Simple harmonic oscillator with load

Thread Starter

Mira7

Joined Jun 26, 2013
10
I tried to construct a harmonic oscillator circuit that is to be loaded with ohmic resistance and integrated in a larger circuit. No oscillation occurred.

So I created a simple version and try to simulate it in Multisim (schematic attached). This version does not oscillate either (in simulation). Switch operations: S1 is closed, S1 is opened, S2 is closed. I expected and wanted oscillation to occur while S1 is open and S2 is closed until all energy is dissipated.

What changes can be made to achieve the desired circuit behavior?
 

Attachments

MikeML

Joined Oct 2, 2009
5,444
The LC ratio is horrible.

I suggest you get a real simulator:D

Here is what it takes to make it sort-of work:

V(a) and V(b) show when the switches are closed.

Both V(c) and V(d) exhibit oscillatory behavior.
 

Attachments

Last edited:

vk6zgo

Joined Jul 21, 2012
677
I sorely doubt if either version will work in practice.

The L/C ratio is too high---C1 looks like a short circuit at the frequencies where L2 reactance looks higher than zero.

The 1.0 resistor is "near as dammit" to a short circuit,too.

f=1/2π√LC is a nice formula,but you can put in many values of L & C.

Many combinations give the same value for "f",but some don't work in the real world!
 

LvW

Joined Jun 13, 2013
1,752
Mira7, I suggest not to waste your (and our) time for designing/analyzing the shown passive circuit.
Instead, it is much more economical to describe your requirements and to show us the selected oscillator circuit. More than that, if it does not oscillate describe your observation (e.g. DC output?).
 

Thread Starter

Mira7

Joined Jun 26, 2013
10
Mira7, I suggest not to waste your (and our) time for designing/analyzing the shown passive circuit.
Instead, it is much more economical to describe your requirements and to show us the selected oscillator circuit. More than that, if it does not oscillate describe your observation (e.g. DC output?).
Ok. I need to dissipate maximum possible energy in a resistive load. The energy should be dissipated in RLOAD as fast as practically possible. Currents should be high as RLOAD will increase with current. I have 250V capacitors available for prototyping.

You might say that a square wave with short pulses and low duty cycle is much better suited for this purpose. It just seem to require a much more complex circuit, but if you have a suggestion for a square wave version rather than a sinusoidal version, then that would be awesome. I'm using Arduino Uno to control switches and it just does not seem to be fast enough to switch in such a way that low currents are avoided (they occur due to depletion of the energy source). Energy transferred at low currents is mostly wasted. Maybe an analog control circuit or a much faster microcontroller is needed.

What are your design recommendations?
 

Thread Starter

Mira7

Joined Jun 26, 2013
10
I sorely doubt if either version will work in practice.

The L/C ratio is too high---C1 looks like a short circuit at the frequencies where L2 reactance looks higher than zero.

The 1.0 resistor is "near as dammit" to a short circuit,too.

f=1/2π√LC is a nice formula,but you can put in many values of L & C.

Many combinations give the same value for "f",but some don't work in the real world!
LC oscillators have turned out to be difficult to get working. For another use case I tried hartley, colpitts and a few other. None of them worked. Eventually I got an astable multivirator working instead.

Perhaps it is time to focus on a sqare wave solution to my current problem.
 

Thread Starter

Mira7

Joined Jun 26, 2013
10
The LC ratio is horrible.

I suggest you get a real simulator:D

Here is what it takes to make it sort-of work:

V(a) and V(b) show when the switches are closed.

Both V(c) and V(d) exhibit oscillatory behavior.
Excellent. I will try it. Which simulator are you using?
 

vk6zgo

Joined Jul 21, 2012
677
LC oscillators have turned out to be difficult to get working. For another use case I tried hartley, colpitts and a few other. None of them worked. Eventually I got an astable multivirator working instead.

Perhaps it is time to focus on a sqare wave solution to my current problem.
Just for laughs,reduce C1 (in your original circuit) to 1nF & try the simulation again,then try 500pf.
 
Top