How to Make a Synth Sound Bigger: Detune, Chorus and Space

To make a synth sound bigger, give your ears several slightly different copies of it: a second oscillator about 10 cents out of tune, a stereo chorus, an octave layer, then a little drive for extra harmonics and some reverb for the sound to sit in. Add them in that order and stop as soon as the notes start to blur. Every one of these moves has a point where big turns into mush, and the last section shows where that point is.

What does detuning a synth do?

A single oscillator is a perfectly steady wave, and perfectly steady sounds small. A choir of twenty singers never lands on exactly the same pitch, and that spread is a big part of why it sounds like a choir.

Detuning a synth copies that spread. You set a second oscillator to the same note, then nudge it a few cents sharp or flat (a cent is one hundredth of a semitone). The two waves slide in and out of step, reinforcing each other and then cancelling, and you hear a shimmer called beating.

The speed of the shimmer is the difference in frequency between the two oscillators. Play A4 at 440 Hz with the second oscillator 10 cents sharp and it sits at about 442.5 Hz, so the sound pulses about 2.5 times a second. Every harmonic beats as well, and faster as you go up (the 2nd harmonic twice as fast, the 3rd three times as fast). That is why a detuned saw wave turns into a moving sheet of sound while a detuned sine only throbs. Saws and squares are packed with harmonics, so they are the waves that reward detune most; the guide to synth waveforms shows what each one contains.

Two habits make detune behave. First, if your synth lets you detune both oscillators, split it: set one 6 cents sharp and the other 6 cents flat, so the average pitch still agrees with a guitar or piano in the same song. Second, remember that a fixed number of cents beats faster on high notes and slower on low ones. Ten cents shimmers about three times a second at C5 and crawls along at well under once a second at C2, so play a new patch from the bottom of the keyboard to the top before you save it.

What does chorus do on a synth?

Chorus makes one sound act like several by blending it with a delayed copy of itself. The delay is 10 to 50 milliseconds, too short to hear as an echo, and a slow LFO keeps stretching and shrinking that delay time. While the delay is growing, the copy sounds a touch flat; while it is shrinking, a touch sharp. The effect is detune that never sits still, applied to everything the synth plays at once.

Stereo chorus is where the width comes from. The left and right copies wobble out of step with each other, so the two speakers never quite agree, and your ear reads the disagreement as space.

The Roland Juno-60 from 1982 is the textbook case. It has one oscillator per voice (plus a sub-oscillator), and Roland’s engineers added the built-in chorus as one of their ways to get thick sounds out of that single oscillator. Its two chorus buttons both run slowly, under 1 Hz. You can hear the same trick on any synth: one saw wave, the filter cutoff about halfway, a 300 ms attack and a slow stereo chorus turn the plainest patch there is into a usable pad.

Starting settings for any chorus:

• Rate: 0.3 to 1 Hz. Slower sounds lush; faster starts to sound like vibrato.
• Depth: low to middle. Deep and fast together is seasick.
• Mix: 30 to 50 percent. At fully wet there is no steady original left, and you hear vibrato instead of chorus.

Chorus and detune stack well because they work in different places. Detune happens inside each voice, before the filter and envelope, and beats at a steady rate for each note. Chorus happens after the synth, on the whole signal, and its movement follows its LFO. A little of each (detune for thickness, chorus for width) usually beats a lot of either one.

Unison vs detune (and how to make a supersaw)

Unison is detune with more copies. Instead of two oscillators, the synth plays several copies of every note at once, spreads them across a detune range, and usually pans them from left to right. Two detuned oscillators sound like one thick instrument; seven in unison sound like a wall.

The cost is processing: seven-voice unison on a four-note chord means 28 copies of the oscillator running at once, so unison is the first thing to turn down when a project starts to crackle.

