/full-range/259293-prv-5mr450-ndy-fast-applications-13.html
Thanks xrk971.If you Google the BBC Research papers on their speaker designs, there is useful information on the work they did with constrained layer damping. Points sometimes missed in discussions here are
- thinner panels worked just as well as thicker when undamped
- the damping was in layers, using 3mm thick bitumin sheets, which enabled joins to be covered without adverse effect (I'm not sure if that was a specific technique or simply because of the availability of the material)
- the thickness of the damping layer matched the thickness of the panel
I use a small, BBC enclosure (9mm ply and 3 layers of 3mm bitumin sheet) which subjectively seems to work very well and makes the sound seem very solid, for want of a better description.
The BBC papers were interested in cost-efctive engineering rather than state of the art but they're an informative starting point.
Hope that helps.
- thinner panels worked just as well as thicker when undamped
- the damping was in layers, using 3mm thick bitumin sheets, which enabled joins to be covered without adverse effect (I'm not sure if that was a specific technique or simply because of the availability of the material)
- the thickness of the damping layer matched the thickness of the panel
I use a small, BBC enclosure (9mm ply and 3 layers of 3mm bitumin sheet) which subjectively seems to work very well and makes the sound seem very solid, for want of a better description.
The BBC papers were interested in cost-efctive engineering rather than state of the art but they're an informative starting point.
Hope that helps.
As I am into car audio as well I follow the DIYMobileAudio forums. There's a massive thread on CLD testing going on with CLD tests on metal (obviously). Still interesting to see how the different CLD tiles compare:
Sound Deadening (CLD) Testing - Car Audio | DiyMobileAudio.com | Car Stereo Forum
Actual tests start at page 9.
I had plans to use automotive CLD tiles in my home speaker build and these tests confirmed that decision. I'll be using Silent Coat 4mm damping on (stacked) Birch Ply. No before and after measurements planned though.
Sound Deadening (CLD) Testing - Car Audio | DiyMobileAudio.com | Car Stereo Forum
Actual tests start at page 9.
I had plans to use automotive CLD tiles in my home speaker build and these tests confirmed that decision. I'll be using Silent Coat 4mm damping on (stacked) Birch Ply. No before and after measurements planned though.
Last edited:
Not to be too pedantic but lets be clear on the definition of constrained layer damping. Non constrained damping is any layer of goo adhered to a rigid structure. Studies show that the approach damps resonance but that a significant amount will be required for effectiveness. Think in terms of damping mass and structure mass being similar. As the structure gets heavier the damping becomes less effective.
Much more effective than surface damping is to split the structure into two layers and apply damping between the layers. (Hence "constrained layer"). This forces more energy through the damping as motion of the structure creates shear motion through the material.
Constrained layer damping works well, even better with thin damping layers. I have seen automotive aftermarket products with a thick layer of goo and a thin layer of foil on the outside, and said to be CLD. They are not. The BBC bitumin examples are also not CLD.
David
Much more effective than surface damping is to split the structure into two layers and apply damping between the layers. (Hence "constrained layer"). This forces more energy through the damping as motion of the structure creates shear motion through the material.
Constrained layer damping works well, even better with thin damping layers. I have seen automotive aftermarket products with a thick layer of goo and a thin layer of foil on the outside, and said to be CLD. They are not. The BBC bitumin examples are also not CLD.
David
This would be good information to see. I think you are probably right to be wary of products designed for different applications. My suspicion is that a loudspeaker cabinet that deflects only a small amount relative to a thin metal panel is going to need a significantly stiffer damping layer to dissipate a worthwhile amount of energy when strained but I haven't looked at the numbers.
