In search of (non-coaxial) speakers with wide vertical dispersion

[carlsnor]

I recently posted a thread discussing why I hesitate to spend heavily on non-coaxial speakers. One of my main gripes is their often poor vertical dispersion. This got me curious: which non-coaxial designs actually offer the widest vertical dispersion?

If we look at good coaxial implementations (like those from Mofi, Genelec, or KEF), they often maintain a 60° or even 100° vertical sweet spot with no more than a 5 dB suckout (excluding the top of the treble, where narrowing dispersion is common anyway). Based on that, I decided to set a new benchmark for non-coaxial speakers: a minimum 60° vertical window with a maximum -5 dB drop (excluding the highest treble frequencies).

In my research, I’ve found a few design methods that help achieve this:

• Lower Crossover Points: Pushing the crossover frequency lower widens the vertical sweet spot. (Ascilab is a notable recent example of this, though spoiler alert: they still don't quite meet my criteria).
• Smaller Dedicated Midrange Drivers: Using a smaller midrange driver—or minimizing the size difference between drivers—widens the vertical window.
• Overlapping/Integrated Waveguides: Blending or physically overlapping the waveguide over the mid/woofer expands the vertical sweet spot. A great example of this approach is the Perlisten A-Series. While I haven't seen Spinorama data for them yet, I suspect they’d pass the test, especially since the Meyer Sound ULTRA-X22 (which uses a similar overlapping layout) misses my criteria by just a hair.

So far, here are the only non-coaxial speakers I’ve found that actually meet my requirements:

• Monoprice (Monolith THX series): Though the upper treble does get a bit narrow.
• Topping MA4: This is probably the best example I’ve found.(although the measurement is made by the manufacturer themself) It maintains an impressive 60-70° sweet spot, likely because it utilizes almost every design method listed above without going to the physical extremes of the Perlisten A-Series.

I'm eager to find more. What other design methods or real-world speaker examples did I miss?

Edit: If we are being a bit forgiving, the Neumann KH 80 almost qualifies; it only dips slightly past the -5 dB mark (reaching about -6 dB) briefly around 2 kHz and 7 kHz. We see a similar behavior with the Ascilab F6B(s), which briefly dips at 1.2 kHz and 1.5 kHz. The Ascilab A6B and C6B perform slightly worse, the Ascilab C5B performs the best out of that bunch, because of the smaller woofer?

 

[AM88]

From what Iunderstand the distance between the tweeter and the other driver, if it is a 2 way helps, not just the driver size. I have been looking for exactly what you are talking about and have found that 4 inch driver or 5 if pushing it give the best results. The ascend acoutics htm 200se2 does quite well and is at the border of what you want but has slightly asymmetrical horizontal directivity, though Dave F. says that is not as bad as it first seems. Good luck in finding more, I will be watching this thread closely.

 

[dualazmak]

Even though this post could/would be a little bit out of the scope of OP @carlsnor, just for your possible reference and interest...

In my PC-DSP-based multichannel multi-SP-driver multi-amplifier fully active audio setup, I intentionally physically "place" tweeter and super-tweeter unusually vertically "apart" while both of them sing quite overlapped Fq zone, and the narrow-directivity highly-efficient metal-horn super-tweeter has unique reverse-reflective sound dispersion material (actually hard-heavy-thick random-surface crystalline cutglass bowl).

If you would be interested in such unique "vertical dispersion approach" for high-Fq sound, please visit my post #931 on my project thread for the details.

Spoiler