Go 4x12 each box 

Keen wrote: Perhaps consider the Ciare 12.00SW, B&C 12BG100/12FG100 or the Celestion FTR12-4080DL, Because these designs will allow the use of a much larger chamber than the LAB SUB, the Lab12 driver is not required. Preference would lye with Ciare or Celestion and make use of the available xmax.

PTSD wrote: Keen wrote: build em like this, easier to load, more chamber etc https://imgur.com/a/1fYwVdm - https://imgur.com/a/1fYwVdm Ah yea true that you would have a bigger chamber altho ya would also have a smaller horn path right needing to make the speaker a bit larger to compensate ?

 

PTSD wrote: Ok Sh**t... Wrote a rly long detailed response and reload the page and all is gone... But happy to tell more this time a bit shorter haha Dutch based and I build as a hobby a Soundsystem with my friends/crew. Currently still at a early stage. Tops: JBL HLA 4895 x4 so currently run 2 pairs in stereo but planning to get more potentially in the future. Amps are all Crest CA but also a few Crest pro series (old once). DSP are London soundweb Blu series but also got a Xilica XP4080 and a XTA448 not yet fully decided what to use in the filed but tend towards London soundweb atm. FOH mixer: Midas U16 System mainly used for electronic music anywhere from slow to very fast (BPM in the hundreds) both low and high distortion music. Subwoofer section is atm still fully missing which is why I am asking here because I was very happy when I heard the old bDEAP subwoofer Tom designed before the BC series. The combination of High sensitivity and directivity without the need to use cardiod... Makes me especially intrested into this design besides the fact that I found the sound characteristic the last time I heard them very convincing them being able to play quite high having for my oppinion a nice transient response and "kick/punch" characteristic (ofcause not unique but still). And the distortion characteristics if I remember where also rather nice not being high especially when crossover the subwoofer somewhat higher. So obviously I can't afford due to it being a hobby besides UNI to buy real BC subs and bDEAPs being not available atall in Europe knowing only a single person who owns them personally. So because of that I look for plans simmilar to these designs. I like the fact that the Bdeap used 12 inch drivers with the only negative aspect being that they are not designed to be layed on their sides like the BC215. (Because the JBL HLA are quite big so I need a bit base to place them on) So anyways I hoped that maybe here someone was able to design a subwoofer of simmilar character that he might be willing to share with me because my personal simulation skills in AKABAK... Are still quite limited and I wouldn't know how to model and design subs like these. Hope this answers some of ya questions if not... please feel free to ask.

Yes but the loss of this first fold is much less significant in these external boundary designs because your horn path is extending outside the box. Also consider that the box is already huge and you’ll only need 2. May as well make them 1220x1220x810 in which case the horn path ends up being something like 3m even more I think. It‘s the least of your worries with this approach.

Based on the patent image, and the manufacturer’s specifications, the “horn” path length is approximately 1.5m. How did you determine that the effective length would be 3m? Can you provide a link or some peer reviewed paper to back up your claim?

The box may be huge in relation to a simple bass reflex, but given that the wavelength at 100Hz is 3.4m, in acoustic terms it is quite small. Most of the boundary effect will be due to the ground plane, something that any cabinet placed on the floor will benefit from. At bass frequencies the cabinet provides insignificant loading.

As I understand it the main point of this approach is that when two boxes are  together with their large side facing forwards (in other words placed on their edge) and the relatively small openings placed together in the centre on the ground, the large outside front face of the combined boxes becomes part of the horn. Therefore it seems appropriate to simulate the arrangement by including this extra section. Effectively the full length from the ground to the top of the box. The effective horn mouth also dramatically increases and imo should be simulated as such.  In the case of the patent image it’s 1.5m inside the box then the extra 1m for the external boundary. As I suggested building a design that’s 1220x1220x810 (to make efficient use of ply wood dimensions) which is a bigger physical box than the box in the patent, then horn path in this case would extend beyond the 3m mark.

The physical opening in the cabinet is too small for this design to be considered as a horn. For the radiation impedance to be to be nearly resistive the effective circumference needs to be greater than one wavelength of the lowest frequency reproduced. A square mouth can be considered to have a circumference equal to a circle of the same area. It is difficult to achieve this in a practical horn which has to be moved, hence the use of multiple horns.

At low frequencies, this is more like a tuned pipe, and the end correction is usually taken as 0.6r, where r is the radius of the pipe. Where the end of the pipe is flanged (as with this design) the end correction goes up to 0.8r At best the the effective length is about 0.25m longer than the physical length.

You might also consider reading up on flare rates and cutoff frequencies. If you put a drive unit in the middle of a 2m square board, you wouldn't model it as a horn with length 1m.