Scoping out an ultrasonic water atomiser

Описание: Although I've featured a few piezo disk water and aroma atomisers in the past - I'm not sure if I've ever scoped one. My main concern was that the act of probing the circuit might damage it by adding rogue capacitance to a resonant circuit.

I decided to remove the inductor and replace it with a resistor to get an easy probe connection point for the oscilloscope. In hindsight the weird waveform across the resistor was probably due to the high frequency making the parallel capacitor value significant with the relatively high resistor value compared to the extremely low inductor impedance.

The actual drive signal is a near 50/50 110kHz square wave. When I searched for 110kHz along with the keyword piezo I found lots of the same water atomiser disks. This is kinda strange, as some of the disks are sold as operating at 2.4MHz. So I wonder if the lower frequency is just plucking a resonant circuit, where the piezo disk capacitance and inductor form that resonant circuit.
On the other hand the 2.4MHz thing might just be an error that has been mass copied by sellers.

The inductor is actually a three terminal transformer that has windings that measure as follows:-

Primary:- 14uH 0.2m 18-turns (inner winding)
Secondary:- 420uH 1.75m 110-turns (outer winding)
Both wires around 38 SWG. About 1 ohm per metre.

The processor in this circuit is surprisingly stable given the lack of supply decoupling and its switching of a fairly nasty inductive load. But it still manages to bang out a 110kHz square wave while still making time to drive the data output to two addressable LEDs. It's possible it just ditches the piezo drive signal briefly while it's sending out the LED data.
It's notable that this circuit doesn't have a capacitor in series between the MOSFET gate and processor drive pin. Many others do that so that if the processor crashes it can't leave the MOSFET turned on continuously, with the risk of high current flow.

With a 4MHz clock and 1 million instructions per second processor like a PIC12F629 it's perfectly viable to create the 100kHz square wave with NOP commands as timing packers, allowing for the extra instruction cycle needed for a jump.

There are many of these water atomisers available, but if you want this exact one then here's a link to the item on AliExpress (not a sponsor).
This is an affiliate link, but that will not affect the price you pay.
https://s.click.aliexpress.com/e/_c3n...
The cost should be around 3 or 4 space-credits.



If you enjoy my videos then the easiest way to support the channel is to subscribe for more good video content. But if you would like to contribute a dollar or two for coffee, cookies and gadgets to take apart, you can do so by supporting me on Patreon. I release videos without adverts for critique and feedback on Patreon as soon as they have been made. I also do Patreon live streams, which are much more relaxed and chatty than the busy public ones.

  / bigclive  

Alternatively, for a single coffee contribution you can use PayPal:-
https://www.paypal.com/paypalme/bigclive


#ElectronicsCreators