Delay — Time, Rhythm, and the Haas Effect

Delay Time: From Width to Echo

The delay time — the gap between the original sound and its repetition — determines what you perceive. Let’s put milliseconds on the map — these are the landmarks you’ll navigate for the rest of the course:

Under ~1 ms: The delayed copy arrives so quickly that your brain doesn’t hear two events. Instead, the delay interacts with the original signal’s phase,

creating comb filtering — a series of peaks and cancellations across the frequency spectrum that makes the sound feel thin or hollow. This is usually a problem, not a tool.

1-30 ms (the Haas zone): Your brain fuses the two signals into one perceived sound, but the delay creates a sense of width and direction. This is the Haas effect (also called the precedence effect): when two identical signals arrive within about 30 ms of each other, your brain perceives them as a single sound coming from the direction of whichever arrived first. Pan the delayed copy to the opposite side and the sound feels wider without adding a distinct echo. This is one of the most powerful stereo widening tools in mixing.

30-50 ms (the transition zone): The delay starts to separate from the original. It’s not quite a distinct echo — more of a “doubling” or “thickening” that adds presence without rhythmic clarity. Slapback delay lives here — the rockabilly vocal effect, the quick repeat that adds energy without cluttering.

50+ ms (echo threshold): The delay is perceived as a separate event — a distinct echo. This is where rhythmic delay effects begin. Set the delay time to match the tempo (quarter notes, eighth notes, dotted eighths) and the echoes create rhythmic patterns that reinforce the groove.

Your echo threshold isn’t a fixed number — it’s personal, and it changes depending on the material. A busy mix might mask individual echoes that would be obvious in a sparse arrangement. The numbers above are landmarks, not laws. Find your own threshold by experimenting: at what delay time do you start hearing a distinct repetition rather than a thickening or widening? (The Echo Threshold video explores this perceptual boundary.)

These millisecond landmarks matter beyond delay effects. Once you internalize where your echo threshold is, you have a reference point for every time-based parameter. A compressor’s attack time of 10 ms? That’s in the Haas zone — fast enough to catch a transient before your brain registers a separate event. A reverb pre-delay of 40 ms? That’s near the slapback zone — your brain hears the dry signal first, then the space fills in behind it. The delay time spectrum is a map you’ll use everywhere.

Delay Parameters

Beyond delay time, most delay plugins offer:

Feedback: How much of the delayed signal gets fed back into the delay input. Zero feedback = one echo. Low feedback = a few fading repeats. High feedback = many repeats that gradually die out. Maximum feedback = infinite repeats that never die (and eventually build into a wall of noise — handle with care).

Mix (wet/dry): The balance between the original signal and the delayed signal. In most mixing applications, the delay is much quieter than the original — a subtle echo, not a competing signal. On a send/return setup (Chapter 11), the mix is 100% wet on the return, and the send level controls the balance.

Filter/EQ: Many delay plugins include filters on the delayed signal. A low-pass filter on the echoes simulates the natural high-frequency absorption that happens in real spaces — each repeat sounds slightly darker than the last, like an echo bouncing off distant walls. This is one of the key differences between delay characters.

Modulation: Some delays modulate the delay time slightly, adding pitch variation to the echoes. This simulates the imprecision of analog and tape delay circuits. Light modulation adds warmth and life. Heavy modulation produces a chorus-like effect.

Delay Characters

Digital delay: Clean, precise, identical repeats. Each echo is a perfect copy of the original. Useful when you want rhythmic clarity without coloration. Can sound sterile if used dry — filtering the repeats helps.

Analog delay: Warm, slightly degraded repeats. Analog circuits introduce noise and high-frequency rolloff with each repeat. The echoes progressively darken and lose definition, which often sounds more musical and less mechanical than digital delay.

Tape delay: The warmest and most characterful. Tape machines introduce wow and flutter (pitch instability), saturation (harmonic distortion), and progressive high-frequency loss. The echoes have a distinctly “vintage” quality — each repeat is noticeably different from the last. Tape delay dominated recorded music from the 1950s through the 1970s and is widely emulated in software.

The character differences are about what happens to the repeats, not the original signal. A tape delay’s first repeat is warmer and wobblier than a digital delay’s first repeat. By the fourth or fifth repeat, the tape delay has degraded significantly while the digital delay is still pristine. Choose based on what serves the song — clarity or character.

Feedback: Delay’s Wild Side

Audio feedback happens when a signal loops: a microphone picks up sound from a speaker that’s reproducing the microphone’s signal, and the cycle repeats — getting louder each time until it howls. It’s a room problem as much as a gain problem. Rooms with strong reflections bounce the speaker output back to the microphone more efficiently.

In the context of delay, feedback is a creative tool, not just a hazard. The feedback knob on a delay plugin controls how much of the delayed signal gets fed back into the input. Low feedback gives you a few fading echoes. High feedback gives you long trails of repeats. Maximum feedback creates infinite repeats that build on themselves — the same principle as a microphone pointed at a speaker, but controlled and intentional.

Ringing out a room — the practice of slowly raising a microphone’s gain until feedback begins, then notching out that frequency with an EQ — is a live sound technique, not a mixing technique. But it illustrates the same principle: feedback reveals the resonant frequencies of a system. In delay, those resonances are the creative material.

Comb Filtering: Delay’s Dark Side

When a delayed signal is mixed with the original at equal levels, and the delay time is very short (under a few milliseconds), the two signals interfere constructively at some frequencies and destructively at others. The result is a series of evenly spaced peaks and notches in the frequency spectrum — it looks like the teeth of a comb, hence the name.

Comb filtering sounds thin, hollow, and phasey. It happens accidentally in recording when two microphones pick up the same source at slightly different distances — each microphone captures the sound at a different time, and when mixed together, the comb filter appears.

In mixing, you can avoid comb filtering by keeping delay times above the Haas zone (above ~30 ms) when blending delayed signals with the original, or by using the delay on a send/return so the original signal isn’t doubled at equal levels.

White Noise and Pink Noise as Diagnostic Tools

White noise contains equal energy at every frequency — it’s useful for revealing comb filtering, room modes, and frequency response problems. Play white noise through a delay and you’ll hear the comb filter clearly — the peaks and notches become audible as a tonal coloration.

Pink noise has equal energy per octave (lower frequencies are louder), which better matches how human hearing perceives frequency balance. It’s useful for calibrating monitoring levels and checking the spectral balance of a room.

Neither is a mixing tool — they’re diagnostic tools. But understanding them helps you hear what’s happening when delay interactions create unexpected tonal changes.

What to Practice

1. Explore the Haas effect. Duplicate a mono track and pan the original hard left, the copy hard right. Add a short delay (10-25 ms) to the copy. Listen to how the sound widens dramatically. Now check in mono — notice how it collapses. This is the Haas effect’s strength and its weakness.
2. Set up a tempo-synced delay. Add a delay on a send, set it to dotted eighth notes at your session tempo, roll off the highs on the repeats, and send a vocal or guitar to it. Listen to how the rhythmic echoes fill space between phrases without cluttering the mix.
3. Compare delay characters. If you have access to digital, analog, and tape delay plugins, run the same signal through each with identical times and feedback. Listen to how the repeats differ — clean vs. warm vs. wobbly. Notice which character fits different musical contexts.
4. Hear comb filtering. Duplicate a track, keep both panned center at equal levels, and add a 2 ms delay to the copy. Listen to the thin, hollow sound. That’s comb filtering. Now increase the delay to 30 ms and hear it transition into doubling. This teaches you why microphone placement matters in recording.