Low-frequency High-amplitude piezo driver using ATTiny 85

[RS98]

Hi everyone,

I am a mechanical engineer who has recently started to experiment in electronics. I've recently started on a simple circuit that is required to drive a piezo disk with a relatively high amplitude (~80 V) at a low frequency, ranging from about 10 Hz to 150 Hz. The circuit is described below.

The main stage is a DC/DC boost converter circuit, where the MOSFET is switched at 1 kHz from an AT-Tiny PWM. The generated DC voltage (about 80VDc), is then modulated at the required low frequency with another N-Channel MOSFET. The frequency is defined by a 47k potentiometer mapped to output a second PWM signal between 1 and 150 Hz using an analog input to the ATTiny.

The circuit works as intended. However, I don't know if there is anything wrong with the circuit that may damage components over time. For example, should I be using P-Channel MOSFETs instead? Also, when I switch the right side MOSFET on, the voltage just goes straight to ground, there is no resistances except for the inductor, am I essentially shorting the circuit?

Again, very new to this, please excuse any dumb mistakes.



Thanks for the help!

 

[Nanren888]

Can you give the source of the circuit?
The placement of the FET relative to the diode is not what I'd expect.

Circuit correct? I'd initially expect the diode between the fet and the capacitor, so it doesn't drain the capacitor.
As it is, the fets are in parallel.
.
Also shorting the stored energy in the capacitor does seem a interesting choice. If your piezo really does not mind one-sided drive you could consider a push-pull driver, but that would bring with it those other fets you mention and more complicated bias.
.
Sorry need someone with more mosfet experience to comment on whether a mosfet straight across the capacitor is workable. 220pF is at least not so large.
.
Intrigued what a piezo driver does at 10Hz.

 

[Nanren888]

I googled for circuits. I no not always a good way to find typical, but anyway

one-sided drive + resistive
Similar with inductive component

or aim for symmetrical drive?

 

[RS98]

Can you give the source of the circuit?
The placement of the FET relative to the diode is not what I'd expect.

Circuit correct? I'd initially expect the diode between the fet and the capacitor, so it doesn't drain the capacitor.
As it is, the fets are in parallel.
.
Also shorting the stored energy in the capacitor does seem a interesting choice. If your piezo really does not mind one-sided drive you could consider a push-pull driver, but that would bring with it those other fets you mention and more complicated bias.
.
Sorry need someone with more mosfet experience to comment on whether a mosfet straight across the capacitor is workable. 220pF is at least not so large.
.
Intrigued what a piezo driver does at 10Hz.

Hi, Thank you for your reply. Yes, that's a mistake in drawing the circuit, apologies. The physical one looks like below:


The boost converter is based off of this instructable: https://www.instructables.com/DIY-Boost-Converter-How-to-Step-Up-DC-Voltage-Effi/ minus the feedback.The rest is my own, which is why I'm suspicious.

He uses a much larger capacitor (47 uF) as you mentioned. However, I noted that the larger cap lowers the driving voltage by quite a bit. I will look at a push-pull driver is, thanks for the tip.

The application is to move water using as much amplitude as possible (sealed off from the voltage of-course). The water really doesn't react well at higher frequencies hence the 10 Hz, which works well.

Thanks again!

 

[Hunter64]

Do you have any specifications of the Piezo element, otherwise it's impossible to answer your question.

 

[RS98]

Do you have any specifications of the Piezo element, otherwise it's impossible to answer your question.

Hi, they're 35 mm generic piezo discs with capacitance of 40 uF at 100 Hz (figure from Amazon: https://www.amazon.co.uk/gp/product/B07P6XCF2M/)

 

[Audioguru]

A piezo transducer is a squeaker, not a speaker. It produces no sound below about 1kHz.

 

[RS98]

A piezo transducer is a squeaker, not a speaker. It produces no sound below about 1kHz.

I don't need it to produce sound, I'm using the mechanical oscillation only to move some water.

 

[Audioguru]

They do not measure a sound producer in water to see how soon it corrodes.
The movement of a piezo disc is tiny then low frequencies will not do much.

 

[RS98]

They do not measure a sound producer in water to see how soon it corrodes.
The movement of a piezo disc is tiny then low frequencies will not do much.

Thanks for the reply, only the brass side of the disc is exposed to the water, corrosion is not an issue. I'm happy with the mechanical aspects of the project, it does what I want it to do. The circuit is the one thing I've doubts about: it works but I'd love the input of someone with electronics experience as to any flaws.

Thanks again

 

[Bluejets]

The mosfets you are using may not be suitable.
The original use logic level mosfets.

 

[Nanren888]

Another reason to consider changing the piezo drive
Regarding the mosfet across the capacitor thing. At 10Hz, you would probably saturate the core of the input inductor given that you are effectively shorting it to across the supply. This would mean a high current through battery, inductor, diode and mosfet.
Also, the rise side of the piezo waveform would follow the boost supply rise, that is in steps. Not sure whether you want any particular waveform on the piezo.