Fullscreen HTML5Med Res (??MB)HTML4

m e r e h T i n i m 2 . k M Just move your fingers near the antennas of this Theremin to create your own electronic music or eerie science fiction sounds. It’s easy to build, easy to set up and easy to play. PART 1: By JOHN CLARKE Main Features • • • • • • • • • External pitch & volume span adjustments Linear pitch change with hand movement over four octaves Linear volume control with hand movement Three sound variation controls Signal level adjustment Internal loudspeaker with headphone listening option Loudspeaker/headphone volume control Line output with muting switch for amplifier connection 12V DC operation from plugpack or battery 24  Silicon Chip siliconchip.com.au min when it comes to playing a Theremin easily. It is also critical to ensure the same range of hand movement for each octave and that none of the octaves are compressed into a tight range (which would make playing difficult). As a result, this unit has been designed to provide excellent linearity when it is adjusted correctly. This has been made easier by a special test circuit that’s used when setting up the Theremin Mk.2. Tonal quality O UR ORIGINAL MK.1 THEREMIN was described in the August 2000 issue and has proved very popular. This new Mk.2 version features a better waveform, has more controls for adjusting the tonal quality and is easier to play, with more progressive hand control. So why are Theremins so popular. The answer is because of their extreme versatility and the uniqueness of the sound they produce, compared to conventional instruments. Even relatively simple hand movements can lead to complex and engrossing performances. Many Theremins produce only simple tones but some Theremins – such as the unit described here – also allow adjustments to the tonal quality, so that the performance can be altered to suit the mood. Typically, the controls allow a range of sounds that can be varied between raspy-edged tones through to pure sinusoidal notes. The resulting sound can consist of smooth gliding tones (glissandi) or it can comprise separate notes (staccato) or a combination of both. siliconchip.com.au It really is a versatile instrument that is only limited by the skill of the player. Controls In order to play a Theremin, you must you must be able to accurately position your hands (and fingers) to produce the required tones. The more accurate a Theremin is in producing the same frequency (or tone) for a given hand position (ie, distance from the antenna), the easier it will be to play. Similarly, the volume control needs to produce a consistent effect in response to hand movements. The SILICON CHIP Theremin Mk.2 has been designed to provide good consistency for both the pitch and volume “antenna controls”. In addition, two external controls have been provided (on the front panel) to adjust the pitch (Pitch Range) and volume (Volume Span) settings. These are required to compensate for any changes that may occur over long periods of time or because of temperature changes. The linearity of the response to hand movements is another critical feature Three further controls are included to adjust the tonal quality or “voice” of the Theremin. The most popular “voice” setting reproduces a cello sound at the lower frequencies, changing to a soprano voice at the upper frequencies. This tonal “voice” creates an interesting backdrop against other instruments, such as a piano or violin. If you are interested in hearing some fine Theremin performances, log onto http://www.peterpringle.com/thereminmp3s.html In operation, the “voicing” can be altered to suit using the Waveform, Symmetry and Skew controls. Each control produces its own characteristic variation in the sound. The Waveform control varies the shape of the signal reproduced by the Theremin. At various settings, the unit produces waveforms that are somewhat triangular in shape, while at other settings it produces either squarer wave shapes or more sinusoidal waves. Each wave shape has its own distinctive timbre, the squarer wave shapes producing a hollow sound similar to that produced by a reed instrument. The more triangular waveforms are less hollow, while the sinusoidal shapes gives a neutral or pure sound. The Symmetry control varies the shape of the waveform below the horizontal centre line. You can vary the waveform shape so that it is symmetrical above and below the centre line or so that the lower half of the waveform becomes more rounded. This rounding produces a sound characteristic of a bowed instrument such as the cello. Finally, the Skew control varies the waveform from being symmetrical July 2006  25 ➊ ➋ ➌ ➍ ➎ Above: these traces show the variety of waveforms that can be reproduced by the Theremin. The top four traces are all at 100Hz and show what