Does a speaker in an anechoic chamber has minimum phase behavior?

[René - Acculution.com]

Here are the on-axis magnitude level responses for a perfect second-order (acoustic) crossover (2,5 kHz) with 1) two perfectly distance-aligned drivers, 2) 1 inch offset between drivers, equivalent to a pure delay on the 'further back' driver, and 3) a second order Bessel allpass compensation on the other driver:

The allpass filter is enough to almost fully compensate for a pure time delay up to 10 kHz or so. For larger distance difference, a second-order allpass filter will have trouble compensating enough. And the type/order of crossover and crossover frequency will also be important. And then there is the implementation of the allpass filter to consider, as a passive version will come with its own problems. And one should also check off-axis responses. I can add step response also, but the LR2 will itself distort such an input, so maybe not of much relevance.

 

[gnarly]

Good stuff Rene,

One inch and acoustic 2nd order though...yikes, that not much of a task.
How realistic is it to think about using all-passes as a 'fixed delay' for say 6 inches, and with a more common LR4 acoustic xover?

 

[René - Acculution.com]

That seems unfeasible at least just playing around a bit. 6 inches with LR2 still, corrected(!) with 2nd Bessel allpass looks like this (with the caveat that I haven't spent much time on checking all variables):

I haven't tried to figure out which order allpass would be needed here. But with an active setup with DSP I guess it is an easy task just to input the corresponding time delay.