All I've ever seen about DIY crossovers is that they're LCR networks, and what I'm used to seeing is an obvious air core inductor. But in the Klipsch design they must be using the self-inductance of the transformer somehow?

That seems pretty difficult, because it seems to me that the inductance in the secondary is going to vary by what tap you use, and the essentially paralled inductors of the primary and secondary are going to have variable interactions depending on the size of the secondary selected?

I found some indication in the patent referenced in the URL that the inductance of the transformer is used as part of the low-pass filter. But I get the impression from the patent that the ratios in that transformer are fixed, not user selectable as in the Klipsch. Does Klipsch simply accept the variations as a compromise to achieve lower losses than they'd have from using resistors to pad the input?

Maybe this is all to complicated for somebody who never got past Ohm's Law!

• http://www.audioannals.com/04481663-Spranger-txt.htm