Imagine a dam holding back a river. The dam's height tells you how much water it can block. Build a taller dam and you hold back more water. Simple enough. But floodwater does not always come over the top. It seeps through cracks in the foundation. It finds gaps where the dam meets the riverbank. And if the river rises slowly with a sustained, low surge, even a tall dam can be overwhelmed in ways the original engineer never anticipated.
Sound insulation works the same way. A wall, floor, or door acts as a dam for noise. STC and Rw are the numbers that tell you how high that dam is. But bass frequencies are the floodwater that finds its way around even the tallest barriers. Understanding what these ratings actually measure, where they come from, and where they fall short is essential for anyone designing spaces where noise control matters.
Two Numbers, Two Standards, One Goal
If you have ever read a building product specification, you have seen one of these numbers. In North America, a wall's ability to block airborne sound is expressed as STC (Sound Transmission Class), defined by ASTM E413. In Europe, Asia, Australia, and most of the rest of the world, the equivalent rating is Rw (Weighted Sound Reduction Index), defined by ISO 717-1.
Both systems exist for the same reason: architects and specifiers need a single number they can compare across products. Nobody wants to read a table of sixteen frequency-specific transmission loss values for every wall assembly in a project. STC and Rw distill that table into one number, measured in decibels. Higher is better. An STC 50 wall blocks noticeably more sound than an STC 40 wall.
But these two numbers are not interchangeable, even when they happen to have the same numeric value. They are measured differently, they cover different frequency ranges, and they handle low-frequency sound in fundamentally different ways. Treating them as identical is a specification error that leads to real-world noise complaints.
How STC Is Measured
The STC rating process begins with a laboratory measurement defined by ASTM E90 (Standard Test Method for Laboratory Measurement of Airborne Sound Transmission Loss of Building Partitions and Elements). Two reverberant test chambers are constructed side by side, separated only by the wall assembly being tested. A calibrated sound source is placed in one room (the source room), and microphones are placed in the other (the receiving room).
The sound source produces noise covering the full frequency spectrum. Measurements are taken at 16 one-third octave band center frequencies from 125 Hz to 4000 Hz: 125, 160, 200, 250, 315, 400, 500, 630, 800, 1000, 1250, 1600, 2000, 2500, 3150, and 4000 Hz. At each frequency, the difference between the sound pressure level in the source room and the receiving room is recorded. This difference, corrected for the area of the test specimen and the absorption in the receiving room, gives the Transmission Loss (TL) at each frequency.
The result is a TL curve: a line plotting how many decibels of sound the wall blocks at each of the 16 frequencies. Walls block high frequencies more effectively than low frequencies, so this curve typically rises from left to right.
The Reference Contour
Here is where STC becomes more than just a measurement. ASTM E413 defines a reference contour — a specific curve shape. The measured TL curve is compared against this reference contour by sliding the contour vertically until two conditions are met:
- No individual TL value falls more than 8 dB below the reference contour at any single frequency.
- The sum of all deficiencies (where measured TL falls below the contour) does not exceed 32 dB total.
This fitting procedure means that STC is not a simple average. It is a curve-matching exercise that tolerates some deviation at individual frequencies as long as the overall shape is close enough. A wall with a deep dip in transmission loss at one particular frequency can still achieve a high STC rating, provided the dip does not exceed 8 dB and the total shortfall stays under 32 dB.
How Rw Is Measured
The Rw measurement process is conceptually similar but differs in several important details. The laboratory test is conducted per ISO 10140 (a multi-part standard that replaced the older ISO 140 series). The test setup — two chambers separated by the specimen — is comparable to the ASTM E90 arrangement.
However, ISO 717-1 uses 16 one-third octave bands from 100 Hz to 3150 Hz: 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1000, 1250, 1600, 2000, 2500, and 3150 Hz. Notice two differences from the STC range: Rw extends one band lower (down to 100 Hz) and stops one band sooner at the high end (3150 Hz instead of 4000 Hz).
The reference curve fitting procedure in ISO 717-1 is also slightly different. The reference curve is shifted in 1 dB steps until the sum of unfavourable deviations (where measured values fall below the shifted reference) is as large as possible without exceeding 32 dB. There is no individual 8 dB deficiency limit as in ASTM E413. The Rw value is read from the shifted reference curve at 500 Hz, just like STC.