David, that's what I was questioning in my original post. The "CLD" that is claimed in car audio is a very different beast to that which is employed by high-end speaker manufacturers and DIY'ers. I was questioning whether anyone had tested this type of damping sheet on MDF or similar. The tests that Wesayso kindly provided the link to are quite exhaustive but (understandably) were aimed at investigating the effectiveness of the sheeting on metal - which is their intended purpose. I will have to set up a piezo guitar pickup and do some comparative tests. I understand that piezo pickups do not exhibit a flat response but I am looking at comparative results. If I set up a control (undamped) panel then measure various configurations of damping I can use a subtractive approach to see what benefits, if any, the damping sheets have on panel resonances in my enclosure. It's a project that's on my to-do list!
I would suggest measuring multiple random locations on that test panel to get a better understanding of how that dampening material works. A single test point may produce misleading results.
Oren42. The pickup I have my eye on has 2 sensors. I guess making multiple tests with the pads in predetermined locations and blending them will be the only way to get proper data.
If you look at the graphs of the test I linked you'll see that the better damping materials (CLD) are very successful in shortening the ringing period of the metal up to 1000 Hz. The biggest peak in the metal is at 80 Hz but there are several smaller ones higher up that don't ring after damping with CLD. The resonances in the building materials we usually use for speakers can vary depending on size and thickness and the amount of braces.
Keep in mind CLD coverage was about 30% in these tests.
One more note about the CLD materials for Cars, most of them are Butyl based products with a thin aluminum layer. There are some that still use bitumen. That's not a good choise in a Car that sees a lot of temperature changes but it might still work well in our home environment. For example some STP products scored very well in the damping tests but are bitumen based. They don't hold up well to heat testing and for that reason they are not recommended for use in a Car. Usually the bitumen based products are cheaper though. And we won't see as severe temperature swings in our listening rooms.
I've seen several threads on this forum using bitumen matt, without the aluminum constraining layer. So if you do plan on a test I'd recommend testing at least 3 different products: Bitumen matt, as often used in speaker builds, a CLD with bitumen as main ingredient and a CLD that is butyl based.
I won't recommend the base speaker material (MDF or Birch Ply or some other variant), that's a discussion that never ends around here. 😉
Keep in mind CLD coverage was about 30% in these tests.
One more note about the CLD materials for Cars, most of them are Butyl based products with a thin aluminum layer. There are some that still use bitumen. That's not a good choise in a Car that sees a lot of temperature changes but it might still work well in our home environment. For example some STP products scored very well in the damping tests but are bitumen based. They don't hold up well to heat testing and for that reason they are not recommended for use in a Car. Usually the bitumen based products are cheaper though. And we won't see as severe temperature swings in our listening rooms.
I've seen several threads on this forum using bitumen matt, without the aluminum constraining layer. So if you do plan on a test I'd recommend testing at least 3 different products: Bitumen matt, as often used in speaker builds, a CLD with bitumen as main ingredient and a CLD that is butyl based.
I won't recommend the base speaker material (MDF or Birch Ply or some other variant), that's a discussion that never ends around here. 😉
Last edited:
Also look here at diy. There are many threads on the subject. Like many things audio, I was interested in this a while back and satisfied my idle curiosity 🙂
The important thing for CLD is that there be (at least) two firm surfaces separated by a gooey wiggly substance. Everything else is details 🙂
The important thing for CLD is that there be (at least) two firm surfaces separated by a gooey wiggly substance. Everything else is details 🙂
I made a test some years ago to test the effect of gluing bitumen sheets to the inside of a cabinet. Not constrained layer though.
To make the test, I built a 200lt cabinet using 20mm birch plywood and made a removable front baffle that had a 40lt subenclosure (cube) with the baffle sealed. A 12" woofer was mounted on the inside face of that cube, firing inside the 200lt cabinet (to the 160 or so liters remaining).
I made frequency response measurements using a warble tone generator and a measurement microphone located at 1m from the enclosure. All tests were done with the amplifier at the same settings.
On the graph attached, the first plot is of a KEF reference 105 speaker - for reference of the measurements (to show any room effects). You can see some of the effects, like a peak at 33Hz, another one at 500Hz, and so forth.
