Clio09,
"Based on what you said it appears to be after the crossover."
Based on what I wrote, the A-C Zobel appears BEFORE the x-over. :-)
Anyway, I'm no expert in this never having built my own speakers. So, it would be safer for me to say that I don't know whether or not before or after the xover makes a diff. I think that we might need a 3rd person in the know to straighten us (atleast me!) out.
Just thinking out loud: they say that the speaker manuf uses a Zobel to make the speaker load freq insensitive to the amplifier. I looked @ the ckt in the link provided by Distortion & if I work out the impedance xfer function of the Zobel & draw its Bode plot I see that there is a pole @ DC (0Hz) & a zero at 1/RC. This zero negates the pole at DC such that the eff freq response is flat over freq. So, if the values of R & C are chosen such that the zero cuts in well below the -3dB point of the speaker, the amplifer will see a constant (over freq) impedance. So, maybe it does make sense to have the Zobel across the speaker binding posts (rather than across the woofer terminals). So, I wonder why, then, all the diagrams in the literature show Zobels directly across the woofer terminals?