🚀 go-pugleaf

>It appears that what I'll get is an am modulated signal that is some 
>combination of the 20khz signal and the 44khz sample rate.

Yup, that's what you'll get, all right.  But now run through the math that
describes how to make this "AM signal," and you'll find that one of the 
components you'll get out of all this is a 20 kHz signal at (or at least in 
proportion to) the original amplitude, plus some other stuff that will be 
above the cutoff of the required LP filter (assume a "brick wall" at 22 kHz if 
you like).  You CAN recover the original.

Sampling theory is not always (some would say "never") intuitive.  But the
math does describe what we should expect out of it, and agrees with what
you'll find in the "real world" pretty well when you go to try it. 


Bob Myers  KC0EW   HP Graphics Tech. Div.|  Opinions expressed here are not
                   Ft. Collins, Colorado |  those of my employer or any other
myers@fc.hp.com                          |  sentient life-form on this planet.