Large Wood Horn Design for B&C DCX464 in 3 way active speaker

Hello all -

I've been lurking here for a long time with some moderate contributions, and have been scouring the forums trying to understand a direction for a custom speaker I'm building around the B&C DCX464. Really, it's being designed around aesthetics - I have always loved the look of big wood horns ala Oswalds Mill Audio and A for Ara. I know the two aforementioned companies build luxury items for a price point that is out of reach for almost everyone, with a sound quality that can likely be achieved for much lower; however, I'm an architect and one of my fatal flaws is that I like how things look as much as how they perform.

The design for this speaker thus started with a visual idea more than an acoustical one - I wanted a big (24" +/-) round wood horn set atop a bass cabinet down below. In looking at several options, it appeared to me the best way to achieve this was to 1. keep the crossover point relatively low (to achieve more flexibility in driver-to-driver spacing) and try to play as much of the frequency range from the horn to keep it as point-source as possible. I saw the DCX464 reviewed relatively favorably on Joseph Crowe's site, and given its broad range and low crossover possibilities, I thought that driver might be the ticket.

I use a program called "Rhino" for work, almost 8 hours a day, and have been for nearly 15 years - so there is nothing outside my capability for modeling. I am also an expert in a parametric program called "grasshopper," so I've been able to translate some of the more complex math involved in various designs into parametric models that allow for extremely fast and quick experimentation (I have built a pretty comprehensive model that allows me to iterate through tractrix horn constructs with petals). I have a 3D printer and a CNC and have been building furniture for 25 years, so there are few things outside my fabrication abilities.

Where I quickly fall apart is trying to determine what exactly the geometry of the actual horn/waveguide should be. I have read a lot of different opinions, but realistically what I'm looking for is good sound that works on a relatively wide dispersion - it doesn't have to be the same EVERYWHERE, but I'd like to avoid a full-range pin-point sweet spot if at all possible.

I've been looking at a lot of other precedents for contemporary horn drivers, including the Oblate Spheroid waveguide, Joseph Crowes ES290 horn, the Advance Transition Horn EXAR, and tractrix horns in order to try to find some commonality between all of them in the design. I have drawn all these in CAD and plotted them for comparison.

I'm also curious about the "pleats" in the X-Shape waveguide (mostly because I could fabricate and think it looks cool) - but don't really understand what it is doing acoustically. https://audiohorn.net/x-shape-horn/

Attached is the curve plot comparison, as well as a couple of early renderings of what I think the cabinet would generally look like. I was looking at the Faital Pro FR400 for the lower end. I'd be doing an active system with maybe a passive crossover for the compression driver. I'm looking for opinions - about horn design, general thoughts on the speaker design, etc. Thank you!
 

Attachments

  • Closeup.png
    Closeup.png
    2.4 MB · Views: 260
  • From Bar.png
    From Bar.png
    2.5 MB · Views: 107
  • Horn Design Comparison Chart.jpg
    Horn Design Comparison Chart.jpg
    49.7 KB · Views: 115
Where I quickly fall apart is trying to determine what exactly the geometry of the actual horn/waveguide should be. I have read a lot of different opinions, but realistically what I'm looking for is good sound that works on a relatively wide dispersion - it doesn't have to be the same EVERYWHERE, but I'd like to avoid a full-range pin-point sweet spot if at all possible.
"A relatively wide dispersion" could mean 180 degrees compared to 360 (omni).
You should determine the dispersion pattern you prefer to determine the what the geometry of the actual horn/waveguide should be.
If you already have the compression driver (or even a small "full range" speaker), you can experiment with low volume, quick iterations of various horn profiles made from foam-core or cardboard and duct tape.
The horn you depict would have what is known as a "beaming" response, -6dB response narrowing from over 120 degrees in the 500Hz range to ~40 degrees at 10kHz.
I've been looking at a lot of other precedents for contemporary horn drivers, including the Oblate Spheroid waveguide, Joseph Crowes ES290 horn, the Advance Transition Horn EXAR, and tractrix horns in order to try to find some commonality between all of them in the design. I have drawn all these in CAD and plotted them for comparison.
Time to start comparing their frequency and polar response, or just decide to use what you find most visually appealing.
I'm also curious about the "pleats" in the X-Shape waveguide (mostly because I could fabricate and think it looks cool) - but don't really understand what it is doing acoustically. https://audiohorn.net/x-shape-horn/
Mabat gives some insight:

Basically the concept is to make the diagonal polar response fall in between that of the horizontal and vertical, though with no actual comparisons of different profiles, can't see (hear) how well the "pleats" work, or how much diffraction/time smear they cause.
I'm looking for opinions - about horn design, general thoughts on the speaker design, etc.
Aside from not liking the aesthetic of the nonagonal petal horn on stalks above the pentagonal woofer cabinet, I acoustically prefer the both the low and high midrange sources to be as close to ear level as possible to avoid the disjointed effect of instrument and vocal point sources changing with the note frequency.

I also generally prefer constant directivity horns of ~90H x 40V pattern compared to beaming round (or nonagonal..) horns, and a much closer listening (and video..) presentation to increase the audio's direct to reflected ratio.

Art
 
Last edited:
  • Like
Reactions: GM
+1 The pioneers got horn systems right ~90 yrs ago, so why mess with success from an industrial engineering POV or accept that while 'sound is round' it doesn't perform all that well as art deco unless the room is modded to shape its response to whatever aspect ratio polar response is required.

Yet to audition, research any 'X' shape horns though, so maybe able to do a 'good enough' rectangular/oval polar response in a round format?