Author Topic: Need help getting Piezo Atomizer circuit working  (Read 1504 times)

0 Members and 1 Guest are viewing this topic.

Offline MatCatTopic starter

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: us
Need help getting Piezo Atomizer circuit working
« on: May 22, 2019, 08:46:59 pm »
Attached is current schematic of the design. 

J6 is the Piezo element, it is rated to run at 1.7Mhz, and should draw about 25 watts at 24v.  C23, 27, 28, 29, and 30 are there for flexibility in trying caps, the circuit calls for 3.3nF, which currently is being handled by a single film cap (C98162 on LCSC).  L4 and L3 are there to allow flexibility in the 2uH specified inductance, currently I am using a 1uH inductor (C165771 on LCSC)  and an adjustable inductor that can go from 0.8uH to 1.6uH.  Trying various cap values, adjusting the inductor, trying a few different transistors, I just can't seem to get this piezo to function as it should (It should atomize water and create fog when operating properly.).    Also of oddness visible in the second attached image is that there is a weird 50Khz oscillation going on as well.
 

Offline MatCatTopic starter

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: us
Re: Need help getting Piezo Atomizer circuit working
« Reply #1 on: May 28, 2019, 05:26:00 pm »
Wow, I have to say, I have been struggling to get a circuit to drive this piezo for months now, and I am not the only one, this forum, and a few others all have dead discussions by others trying to get it to work, come on guys, take up  a challenge, let's get a workable circuit figured out that the entire internet can use :).
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Need help getting Piezo Atomizer circuit working
« Reply #2 on: May 28, 2019, 06:27:37 pm »
Is this circuit from an app note?

A few things:

- I don't see any bypassing on +24V to GND.  I would add a couple of 100n as well as a 10 or 22 uF electrolytic.

- R25 provides a base current to Q14 of 24 V / 470R = 51 mA  The minimum hFE is on the order of 50 so the collector current would be something like 2.5 A.  This would give a static dissipation of 60 W which is almost at its dissipation limit.  I would increase R25 to something like 2K2 or 3K3 and see what happens.

- R30 seems a bit low.

- what are L3 & L4? Unless they are toroids, is there any chance their magnetic fields are interacting?

If you have a signal source I suggesting breaking the feedback loop at C22 & L2 and injecting a sine wave into it and then observe the voltage on R30 while varying the frequency.
 

Offline MatCatTopic starter

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: us
Re: Need help getting Piezo Atomizer circuit working
« Reply #3 on: May 29, 2019, 07:03:08 pm »
I actually paid someone a small fee to help me figure out a circuit from an app note, he came up with this circuit and managed to make it work with random bits and bobs he had and DIY inductors, I am trying to get it to work with readily available components (mostly SMD).

The 24V rail has bypass elsewhere, you are only seeing the atomizer specific aspect of the circuit.

The piezo is spec'd to run at 25W, I will have to look at that on the resistor...

I honestly don't know enough about this circuits operation to fully understand what R30 is doing.

L3 is C165771 on LCSC, and L4 is typically an adjustable inductor.

I do have a signal source, I will have to experiment with that when I get the time in a few days.
 

Offline MatCatTopic starter

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: us
Re: Need help getting Piezo Atomizer circuit working
« Reply #4 on: June 18, 2019, 06:09:12 pm »
Been a WHILE since I posted, going to bump this thread.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Need help getting Piezo Atomizer circuit working
« Reply #5 on: June 18, 2019, 08:08:59 pm »
MatCat, have you found anything new?  Did you try any of the things suggested?
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Need help getting Piezo Atomizer circuit working
« Reply #6 on: June 18, 2019, 08:39:59 pm »
Those class E circuits are voodoo to me.

Why not just buy something which works and reverse engineer that?
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Need help getting Piezo Atomizer circuit working
« Reply #7 on: June 18, 2019, 08:41:42 pm »
Check circuits for piezo humidifiers, there is a lot of info out there. Google shows these circuits.

https://www.google.com/search?q=piezo+humidifier+circuit&source=lnms&tbm=isch&sa=X&ved=0ahUKEwiEvafW7_PiAhURLK0KHTApBUMQ_AUIECgB&biw=1120&bih=534
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7043
  • Country: nl
Re: Need help getting Piezo Atomizer circuit working
« Reply #8 on: June 19, 2019, 05:23:42 am »
Check circuits for piezo humidifiers, there is a lot of info out there. Google shows these circuits.

Sometimes I suspect designers just threw components at a circuit till it worked. What the hell is C2 doing in the circuit here? It's not going to resonate with L1, not at 1.7 MHz at any rate, all it can do is massively increase L1 current for no good reason AFAICS.

This design works in simulation, minus the extra capacitor which I assume was simply a copy paste error (also put the DC bias before the tank inductor, but it doesn't really matter). But I have no idea why it works ... why does L3 swing up the gate voltage? How do you derive the magic values for R5/L3/C4? (You can't change them too much before the design stops working in simulation.) L1 is simple, it needs to resonate with the parallel capacitance of the piezo, the rest is voodoo to me.
« Last Edit: June 19, 2019, 03:10:52 pm by Marco »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf