Chorus

Module ID fx.chorus · Category Effect

Chorus Module Toms on every beat through three voices: the braid weaves slowly and lights where each hit passed

The Chorus thickens a sound by layering copies of it, each delayed by a few milliseconds that keep changing. The moving delays bend each copy's pitch slightly up and down, and the copies beat against each other and against the original, the way several players on the same part never quite line up.

It is true stereo, and it makes a mono source wide on its own. Short delays and a little feedback take it into flanging.

How it works

  1. Each channel has its own delay line: left stays left, right stays right.
  2. Each voice reads both lines through a pair of taps, and an LFO sweeps each tap's delay around Delay. At full Depth the delay swings all the way from zero to twice the Delay time.
  3. The voices are spaced evenly around the LFO cycle, so they never move in step.
  4. The voices are summed and blended with the dry signal by Mix.

True stereo

A stereo source keeps its image: nothing patched into In L ever reaches Out R.

With In R unpatched, In L feeds both sides and the chorus creates the width itself. Each right tap's LFO sits halfway between two left taps (180° away with one voice, 90° with two), so the two sides always bend in pitch in opposite directions.

Changing Voices fades voices in or out and slides the rest to their new places on the cycle, so it never clicks.

The display

The display draws the voices sweeping over the last two seconds, newest at the right edge. Up and down is how far each tap reads behind Delay: the centre line is the Delay itself, and the faint lines either side are as far as Depth lets the taps swing.

  • Strands. Each voice is a strand in its own colour. The solid strand is its left tap, and the fainter one beside it is its right tap, half a voice-spacing behind. One voice draws a single wave, and four weave a braid. Rate sets how many cycles cross the panel, and Shape bends them from sine to triangle. A dot at the right edge marks where each voice is now.
  • Light. The strands light up where sound came in, so the braid glows with whatever passes through it.
  • Feedback. Positive FB thickens the strands. Negative FB, the hollow flanger sound, turns them hollow and violet.
  • Words. The top left gives the rate and delay, with the rate in orange while Rate CV moves it. The top right gives the number of voices and the feedback.

Inputs

PortSignal TypeDescription
In LAudio (Blue)Left input
In RAudio (Blue)Right input. When unpatched, it copies In L
Rate CVControl (Orange)Speeds up or slows down the sweep: ±1 changes Rate by ±50%
Depth CVControl (Orange)Adds to Depth: ±1 adds ±50%

Outputs

PortSignal TypeDescription
Out LAudio (Blue)Left chorus, mixed with the dry signal
Out RAudio (Blue)Right chorus, mixed with the dry signal

Parameters

ControlRangeDefaultDescription
Rate0.1 Hz – 10 Hz1 HzSpeed of the sweep
Depth0 – 100%50%How far the delay sweeps around Delay
Delay1 ms – 30 ms10 msCenter delay time
FB (Feedback)−0.5 – +0.50Feeds the voices back into the delay lines, for flanging. Negative values sound hollower
Mix0 – 100%50%Dry (0%) to chorus only (100%)
Voices1 – 42Number of voices (dropdown on the node)
ShapeSine / TriSineLFO waveform (dropdown on the node)

While Rate CV or Depth CV is patched, its knob stays live and sets the center the CV works around, so you can keep turning it as the sweep plays.

Shaping the sound

Rate and Depth. Slow rates sway gently; fast rates turn into vibrato. Light depth thickens without drawing attention; heavy depth wobbles until it sounds out of tune. The two work together, since the pitch bend depends on how fast the delay is changing: a fast rate needs less depth.

Delay. Below about 5 ms the copies comb-filter against the dry sound, and with some feedback it becomes a flanger. From 5 to 15 ms is classic chorus. From 15 to 30 ms the copies separate into a wide, doubled sound.

Voices. One voice is the simplest, almost flanger-like, with the two sides sweeping in opposite directions. Two is the classic stereo chorus. Three and four build toward a thick ensemble, at some cost in clarity. The output stays at about the same loudness whatever the voice count.

Shape.

ShapeCharacter
SineSmooth and vocal: the pitch eases in and out of each bend
TriThe delay sweeps at a constant speed, so each half-cycle holds one steady detune. Glassier, closer to the classic bucket-brigade choruses

Switching shape morphs between the two over 20 ms.

Feedback. Feedback deepens the comb-filter peaks and makes the sweep ring. The loop uses the voices' average, so its gain never exceeds the knob however many voices are running.

Bypass

Click the power switch in the node header, press Ctrl+B with the module selected, or choose Bypass from its right-click menu. In L passes straight to Out L and In R to Out R (In R still copies In L when unpatched). The switch crossfades over 20 ms.

Starting points

SoundRateDepthDelayFBVoicesMix
Subtle thickening0.3 Hz20%5 ms0230%
Classic chorus0.5 Hz40%8 ms0250%
String ensemble0.8 Hz60%15 ms0460%
Flanger0.2 Hz70%2 ms+0.4150%
Vibrato5 Hz30%5 ms01100%

At 100% Mix you hear only the moving copies, without the dry signal to beat against, so the chorus becomes pure vibrato.

On bass, keep Depth and Mix low, or the low end turns vague.

Patch ideas

Widen a mono voice. Patch only In L and take both outputs:

[VCA Out] ──> [Chorus In L]
[Chorus Out L] ──> [Audio Output Left]
[Chorus Out R] ──> [Audio Output Right]

Before the space. Chorus usually goes before time-based effects, so the reverb smears the movement together:

[VCA Out] ──> [Chorus In L]
[Chorus Out L] ──> [Reverb In L]
[Chorus Out R] ──> [Reverb In R]

Drifting chorus. A very slow LFO into Rate CV makes the chorus speed itself wander.

  • Delay: longer echoes
  • Reverb: chorus into reverb for ambient washes
  • Oscillator: unison detune thickens at the source
  • LFO: modulates Rate and Depth