Excel Theremin Simulator v4

Given two identical coils magnetically coupled on the same core:

Wiring them in series gives single winding with twice the number of turns.

Inductance is proportional to the number of turns squared, so doubling the turns should give 4 times the inductance of a single winding.

4 * 160uH ~= 600uH so that makes sense.

Do you think the capacitor is located between the coils in the antenna resonator for a reason, or can it be placed anywhere as long as it is in series? 

It seems to me that this series capacitance is perhaps one way to improve linearity?

As for the tank, since the windings are magnetically coupled, I would think whatever one winding "sees" the other will "see" as well, so they could be lumped together for a first order analysis (but my coil intuition is still taking baby steps).

What I don't get is how the oscillator works.  If drive and sense are on either side of the capacitor, at resonance you get a 90 degree phase difference, but the tube gives 180, correct?  [EDIT] Oh, I think I see it now.  The coil ends are shorted at AC, so it looks like a parallel tank, with 0 degree drive/feedback.

"Do you think the capacitor is located between the coils in the antenna resonator for a reason, or can it be placed anywhere as long as it is in series?"

It must be placed where it is, as I see it - It is the "transformer" aspect of the tank coil which provides the feedback, not the capacitor.. Yes, both ends of the inductors are effectively "grounded", so viewing it this way, only the transformer coupling is relevant really (waveforms across the capacitor are 180 degrees out of phase) .. But from a perspective of determining resonant frequency it is a series resonant circuit.. because waveforms across the capacitor are 180 degrees out of phase.

It fills me with awe!

Fred.

 (I dont want to divert this thread too much with discussion about the Lev oscillator, so will update on the RCA thread - I have just tried some passive frequency response simulations and - well - at the moment I dont understand what I am seeing..)

To better understand the capacitor that is often found in series with the linearizing inductor in many Theremins, I added the element "Cser" to my spreadsheet simulator:

http://www.mediafire.com/?wozcnx2vvb7j3z6

The upshot: I think it's mainly there to block DC.  It doesn't seem to help linearity, and it hurts sensitivity if the value is too small.

I have just understood the puzzle I had when doing passive frequency response analysis of the LEV oscillator..

It looks like "The coil ends are shorted at AC" - but they are not! The capacitor from +V to ground does not in fact short the signal to ground in the way one sees with the EW oscillator - because this point is the current source for the whole series LC circuit - it is as if one was applying the excitation to this point!

This occurs because the Triode / Fet pulls current through the upper inductor from the +V supply, which induces current in the lower inductor, and the capacitor in the middle simply sets the resonant frequency.. It is always, purely, a series resonant circuit!

I think the implication of this is that both the upper and lower inductances do need to be the same value.

Fred.

"It is always, purely, a series resonant circuit!"  - FredM

I agree (and I was wrong) series resonant.  But C15 and C21 form an AC ground to both ends of the coupled coils, no?  I don't know tubes, so I'm not sure what the DC operating points of the ends of C22||C23 are - do you know?  (With no tube present it would obviously be Vcc & ground.)

Ah, Fred, I see you have answered my questions over on page 6 of the RCA THEREMIN thread with a most useful schematic and simulation, thanks!

The grounding of the coil ends I was talking about was not so much about diverting a signal to ground, but more a way to anchor the oscillation center points.  Indeed, your simulation shows the coil ends oscillating about +12 and ground.  Coupling the coils and going series resonant is a clever method that allows the use of an inverting gain element (tube) to be used, and employs the gain element quite simply about its natural bias points.  The coil is more complex, but the overall circuit is symmetrical and quite elegant.

Woody Allen said "90% of life is just showing up."  IMO 90% of the engineering game is just finding the right circuit topology.