Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Ultrasonic fogger

Status
Not open for further replies.

nikt0

New Member
I've always been fascinated by the ultrasonic fog generators you see in pet shops (and this link), but they tend to be kind of pricey. I think it would be fun to build my own from scratch, but there are a few operating principles I need some clarification on.

My understanding is that most of these fog units (and cool mist humidifiers and medical nebulizers) use a piezoelectric transducer with a resonant frequency of ~1.6MHz, and this high energy vibration nebulizes the water on top of the disc, creating a cool looking fog effect.

So, basically, all I'd need to do is build a 1.6MHz function generator (555 timer-based?).....right? I'm still not sure how much power it takes to drive one of these piezo transducers, however. Should I build my function generator and use its output to drive a power transistor (like, say an IRF510 MOSFET), which in turn, would drive the transducer, allowing me to experiment with higher voltages?
 
Why don't you first start with a transducer, find its operating characteristics on the data sheet, then design for that?
 
Yes try to find a 1.6 Mhz transducer on net. Then design the driving circuits.. When completed share your fog with us :p

Aily
 
Attached is Xtal Oscillator. Shown crystal is of 1.84 Mhz. However 1.6 Mhz crystal can be used in place of it.

Aily
 

Attachments

  • Xtal_osc.jpg
    Xtal_osc.jpg
    10.3 KB · Views: 7,893
They do make ultrasonic humidifiers too so googling for that info might help. If you could identify a spare part you'd be able to zero in on something a little quicker. If you've got an appliance parts dealer or store nearby you might find they can show you the part - the cost might be right.
 
I just built a fogger project just a week or so ago, from parts I salvaged from a humidifer. Need to do more work on it, just through it together with hot glue and a little solder in about an hour, so kind of rough but it worked okay. Not quiet the volume of fog I had hoped for, but not bad. Search for the thread...
 
I just got a package of transducers in today (bought off eBay), but unfortunately, there are no identifying marks on the discs or the packaging. They were sold as replacement parts for foggers like the one I linked to, and come with a metal key that you use to open the unit.

So far, I haven't had much luck googling for information. Most of my searches turn up merchants trying to sell foggers or medical nebulizers or asthma support groups. If I try looking for parts or part numbers, I mainly get Chinese import/export companies or large OEMs, and even then I can't find part numbers or datasheets.

Is there any general information about driving ultrasonic transducers that I could perhaps apply in my project? Is it simply a matter of building an oscillator at the correct frequency and using it to drive the required voltage to the transducer, or is there more to it?
 
The most important thing with a piezo device is it's resonant frequency. Obviously power output is going to be an issue but these devices don't generally put out a huge amount. They rely themselves on the fact that they generate a frequency which is at some natural harmonic of the substance. You don't necessarily need to generate it's resonant frequency directly, especially with a square wave drive (like a micro controller or fast logic) BTW, a 555 won't cut it, they top out in the hundreds of kilohertz. Square waves contain an infinite (but amplitude decreasing) number of odd harmonics at whatever frequency it's running at.
 
Could you connect a signal generator that goes from audio to whatever you think the max might be and monitor the voltage. It would seem that you'd see a change in behavior as you approach or arrive at resonance. It's possible that there are multiple points of resonance.

Inverting this approach - can you find at least one working "fogger" and look at the transducer terminals - with a scope or counter. The unit you measure may not represent the transducers you have in hand but it's a start.
 
Ok, how about I design a PWM circuit and filter the output with a capacitor to approximate a sine wave. Would that work, provided I gave the transducer enough voltage and generated a frequency that's a harmonic of the transducer's resonant frequency?

Again, any gentle nudges or hints to steer me in the right direction would be much appreciated, as I'm having trouble finding general theory or design guidelines one must consider when designing an ultrasonic piezo driver.
 
Actually the ultrasonic fogger is a LOT of power! I think I saw one with a 40W transformer, I don't really remember for sure. They usually have a sensor to turn it off if the water gets low so it won't burn up.

The fogger parts I've seen- a silver puck with the transducer in the center. The driver is built into them, they take in DC.

They are NOT in the 1.6MHz range! UltraSONIC... think like 40KHz. No mechanical wave works in the megahertz range.

In order to produce fog effectively, the water depth needs to be within a certain range. Too deep or too shallow is a prob.

Wanna feel something odd, put your finger in the water. Damn feels like an electric shock.
 
