[JohnR]

All other things being equal, would it be better to construct the bracing in a cabinet so that the unbraced sections of panel are:

a) 12" x 12", or
b) 8" x 24"

?

[jules]

Hi John,

Rythmik seem to think internal bracing is best done along the length of the cabinet rather than across it. Here's their 12" plan:

http://www.rythmikaudio.com/download/DS1200s_sub_plan.pdf

The 15" driver plan is slightly different though probably helpful.

jules

[BPT]

John:
Go with the 12" x12". Many years ago I was into speaker building with a group of guys that were pretty fanatical (at the time, there weren't (m)any speakers being built with extensive bracing and 1.5"--3" thick walls). After much trial and error experimentation, they concluded not to have any panel area larger than 6" without bracing when using using 3/4" MDF and nothing longer than 12" for double layer MDF (1 1/2" MDF). This produced some of the least colored sound, I've experienced.
Chris H.

[JohnR]

they concluded not to have any panel area larger than 6" without bracing

Well, a panel has two dimensions... wouldn't the short one be the important one?

[JohnR]

Rythmik seem to think internal bracing is best done along the length of the cabinet rather than across it. Here's their 12" plan:

http://www.rythmikaudio.com/download/DS1200s_sub_plan.pdf

The 15" driver plan is slightly different though probably helpful.

Thanks jules - took me a while to visualize the 15" one...!

I'm thinking of a bit of an experiment in trading off wall thickness for bracing.

[Steidl Guitars]

A few (hopefully relevant) thoughts about bracing.

For acoustic guitars, bracing the top performs two functions:  mechanical strength to resist the pull of the strings and resonance control.  Assuming that two different bracing designs provide the same mechanical strength, the designs will likely differ in their fundamental resonant frequencies.

Of course, bracing on loudspeakers and guitars perform different functions. 

For acoustic guitar bodies, we want them to vibrate relatively evenly across the range of frequencies they are designed to create, so we brace them so that no single frequency is overly dominant. 

For loudspeakers, we want then to be inert, so I'd think we'd want to brace them so that their resonant frequencies are as far as possible from the frequencies they are trying to reproduce. 

With subs, I'd think that would mean to brace so that any resonant frequencies would be as high as possible, which means to create more rather than fewer nodal points on the box (points where the box is in contact with a brace) -- that means minimizing locations on the box where there is no brace.  If that logic is correct, something like a honeycomb or lattice might be ideal, at least so far as resonances.  There are likely issues with air movement too, which this doesn't consider. 

That means I agree with Chris above:  I'd guess that 12 x 12 would likely be preferred.

That said, with such long wavelengths, I bet it matters very little in the end...

To see how bracing affects vibrational models, you might have a look at these photos of Chladni patterns:

http://www.phys.unsw.edu.au/music/guitar/patterns.html

And I'd bet that guitar builders are even crazier than loudspeaker builders

[planet10]

One should divide panels such that the aspect ratio of the subpanels is greater than the panel being braced.
The subpanels should not have the same aspect ratio (ie the centre of a panel is not a good place to put a brace.

The Rythmk looks almost like something i would design, except that i'd have the bracing just off centre (brace edge on centre line)

It is unclear whether John's unbraced panel is 12x24 (actually 24 3/4 assumming 3/4" material) or 16x24. Assumming the first, i'd end up with subpanels 6x24.75 and 5.25x24.75.

I would also, never, ever use MDF.

dave

[jules]

John,

I've tried external bracing, in conjunction with internal bracing on mine. The external bracing doesn't mess with activity inside the box and it gives a rather different visual result. I went lengthwise inside and circumferential outside. It's certainly different though it also has some logic going for it.

jules

ps agree with planet 7 on making things slightly off centre.

[Mr Content]

John I agree that the sub panels should be different sizes, and therefore have different resonate frequencies. I always place them "off centre"  and depending on the depth of the sub I may go with 2 sets of bracing. I believe that the box should be as inert as we can make it, I use 3 different substrates with dissimilar resonances glued with a rubber floor tile glue, a finished thickness of 48mm, with internal bracing. Wrapping another layer on a box is not that hard. 

Mr C

[srb]

I would also, never, ever use MDF.

