Sunday, 8 November 2009
The audio signal moves from the theremin to the amplifier from right to left, starting with a delay pedal in ping-pong configuration. The delay is, for the sake of argument, 1 second. A sound appears at both outputs (A and B) instantaneously, and is then repeated at A one second later, then again at B one second after that.
Both the signals from the delay are fed into the ring-mod where one amplitude modulates the other, and the resulting signal (C) exits the box at the centre of the right side. The knob at the centre on the left side controls its volume.
Signal C then passes through a low pass filter. The knob at the top in the centre controls the operating frequency of the filter, and the knob on the left side controls the filter's resonance.
Meanwhile copies of the signals from the delay pedal A and B exit the ring-modulator unaffected on the sides, and are mixed with the ring-modulated and low-passed signal C. The knobs at the sides of the top of the box control the volume of A and B.
The combined signals then pass through a reverb pedal.
The Ring Modulator
It is sufficient to consider the simple case of two sine waves being fed into the ring modulator.
A signal of frequency X amplitude modulated against a signal of frequency Y creates a new signal Z with two component sine waves which are the sum of X and Y and the difference of X and Y.
The pitch of the sum component Z(sum) is between the pitches of X and Y, but one octave higher. The exact relationship gives ring-modulation its characteristic and familiar sound.
Where X and Y are consonant with respect to each other, Z(sum) will be consonant with respect to both. Likewise, given a dissonant X and Y, Z will be dissonant with respect to both.
Where X and Y are in unison Z(sum) will be one octave higher. Z(diff) - the difference component - will be silent, zero.
When the difference between A and B is small, Z(diff) will be low in pitch. When it is larger Z(diff) will be higher in pitch. The same considerations about consonance and dissonance apply as with Z(sum).
In short, it augments consonance and dissonance alike.
Z(diff) becomes more interesting when it is fed via a ping-pong delay, giving a relationship between A and B over time.
It is sufficient to consider the simple case of a signal going first to A and then to B one second later.
A slow sweep upwards through the pitch field, varying the frequency at a constant rate (note - not the pitch - in practice this means slowing down logarithmically as pitch increases if the field is linear with respect to pitch to maintain a constant rate of change) will create a Z(diff) of constant low pitch, while Z(sum) rises with A and B.
In short, the pitch of Z(sum) is dependant on the player's pitch hand's distances relative to the pitch rod, and Z(diff) is dependant on its velocity relative to the pitch rod.
In this configuration the panic box can be thought of in part as a simple audio "difference engine."
The piece Gently Drowning is an example of this configuration. On YouTube here. Very slow changes in pitch by overlapping notes or slow glissandi create very low sounds.
By inserting a pitch shift between the delay pedal and the ring-mod on one of the channels increases the frequency of Z(diff) by a constant amount throughout, giving higher pitches.
The piece Iron Sun is an example of this configuration. On YouTube here.
By replacing the delay pedal with a dual output pitch-shifter, with one output giving the dry signal, and the other a signal a fixed interval higher (or lower) than the dry signal, a new effect is created. Note that the interval is constant with respect to pitch. This means it varies with respect to frequency. The same interval will create a Z(diff) of low frequency when the dry signal is low pitched, and a higher frequency when it is high pitched.
The piece The New Consonance is an example of this configuration. On YouTube here. In this case the interval created by the pitch-shifter is very small - 30 cents, so Z(diff) is subsonic, and is audible as beats - low notes having slow beats and higher notes, faster beats. To add to the fun there is a delay after the panic box, so sounds can overlap.
Note that all three pieces have significant low frequency content, so are not well reproduced by small computer speakers.