Spectrum Adaptation Terms: C and Ctr
This is where Rw pulls ahead of STC in practical usefulness. ISO 717-1 defines two spectrum adaptation terms that modify the single-number rating to account for different types of noise:
- Rw + C — adjusts for broadband noise sources like speech, children playing, railway traffic at medium/high speed, highway traffic above 80 km/h, jet aircraft at short distance, and factory noise with predominantly mid/high frequency energy. The C term is calculated using a pink noise spectrum weighting.
- Rw + Ctr — adjusts for low-frequency dominant noise sources like urban road traffic, disco music, mechanical plant noise, and aircraft at long distance. The Ctr term uses a traffic noise spectrum weighting that emphasizes frequencies below 500 Hz.
STC has no equivalent mechanism. It produces one number. Period.
STC vs Rw: Side-by-Side Comparison
| Feature | STC (ASTM E413) | Rw (ISO 717-1) |
|---|---|---|
| Region | North America | Europe, Asia, Australia, most of world |
| Frequency range | 125 -- 4000 Hz | 100 -- 3150 Hz |
| Laboratory test standard | ASTM E90 | ISO 10140 |
| Field test standard | ASTM E336 (gives FSTC) | ISO 16283-1 (gives R'w) |
| Reference curve fitting | Max 8 dB single deficiency, max 32 dB total | Max 32 dB total (no single-band limit) |
| Adaptation terms | None | C (pink noise), Ctr (traffic noise) |
| Typical range for walls | STC 33 -- 65+ | Rw 33 -- 65+ |
| Numeric relationship | For most assemblies, STC is approximately equal to Rw | Rw is approximately equal to STC |
For most conventional wall assemblies, the STC and Rw values are within 1-2 dB of each other. This numerical similarity is coincidental — it results from the fact that both reference curves have similar shapes and both are read at 500 Hz. It does not mean the ratings are measuring the same thing, and the difference grows for assemblies with unusual frequency-dependent behavior such as laminated glass or mass-spring-mass systems.
Common Wall Assemblies and Their Ratings
Here is where theory meets the specification sheet. These are representative values for common construction types. Actual values depend on workmanship, edge conditions, and specific products used.
| Wall Assembly | STC | Rw | Rw + Ctr | Typical Use Case |
|---|---|---|---|---|
| Single 13 mm plasterboard on wood studs, no insulation | 33 | 33 | 27 | Interior partition (inadequate for privacy) |
| Single 13 mm plasterboard on steel studs, 50 mm glass wool | 39 | 39 | 33 | Light interior partition |
| Double 13 mm plasterboard on steel studs, 50 mm glass wool | 48 | 48 | 42 | Standard office partition |
| Double 13 mm plasterboard on staggered steel studs, 2x 50 mm glass wool | 53 | 52 | 46 | Conference room wall |
| Double stud wall, insulation in both cavities, quad plasterboard | 58 | 57 | 51 | Hotel room separating wall |
| 150 mm cast concrete (2300 kg/m3) | 50 | 50 | 46 | Structural separating wall |
| 200 mm cast concrete (2300 kg/m3) | 55 | 55 | 50 | Apartment party wall |
| 200 mm concrete masonry unit, plastered both sides | 52 | 52 | 47 | Residential separating wall |
| Acoustic door with perimeter seals and drop seal | 32 -- 42 | 32 -- 42 | 26 -- 36 | Studio, meeting room, clinic |
| 6.38 mm laminated glass (acoustic interlayer) | 34 | 34 | 31 | Office glazing |
| Double glazing (6 mm - 100 mm air - 6 mm) | 40 | 40 | 35 | Facade glazing |
Notice the Rw + Ctr column. Every assembly loses performance against low-frequency noise. The gap between Rw and Rw + Ctr is typically 4 to 7 dB. For the single plasterboard partition, the Ctr penalty is 6 dB, dropping effective performance from 33 to 27 dB against traffic noise. That is the difference between marginal and useless.
The Mass Law: Why Heavier Walls Block More Sound
The most fundamental principle in sound insulation is the mass law: at a given frequency, doubling the surface mass of a wall increases its transmission loss by approximately 6 dB. A 200 kg/m2 concrete wall blocks about 6 dB more sound than a 100 kg/m2 wall, all else being equal.
