The chorus effect is a time-based modulation effect that makes one source sound thicker, wider, and more like several similar performances playing together. In its simplest electronic form, the input is split into a dry path and a short delay path whose delay time is continuously varied by a low-frequency modulator. The moving delayed signal is then combined with the dry signal.
The important detail is that chorus does not require “two or more shifting delays.” A basic chorus can be built from one modulated delay line plus the dry signal. Multi-voice and ensemble choruses may use several delay lines, but those are expanded architectures, not the minimum definition of the effect. The motion is also not the same as a static detune: changing delay time continuously changes the playback timing and creates a continuously moving pitch deviation.
Chorus Effect: Table of Contents
What is the chorus effect?
Natural chorusing happens whenever multiple performers attempt to play or sing the same part together. Their timing, tuning, articulation, and level never match perfectly. Those tiny differences create a larger, animated sound that is recognizably different from one perfectly duplicated performance.
An electronic chorus recreates part of that behavior with controlled time variation. A delayed copy of the source arrives slightly later than the dry signal, and the amount of delay is continuously moved. Because the delay time is changing rather than staying fixed, the processed path also moves slightly in pitch. When dry and wet are combined, the result contains continuously changing timing, pitch, and phase relationships.
Most chorus designs operate with delays measured in tens of milliseconds rather than the ultra-short delay times associated with obvious flanging. The exact ranges overlap between products, so delay time alone does not define the category. Chorus is better identified by the combination of short variable delay, modest pitch movement, dry/wet summing, and an intended doubling or ensemble-like result.
How chorus works
The core signal flow is simple:
- The input signal is split into a dry path and a processed path.
- The processed path enters a short variable delay line.
- An LFO or another slow modulation source continuously changes the delay time.
- The moving delayed signal is mixed with the dry signal, or distributed across stereo channels.
That moving delay is the defining mechanism. If the delay stayed fixed, the processor would behave more like static doubling or a short-delay widening effect. Once the delay time moves, the processed path alternately falls farther behind and catches up to the incoming signal. That changing read speed produces a pitch shift.
Physical Audio Signal Processing describes chorus as a process that makes one source resemble several sources performing in unison, commonly using time-varying delay lines. The same text also treats variable delay as a way to produce frequency shift through the Doppler mechanism.
The dry and delayed paths also interfere when they are summed. Their relative phase changes with frequency and with the instantaneous delay time, so the peaks and notches in the combined spectrum move continuously rather than remaining fixed. That movement is part of the liquid, shimmering quality associated with chorus.
Why changing delay time changes pitch
The pitch movement in a traditional chorus is not a separate pitch-shifter inserted after the delay. It is a direct consequence of moving the delay time itself.
If a continuous input is read from a time-varying delay line, it can be written as:
y(t) = x(t – D(t))
where D(t) is the delay time as it changes over time. For a sinusoidal component, the instantaneous output frequency follows:
fout = fin [1 – dD(t)/dt]
This relationship is the practical reason chorus pitch rises when the delay gets shorter and falls when the delay gets longer. The effect depends on the rate of change of the delay, not merely on the absolute delay value.
If the delay is being moved by a sinusoidal LFO, both modulation depth and modulation rate influence the maximum pitch excursion. A deeper sweep covers more delay time, while a faster LFO changes that delay more quickly. This is why increasing Rate can make a chorus sound not only faster but also more obviously pitch-modulated, especially when Depth is already high.
Digital implementations add another problem: the desired read position often lands between stored audio samples. High-quality digital chorus therefore uses fractional-delay interpolation to estimate values between sample positions as the delay moves. Without interpolation, the read position would jump from sample to sample and create stepping or zipper-like artifacts.
