TL;DR
Three single-number ratings exist for sound absorption, and specifying the wrong one on a project can cause confusion, re-testing, or incorrect material selection. NRC (Noise Reduction Coefficient) is the North American standard per ASTM C423, averaging absorption at 250, 500, 1000, and 2000 Hz. SAA (Sound Absorption Average) is NRC's modern replacement in the same standard, using twelve third-octave bands from 200 to 2500 Hz. αw (weighted sound absorption coefficient) is the European/ISO metric per ISO 11654:1997, using a reference curve method with a shape indicator letter. All three reduce a frequency-dependent phenomenon to a single number. All three lose critical information in the process. This article explains what each metric captures, what it hides, and when to insist on full octave-band data instead.
The Tender That Specified the Wrong Rating
A London architecture firm designing a 120-seat lecture theatre in Dubai specified ceiling tiles with "NRC ≥ 0.85" in the tender documents. The acoustic consultant was British-trained and accustomed to αw ratings. The ceiling manufacturer was European and published only αw values. The product they proposed had αw 0.80 (Class A) — which the project manager rejected because "0.80 is less than 0.85."
In fact, the product had octave-band absorption coefficients of [0.30, 0.75, 0.95, 1.00, 0.95, 0.85] at the standard six frequencies. Its NRC (average of 250-2000 Hz values) was 0.91 — well above the 0.85 requirement. The αw of 0.80 was lower because the ISO 11654 reference curve method penalises low-frequency deficiency more heavily than NRC's simple average. A three-week procurement delay and rebid cost the project approximately £12,000 in additional consultant fees — all because two different rating systems were mixed in the same specification.
NRC: Noise Reduction Coefficient (ASTM C423)
NRC is the arithmetic mean of absorption coefficients at four frequencies, rounded to the nearest 0.05:
NRC = (α₂₅₀ + α₅₀₀ + α₁₀₀₀ + α₂₀₀₀) / 4
Per ASTM C423-17, absorption coefficients are measured in a reverberation chamber using the Sabine equation. The test sample is typically 6.69 m² (72 ft²) placed on the chamber floor.
Strengths
- Simple to calculate and compare
- Widely published by North American manufacturers
- Decades of familiarity among architects and specifiers
Weaknesses
- Ignores 125 Hz entirely — critical for music rooms and spaces near mechanical equipment
- Ignores 4000 Hz — important for speech intelligibility
- Rounding to 0.05 means products with materially different performance can have the same NRC
- Values above 1.0 are common and confuse non-specialists
NRC Values for Common Materials
| Material | NRC | α₁₂₅ | α₂₅₀ | α₅₀₀ | α₁₀₀₀ | α₂₀₀₀ | α₄₀₀₀ |
|---|---|---|---|---|---|---|---|
| 25 mm mineral wool (wall) | 0.75 | 0.10 | 0.45 | 0.80 | 0.90 | 0.85 | 0.80 |
| 50 mm mineral wool (wall) | 0.95 | 0.20 | 0.70 | 0.95 | 1.00 | 0.95 | 0.90 |
| Suspended acoustic ceiling | 0.70 | 0.25 | 0.50 | 0.70 | 0.80 | 0.80 | 0.75 |
| Heavy curtain (draped) | 0.55 | 0.10 | 0.30 | 0.55 | 0.70 | 0.70 | 0.65 |
| 6 mm carpet on concrete | 0.30 | 0.05 | 0.10 | 0.25 | 0.40 | 0.45 | 0.50 |
| Painted concrete | 0.05 | 0.01 | 0.02 | 0.03 | 0.04 | 0.05 | 0.05 |
Notice that 25 mm and 50 mm mineral wool differ by NRC 0.20, but the real story is at 125 Hz: 0.10 versus 0.20. For a room with significant low-frequency content — a music rehearsal space or a room adjacent to a plant room — this factor-of-two difference at 125 Hz is invisible in the NRC rating.
SAA: Sound Absorption Average (ASTM C423)
SAA was introduced in the 2009 revision of ASTM C423 as a refinement of NRC. It averages twelve third-octave band absorption coefficients from 200 Hz to 2500 Hz:
SAA = (α₂₀₀ + α₂₅₀ + α₃₁₅ + α₄₀₀ + α₅₀₀ + α₆₃₀ + α₈₀₀ + α₁₀₀₀ + α₁₂₅₀ + α₁₆₀₀ + α₂₀₀₀ + α₂₅₀₀) / 12
SAA includes more frequencies than NRC (twelve vs four), giving a finer-grained average. It is also rounded to the nearest 0.01 rather than 0.05, providing better resolution. In practice, SAA and NRC values for the same product rarely differ by more than 0.05. Many manufacturers publish both; some publish only NRC for backward compatibility.
