The coax effect should be most dramatic up high, so a tweeter/mid-coaxial would make more sense.
Yes, I essentially agree with you.
You would please let me having a little bit off-topic issue. In case if we would have slight to considerable offset between tweeter/super-tweeter and midrange, to which our ears and brain would be more sensitive between right-left offset and up-down offset?
I assume you would agree with my thoughts that our ears are located left and right of our head, and hence we are rather sensitive for left-right sound image/allocation. On the other hand, our ears and brain would be relatively gullible with up-and-down sound staging, if both of the drivers are completely time-aligned (phase-matched). I recently discussed this
here in terms of extent of "speaker disappearance".
Just for your reference, as I shared
here, my tweeter and super-tweeter (singing together over 8,200 Hz, but tweeter tends to drop over 14 kHz; see
here) are intentionally allocated uniquely far away up and down; I have my super-tweeter under the woofer which is/was found to be the best (magical for me) position through my intensive evaluation of 8 different positions around the main SP cabinet. In that physical alignment, midrange and woofer are sandwiched by tweeter and super-tweeter from up and down, respectively. All the SP drivers are fully time-aligned in 0.1 msec precision as shared/summarized
here and
here.
Very interestingly, at least in my fully time-aligned (0.1 msec precision) system, this positioning of tweeter (up) plus super-tweeter (down) (singing together at 8,200 Hz to 20,000 Hz) enables very nice "wider/natural" pseud-coaxial (or quasi-coaxial) sound image/stage with almost complete SP disappearance. I may express it would be "
planar coaxial sound image" rather than "
point coaxial sound image".
Maybe, I am going out of the topic of this thread which is "pure coaxial PA SPs". If needed I will continue this discussion on
my project thread.