The document below was found by John Snyder during his Caramoor research many years ago. Lucy Rosen owned that country estate, and maintained two custom theremins made for her by Professor Theremin. The typewritten photocopies John rescued were apparently originally transcribed from shorthand notes. The document contains numerous grammatical errors. I edited here and there for better clarity. Lucy Rosen obviously cared for her instruments. She created not just a repair record, but a theremin primer to the benefit of posterity. Major thanks to John Snyder for rescuing this lost information, and to the late Lucy Bigelow Rosen for committing this data to paper. (Reid Welch)
Pitch Antenna Tuning
The tuning method is similar to the previous R.C.A. model.
Frequency is slightly higher and capacity of the resonating coil is smaller. That increases the possibility of making the playable range bigger (near five octaves). To obtain correct tuning, the antenna coil must be tuned on a slightly lower frequency than the variable oscillator. When the hand reaches the greatest distance from the antenna, the system comes close to resonance. If on these conditions the circuit comes too close to resonance, breaking out in high frequency oscillations can occur. To obtain maximum musical range, this resonance must be close as possible to critical, but still far enough to prevent sudden periodic breakout of high frequencies. The correct tuning can be obtained by means of the variable trimming condenser in the antenna oscillatory system. Increase of that condenser increases antenna sensitivity (making the resonance lower)
Constant Oscillator Tuning
The constant oscillator must be tuned in such a way that when the position of the pitch tuning condenser on the front panel is approximately in the middle of the dial, both oscillators (constant and variable) will be in resonance (zero beat). This can be obtained by means of a trimming condenser in the constant oscillator circuit. To obtain the best possible stability of both antenna and constant frequency oscillators, grid resistances have a certain critical value corresponding to a maximum negative potential on the grids. Further increasing these resistances will create an interruption in high frequencies related to acoustical pitch.
Inter-Coupling Between Oscillating Circuits (antenna and constant)
A double-grid tube detects and mixes the two high frequencies for further amplification and volume control. …(sic) which creates coupling of both oscillators. Correct value of this coupling is very important to obtain a desirable tone-color and correct distribution of the musical intervals in the space around the antenna. By means of the phenomenon of pulling into synchronization of both high frequency oscillators (antenna and constant) whenever they get near the resonant point, certain distortions of the sinusoidal wave form occur, as evidenced when differential (beat) frequencies appear. These distortions create desirable acoustical harmonics. Besides this effect, when the antenna oscillator, by playing low acoustical frequencies, approaches the resonant point with the antenna coil system, the tendency of synchronism decreases, which tends to increase the distance between the intervals in the space around the antenna.
Therefore, the correct value of resistance in grids of mixer tube controls tone color, and distribution of intervals in space and the optimum value to obtain enough output from mixer tube.
Tuning of Volume Control System
The left antenna of the instrument is attached to a high inductance/low capacity coil to the oscillating circuit. Resonating circuitry is placed in the grid circuit of the detector tube. Near the resonance point the current in its plate circuit increases. This increases drop of the potential in the cathode resistance. The cathode potential is used to supply plate or screen current for the first amplifier tube. To obtain more sharpness of tuning, a grid of the detector tube (24A) is supplied with the negative potential from dry cell battery. The other tap of the same battery serves the purpose of supplying the plate (or screen) of the amplifier tube with a negative potential when no drop of voltage occurs in the detector tube cathode resistance. To obtain correct tuning, the resonating circuit must be adjusted by means of the trimming condenser on the front panel of the instrument, which is connected in series with the other trimming condenser for spreading its range. Auto transformer coupling of the volume antenna is tuned with the oscillator circuit, similar to the pitch system.
To indicate correct pitch of the instrument before any sound is produced, a little neon lamp is used. Two stages of amplification are attached to the plate of the mixer tube. The second tube has in its grid circuit a choke and condenser in parallel, resonating on the frequency corresponding to the musical pitch “A”. Exact tuning can be obtained by adjusting the value of this condenser to proper resonance with the
frequency of “A”.
Tone Color Control
On the front of the instrument is located a ten position switch which changes the value of capacity in the plate circuits of the output tubes.
The first (left) position corresponds to zero additional capacity, which produces sound with the maximum of high frequency harmonics. Rotation of the switch to the right connects larger capacitances, and in this way limits the high frequency harmonics. If it is necessary to make other changes in the tone color of the instrument, it is possible to use some of the last positions, which are at present left free, so that some other combinations of capacitance, or capacitance with resistance may be used.
Automatic voltage regulation is important to eliminate changing of the pitch or volume of the sound due to unstable line voltage, or to compensate for the changing load on the power unit with audio volume alteration.