The best-known unison sound is the supersaw. Roland’s JP-8000, from the late 1990s, had a waveform by that name that resembles seven sawtooth oscillators detuned against each other, and it became a signature sound of trance. To build one on any synth with a unison section:

1. Pick a saw wave and set unison to 7 voices.
2. Start unison detune at about a quarter of its range. If it still sounds like one thick saw, raise it; if a held chord sounds out of tune, lower it.
3. Turn spread (sometimes called width or stereo) most of the way up.
4. Leave the low-pass filter mostly open with little resonance, and cut everything below about 150 Hz with a high-pass so the bass has room.
5. Set the amp envelope to attack near zero, sustain at full, release around 300 ms.
6. Add reverb and play a triad in the middle of the keyboard, such as A minor (A3, C4, E4).

If your synth has no unison, two saws at plus and minus 7 cents through a stereo chorus get you most of the way there.

Octave layers, drive and reverb for a fuller sound

Octave layers. A second oscillator one octave down adds weight, and one an octave up, at roughly half the level of the main oscillator, adds air. Use a sine or triangle for the low layer so it stays clean. A layer a fifth up (7 semitones) sounds huge on single-note leads, but on chords it adds new notes: a C major triad (C, E, G) with fifths stacked on top gains G, B and D, which turns it into something close to Cmaj9. Save fifth layers for melodies.

Drive. Drive pushes the signal into soft clipping, which rounds off the peaks of the wave and adds new harmonics. Those harmonics help a sound hold its place in a busy mix, and they keep a bass audible on small speakers, because you hear the upper harmonics and your ear fills in the low note. Turn drive up until you hear grit, then back it off a notch.

Reverb and delay. Reverb puts the synth in a room, and a room is what makes a sound feel large. For pads, try a decay of 2 to 3 seconds at 20 to 30 percent mix. A pre-delay of 20 to 40 ms lets the start of each note through dry before the reverb arrives, so the sound stays clear in front and grows behind. A stereo delay widens without washing things out: at 120 BPM an eighth note is 250 ms and a dotted eighth is 375 ms, and a ping-pong delay that bounces between the speakers adds width you notice on headphones right away.

Gelato Synth has no unison control, so it does all of this with two oscillators. Start from the blank Init sound, set Shape to saw on both Osc A and Osc B, leave Ratio at x1, raise Hear B until B is as loud as A, and set Detune to about 10 cents (the slider reaches 50 either way). The Chorus slider sits on the same Osc B page. For an octave-down layer instead, set Ratio to x0.5. Drive is on the Filter page, and the Space page has Echo, Echo Time, Reverb, Size and Decay.

When bigger is worse: keeping bass and chords clear

Every trick above adds copies or movement, and the low end and dense chords are where copies start to fight.

Keep the sub clean. On a low note, detune beats slowly enough to hear as a change in volume: at C2 (about 65 Hz), 10 cents makes the bass swell and dip about once every 2.6 seconds, out of time with the song. Stereo chorus and wide unison down there can also cancel out on a mono speaker. The usual fix is to split the bass into two layers: a plain sine or triangle for the sub, kept centered and dry, and any detune, chorus or drive on a higher layer. The synth bass guide has recipes for both halves.

Thin out big chords. Three detuned oscillators playing a four-note Cmaj7 through chorus and a long reverb is a lot of beating at once, and the chord starts to sound out of tune. For pads playing 7th and 9th chords, halve the detune, drop the octave-up layer, and high-pass the pad around 150 to 200 Hz so the bass and kick own the bottom. Spreading the voicing helps too: C3, G3, E4, B4 is clearer than C3, E3, G3, B3 bunched together low down.

Check in mono, at matched volume. Switch your mix to mono now and then and listen for parts that shrink or vanish; those have too much stereo width. When you compare before and after any of these tricks (drive especially), match the levels, because the louder version will always sound bigger.

Leave something small. A dry, narrow verse makes the wide section after it feel wider. If every part has detune and a long reverb, nothing stands out, so pick one big sound per section and keep the rest tight.