can be done with just the waveform and symmetry controls, with the skew control set to minimum. Waveform 1 shows the output when the Theremin is set to produce a sinewave. The next waveform (2) is more distorted, with more triangle characteristics, while waveform 3 has a squarer wave shape. Waveform 4 shows what the skew control does to the signal at around 100Hz. As can be seen, it becomes very asymmetrical about the horizontal and vertical axis, exhibiting a more sawtooth wave shape. The final waveform (5) was obtained using the same settings that gave the waveform 4 but at a higher frequency of 250Hz, making it more sinusoidal in shape. This characteristic occurs for all waveform shapes at the higher frequencies. A filter adjustment sets the threshold point where the tone becomes more sinusoidal. to asymmetrical (ie, more sawtooth in shape) about the vertical axis. An asymmetrical wave shape produces a brighter, richer sound. 26  Silicon Chip Waveforms 1-5 show just some of the variety of waveforms that can be produced by the Theremin Mk.2. Note that all three tonal controls interact with one another, so that a whole array of subtle sound variations can be reproduced. These variations in the sounds are due to the harmonic content of the waveform. A pure sinewave comprises only the fundamental frequency and that is the only tone that you hear. If, for example, you play note A4, then you will hear a pure tone at 440Hz. Waveforms that are not pure sine­ waves include extra signals called harmonics. Harmonics are additional tones that are multiples of the fundamental frequency. So, for example, if you play note A4 again but produce a square wave, you will hear the fundamental 440Hz plus multiples of that frequency. Square wave harmonics are always odd and so you will hear the third harmonic (3 x 440Hz or 1320Hz), the fifth harmonic at 2200Hz and the seventh and ninth harmonics, etc. Note that the harmonic signal level is lower than that of the fundamental and diminishes with increasing frequency. In fact, the third harmonic is one third the level of the fundamental, while the fifth harmonic is one fifth the level, etc. Triangle waves also have only odd harmonics but the harmonic levels drop off much faster than the square wave harmonics. The third harmonic, for example, is only 1/9th the level of the fundamental and the fifth harmonic is 1/25th of the fundamental’s level. When the wave shape is skewed about the vertical axis to produce a sawtooth shape, or if the symmetry is altered about the horizontal axis, the harmonic content will include even and perhaps odd harmonics, depending on the waveform. Even harmonics are those that are twice the fundamental frequency, four times the fundamental, etc. These even harmonics give a stringed instrument sound effect and can enrich the sound produced by a square or triangle wave. Gain control The next control in the lineup is the Gain control. This is included to adjust the audio output level on the Theremin’s line output socket. It basically allows the output level to be correctly adjusted in response to different wave shapes. In practice, the line level output signal is fed out via a 6.35mm mono siliconchip.com.au Fig.1: the Theremin is based on three virtually identical oscillators plus a balanced mixer (IC1). The mixer accepts the signals from the pitch and reference oscillators and generates difference signals to produce the tones. These are then fed to the output stages via a voltage-controlled attenuator stage. jack socket. It can be switched on or off using the Muting switch. Power for the circuit comes from a DC plugpack. Monitoring Block diagram Normally, you would use the line output from the Theremin to feed an external amplifier and loudspeakers. However, the unit also features an internal amplifier and loudspeaker, which can be used for practice sessions (or as a foldback monitor during live performances). A headphone socket is also provided and this automatically disconnects the internal loudspeaker when the headphones are plugged in. OK, let’s now take a look at how the unit works. We’ll start with the block diagram which is shown in Fig.1. A balanced mixer (IC1) is at the heart of the operation. This accepts two signals: one from a reference oscillator (based on coil T1 and Mosfet Q1) and the second from a pitch oscillator (based on coil T2 and Mosfet Q3). The latter’s frequency is controlled using the pitch antenna, which is connected to the oscillator via an equalising coil. Typically, both the reference oscillator and the pitch oscillator are set to the same frequency, at about 455kHz. Any movement of the hand near to the pitch antenna will then alter its capacitance to ground and change the pitch oscillator’s frequency. In operation, the mixer produces several