76.22.120.252
In Reply to: RE: baffle shape for horns posted by hollowboy on June 10, 2008 at 17:30:39
It's a pretty well documented effect. Diffraction from the edges of the horn mouth create not only an axial resonance, but also a "wobble" in the directivity.
For normal horns (non-conical and not CD types) the directivity narrows with increasing frequency, and once it is significantly narrower than the horn flare at the mouth, the acoustic wave might be said to not feel the mouth - it has detached from the horn walls, so to speak. So the baffle/edge effect is confined to frequencies near the horn and/or mouth cutoff, and with a suitable crossover its effects can often be minimized.
Non-circular horns produce their own weirdness in directivity vs. frequency, as the higher-order circumferential modes interfere with each other. I think Earl Geddes makes this point in his writings fairly clearly. Avoiding the circular shape problems produces different problems rather than solving the original one. (Actually the same is true of point-source direct radiators, though that problem is not widely appreciated.)
Follow Ups: