An amplifier circuit for a piezoelectric disk for ultrasound

[Isabella_]

Hey guys,

I need your help. I am a bit lost. I have to build an amplifier circuit for a piezoelectric disk for ultrasound generation in university.
The technical data (piezoelectric disk):

Resonant frequency = 80 kHz
Electrical capacity = 35 nF
Max. Supply voltage = 400 V

In the university I have a function generator which supplies me with an alternating voltage with the desired frequency (70 kHz). I think it would make most sense to build a driver circuit.

I would be very grateful if you could explain me how to proceed.
How should the amplifier circuit look like, which components and what else I have to consider.


Thanks in advance

 

[dick_freebird]

One key question is, "can you drive it with a square wave?" (with
acceptable results). A resonant load might reject the upper
harmonics and give adequate "spectral purity" in the acoustic
domain (or not, depending on Q and the piezo slab's ability
to translate higher electrical frequencies.

35nF is chubby and 400V is pretty large swing. You could find
half-bridge power stages that will go bang-bang (switching)
at those voltages. Finding a good linear amp, with voltage
gain of (what, 10-50?) and a good RMS current capability
and the thermal attributes to support that all, will be a
narrower field.

Companies like Apex make hybrids for this kind of thing
(whether ot not to the load and voltage you state). A
Class D audio amp might be another way to go if you
need spectral purity, but not too many audio systems use
a 400V+ supply (being for 2-8 ohms resistive, this would be
something like a 10kW - 2.5kW audio channel where norms
are 10-100W).

And you might need to remove some input filter (> 20kHz) to get
what you want, out of the PWM.

I'd bet there are interesting papers out there that speak
better to the "care-abouts" and "gotchas". Might have to
try a few keywords to get at what is pertinent. Maybe you
have some in mind already; if not, look for products in that
application-niche and follow the buzzwords as breadcrumbs.

 

[Isabella_]

Hey guys,

I need your help. I am a bit lost. I have to build an amplifier circuit for a piezoelectric disk for ultrasound generation in university.
The technical data (piezoelectric disk):

Resonant frequency = 80 kHz
Electrical capacity = 35 nF
Max. Supply voltage = 400 V

In the university I have a function generator which supplies me with an alternating voltage with the desired frequency (70 kHz). I think it would make most sense to build a driver circuit.

I would be very grateful if you could explain me how to proceed.
How should the amplifier circuit look like, which components and what else I have to consider.


Thanks in advance

First of all thank you for your answer.

I can't use a square voltage thats why i can't use FET or a Mosfet. The disc behave like a capacitor.
When i want o graduate a ultrasound with a frequency with 70 kHz how high must be the current when the input voltage is max 400 V and the capacitance 35 000 n?

 

[FvM]

A piezo disc does not behave like a capacitor in the operating frequency range. It behaves like a LC+C series/parallel resonator with real impedance in both resonance points. Ultrasonic power sources are mostly operated in series resonance.

Curiously you aren't yet talking about input power.

 

[Warpspeed]

Think of it more like a resonant tuning fork. The impedance varies hugely just around resonance.

Best way to start out, run your function generator into it with a sine wave at just a few volts and plot the impedance change, and be prepared to be amazed.
You cannot design a driver for it unless you know the load impedance and required power level and the exact frequency.

To feed power into it optimally, its a lot more practical to choose the lowest dip in impedance, and drive it at that exact frequency.
Even a few hundred Hz either side of 70Khz will make a big difference.
Square wave drive will be fine.

Just because the disc will probably fracture at it 400v max rating, does not mean you must run it up that high.
A few tens of volts will probably be enough if you can get it to absorb enough drive current.