signals, depending on the incoming signals from the reference and pitch oscillators. These signals are: (1) the original reference oscillator signal (455kHz); (2) the sum of the reference and pitch oscillator frequencies; and (3) the difference between the pitch and reference oscillator frequencies. When the two oscillators are at the Presentation As shown in the accompanying photos, the SILICON CHIP Theremin Mk.2 is housed in a plastic case which in turn is mounted on a small camera tripod which serves as a desk stand. The pitch antenna sits vertically in the rear righthand corner of the box, while the volume antenna lies nearly horizontally on the lefthand side of the box. The various controls are arranged along the front face of the box, while the line output, headphone and DC supply sockets are located on the righthand side of the box, along with the power switch and a muting switch. siliconchip.com.au same frequency (eg, 455kHz), then the sum of the two frequencies will be 910kHz while the difference frequency will be close to zero. The mixer’s output is then fed to a low-pass filter which has a cutoff frequency of 3kHz. As a result, the 455kHz and 910kHz signals are filtered out, leaving only the difference signal. In this case, however, there will be no output since the difference signal is zero. However, when the pitch oscillator’s output frequency is lowered by moving the hand closer to the pitch antenna, the difference between the reference and pitch oscillators increases and we get an audible output. The lower the pitch oscillator’s frequency, the greater the difference frequency from the mixer and the higher the tone fed to the amplifier stages. For example, if the pitch oscillator is reduced to 454kHz, the difference frequency will be 1kHz and so we get a 1kHz audio output from the low-pass filter. If it goes down to 453kHz, we get a 2kHz audio output signal. In practice, the difference signal from the mixer ranges in frequency from 65.4Hz to 2093Hz, which is equivalent to five octaves. Equalising coil The equalising coil in series with July 2006  27 The Origin Of The Theremin I N 1919, A RUSSIAN PHYSICIST named Lev Termen (or Leon Theremin as he is called in the west) invented an electronic musical instrument called the “Theremin”. At that time, the Theremin was innovative and unique in the musical world and was essentially the first electronic instrument of its kind. Playing it was also unique, the technique relying solely on hand movements in the vicinity of two antennas to control two electronic oscillators – one antenna to vary the pitch of the sound and the other to change the volume. In operation, the pitch change afforded by the antenna is infinitely adjustable over several octaves, with the frequency increasing as the hand is brought closer to the antenna. An ear for pitch and fine hand control are essential requirements to become proficient at playing the Theremin. To a large extent, the Theremin has been made famous by recitalist Clara Rockmore. Born in Lithuania in 1911, she was an accomplished violinist by 5-years old. She began to learn to play the Theremin after meeting Leon Theremin in 1927 and ultimately developed a unique technique for playing the instrument. This technique involved minute finger movements to capture and modulate the tone of the note and enabled her to play the instrument with great precision. The Theremin was subsequently further developed and manufactured by the Radio Corporation of America (RCA) around 1929. This design consisted of a large box with an attached antenna and wire loop. The antenna provided the control for the pitch while the loop enabled the volume to be adjusted. In practice, the pitch control antenna was mounted vertically while the volume loop sat horizontally, to minimise interaction between them. And of course, the circuit used valves. General Electric (GE) and Westinghouse also made Theremins in the 1920s. However, the number of units produced was quite modest, with only about 500 units being made. Today, the Theremin is hailed as the forerunner to modern synthesised music and was instrumental in the development of the famous Moog synthesisers. Because of its unique sound, it has been popular with music producers for both film and live performances. For example, is was used to produce background music in “The Ten Commandments” feature film by Cecil B DeMille. Its eerie sounds have also made it ideal for science fiction movies, including “The Day the Earth Stood Still” and “ It Came From Outer Space”, and in thriller movies such as “Spellbound” and “Lost Weekend”. In addition, Bands such as the Bonzo Dog Band and Led Zeppelin have embraced the Theremin. The Beach Boys used an instrument similar to the Theremin – called an Electro-Theremin (also named a