"Ben Bradley" wrote in message


Knowing that the amplitude of the sidebands should be proportional
(or at least some type of direct relation) to the amplitude of the 50
Hz signal. lower it by 10 or 20dB, and let's see how the sidebands
drop, hoping they don't fall into the noise.

OK, the data from 10 dB lower input is the second set of triple-tone
experimental data (light yellow background) at
http://www.pcavtech.com/techtalk/doppler/

Regardless, you already
have me convinced you're measuring Doppler distortion.

Well, a mixture. I'm convinced that AM distortion usually dominates, but
that Doppler is also always there, if its large enough to find.