Possibly a "new" theremin topology, and links to related background

Christopher wrote:

If the ghost tones go away when using batteries on your EtherWave Standard my experimental fix would be to put a 100 ohm resistor in series with the +12v connection above each oscillator section. Add a 470 uf capacitor below the 100 ohm at the positive side of the osc circuit to the negative rail. This is a very low current spot and this extra filtering can work. I use this added method in my theremin designs to completely filter out noise using even the cheapest 12v switching wall-warts. Split power supplies are an engineer thing to make things complicated, often not needed. That is what makes battery powered busking more complicated with the EWS.

The problem which Ruslan has to deal with in Russia is poor grounding. That means that the ground wire in most wall plugs (if it exists) is either directly connected to the Null wire or has a high resistance towards true ground. In both cases, the ground wire carries dirty signals which couple through every transformer and power supply. Thus, operating it on batteries is most probably the only solution.

I strongly object to the kind of additional filtering which Christopher suggested, at least for the fixed pitch and the  volume oscillators of the EWS or Plus because this will make the active tuning stages unstable, causing them to oscillate themselves around 300Hz which gives horrible results. 

Thierry said: "The problem which Ruslan has to deal with in Russia is poor grounding. That means that the ground wire in most wall plugs (if it exists) is either directly connected to the Null wire or has a high resistance towards true ground."

Now that would be very frustrating, I believe a good ground is key to a theremin operating at its best. My solid state design will only give me the audio wave shape I like with a good earth ground where by my tube/valve front end oscillators don't care as much about ground, this is something I always found interesting. I always attributed this to solid state responds fast or sharp and tube/valves have a sluggish behavior.

I still think my experimental fix might help the Moog oscillator design, I will admit I don't have a clue how that oscillator design works. My whole life is about staying well grounded. (-'

Christopher

What is that all about, did you have enough raisins in your oatmeal this morning?

Christopher

"If I am correct, I believe your digital theremin operates with constant frequency antennas (?)"  - FredM

I hope I'm not forced at some point into going there.  That would mean some kind of tank tuning mechanism and I'm really trying to avoid that.  I want to be able to stick any random antenna on it and have it just work at whatever frequency gives maximum voltage swing (pure LC resonance).  Interference might be dealt with by changing the antenna length or similar (a miracle occurs here). 

It's likely that I'll sample the DPLL offset and run the PWM clock at the audio sample rate - those kinds of things.

I do think the upside down oscillator makes a heck of a lot of sense though.  And your name for it is cool!

I'm not sure that the upside-down concept especially with pitch and volume circuits operating at the same frequency will work in reality. A PLL running at several hundreds of kHz will forcibly have some frequency and phase jitter which risks to add a low frequent "blubber" to the audio signal. And the two VCOs (volume and pitch) would probably tend to lock together through stray capacitance. 

The 4046 PLL chip existed already in the late eighties and it was cheap, but Robert Moog decided to go rather through analogue components with large and expensive coils and variable capacitors, analogue phase and amplitude detectors, etc. just to get two simple but clean CVs for pitch and volume in the 91 Series without too much low pass filtering which would have given a delay in reactivity. There must have been a reason...

There must also be a reason why music from a FM tuner with quadrature demodulator sounds more "round" while PLL demodulators have sometimes the tendency to sound somewhat harsh. And that's with a well filtered audio range from 50 to 18000Hz only. But in a theremin volume circuit, we'd have to go down to DC.

"You are  "sweeping" the PLL "reference frequency" until you obtain a lock, and then tracking changes to the antenna frequency by adjusting the DPLL "reference frequency" (which is, in fact not a frequency, but a numerical operator) to keep it in lock. (?)"  - FredM

Yes.  The DPLL NCO kicks out a wave which is used to drive the series tank L.  The DPLL phase detector compares the phase of this to the output of a secondary winding on the series tank L, which is 90 degrees delayed at LC resonance.  The DPLL loop filter has a phase error integrator which is conveniently low pass filtered by the DPLL construct.  It's pretty much crying out to be used in a Theremin.  The whole thing is actually more of delay locked loop and could perhaps be implemented that way if one could get sufficient long-term error information (NCOs are handy here because they can be made arbitrarily accurate by simply increasing the bit width).

Crazy idea #10665:

Fred, it occurs to me that if you are going the upside down shielded route then you could perhaps somehow vary the shield / antenna capacitance to tune the antenna C and series tank L to the fixed frequency your are driving it at?  If done actively and at least somewhat linearly then you could use the "error" from nominal (null) for this control as your pitch indicator.  Crazy?

I have this crazy feeling that all of these crazy notions are so beautifully crazy I'll be forced to use them myself!

Crazy idea #10666:

With antenna shielding, does it really matter all that much (in terms of unwanted interaction between volume and pitch, or between multiple Theremins physically close together) if the antenna frequencies aren't locked?

Ok, too much crazy ideas, I'm out from here for the moment. Hell, we are talking about a music instrument (the theremin if I remember well) in which the electronics are only means to an end, where musicians will not care about frequencies, phase comparators and mixer topologies, but they'll judge if the instrument allows them to give their art an expressive voice...

I fear that "we engineers" (I exceptionally allow to include myself although I studied mathematics and not engineering) tend sometimes to loose the professional musician as the final target or client of our art from sight. Or did I miss something? Is it now about building penny whistles for everybody?

"Is it now about building penny whistles for everybody?"  - Thierry

Perhaps truly modern Theremins are like laws and sausage, where one is better off not seeing them be made.

"By any means necessary" is my motto (as long as it isn't too difficult).

I kind of don't get your argument, which seems to be that too much deep thought about technical matters somehow takes the designer's eye off the ball.  Not getting the technical part right means, by definition, that one is incapable of catering to the wants and needs of professional musicians.  All the caring in the world won't make a good Theremin.

The great flaw in theremin designers and engineers (and I include the late Dr. Bob in this) is that they THINK they know what musicians want.

They don't.

You have only to look at the Etherwave Pro debacle. The instrument had more design mistakes in it than the Titanic. 

The great effort at consultation with players that builders go through is an empty exercise in theremin fellowship and camaraderie - it's terrific fun but it is empty and has borne no fruit.

I think all the talk about theremin building and the "new topology" and revolutionary "mostly digital" designs etc., is GREAT and a lot of people really love it, but it has been going on for years and has produced nothing of value that I am aware of other than pleasure for those who participate in it.

There's nothing at all wrong with that but I think we need to step back and recognize it for what it is.