Tannerin) – in their famous “Good Vibrations” hit from the 1960s. The Electro-Theremin differs from the Theremin in that it incorporates a mechanical controller to adjust the pitch rather than hand movements relative to an antenna. Many commercial Theremins are available on the market today, including the Etherwave series from Moog Music Inc, PaiA’s Theremax and Wavefront’s Classic and Travel-Case Theremins. SILICON CHIP has also published two previous designs for home construction – ie, a basic Theremin in August 2000 and a MIDI Theremin in April and May 2005. the pitch antenna vastly improves the linearity of any frequency changes with hand movement. Without it, these frequency changes would be very non-linear – very large hand movements would be required to produce pitch changes at the low-frequency end, while only minute hand movements would be required at the highfrequency end. This “compression” of the frequency range for hand positions close to the antenna is due to the way a tuned circuit works. The variations 28  Silicon Chip in capacitance with hand movement are linear with the distance from the antenna. However, the frequency of a tuned circuit is inversely proportional to the square root of the capacitance. As a result, greater pitch variations occur for a given hand movement the closer we get to the antenna. Adding the equalising coil has the effect of reducing the hand movements required for the lower octaves and increasing the hand movements required for the upper octaves, so that the overall response is much more linear. When adjusted correctly, the resulting improvement in linearity is almost magical! Basically, the equalising coil works by setting up a resonant circuit. The resonant frequency is set to be just below the “at rest” frequency of the pitch oscillator and is based on the coil’s the inductance (about 10mH) and the capacitance of the antenna (about 12pF). The corresponding components in the pitch oscillator have an inductance of 560mH and a capacitance of 220pF (both inside coil T2). Any hand movement near the antenna will increase its capacitance and thus cause a reduction in the resonant frequency. However, this frequency shift will be much greater than the corresponding frequency change of the pitch oscillator. That’s because the effect of hand capacitance (just a few picofarads) is far greater on the 12pF antenna capacitance than it is on the much larger 220pF capacitor in parallel with the pitch oscillator coil. The overall effect is that your hand has a progressively smaller effect on the pitch oscillator as it is brought closer to the antenna. This introduced non-linearity counteracts the inherent non-linearity of the pitch oscillator and makes pitch changes much more linear for given hand movements. Waveform shaping The wave shape of the output is controlled using the Waveform, Symmetry and Skew potentiometers (VR3-VR5). Both the Waveform and Symmetry controls work by changing the DC bias levels on both the signal and carrier inputs of the mixer. A different bias voltage affects the wave shape that’s applied to a particular input of the mixer and this changes the resulting output waveform. Note that buffer stages (Q2 & Q4) are included in series with the outputs of the reference and pitch oscillators before the signal is applied to the mixer. These isolate the oscillators from the DC bias voltages at the mixer inputs, to prevent unwanted changes to the oscillator frequencies. The Skew adjustment varies the coupling between the pitch and reference oscillators, in turn varying their tendency to lock to the same frequency. When both oscillators are running close to the same frequency, increasing the skew control will cause the two oscillators to lock and so their siliconchip.com.au Fig.2: this graph shows the response of the bandpass filter. It reduces the signal level as the frequency of the volume oscillator decreases. output frequencies will be the same. However, if extra hand capacitance forces the pitch oscillator to change, it will suddenly “snap” to a different frequency. At the same time, the reference oscillator will continue to have an affect and so the resulting output waveform from the mixer will be skewed. Low-pass filter As mentioned, the adjustable lowpass filter following the mixer output removes the higher frequencies from the mixing process, leaving only the difference frequency. Its frequency of roll-off can be varied from 3.3kHz down to 592Hz, depending on the effect required. Following this filter, the signal is fed to an attenuator and then to amplifier stage IC3. This stage has a gain of between two and seven, depending on the setting of gain control VR6. The output from IC3 then goes to the Line Out socket via Muting switch S2. It also goes to power amplifier IC4 via