This is why concrete and masonry walls provide good sound insulation without any special engineering — they are simply heavy. A 200 mm concrete slab weighs approximately 480 kg/m2 and achieves Rw 55 through mass alone.
But the mass law has a critical exception.
The Coincidence Dip
Every rigid panel has a critical frequency at which the wavelength of bending waves in the panel matches the wavelength of sound in air. At this frequency, the panel vibrates efficiently in sympathy with the incoming sound wave, and transmission loss drops sharply. This is called the coincidence dip, and it can reduce insulation by 10 dB or more at the affected frequency.
For a 13 mm plasterboard sheet, the critical frequency is approximately 2500-3000 Hz. For 150 mm concrete, it drops to around 100-120 Hz. For 6 mm glass, it sits around 2000 Hz. The critical frequency is inversely proportional to the panel thickness — thinner panels have higher critical frequencies.
This means you cannot simply increase mass indefinitely and expect proportional improvement. Every material has a frequency band where it underperforms its mass-law prediction. Acoustic engineers design around coincidence dips by using composite constructions (mass-spring-mass systems) that shift the resonant behavior into less problematic frequency ranges.
Mass-Spring-Mass: The Double Wall Advantage
A double wall — two separate panels with an air gap between them — behaves as a mass-spring-mass system. Below a resonant frequency (determined by the mass of the two panels and the width of the air gap), the system performs worse than a single wall of equivalent total mass. Above that resonant frequency, performance improves dramatically — by approximately 12 dB per doubling of frequency rather than the 6 dB predicted by the mass law for a single panel.
This is why a double stud wall with two 13 mm plasterboard sheets per side (total mass approximately 40 kg/m2) can achieve STC 58, while a single concrete wall of much greater mass (200 kg/m2) achieves only STC 55. The double wall exploits the mass-spring-mass principle to punch above its weight class.
Adding glass wool or mineral fiber insulation in the cavity improves performance further. The insulation does not add significant mass. Instead, it dampens the air spring, reducing resonance effects and absorbing sound energy that would otherwise bounce between the two panels. Adding 50 mm of glass wool to a double wall cavity typically adds 5 to 8 STC/Rw points.
Flanking Transmission: The Wall Is Only Part of the Story
Here is the uncomfortable truth that no data sheet will tell you. Even an STC 60 wall can be defeated by flanking transmission — sound that bypasses the wall entirely by traveling through connected structures.
The most common flanking paths include:
Ceiling plenum. In offices with suspended ceilings, the space above the ceiling tiles is often continuous across partition walls. Sound enters the ceiling void on one side, travels across, and re-enters the room on the other side. A wall built only from floor to suspended ceiling (not to the structural slab above) will never achieve its rated STC, regardless of its construction. The effective insulation can drop by 10 to 20 dB.
Floor slab continuity. In concrete-framed buildings, the floor slab runs continuously under partition walls. Sound energy enters the slab on one side, propagates through the concrete, and radiates into the adjacent room. This is structure-borne flanking, and it sets a practical upper limit on the insulation achievable by improving the wall alone.
Ductwork and services. HVAC ducts that serve both rooms create a direct acoustic path. Even sealed duct penetrations transmit sound through the duct walls. Electrical back-to-back outlet boxes in a partition wall create a gap in the insulation where sound leaks through.
Doors and glazing. A wall is only as good as its weakest element. An STC 55 wall with an STC 32 door produces a composite insulation value of approximately STC 31 if the door is 2 m2 in a 15 m2 wall. The door dominates because sound follows the path of least resistance.
This is why field measurements (FSTC per ASTM E336, or R'w per ISO 16283-1) are always lower than laboratory measurements. The laboratory test eliminates flanking by design. The real building has flanking everywhere. A typical field-to-lab difference is 3 to 7 dB, but poorly detailed constructions can lose 15 dB or more.
When STC and Rw Are Not Enough
Both STC and Rw share a fundamental limitation: they are weighted toward mid-frequency performance. The reference curves used in both systems assign the most weight to the 500-2000 Hz range, which corresponds to the frequency range of conversational speech. This makes sense for the most common sound insulation problem — speech privacy between adjacent rooms.