The stability of high frequency circuits independent of voltage drop is very good, but still, in the low range of the instrument, pulling into synchronism is considerably affected by slight voltage variation. This causes changes in pitch related to volume. To eliminate this effect, a series pair of voltage regulating tubes (874) are used, or an additional loading tube (43) is applied, in which its grid is connected to the fluctuating line through a condenser and plate, producing a load on the power supply through a resistance leading to the high frequency circuits.
Change of this condenser in the grid of the tube compensates for fluctuating rail voltages.
Residual Tone – The Hum
Shown in the diagram of the latest instrument of Sept. 1938, a shunt condenser is installd across the 27 to shut off (or allow) leakage from plate at the will of the player. The object is to have sufficient leakage tone during concert hall work so that the artist can hear the pitch before the note is played, but the audience cannot.
Questions Answered by Professor Theremin Regarding Care of His Instruments:
I. What tube, or other things, control the neon pilot light?
On the AC Theremin instrument of January 1938, two tubes, type 57 on the diagram; on the September 1938 instrument, two tubes- one 57 and one 58.
The neon light should be tuned to accuracy by the maker of the instrument (see Professor’s own notes on tuning) and the modern instrument should only require changing the bulb when it burns out.
On the Jan. ’38 Theremin, the .05 variable condenser parallel to the choke regulates the resonant pitch determining when the lamp lights: for higher resonant pitch, use smaller capacity.
The .001 condenser in the input, and the .002 in the grid circuit control the sharpness of the neon light tuning.
Increase capacity for more brilliance, but less sharpness.
Decrease the capacity for sharper tuning (i.e. accuracy), but you give up some brilliancy for it.
Hardness of the ageing tube will in time cause diminished brilliance, but increased sharpness, till the light is entirely lost and useless for a guide. You can change the tube before this; but if both light and sharpness are defective with a new tube, the trouble is proved not with the tube, but with the circuit.
II. How are the tone colors made?
By a group of condensers beside the battery, which add overtones according to their arrangement. A knob in front of the player is connected with each one of these, and can be turned to call forth any one of them.
Or in technical language, the switch is connected with different capacities paralleled to plates in the output transformer, through cables nos. 10 and 11.
Note: Number 1 position on switch has no capacity in parallel, and No. 7 has .1mF.
In the cabinet of the September ’38 instrument, the capacities are in order of magnitude from front to back.
The 1,000 Ohm and 15,000 Ohm resistances in the grid of th mixer tube 24A control the harmonics on the original, or “natural” sound of the instrument. Therefore, the tube affecting tone color is the 24A. To get a more brassy sound increase the number of high frequency harmonics in the
output circuitry removing shunting resistance and capacity out, or by diminishing them.
Other changes in tome-color are affected by the player’s right hand vibrato, or the left hand volume control: or again, by the condition of the tubes. If the tubes or condensers deteriorate, the weak ones must be found and changed.
Removing the 100,000 Ohms resistance in the 1938 instrument will give a trumpet tone.
III. What are the signs, i.e. sounds, that show a deteriorating condition of the battery? (A radio “C” battery, D>C> 22.5 volts is used for regulation of the volume control system)
As the battery gets near discharge, the sound grows louder and unequal, or it starts suddenly. The left hand response is spasmodic, and the volume knob requires constantly to be turned further and further down in the reducing direction. A sharp breaking in volume with no movement of the left hand indicates a defective battery. If the battery is completely worn out, the sound cannot be stopped at all by the left hand. If the battery is merely low, it is impossible to reduce the volume by turning the volume control knob to the right.
Scratching sounds can be caused by defective contacts in a worn battery.
IV. What is the cause and use of audio leakage, the “hum of the electric instrument?
The “hum” is caused by capacity resistance interconnection in the volume modulating 27 tube. To regulate the hum, a condenser is put between the grid of the 27 tube and the plate of the previous stage, mixer tube 24A. To increase the hum, increase the capacity. In the latest instrument (September 1938), the condenser is adjusted to a point near to neutralization, i.e. very small hum.
The ideal is to have a nearly silent instrument, but to be able to increase the hum in case of playing to a great distance through many intervening sounds, or with the noise of an orchestra beside the instrument.
V. What part of the instrument has to do with the staccato, or speed of the left hand, and the separation of notes, i.e. sharp attack?
This depends upon (a) voltage of the “C” battery. With higher voltage and the same height of the left hand, one obtains a sharper staccato, i.e. influence on volume.
(b) Also, if the capacity of the one microfarad blocking condenser is decreased in the circuit of the type 53 tube, a sharper staccato is obtained. Too much staccato will give a knocking effect upon fast rise of the left hand, so you have a limit here. To slow down the staccato, increase the capacity of the one microfarad condenser.
