I've designed this ugly monstrosity. I call it The Off Axis. It uses 9 https://www.parts-express.com/pedocs/specs/297-2164--tebm65c20f-4-spec-sheet.pdf and 1 https://www.parts-express.com/pedocs/specs/Spec Sheets/292-816--grs-8sw-4he-spec-sheet.pdf
Tell me, do you think this abomination would have too many canceling waves and sound like poop, or would it be awesome paired with a KABD4100 for full room filling action?
Despite the technical challenges I'm sure it won't be like that. It's easy to say that things won't come together in an ideal way, but when you go looking for reflections there can be good in diversity. It may even offer some control benefits in the lower midrange. Perhaps some dynamic range.
Perhaps running a time aligned channel for each tier would alleviate some of the vertical combing issues at the listing level, while stationary and seated.
There are enough drivers to likely blend the constructive/destructive interactions, I didn’t think about that aspect.
The last time I checked the general book of acoustics, it very clearly mentioned that ALL sources will interfere (= combing) with each other.
It's just a matter of path length difference (= distance and angle) for a certain frequency.
I guess an argument can be made that the more random a pattern of sources is, the less this effect will be.
Or when the beaming or pattern of the source (cardioid/dipole) is so strong that they will barely overlap off-axis.
Or when a low-pass filter is being used well below the frequency of interference.
This design is quite 3 dimensional, so I have my doubts simple time alignement (I assume with delays etc) is going help all that much.
It's going to be a lot more complicated.
That's a good start yes 🙂
These systems are just not all that evident and simple to design if you want to do it really well.
Creating arrays at random is a little like comparing an open-baffle speaker to a dipole.
In any case, Don Keele was a proponent of curved three dimensional arrays..
https://www.keele-omholt-technologi...ation-of-CBT-Theory-to-Loudspeaker-Arrays.pdf
In any case, Don Keele was a proponent of curved three dimensional arrays..
https://www.keele-omholt-technologi...ation-of-CBT-Theory-to-Loudspeaker-Arrays.pdf
..and your point is?
I'm not averse to discussing the specifics if it comes to that, as Keele wasn't to entertaining flat baffle placement for various delay and shading schemes.
I'm not averse to discussing the specifics if it comes to that, as Keele wasn't to entertaining flat baffle placement for various delay and shading schemes.
I would love to watch that discussion play out. It generally seemed sufficient to apply baffle correction and round overs on the edges. I’m not at all familiar with shading. I’ve been considering trying a 1/2” felt absorption panel with speaker cutouts and covering it with grill cloth to absorb the baffle reflection, at least in the wavelengths that are involved in diffraction.
For questions like this, I'll often start as simply as possible to see if things sound as bad as simple driver models show. 2 or 3 drivers taped on a FOME-COR baffle of appropriate shape will likely give you a feel for high frequency lobing/combing issues. And if you want to change the shape, you can do so in a couple minutes.
Get the DMLs as close as possible to push combing up in frequency and if still an issue you might want to try LPing the bottom 6 DMLs
Looking at each array of 3 wire in series and add a shunt cap the bottom 2 (what i’d do).
dave
Looking at each array of 3 wire in series and add a shunt cap the bottom 2 (what i’d do).
dave
When a wavefront runs normal to the baffle it can remain uninterrupted. Using absorption can create a disconnect, causing diffraction then and potentially more as it continues unsupported. This is not unlike what happens when it encounters a cabinet roundover and the baffle falls away from under it.