Home-brewed Theremins ... and other strange devices

Theremin Instruments - Questions and Answers (Q & A)

The author has compiled some mostly Theremin construction related questions and answers. If you have further questions, feel free to ask sending an e-mail. Your relevant question will be added.

Note: 'Etherwave' is registered trade mark of Moog Music Inc. 554C Riverside Drive, Asheville, NC 28801 [69]

  1. Question: Can the sound of Clara Rockmore’s Theremin instrument be recreated?
    Answer: Yes, it can be recreated by various circuits (tube or silicon based) when observing the correct volume depended pulse-pause ratios of the generated rectangular wave-form, thus producing the correct spectrum.

  2. Question: Does a scientific paper on the timbre of the Theremin instrument exist?
    Yes, a paper [74] by the Department of Musicology (Historical Musicology, Ethnomusicology and Systematic Musicology), University of Vienna, does exist, which erroneously focuses on the sound spectrum of wave forms similar to the EW Standard Theremin of R. Moog. Similarities of soprano vocals and the ‘a’ formant are emphasized. In addition, the main instrument of C. Rockmore, used for her famous recordings, is denoted as modified RCA Theremin – which is of course not the case. Even if the paper claims to study the timbre of the Theremin, a large part of the discussion is about Leon Theremin’s life and patents. The authors haven’t been aware of the fact that the early patents do not describe the circuits and principles as realized in the Rockmore/Rosen instruments of the 1930’s. The principles of these instruments are described in detail by the author of in 'General Functionality of the Rockmore- and Rosen Instruments according to Leon Theremin’s Design’. The spectral diagrams of the Theremin sound presented in the paper have no expressiveness, because the resolution is very low and the frequency graduation is linear. A significant frequency analysis, as common in AF and RF engineering, is presented by the author of in 'General Functionality of the Rockmore- and Rosen Instruments according to Leon Theremin’s Design’. Apologetically must be said that this paper was compiled in 1999 by musicologists which of course had only very little knowledge in the electronical principles used by Leon Theremin and in electrical engineering. A truly meaningful research on the Theremin can only be established by close cooperation of musicians,  musicologists, and trained (RF-) engineers.

  3. Question: Why doesn’t the Etherwave Standard (EWS) Theremin of R. Moog create a sound alike the Rockmore theremin even if a somewhat rectangular waveform can be produced (depending on the setting of the waveform and brightness knobs)?
    Answer: The rise- and fall-time of the slopes of the generated waveform is to low and the pulse-pause ratio cannot be adjusted to the required times. In addition, there isn’t the appropriate coupling between the pitch oscillators and the pitch antenna extension coils provided, which is essential for the pulse-pause-ratio distribution over the whole pitch range. Another missing factor is that the resulting spectrum is (mostly) volume independent.

  4. Question: Can the Etherwave Standard (EWS) Theremin of R. Moog be modified to achieve a sound close to the timbre of C. Rockmore’s instrument?
    Answer: Yes, this is possible as described by the author in 'Etherwave Standard Theremin Rockmore-Sound Modification (RS-Mod.)'. Several major changes in the EW-Theremin’s circuitry are necessary, the modified sound cannot brought back to the original EW sound by simply adjusting the waveform and brightness knobs!

  5. Question: Is there a simple solution to extend the bass range of the Etherwave Standard (EWS) Theremin by largely maintaining the original sound, which is adjustable?
    Answer: Yes, the solution consists of a simple standard resistor and a small adjustable capacitor as described by the author in 'Etherwave Standard Theremin Modifications'.

  6. Question: It seems that each of the vintage RCA Theremin instruments as well as the Rockmore and the Rosen Theremin have different timbres, even when the circuit is identical. Why is this the case?
    Answer: The Rockmore and Rosen Theremin schematics only differ slightly (while the series fabricated RCAs should be completely identical), but when observing the schematics closely, e.g. the values of the resistors determining the RF amplitude of the oscillators are not identical. In general it must be said that the timbre depends mainly on the pulse-pause ratio of the generated rectangular audio wave-form, which is determined by at least the following factors: RF-amplitudes of the oscillators in conjunction with their generated negative bias, resistance values at the control and screen grids of the phase detector tube, inductance and lumped capacitance of the interstage transformers, conductance/amplification values of the tubes (tolerances), quality factors of the pitch oscillator- and pitch-antenna-extension coils (tolerances) and varying coupling between these coils.

  7. Question: Would it have been possible already in the 1920s to make each RCA Theremin sound nearly identical e.g. like the EW-Theremin, where each of them sounds identical?
    Answer: Yes, but there have to be means included to adjust the instruments. For instance, the working point of the pitch oscillators have to be adjustable - rheostats have to be added which increase cost. Same for rheostats at the control and screen grids of the phase detector tube. All this adjustments mainly compensate the tolerances of the tubes. In addition, to initially adjust the timbre, at least one of the pitch oscillator coils has to be mechanically aligned to adjust the coupling effects of the coils for correct audio pulse-pause ratio. These adjustments would have to be performed by trained persons which had to control the audio pulse e.g. at an oscilloscope screen. All this measures seem to be quite expensive and complicated (the production of the RCA Theremin in 1928/29 was already quite complex and costly compared to a radio-set), so these improvements have obviously been omitted.