– July 15, (1940) – The September Concert Theremin had stopped one day, gently but firmly, for no apparent reason after playing beautifully for weeks with full, clear tone, good pitch and easy left hand action. We had recently changed all tubes, and could not think where to look for trouble, so brought the instrument into town to test by a radio engineer.
The tests were made with a Tripplet Model 1200A Volt-Ohm-Milliammeter for DC and AC currents.
We began by testing the amplifier. The first tests showed at once no “B” or plate supply voltage for the power tubes. This would make the theremin silent. It might be caused by a defective choke or condenser in the power circuit. The power circuit has a choke for the two 6L6 tubes, containing resistance of 200 ohms, an inductance of 8 Henrys rated for a current of 150 mA. This choke was worn out, so the 6L6 tubes did not heat.
The power circuit has also four condensers in the amplifier, two of 40mF, one of 20mF and one of 30mF, which was burned out, and changed with the choke.
Continuing the tests in the cabinet, one resistor of 800 Ohms in the back was found burned out, connected with the output circuit (choke and metal tubes) in the amplifier. This resistor was replaced, and another near it, also of 800 ohms, to make sure of the equal strength of both.
Following these replacements with a soldering iron, the instrument recovered its voice with an especially strong, clear tone. The work took a little more than two hours.
– July 22, 1940 – The January Theremin suddenly made scratching, crackling noises when warmed to play on saturday (two days ago), so today we sent for the local radio engineer, Mr. Doyle of Mount Kisco, to test tubes and other sources of trouble. We found three defective tubes- one 59 and one voltage regulator type 874 in the amplifier, and a 24A in the cabinet of the instrument.
The fresh tubes much reduced the offending noises, but they remained potentially audible, to be heard again with every left hand movement increasing the volume. This indicated that the trouble might be in the battery or its contacts, which we had found before to be the cause of the same scratching noises- from grit on the contacts, for instance, or corrosion. We had recently renewed the battery, so we tried polishing the prongs, stopping to hear the sound from time to time until all noise disappeared except the musical sound of the instrument.
The noises did not return, so we have apparently done all that is necessary this time. The instrument is playing with full force and clearness.
Professor Theremin advised keeping strips of emery paper for such polishing of corroded surfaces, and Mr. Doyle says that to-day, most radio engineers use Carbon Tetrachloride from a screw-top bottle (it evaporates quickly) applied with a small paint brush. It is completely effective, and leaves no grits.
(Editor’s note: Carbon tete is poisonous. It has been banned for decades)
– July 24, 1940 – Decrease in volume. This may be caused by several things, and is important because of the effect on tone quality of the instrument.
There should always be power beyond what the player’s hand would release or the sensitivity to the hand is destroyed.
The volume is controlled in playing by the left hand, and like the pitch for the right hand, has its own special field or circuit. Loss of volume may be caused by low voltage, or drop in voltage from normal or tubes wearing out; see particularly the power tubes in the amplifier; or, at more rare intervals- the need to readjust the set screw of the volume regulating condenser in the cabinet.
This week brought a noticeable change of tone-color of the September Theremin, which, after listening attentively for several days, was definitely connected with loss of power. It was observed that the player’s volume control knob in the front panel was pushed in to its extreme degree, with only diminishing power for the result. Examination was needed.
After testing our AC voltage lines and the tubes (which were, anyway, new), we believed that the violent change in temperature to extreme heat and humidity had either altered, or made inadequate, the setting of the volume screw.
This screw was identified, its position marked with a pencil, and a very delicate hairline alteration (after some experiments), gave us the needed increase in volume, and with it, depth of tone.
Note: the tubes especially connected with power in the September theremin are the two metal 6L6 output tubes.
– Nov. 25, 1940 – Sent the January Theremin in for examination on account of escaping sound even when switches were down and closed. Had tried changing the battery in the cabinet, but it made no difference, and clearly was not the cause of the trouble.
The mechanic’s tests in New York showed the volume circuit to be out of order, or “de-tuned”, and a defective switch was changed.
Other extra sounds or “hum” besides the playing note suggested examination of the amplifier for general tests. We changed (1) a 200 ohm resistor, (2) a 50,000 ohm resistor, and (3) the output transformer, as all showed deterioration and change from original strength.
The tone quality remained unsatisfactory, with occasional distortion.
We checked inside the cabinet to find the reason for this.
We adjusted the mixing action of the 24A tube, located between the two 27 oscillator tubes, by taking off the resistor which had grown too high by deterioration. This
entirely cleared the tone.