Originally Posted by Gouie
Very interesting concept. Is there a recommended distance between the pieces of bracing (in your diagram, left hand side, distance from braces where they meet the side panel)? What about the recommended distance between full brace panels themselves - for example, on a larger 4 ft long box, how many brace panels would be required?
I haven't checked on this forum for a year or so (building snowmobile and writing a book). But I thought I'd answer you although it may be long since you decided. You can always add more bracing if you determine the need.
The maximum distance between bracing points depends primarily on four things:
-Panel thickness and panel material (ie how stiff the panels are).
-Upper frequency range of the enclosure.
-Volume you will be pumping out (decibels).
-Panel sizes (larger speakers with larger panels require more bracing).
lets take the worst case that require a ton of bracing:
An 80 liter speaker with 15" woofer that is expected to run between 100 and 1500hz, pumping out 130db at 1 meter with full power (1700 watts). To achieve this decibel output one needs minimal use of stuffing to maintain enclosure efficiency near port tune. To use very little stuff we need to use high strength panels of high density particle-board (the kind used for floors, when you tap them they ring like hardwood and they are extremely strong where they are glued together), that is 22-25mm thick (25mm is one inch).
With these requirements and this panel thickness and strength, I would not have more than about 3-4 inches between bracing points. Resulting in a difficult enclosure to brace properly, but it is possible, here's the woofer section:
Note that I had to use some cross-bracing (white long pieces), which is inefficient, and problematic because you have to brace the braces, the cross braces now having surfaces more than 3-4 inches from bracing points. If I ran a resonance simulation you'd see the white cross-braces flop about like week old celery without something tying them together. Even the dark red/orange braces will need one brace running across them diagonally, but that can be a very thin brace. Or else we get this issue:
Instead we want this:
If I had the option I would have the rear panel be the opening and then brace the internal braces directly to the basket and magnet assembly of the speaker itself. The basket and magnet is the most solid piece in the entire speaker, its a shame not to use it for strength, but for this particular design I wanted the option to quickly change blown elements.
In this speaker the noise from the enclosure panels moving would be about 30db below the output of the speaker, at 1000hz. Lower db on lower frequencies. So that's a 30 db signal to noise ratio. Remember that the noise of the speaker itself will be reproduced by the speaker panels as well, 30db below the speaker output. And the noise from the panels not only comes off the external surfaces, but the internal ones, and hits the speaker surface internally, so it is reproduced to some degree by the speaker itself as well, especially where the distances coincide with certain frequencies. With this particular speaker the depth is about two feet if I recall correctly, and the speed of sound is just over 1000 feet per second so if this enclosure was just one foot deep it would need extra stuffing at the rear regardless of how strong the bracing is. It still likely needs a thin layer of stuffing at the rear panel since any frequency at about 1200hz or there about that lasts for more than two cycles will add to itself when it is reflected back.
Which is why REALLY expensive speakers are braced beyond belief or made by extremely solid materials like really thick polyester and glassfiber composites.
Compare the 30 db signal to noise ratio of the enclosure to the signal to noise ratio of the amplifier which is about 100db, which is why I basically ignore that spec on amplifiers, all that matters is that the amplifier can deliver the required wattage with better than or equal to 30db signal to noise ratio, in this case.
At 1000hz distortion is very audible to the human ear, so chances are if you slap together a system with any intention of going above 100db at those sorts of frequencies then it will probably be very unpleasant to listen to it without a very over-engineered bracing scheme. But if the bracing is well engineered, it could be possible to listen to the system at enormous volume without it being unpleasant. If the rear panel is slanted or split up into more than 1 face, you improve the reflective noise internally.
Too many invest in extremely expensive speaker elements and then spend no more time or effort to think through the bracing (window bracing, ugh, so inefficient, so hard to make, so difficult to scale beyond a couple window braces).
In a subwoofer on a very low frequency however, you can get away with murder, practically. But you quickly find out if you have enough bracing or not, by listening to the enclosure before putting stuffing in it. And feeling the panels. Any significant movement in the panels is going to make the bass sound washy and overbearing, rather than tight and controlled. Basically my rule of thumb with subwoofers for low frequencies is that I want to not be able to destroy the enclosure with a sledgehammer before giving up, so I make the enclosure and then brace it sufficiently with diagonal bracing so that I'd give up before the enclosure did. The bracing is also made from offcuts so its cheaper to brace than stuff it full of stuffing, just remember to glue (and MDF requires some screws or staples, because MDF is weak).