And how do you suppose I would integrate a super-tweeter? Drill a hole and friction fit?

Also, first order crossovers FTW. Easier to integrate and a more coherent sound. I'm a believer.

Drill hole. Nail / screw in retention bracket to keep tweeter cylinder in. Wire tie to keep in place.

First order crossovers good for certain drivers like Fostex. Bad for metal drivers which ring like below - where fourth order is almost necessitated. That's why these metal cone driver speakers you see at the audio shows sound flat and lack realism. Too many big caps and coils in the way.

Sunday morning EQ session:

Here's the response again un-EQed. Blue curve is response with mid and woofer tuned to match sensitivity. Obviously last two octaves are shelved. Teal curve is tuned to bring the treble up in level at the expensive of having a mid and upper-mid emphasis:



PEQ settings were arrived at with pink noise at listening position. Curve goes from teal to purple. Aiming for ~5dB slope:





This is the difference between leaning back in the sofa and standing up. Note different smoothing here to highlight differences. Not too shabby though:

Update:

I've rearranged the room so that my speakers are now on the right side. Some things have improved, others have become worse. I seem to be getting better overall tonal balance at the expense of some room mode problems. A general muddiness really. Still tweaking. I'm also "biamping" now; that is I have my Parasound running the FE166 full-range and the Crest running the 15A with a 2nd order passive filter in-line.

Here's FR, no EQ.





Here's the ETC. It's actually much cleaner past 20ms, but still got some stuff hanging around in that critical region (2.9ms, 6.7ms, 8ms). They also seem to be higher gain now. Likely from the new sidewall that is much larger and more reflective. More investigation needed into this.



Spectrogram shows less energy being stored at that 35Hz mode (previously the length mode) but more energy elsewhere. Definitely less clean here. 75Hz mode is the harmonic of the new axial width mode. Need to measure outside for good reference.

lol, well it's more muddy in this new position