Boosting function generator voltage

ronsimpson

Joined Oct 7, 2019
1,677
It's drawn backwards and I didn't hear him explain why L1 and C1 were required
C1 is there to ac couple the pieso. No dc will be on the Pieso.
L1 is there to increase the output voltage. There can be ac voltage across a coil but no dc voltage. The average voltage across L1 is zero. So if Q1 is on for 50% of the time there is 24V and some time across L1. When Q1 is off the voltage will increase to about 50 volts for about the same time.

alangibson

Joined Sep 2, 2021
31
Put a scope on the collector of the transistor. If you don't have one, connect the input end of the resistor to 5V and measure the voltage on the collector.
I've attached a scope snapshot from collector to ground with the piezo in place. With the piezo removed, it's a relatively steady 24 V.

Attachments

• 90.5 KB Views: 6

alangibson

Joined Sep 2, 2021
31
C1 is there to ac couple the pieso. No dc will be on the Pieso.
L1 is there to increase the output voltage. There can be ac voltage across a coil but no dc voltage. The average voltage across
Do you have any idea what would cause the inductor to overheat in this circuit? Many people that have put it together, including myself, have said that their inductor melts down.

ericgibbs

Joined Jan 29, 2010
14,234
hi alan,
Which post has the circuit you are using.?
I could create a LTSpice simulation, see if that shows the problem.
E

alangibson

Joined Sep 2, 2021
31
Which post has the circuit you are using.?
I've attached a schematic of what I have right now for the voltage boosting circuit (not the transistor switch). V1 is a small function generator with voltage adjustable between 0.05 V and 5 V peak. V2 is a regulated lab power supply. L1 is now a 100 uH toroidal inductor, not the radial one I was previously using

I could create a LTSpice simulation, see if that shows the problem.
Thanks for the offer.

A few notes:

- I've managed to get it to the point where it is merely too hot to touch by replacing the radial inductor with a toroidal one.
- It's putting out the amount of mist I expected now (for as long as I'm brave enough to let it run for). I've attached a capture of the wave form I've got now.
- These piezos resonate at 113 kHz and 108 kHz. By playing around I've found that it hits its maximum mist output at 108 kHz.
- Current draw on the lab power supply now reads 0.28 A

Attachments

• 124.3 KB Views: 8
• 31 KB Views: 8

ronsimpson

Joined Oct 7, 2019
1,677
what would cause the inductor to overheat in this circuit?
Too much current. Heat from core loss.
Too much current. Heat from copper loss.
Reduce the duty cycle. Get a larger core.
------
Send a picture of the coil. Size of core, size of wire. Any data you have.

Last edited:

ericgibbs

Joined Jan 29, 2010
14,234
hi alan,
As per your circuit, LTSpice shows this plot.
E

BTW: How was a piezo value of 100R chosen.?

eetech00

Joined Jun 8, 2013
2,658
Simulation looks pretty close....

Attachments

• 226.4 KB Views: 2
Last edited:

alangibson

Joined Sep 2, 2021
31

alangibson

Joined Sep 2, 2021
31
How was a piezo value of 100R chosen.?
100R is my best guess as to the actual impedance while resonating. The union of all the specs I've seen from the many offers they are available from on Amazon and AliExpress are:

Max voltage : 70 V
Max voltage : 60 V
Voltage: 3 - 12 V
Operating power: 1.5 W
Max power 2.5 W
Resonating impedance: <= 100 Ohm
Resonating impedance: 60 Ohm
Input impedance: 200 Ohm
Output impedance: 200 Ohm

alangibson

Joined Sep 2, 2021
31
I wired up two of the 100 uH inductors in series for 200 uH, and now I'm getting a current draw of 0.15 A on the lab power supply. Temps for the inductor and MOSFET are down from 100 - 115 C to 50 60 C.

Mist output is the same, so I'm going to say that the problem was indeed using a small radial inductor instead of a larger toroidal one. I think the next step is to get a name-brand 220 uH one like this: https://www.mouser.at/ProductDetail/Murata-Power-Solutions/32221C?qs=94qWIstW55VRKB8JdQNyVg==

ApacheKid

Joined Jan 12, 2015
442
Hello all. I'm generating a 113 kHz zero-crossing sine wave at peak voltage of 5 V with a cheapo function generator. My target load is a piezoelectric mister that looks to be about 400 Ohm resistance when resonating.

I would like to boost the voltage to 50 V peak since this is the most that I'm sure the mister can take. Can anyone recommend a good method for boosting the voltage to 50 V while retaining the waveform and using only a max 24 VDC power suppy?
If its a sine wave then why not use a step-up transformer?

alangibson

Joined Sep 2, 2021
31
I added one more inductor in series (for a total of 300 uH). Now the inductors are basically room temperature and the MOSFET is 41 C.

Output voltage seems to drop about 3 V per additional inductor. It's currently about 46 V peak which is indistinguishable from the single 100 uH inductor in terms of mist output.

To any future readers finding this thread because they're having overheating issues on their piezoelectric ultrasonic mister driver, use a 330 uH toroidal inductor and that should fix your problem.

Example inductor: https://www.amazon.de/gp/product/B087CKSGHC
Mister I used: https://www.amazon.de/gp/product/B07V72T1T9

Sources I used to build the original circuit:

https://www.instructables.com/Make-Your-Own-Super-Simple-Ultrasonic-Mist-Maker/
https://www.electroschematics.com/ultrasonic-mist-maker/
https://edisonsciencecorner.blogspot.com/2020/07/diy-mist-maker.html

alangibson

Joined Sep 2, 2021
31
I've attached the final working schematic I ended up with.

The addition of R4 changes the NE555 duty cycle from 70% to 50%. This fixes the other problem I had where the circuit would try to pull well over 1 A when I connected the timer circuit to the gate of the MOSFET.

Here's the BOM as well:

Code:
V1  : 5 VDC power supply
V2  : 24 VDC power supply
IC1 : NE555 timer
https://www.amazon.de/product/dp/B086J46GSG
Q1  : IRFZ44N MOSFET
https://www.amazon.de/gp/product/B0831NZHNW
L1  : 330 uH toroidal inductor (Other types are likely to overheat)
https://www.amazon.de/gp/product/B087CKSGHC
PZ1 : 113 kHz piezoelectric ultrasonic mesh mister
https://www.amazon.de/gp/product/B07V72T1T9
R1  : 10 Ohm 0.5 W resistor
https://www.amazon.de/gp/product/B08FD4KHP2
R2  : 5 kOhm multi-turn potentiometer
https://www.amazon.de/gp/product/B07X6RR6F8
R3  : 10 kOhm 0.5 W resistor
R4  : 10 kOhm 0.5 W resistor
C1  : 100 nF ceramic capacitor
https://www.amazon.de/product/dp/B07PP7SFY8
C2  : 10 nF ceramic capacitor
C3  : 100 nF ceramic capacitor
C4  : 100 nF ceramic capacitor

Attachments

• 43.1 KB Views: 10

ericgibbs

Joined Jan 29, 2010
14,234
hi alan,
A simple model of the 113kHz Piezo.
Try these values in place of the 100R fixed resistor in your circuit.
E

Attachments

• 28.4 KB Views: 4
• 28 KB Views: 4

ronsimpson

Joined Oct 7, 2019
1,677
the problem was indeed using a small radial inductor
If you are using 2 or 3 inductors and they are cool. Then you are at a point of having 2 or 3x the area of the single core.
Here is a picture where some one doubled up the cores to get twice the area.

I don't care much about the Diameter, mostly the cross sectional area.

Because the 3 core option is cool then the heat is not coming from copper loss.
RonS.