What Kef describe in LS50 whitepaper. I don't think decoupling being the purpose though.
How i see thing is that the mechanical impedance of each layer (of kind of cld) is different for a given range of frequency of interest.
This is what Tannoy DMT is about too. And i agree the 'matrix' help to spread energy around whole box in your description.
It is more a damping process than decoupling in my view.
Decoupling mean a very high transmission level at a given frequency (fc) then (a fast but) gradual attenuation where no transmission occurs ( and why you need subhertz freq for architectural structural decoupling).
If you use bolt then you'll have to decouple them...
This is how Bon did:
https://www.diyaudio.com/community/...ained-layer-construction.218437/#post-3140184
Last edited:
I am willing to tell all that I know.
But, the last time I bought micro-spheres was > 10 years ago. I simply don't remember what I bought. I know that I used hollow spheres as opposed to solid ones because they worked better. But what size, I don't recall. I don't think that it is critical.
I agree, that's why I used no fasteners in my speakers (except for drivers,) Only well damped glue.
It’s no dissipation but reflection. Mass law presumes zero stiffness. So that alone doesn’t increase damping. You could argue a relatively thick constrained layer with profound plasticity combined with two relatively thin outer layers would work better than thick outer layers and a thin constrained layer, both the same total thickness, qualities of the outer layers and total mass.
I take your point. But in this case we are using 10mm aluminium which is exceptionally stiff. So while the bolts are useful for structural reasons, there may be a lot less vibration passed to the plywood than if you were using an all plywood construction.
Given the fact that this would be bolted, does it even matter much what you put between the aluminium and the outer ply? Silicone would be more of a filler/damper, while Tightbond Melamine glue would be more a glue. Not sure which of the Sikaflexes do what.
Well, then…
Do you recall what shore the pu was that you used for the damping layer? Is a vacuum bag necessary when casting it?
Thanks!
Your approach sound a lot like Enhanced Audio's microphone suspension system to me:
http://www.enhancedaudio.ie/technology.html
Yes - I am very inexperienced in speaker building but I've been reading a lot about aluminium enclosures like Magico, Piega and others and my interpretation was to go for rigidity. I have a couple of aluminium enclosure projects I'm playing with. No doubt damping aluminium isn't the same as damping wood, but you don't want aluminium to ring so some damping is involved, whether CLD or attaching self-adhesive butyl/bitumen to the inner surfaces. I'm trying to work out which way to go here. There's not much on aluminium enclosures on DIY Audio.
Oh yes I get your point. If you go to thin, the constrained layer stiffness kicks in and you're back to a plank of wood. So it's a balancing act.
That's the issue :s
Yes but too thick and the layer won't get out of its elastic region (not enough shear) and will not dissipate any energy. Then it's only mass loading, it will only lower the resonant frequency of the panel.
I did a quick mockup of a CLD in F360. All values are arbitrary.
I don't know how shear is measured so bear with me. On the 5mm example, the disformation of the damping layer is about 15% (from 5mm to 5.7mm). On the 1mm example we get a ridiculous 125%.
This is a vastly exaggerated example and I don't think a real panel movement will be more than +/-100µm without damping. Meaning, do you even get out of the elastic region with a real 5mm thick layer?
It's all about acoustic resonance (frequency) and energy.
With more mass, more energy is needed to get something to move (accelerate).
That being said, with a very un-dampened system, this will eventually also resonate, as long as you hit the right frequency for long enough. An additional problem can be at the higher order harmonics of this system.
Another strategy is to push this frequency entirely out of the frequency band, this can be done with stiffness.
What works "best" REALLY depends on context.
Although in general I like to push the resonance frequency as high as possible.
With adequate internal absorption it is possible to dampen those frequencies enough to not be able to hit those frequencies much anymore.
On top of that, there are internal resonances in the material itself, those are called coincidence frequencies.
Typically those are just dampened by adding (a bit of) mass.
I don't think those are that big of a deal for MDF or Plywood, it's a bit of a different story for metal or glass etc.
But yeah, this is one of the many big compromises of a fullrange two-way system.
Because you can't put the resonance frequency to low (works well for pure mids) and to high either (works well for woofers)
Btw, although not related to side wall resonances, the stiffness of a cabinet also has effect on the Qt of the cabinet (+ speaker). Or rather adding a bit of loss in the piston range performance.
Depending on the situation, but in some cases this can be definitely be audible!
And often mistaken for lack of panel resonances btw!!
Also, there is the issue of direct coupling of energy from speaker basket/frame to the cabinet itself as well as the reaction force from the woofer on the cabinet.
VERY often people mistake those for side wall resonances.
Problem is that as soon as you add mass to the panels, this will also change.
Therefor not comparing apples with apples anymore aka BS measurements and conclusions.
After 10 years, no I don't remember the shore number. It was ie40 from innovative polymers if I remember correctly. I did not use a bag, but my only open surface was machined and it always had bubbles. For no bubbles, you would have to evacuate.
The spheres rub against each other creating friction, which enhances the damping.
I have been imagining they are each coated if they have been mixed well, and are in fact causing many small shear events in the polyurethane. Would that be a reasonable assessment?
As Earl says, it is internal friction that causes the damping. In most materials that have a big stress-strain hysteresis curve, it is internal friction between grain crystals (or some other microscopic structure) that causes the hysteresis curve. Some polymers have such long molecular chains that the coiling and uncoiling of the molecules results in slippage. It is a bit of an art to design a material with high damping, and also have other desirable properties such as a particular stiffness, strength, adhesion, etc.
Metals have this behavior too, but only when strained above the yield point. They will eventually fracture if repeatedly pushed beyond yield.
Using microballoons / microspheres in an adhesive is a clever way to get damping.
Metals have this behavior too, but only when strained above the yield point. They will eventually fracture if repeatedly pushed beyond yield.
Using microballoons / microspheres in an adhesive is a clever way to get damping.
I learnt that aluminium and other materials don’t even have a formal yield point like steel has (picture), but that we agreed on an arbitrary value that keeps things together long enough.