volume control VR7. The power amplifier then drives an internal loudspeaker or a pair of headphones. Volume control oscillator The volume oscillator is based on transistor Q5 and transformer coil T3. As with the pitch oscillator, its frequency varies in response to hand movement. In operation, its frequency reduces as the hand is brought closer to the volume antenna. The resulting signal is then fed to a bandpass filter that rolls off signals above and below its centre frequency. Fig.2 shows the response siliconchip.com.au Fig.3: this is the basic arrangement for the equalising coil tester. It allows the pitch oscillator to be correctly adjusted so that the equalising coil and pitch antenna resonate, as indicated by a voltage dip on the output meter. shape of this bandpass filter. The bandpass filter is set so that its centre frequency is above the frequency range of the volume oscillator – ie, the frequency of the volume oscillator is to the left (or lower frequency side) of the peak in the filter response curve. Therefore, as the oscillator’s frequency decreases, the filter reduces the signal level. The output from the bandpass filter is fed to a slope detector based on diode D3. This converts the signal to a DC voltage which is then applied to a level-shifting amplifier based on IC5. Its output in turn controls the audio attenuator stage. Equalising coil tester Finally, we come to the equalising coil tester which is attached to the pitch oscillator. This tester allows the equaliser coil to be checked with the pitch oscillator, to verify that its value is correct. Fig.3 shows the basic arrangement. In operation, the pitch oscillator’s output is lightly coupled to the pitch antenna via the equalising coil. The oscillator frequency is then is adjusted until the following level detector circuit detects the resonance, as indicated by a voltage dip on the meter. Circuit details OK, that covers the basics. Now let’s take a look at the complete circuit diagram – see Fig.4. The first thing to note is that all three oscillators (Reference, Pitch and Volume) are virtually identical. Each oscillator is based on a junction FET (Q1, Q3 & Q5) and a standard IF (intermediate frequency) transformer coil (T1-T3), as used in low-cost AM radio tuners. Each transformer includes a tapped primary winding and a parallelconnected capacitor to form a tuned circuit. Its corresponding JFET drives a portion of the primary winding (ie, between the pin 2 tap connection and ground), while the signal at the top of the primary is coupled to the gate (which is self-biased) via a 68pF capacitor. This arrangement provides positive feedback to maintain oscillation at the tuned frequency. In the case of the reference oscillator, transformer T1 is tuned to produce an output frequency of about 455kHz. Power for the circuit comes from a +8V rail and this is applied to Q1’s drain via potentiometer VR2 and a 220W resistor. VR2 provides pitch adjustment by varying the drain to source current flow through Q1. This alters the gateto-source voltage and thus Q1’s gateto-source capacitance. And this in turn alters the tuned frequency. T1’s secondary winding at pins 4 & 6 provides a low impedance output from the oscillator. This output is then further buffered using an amplifier stage based on JFET Q2 which is configured as a source follower. This buffering is essential to isolate the oscillator from the following stages, so that it is immune to any capacitance changes caused by varying the bias levels at the inputs to IC1. The pitch oscillator is almost identical, the main difference being the use of a fixed 680W resistor in Q3’s drain circuit. In addition, the pitch antenna July 2006  29 Fig.4: the complete circuit for the Theremin Mk.2. Each oscillator is based on a junction FET (Q1, Q3 & Q5) and a standard IF transformer coil (T1-T3). IC1 is the balanced mixer – it produces the difference signal and feeds this to the audio output stages via an attenuator (OPTO1), in turn controlled by the volume oscillator and its following stages. 30  Silicon Chip siliconchip.com.au is connected to pin 1 of transformer T2 via a 1nF capacitor and the equalising coil (L1). The equalising coil test circuit is also attached to this part of the oscillator circuit during testing. In this case, the oscillator signal at pin 1 of transformer T2 (marked “Test”) is coupled into the pitch antenna via a 100kW resistor. This resistor ensures only minimal loading of the equalising coil and antenna tuned circuit. Diode D2 and the 10nF capacitor form a peak detector and this allows us to measure the relative level of the signal across the equalising coil and pitch antenna. The associated 100kW resistor across the 10nF capacitor helps to discharge the capacitor, so that the voltage on D2’s cathode