Piezo Tweeters

[subwo1]

Hi Everyone. I thought I would get some opinions on a project I am
working on. I am working on an ultrasonic varmint repeller to discourage
cats, racoons, oppossums, etc. and am not sure how loud I
can make it before blowing the tweeters. I remember a rather universally
characteristic figure of around 28 VRMS.

I think I could drive one at a few hundred watts if I had a transducer
that could withstand it. however, I don't want to run up my power bill,
and I want to maintain simplicity, so I don't want to bother with a
proximity detector. The present version runs on a standard 120 VAC timer
set to run dusk to dawn.

Therefore, I am considering driving it with only around 40 volts peak to
peak, but tuning the physical sonic area around the piezo element so
that it can resonate somewhere around 17 or 20 Kilohertz.

So the main thing I am wondering about is whether peaking the output for
a certain frequency will physically damage the piezo element inside the
tweeter. If the effect is anything like a cone speaker in a speaker box,
then my impression is there should be no problem. Howerver, the piezo
element is a crystaline structure, and I suspect that it may crack.
Thanks in advance, everyone.

 

[alfsch]

Hi,
the piezo has high impedanze, so you need not much power to drive , but i ve seen some cracks in heavy used piezos...but you should take them, because theres nothing cheaper, i think, to make high sound pressure at >20kHz.
greetings
alf

 

[Circlotron]

Since the tweeter will appear predominantly capacitive, you could try putting a suitable value inductor in series with the tweeter and driving it from a pair of mosfets in a half-bridge configuration. Use a square wave drive - the LC combination will turn it into quite a reasonable sinewave. Start with 10v p/p squarewave. Monitor the sinewave voltage across the tweeter with a scope and start from above resonance and work downward till you get the the required amount of voltage drive across it as you tune up the resonant slope. (i.e. decreasing frequency) If you run above the LC resonant frequency it will enable the mosfets to soft-switch but that's not really an issue at these power levels, just nice to have if all it needs is a stroke of the pen. You can put extra capacitance across the tweeter if you need to for the sake of tuning. Use a polypropylene (FKP or MKP) or polycarbonate (FKC or MKC) rather than a more lossy polyester (FKS or MKS) for both the series resonant cap and the tweeter shunt cap.

GP.

 

[tiroth]

Therefore, I am considering driving it with only around 40 volts peak to
peak, but tuning the physical sonic area around the piezo element so
that it can resonate somewhere around 17 or 20 Kilohertz.

Man, that would also be an effective me-repeller.

 

[Bill Fitzpatrick]

It's my understanding that ultrasonic varmit repellers don't work.

You might want to do an internet search on the subject before you waste a lot of your time.

 

[subwo1]

Bill,
so far, it appears to be helping some.

tiroth,
sorry.

Circlotron.
that soft switching explains why a switching power supply I am working on keeps the mosfets cooler.


alfsch,
makes sense.

 

[Electro]

Ultrasonic repellers that uses peizo speakers don't work because they are directional. The best way is use an inductor and oscillating it.