The tweeters response is pretty standard, a rising response towards the low end, coupled with a bit of a hump. This is due to the waveguide loading as well as the possibility of a high Q bump. Neo tweeters often have a high Q bump on their low end due to insufficient space beneath the dome/surround for a more optimum alignment. A tweeter is basically a tiny sealed box loudspeaker. A high Q alignment isn't going to make the tweeter sound worse in comparison, it simply requires more xover work to hammer it flat.
The woofer has a nice flat extended response save from the diffraction bump at 1700Hz.
Given that the tweeter isn't particularly well extended towards the low frequencies it would definitely make sense to measure its distortion. This is something that I did but unfortunately I didn't save. Needless to say the tweeter is perfectly capable, but from memory this was only down till around 2.5kHz. Certainly not one of the better neo domes out there but usable and due to the waveguide loading should really be crossed over at around 3-4kHz for a directivity match to a suitable woofer or midrange.
Unsurprisingly that's exactly what I did here...
...placing the xover at just over 3kHz. There is considerable asymmetry here that was necessary, in part to bring the drivers into correct phase alighment, but also because I was recycling parts from my parts bin and had to make do with the component values that I had. For all its issues the xover works very well.
As noted solving gross linearity issues was not a target of this xover and a notch filter would most certainly be required on the woofer, in a fully passive loudspeaker, as would more attenuation on the tweeter. This was left deliberately hot so that the DSP would only be required to attenuate should the tonal balance need to be adjusted.
When it comes to the EQ a starting point is necessary and the frequency response below highlights the frequency response of the loudspeaker with it in place.
A slight smiley faced frequency response was the target. This usually proves successful in providing a decent amount of air, for excitement in applications where the speakers would be listened to off axis with the raised lower end giving warmth to the sound and a listen all day, non fatiguing, sound quality. Maybe not the utmost in neutrality but certainly enjoyable for extended periods of time, even when tired and possibly with a bit of a headache! (Soldering remember!)
The crossovers, transfer functions and impedance are shown below.