Is that a secret magic sauce or have you industrial suppliers with glue reference to advice, please ?
O assume all has that petrolueum base made of PU, what more ? Air is a stiffener and at least there must be a gluing behavior. Is the stiffening with age something possiblle to avoid ?
The only possibility I see is a very thin inbetween layer of stabilased layer à la sorbothane glued to two solid layer. Not sure any loudspeaker maker go as far...
O assume all has that petrolueum base made of PU, what more ? Air is a stiffener and at least there must be a gluing behavior. Is the stiffening with age something possiblle to avoid ?
The only possibility I see is a very thin inbetween layer of stabilased layer à la sorbothane glued to two solid layer. Not sure any loudspeaker maker go as far...
The damping comes from the viscosity, not the elasticity, right? You don't want to damp anything with a superball; that's so elastic it returns 99% of the energy you wanted to dissipate.
If the material is so overdamped that a small offset remains when the audio stops, what harm occurs? Are you saying that the damping material can be stretched so far that it stiffens & stops damping? Can alternating pressure actually move CLD panels so far in one direction?
I read that assumption about thin constrained layers a lot. Anyone out there that can explain why the layer would have to be thin?
^ i suppose it comes from the fact some glue are used as shear layer in some example? Iirc it was emphasized on some technical description of the principle for metal sheet cld too.
Anyway it depend of the CLD you try to do and material you use: some automotive dedicated products could be used in some case but nothing garantee it'll works fine with wood ( depend of depth/height of it).
Anyway it depend of the CLD you try to do and material you use: some automotive dedicated products could be used in some case but nothing garantee it'll works fine with wood ( depend of depth/height of it).
in mechanic : it is elastic till the resilience point is not reached ! After there is a break with no return to the initial shape.
Ceratinly, plastic could refer to the shearing effect that is the way the material moves with efficienty inside to transform into heat... (the way it recovers... same same but different from elasticity... last outrages to drosophils vocabulary !)
I think I stop Hifi....
Ceratinly, plastic could refer to the shearing effect that is the way the material moves with efficienty inside to transform into heat... (the way it recovers... same same but different from elasticity... last outrages to drosophils vocabulary !)
I think I stop Hifi....
It's everyday PU that I get from a local supplier called Innovative Polymers. It's now owned by a German company. The key however is the 3M micro ballon additive. This yields most of the internal damping.
For any situation there is an optimal inner layer thickness, but it is never "as thin as possible." There are equations that can be used, but experimenting is probably the best idea for DIY.
Thanks. 3M and Sika...two serious companies that share half 90% of the market in adhesiv tech.
Viscoelastic materials are typically used for the damping layer in CLD structures. As the name implies such materials create an elastic force proportional to displacement and a damping force proportional to velocity. The elastic force is typically larger than the damping force and follows from the damping coefficient (or whatever coefficient is used to express the size of the damping compared to the elasticity). The bottom and top layers will typically be statically offset a small amount due to gravity which is not normally problem in practise.
If you want to wrap your mind around how a viscous material with zero stiffness would behave consider a viscous oil. With a bit of support you might get something to work but the damping force is going to be small in size and the thinner you make the layer in order to increase the shearing the more there will be clearance issues and the more the oil is going to heat up and become less viscous.
To get a high damping force with a viscoelastic material one will get (almost always) an even higher stiffness force. However the damping force is not what needs to be maximised but the power dissipated by the damping force which is the product of the damping force and the velocity. Low stiffness means low damping force and hence low power dissipation. A high stiffness approaching that of the structural layer would start to lower the velocity so there is an upper limit. Of course if there was a material with a high damping coefficient and a high stiffness one wouldn't bother with CLD but simply use it in a single layer structure.
