Hi
(this a spin off from www.diyaudio.com/forums/showthread.php?s=&threadid=98179)
There seem to be two approaches to Constrained Layer Construction:
Wilson Audio say the best form of CLD is twin identical panels of the same mass and stiffness. Like two sheets of equal thickness MDF or Plywood. Then the glue between the sheets has to remain soft and plyable but not too thick
More in post 12 here www.diyaudio.com/forums/showthread.php?s=&threadid=98179
Whereas a local who’s been DIYing for 30 years, uses a composite of 2 - 6! bonded layers
“which have different thicknesses, materials, densities and therefore resonant behavior.
A single layer cannot vibrate at its particular resonant frequency because it is so heavily damped by the other layers.
. .
“Good materials I’ve used for constructing composite box walls are:
1. Chipboard flooring (yellow & red tongue) - it has 3 pronounced layers of differing densities, and it’s fairly cheap
2. Plywood (at least 18 mm thick, and with plys at least 3 mm thick). Small air voids would make no difference. Expensive.
3. Masonite (a low density thin layer which helps to break up resonances). Cheap.- Ordinary chipboard. Cheap
4. 5 mm bitumen impregnated vinyl (used for deadening sound transfer through panels in cars etc). Can use 1, 2 or 3 layers stuck together.
5. Sand and bitumen paint mixed to make a render. Very messy, smelly and takes a long time to dry.
http://groups.yahoo.com/group/maclub/message/26753
His latest evolution was
1. 1 mm Walnut wood veneer,
2. 4.8 mm Masonite,
3. 2 layers of 3 mm car sound deadening panels,
4. 19.5 mm "yellow tongue structafloor" chipboard,
5. 1 layer of Bostic (car sound) deadening panel,
6. 14 mm poly foam.
(he uses a contact adhesive, and small zinc plated nails to hold on bitumen or lead panels while the glue dries)
with modest drivers, sounded Very good
Tho I'd only do 6 layers for an "Ultimate" project
The logic of “different thicknesses, materials, densities and therefore resonant behavior, and
A single layer cannot vibrate at its particular resonant frequency because it is so heavily damped by the other layers.”
- to me sounds more convincing .
Any other approaches or views?
Cheers
(this a spin off from www.diyaudio.com/forums/showthread.php?s=&threadid=98179)
There seem to be two approaches to Constrained Layer Construction:
Wilson Audio say the best form of CLD is twin identical panels of the same mass and stiffness. Like two sheets of equal thickness MDF or Plywood. Then the glue between the sheets has to remain soft and plyable but not too thick
More in post 12 here www.diyaudio.com/forums/showthread.php?s=&threadid=98179
Whereas a local who’s been DIYing for 30 years, uses a composite of 2 - 6! bonded layers
“which have different thicknesses, materials, densities and therefore resonant behavior.
A single layer cannot vibrate at its particular resonant frequency because it is so heavily damped by the other layers.
. .
“Good materials I’ve used for constructing composite box walls are:
1. Chipboard flooring (yellow & red tongue) - it has 3 pronounced layers of differing densities, and it’s fairly cheap
2. Plywood (at least 18 mm thick, and with plys at least 3 mm thick). Small air voids would make no difference. Expensive.
3. Masonite (a low density thin layer which helps to break up resonances). Cheap.- Ordinary chipboard. Cheap
4. 5 mm bitumen impregnated vinyl (used for deadening sound transfer through panels in cars etc). Can use 1, 2 or 3 layers stuck together.
5. Sand and bitumen paint mixed to make a render. Very messy, smelly and takes a long time to dry.
http://groups.yahoo.com/group/maclub/message/26753
His latest evolution was
1. 1 mm Walnut wood veneer,
2. 4.8 mm Masonite,
3. 2 layers of 3 mm car sound deadening panels,
4. 19.5 mm "yellow tongue structafloor" chipboard,
5. 1 layer of Bostic (car sound) deadening panel,
6. 14 mm poly foam.
(he uses a contact adhesive, and small zinc plated nails to hold on bitumen or lead panels while the glue dries)
with modest drivers, sounded Very good
Tho I'd only do 6 layers for an "Ultimate" project
The logic of “different thicknesses, materials, densities and therefore resonant behavior, and
A single layer cannot vibrate at its particular resonant frequency because it is so heavily damped by the other layers.”
