3 coil toroid

[mysticalchemist]

While waiting for HV parts to arrive, I had a play today with a TPU sort of idea I've had floating around for a while. The idea is to wind three equal coils on a toroid, and to drive them with a ring oscillator at their self-resonant frequency. A pickup loop is wound around the outer circumference of the toroid.

I wound an unknown (think it is powdered iron) core with a two turn pickup loop (red wire), and then three conventional wound coils (white, green, orange wire).

To drive the three coils, I used a circuit that I saw at JLN labs - http://jnaudin.free.fr/html/emvtxgen.htm

Today was just a rough breadboard build, and it did not work as expected. The transistor collector waveforms were very similar to those on the JNL page but not quite what I wanted. There were not enough turns on the coils, so there was a lot of dead time as far as flux change was concerned, and therefore a very high current draw. (5 volts was about the maximum supply that could be used.)

The output loop signal (yellow trace) was also not quite as expected. It had a spike at the transistor switching points (red trace) with a DC component that was different for each of the three phases. (I'm not worrying about this right now until I get the drive doing what I want.)

Next weekend I'll rebuild the circuit in a more symmetrical manner, and also rewind the three toroid coils with a lot more turns.

[EMdevices]

@mysticalchemist

I believe the ring oscillator needs independent inductors, not mutually coupled ones.  The phase changes from stage to stage and when the phase is 360 deg it is fed back to the front end (or becomes 360 after feedback)  and then the oscillator can sustain itself.   If the inductors are coupled the interaction is more involved and harder to figure out.


EM

[mysticalchemist]

Thanks EM.

I spent quite a while looking at the base and collector waveforms trying to work out what exactly was happening.

With fresh mind I have decided to try breaking the ring and driving one stage with a pulse gen and seeing what happens in the three stages while I try different coupling arrangements. That is my next test when I get some more time to play.

Peter

[EMdevices]

I simulated this circuit in LTspice IV,  to illustrate the phase change from stage to stage.

All such "rings"  oscillate at the frequency that gives a 360 degree phase change around the loop, (and the loop gain has to be greater than 1 as well)    So if we have 3 stages, the phase progression is 360/3 from each stage, if 4 stages are used, then each stage changes the phase by 360/4,  etc..   (assuming all the stages are identical)   If two stages are used, each stage has to be an inverting amplifier, i.e. produce a 180 deg phase change.  And that won't happen with inductors and capacitors, I don't think, the components must be resistive, or operate at resonance where the circuit is resistive.

EM