I think it's done pretty well for the last 100 years.

.
I have a reasonable idea about how AM works, that's not the issue I'm trying to bring up.
My concern has to do with the fact that this project is dealing with a
mechanically "resonating" "Transducer",
and not strictly an Electrical/Magnetic-Waveform, as in proper AM-Radio-Broadcasting.

The ~500Hz bandwidth mentioned earlier seems to be rather optimistic in my opinion,
vaguely similar to 2-Tin-Cans with a String stretched between them.

I can't imagine a practical use for such a system,
but it is an interesting Project none the less.

BTW,
mounting the Transducers on a curve, having the radius of the desired projection distance
should substantially improve the maximum projection distance that can be attained,
and doubling the number of Transducers should generate much more Signal-Gain than
doubling the amount of Power pushing half the number of Transducers.
.
.
.

 

.
I have a reasonable idea about how AM works, that's not the issue I'm trying to bring up.
My concern has to do with the fact that this project is dealing with a
mechanically "resonating" "Transducer",
and not strictly an Electrical/Magnetic-Waveform, as in proper AM-Radio-Broadcasting.

The ~500Hz bandwidth mentioned earlier seems to be rather optimistic in my opinion,
vaguely similar to 2-Tin-Cans with a String stretched between them.

I can't imagine a practical use for such a system,
but it is an interesting Project none the less.

BTW,
mounting the Transducers on a curve, having the radius of the desired projection distance
should substantially improve the maximum projection distance that can be attained,
and doubling the number of Transducers should generate much more Signal-Gain than
doubling the amount of Power pushing half the number of Transducers.
.
.
.

500Hz end-to-end is optimistic as it depends of near perfect demodulation at the target. Our brains can usually reconstruct (with low background noise) human speech from 500Hz BW channels for common words with deep vowel sounds as it can predict the words consonants from low frequency clues.

 

L QC is correct about increasing the effectiveness! As for driving resonance, mechanical resonance is a bit different from electrical resonance, so I suggest starting out at a lower power level, and observing carefully as you raise it.