Use cone or motor distance for estimating time alignment?

[owdi]

When modeling an MT on a flat baffle, common practice appears to be to estimate the woofer delay by measuring the difference in distance to the listener between the voice coil of the tweeter and woofer. I can't site any sources right now, but I have read this on multiple sites and forums.

This practice has never sat well with me, it doesn't make sense. AFAIK, sound energy travels through the voice coil and cone much faster than it does through air. If the cone isn't breaking up, then it's moving like a piston, instantly*** transferring energy from the voice coil to the air. If that's true, then the best estimate of woofer delay is to look at the difference between the cone of the tweeter and woofer.

So which is the better way? Taking the difference between cones, or voice coils? What am I missing?

Dan

***I know it's not speed-of-light instant, but I assume it's much much faster than sound through air.

[Dennis H]

Theory aside, in practice it turns out that the acoustical center of a woofer is somewhere near where the voice coil joins the cone. But if you do measurements that include real phase, you don't need to worry about it. The XO software will work just fine with the measurements and you don't need to worry about where the 'real' AC is because the software can calculate the difference in phase between the woofer and tweeter.