Guidance on resonant freq scanning circuit

[chadj2]

I was hoping to get some guidance on which direction I should go first on designing a circuit for finding LC resonant frequency. I am not sure if such a circuit is in the public domain somewhere, but I have been unable to find one so far. What I am trying to do is build a circuit which is able to maintain resonant frequency in a LC tank even if the capacitance or inductance changes. For example, a circuit which constantly samples the voltage across the capacitor and inductor for peak voltages and makes adjustments in frequency to maintain peak voltages. Thanks in advance.

Chad

 

[k7elp60]

If you are talking about an oscillator tank circuit, then a phase lock loop would solve the problem. If you mean a receiver tank circuit a possible solution would be possibly an FM AFC circuit.

 

[chadj2]

Thanks for the response. I am just using a tank circuit. I wanted to keep the AC voltage across the capacitor and inductor at maximum. I will look into phase loop lock circuits for now then. Thanks.

Chad

 

[crutschow]

Why not just make the tank into an oscillator circuit? That will automatically generate the frequency at the resonant point.

 

[chadj2]

I wish that I knew how do design a circuit that does that. What I was hoping to do is drive a LC tank at its resonant frequency. The problem is that the inductance is going to be changing somewhat periodically. So I need a circuit that will detect the change in the resonant frequency of the tank and adjust the driving frequency up or down and drive the tank at the new resonant frequency.

Chad

 

[crutschow]

What is the frequency of the tank? That will affect the type of circuit needed.

 

[chadj2]

I haven't built it yet but the resonant frequency should vary between 5Khz and 40Khz. Thanks for the help.

Chad

 

[crutschow]

Perhaps you could something similar to this LC oscillator circuit. It should oscillate at the resonant frequency of the tank.

 

[chadj2]

I am not sure if it matters that the tank I am using is going to be in series resonance. But the big problem is that I am not exactly sure how to integrate that op amp into a circuit that drives a frequency back through the op amps input tank. Thanks again.

Chad

 

[crutschow]

I don't understand your concern. The circuit will automatically oscillate at the resonant frequency, which is what I thought you were looking for. Are you trying to do this while the tank is part of another circuit and that's why it's in series resonance? Otherwise the resonant frequency doesn't change if the tank is in series or parallel.

Perhaps if you explained how this tank is being used or what you're trying to achieve, we would have better ideas on what to do. It's difficult to come up with a solution when we only have bits and pieces of your project.

 

[chadj2]

Thanks again for the response. I am using the series resonant circuit for a voltage gain circuit. But the inductor will also have a secondary winding which will sometimes be lightly loaded but I want the circuit to remain in resonance. I have heard that there was a scanning circuit that will find a resonance frequency. I stumbled across a patent that seems to be doing something similar to what I am trying to do. http://www.rexresearch.com/meyerhy/wo92.htm
In that patent he talks about a resonant scanning circuit that is able to keep his tank in resonance. It would appear the resonant scanning circuit is able to detect resonance by detecting peak voltage and locking into that frequency. I think figure 8 shows the block diagram for how it connects. Unfortunately, the guy didnt say what IC he was using to do the scanning. Thanks

Chad

 

[crutschow]

Certainly it's relatively easy to vary the frequency up and down and look for the peak voltage. This can readily be done with a microprocessor. It would be programmed to scan or dither the frequency slightly and observe the peak voltage to keep the frequency at or very near resonance. As soon as it detects a drop in voltage, it reverses the frequency scan direction. (A similar method is used to keep scientific satellite dishes aimed for maximum signal). Many µP's have a D/A output which could control a VCO (voltage controlled oscillator).

That would likely be a lot easier than trying to do it with discrete IC circuits.

 

[chadj2]

I really probably need to learn how to use microprocesors it sounds like they can do anything. How long does it usually take to learn how to program one? So if I understand you correctly you are saying I should cease looking at discrete components to do this task and concentrate on learning how to use a microprocessor to do it.

Chad

 

[crutschow]

How long does it usually take to learn how to program one? So if I understand you correctly you are saying I should cease looking at discrete components to do this task and concentrate on learning how to use a microprocessor to do it.

Well, that's my thought, but I'm somewhat familar with microprocessors.

It's difficult to say how long it would take. Depends upon if you have ever done any programming. Basic is probably the easiest to learn as it uses common English words and syntax that's relatively easy to understand. It's the only thing I've used for µPs. You might Google some tutorials on Basic Programming Language to see how it looks to you. I leave C and Assembly to the experts. Those are like learing a foreign language.

My feeling is that, since there's a sequence of events that need to be perfomed involving making decisions, it seems like a discrete design would be fairly complicated. Offhand, it would require a sweep circuit, a method of detecting the peak voltage (perhaps a differentiator), some comparators to control sweep direction, etc.

To get an idea of what's involved in programming you can also look at

PIC tutorials:
Gramo's: www.digital-diy.net/
Bill's: www.blueroomelectronics.com/

or at www.Parallax.com for info on Stamp processors. I've used a Stamp processor to program a project in Basic and it was reasonably simple. You just plug the Stamp board into you computer and start programming. PIC has programming tools that are similar I believe.

If you think you will be doing more circuit design in the future you should strongly consider learning about microprocessors, since there's so much that's easy to to with a µP but difficult or perhaps impossible to do with discrete circuits. Basically you substitute software for hardware.

 

[Willbe]

the inductance is going to be changing somewhat periodically
a circuit that will detect the change in the resonant frequency of the tank and adjust the driving frequency up or down and drive the tank at the new resonant frequency.

Does the inductance change in steps or linearly with time?

 

[chadj2]

The secondary winding of the inductance would be charging a capacitor. I am not exactly sure if the inductance change could be specified that precisely. The capacitor would also be discharging at various rates. So, I would say the inductance change would be quite random. However, I am planning on limiting the range of inductance range. For example, I will not let the secondary winding to become fully loaded.

Chad

 

[chadj2]

I think I found what I am looking for at this link. **broken link removed**

Thanks

Chad

 

[Nigel Goodwin]

Well it probably explains why you're asking stange questions that make little sense, Stan Meyer was a fraudster and anything to do with him doesn't work and makes no sense at all. It's just a load of gibberish designed to separate gullible investors from their money.

 

[eblc1388]

I wish I know why people patented a 555 oscillator circuit.

 

[Willbe]

I wish I know why people patented a 555 oscillator circuit.

In general, the advancement of science is good for the public. The USPTO figures if people could make money off their invention they would be more motivated to invent.

An invention has to be "novel" or it has to be "not obvious to one of ordinary skill in the [science of oscillators]."
https://www.google.com/search?hl=en...uld+have+been+obvious+to+one+of+"&btnG=Search
Yet, I saw an patent application for a paper clip with ridges. It seems to me obvious to add ridges to a paper clip, but I only worked in that awful place (20% turnover/year among patent examiners) for three years so what do I know?