Chorus controls and what they actually change
| Control | What it changes | Typical audible result |
|---|---|---|
| Rate | Speed of the modulation source | Faster cyclic movement. With sufficient depth, it also increases the speed of pitch change. |
| Depth | How far the delay time moves around its center value | More obvious time and pitch modulation, wider or more seasick motion at extreme settings. |
| Delay / Center Delay | The baseline delay before modulation is applied | Shorter values move toward tighter comb-filter coloration; longer values move toward looser doubling. |
| Mix / Wet-Dry | Balance between original and modulated paths | Controls how strongly the moving copy interacts with the dry source. |
| Feedback / Regeneration | Returns some processed signal to the delay input | Adds resonance and stronger comb-filter character. Heavy feedback can move the sound toward flanging. |
| Stereo / Spread | Distribution or phase relationship of modulation across channels | Changes width and movement across the stereo image. |
Not every chorus exposes all of these parameters. Some classic designs intentionally hide most of them. A Dimension-style processor, for example, can offer fixed button combinations instead of conventional Rate and Depth knobs. The underlying signal path still contains delay and modulation even when the interface does not reveal those values directly.
The wet/dry control is especially revealing. In a basic chorus, removing the dry path leaves only the time-varying delayed signal. The moving pitch remains, but the dry/wet interference that creates the classic thickening disappears. The result behaves like a vibrato effect rather than conventional chorus.
What chorus sounds like
A well-set chorus can sound wider and richer without calling attention to itself. At more obvious settings it becomes watery, shimmering, swirly, glassy, animated, or deliberately unstable. Those descriptions are subjective, but they correspond to specific technical changes: moving arrival time, moving pitch, and moving interference between related signals.
Slow Rate with moderate Depth usually produces broad, gradual motion. Faster Rate creates a more obvious cyclical wobble. More Mix makes the effect easier to identify. Longer center delays can feel more like a second performance, while shorter delays emphasize spectral coloration. Feedback sharpens the resonance and can make the movement more metallic.
The source matters just as much as the settings. Chorus on a sustained synth pad exposes smooth motion because the source lasts long enough for the modulation cycle to unfold. On a short percussion hit, the same settings may be heard mostly as width or a change in attack. On bass, the movement can feel luxurious in stereo but unstable in mono if the low frequencies are being decorrelated aggressively.
Stereo chorus is not one fixed architecture
“Stereo chorus” describes a result, not one mandatory circuit. Designers have used several architectures to create width.
- Dry on one side, modulated signal on the other: the classic Roland JC-120 approach separates the direct and modulated paths across two speakers so part of the mixing occurs acoustically in the room.
- Two modulated delay paths: two wet voices can use LFOs with different phase offsets so one side is moving differently from the other.
- Cross-mixed architectures: Dimension-style designs use more complex routing to create width while reducing the obvious cyclical wobble associated with a conventional chorus.
- Multi-voice ensemble chorus: several moving delay voices are distributed across the stereo field to create a denser, less obviously repetitive texture.
This is why two products labeled “stereo chorus” can sound radically different even when their nominal Rate and Depth values appear similar. The number of delay voices, LFO relationships, polarity, cross-mixing, filtering, and wet/dry routing all affect the stereo image.
Chorus vs flanger, phaser, vibrato, doubling, and detune
| Effect | Core mechanism | Key distinction from chorus |
|---|---|---|
| Chorus | Short time-varying delay mixed with dry signal | Designed primarily for moving doubling, thickness, and width. |
| Flanger | Shorter time-varying delay, commonly with more feedback | Produces stronger moving comb filtering and the familiar metallic or jet-like sweep. |
| Phaser | All-pass filter stages create frequency-dependent phase shifts | Creates moving notches without relying on the same variable-delay pitch mechanism. |
| Vibrato | Pitch modulation without a dry reference path | In a chorus circuit, a fully wet modulated delay produces vibrato-like pitch movement instead of dry/wet thickening. |
| Doubling / ADT | A second performance or delayed copy with static or less cyclic variation | Aims to resemble another take rather than an obvious repeating LFO cycle. |
| Static detune / MicroPitch | Small fixed pitch offsets, often combined with short delay | The pitch offset can remain relatively stable instead of continually rising and falling with a chorus LFO. |
Chorus and flanging are especially easy to confuse because they belong to the same broad family of variable-delay effects. The boundary is not a perfect millisecond cutoff. Flangers generally use shorter delays and stronger feedback so the comb-filter notches become more obvious and resonant. Chorus generally uses longer short delays and less feedback so the ear focuses more on doubling and pitch movement.
