PLL Oscillator, Resonance?

bertus

Joined Apr 5, 2008
22,270
Hello,

The schematic shown is a control loop for an oscillator.
The oscillator itself is not shown.
Th HF signal should be taken from the oscillator.
The contol voltage should go to a varicap diode in the oscillator.

Greetings,
Bertus
 

leftyretro

Joined Nov 25, 2008
395
Could this circuit be used to automatically tune a circuit to its resonant frequency?

http://electronics-diy.com/electronic_schematic.php?id=45
Even with the not shown voltage controlled oscillator this circuit only generates switched fixed frequencies in 100khz steps between it's oscillators total range. First you have to say if that is fine enough frequency resolution for what you require? I suspect this circuit was part of the tuning section to generate a LO frequency for a FM broadchast band receiver.

Lefty
 

Thread Starter

Farlander

Joined Oct 14, 2008
158
What I require is a frequency generator which automatically sets itself to the circuit's resonant frequency.

The circuit is an LC tank circuit, so it will resonate at its natural frequency. In this case however, the capacitor has variable capacitance.

I was thinking that a pickup coil might be able to detect oscillations, and send the signal back to some kind of PLL or something... any ideas?
 

leftyretro

Joined Nov 25, 2008
395
What I require is a frequency generator which automatically sets itself to the circuit's resonant frequency.

The circuit is an LC tank circuit, so it will resonate at its natural frequency. In this case however, the capacitor has variable capacitance.

I was thinking that a pickup coil might be able to detect oscillations, and send the signal back to some kind of PLL or something... any ideas?
Measuring the resonant frequency of the LC network would require the use of a sweep generator (oscillator) and a wide frequency range AC voltmeter. When voltage is max or min (depending on series LC or parallel LC) then that is the tanks resonant frequency. Automating such a test apparatus is not simple of cheap. The industry has for instance Spectrum analyzers with tracking generator that can semi-automate such measurements. Do you really require such an automatic system?

A pick-up coil would not do any good as a LC tank, unless excited by a voltage, amplified and using feedback, will not oscillate on it's own, therefore there is no signal to pick up.

Lefty
 

Wendy

Joined Mar 24, 2008
23,415
As I understand it phase locked loops have more to do with pulling a frequency out of noise than resonance. It can mimic resonance to some extent, but it isn't the same thing.

It's best application is to synthisize a new frequency from an established time base with excellent accuracy.
 

Ron H

Joined Apr 14, 2005
7,063
As SgtWookie indicated, the simplest way to drive a circuit at its resonant frequency is to put into an oscillator, which is what his circuit is. There are myriad ways of making LC oscillators. You need to provide more information. Why do you want to do this?
 

SgtWookie

Joined Jul 17, 2007
22,230
Why do you want to do this?
He wants to find a way to cause a hydrogen/oxygen "HHO"/"hydroxy" cell resonate at it's natural frequency.

This is of course difficult, because as the gasses are formed in the cell, the dielectric constant between the plates changes. Since the dielectric constant of pure water at room temperature is around 78, and the dielectric constants of both hydrogen and oxygen are very slightly over 1, tracking a particularly desireable resonant frequency could be somewhat problematic.

I've informed our OP several times previously that either external capacitance needs to be added or the value of the inductor needs to be increased to compensate, but I guess he still hasn't grasped that concept yet.
 

Thread Starter

Farlander

Joined Oct 14, 2008
158
Yes the frequency will likely have to adapt to a very wide range.

Wookie at R1 on your schematic is that a potentiometer? Why does it say 90%?

Could the output of the op amp be used to swtich a fet for increased power to the circuit?

And yes the application is for hydrogren experimentation though I wish not to disclose much for fear of disapproval.
 

SgtWookie

Joined Jul 17, 2007
22,230
Yes the frequency will likely have to adapt to a very wide range.
It'll need to adapt to a wide range of capacitance. The easy way would be to use a moveable ferrite core in the inductor.

Wookie at R1 on your schematic is that a potentiometer?
Yes.
Why does it say 90%?
That's because between the wiper and the right side (towards the output of the opamp) is 9k Ohms and between the wiper and the ground side is 1k Ohms. If the sizes of the capacitor and/or the inductor are radically changed, it may require adjustment of both R1 and R2 to make the circuit oscillate properly.

Could the output of the op amp be used to switch a fet for increased power to the circuit?
No, not really - the rise time is too slow, and it's not rail-to-rail. However, as it is, it's a simple way for you to determine your cell's resonant frequency. Just connect the output up to an o'scope or frequency counter.

I'm really too tired to think about it at the moment, but I suppose a change to a comparator circuit may prove beneficial. Trouble is, MOSFETs take some time to switch on/off.

And yes the application is for hydrogren experimentation though I wish not to disclose much for fear of disapproval.
You'd be doing well to keep things inexpensive then. ;)

Nothing wrong with experimenting. Going broke while experimenting is different.
 

flat5

Joined Nov 13, 2008
403
In general, to sniff a tank circuit you can use a Grid Dip Oscillator. I only mention it because no one else has :)
 

Wendy

Joined Mar 24, 2008
23,415
I was thinking about this, I don't think a PLL will work in the classic sense, but you could use a phase detector to detect when resonance has been reached. It will also tell you if the frequency needs to be moved up or down. This is a job for a micro controller though, heavy duty circuitry (I wouldn't even attempt to use a PCB until a prototype is built). The problem is the range of frequencies, from what I've been able to infer the range is just too wide.
 

beenthere

Joined Apr 20, 2004
15,819
Electronically, that is true. But simple observation of the electrolytic cell will also demonstrate unusual activity. If some resonant activity does take place, the frequency will be obvious, and making up a tuning circuit that only has to cover a few hundred or thousand Hz rather than DC to 2.4 GHz is a bit more realistic.
 

Thread Starter

Farlander

Joined Oct 14, 2008
158
Thanks for all the input...

It's not the range of frequency that is the problem so much as constantly adjusting it. I already built a circuit that can switch a fet anywhere from 8khz to 40khz but without automated resonant detection, all I can do is watch an ammeter and aim for minimum amps. This is useful for about a milisecond, then the capacitance changes and subsequently, resonant frequency...

If the circuit could detect low amps or high volts and aim for the frequency which produces either of those effects the best, that would be ideal. It does not need to be so smart that it detects phase.

Could this work?
http://www.youtube.com/watch?v=Ru8YQ6HUwbU&watch_response
 

Wendy

Joined Mar 24, 2008
23,415
With a name like "Self Powering Resonant circuit for HHO Cell" my enthusiasm isn't there. Look up Wikipedia for Armstrong Oscillator, which is what this is.

There is a 100Ω resistor in series, this sets up an absolute current right off of 0.12A, if there is resonance involved it goes down.

I'd be very suprised if it oscillated at all. You remove energy from a tank circuit (which is what this does) and it tends to kill the oscillator.
 
Last edited:

Thread Starter

Farlander

Joined Oct 14, 2008
158
Thanks guys I appreciate all who bear with me. My enthusiasm precedes my knowledge on many occasions...

Ok in regards to this photo... assuming I had a capacitor with value of 11uF (the water tank cell) and an inductance, of oh say, 1H (to keep it simple) that would put my initial resonating frequency somewhere around 14khz.



The question: If I were to say use 12V input, what type of FET should be utilized? I don't know what kind of voltage to expect at the gate... thanks
 
Last edited:
Top