Making sounds with analogue electronics - Part 7: Topology of a typical synthesizer

Polyphonic analogue synthesizers are often implemented as several monophonic synthesis 'engines' or 'voices' connected to a common polyphonic keyboard. Each of these 'voices' receives monophonic note-pitch voltage, gate and trigger information, and performance controller information. It is usual for each voice to produce the same sound or timbre: multi-timbrality is normally commonly found in digital instruments only.

The assignment of the voices to the keys which are played on the keyboard is carried out by key assignment circuitry or software in the polyphonic keyboard. This deals with the reassignment of notes which are playing (note stealing) and the method of assigning notes to the voices (last-note priority, etc.). (More details of keyboards can be found in Chapter 7.)

Controlling portamento on a polyphonic synthesizer is much more complex than on a monophonic synthesizer. The transitions between several notes can be made using several portamento algorithms. These are often named according to the effective polyphony that they produce, although in practice, only short portamento times are used to give a slight movement of pitch at the beginning of notes: this is frequently used for vocal, brass and string sounds.

Longer portamento times do not suit polyphonic keyboard technique, except for special effect usage with block chords, and often a glissando is more musically useful – where all the notes in between the last-note played and the next are played in sequence.

Memory stores seem to be more widespread in early polyphonic analogue synthesizers than in their monophonic equivalents. Initially, some manufacturers produced low-cost polyphonic synthesizers without memories, but these were not very popular in comparison with their more expensive memory-equipped versions.

The designers of polyphonic synthesizers seem to have placed more emphasis on the accessibility of the memory recall controls than on the front panel controls for programming the synthesizer voices. The Yamaha CS-80 demonstrates this principle in its design: the programmable memories are hidden underneath a flap, and have tiny controls, whilst the large, colorful memory recall buttons are handily placed right at the front of the control panel.

The performance controls on polyphonic analogue synthesizer tend to be optimized for polyphonic playing techniques. Pitch-bend is normally only a semitone, and can often be applied to only the top note or the last note which has been played on the keyboard. The modulation wheel is often replaced or paralleled by a foot pedal and often controls timbre through the VCF cut-off rather than vibrato. After-touch is almost invariably used to control vibrato or tremolo, and some instruments provide polyphonic after-touch pressure sensing instead of the easier-to-implement global version.

Some instruments have an LFO which is common to all the voices, and so vibrato or tremolo modulation is applied at exactly the same frequency and phase to all the voices. In contrast, instruments which use separate LFOs for each voice circuit will have slightly different frequencies and phases, and this can greatly improve string and vocal sounds.

Real-time changes to the timbres are normally made using additional controllers: foot pedals, foot switches and breath controllers.

Manipulating front panel controls whilst playing with both hands on a polyphonic keyboard seems to be unpopular, and if front panel controls are used then the playing technique used often reverts to monophonic usage, as described earlier – although polyphonic keyboards almost always have retriggered envelopes, which restrict some performance techniques. The performance controls are placed on the left-hand side of the keyboard, just as with monophonic synthesizers (Figure 3.6.4).

FIGURE 3.6.4 A summary of the main features of a typical analogue polyphonic synthesizer of the 1980s.