>I think that what David is suggesting is right.  Perhaps one could think of
it
>as a first approximation of a sinewave that is being sampled at the peaks
and
>the zero-crossings?  Since the slope of the sinewave at the zero-crossing
is
>more steep than the slope at the peaks, the 0-0 period could be very small
in
>respect to the period of the signal.


Interesting, that wasn't my intention, but that is the result.    I've
scoped the lines
on a piezo driven with the "wrong" cadence, and I have seen the voltage
doubling
effect working. It wasn't double, as it was dumpng into the uP protection
diodes,
but that's the whole point. We don't want to put much current into those
diodes,
and the basic limit here is the uP's own output current.  I think if you ask
your
uP vendor, they won't reccomend dumping 20mA into the protection diodes :-P

>I have not used this idea, but I will incorporate it the next piezo element
that
>I drive.  Thank you, David.


:)


An additional note, someone suggested that impact or vibration could cause
the
same effect, and I agree.  I'm leaning twoard a pair of zeners out there.

A magnetic beeper wouldn't have as many potential problems, maybe that's a
better
path.