New crossover for speaker with ribbon tweeter and dome midrange

[ChrisK15]

Hi all - I have some speakers that I made almost 15 years ago which sound great, but only after some hefty EQ. At the time I was hesitant to make my own crossover due to lack of experience/knowledge so I got an off-the-shelf Dayton 3-way crossover (either 500/4K or 750/5.6K). It works, but because of the driver sensitivities/impedances the balance is pretty off.

Tweeter: Aurum Cantus G2si (6 ohm, 96dB) - with 2-3ohm resistor
Midrange: Morel MDM 55 (8 ohm, 90.5dB)
Woofer: Aurum Cantus AC250-F1 (8 ohm, 87dB)

I'd like to completely redo the crossover with values designed specifically for these. I've plugged the base values into a crossover calculator but I lack the knowhow to properly compensate for the mixed driver sensitivities, particularly the woofer. An L-pad I assume would work well for at least the tweeter while testing, but not sure if it would also work or be recommended for the midrange. These won't see particularly high power/volume as I don't listen super loud, but I don't want to risk damaging anything either.

Any thoughts on how to start with this? I do plan on continuing to use DSP, particularly for the low end, but at the very least I want to get everything as electrically aligned as possible first.

Thanks!

 

[ThatSoundsGood]

Do you have the ability to measure each individual driver separately in the box? That is really the only way to get an accurate crossover made.

 

[ChrisK15]

Yes I do, I have REW set up so I can definitely do that. Is it recommended to measure at the listening position or centered on each driver?

 

[ThatSoundsGood]

1 meter away-directly on axis with the tweeter and the speaker should be as far away from boundaries as possible. People often like to do this outside because any reflections will cause changes in the frequency response.

 

[eriksquires]

I like OmniMic for it's ease of use and simplicity over REW, but of course REW is what most use.

The other two tools you need is Dayton DATS or willingness to use the REW impedance measurement and something like XSim or Vituix CAD