Music and Noise

Bazaar, fairs, displays and similar functions are usually noted for noise, and sound-producing electronic gadgets of various types can be made effectively to compel attention. The Theremin in one form or another is possibly the most popular since it combines a degree of mystery with its wide range of notes.

The Theremin (named after its inventor), has a circuit such as that shown in Fig. 9. It can be seen that the basis of the device consists of a pair of oscillators heterodyned together to produce an audible beat note. The controlled oscillator, built round V₁ in Fig. 9, has a capacitance control rod connected to the tuned circuit in much the same way that a capacitance relay has an aerial or plate. In the case of the Theremin, however, the aim is to vary the frequency of the oscillator rather than its amplitude. The second oscillator, built round the triode section of the frequency-changer type valve V₂ must be very stable in operation. Both oscillators are tuned to identically the same frequency with the control rod of V₁ left clear and open, a zero-setting tuner, C₁₀, being provided for a frequency adjustment on the fixed tune oscillator. Under these conditions no audible beat note is produced.

When the hand approaches the control rod of V₁ the frequency of the V₁ oscillator varies, and the frequency change increases with the nearness of the hand to the rod. The oscillator outputs then combine to produce a heterodyne note in the hexode section of V₂, this audio output being passed via C₁₃ to the audio amplifier V₃, and from there to the output stage V₅. Unwanted high frequency components are filtered out by R₈, C₁₂ and C₁₄.

A normal volume control can be fitted to the Theremin and operated by the other hand or by foot control, but a capacitance volume control adds to the novelty of the circuit. This is provided by a further oscillator built round V₄ and fitted with a second control rod, V₄ acting in much the same manner as a capacitance relay valve. One section of the double triode is connected as a diode and oscillator voltage is fed to this valve, is rectified and appears as a bias across R₁₆ from which the required controlling voltage can be tapped off and applied across R₁₂. With V₄ oscillating and its control left clear and open R₁₆ is set to give very low volume from the speaker, the preset volume control R₁₀ having previously been adjusted for the maximum required volume without distortion. If the hand approaches the V₄ control rod the oscillatory voltage is reduced; the bias on V₃ falls, and the volume rises, the output being at maximum when the hand is close enough to the V₄ rod to cut off the oscillator completely. The V₄ oscillator is set for ultimate sensitivity by adjusting C₁₇; L₃ can be tuned by a 100 pF variable tuner if desired.

With considerable practice it is possible to produce a recognizable air from the Theremin, and experiments with tone control networks or filters between the audio amplifier and output stages make it possible to obtain various tonal effects; but perhaps the most simple and satisfactory way of using the circuit in a display item is to preset the volume to a bearable level and make the note control rod part of some novelty. One idea which has been tried with complete success is to conceal the Theremin with its loudspeaker behind a large teddy bear, the control rod being a staff held in the bear's paw. Place at its end the notice "Don't Touch, I Growl", which invariably ensures that every passer-by reaches out to touch it. As the hand approaches the staff in front of the bear a low growling note is heard from the loudspeaker, rising to a penetrating scream if the hand is further extended. For a purpose such as this the volume control section of Fig. 9 can be completely omitted with the earthy side of R₁₀ taken to the chassis and R₁₁ and R₁₂ deleted from the circuit along with C₁₄ and C₁₅, V₄, and all its associated components.

A display on these lines will be very popular with children, who will show a marked tendency to investigate every part of the circuit. Accordingly it is felt that a full-wave power supply system, with an earthed chassis and control rods is essential for safety. The wiring of the Theremin should be concealed below the chassis as should the terminals of the mains transformer, but a juvenile audience will always appreciate being allowed to see the valves, etc., on the chassis top.

If the Theremin in any form is to be operated for long periods a thorough check should first be made to test the effects of temperature rises and changes so that any zero drift may be discovered and corrected. If the top of the chassis is left open and well ventilated the oscillators should settle down quickly to stable operation, but if the chassis is enclosed in a case a long slow drift might become evident. In such circumstances it would be found helpful to use for C₉ a capacitor whose value decreases when its temperature increases. This property is known as a negative temperature coefficient, and many types of ceramic capacitors possess it. If the drift is rapid the whole value of C₉ should consist of a ceramic capacitor but for slow drifting a proportion of C₉ might be made ceramic, the remaining capacitance being provided by an ordinary mica capacitor, the two capacitors being connected in parallel. The two proportions would have to be determined by trial.

The two main oscillators should be screened from each other; a chassis layout such as that shown in Fig. 10 is suitable, under-chassis compartments being made by partitions of sheet metal.

As in the capacitance relay the control rod (or rods, if the capacitance volume control is included) should have a stray capacitance to other earthed objects as small as possible, and when the Theremin is used as a musical instrument the rods should be mounted directly on the chassis or case by means of small stand-off insulators. In a display such as the teddy bear novelty the lead from the chassis to the note control rod should be kept short and as clear and open as is practicable.

In Fig. 9 C₂₀ is found by trial to suit the loudspeaker and the constructor's requirements, and is shunted across the primary of the output transformer to by-pass the increased high frequencies common to this particular type of output stage.

It is a simple matter to set up the Theremin when it is first put into operation, although it may be found beneficial to remove the output valve and to use headphones connected in from the grid pin of V₅ to earth, the preset volume control R₁₀ being set well back. Switch on and allow the valves to warm up for some minutes, by which time the circuit will possibly be howling strongly. Set C₂ to almost its full capacitance, set the zero control C₁₀ to its mid-point position (i.e. to approximately 25 pF) and then adjust the trimmer C₈ for silence or no output. Turning C₁₀ in either direction should then cause a low note to be heard. As the unit warms up a touch to C₁₀ may be necessary to give silence with no earthed body near to the V₁ control rod. Serious drifting which makes C₁₀ ineffective as a control, a note being heard no matter what the value of C₁₀, means that a correcting capacitance must be used in the C₉ position as already described.

It is assumed throughout the foregoing that the values of C₁ and C₉ are fairly close. If one capacitor is rather wide in its tolerance it may not be possible to balance the circuit at all at the first trial, no silent point being found. If this is the case note the effect of increasing the capacitance of C₈ from minimum towards maximum. If the note rises in pitch C₉ is higher in value than C₁ and a balancing capacitance of up to 50 pF or so must be added in parallel with C₁. If the note falls in pitch the value of C₉ is lower than that of C₁ and extra capacitance must be added across C₉. It should be possible to produce a really low note with the hand just approaching the control rod. If, from silence, a relatively high note is suddenly obtained as the hand is brought in towards the rod one oscillator is "pulling" the other, and causing the frequency difference between the oscillators to jump rather than move smoothly away from the zero point. Extra screening between the V₁ and V₂ circuits, with extra decoupling if necessary, is the cure. In a display novelty such as that of the bear a really low note is very desirable.

First published 1953, eighth impression 1963
London: Norman Price (Publishers) Ltd.

Schematics

Newspaper & Magazine

Technical