> I'm afraid I don't understand the premise behind this at all. It seems to
> me that you are talking about creating a tiny FM transmitter with very
> little deviation,and I don't see how that will decrease harmonic content.
> The signal with no harmonics is a sine wave at a constant frequency. Doing
> ANYTHING to alter this INTRODUCES harmonics. It seems to me that shifting
> the osc. frequency will mainly shift the rest of the harmonics with it. I
> don't want to take the time to actually try to figure out the result in
> full detail,but even if it DOES spread the signal out wider, it won't be
by
> much, and any unintentional receiver which received the original signal
> would probably have enough bandwidth to receive the slightly spread
signal,
> too.


His premise is that if you were able to shift the osc by a lot, then it
would reduce the average power at any point. Unfortunately, EMC scans are
not done with an infinitely narrow bandwidth. They are rather broad, but the
exact number I can't recall. You'd have to sweep the osc rather far.   The
osc would average out to the right number of cycles within any reasonable
time, like a bit-time for serial comms, but would never be on any one
frequency for very long.

As you point out in the end, it's not likely to spread the osc by much at
all, even using very sensitive varicaps for the capacitors. I leave it to
the student to figure what the relatively huge oscillation waveform does to
their capaacitance values ove time.

In short, coupling the osc caps to VCC will never be as good as coupling
them to GND. File this one under "stupid uP tricks".