Did someone try to use PLL in volume antenna cirquit?

Since variable pitch (antenna) oscillator output is basically FM signal, why not use PLL like CD4046B in FM demodulator mode to decode its signal?

Set PLL parameters to track input frequency and produce output voltage e.g. in range 0.2..0..8Vcc for different hand positions.

This signal is easy to measure by digital theremins, and easy to use in analog theremin (move its range, and even do some linearity correction, and then feed to VCO).

Yes, FM, exactly.  You can do this, but you're effectively using an RC oscillator as a reference for an LC oscillator.  The RC is less stable, and I think that would lead to tearing and drift problems.  Better to use a tunable LC as the VCO in the PLL, which would work much better IMO.

Our most experienced embedded engineer, FredM, who unfortunately passed away too early, always said that a PLL in a theremin was an overkill.

What Bob Moog did in the series 91 theremins could be of interest for you: He started from a 2.8MHz crystal oscillator, did some uneven or prime number dividing to obtain two reference frequencies where none would intersect withe the harmonics of each other. These square signals were fed via small (22pF) capacitors into parallel resonant circuits with relatively small L and big C (and good old variable capacitors for tuning) to which the antennas were connected through linearization coils as in a classic heterodyning theremin.

Analog phase comparators were used between the original and the dephased signal (before and after the small series capacitor) to obtain the analog pitch and volume CVs. The volume CV went to a LM13700 VCA and the pitch CV controlled another LM13700 as an audio oscillator, outputting a triangle wave which went then through the classic Moog style non-linear filter wave shaping.

"Yes, FM, exactly.  You can do this, but you're effectively using an RC oscillator as a reference for an LC oscillator.  The RC is less stable, and I think that would lead to tearing and drift problems.  Better to use a tunable LC as the VCO in the PLL, which would work much better IMO."

My idea is to use LC oscillator identical to one used for variable pitch, as a VCO by adding of varactor. The it would be as stable as antenna oscillator. Will try to test this in LTSpice.

"Our most experienced embedded engineer, FredM, who unfortunately passed away too early, always said that a PLL in a theremin was an overkill. What Bob Moog did in the series 91 theremins could be of interest for you: He started from a 2.8MHz crystal oscillator, did some uneven or prime number dividing to obtain two reference frequencies where none would intersect withe the harmonics of each other. These square signals were fed via small (22pF) capacitors into parallel resonant circuits with relatively small L and big C (and good old variable capacitors for tuning) to which the antennas were connected through linearization coils as in a classic heterodyning theremin. Analog phase comparators were used between the original and the dephased signal (before and after the small series capacitor) to obtain the analog pitch and volume CVs. The volume CV went to a LM13700 VCA and the pitch CV controlled another LM13700 as an audio oscillator, outputting a triangle wave which went then through the classic Moog style non-linear filter wave shaping."

Thank you, Thierry!

It's very useful information.