RT60 — Reverberation Time

Reverberation Time (60 dB decay)

For room acoustics design principles and acoustic treatment, see audio/room-acoustics. For speech intelligibility measurement, see glossary/sti.

RT60 is the time in seconds required for a sound to decay 60 dB below its initial level after the source ceases. It is the single most important acoustic measurement for characterizing a room's suitability for speech or music reproduction. Long RT60 (high reverberation) benefits certain music performances but degrades speech intelligibility. Short RT60 (acoustically dead rooms) is ideal for speech but sounds unnatural for music and produces listener fatigue in larger spaces. Every installed AV system — conferencing, lecture capture, sound reinforcement, recording — must be designed in context with the room's RT60.

Sabine's Formula

RT60 is predicted using Sabine's equation (valid for diffuse sound fields in rooms with relatively low absorption):

RT60 = 0.161 × V / A

Where:

• V = room volume in cubic meters (m³)
• A = total absorption in sabins = Σ(Surface area × absorption coefficient)
• 0.161 = a constant derived from the speed of sound

Example: A conference room measuring 8 m × 5 m × 3 m (volume = 120 m³) with carpeted floor (α ≈ 0.35), acoustic tile ceiling (α ≈ 0.70), and hard walls (α ≈ 0.05):

• Floor: 40 m² × 0.35 = 14 sabins
• Ceiling: 40 m² × 0.70 = 28 sabins
• Walls: 76 m² × 0.05 = 3.8 sabins
• Total A = 45.8 sabins
• RT60 = 0.161 × 120 / 45.8 ≈ 0.42 seconds

This is a reasonable value for a conference room. Adding hard-surfaced furniture or removing acoustic tile would increase RT60 significantly.

For rooms with high absorption where Sabine's equation over-predicts, use the Eyring equation instead:

RT60 = -0.161 × V / (S × ln(1 - ᾱ))

where S = total surface area and ᾱ = average absorption coefficient.

Target RT60 Values by Application

ANSI S12.60 (Classroom Acoustics) requires RT60 ≤ 0.6 seconds for classrooms with volumes under 283 m³ (10,000 ft³), and ≤ 0.7 seconds for larger classrooms. This standard is referenced in many school construction specifications.

Measuring RT60

RT60 is measured using one of two methods:

1. Interrupted noise method: play broadband pink noise through the room's loudspeakers (or an omnidirectional source), abruptly stop the signal, and measure the decay rate with a sound level meter or measurement software (Smaart, Room EQ Wizard, NTI XL2 with acoustic module)
2. Impulse response method: fire a starter pistol, pop a balloon, or use a swept sine burst; record the impulse response and calculate decay time from the room's energy time curve (ETC)

Results are reported per octave band (typically 125 Hz, 250 Hz, 500 Hz, 1 kHz, 2 kHz, 4 kHz). RT60 at 500 Hz and 1 kHz are the most speech-relevant. Low-frequency RT60 (125–250 Hz) is typically longer than mid-frequency in most rooms — excessive low-frequency reverberation produces a boomy character that degrades both intelligibility and music clarity.

Because achieving a true 60 dB decay is often masked by background noise, practitioners commonly measure T20 (20 dB decay) or T30 (30 dB decay) and extrapolate to T60. Measurement software handles this automatically.

RT60 and Speaker System Design

In reverberant rooms (RT60 > 1.0 s), distributed speaker systems (many small speakers close to listeners) achieve better STI than a few large speakers at the front. Each listener receives high direct-to-reverberant ratio from the nearby speaker. Delay settings between zones must be carefully aligned to avoid echoes from speakers arriving out of phase — delay should be set so far speakers fire before (or at the same time as) near speakers from the listener's perspective.

Acoustic treatment (absorption panels, baffles, ceiling clouds) is the most effective way to reduce RT60 in an existing room. Each doubling of absorption approximately halves RT60. Common installed treatments: fabric-wrapped fiberglass panels (125–4000 Hz range), perforated wood panels with air cavity (low-frequency resonant absorbers), ceiling acoustic baffles suspended in open spaces.

Common Pitfalls

• Designing for average RT60 without checking per-band variation. A room with RT60 = 0.5 s at 1 kHz but RT60 = 1.8 s at 250 Hz will sound boomy and reduce intelligibility of low-frequency consonants. Fix: always check RT60 across all octave bands; treat low-frequency excess with bass traps or resonant panel absorbers if needed.

• Ignoring RT60 when specifying a conferencing DSP. AEC (Acoustic Echo Cancellation) tail length must exceed the room's RT60. If RT60 is 0.8 s but the DSP's AEC tail length is only 0.5 s, residual echo is present after the direct sound is cancelled. Fix: select a DSP with AEC tail length matching or exceeding the room's RT60 (Biamp Tesira AEC supports up to 200 ms; Q-SYS AEC room size setting must cover the full reverberant tail).

• Measuring RT60 in an unfinished room and designing to that value. An empty concrete room has RT60 of 2–4 seconds; the finished room with carpet, furniture, and occupants may have RT60 of 0.5–0.8 s. Designing a speaker system for an unfinished room produces massive over-engineering. Fix: use predicted RT60 from Sabine/Eyring calculations based on planned finishes, or measure after construction is complete.

• Confusing RT60 with noise floor. A room with a low noise floor but high RT60 still has poor intelligibility. A quiet, reverberant room is a common trap — it feels acoustically pleasant but produces STI in the poor range. Fix: measure both RT60 and background noise level; optimize for STI, which accounts for both variables.