An experiment to measure the effectiveness of CLD for a speaker cabinet is far from straightforward and not something I would consider without a robot controlled laser vibrometer. An experiment involving a sealed box, a shaker, a microphone and a lot of patience might produce something qualitatively useful but much of the relevant physics in the motion of a speaker cabinet will be missing. It would also be substantially more expensive and time consuming than a parameter study of a full CLD 3D speaker cabinet using simulation typically used by industry (or at least the more engineering orientated parts of it).
If you want to wrap your mind around how a viscous material with zero stiffness would behave consider a viscous oil. With a bit of support you might get something to work but the damping force is going to be small in size and the thinner you make the layer in order to increase the shearing the more there will be clearance issues and the more the oil is going to heat up and become less viscous.
To get a high damping force with a viscoelastic material one will get (almost always) an even higher stiffness force. However the damping force is not what needs to be maximised but the power dissipated by the damping force which is the product of the damping force and the velocity. Low stiffness means low damping force and hence low power dissipation. A high stiffness approaching that of the structural layer would start to lower the velocity so there is an upper limit. Of course if there was a material with a high damping coefficient and a high stiffness one wouldn't bother with CLD but simply use it in a single layer structure.
An experiment to measure the effectiveness of CLD for a speaker cabinet is far from straightforward and not something I would consider without a robot controlled laser vibrometer. An experiment involving a sealed box, a shaker, a microphone and a lot of patience might produce something qualitatively useful but much of the relevant physics in the motion of a speaker cabinet will be missing. It would also be substantially more expensive and time consuming than a parameter study of a full CLD 3D speaker cabinet using simulation typically used by industry (or at least the more engineering orientated parts of it).
Dr. Geddes,
Kindest regards,
M
I looked at the 3M web-site, which lists many options: https://www.3m.com/3M/en_US/p/c/advanced-materials/glass-bubbles/. Which ones did you use?
Kindest regards,
M
For the glass beads, isn't there some cyanoacrylate fillers you can buy from RC Model stores? Here's an example: https://micro-modele.fr/fr/colles-professionnelles/2206-micro-billes-de-verre-17-ml.html
The way I see CLD right now is like this:
You basically want your constrained layer to get into plastic deformation as soon possible. You want it to be a compound that is stable in time. It shouldn't dry, change its properties, impregnate your substrate. You want something that has either good adhesion with your substrate or that is easy to glue to the substrate. It already sounds like the unobtanium 3000 marketing...
The shear gets greater the thinner the layer is (simple geometry here). Thinner is assuredly better for the constrained layer deformation, but that's until you reach a limit where the constrained layer separates into two parts that slides on top of each other (delamination). So, the layer thickness should be adjusted based on the maximum shear it can support and/or the maximum offending displacement you are expecting. Seeing how little the displacements are for a speaker panels, I don't think thick layers are a good choice and they may act more as orthotropic layering and/or mass loading than CLD. But how do you check that before building anything?
For the moment, the best method I can think of to make a preliminary study would be to make some thorough FEM analysis. But it's time consuming with open source software, expensive with paid software, and I'm a noob for this, so it's also prone to huge errors.
All of those discussions got me to the conclusion that it may simply be overkill to even start to try to make it perfectly right. There is so much parameters involved. The documentation is scarce, the equations are way way over my league (even as a signal processing geek).
I will still get this accelerometer in the mail in some days, and I'm still willing to test some stuffs. I'm going to open another thread and try to list what could be interesting to test. I hope I'll find the time and motivation for this.
The way I see CLD right now is like this:
You basically want your constrained layer to get into plastic deformation as soon possible. You want it to be a compound that is stable in time. It shouldn't dry, change its properties, impregnate your substrate. You want something that has either good adhesion with your substrate or that is easy to glue to the substrate. It already sounds like the unobtanium 3000 marketing...
The shear gets greater the thinner the layer is (simple geometry here). Thinner is assuredly better for the constrained layer deformation, but that's until you reach a limit where the constrained layer separates into two parts that slides on top of each other (delamination). So, the layer thickness should be adjusted based on the maximum shear it can support and/or the maximum offending displacement you are expecting. Seeing how little the displacements are for a speaker panels, I don't think thick layers are a good choice and they may act more as orthotropic layering and/or mass loading than CLD. But how do you check that before building anything?
