Discussion on what materials to build speakers out of

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi pietschu,
Where do you find all of this data? You are good with the research ;)

It's one of the things I'm fairly good at. Having electronic access to a University Library also helps. I sometimes go a bit far though, or so I'm told. Once got the fan curves for built in vacuum cleaners from factory engineers when I was deciding whcih brand to get, and created a spreadsheet to compare. Just after that someone posted a vacuum cleaner qustion on the companies intranet and I replied and posted the spreadsheet. Never did live that down. Still get : "Got a spreadsheet for that?"

I knew a metal skinned sandwich was stiff but I had no idea it was that much stiffer than a comparable thickness of plywood - wow!
Wheels are turning...

I also didn't realize that and those stiffness value are for thin aluminium and foam! Now to get those values the bond between the thin metal and the filler material has to be perfect and able to handle the stresses applied. Luckily for speakers the stresses are very small so the non-environmentally contact cement and some decent pressure (I once resorted to driving over the glued together panels with a wheel of the minivan, worked great) should do.


The ducting sheet metal is rather thin also, unless it is a large piece (the bigger the duct, the thicker the material).

I'll definitely look into a metal shop if I decide to go the metal composite way. Even thin ain't so bad though.
 
Last edited:
Disabled Account
Joined 2007
I like the spreadsheet story. :)

One way to bond the skins together is to use spray foam as the core - it will adhere to the skins as it is expanding and curing.
To construct this way you can start with the inside, a complete rectangular box made of pieces of flat sheet metal and epoxy the seems together. Then use 6 more sheets that are slightly bigger than the first ones and glue those onto the box with silicone, making the layer of silicone about 1/16" thick. Last build a bigger box, about 1" bigger all round than the inner box but leave the top off so that it can be filled with spray foam.

117.PNG

Fill it with spray foam, let it cure for a few days then cut off the excess from the top and glue on the metal sheet that forms the top.
Sounds like fun. :)
 
Can you define compliance?

I'm no expert but the article I found does not seem to discuss any foam properties. Maybe the types they looked at were all very similar. I think you'd want the smallest gas bubble sizes possible and the foam must be able to withstand the compressive forces without crushing. I'm guessing that for speakers this would only be an issue anywhere a screw or other fastner is driven through the composite sheet.
 
As per recommendations from here i am to increase box rigidity. I am going to use some right angle steel/ally, and some flat square section tube i have lying around. glueing with epoxy and screws, i hope should be enough.

Following the likes of Daves regime, i am planning to brace vertically along panels, an to brace every 4" or so, maybe slightly more or less.

Will this increase rigidity sufficiently? should i also brace at right angles, ie horizontally as wel as vertically?

Or... any critisisms with this method?
 
Interesting, I was reading about sandwich cores of wood/fiberglass and foam last night as used in boat hauls. I was amazed at increase in rigidity. God only knows how something like that sounds.

So much complication jeez I might as well well go IB and forget about this :) Although that baffle would have issues as well.
 
Disabled Account
Joined 2007
Here's an interesting panel building method:

117.PNG

It has 2 skins, bowed away from each other. The skins could be 1/4" hardboard and the space left between could be filled with spray foam. The advantage for this is vastly increased stiffness and the loads would be transfered and concentrated at the strongest part of the box - the corners.
No need for internal bracing.
It would make for an interesting build but it might be worth the effort.


BTW, what happened to Dave? He ran away?
:)
 
Last edited:
diyAudio Editor
Joined 2001
Paid Member
That is basically the same as I described it - my lossy medium (silicone) is the damping layer in their illustration.

I'm interested in another option: layers that are of different characteristics but glued rigidly together. Turntable plinths are often done this way.

Say, two exterior 1/8" plates of aluminum, epoxied to a center layer of 1/2" Corian

Clearly the characteristics of this are going to be different that either material alone, and "damped" due to the two materials resonant frequencies constraining each other, but maybe not as lossy as the kind with a flexible layer.

The flexible layer actually consumes energy that could be creating sound I think.
The very rigid cabinets seem to have no down side..

If you built up your tapered foam panels by clamping them on a flat table you could have the inner layer flat and easy to assemble and the outer layer bowed out
 
Last edited:
The very rigid cabinets seem to have no down side..
t

That's one of the things I am not so sure about after reading the BBC paper.
Dave has certainly had a lot of experience on the subject and thinks this is the way to go...