What's the best material?  Baltic Birch plywood?  Bamboo plywood (Cali Bamboo, Plyboo)?
 
Steve
 
 

[planet10]

ps agree with planet 7 on making things slightly off centre.

Planet 7 would be Neptune... i'm a bit further out than that

dave

[planet10]

What's the best material?  Baltic Birch plywood?  Bamboo plywood (Cali Bamboo, Plyboo)?

If you are talking about a single layer of a relatively affordable wood material, yes some sort of high count, void free multiply, of which baltic birch is a good, cheap choice.

Bamboo, really a grass, is a fantastic material (you want to stick with the the 3-layer with a strand external sheath) for building speakers. More care needs to be taken, and it is not cheap (we just bought 2 4x6 sheets for a couple projects, the same cash would buy close to a lift of the 5x5 utility grade BB we usually use)



There are quite a few suppliers of bamboo ply, Cali & Plyboo 2 of the bigger ones.

dave

[ctviggen]

One should divide panels such that the aspect ratio of the subpanels is greater than the panel being braced.
The subpanels should not have the same aspect ratio (ie the centre of a panel is not a good place to put a brace.

The Rythmk looks almost like something i would design, except that i'd have the bracing just off centre (brace edge on centre line)

It is unclear whether John's unbraced panel is 12x24 (actually 24 3/4 assumming 3/4" material) or 16x24. Assumming the first, i'd end up with subpanels 6x24.75 and 5.25x24.75.

I would also, never, ever use MDF.

dave

Why no MDF?  I think every speaker/sub I've ever owned has been MDF. 

[NagysAudio]

The brace should go between the top panel of the subwoofer and the bottom panel.

[JohnR]

That said, with such long wavelengths, I bet it matters very little in the end...

To see how bracing affects vibrational models, you might have a look at these photos of Chladni patterns:

http://www.phys.unsw.edu.au/music/guitar/patterns.html

And I'd bet that guitar builders are even crazier than loudspeaker builders

Hey, that's cool! I followed a link there to a violin one as well. Someone who was really serious about cabinet vibration could use that - idea being not to have any modes I suppose. However as you suggest for a sub frequencies will be below the resonance modes - one hopes! I think it's more a question of acoustic loss through panel flexing.

[JohnR]

One should divide panels such that the aspect ratio of the subpanels is greater than the panel being braced.
The subpanels should not have the same aspect ratio (ie the centre of a panel is not a good place to put a brace.

Cool - I'm going that way.

It is unclear whether John's unbraced panel is 12x24 (actually 24 3/4 assumming 3/4" material) or 16x24. Assumming the first, i'd end up with subpanels 6x24.75 and 5.25x24.75.

Oh, I should have said, it was a hypothetical 24x24 panel. Then with two bracing panels the choices were to divide it into 12x12 or 8x24 panels (more or less). It was a thought experiment into "optimum" bracing usage i.e. given a certain amount of bracing how best to place it.

[JohnR]

I believe that the box should be as inert as we can make it, I use 3 different substrates with dissimilar resonances glued with a rubber floor tile glue, a finished thickness of 48mm, with internal bracing. Wrapping another layer on a box is not that hard. 

Hey Nig - what if I ever move, how am I going to get it down the stairs?!

Actually I'm going to try an experiment in light-but-stiff. I can get marine ply pretty easily locally. The bamboo idea sounds interesting, I see there's a place in Queensland that makes bamboo ply and panels, perhaps something for a future project.

I like jules' idea of external bracing - could get quite creative that way (fins?)

[planet10]

Actually I'm going to try an experiment in light-but-stiff.

Both lighter, and stiffer push the panel resonance up. As does panel thickness, and shorter panel spans.

dave

[planet10]

Why no MDF?  I think every speaker/sub I've ever owned has been MDF.

That is a whole 'nuther long thread... http://www.diyaudio.com/forums/construction-tips/98834-discussion-what-materials-build-speakers-out.html

Just let me say that, IMHO, MDF is not a suitable material to build speakers out of. It's only redeeming quality is that it is cheap.

dave

[dvenardos]

Sandbox anyone? 
http://www.audiocircle.com/index.php?topic=53675.0