The second plot (as per the legend) is of the test enclosure without any layering. The levels (save for the resonance at 120Hz) are lower than the free firing R105 since the 12" woofer is totally enclosed in the 20mm cabinet. However, the levels aren't that low and I could clearly hear the tones.
Third plot is of the enclosure treated with a single layer of 3mm bitumen sheets fully covering the inside of the side panels only. Some resonances are tamed and the main resonance at 120Hz drops both in level (-6dB) and frequency (-20Hz).
Fourth plot is after applying a single layer to the back panel as well. There is still a noticeable drop in the main resonance as well as throughout the spectrum (useful spectrum that is, 20Hz ~ 300Hz).
Fifth plot is after applying a full second layer to the inner surfaces (except the front) and the next one after applying 3 layers.
Finally I braced the cabinet just below the middle of the height. That gave some effect but IMHO not that much. I must point out that at the third layer (9mm), I had issues doing the measurements and had to increase the level of the amplifier by 12dB (as measured with reference to the R105 I used) and then subtracted them from the plots. The level produced otherwise was so low the meter measured the sound of the leaves brushed by the wind outside through closed windows...
My next experiment will be to build an enclosure out of 4mm plywood alternated with 3mm bitumen sheets. It's a simple idea but I've already found several technical issues to construct - and I haven't even started the process yet.
I do agree that the thicker the wooden panels, the more the damping material needs to be for effect. And believe me the last layer of bitumen was applied with the cabinet resting on it's back on the floor. I could not lift it up on the bench by myself.
To make the test, I built a 200lt cabinet using 20mm birch plywood and made a removable front baffle that had a 40lt subenclosure (cube) with the baffle sealed. A 12" woofer was mounted on the inside face of that cube, firing inside the 200lt cabinet (to the 160 or so liters remaining).
I made frequency response measurements using a warble tone generator and a measurement microphone located at 1m from the enclosure. All tests were done with the amplifier at the same settings.
On the graph attached, the first plot is of a KEF reference 105 speaker - for reference of the measurements (to show any room effects). You can see some of the effects, like a peak at 33Hz, another one at 500Hz, and so forth.
The second plot (as per the legend) is of the test enclosure without any layering. The levels (save for the resonance at 120Hz) are lower than the free firing R105 since the 12" woofer is totally enclosed in the 20mm cabinet. However, the levels aren't that low and I could clearly hear the tones.
Third plot is of the enclosure treated with a single layer of 3mm bitumen sheets fully covering the inside of the side panels only. Some resonances are tamed and the main resonance at 120Hz drops both in level (-6dB) and frequency (-20Hz).
Fourth plot is after applying a single layer to the back panel as well. There is still a noticeable drop in the main resonance as well as throughout the spectrum (useful spectrum that is, 20Hz ~ 300Hz).
Fifth plot is after applying a full second layer to the inner surfaces (except the front) and the next one after applying 3 layers.
Finally I braced the cabinet just below the middle of the height. That gave some effect but IMHO not that much. I must point out that at the third layer (9mm), I had issues doing the measurements and had to increase the level of the amplifier by 12dB (as measured with reference to the R105 I used) and then subtracted them from the plots. The level produced otherwise was so low the meter measured the sound of the leaves brushed by the wind outside through closed windows...
My next experiment will be to build an enclosure out of 4mm plywood alternated with 3mm bitumen sheets. It's a simple idea but I've already found several technical issues to construct - and I haven't even started the process yet.
I do agree that the thicker the wooden panels, the more the damping material needs to be for effect. And believe me the last layer of bitumen was applied with the cabinet resting on it's back on the floor. I could not lift it up on the bench by myself.
Attachments
Thanks everyone for their input. Lots to think about here. I ordered some additional Dynamat (as well as the Megasorber D10 that I already have). I'm hoping to do some tests with undamped, D10 only, Dynamat Xtreme only and both. I'm running out of time though as it's approaching the end of the year and I'm not going to have access to the speaker whilst it gets assembled and spray painted over the break.
- Status
- Not open for further replies.