A few things that make this less certain are that in the paper they found that Q of the panel is frequency dependent. For example 17mm BB was a 3-4 times stiffer (Elastic Modulus along the grain) at low freq. (62.5 Hz) than it was at the 6khz.

Additionally, they found that vibrations in the lower freq. 100-500 (approx) were less audible than those in the 2-7khz. So pushing vibrations up with bracing can reduce low freq. vibration it can increase the audible vibration higher up.

Dave you say there is less energy in this region but both panel stiffness and human perception make this less clear cut.

They also did not conclude that thicker is better, Actually, they found double thickness BB 9-18mm caused more audible vibrations.

Not one of the materials however was good without a lot of damping material to lower the Q.

I think that it is very difficult to prescribe which way to go without measurements on the complete speaker. It does however look as though panel damping is useful. I know Dave you don't but I would love to see the measurements to back it up.
 
Disabled Account
Joined 2007
I'm interested in another option: layers that are of different characteristics but glued rigidly together. Turntable plinths are often done this way.

Say, two exterior 1/8" plates of aluminum, epoxied to a center layer of 1/2" Corian

If you built up your tapered foam panels by clamping them on a flat table you could have the inner layer flat and easy to assemble and the outer layer bowed out

Hi Mark,
I see that layer of Corian as being stiff enough on its own. The advantage of a sandwich is high strength, stiffness in a lower weight panel. Using a heavy, stiff core negates that advantage. There is some benefit at some frequencies due to the layering of different materials.
I thought about doing the panels flat on one side but I couldn't see any advantage to it. The flat part would need to be a stiffer (most likely thicker material to withstand the tension from the other skin, even if it was clamped in this position while the glue dried. The addition of ribs to the mix to ensure the flat side stays flat would complicate the construction quite a bit.

That's one of the things I am not so sure about after reading the BBC paper.

Not one of the materials however was good without a lot of damping material to lower the Q.

Dave says that adding damping material (specifically mass damping) lowers the resonant frequency. I don't see it this way. If the panel is stiff, you can't take that away by adding mass damping - the panel is still as stiff as it was, it just has something added to lower the Q of and resonance and help dissipate other vibrations.
 
Dave says that adding damping material (specifically mass damping) lowers the resonant frequency. I don't see it this way. If the panel is stiff, you can't take that away by adding mass damping - the panel is still as stiff as it was, it just has something added to lower the Q of and resonance and help dissipate other vibrations.

The math says Dave is correct. The equation for resonant frequency of a panel has the stiffness term divided by the mass term. For two panels to have the resonant frequency, the more heavier one has to be stiffer.
 
Disabled Account
Joined 2007
Fair enough, but...
Everything in proportion: How much mass would you need to add to drive the resonant frequency (let's call this RF from now on, ok?) down to where it is equal to where it was before bracing? For example, bracing drives the RF up from 300hz to 1000hz, how much mass would it take to bring it back down to 300hz? I believe that some effective amount of mass damping can be added without adversely effecting the RF.
 
How much mass would you need to add to drive the resonant frequency (let's call this RF from now on, ok?) down to where it is equal to where it was before bracing? For example, bracing drives the RF up from 300hz to 1000hz, how much mass would it take to bring it back down to 300hz?

OK here's a simple example (although since RF usually means radio frequency, I'll be using Fr) for 1/2" MDF (the good kind).

For a 11" by 11" panel, Fr = 308 Hz

By bracing it every 5.5" in both directions, Fr = 1233 Hz

You'd have to bring the effective density up from 769 kg/m3 to 12304 kg/m3 (16 times more mass) to get Fr=308 Hz (assuming limp mass which adds no stiffness)

Doubling the effective density results in Fr=872 Hz
 
Last edited:
frugal-phile™
Joined 2001
Paid Member
I thought about doing the panels flat on one side but I couldn't see any advantage to it.

And lots of disadvantages. You are going about it the right way. My only concerns are possibilities of the foam having its own parasitic resonances. The only way you are going to find out is to build it. I am following that thread with interest.

dave
 
Disabled Account
Joined 2007
Adding mass without adding stiffness lowers the panel resonance. That is straight text book.

And by lowering the Q of the resonance you make it more audible (Toole) and more likely to be excited.

dave

Back up a bit here - I said "add stiffness then add mass".

Increasing stiffness drives the RF up, damping lowers the Q AND shortens the overall decay time for all vibration, not just at the RF.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.