** Interesting fact:

Testing a ( moving coil) driver with a battery establishes a polarity that
is the reverse of the "real" one during normal operation.

The *actual cone motion* in response to alternating current drive (within
a driver's operating band) involves a double integration of the current
ave [???] -- hence a 180 degree phase shift.

The same fact also means that a cone reproducing a square wave (within
it's
operating band ) follows a motion very close to that when reproducing a
sine
wave at the same fundamental frequency. IOW the third and higher order odd
harmonics of the square wave cause very little actual cone motion.

Bet anything this starts an argument.

There's no argument. As a degreed EE, I'm reasonably certain everything Mr
Clavin here has said is either wrong or a misinterpretation.

The claim of polarity inversion could easily be checked by monitoring the
speaker's output with a mic of known polarity. Or is Cliffy going to claim
that mics, too, have the "wrong" polarity?

There are two likely points of confusion. First, the current flow through an
inductor lags the voltage across the inductor by 90 degrees. (How a "double
integration" occurs is not immediately obvious.) Second, most cone-type
drivers are more resistive than reactive, so you're not going to get 90
degrees, anyway.

Assuming (incorrectly) that the driver were wholly inductive, a double
integration of the current would cause the current to fall at 12dB/8ve.
Assuming the driver is working "constantly velocity", you'd still wind up
with a 6dB/8ve net rolloff. (I think.)

I have no objection to someone saying "Everything you know is wrong!". I do
it all the time. But no one ever gives ME serious consideration, even when
I'm dead right. Cliff, here, should be grateful I treated his claims with a
degree of seriousness. It's far, far more than he deserves.