mike wrote:
John Popelish wrote:
I think you can mechanically broaden the response of the piezo and
also increase its sensitivity by putting an exponential horn on it.
Agree. Fixing it mechanically is likely better than trying to patch
up a high Q response after the fact.
OK, that's two votes for the horn, so it's probably not a bad idea at
all. However, I have both a space limitation and a horn-manufacturing
limitation, not to mention a limitation in my understanding of acoustics
which I figure would help me work out the size needed for best
performance from ~30kHz to ~100kHz. Quicky paper horns (pretty much
linear) seem to help with the sensitivity, though.
First thing I'd try is to run the piezo into the virtual ground of an
op-amp. But the intrinsic series impedance of the transducer may make
this unfruitful.
I'll give that a shot on my breadboard. I still have to check the actual
response of the transducer; I've been assuming so far that it has a
fairly tight spike around 40kHz, but I've also read that some work
just fine with no detuning at all. Given that I'm already fairly
happy with the sound, what I'm trying for now is to solidify my
understanding of the detuning. I'm currently under the (probably
wrong) impression that it's mostly phase games. Am I close?
May have to resort to mechanical damping. Glue
the piezo to a piece of foam rubber and load it with a horn as
described above.
Cool.
I have a commercial ultrasonic leak detector. Can't see in the hole too
well, but it looks more like a plastic dome on a voice coil than a piezo
transducer.
The transmitter and receiver transducers "look" to be the same, but
that's not necessarily a valid conclusion. I never found any data on
the setup, so don't know what the claimed bandwidth is.
Might be interesting to plug on a capacitor microphone element and see
what that does. Most of the microphone specs I've seen are still going
strong at the 20 KHz. end of the graph.
mike
Hm...that sounds neat. I'll see if I can find one.
Thanks for the tips,
Torben
|