Question 1

Why does my RT60 measurement give different results each time?

Accepted Answer

Variation between RT60 measurements is normal and depends on several factors. Position-to-position variation of ±10-15% is typical due to the room not being a perfect diffuse field. Time-varying noise (HVAC cycling, traffic) can contaminate the decay curve. Insufficient INR causes the noise floor to corrupt the lower portion of the decay, leading to underestimated RT60. Door/window positions changing between measurements alter the room absorption. Temperature and humidity changes affect air absorption at high frequencies. To improve consistency: use swept signal (not impulsive) sources, ensure INR exceeds 45 dB for T30, keep conditions constant between measurements, average multiple positions, and use SonaVyx r-squared validation to flag unreliable decays.

Question 2

My RT60 at low frequencies seems unrealistically long. What is wrong?

Accepted Answer

Long low-frequency RT60 readings have several possible causes. First, room modes create standing waves that decay independently and slowly, especially at frequencies near the room resonance. At 63 Hz, the wavelength is 5.4 m, so even moderate rooms have only a few modes. Second, insufficient source energy at low frequencies means the sweep does not excite the room enough, leading to low INR and unreliable decay fitting. Third, HVAC and traffic noise can mask the low-frequency decay. Fourth, the microphone proximity to a wall node can cause level anomalies. To improve: use a larger speaker or subwoofer for the source, increase sweep duration to 15-20 seconds, measure away from walls, and check that INR exceeds 35 dB in the 63 and 125 Hz bands. SonaVyx flags bands with low INR.

Question 3

Why is my EDT much shorter or longer than T30?

Accepted Answer

EDT and T30 measure different parts of the decay curve, so they naturally differ. EDT shorter than T30 occurs when there are strong early reflections that cause rapid initial decay, but the late reverberant tail decays slowly. This is common in rooms with nearby reflective surfaces (low ceiling, close side walls) combined with large volume. The sound feels clear but reverberant. EDT longer than T30 occurs when there is a long early buildup before the decay begins, or when the room has coupled volumes (e.g., a stage house connected to an auditorium). Ideally, EDT/T30 should be between 0.8 and 1.2. Ratios outside this range indicate non-exponential decay, which SonaVyx flags in the analysis summary.

Question 4

How do I handle background noise contaminating my RT60 measurement?

Accepted Answer

Background noise sets a floor on the decay curve, limiting the usable dynamic range. Steps to minimize contamination: measure during the quietest time possible (early morning, weekend). Turn off HVAC systems if building management allows it. Close all doors and windows. Use a louder source (bigger speaker, higher volume). Use a longer sweep signal (concentrates more energy). In post-processing, SonaVyx applies the Lundeby noise floor detection algorithm to identify where the decay merges with the noise and truncates the impulse response at that point before applying Schroeder integration. If background noise is unavoidable, rely on T20 (which needs less dynamic range) rather than T30, and note the limitation in your report.

Question 5

My RT60 measurement shows a double slope. What causes this?

Accepted Answer

A double slope (also called non-exponential decay or coupled-volume decay) appears when two acoustically connected spaces have different reverberation times. Common examples: a church nave coupled to a transept, a theater auditorium coupled to a stage house, or a room with a dropped ceiling creating a plenum cavity above. The initial decay follows the shorter RT60 of the primary space, then transitions to the longer RT60 of the coupled space. ISO 3382 acknowledges this phenomenon and recommends fitting separate linear regressions to each slope segment. SonaVyx detects non-linear decay curves through r-squared validation. When r-squared drops below 0.95, it flags a potential double slope and reports both apparent decay rates.

Question 6

Why does RT60 differ between the occupied and unoccupied room?

Accepted Answer

People absorb sound significantly. Each seated person adds approximately 0.4-0.6 m² Sabins of equivalent absorption area (ISO 354 laboratory data). In a 500-seat venue, the audience adds 200-300 m² Sabins, which can reduce RT60 by 30-50% compared to the empty room. The effect is largest in rooms that are otherwise reverberant (few absorptive surfaces). In well-treated rooms with carpet, ceiling tiles, and wall panels, the audience makes less relative difference because the room already has substantial absorption. Chairs also matter: upholstered chairs absorb almost as much as occupied chairs, while unupholstered chairs add very little. This is why good venue design uses upholstered seating to maintain consistent acoustics regardless of attendance. SonaVyx lets you record measurements for both conditions and compare.

Question 7

The r-squared value of my RT60 decay is below 0.95. Is the result valid?

Accepted Answer

An r-squared below 0.95 indicates that the linear regression fit to the energy decay curve is poor. This means the decay is not following a simple exponential pattern, and the reported RT60 may not be reliable. Common causes include: insufficient INR (noise floor contaminating the lower part of the decay), double-slope decay from coupled volumes, discrete echoes creating steps in the decay, or non-diffuse sound field. ISO 3382 recommends r-squared of at least 0.95 (T20) or 0.99 (T30) for valid results. If you cannot achieve this, try: increasing source level, using a longer sweep, measuring in more positions, or checking for obvious acoustic anomalies (flutter echo, coupled spaces). SonaVyx flags bands with low r-squared in red.

Question 8

How do I know if my measurement microphone position is correct?

Accepted Answer

ISO 3382 specifies: microphone at least 1 m from any reflecting surface (walls, furniture), at least 2 m from the sound source, and at a height of 1.2 m (seated ear height) or 1.5 m (standing ear height). Different microphone positions should be separated by at least half a wavelength of the lowest frequency of interest. Avoid positions near room corners (where modes are strongest and least representative), directly on the axis between parallel walls (flutter echo paths), or close to the sound source (direct sound dominates over reverberant field). For system tuning, measure at the primary listening position plus 2-4 additional positions representing the audience area. SonaVyx measurement wizard guides you through positioning requirements step by step.

RT60 Measurement Troubleshooting: Common Problems and Solutions

Frequently Asked Questions