How this instrument works
The absorption coefficient α is the fraction of incident sound energy a surface does not reflect. Send a known intensity of sound at a material, measure how much intensity comes back, and the absorbed share is simply what did not return: α = absorbed intensity ⁄ incident intensity. Because it is a ratio of two intensities in the same units, α is dimensionless and bounded between 0, a perfect reflector, and 1, a surface that returns nothing at all — an open window is the textbook example, since sound leaving through it never comes back to be measured.
The formula falls straight out of energy conservation at a boundary. Sound striking a surface splits into a reflected portion and everything else — energy that enters the material and never re-emerges, lost to friction as air squeezes through a porous structure like foam or fiberglass, or to bending losses in a limp panel, ultimately turning into a trace of heat. If nothing transmits through to the far side, absorbed and reflected intensity must add back up to the incident intensity, so the reflection coefficient is just 1 − α. A material described as having α = 0.30 is, in that sense, exactly 70% mirror.
One α value never tells the whole story, because absorption is strongly frequency-dependent: a foam wedge that swallows 95% of a cymbal's energy might reflect most of a kick drum's low thump straight back. Published single-number ratings like the Noise Reduction Coefficient average four mid-band values and hide that shape entirely. A subtler limit shows up in lab practice — reverberation-chamber measurements occasionally return α fractionally above 1 for small samples, an artifact of sound diffracting around the panel's edges rather than a real violation of energy conservation.
- Enter the Incident sound intensity striking the test surface — any consistent unit works, since only the ratio matters.
- Enter the Absorbed sound intensity, the portion that does not reflect back, in that same unit.
- Read the Absorption coefficient, α, a dimensionless figure between 0 and 1.
- Compare α across materials or frequencies to judge which surface tames a room, or check a manufacturer's rating against the raw energy numbers behind it.
Worked example — a carpet swatch at α = 0.30
A lab fires 100 units of sound intensity at a thick carpet-over-concrete sample and measures 30 units absorbed, the rest reflected. The formula gives α = 30 ⁄ 100 = 0.30: the carpet keeps three-tenths of every joule of sound energy that reaches it and sends the other seven-tenths bouncing back into the room. That single number is what an acoustician plugs into a room's total absorption tally when deciding whether a space needs more treatment.
Set the same carpet beside bare painted concrete, which typically returns an α closer to 0.02 to 0.05 at the same mid frequencies. The gap is roughly a factor of ten, which is why swapping a hard floor for carpet, or hanging heavy drapes over glass, makes such an audible difference to a room's liveliness — far more than the eye would guess from how thin the extra layer looks.
Questions
What do α = 0 and α = 1 mean physically?
α = 0 describes a perfect reflector: every unit of incident sound intensity bounces back, nothing is absorbed, the textbook case of a rigid, hard, sealed wall. α = 1 describes a perfect absorber, where all incident energy is retained and none returns — an open window is the standard real-world approximation, since sound passing through it simply leaves and is never measured coming back.
How is this different from reverberation time?
Absorption coefficient is a property of one surface at one frequency; reverberation time is an outcome for an entire room, built by summing every surface's area times its α and feeding that total into Sabine's formula. Raising a single wall's α from 0.05 to 0.3 changes a room's decay time, but α itself is the input measurement, not the room-level result.
Does a high absorption coefficient mean a material blocks sound from a neighbor?
No, and this is the most common mix-up in the field. Absorption keeps sound energy from bouncing back into the same room, cutting echo and reverberation; it says nothing about how much energy leaks through the material into the next room, which is governed instead by sound transmission loss. Soft foam panels absorb well yet transmit sound easily, so they quiet a room's echo without soundproofing it.
Why do product datasheets quote one NRC number instead of α at each frequency?
The Noise Reduction Coefficient averages a material's α values at 250, 500, 1000, and 2000 Hz and rounds to the nearest 0.05, giving buyers one comparable figure. It is convenient but hides frequency shape: two products can share an identical NRC of 0.70 while one absorbs bass poorly and the other absorbs treble poorly, which matters for a room with a specific noise problem.
Can a measured absorption coefficient come out above 1?
Occasionally, in reverberation-chamber testing of small samples, and it is a measurement artifact rather than a physical impossibility. Sound diffracts around the exposed edges of a test panel and adds extra apparent absorption beyond the panel's own area, inflating the calculated α slightly past 1. For the intensity ratio this instrument computes directly, α is always bounded between 0 and 1 by definition.
What are typical α values for common building surfaces?
Bare painted concrete or brick sits around 0.02 to 0.05, glass windows near 0.03 to 0.04, thick carpet over an underlay around 0.25 to 0.35, and 5 cm mineral-wool or fiberglass panels can reach 0.8 to 0.95 at mid frequencies. The roughly twenty-fold spread between hard and soft surfaces is exactly what architectural acousticians balance when treating a room.