drops with decreasing signal level. In practice, the ferrite slug inside T2 is adjusted so that pitch antenna and equalising coil resonate. We’ll describe how this is done in Pt.2. siliconchip.com.au The volume oscillator is similar to the pitch oscillator but also includes a variable drain supply. This is provided by potentiometer VR1, which is the volume range (or volume span) adjustment. The volume antenna is connected to pin 1 of T3 via a 1nF capacitor. Mixer The reference and pitch oscillator signals from buffer stages Q2 and Q4 are applied via 1nF capacitors to pins 1 (signal) and 10 (carrier) of IC1 respectively. However, the signal applied to the carrier input is reduced to around 50mV using a resistive divider at the source of Q4. This reduction in signal level is necessary to prevent overloading the mixer stages of IC1. IC1’s signal inputs at pins 1 & 4 and its carrier inputs at pins 8 & 10 are biased using potentiometers VR3 and VR4 respectively. Note, however, that any signal applied to pins 4 and 8 is shunted to ground via 100nF capacitor. In other words, these pins are simply DC biased. The DC bias range provided by VR3 & VR4 is set by the outputs of buffer stages IC2a-IC2d. These op amps are all wired as voltage followers and each buffers a sampled voltage from the +9V rail, as set by trimpots VR8-VR11. In effect, each buffer pair sets the maximum and minimum bias voltages and applies these to its corresponding potentiometer (VR3 & VR4). This ensure that VR3 and VR4 only provide the range of control that is necessary to produce the varied waveforms. A 1kW resistor between pins 2 and 3 of IC1 sets the gain of the mixer, while the bias voltage at pin 5 sets output signal level. The balanced mixer outputs appear at pins 6 and 12. Each output is biased on using 2.2kW pull-up resistors and is filtered to remove the high-frequency components. The output at pin 6 has a fixed July 2006  31 Par t s Lis t 1 PC board coded 01207061, 188 x 103mm 1 plastic UB2 utility case, 197 x 113 x 63mm 1 12V DC 450mA plugpack 1 190 x 105mm aluminium sheet (1mm thick) 1 100mm 4W 2W loudspeaker 2 high-quality stereo switched 6.35mm jack sockets, PCmount (Jaycar PS-0195) 1 2.5mm PC-mount DC socket 2 SPST ultra-mini rocker switches (S1,S2) 1 mini tripod (Jaycar AM-4112 or similar) 1 mini heatsink, 19 x 19 x 10mm 1 M4 x 25mm Nylon screw 1 M4 nut 9 M3 x 6mm screws 9 M3 nuts 3 4mm eyelet crimp connectors 4 4.8mm female spade connectors 7 plastic knobs to suit (do not use metal knobs) 20 PC stakes 1 400mm length of 0.7mm tinned copper wire 1 12m length of 0.25mm enamelled copper wire 1 250mm length of medium duty hook-up wire 1 green banana socket 1 11mm OD x 4mm ID x 2.5mm Nylon spacer or similar (eg, 3 x M4 Nylon washers) 1 300mm length of green hook-up wire Transformers and ferrites 2 pot cores, 26 x 11.5 x 8mm (Al of 4740) (Jaycar Cat. LF-1060 low-pass filter consisting of a 22nF capacitor to ground. By contrast, the output at pin 12 is connected to an adjustable low-pass filter consisting of VR13 and a 22nF capacitor. As stated previously, its roll-off frequency can be continuously adjusted from 3.3kHz (VR13 set to 0W) down to 592Hz (VR2 at 10kW). Volume control Following the low-pass filter, the signal is AC-coupled to a 100kW resistor in series with the pin 3 input of amplifier stage IC3. This input is biased 32  Silicon Chip or equivalent) (L1) 1 bobbin to suit above cores (Jaycar Cat. LF-1062 or equivalent) 3 low-cost 455kHz 2nd IF transformers (white slug) 1 low-cost 455kHz 3rd IF transformer (black slug) Potentiometers & trimpots 2 16mm 1kW linear PC-mount potentiometers (VR1,VR2) 3 16mm 5kW linear PC-mount potentiometers (VR3,VR4,VR6) 1 16mm 10kW linear PC-mount potentiometer (VR5) 1 16mm 10kW log PC-mount potentiometer (VR7) 4 10kW multi-turn top-adjust trimpots (code 103)(VR8-VR11) 1 5kW multi-turn top-adjust trimpot (code 503) (VR12) 1 10kW horizontal trimpot (code 103) (VR13) 1 2kW multi-turn top-adjust trimpot (code 203) (VR14) Antenna Parts 1 375mm length of 16mm dia­meter plated steel or stainless steel tubing 1 125mm length of 16mm dia­meter plated steel or stainless steel tubing 2 chromed towel rail end brackets to suit above tubing 2 16mm ID plastic end caps 1 miniature tripod with ¼-inch mount (Jaycar AM-4112 or AM4110) 2 M4 x 10mm screws 2 M4 x 15mm screws 4 M4 nuts 1 ¼-inch Tee nut to 4.5V via the 10kW voltage divider resistors across the 9V supply and a second 100kW resistor. This allows the op amp to produce a symmetrical output voltage swing before clipping. The 4.5V bias supply is decoupled using a 100mF electrolytic capacitor to remove any signal ripple. The signal level applied to pin 3 of IC3 is controlled by OPTO1 which is an opto-coupled LDR (light dependent resistor). This in turn is controlled by the volume