Andy touched on mass-loading, in addition to damping and bracing, and I find that a very valuable tool as well. (If you haven't already looked up @augerpro thread on empirical bracing research, please do so.)
To brace is to stiffen, damp to absorb, and mass-load to slow or increase the energy required to make a panel move. To mass load is basically as it sounds, to add mass to panels. I've used a bunch of 4" tiles from the local hardware, marble, ceramic, glass, whatever. Just buy a box. Then I space them out in a line and glue them to the largest panels, say 3-4 of them on each of the inner side walls of a tower. E6000 to glue them to the surface ugly side down, and run an X of E6000 lines over top of them to where the X glue joins the under glue on the cabinet wall. Once this dries, you will notice the difference. Place wall bracing between them as in B.
To brace is to stiffen, damp to absorb, and mass-load to slow or increase the energy required to make a panel move. To mass load is basically as it sounds, to add mass to panels. I've used a bunch of 4" tiles from the local hardware, marble, ceramic, glass, whatever. Just buy a box. Then I space them out in a line and glue them to the largest panels, say 3-4 of them on each of the inner side walls of a tower. E6000 to glue them to the surface ugly side down, and run an X of E6000 lines over top of them to where the X glue joins the under glue on the cabinet wall. Once this dries, you will notice the difference. Place wall bracing between them as in B.
mass loading can be a false friend who makes up into your real friend : stifness. Most of the time thickness is added to enhance the stifness. But the mass decreases the resonance. SO you want stiff and ligth for the bass to push out the structure resonances higher than the bandpass of the bass driver, and weigth for the the mid to trebles to push the resonance lower. But gravity makes the story difficult as most of the time the mid/treble cabinet stand on the bass cabinet or simply is in the same standalone cabinet. I loved btw what Linkwitz made with his design where the open bafle doesn't stand on the bass cabinet but on a structure standing on the soil around it. When a standalone cabinet you have a trade off to choose and also in the internal longer size of the bass cabinet as a friend tool as well.
Mass, as far I know is not that interresting for dampening but to block bass frequency to pass trough (hence heavy walls in house works better for isolation than wood vibrating walls.
What you better want is a dampening with a larger bandwidth and play with the mass spring effect to translate energy into heat if I understood it well enough. The challenge is to make such a material with opposite properties : stifness VS damping. Hence the material should not be the same according the frequencies played by the transducers and the cut offs frequencies I assume. Hence also the ubiquitpus composite materials talked already.
This contradiction is well illustrated by constraining layers damping technic in hifi where "elasticity" must be targeted to a point that does not weak the stifness too much to be counterproductive.
Energy must out somewhere.
Mass, as far I know is not that interresting for dampening but to block bass frequency to pass trough (hence heavy walls in house works better for isolation than wood vibrating walls.
What you better want is a dampening with a larger bandwidth and play with the mass spring effect to translate energy into heat if I understood it well enough. The challenge is to make such a material with opposite properties : stifness VS damping. Hence the material should not be the same according the frequencies played by the transducers and the cut offs frequencies I assume. Hence also the ubiquitpus composite materials talked already.
This contradiction is well illustrated by constraining layers damping technic in hifi where "elasticity" must be targeted to a point that does not weak the stifness too much to be counterproductive.
Energy must out somewhere.
Last edited:
Measure is one thing human perception an other. The thinwalled damped BBC speakers (LS3/5A and Spendor BC1, Harbeth and others) They use massdamping and thin/weak walls to bring the resonances out of the critical midrange down in the bass range. Their take is that apparent harm to the music is less for resonances in the bass than in the midrange. Other manufacturers go for dense and stiff materials.
It is all trade off between different accoustic parameters as well as cost and size.
It is all trade off between different accoustic parameters as well as cost and size.
Now we had a lot of wise talk and no one knows what to do in practice... We only know that obviously most speaker cabinets are build wrong and must sound pretty ugly. I go and take a depression.
Lol... It seemed to me evident ! After several breakdowns, solution is obvious : supress the cabinet and go open baffle !🤣
If a doubt yet : go baffle-less
If a doubt yet : go baffle-less
We've got a question (bracing X dumping), then we went through theoretical discussion to verify, clarify and we've got solid conclusions.
For me, it's clear: the best cost/benefit is simple bracing with the same wood which panels are made.
We move the resonances to higher frequencies (out of main woofer pressure) and problem is solved.
In last months, I had focus on removing speaker resonances of some prototypes I've built and I've could remove the resonances with simple transversal bracing. not even as effective as longitudinal but still good enough to resolve the problem.
Things like exotic dumping with multiple type of special materials are effective too but are complex and expensive.
It's up to the builder to decide.