  8. Question: Is it possible to adjust the timbre of the RCA or Rockmore/Rosen Theremin instruments without major changes?
    Answer: Yes, there are at least two possible alternatives: The simplest solution is to swap positions of  tubes or replacing tubes with others as long as a tube combination is found which gives the desired sound. This procedure is a quite expensive because a bunch of working #27 and #24 tubes have to be at hand, furthermore it is very time consuming finding the right combination of tubes. Another solution is to slightly change the positions of the oscillator coils to each other or reverse their connections. It might be, that during production, it only has been checked that the oscillators had been working, regardless of the phase relation between both oscillator coils. Furthermore, it is important, that the original interstage transformers are present and their winding connections have the correct phase relation. When the sound is not sufficient, it might help to reverse e.g. the secondary connection wires. The right inductance and lumped capacitance values of the interstage transformers are important to generate the proper sound.

  9. Question: Why is it not irrelevant how the first interstage transformer of a classic tube Theremin between the phase detector #24 tube and the voltage controlled limiter amplifier #27 tube is connected in regard to the phase of the audio pulse signal, even if this is an AC voltage?
    Answer: The secondary winding must be connected in such a way to the limiter amplifier #27 tube that the limited rectangular pulse, which can be measured at the anode (plate) of this tube, has the sufficient pulse pause ratio, thus producing the desired timbre. Usually, the single-sine pulse half-waves are of unequal amplitude and duration. The transformer has to be connected in such a way that the larger half-wave with shorter duration produces the larger output rectangular pulse.

  10. Question: Why do you denote the #24 tube of a classic tube Theremin, which is often referred to as Mixer, as a ‘Phase Detector’?
    Answer: The action of this tetrode tube is to produce a current in the primary winding of the first interstage-transformer which is mainly dependent on the phase of the two pitch oscillator RF voltages. This current of RF pulses gets integrated and produces a single-sine like output at the secondary of the transformer, whereas one of the half waves is mainly caused by the dampened fly-back action (collapsing magnetic field when tube is not conducting). Because of the imperfect phase detection of the tube (which it is not intended to serve for), the stage is somehow an additive mixer, but it has nothing to do with a multiplicative mixer and shall not be confused with it. The multiplicative mixer does only produce minor distortions (RF harmonics) whereas the additive mixer does. The additive mixer is therefore unsuitable for AM radio application and was replaced as soon as multiplicative mixer tubes (e.g. ECH81/6AJ8 Triode-Heptode or 6BA7 Pentagrid-Converter Heptode) had been available.

  11. Question: Why do you denote the voltage controlled amplifier #27 tube of a classic tube Theremin, which is often referred to as VCA, as a ‘Voltage Controlled Limiter Amplifier’?
    Answer: The circuit applied by Leon Theremin is not a simple VCA (Voltage Controlled Amplifier). A VCA, as it is known today, is an amplifier of which it’s amplification factor is dependent on a control voltage, producing only minor distortions. The tube stage, as it is applied in e.g. the Rockmore/Rosen Theremin instruments, is totally over-driven by the large single-sine audio-pulse signal of the first interstage-transformer. This leads to a rectangular wave-shape of the audio output signal. A variation of the input amplitude has only minor effect on the output signal, similar to an limiter-amplifier (e.g. used in FM IF amplifiers). The output amplitude is controlled by varying the anode (plate) voltage of the triode tube, from strongly negative (totally cutting-off and blocking of the tube) to positive (large output), which is quite uncommon but possible because the output is nearly rectangular and residual feedthrough of the control (anode) voltage is therefore no issue (the rectangular voltage rises from the negative anode blocking voltage to the desired positive level; there is only minor voltage offset present). It is not a (soft-) clipping or limiter-amplifier stage, as it is generally applied in audio today, because the intention is not to distort e.g. a sine-wave somehow or limit the signal amplitude with more or less distortion, but to generate a nearly rectangular wave-form, which is amplitude controlled by a steering voltage.