oscillator and its following circuitry. As shown, the LDR is connected between pin 3 of IC3 and Semiconductors 1 MC1496 balanced modulator (IC1) 1 LM324 quad op amp (IC2) 1 TL071 JFET input op amp (IC3) 1 LM386 1W power amplifier (IC4) 1 7809 9V 1A regulator (REG1) 1 7808 8V 1A regulator (REG2) 5 2N5484 or 2N5485 JFETs (Q1-Q5) 1 BC337 NPN transistor (Q6) 1 BC327 PNP transistor (Q7) 3 1N4148 diodes (D1-D3) 1 1N4004 1A diode (D4) 1 NSL-32SR3 optocoupler (Silonex) (OPTO1) Farnell Cat. 369-2218 1 5mm red LED (LED1) Capacitors 1 1000mF 16V PC electrolytic 1 470mF 16V PC electrolytic 3 100mF 16V PC electrolytic 7 10mF 16V PC electrolytic 1 2.2mF 16V PC electrolytic 1 220nF MKT polyester 8 100nF MKT polyester 1 47nF MKT polyester 2 22nF MKT polyester 6 10nF MKT polyester 4 1nF MKT polyester 1 470pF ceramic 3 330pF ceramic 3 68pF ceramic Resistors (0.25W 1%) 1 330kW 2 680W 10 100kW 1 330W 1 47kW 2 220W 2 22kW 1 150W 4 10kW 3 100W 1 4.7kW 1 39W 4 2.2kW 3 10W 6 1kW the 4.5V rail. Normally, the LDR has negligible effect on the signal since its resistance is considerably higher than the 100kW resistor at pin 3 of IC1. However, when current flows through the LED in OPTO1, the resistance of the LDR falls. This shunts signal from pin 3 to the 4.5V rail. In operation, the LDR has a resistance range from about 25MW down to 60W, giving an attenuation range from 0db to -64dB. As discussed previously, hand move­ments over the volume antenna siliconchip.com.au Building the Theremin Mk.2 is straightforward, with all but a few parts mounted on a single PC board. The full constructional details will be in Pt.2 next month. control the amount of attenuation. It works as follows. First, the signal from the volume oscillator (T3 & Q5) is fed to the bandpass filter which is based on transistor Q6 and tuning coil T4. T4 is tuned so that the output signal at its pin 6 decreases in level as the volume oscillator frequency decreases (ie, as the hand moves closer to the volume antenna). Diode D3 rectifies this signal and its output is filtered using a 2.2mF capacitor to provide a DC voltage. The associated 10kW resistor across the capacitor provides a discharge path, thus allowing the voltage across the capacitor to fall if the signal level falls. This voltage is fed to IC5b which is wired as a non-inverting amplifier and level shifter. Trimpot VR14 adjusts the output offset of the amplifier, so that it can be set to vary from about 8V down to nearly 0V with hand movement. IC5b’s output appears at pin 7 and drives PNP transistor Q7 which is wired as an emitter follower. This in turn drives the LED within OPTO1 via a 330W resistor. The anode side of the LED is connected to the 8V supply. Note that there are two supply rails – siliconchip.com.au ie, 8V and 9V. Op amp IC5 is powered from the 9V supply while the LED is powered from the 8V supply. The 1V extra for the op amp is to ensure that IC5b’s output can swing high enough to switch Q7 and the LED fully off. Power amplifier IC4 is also powered from the 9V rail while the more “voltage sensitive” sections of the circuit are powered from the 8V rail. This separation of supply rails ensures that IC4 can drive the loudspeaker at full power without affecting other parts of the circuit. Op amp IC3 buffers the attenuated signal and also provides gain that can be varied from 2-7, depending on the setting of Gain control VR6. The 10nF capacitor between pins 2 & 6 rolls off high frequencies to prevent instability. IC3’s output appears at pin 6 and is AC-coupled to mute switch S2 in series with the line output socket (CON1). The series 150W resistor serves to isolate IC3’s output from the load connected to the line out. IC3’s output is also fed to volume control VR7 via a 10mF capacitor. From there, the signal goes to pin 3 of power amplifier IC4 via a 220nF ca- pacitor. IC4 then drives the Theremin’s loudspeaker or a pair of headphones via a 1000mF capacitor and connector CON2. Plugging the headphones into CON2 automatically disconnects the loudspeaker. Note that a Zobel network comprising a 10W resistor and 47nF capacitor is connected across IC4’s output. This is done to prevent oscillation in the amplifier. Power supply Power for the circuit is derived from a 12V DC plugpack. This is fed in via power switch S1 and diode D4 which provides reverse polarity protection. LED 1 provides power on/off indication, while a 470mF electrolytic capacitor filters the supply rail before it is applied to 3-terminal regulators REG1 and REG2. REG1 and REG2 respectively provide the regulated +9V and +8V supply rails. Their outputs are decoupled using 10mF capacitors. That’s all we have space for this month. Next month, we will give the full construction details and describe the setting-up and adjustment proceSC dures. July 2006  33