Notes about the RCA theremin oscillators by Reid Welch.
Both RCA oscillators are identical, two-layer coils wound on a tubular fiber form about 1.5″ diameter by 1 3/4″ long.
The wire is single silk covered (“SSC”) enameled wire approximately #30 B & S gauge.
Turns per layer- about 75.
The winding length is around 1 3/4″
A couple of turns of black “book cover cloth” is wrapped between layers.
Inductances: inner layer =158uH actual. Outer layer= 177uH actual.
Note: The inner and outer inductances are unmatched because the outer layer, having a greater diameter, happens to offer more inductance.
General hook up: The coil layers are wound in opposite directions. To wit:
If the inner layer is CW, the outer layer is CCW. Also, the oscillator system will not work if any one layer is connected in reverse.
RCA stock connections:
Inner layer “start” goes to cathode. The ’27 cathode is standing on a bypassed 5k resistor to ground.
Inner layer “finish” goes to the grid.
Outer layer “start” goes to the bypassed B+ rail.
Outer layer “finish” goes to the plate.
Note: As said before, if connections of either layer are reversed there can be no oscillation. OTOH, the layers are interchangeable, but that’s not necessarily optimal.
RCA uses a cap of about 1100 pF between grid and plate on each of the fixed and variable oscillators. About 50pF of adjustable capacitance is in parallel with the larger C for the purpose of trimming oscillator frequency. The frequency runs about 175kHz.
RCA-style oscillators are easy to make. The first (and only) ones I’ve made were wound on the paper tube core from a roll of towels. These operate today in the Stout. I merely approximated the RCA specs, using smaller wire (32 DCC) and a little less inductance (150uH) than the RCA and a C of about 900pF. The frequency is 212kHz.
Larger wire is surely a good idea.
An LCR meter is handy if you work with coils. Equally valuable is a frequency counter. The Radio Shack 22-168A multimeter measures frequencies up to 1mHz.
Place theremin oscillator coils as far apart as possible (six inches minimum) and preferably not on or near a metal chassis. Orient the coils 90 degrees from each other. Oscillators drift with the slightest temperature change, so if there is a slow, continuing temperature rise inside the cabinet pitch drift will continue indefinitely.
[Copyright 1997 by Reid Welch]