Chorus and phasing can both create a sweeping sound, but the signal structures are different. A phaser builds its notches from all-pass filter phase shifts. Traditional chorus creates motion by varying delay time, which also varies pitch.
Analog BBD chorus vs digital chorus
Bucket-brigade chorus
Many classic analog choruses use a bucket-brigade device (BBD) as the variable delay line. A BBD contains a chain of analog storage stages that pass charge from one stage to the next under control of a clock. Changing the clock speed changes the time required for the signal to travel through the device, which changes the delay time.
The Panasonic MN3007, one of the best-known BBDs used in modulation effects, is a 1,024-stage device. Classic BBD circuits also need filtering and careful gain management because the clocked delay process has limited bandwidth and introduces noise and clock-related artifacts. Those surrounding filters, companding stages, gain choices, and nonlinearities contribute significantly to what listeners describe as the darker or softer character of vintage analog chorus.
It is therefore misleading to describe a BBD as a tiny piece of tape. Tape and BBD delays can both sound bandwidth-limited and imperfect, but they store and move the signal in fundamentally different ways.
Digital chorus
A digital chorus stores samples in memory and moves one or more read positions through that buffer. Because the desired delay is continuously changing, the processor needs interpolation whenever the read position falls between discrete samples. Different interpolation methods trade computational cost, magnitude accuracy, phase behavior, and high-frequency response.
Digital chorus is not inherently cleaner-sounding in a musical sense. It can be designed for transparent modulation or deliberately model BBD filtering, companding, saturation, clock leakage, and other analog behavior. Digital processing also makes complex multi-voice modulation, randomization, tempo-aware control, and elaborate stereo routing much easier to implement.
Using chorus in a mix
Guitar
Chorus can make a clean guitar feel much larger without requiring another recorded take. Slow modulation and a moderate mix can add width while preserving the pick attack. More Depth and Rate move the result toward the recognizable liquid guitar chorus associated with many late-1970s and 1980s productions.
Synths and pads
Synthesizers are ideal chorus sources because sustained oscillators make slow modulation easy to hear. A relatively simple sawtooth or pulse-wave patch can become much wider and more animated when a stereo chorus adds slightly different moving copies. This is one reason chorus became inseparable from the sound of classic string machines and later Roland polysynths.
Vocals
On vocals, subtle chorus can create width or a synthetic doubled texture, especially on backing parts. On a exposed lead vocal, obvious cyclic pitch motion can quickly become noticeable, so lower Mix and slower modulation are common starting points. When the goal is a more stable double, dedicated doubling or static micro-pitch processing may be a better fit.
Parallel chorus
Placing chorus on a send can separate the dry source from the processed layer. The return can then be filtered, automated, widened, or compressed independently. This is useful when the source needs the animation of chorus but must retain a stable center image and clean transient.
The same principle works inside a plugin that provides separate dry and wet controls. The practical goal is not to maximize width. It is to decide which part of the source should stay stable and which part can move.
Chorus on bass and mono compatibility
Chorus on bass can sound excellent. It is a major part of many fretless, post-punk, new wave, synth-bass, and atmospheric bass textures. The risk is not that bass “must always be mono.” The risk is that strong time and phase differences in the lowest frequencies can make the fundamental change dramatically when the signal is summed to mono.
A practical workflow is to keep the deepest low end more stable while allowing upper harmonics to carry most of the stereo modulation. That can be done in several ways:
- Use a parallel chorus return and high-pass the wet path so the sub region stays mostly dry.