3D picture of my simple bracing (cheap 18mm MDF):
Without damping none of the work put into the cabinet by the driver will be dissipated and the magnitude at resonance will keep getting larger and larger until the speaker breaks. Doesn't matter how heavy or stiff without damping the potential magnitude of resonances is infinity. However, for a resonance at a particular frequency a stiff heavy cabinet will have more intertia compared to a floppy light one and so the magnitude of resonances will increase towards infinity more slowly for the same driver pumping energy into the cabinet. (Resonant frequency scales with the square root of stiffness over mass).
I do think that dampening and bracing are "opposite" and both needs to be done. We need to make two things in order to have the best enclosure.
1) bracing to make it stiff. This will increase bass response and avoid unwanted resonnance of panels that will generate unwanted sound wave inside AND OUTSIDE the cabinet.
2) Damping internal sound wave this is usually done with felt whool and the like. I never done it but a sandwich of Panel | small felt | tar plate | felt would be efficient for this.
Damping the panel with elastic braces means that the lower frequencies WILL activate resonance (like dampers in cars)
1) bracing to make it stiff. This will increase bass response and avoid unwanted resonnance of panels that will generate unwanted sound wave inside AND OUTSIDE the cabinet.
2) Damping internal sound wave this is usually done with felt whool and the like. I never done it but a sandwich of Panel | small felt | tar plate | felt would be efficient for this.
Damping the panel with elastic braces means that the lower frequencies WILL activate resonance (like dampers in cars)
This thread by Augerpro, could be an idea to look at 🤓
https://www.diyaudio.com/community/threads/a-monster-construction-methods-shootout-thread.356130/
https://www.diyaudio.com/community/threads/a-monster-construction-methods-shootout-thread.356130/
They are opposite in a material but not in the strategy of making loudspeakers. Stifness helps more in the lows, damping more in tthe highs to tame what disturb the cabinet. There are structural weakness areas too, like the gluing junction, in the stifness efficienty.
Does tension add anything here?
for instance, I used 2 layers of 3/8" bending birch ply to create a curved side surface, with a lot of wood glue between, and a lot of weight on top and a curved frame underneath to conform each side to the target curved state. in total that is 3/4", and the ply is still under tension across that curve. it seems really stiff.
for instance, I used 2 layers of 3/8" bending birch ply to create a curved side surface, with a lot of wood glue between, and a lot of weight on top and a curved frame underneath to conform each side to the target curved state. in total that is 3/4", and the ply is still under tension across that curve. it seems really stiff.
Just wading in.
Yes it does. If you can get resonances out of the bandwidth of the cabinet driver. This is easy with woofers/subwoofers (particularily if also push-push). With a midWoofer it is more difficult, but it is easy to show that above the driver’s mass corner, the HF energy falls 2nd order, and one can argue for 4th order. So if one can push any (potential) resonances high enuff that they are not excited, it is as if theydon’t exist.
This is the approach i take with FR/single driver boxes.
In a midRange/MidTweeter box one can push any (potential) box resonances below the bandwidth that works too. The BBC has a commony used approach for this. @waxx did a good example with his Alpair 10.3 midTweeter enclosures.
http://www.p10hifi.net/forum/waxx-a103-sat-220121.pdf
dave
Last edited:
A work colleaguesaid that when he was a kid, he used to roll a marble on a windowsill, it drove the neighbors mental, because their windowsills had the exact natural frequency, and so picked up the vibrations, even though the houses were detached - although I think he said that there was only inches between them. I've heard that the same effect works on the treads of stairs in terrace houses.
It might be an idea to stiffen the enclosure out of the bass driver range, then mechanically de-couple the other drivers.
Damping will, by it's very nature, to flexible to stiffen. A felt pad glued to the cabinet, with carpet tile glued to that works well, it also dampens sound reflections on the inside, and gives a double glazing effect, stopping the internal sound passing through the walls.
Found this little tidbit that looked interesting. This is a KEF speaker. They use a multimaterial solution. My guess is thats just rubber stripping. Interesting that it doesn't go across the entire brace but is only a few strips here and there
I will try something like this
I will try something like this
You might also consider how and if the panels are excited. I suppose the enclosure you show in post #1 is sealed and completely stuffed. Then the main exciting factor is probably the driver mounted (stiff or flexibly) on the baffle.
I have a bunch of weights here at work that I will be adding to the bottom of this little 2.5 cabinet. Mostly because it is top heavy and that will accacerbate the rocking from the speakers punching.
Tiles is a cool idea. I know on some brake calipers they literally put wheel weights on them and this prevents high pitched squeaking.
I will have to see if this is part of my sim package here at work. It probably is but I only ever run tensile sims. Can't be too hard to learn.
Look at the brace, it is not glued solid to the sides, but uses a flexible glue to 'eat' up vibrations. At least that's what they do in the LS50 Meta.