Piezo transducer questions

Hi All,

I'm looking around at piezo transducers for a project I'm starting on and I thought I'd check here for a bit of advice before buying. I'm looking to build a keyboard that plays wine glasses. Basically the idea is oscillators into piezo units attached to wine glasses.

Many transducers I come across list their resonant frequency. I don't know how important a spec that is for my needs. What does this equate to in actual use? Something I need to concern myself with when running sine waves into them? Most of the resonant frequencies seem to lie above where I'll be using them at, between 2k and 5k, whereas I'm working with glasses who's fundamental is in the range of 300 to 1500hz.

Anyone know of a good type that's findable in the US that would work well for sticking to wine glasses?

Hi

I actually design piezo transducers (from time to time).

What transducers are you looking at? Piezo buzzer disks?

In any event, What you want to do is not a problem. A bender element can easily be made to respond to all audio frequencies of the glass.

Just need some more information...

Les
L M Watts Technology

I think the ones I've seen are buzzer discs, basically. I'm not totally sure what the options available are out there. I'm just looking through the Mouser catalog and a few other places online. Now that I look some more, I'm seeing some benders in the catalog. What is that, a flexible element?

Basically I just need something small that can be easily and securely adhered to the side of a wine glass, and that I can feed audio signals to in the frequency range described above. Sounds like you're saying I'd be safe to go with most any variety of bender element?

Hi Bob

Yes Mouser has bender elements . These are Brass/tapecast soft PZT unimorphs. Of course they are big and flat.

You will be needing the piezo material without the brass. A good size might be 1X3 mm...something like that.

If this is a commercial thing I can make any transducer you want.

If it's a hobby/fun thing you can actually separate the PZT from the brass with mild heat, and score and snap it with a diamond or carbide scribe just like cutting glass.

You'll need some copper foil (.001" or so) and some air cure silver filled conductive epoxy. You can then glue them right to the glass. They'll operate in 1-3 mode, with the glass getting the neutral axis outside the PZT. (that's why you don't need the brass part)

At DC the things will appear capacitive, but will have resistive and inductive motional impedance as they go through the glass resonances. A FET op amp like TL072 will be a good start to amplify the signal.
You'll be able to get the full audio range...even a good ways beyond it.

If you are going to go the hobby route get Mouser 665-AB2065B. That's the 6500Hz bender element. Twenty cents each. Don't concern yourself with the resonance frequency. That means nothing in your application.

Then we can go from there, unless you want me to build transducers for you.

Les
L M Watts Technology

Just a hobby for the moment, though if it works out well I may entertain the notion of marketing these, thought it would be in very small quantity.

Your last post was very helpful, thanks a bunch! Once I get the thing up and working I'll be sure to make a post on the board about it.

YW!
Any further questions...just ask.

Les
L M Watts Technology

leswatts said:

I actually design piezo transducers (from time to time).

Click to expand...

Hi Les, Just for my own enlightenment, I am a little curious about piezo-electric transducers. I thought they were more chemistry than electronics, and I had the feeling it was a big industrial thing. Are you actually making the cristals, or buying them raw and conditioning them? (I don't want you to disclose your industrial secrets, just curious)

Hey Abbey

I buy the lead zirconate titanate material in pre poled sheets made to my specifications. I then use these along with other materials to make transducers.

For the last couple years I have had a research project to design special piezo microphones used in an energy harvesting application. The microphones actually generate enough power to run electronics.

The best I have done is about one watt from an 8 mm cone microphone. The sound field is powerful...about 10^5 Pascal overpressure.

Hooked in series, arrays of my microphones put out several hundred volts and will shock the daylights out of you!

I also design transducers for musical instruments and such. Lots of guitar stuff. I have some ultrasonic projects as well from time to time. Some are piezo and some are magnetic.

It's a lot of fiddly watchmaker stuff. I often machine  parts smaller that a mm. I have end mills in the shop that are only 0.01" in diameter!

The picture in the topic "show your bench" is actually my setup for adjusting impedance of the microphones, using an Agilent impedance analyzer. Delivering power is all about impedance matching of course.

Les
L M Watts Technology

I guess a little more clarification on some of the transducer's physical aspects of your earlier post... I'm really new to the physics of these things.

So reading up a bit more, it basically looks like what I need to do is to cut smaller pieces out of the larger piezo material, epoxy the foil to both flat sides of the element, and then epoxy this whole deal to the glass and use the foil as leads. Correct?

Is the brass plate in the one you suggest from Mouser primarily used to impedance match when coupled to whatever one might be sticking the element to?

Does saying that it'll operate in 1-3 mode mean that the material will be expanding and contracting along all 3 of its axes? And the neutral axis is perpendicular to its thinnest axis, which would be the most efficient surface to couple to?

Thanks for the explanation, Les!

> Is the brass plate in the one you suggest from Mouser primarily used to impedance match when coupled to whatever one might be sticking the element to?

I think it is the "wine glass". That the brass disk is tuned to sing a specific pitch. When driven at that pitch, it vibrates violently. What it really is, is a tuned circuit doing an impedance transformation between the very stiff piezo and our very thin air.

You could probably just stick the whole thing to a big bassy burgundy glass. The combination WILL ring at burgundy-glass pitch, pretty-near. A little bit flat because of the added mass. But you are probably water-tuning, so this tunes-out.

The hunk of brass might load a super-thin wedding-toast thimble so much it won't ring well. Which would be why Les tells you how to peel and split a lot of un-needed mass away from the commodity peeper-assembly.

http://www.youtube.com/watch?v=-WZExmtIJkI
http://www.youtube.com/watch?v=sXSnI8VTIDM
http://www.youtube.com/watch?v=A9t1jc6pIuc
http://www.youtube.com/watch?v=8D9BBMDWoNM
http://www.youtube.com/watch?v=sBdMbrvZKgE
http://www.glassarmonica.com/
http://www.finkenbeiner.com/gh.html

I actually have found glasses on pitch, so hopefully no post-tuning will be needed. Thanks for the links!

So reading up a bit more, it basically looks like what I need to do is to cut smaller pieces out of the larger piezo material, epoxy the foil to both flat sides of the element, and then epoxy this whole deal to the glass and use the foil as leads. Correct?

Click to expand...

Correct.

The foil/ conductive epoxy is to terminate the device. You can't really solder to them because they have a rather low curie temperature. They'll depole. For this reason it's a fine line to get enough heat to debond the brass and PZT without ruining them. You'll break or depole many. A temperature controlled hot plate is handy. Try about 150 degrees centigrade.

The PZT creates a potential by squeezing it in the thickness direction, and by stretching due to the poisson effect. Putting them in a beam or plate away from the neutral axis gives great mechanical advantage. That's what the brass is for. But the glass itself will take the place of the brass in your application.

I have some resource material about piezo , but like me it tends to be mathy. I'll dig them up if anyone wants/needs to know.

Les
L M Watts Technology