Well, as we mourn the death of non-axisymetric Ath, I have a question..
Companies like Revel have shown that the cabinet is very much a part of wave control as is the 'actual' waveguide. Could it be argued that a rectangular WG better prepares the wave to wrap around a rectangular cabinet?
Companies like Revel have shown that the cabinet is very much a part of wave control as is the 'actual' waveguide. Could it be argued that a rectangular WG better prepares the wave to wrap around a rectangular cabinet?
Well the non-axisymmetric Ath is not dead, it's that only I can use it 
I would say that what counts the most in the end is really the round-over (the radius). Whether is the waveguide rectangular or circular won't make a dramatic difference, IMO. The smoother the roundover, the better.
Indeed, it seems better to have the horn "shadowed" by (an array of) corner mounted midwoofers.
It works! So I don't need the Gmsh SDK anymore. All the tool does now is that it writes *.geo scripts that can be converted to *.MSH (or whatever) with gmsh command (either manually or by calling gmsg directly from Ath - that shouldn't violate anything).
You saved the day
No intrusive big loudspeaker system in the room?
The room partly integrated in the loudspeaker system (or the other way around) > the walls as horn extension in the horizontal plane.
However, I think much depends on the shape and size of the room as well as the interior in general (objects).
The BD Orelo system shown in the picture works wonderfully, but that's an OB with 'wings'.
Such a setup needs dsp.
Last edited:
What I have in mind is something like this sketch (if it attached). A shallow rectangular-ish box with the tweeter in one corner and a MEH style woofer. The tweeter needs to be flush with the corner. If we choose the direction it is facing as 0 degrees then the HF horizontal polars should approximate a tweeter twice as long and high (with the walls removed).
The opposite edges of the box could be blended into the ceiling/floor to minimise diffraction. That would be the compromise to get enough volume in the box for the woofer.
Lots of 90 degree angles so not so much in common with the build of other waveguides - but based on the same principle of minimising diffraction, I would argue.
Attachments
Last edited by a moderator:
This picture of a test with a cardboard box might be better illustration.
If the tweeter is flush with the corner then its acoustic mirrors from the walls constraining its radiation will extend its effective size. I'm more interested in the effect that this has on the radiation pattern than the gain - so I was a little sloppy in referring to that as reinforcement.
In practice you won't be able to get it completely flush, so the tweeter's acoustic mirrors will behave more like an array. You need the tweeter to be small if you don't want a narrow radiation pattern. Said another way, you want the acoustic mirrors to be within a quarter wavelength through the operating range.
A false wall would be ideal for this, but a compromise is to make the dimensions of the speaker on the face containing the drivers large enough that the baffle step/taper to the wall is relatively small. This is the equivalent of the mouth of this when considered as a waveguide.
Unsure if that makes more sense. Pictures are probably better than words in this case.
If the tweeter is flush with the corner then its acoustic mirrors from the walls constraining its radiation will extend its effective size. I'm more interested in the effect that this has on the radiation pattern than the gain - so I was a little sloppy in referring to that as reinforcement.
In practice you won't be able to get it completely flush, so the tweeter's acoustic mirrors will behave more like an array. You need the tweeter to be small if you don't want a narrow radiation pattern. Said another way, you want the acoustic mirrors to be within a quarter wavelength through the operating range.
A false wall would be ideal for this, but a compromise is to make the dimensions of the speaker on the face containing the drivers large enough that the baffle step/taper to the wall is relatively small. This is the equivalent of the mouth of this when considered as a waveguide.
Unsure if that makes more sense. Pictures are probably better than words in this case.
Attachments
Much clearer, thanks. Ideally, if you can manage to do it right, this radiation would be heading for a quarter of a hemisphere. As it stands I might guess that the higher frequencies would be more reluctant to travel over your set square.
I'd question comparing these methods as they are probably quite different. The upfiring philosophy has been more omni, with diffusion, or a diverse mix of reflections wanted.