Ultra-steep crossovers (like 100dB/oct) - smart or crazy?

[kemmler3D]

creates alot of odd sounding artifacts, sort of like a selective on/off switching of specific frequencies

I can see how this would be a problem unless each driver basically sounds indistinguishable at the xover point. So you'll want each driver to be very well behaved with very low distortion at the xover, or else you'll be able to pass the ABX test no problem... and they'll need to be very close together so the apparent source doesn't jump around.

I'm thinking of putting 2x Purifi ptt4-0x04-nac-03 and 1x Bliesma T34A basically touching each other, crossed around 1500-1700 (maybe 2khz if I change my mind about waveguide) or so, so I think the steep xover is worth trying. But, I am not sure if it will have any benefit over an LR4. We shall see...

 

[Juhazi]

Perhaps someone could make a sim with Vituixcad for a 3-way with steep filters. Horizontal directivity matching should be rather easy, but I'd like to see how vertical off-axis response and impulse behave. Then time alignment gets broken right away (unless it is coaxial)! And it would be interesting to listen to this speaker (also off-axis) making a slow sine sweep and playing piano music! I have several coaxial speakers and there is some magic in them, really!

Low or no group delay/phase shift preserves the timbre of an instrument, it's fundamental note and harmonics relation.

I have two dsp-multiway speakers at home, other with minidsp, other with Hypex FA. Perhaps next year I'll have more time for this kind of tests.. So far I prefer LR2 over LR4. Tests with asymmetric and very steep are waiting... Now with Minidsp 2x4HD I can start learning FIR too. (Speakers with minidsp and coaxial mid-tweeter are at my daughter's)