I'm making some ultrasonic 'buzzers'. It is essentially a pzt ring that is glued with an epoxy resin to a thin steel diaphragm. I'm sure most of you have seen the audible buzzers you find in alarm bells, toys and the like, and this is similar. My buzzer is ultrasonic and operates at around 40KHz for resonance so theoretically it should be silent to the human ear, but there is a very noticeable audible ticking that I need to get rid of. What can you suggest?
Get another piezo and put the two close together and look at the receiver on a CRO.
This will confirm if you are actually transmitting at 40kHz.
If you are creating the 40kHz via a micro, it is the cycling of the sub-routine.
I checked and the CRO is 40KHz for transmit and receive. I have had other piezos buzz so I think it is a common problem, it may be like white noise or inteference. I think it has something to do with the gluing or steel material, perhaps I could dampen the audible buzz or glue more accurately? The drive circuit is a signal generator inputting to an amp outputting into a transformer. I've used this circuit on a different piezo without the transformer and it buzzed it the same way, maybe I could try a different amp?
I've attached a picture of the front and reverse of one I made some time ago, but the one I'm working with is almost identical. The pictured one is a bit crooked, I usually put them on better than that.
Then the "ticking" is probably from switching the sinewave on and off at points that are not the zero-cross point. That causes a square-wave type of edge that will make a click sound.
Then the "ticking" is probably from switching the sinewave on and off at points that are not the zero-cross point. That causes a square-wave type of edge that will make a click sound.
It is a quite distinctive ticking sound, so you are right. I'm trying to understand where exactly a square wave edge could be coming from, I can't see one on the scope - but I have a very cheap scope with low resolution so it wouldn't give me any clues. I'm wondering if the power board I'm using is a problem?
That is an interesting question. The ticking frequency varies with the voltage, at start up (~300V) the ticking is around 3Hz at maximum volts (~600V) it is around 1KHz.
That is an interesting question. The ticking frequency varies with the voltage, at start up (~300V) the ticking is around 3Hz at maximum volts (~600V) it is around 1KHz.
I mentioned the general circuit earlier in the thread, in that reply it was slack wording (apologies). So the answer to your question is yes.
My setup is this: signal generator
Mains transformer feeding power board
Power board from chipamp as noted overleaf
Power board supplying +26 and -28 VDC into amp
Amp is OPA548 outputting into a transformer
Transformer is 1:20 miniature audio transformer
(I have this problem even without using the transformer)
What I was wondering when I posted the power board schematic/details was whether there was some kind of offset in the setup at the zero cross point leading to the square wave edge you mentioned.
I'm not sure what the dielectric breakdown would be? I can't see any sparking unless I put the volts up too much, then sometimes the conducting epoxy I use melts and it'll spark across the epoxy, a few minutes later the pzt cracks.
I tried giving the signal from the generator an offset, keeping the sine wave above gnd. I also tried using a RC snubber network as mentioned in the amplifier notes, one shouldn't be needed to drive such a low capacitance and it didn't make any difference.
Is it worth trying to filter the output from the amplifier? How would I best do that?
Well you did say you were using a small audio transformer. Unless that's specifically designed for high voltage the winding insulation might not like > 300V ! Or there could be inter-winding arcing.
On the other hand, you said you were getting the noise also without the tranny. So that would tend to rule out arcing if the noise is still there at 28V.