Driving a Resonator

Thread Starter

kaosad

Joined Apr 14, 2008
21
Hi,

I have a simple resonator tank with an inductor and a capacitor connected in parallel. I try to keep it resonating at vey high voltage (several kV) between kHz to MHz. The capacitor and inductor are small, in terms of nH and uF, respectively; so the enegy and the maximum current in the tank are small.

Is there a simple and power efficient way to inject sufficient energy into the tank to keep it resonating?

I was thinking of using Colpitt's Oscillator circuit, but I don't think it can handle above several kV range due to the limitation of transistor.

Anyone has good suggestion?
 

scubasteve_911

Joined Dec 27, 2007
1,203
I'm a bit stuck with your question.. Is the 'resonator' actually the tuned part of the circuit that dictates frequency? It doesn't appear this way, since you are indicating that you wish to vary this frequency significantly.

Oscillators typically exploit negative resistance with such a tank circuit, how are you accomplishing this? Or, are you merely driving a variable frequency into some sort of piezo device that can be brought to mechanical resonance?

The high voltage stuff is going to be very tricky to do. You might want to look into apex micro, since they deal with high voltage opamps, but I am not sure how high. You can try to do a bridge configuration to double your voltage swing. Other than that, you will need to do everything discretely.

Steve

Steve
 

Wendy

Joined Mar 24, 2008
23,408
How about a transformer configuration? The catch is the LC circuit looks like a series circuit, I'm not sure how you would get around it.
 

beenthere

Joined Apr 20, 2004
15,819
This is a case where a circuit diagram would really be helpful, plus some idea of what the circuit is doing.

Resonators, by their nature, are passive and take very little energy to operate. You are speaking of active devices, like the Colpitts oscillator. You should be able to run an oscillator to generate the frequency and then use other devices to amplify the signal to your desired voltage and power level.
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
Sorry for not being concise. The frequency is selected based on the oscillating energy in the tank so that the system does not melt the components. There are some dielectric materials to be studied and all of them are below several nH. Each material is tested one at a time, so the value is fixed for each experiment. The desired frequency is attained by manipulating the inductance.

The LC in the tank can be in parallel or in series; the tank configuration is not important. The important things are:

  1. it should oscillate at kHz - MHz (selected based on the energy in the tank)
  2. the amplitude should be over several kV.
Bill_Marsden, can you tell me more about the transformer configuration?
 

scubasteve_911

Joined Dec 27, 2007
1,203
I'm still baffled by your question.

Your saying that operating frequency is a function of two variables: 'energy' in tank and inductance. How exactly does on monitor this tank energy, by amplitude and knowing inductance, resistance, and capacitance? If you want to change your frequency on account of too much/little energy, then how are you planning to vary the inductance?

I was under the impression that tank circuits require negative resistance to help sustain oscillation. The only way to change frequency with a circuit like this is to change one of the circuit's parameters, like inductance, as you had mentioned. You should highlight exactly what you're doing with this to give us a better picture.

Steve
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
My experiment deals with dielectric materials. Say, I use material A which is placed between two conducting plates to make a capacitor. Provided that I have the capacitance of this capacitor setting, then I choose an inductor with an inductance with certain resonant frequency such that it falls in between kHz and MHz. Am I clear now?

My question is, how do I pump in energy into the tank efficiently such that they resonate at kV amplitude?
 

SgtWookie

Joined Jul 17, 2007
22,230
Actually, an LC tank circuit can have a fairly narrow, or quite broad resonant frequency range depending upon the ratio of one to the other. Sure, there will be a center frequency (Fc) but if the value of one component is significantly skewed from the other, the frequency range can be quite large. This becomes apparent once you've spent a few years with LC filters.

As Bill Marsden mentioned, inductive coupling (ie: transformer) is the way I would approach the problem. One might simply provide an occasional pulse to keep the LC tank ringing, if frequency weren't an issue.
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
Scubasteve_911, I think I know what your confusion is...

I choose one fixed resonant frequency-- but that frequency should fall in between kHz - MHz. (Once selected it stays and never to be changed during the test)

I can tolerate noise in the resonant frequency and it needs not be precise (+/- df around the center is fine).
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
SgtWookie, do you have a schematic diagram for it (in particular, the circuit that pumps in energy at the correct timing)?
 

SgtWookie

Joined Jul 17, 2007
22,230
SgtWookie, do you have a schematic diagram for it (in particular, the circuit that pumps in energy at the correct timing)?
Do you mean a generalized circuit?

Better to just concentrate on the frequency you have in mind.

This is starting to sound like a homework assignment. Is it?
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
No, of course not. Why would you think it is a homework? Someone else suggested using transfomer, remember? I am just asking how is it supposed to be done, that is all. Also, if it were a homework, why should I care about avoiding components from being melted?
 

kubeek

Joined Sep 20, 2005
5,793
One important question needed for the design of the driver is what the current fed to the tank should be, or the dissipated power in the tank.
 

Thread Starter

kaosad

Joined Apr 14, 2008
21
Can it be done with feedback? Probably the driver pumps in sufficient energy so as to maintain the desired peak voltage.
 
Top