Hi all. I am doing experiment about ultrasound on mircofluidic but have no idea of how to drive a piezo disc.

I have a 1 MHz piezo disc (STEMiNC SMD20T21F1000R), signal generator, oscilloscope, and three amplifier (Trek Model 2100HF; Mini-Circuits TVA-R5-13A+; T&CPOWER AG 1017L).

When I connected the generator, amplifier, and piezo disc in sequence and adjusted the AC frequency to 1MHz, I did not feel the piezo vibration or the water atomization in the nearby cup. The voltage on the piezo is increased up to 100V, but it just heats up.

I used T&CPOWER AG 1017L amplifier which shows forward power, reverse power and load power. When the frequency is adjusted to about 1.06 MHz, reverse power = 0 and load power = forward power. Does this mean the impedance matching or the frequency is right?

Thanks for your kind help.

 

Welcome to AAC.

PZTs are not necessarily simple to drive. They need a signal and amp with a very fast slew rate, and they need are very low impedance (~2Ω is typical). As you have implied, they need a high voltage as well.

Have you measured the signal at the transducer? If you are counting on the knobs on the instruments you are flying blind.
If you have checked, what does the waveform look like; is it a clean square wave with crisp rise and fall times?

How does the PZT perform at much lower frequencies? For example 10kHz, assuming that lies in the amps passband. If not, whatever the lower end of the passband is, +20% would be a good test.

You will need to systematically track down the problem, starting at the function generator and moving to the PZT.

 

Welcome to AAC.

PZTs are not necessarily simple to drive. They need a signal and amp with a very fast slew rate, and they need are very low impedance (~2Ω is typical). As you have implied, they need a high voltage as well.

Have you measured the signal at the transducer? If you are counting on the knobs on the instruments you are flying blind.
If you have checked, what does the waveform look like; is it a clean square wave with crisp rise and fall times?

How does the PZT perform at much lower frequencies? For example 10kHz, assuming that lies in the amps passband. If not, whatever the lower end of the passband is, +20% would be a good test.

You will need to systematically track down the problem, starting at the function generator and moving to the PZT.

Hi Ya’akov,

Thanks for your kind advice.

Yes the piezo's resonant impedance Zm is ≤4 Ω according to its webpage. Square wave was always distorted with uneven peaks, so sine wave was used which seems pretty clean.

I am confusing about the impedance matching of the circuit. The generator, amplifier and BNC plugs all have 50 Ω impedance, so how them matches with the PZT which seems has a very low impedance?

I measured the voltage across the PZT. When only one PZT is connected to the circuit, the sine wave is always clean. The voltage value will change when adjusting the frequency, and will obviously decrease to a minimum value near 1.06MHz.

When PZT is connected in series with a 50 Ω resistor, the signal on the oscilloscope has been oscillating and seems to be a superposition of multiple sine waves (with 45 degrees offsets). At the same time, the FP, LP and RP on the amplifier are also constantly changing. When adjusting the frequency to about 1.6MHz, a stable and clean signal is displayed on the oscilloscope.

 

How much I had repair such systems, them always contain a resonant tank to match an impedance. Because in the conjugation 50 Ohms with the 4 Ohms, clear, the whole energy is going to be lost. Wonder only why the measurer of generator don`t see it?