Hello everyone

Finding the resonant frequency of a 40kHz transducer using a sine wave on the frequency generator is no problem. When I switch to a square wave, everything seems to go haywire on the oscilloscope. There seems to be no way to find the resonant frequency. This wouldn't be an issue but my ultrasonic generator board is square wave.

3 questions:
Why can't an ultrasonic transducer be tested with a square wave?
Why does it work so well with a sine wave?
Why does an ultrasonic cleaner use a square wave and not a sine wave? It would seem to me a physical device would respond better to the voltage increase and decrease of a sine wave better than the abrupt change of a square wave. Testing proves this.

Equipment used: 200MHz oscilloscope and a 25MHz frequency generator. Device was tested at 40kHz

DUT: 40kHz ultrasonic transducer with a 100ohm resistor. Voltage used was 10vPP and the DUT was under load

 

Hello everyone

Finding the resonant frequency of a 40kHz transducer using a sine wave on the frequency generator is no problem. When I switch to a square wave, everything seems to go haywire on the oscilloscope. There seems to be no way to find the resonant frequency. This wouldn't be an issue but my ultrasonic generator board is square wave.

3 questions:
Why can't an ultrasonic transducer be tested with a square wave?
Why does it work so well with a sine wave?
Why does an ultrasonic cleaner use a square wave and not a sine wave? It would seem to me a physical device would respond better to the voltage increase and decrease of a sine wave better than the abrupt change of a square wave. Testing proves this.

Equipment used: 200MHz oscilloscope and a 25MHz frequency generator. Device was tested at 40kHz

DUT: 40kHz ultrasonic transducer with a 100ohm resistor. Voltage used was 10vPP and the DUT was under load

Ceramic ultrasonic transducers are fairly complex devices. They have have both series and parallel resonant frequencies. It is difficult, if not impossible to find the resonant frequency of an ultrasonic transducer using a swept square wave. The high transition rate of a square wave will cause the transducer to resonate at its natural frequency and at harmonics. All you will see on a scope is an odd shaped waveform. The amplitude will vary seemingly at random with the sum of the total of the harmonics produced as the applied frequency varies.
Once you know the resonant frequency, you can drive it successfully with a square wave but you have to be careful not to overdrive it because high transition currents can overload it and cause the ceramic material to crack. This can be controlled by slowing the rise and fall times of the square wave.

 

Yes, I knew there would be a problem running 2 transducers from one circuit board so I'll be using one board per transducer. The circuit board's frequency can be adjusted to the resonant frequency of the transducer which is easy to find with a sine wave.

You stated 'This can be controlled by slowing the rise and fall times of the square'. Wouldn't this for all practical purposes be a sine wave?
It just seems strange to me to have a square wave for and ultrasonic transducer.

Regardless, thanks for the info. I thought it was just me making a mistake. Everything worked so well with a sine wave and went to a mess with a square wave.

 

It just seems strange to me to have a square wave for and ultrasonic transducer.

High power square waves are simpler/cheaper to generate than sine waves.

 

hi,
All my commercial ultrasonic equipment used a square wave drive.
E

 

Yes, I knew there would be a problem running 2 transducers from one circuit board so I'll be using one board per transducer. The circuit board's frequency can be adjusted to the resonant frequency of the transducer which is easy to find with a sine wave.

You stated 'This can be controlled by slowing the rise and fall times of the square'. Wouldn't this for all practical purposes be a sine wave?
It just seems strange to me to have a square wave for and ultrasonic transducer.

Regardless, thanks for the info. I thought it was just me making a mistake. Everything worked so well with a sine wave and went to a mess with a square wave.

Commercial high powered square wave ultrasonic transducer drive circuits usually include an inductor in series with the output to reduce the rise time of the signal at the transducer. This does not make the signal into a sine wave. It just slows down the positive and negative transitions.

 

If I wanted to add an inductor, what would it look like? Would the inductor go on the positive output, negative4 output or both? It may be a dumb question to most here but I'm a boat Captain and not an EE.

The circuit board is here and its output can be measured. Slowing down the + and - transitions would seem like a good idea. If nothing else, it would be nice to experiment with.

 

hi Eli,
Do you have a model number or link for the ultra sonic transducer.?
E

 

You don't detail the actual test circuit.

If you are testing by applying the low-impedance output of the signal generator directly to the transducer you must remember that components of the applied waveform don't disappear just because they are outside of the resonant response of the transducer. If you apply a square wave, the voltage you see on the scope isn't going to turn into a sine wave. At resonance there will be a strong sinusoidal component essentially superimposed on the square wave. If you use a sine wave, then there is "nothing there" other than the fundamental frequency so you will see a sharp increase in amplitude of a clean sinewave at resonance.

It can be much more instructive to look at the current through the transducer. Since you probably don't have an oscilloscope current probe, the best way is to use a low value resistor between "ground" and the transducer and look at the voltage across that.