SAA vs NRC: When It Matters
For broadband absorbers (mineral wool, polyester fibre), SAA ≈ NRC ± 0.02. The difference is negligible for specification purposes.
For resonant absorbers (perforated panels, membrane absorbers, Helmholtz resonators), the difference can be larger because these products have narrow absorption peaks that may fall between NRC's four test frequencies but are captured by SAA's twelve bands. A Helmholtz resonator tuned to 315 Hz would register nothing in NRC (nearest frequencies are 250 and 500 Hz) but would contribute to SAA.
Model your room with full octave-band absorption data → AcousPlan Calculator
αw: Weighted Sound Absorption Coefficient (ISO 11654)
The European approach, codified in ISO 11654:1997, takes a fundamentally different mathematical path. Instead of averaging, it fits a reference curve to the measured data:
- Measure absorption coefficients in third-octave bands from 200 Hz to 5000 Hz per ISO 354:2003
- Convert to practical absorption coefficients αp (octave-band values rounded to nearest 0.05)
- Shift a standardised reference curve downward in 0.05 steps until the sum of unfavourable deviations (where measured values fall below the shifted reference curve) does not exceed 0.10
- The value of the shifted reference curve at 500 Hz is αw
Shape Indicators
When measured absorption deviates significantly from the reference curve shape, ISO 11654 appends a letter indicating where:
| Indicator | Meaning | Excess Absorption At |
|---|---|---|
| L | Low frequency | 250 Hz band |
| M | Medium frequency | 500 or 1000 Hz band |
| H | High frequency | 2000 or 4000 Hz band |
A product rated αw 0.80 (L) absorbs significantly more at low frequencies than the reference curve shape predicts. This is valuable information that NRC discards entirely.
Absorption Classes
ISO 11654 also assigns letter classes:
| Class | αw Range | Description |
|---|---|---|
| A | 0.90 - 1.00 | Highly absorptive |
| B | 0.80 - 0.85 | Highly absorptive |
| C | 0.60 - 0.75 | Absorptive |
| D | 0.30 - 0.55 | Absorptive |
| E | 0.15 - 0.25 | Reflective |
| Not classified | 0.00 - 0.10 | Reflective |
Head-to-Head Comparison
| Feature | NRC | SAA | αw |
|---|---|---|---|
| Standard | ASTM C423 | ASTM C423 | ISO 11654 |
| Frequency range | 250-2000 Hz | 200-2500 Hz | 200-5000 Hz |
| Number of bands | 4 octave | 12 third-octave | 18 third-octave (input) |
| Rounding | 0.05 | 0.01 | 0.05 |
| Method | Arithmetic mean | Arithmetic mean | Reference curve fitting |
| Maximum value | > 1.0 possible | > 1.0 possible | 1.00 (capped) |
| Shape information | None | None | L, M, H indicators |
| Classification system | None | None | Class A-E |
| Regional use | North America, Asia | North America | Europe, Middle East |
| 125 Hz included? | No | No | No (starts at 200 Hz) |
When Single Numbers Are Not Enough
All three ratings serve one purpose: simplifying product comparison for general specification. They are adequate for:
- Suspended ceiling tiles in standard offices
- Wall panels in corridors and circulation areas
- General-purpose acoustic treatment in non-critical spaces
- Recording studios and broadcast facilities (need full octave-band data)
- Music performance spaces (need 125 Hz and below)
- Spaces with specific background noise spectra (need frequency-matched treatment)
- Rooms requiring compliance with BB93, DIN 4109, or ANSI S12.60 (standards specify frequency-dependent requirements)
- Any project where the acoustic consultant needs to run RT60 calculations (these require per-frequency absorption coefficients)
The Practical Specification Guide
For North American projects: Specify SAA per ASTM C423, with NRC as an accepted alternative. Always require the manufacturer to provide full third-octave-band data in addition to the single-number rating.
For European/Middle East projects: Specify αw with class per ISO 11654, with αp (practical absorption coefficient) data at octave-band frequencies.
For international projects: Specify both NRC and αw, noting that they are not equivalent. Include a clause requiring full frequency-dependent data for acoustic design verification.
For critical applications: Skip single-number ratings entirely. Specify minimum absorption coefficients at each octave band: "α ≥ 0.80 at 500, 1000, and 2000 Hz; α ≥ 0.50 at 250 Hz; α ≥ 0.20 at 125 Hz."
Summary
NRC, SAA, and αw are different mathematical summaries of the same physical data. They are not interchangeable, and mixing them in a single specification is a common source of procurement errors. For routine commercial interiors, any single-number rating is adequate for initial product selection. For acoustic design calculations — RT60 prediction, STI modelling, compliance verification — you need the full frequency-dependent absorption coefficients that sit behind these ratings.
Compare materials by their full absorption spectra → AcousPlan Material Database