At the time I was testing these I had a few waveguides stacked about and grabbed a couple to look at how the tweeter fit placed on them. To my surprise it seemed to fit quite well on the standard 1" bolt-on opening. After seeing this I modeled a little adapter/bracket and 3d printed it to try proper mounting to one of these waveguides. The SEOS-66 looked like the best fit as the throat is fairly shallow and I don't expect to use this with anything larger then a 6.5" woofer. IIRC the SEOS-66 is either 70x70 or 80x80 dispersion.

After a couple attempts I had a nice fitting adapter made up which mounts directly to the waveguides in the 2-bolt fashion use a pair of m6 bolts. I made sure to add a beveled ring on the front which fits over the outer half of the surround of the dome filling in the empty space there which helps minimize the reflection and interference response that would occur in the high frequencies if the space had been left open.

The downside at least with this version of the adapter is that it does not make an airtight seal against the waveguide mounting flange. It could be glued or sealed with some kind of adhesive but I do intend to modify the design of the adapter to beef it up adding a space for installing an o-ring or small gasket to make the seal airtight.

The Measurements

This is probably what you clicked on the link looking for.

Top is the response on axis, no smoothing or gating (why it's kind of noisy). Below are a series of off axis plots starting at 0 and measured at every 10 degrees ending at 90. It gets kind of weird past 70 degrees because I did not have the waveguide mounted in a baffle. I also believe the edge diffraction of the unmounted waveguide may contribute to the small ripples you can see in the response starting at 3k.

But overall not bad at all, very smooth with just a small bump around 10k and some lift above 18k. Off axis behavior is excellent directivity control is good down to 1600-1700 Hz with very even coverage and only very slight narrowing of the pattern up until about 18kHz.

Normalized and Polar plot:

Distortion

Distortion is also quite good in the waveguide, I feel like it could be usable down to 1700-1800hz, maybe around 2k if using it for a higher output design. I didn't crank it up too high but I did do a measurement at 100dB at ~1/2 meter and 3rd order distortion performance was better then expected measuring a hair above 0.1% at the low end. 2nd order harmonics start to rise as they approach and take off below 2k but remember the response shape also peaks around 2k. When the response is flattened and with a crossover in place I expect distortion will actually be lowest in that range.