Resistor values for voltage divider biased common drain/collector amp

Thread Starter

alangibson

Joined Sep 2, 2021
31
Hello all. I am trying to build an amplifier to drive a 113 kHz piezoelectric mister with 24 VDC. Signal is provided by a cheapo function generator capable of providing up to 5 VDC. So far I'm having no luck figuring out the resistor values to build a voltage divider biased common drain/emitter for a either the IRFZ44N MOSFET or the BC547 BJT.

I haven't found a single resource yet that gives a clear way of deriving the appropriate values. They almost always either assume you already know the resistor values.

Does anyone know of a straightforward method of determining these values?
 
Last edited:

Thread Starter

alangibson

Joined Sep 2, 2021
31

Ian0

Joined Aug 7, 2020
9,621
Do you need to drive it with 113kHz, or just to switch it on and off?
(If you just need to switch it on and off, what's the relevance of 113kHz?)
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
Do you need to drive it with 113kHz, or just to switch it on and off?
(If you just need to switch it on and off, what's the relevance of 113kHz?)
It needs to be driven at 113 kHz because that is the crystal's resonant frequency. Supplying a constant DC current would just heat it up.
 

Papabravo

Joined Feb 24, 2006
21,094
Hello all. I am trying to build an amplifier to drive a 113 kHz piezoelectric mister with 24 VDC. Signal is provided by a cheapo function generator capable of providing up to 5 VDC. So far I'm having no luck figuring out the resistor values to build a voltage divider biased common drain/collector for a either the IRFZ44N MOSFET or the BC547 BJT.

I haven't found a single resource yet that gives a clear way of deriving the appropriate values. They almost always either assume you already know the resistor values.

Does anyone know of a straightforward method of determining these values?
It depends entirely on your idea of straightforward. I have a written design procedure for designing a common collector (aka emitter follower) amplifier. It would require you to specify and calculate things that you have not mentioned so I doubt that it would be of much use to you, since it seems unlikely that your proposed solution is applicable to the problem you are trying to solve.

Maybe we could back up, understand the problem, and consider alternative solutions.
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
I assumed it was a square/rectangular wave from 0-5V. Apparently we need to keep peeling the onion...
I should have included this in the first post. The output of the function generator is zero-crossing and adjustable between 0 - 5 V peak. So the 5 V sine wave is 10 V peak to peak. It'll do square, triangle and sine waves.
 

Ian0

Joined Aug 7, 2020
9,621
No-one has mentioned current yet. I've seen piezo transducers for dehumidifiers that need 60 Watts.
How much current does it have to switch?
And do you have to drive it at 113kHz, or do you have to drive it at its resonant frequency?
 

dl324

Joined Mar 30, 2015
16,788
I should have included this in the first post. The output of the function generator is zero-crossing and adjustable between 0 - 5 V peak. So the 5 V sine wave is 10 V peak to peak. It'll do square, triangle and sine waves.
It will be simpler if you use a square wave from 0-5V. If your function generator doesn't have a DC offset, put a diode across the BE junction to keep breaking down and killing the transistor beta.
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
Maybe we could back up, understand the problem, and consider alternative solutions.
Indeed it's possible I've picked the wrong approach.

I've got a 113 kHz mesh piezo mister (https://www.aliexpress.com/item/4000498578805.html). I've got a function generator capable of outputting zero-crossing square, triangle and sine waves at 5 V peak voltage (10 peak to peak). I've also got a power supply capable of outputting 24 V. So what I'd like to do is drive the piezo at 113 kHz with 24 V. I'd love to hear any suggestions about the best way to do that.
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
No-one has mentioned current yet. I've seen piezo transducers for dehumidifiers that need 60 Watts.
How much current does it have to switch?
And do you have to drive it at 113kHz, or do you have to drive it at its resonant frequency?
This one is normally 1.5 W, peak 2.5 W. Current should be 50 - 100 mA I believe. 113 kHz is the published resonant frequency.
 

atferrari

Joined Jan 6, 2004
4,763
Alan, sorry to tell you but last year I decided to revisit the resistor divider polarization; got tens of pages quite well explained just going through Google.
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
Alan, sorry to tell you but last year I decided to revisit the resistor divider polarization; got tens of pages quite well explained just going through Google.
I've read every page Google has to offer, and as far as I can tell they all have the same problem of assuming magic resistor values of including unexplained magic numbers. You'll find *a lot* of people making the same complaints in comment sections and on YouTube.
 

Ian0

Joined Aug 7, 2020
9,621
Indeed it's possible I've picked the wrong approach.

I've got a 113 kHz mesh piezo mister (https://www.aliexpress.com/item/4000498578805.html). I've got a function generator capable of outputting zero-crossing square, triangle and sine waves at 5 V peak voltage (10 peak to peak). I've also got a power supply capable of outputting 24 V. So what I'd like to do is drive the piezo at 113 kHz with 24 V. I'd love to hear any suggestions about the best way to do that.
So, if you just switch it with a transistor at 113kHz, it will end up with an average DC bias of 12V, with a 12V peak 113kHz signal superimposed on it. Is that what is required? Or should it have 24V peak with no DC bias?
There seems to plenty of specifications, some of which make more sense than others:
Adjust the frequency difference: 0.003 MHz
Temperature frequency difference: 0.003 MHz
Total frequency difference: 0.003 MHz
Load capacitance: 3200pF
Load resonance resistance: 60Ω
Excitation level: 0.5mW
Reference temperature: 26℃/78.8℉
Insertion loss: 3 dB
Stopband attenuation: 2dB
Input impedance: 0.2kΩ
Output impedance: 0.2kΩ
I think I can understand "load capacitance".
 

Thread Starter

alangibson

Joined Sep 2, 2021
31
So, if you just switch it with a transistor at 113kHz, it will end up with an average DC bias of 12V, with a 12V peak 113kHz signal superimposed on it. Is that what is required? Or should it have 24V peak with no DC bias?
From past experimentation it seems that they're not very picky about the voltage you feed them, so I think they both should work. 24 V zero-crossing (48 peak to peak) should probably work too since it has a max working voltage of 60 V.
 
Top