  12. Question: Why do you denote the large coils of a classic tube Theremin connected between the oscillators and the antennas, as ‘antenna extension coils’ (or ‘loading coils’), even if this coils are usually named ‘linearization coils’ (or ‘equalization coils’)?
    Answer: The term ‘antenna extension coil’ (or ‘loading coil’) is a fundamental definition in RF engineering. It denotes a coil which is connected in series to the antenna to electrically match a mechanically short antenna to the wavelength of the radio-frequency it shall operate on. A ‘linearization coil’ is a device (magnetically biased saturable inductor) applied in (vintage) picture tube TV-sets or monitors to shape the current for horizontally deflecting the electron-beam in such a way that a undistorted picture is generated on the screen. As you can see, the term ‘antenna extension coil’ fits best to the functionality of the coil applied to the Theremin. An antenna system applying a coil connected in series to the antenna and the calculation of it is described in: [76] ‘Antenna with Series Inductance’, Radio Instruments and Measurements, USA Bureau of Standards Circular C74, 1924/1937, p.76

  13. Question: How can the value of the ‘antenna extension coil’ (or ‘loading coil’) of a Theremin be calculated?
    Typically, the length of an antenna is ¼ of the wavelength. In terms of the Rockmore/Rosen Theremin instruments, with the pitch oscillators operating around 170kHz (wavelength of 1764m), the antenna rod has to be 441m to match ¼ of the wavelength. This is not possible because of obvious reasons (the length of the RCA pitch antenna rod is about 0.45m). Of course, the antenna rod has a certain capacity to ground (typically about 7..10pF in terms of the Theremin). This capacity determines in conjunction with the series connected extension coil the resonance frequency of the antenna-system. At the Theremin, this resonance frequency is not equal to the frequency of the connected oscillator, it is slightly lower to just pull the oscillator and not lock it to it’s own resonance frequency. An antenna circuit resonance-frequency of e.g. 165kHz at an antenna capacitance of 7..10pF would lead to a necessary extension coil inductance of 133..93mH. A usable calculator for antenna loading coils can be found here: ([75] 'Off-center loaded dipole antennas', J. Hall, QST Sept. 1974, p.28-34).
    At the Theremin, assuming a (pitch-) antenna length of 0,4m, this leads to a ¼ wavelength (441m) to antenna length ratio of approx. 1/1100 which is extremely high. Usually, for RF antennas this ratio is quite low (e.g. 1/5). There have been calculations published which are only usable for low ratios [79] ‘Hochfrequenztechnik I.’, J. Kammerloher, 5. Auflage, Lehrbücher der Feinwektechnik/K. Gehlhoff/Berlin, Winter’sche Verlagshandlung, Leipzig, 1941, page 141. At the Theremin, because of the very short vertical antenna rod length compared to the wavelength, the current and voltage distribution at the antenna rod has little effect. The current and voltage distribution is concentrated to the antenna extension coil. Therefore, this coil can be regarded as sort of top loaded tesla-coil. The coils roof/top capacity (to ground) is formed by the antenna rod.
    When assuming an antenna rod capacitance of 7.6pF, the extension coil value can be calculated to be 115mH at a frequency of 170kHz using the well known formula for the series resonance circuit. The ‘Off-center loaded dipole antenna’ approach leads to an inductance of 115mH for a antenna length of 0,5m with diameter 9.5mm at a frequency of 170kHz (coil position, distance from feed-point = 0m). It must be noted, that the calculated values can only be a rough estimation for the coil value because various influences like the distance to (virtually) grounded objects at the Theremin housing (effective extension coil roof capacitance) usually reduce the required coil inductance value. It is recommendable to use adjustable coils (e.g. by ferrite core or trappings at the cold oscillator side of the coil).

  14. Question: Does the Theremin radiate RF energy?
    Answer: Yes, it produces electromagnetic fields. The radiated RF energy is quite low due the low power oscillators and the antenna resonators not tuned to the resonance peak. Tuning to the resonance peak is normally applied in RF transmitters for maximum RF radiation.

  15. Question: Is the RF energy which is radiated by the Theremin antennas important for the correct function (e.g. pitch range, volume range, sensitivity to disturbances)?
    Yes, the voltage at the antennas and thus the RF current flow has to be of appropriate height to achieve a sufficient signal to noise ratio (S/N).
  16. Question: What does electromagnetic field’ mean?
    The term electromagnetic field includes the electrical (capacitive) and magnetic (inductive) fields
    , which are inseparably linked. At short distance of the Theremin antenna, the magnetic (inductive) field is extremely small with respect to the electric (capacitive) field which dominates, refer to [78] Ch.23 p.664 and [47] p.32. Therefore, the antenna of the Theremin can be regarded as one plate of the capacitor formed by the antenna and surrounding grounded objects like mains protection earth (PE), where PE serves as the second capacitor plate - just as well as the virtually PE connected hand of the player.
  17. Question: Why is there no circuit of the ‘Rockmorizer Silicon Theremin’ provided yet?

    Answer: A description of this Theremin prototype including schematic will be prepared in the near future. There’s a schematic existing in an EDA CAD format which is abandoned by the author; it must be converted or redrawn before presentation.

  18. Question: Why does it take so long to update the homepage, could the design and presentation of working instruments be somewhat accelerated?
    Answer: No, all must be regarded as a hobby. Research, development and documentation work is very time consuming, where documentation is the most exhausting part (and therefore often omitted). The author has to spend his main time in earning his money at his daily job, which is not connected to the research and development of Theremin instruments.