- Use a multiband or frequency-selective chorus if the processor provides one.
- Use Mid/Side processing after the chorus to reduce very low-frequency Side energy when the mix benefits from it.
- Check the mono sum instead of relying only on a correlation meter or a visual stereo display.
There is no universal crossover such as “everything below 150 Hz must be mono.” The useful frequency depends on the part, arrangement, playback context, and what the chorus is doing. The goal is simply to keep important low-frequency weight from disappearing when stereo differences collapse.
A short history of the chorus effect
The electronic chorus effect grew from a long-standing goal: make one electronic or recorded voice feel like an ensemble. Organ and early electronic-instrument designers experimented with multiple voices, moving delay, and related modulation techniques before the compact chorus pedal existed.
String machines pushed the idea forward dramatically. The Solina String Ensemble, introduced in 1974, used multiple modulated delay paths to transform a relatively simple divide-down sound source into the lush ensemble texture that became one of the defining keyboard sounds of the decade. Sound On Sound’s technical history of the Eminent 310 and Solina describes the Solina’s three delay lines and phase-offset modulation as central to that sound.
Roland then made chorus a major standalone effect. The JC-120 Jazz Chorus amplifier arrived in 1975 with the company’s “Dimensional Space Chorus” concept. In 1976, BOSS released the CE-1 Chorus Ensemble, derived from the JC-120 circuitry. BOSS identifies the CE-1 as the world’s first chorus effect pedal and the first BOSS pedal. The smaller CE-2 followed in 1979.
BOSS documents this lineage directly, including the JC-120, CE-1, CE-2, the analog BBD architecture, and the distinction between chorus and vibrato modes. Also in 1979, Roland introduced the SDD-320 Dimension D, a studio stereo chorus known for creating width with less obvious cyclic modulation than a conventional chorus.
The importance of these designs is not only historical. Modern pedals, plugins, synthesizers, and DAW effects still organize chorus around the same engineering choices: delay time, modulation rate and depth, wet/dry routing, feedback, number of voices, filtering, and stereo topology.
Common chorus mistakes
“Chorus needs two or more modulated delays”
No. A single variable delay path mixed with the dry source is enough to produce a basic chorus effect. Multi-delay designs create richer or more complex versions, but they are not the minimum requirement.
“Chorus is just detuning”
Not in the normal engineering sense. Traditional chorus produces a moving pitch deviation because the delay time is moving. Static detune processors can create width and thickness without the same cyclic variable-delay behavior.
“Chorus and flanger are the same effect with different names”
They share the same broad variable-delay family, but typical parameter ranges and design priorities differ. Flanging emphasizes shorter delays, stronger comb filtering, and often more feedback. Chorus emphasizes moving doubling and pitch variation.
“Analog chorus sounds warm because BBDs behave like tape”
BBDs are not tape devices. Their character comes from a clocked analog delay architecture plus the filtering, noise-management circuitry, gain structure, distortion, and component behavior around the delay line.
“More stereo width is always better”
A spectacular soloed chorus can destabilize the center of a mix or lose important information in mono. Width is useful only when the musical source still performs its intended role in the complete mix.
Chorus-related terms
- Delay: the underlying time-storage process used by a conventional chorus.
- Phase: the frequency-dependent relationship that changes continuously when dry and delayed signals are combined.
- Flanger: a closely related variable-delay effect that generally uses shorter delays and stronger feedback.
- Phaser: a modulation effect based primarily on all-pass filter phase shift rather than variable delay.
- Haas Effect / Precedence Effect: psychoacoustic behavior associated with closely spaced arrivals and localization.
- Vibrato: periodic pitch modulation. A fully wet variable delay path can produce vibrato-like pitch movement.
- Bucket-Brigade Device (BBD): a clocked analog delay technology used in many classic chorus circuits.