- to me sounds more convincing .
Any other approaches or views?
Cheers
Rick,
For CLD to work properly, there needs to be an elastic damping layer sandwiched (bonded) between two more ridgid layers. At certain frequencies the two outer layers will be de-coupled from each other and will vibrate and be damped independently according to their separate properties. At these frequencies, the elastic layer does not pass the vibrational energy from one outer layer to the other.
The three layers acts as a lowpass filter so at lower frequencies and all three layers will vibrate together. The cutoff of this filter should be low enough to be below audible range or be damped enough to minimize excessive vibration. However, at certian frequencies, the outer layer masses will resonate against the spring of the elastic layer. This would a good reason for selecting two outer layers with different properties. The outer layers should be selected with the stiffness, mass, and internal damping to resist vibrating at these resonant frequencies.
Selecting the correct materials and construction for proper CLD methods is not trival and will require some advanced analysis or a lot of experimentation, or both. To complicate matters significantly, more layers can be added including the addition of Free Layer Damping (coatings) methods and different panel geometry.
For CLD to work properly, there needs to be an elastic damping layer sandwiched (bonded) between two more ridgid layers. At certain frequencies the two outer layers will be de-coupled from each other and will vibrate and be damped independently according to their separate properties. At these frequencies, the elastic layer does not pass the vibrational energy from one outer layer to the other.
The three layers acts as a lowpass filter so at lower frequencies and all three layers will vibrate together. The cutoff of this filter should be low enough to be below audible range or be damped enough to minimize excessive vibration. However, at certian frequencies, the outer layer masses will resonate against the spring of the elastic layer. This would a good reason for selecting two outer layers with different properties. The outer layers should be selected with the stiffness, mass, and internal damping to resist vibrating at these resonant frequencies.
Selecting the correct materials and construction for proper CLD methods is not trival and will require some advanced analysis or a lot of experimentation, or both. To complicate matters significantly, more layers can be added including the addition of Free Layer Damping (coatings) methods and different panel geometry.
Personally, I would build two test boxes. Both very cheap.
1.
yellow tongue chipbord- silicone bathroom sealer-yellow tongue chipboard.
I can imagine the resonance of both inner and outer panels to be around equal, and with the silicone acting as a compliance material, the second outer board may (or may not) resonate in anti phase with the inner wall, cancelling overall resonance to some degree, and at some frequencies.
Gotta be worth a test box!
2.
Masonite-siliconed to an open timber frame- siliconed to masonite, voids filled with sand.
Energy travelling through the inner masonite panel would be absorbed by the loose sand before reaching the outer skin.
Any energy transmitted through the silicone bond into the timber frame would need to travel through another silicone bond before transfer to the outer skin.
the frame could even be made from chipboard with slots or holes cut into it.
Mind you, there is a good chance neither would work in practice, but it would be a nice and fairly cheap study if I had the time!
1.
yellow tongue chipbord- silicone bathroom sealer-yellow tongue chipboard.
I can imagine the resonance of both inner and outer panels to be around equal, and with the silicone acting as a compliance material, the second outer board may (or may not) resonate in anti phase with the inner wall, cancelling overall resonance to some degree, and at some frequencies.
Gotta be worth a test box!
2.
Masonite-siliconed to an open timber frame- siliconed to masonite, voids filled with sand.
Energy travelling through the inner masonite panel would be absorbed by the loose sand before reaching the outer skin.
Any energy transmitted through the silicone bond into the timber frame would need to travel through another silicone bond before transfer to the outer skin.
the frame could even be made from chipboard with slots or holes cut into it.
Mind you, there is a good chance neither would work in practice, but it would be a nice and fairly cheap study if I had the time!
I had professor who worked for the school only one term due to staffing issues, he normally works for a company that make kinematic mounts. While he mostly deals with XY stages he has done extensive work with vibrations and more importantly damping them.
The definition of CLD is two panels separated by a thin viscous layer.
I asked him why they needed to be equal and he gave me this answer, if you have a rigid substrate and constraining layer (i.e. thick wood) then when one is set into a vibratory mode it will only resonate at higher frequencies due to its stiffness, high frequency resonance causes smaller ripples when compared to low frequency resonance. This ripple causes "shear strain" in the damping layer, which is converted into low level latent heat, instead of transmitting or reflecting it.