For the moment, the best method I can think of to make a preliminary study would be to make some thorough FEM analysis. But it's time consuming with open source software, expensive with paid software, and I'm a noob for this, so it's also prone to huge errors.
All of those discussions got me to the conclusion that it may simply be overkill to even start to try to make it perfectly right. There is so much parameters involved. The documentation is scarce, the equations are way way over my league (even as a signal processing geek).
I will still get this accelerometer in the mail in some days, and I'm still willing to test some stuffs. I'm going to open another thread and try to list what could be interesting to test. I hope I'll find the time and motivation for this.
He previously said that he has to guard the exact composition of the damping compound for commercial reasons.
Hi maximax77,
Kindest regards,
M
Thank you for the information, but, as I understand he is no longer in such a business, so I do not think that my question was unreasonable.
Kindest regards,
M
The whole idea of CLD is that the outer layers move independently. Otherwise there would be no shear. Thinner is not better if you consider full independence, but there lies the challenge. It actually is a mix of elasticity in the outer materials, combined with the plastic behaviour of the constrained layer.
In everyday life the availability of materials with sufficient plastic behaviour is limited. Measurements are available, see one of my earlier posts. The myth of the green glue has somewhat been debunked in those: such a construction acts more or less like one panel. And my gut feeling says a lot of constructions with bitumen or otherwise thicker layers do, too. But maybe less so. Bitumen and the like however have better plastic behaviour than most If not all glues, which are merely (visco-)elastic than plastic by nature. If glue would behave merely plastic, it would not glue that well, would it?
Mass loading actually helps reducing transmission, even so in a CLD system. So thick heavy layers with plastic behaviour do make sense.
Can you expand on this with respect to how the motion of the structural layer (kinetic energy) is opposed and converted into heat. Mass in itself doesn't dissipate energy and plastic deformation reduces stress relative to what it would have been if linear which one might expect to reduce energy dissipation. Not saying it doesn't increase the damping force opposing the motion of the structural layer in some way but it isn't clear to me how.
Imo it is rigth.
Those bracings made of one frame with holes and coupling all the sides of the cabinet are just making the enclosure stiff but distribute a d couple resonances between wall panels. There are indeed no sliding panels. Bracings should be made not to stiff but to eat vibration. Meaning the braces should be done with multiple pieces glued with those PU damped glues for sheraings occur. That mean a pary of the bracing being totally decoupled from the cabinet by several glue interfaced. IMO.
Those bracings made of one frame with holes and coupling all the sides of the cabinet are just making the enclosure stiff but distribute a d couple resonances between wall panels. There are indeed no sliding panels. Bracings should be made not to stiff but to eat vibration. Meaning the braces should be done with multiple pieces glued with those PU damped glues for sheraings occur. That mean a pary of the bracing being totally decoupled from the cabinet by several glue interfaced. IMO.
I would like your opinions on what glue/damper to use between aluminium and ply. These will be solidly bolted together with 10mm bolts. I have a very solid aluminium face plate and the rest of the cabinet will be 12mm plywood. Picture shows how the sides will be attached to the U section faceplate. Faceplate is 6" across and 15" high and 3" front to back.
I've seen suggestions for Tightbond Melamine glue and Sikaflex. On ebay here in UK I see Sikaflex 221 and EBT+. I suppose bitumen and butyl sheets are possible but I doubt I want any thickness of filler here - rigidity is the objective. Feel free to suggest any product you think would work here.
I've seen suggestions for Tightbond Melamine glue and Sikaflex. On ebay here in UK I see Sikaflex 221 and EBT+. I suppose bitumen and butyl sheets are possible but I doubt I want any thickness of filler here - rigidity is the objective. Feel free to suggest any product you think would work here.