Kosst_amojan wrote: "There’s all kinds of ways of managing edge diffraction! On narrow speakers the time delay is virtually insignificant and it’s really only a problem in the treble region... It’s just technically convenient to drive the diffraction problem up the spectrum because higher frequencies are easier to manage. "
I think you have a misconception. It sounds like you think diffraction can be limited to high frequencies by using a narrow baffle. This is not the case. What a narrow baffle does is, it reduces the time delay between the arrival of the non-diffracted direct sound and the arrival of the diffracted sound.
(One thing a narrow baffle does do is, it raises the "baffle step" frequency, which is the frequency at which the baffle’s face no longer acts like a 180-degree horn. Perhaps this is what you were thinking of?)
I think a brief explanation of how the time delay of a diffracted signal impacts imaging is called for. Apologies in advance for getting technical here; to anyone who dislikes technical discussion, please avoid the next paragraph:
The ear derives directional cues primarily from the first .68 milliseconds of a signal. This is the time it takes for sound to travel about nine inches, and correlates to the distance around the head from one ear to the other. Diffraction or reflections arriving within that first .68 milliseconds tend to degrade the imaging because the ear gets a false secondary early-arrival cue that normally would correlate with sound that had reached one ear first and then travelled around the head to the other ear, diffraction’s time delay mimicking the arrival at the second ear. A small time delay (narrow baffle) would correspond to a smaller false angle for this false cue, while a larger time delay (wide baffle) would correspond to a larger false angle. So with a narrow baffle, the false cues are not as drastic. This is why, all else being equal, a narrow baffle generally has better imaging than a wider baffle. If the baffle is wide enough that NO reflections occur within that first .68 milliseconds, then the imaging should be excellent (which is what happens with precisely flush-mounted studio main monitors).
Diffraction has other negative effects which are beyond the scope of this post.
Kosst also said: "There are simple solutions to that problem too. Lenses, damping, horns of some sort."
I don’t know what you mean by "lenses" in this context.
Damping material has virtually no ability to absorb a sound wave travelling parallel to its surface; the sound wave has to strike the damping material at an angle in order to be absorbed by it. And the damping material has to be thick enough relative to those frequencies to absorb them effectively. If we want effective absorption the sound must strike the damping material at an angle, and if we want absorption down low enough in frequency then the damping material must be fairly thick in both width and depth. A thick felt donut can help in the highs, but isn't going to do much in the mids.
Assuming the horn itself is not a source of diffraction (most are), in order for a horn to have good radiation pattern control down low enough to usefully minimize cabinet edge diffraction, it must be fairly large... and now we’re back to having a wide baffle again, especially once we factor in the fairly large-radius lips the horn will need to avoid diffraction at its mouth. This is actually the technique that I use, but the result is not a narrow baffle.
Of the techniques you mention, aggressive use of damping material (combined with a small enough midrange driver to get enough damping material between edge of driver and edge of enclosure) sounds to me like the most promising for effectively minimizing diffraction in a relatively narrow-baffle speaker. I do not recall ever seeing this approach on any narrow tower speaker, probably because it would run counter to the primary purpose for using the narrow tower format in the first place: Aesthetic appeal.
Duke