The problem with thin constraining layers is that they don't resist low frequency modes which when taken in the context of a composite structure, the low freq. modes appear as thin panels bending towards the thicker panels, the thin damping layer cannot convert this compression strain into heat as efficiently so it transmits it to the next layer. It will still work, but efficiency is lost. The important thing is to have a constraining layer, free layers are not any where near as efficient.
Not the most descriptive or technical answer but...
for more info:CLD
Hi Rick,
i have previously used silicone and a water based contact cement for CLD panels and been underwhelmed with the results.
I've just used Green Gluehttp://www.greengluecompany.com to make an open baffle speaker with 15mm birch ply and 12mm viroc (a cement wood fibre board). I also used it to quieten some portable humidifying machines at work. I am very pleased with the results the green glue does seem to be the right stuff to use.
If you can get hold of it in oz I would definitely use it, i's primary use is for applying between plasterboard for sound proofing but it works just as well with wood.
The things to note is that you have to use screws or nails to join the panels. I used a few 18Ga brads from a nail gun, also make sure you find something to clean it up before you start, it flows quite easily when first applied but then remains permanently tacky. I used a citrus based glue remover.
If I ever make any more boxes I would definitely use it.
regards
Khush
i have previously used silicone and a water based contact cement for CLD panels and been underwhelmed with the results.
I've just used Green Gluehttp://www.greengluecompany.com to make an open baffle speaker with 15mm birch ply and 12mm viroc (a cement wood fibre board). I also used it to quieten some portable humidifying machines at work. I am very pleased with the results the green glue does seem to be the right stuff to use.
If you can get hold of it in oz I would definitely use it, i's primary use is for applying between plasterboard for sound proofing but it works just as well with wood.
The things to note is that you have to use screws or nails to join the panels. I used a few 18Ga brads from a nail gun, also make sure you find something to clean it up before you start, it flows quite easily when first applied but then remains permanently tacky. I used a citrus based glue remover.
If I ever make any more boxes I would definitely use it.
regards
Khush
Richard,
Tony Gee used a 3-layer approach to both boxes for his Andromeda speakers. 22mm MDF, then 4mm lead-bitumen, then 18mm MDF. Worth a read.
But it's interesting to note that he hasn't gone back to this approach since, even for his Soup speakers. Possibly the extra work involved doesn't produce the improvement desired. You would have to ask him yourself - he is a member here "Geenius".
Tony Gee used a 3-layer approach to both boxes for his Andromeda speakers. 22mm MDF, then 4mm lead-bitumen, then 18mm MDF. Worth a read.
But it's interesting to note that he hasn't gone back to this approach since, even for his Soup speakers. Possibly the extra work involved doesn't produce the improvement desired. You would have to ask him yourself - he is a member here "Geenius".
Hi Everyone,
The Andromeda construction is still hard to beat. The reason why the Soup loudspeaker didn't use this technique is due to the fact that the outside dimensions were already determined, I needed a certain amount of litres volume that lead to the resulting 30mm mdf panel thickness. I could have made it (for example), 18mm mdf + 4mm dampingpanels + 8mm mdf, but with all the angles it would have got a bit too complex. Saying all this, it is possible to make the Soup better if you use the layered technique.
Thanks for the explanation, Tony. I was wondering if it was "just too complex" for use in all your other speaker designs. I think the only other build I've read about that came close (in complexity) was Lars Mytting's "Toccata Grande" build (on the SpeakerBuilding.com site).Geenius said:
Both Andromeda and Toccata made me feel tired just reading about them - I'm too lazy to go to all that trouble (unless I win the lottery!)
I'm thinking now, if time were not a scarce resource, CLD could well be the way to go . . but in the real world, or at least my part of it, the ideas back in
www.diyaudio.com/forums/showthread.php?s=&threadid=98179
are more doable, in particular low mass , maybe aperiodic.
If SL and a local former speaker teacher endorse it . .
Cheers
www.diyaudio.com/forums/showthread.php?s=&threadid=98179
are more doable, in particular low mass
, maybe aperiodic. If SL and a local former speaker teacher endorse it . .
Cheers
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.