You don't need a sine wave for piezo's they are just happy as can be to run on a square wave, as long as you're not trying to produce an audio signal on a piezo speaker that is =) You just want power out at a primary frequency, a square wave will do that fine. Square waves have a lot of natural harmonics as well. You could make a simple oscilator using a schmitt triggered inverter and a resistor/capacitor, a second inverter would allow you to buffer the signal and allow you to drive a few mw's into a load. A simple transistor or FET stage would give you up to a few watts of power easily.
 
To properly drive one though you really need to know the impedance at it's resonant frequency so you know what voltage to use.
 
Here are a few quick pictures. Not sure if they will be of much help. The TO-220 device got hot enough to melt the plastic container I put the electronics in to protect from splashing. Yes, it was on a nice sized heatsink, which I need to remove to get the PCB out.
 

Attachments

  • PIC00010.JPG
    PIC00010.JPG
    667.5 KB · Views: 5,696
  • PIC00011.JPG
    PIC00011.JPG
    572.7 KB · Views: 2,351
  • PIC00012.JPG
    PIC00012.JPG
    535.3 KB · Views: 2,346
HarveyH42 said:
Here are a few quick pictures. Not sure if they will be of much help. The TO-220 device got hot enough to melt the plastic container I put the electronics in to protect from splashing. Yes, it was on a nice sized heatsink, which I need to remove to get the PCB out.

Wow, thank you! The pictures are very helpful. The TO-220 component unfortunately appears to be a custom job, and I can't find a datasheet for it. I'm assuming it's a high frequency, high power FET of some sort.

I still have a few questions, however: in your picture of the component side of the board, there appears to be another component hidden by the large red capacitor in the upper left corner, probably a resistor? Also, what are the color stripes printed on the inductor that's partly obscured by the same capacitor?

Does the white socket connector have a single wire leading from it? What does it lead to? And the wires soldered to the trace side of the board - they lead to the transducer, I presume?


Thank you so much to everyone who's replied!
It's kind of frustrating knowing enough about electronics to be dangerous, but still missing some of the fact and theory necessary to design many of the projects I want to try. I've picked up a little bit of electronic theory through various projects I've undertaken, but have no formal training. I've found microcontrollers to be a great crutch, since I find it easier to kludge together firmware whose functionality could probably easily be implemented with a few transistors or op-amps by an experienced engineer, but sometimes you don't have the option of using an MC.

And I'm rambling.... I guess what I'm trying to say is, I really appreciate your helping me to fill the gaps in my knowledge.



Oh yeah, one other thing: what is impedance? I found the wikipedia entry on it rather confusing. It sounds to me like resistance that's generated by a component when an AC waveform is run through it. Is this right, or at least on the right track?
 
Last edited:
nikt0 said:
Wow, thank you! The pictures are very helpful. The TO-220 component unfortunately appears to be a custom job, and I can't find a datasheet for it. I'm assuming it's a high frequency, high power FET of some sort.

I still have a few questions, however: in your picture of the component side of the board, there appears to be another component hidden by the large red capacitor in the upper left corner, probably a resistor? Also, what are the color stripes printed on the inductor that's partly obscured by the same capacitor?

Does the white socket connector have a single wire leading from it? What does it lead to? And the wires soldered to the trace side of the board - they lead to the transducer, I presume?




The inductor, brown-black-brown-silver. The white connector is a single wire, for sensing water level, auto-shut off. The two wires on the trace side go to the transducer. Thought about hooking up the meter and getting a frequency, but no idea where I the manual is, little concerned it might be a little much. Considering that the transducer is the same size and appearence as any other piezo disc, with exception of a stainless surface, I'd guess that its 40 kHz or less and under 50V. Should be able to get a 555 to generate the pulses, and a high power switch. The transducer will get hot, definately keep it wet.
 
Oznog said:
They are NOT in the 1.6MHz range! UltraSONIC... think like 40KHz. No mechanical wave works in the megahertz range.

Not quite right. Medical ultrasonics, as used for imaging the internals of human bodies for cancers, babies etc, use frequencies of 4.5 and 13 Mhz.
The have to use such frequencies to get the resolution.

JimB
 
Medical scanners work as high as **broken link removed** Crystals in oscillators can work as high as 200MHz. The maximum frequency of sound in any medium is about 12.5THz, though I don't know if sound waves of this frequency exist in nature, maybe in nuclear reactors or on stars.
 
So what is need to make the fogger??? any one can please put here a schematic of the fogger please :( :(
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top