But it fails for three important scenarios:
Bass-Heavy Sources
Music venues, nightclubs, home theaters, and mechanical plant rooms generate significant low-frequency energy below 250 Hz. A wall rated STC 55 might provide only 35 dB of actual transmission loss at 63 Hz. The STC rating gives no indication of this weakness because 63 Hz is not even in the test range.
Rw handles this somewhat better through the Ctr adaptation term. An Rw 55 (Ctr -8) wall tells you that effective performance drops to 47 dB against traffic-type noise. But even Ctr does not extend below 100 Hz, where the deepest bass energy lives.
Impact Sound
STC and Rw measure airborne sound insulation only. They tell you nothing about impact sound — footsteps, dropped objects, vibrating machinery. Impact sound insulation is measured by a separate set of metrics: IIC (Impact Insulation Class, per ASTM E989) in North America and Ln,w (Weighted Normalized Impact Sound Pressure Level, per ISO 717-2) internationally. These are whole separate rating systems with their own test standards.
Low-Frequency Resonance
Some wall assemblies have structural resonances at specific low frequencies that create severe dips in their transmission loss curves. A mass-spring-mass double wall, for instance, has a resonant frequency typically between 50 and 100 Hz where it actually transmits more sound than a single wall of equivalent mass. This resonance falls below the STC and Rw test ranges and is invisible in the single-number rating.
Building Code Requirements
Every building code specifies minimum sound insulation between certain room types. Knowing which standard applies to your project determines whether you need STC or Rw.
North America: IBC 2021
The International Building Code, Section 1207, requires:
- STC 50 (laboratory) or FSTC 45 (field) between dwelling units and between dwelling units and public or service areas
- IIC 50 (laboratory) or FIIC 45 (field) for floor-ceiling assemblies between dwelling units
- These requirements apply to walls, partitions, and floor-ceiling assemblies separating dwelling units
Germany: DIN 4109:2018
The German standard specifies minimum weighted sound reduction index:
- Rw 53 dB between apartments (airborne)
- Rw 57 dB between apartments and particularly noisy commercial premises
- L'n,w 53 dB maximum impact sound level between apartments
- Enhanced requirements available in DIN 4109-5 for higher comfort levels (SSt II: Rw 56, SSt III: Rw 59)
Australia: NCC 2022 (Volume 1, Part F5)
The National Construction Code requires:
- Rw + Ctr 50 between sole-occupancy units (the explicit use of Ctr is notable — Australia requires the traffic noise adaptation term, not bare Rw)
- Ln,w + CI 62 maximum impact sound (with impact spectrum adaptation term)
- These are among the most stringent residential sound insulation requirements in the world
United Kingdom: Approved Document E
- DnT,w + Ctr 45 dB minimum airborne insulation between dwellings (the DnT,w notation indicates a field-standardized level difference, not a laboratory R'w)
- L'nT,w 62 dB maximum impact sound level
France: NRA 2000
- DnT,A 53 dB between dwellings
- DnT,A 58 dB between a dwelling and a commercial space
- L'nT,w 58 dB maximum impact sound level between dwellings
Practical Guidance: Choosing and Specifying Sound Insulation
If you are specifying a wall assembly for sound insulation, here are the questions to ask in order:
1. Which standard applies? If your project is in North America, you need STC. If it is in Europe, Asia, or Australia, you need Rw. If your client operates internationally, specify both.
2. What is the noise source? For speech privacy (offices, healthcare, hotels), the standard STC/Rw rating is usually adequate. For music, mechanical equipment, or traffic noise, you need Rw + Ctr or a full frequency-by-frequency analysis that extends below 125 Hz.
3. What about flanking? Laboratory ratings assume zero flanking. Your building has flanking. Budget a 5 dB reduction from lab to field as a minimum, and more if the construction details are not carefully controlled. If you need FSTC 50 in the field, specify STC 55 or higher in the lab.
4. What is the weakest element? Calculate the composite STC/Rw of the entire partition including doors, glazing, and penetrations. The weakest element dominates. There is no point specifying an STC 60 wall if the door is STC 30.
5. Does the assembly address the critical frequency band? Check the transmission loss data at the specific frequencies that matter for your noise source. Do not rely on the single-number rating alone. A good manufacturer publishes full one-third octave band TL data — if they only publish STC/Rw, ask for the underlying test report.
Try It Yourself
AcousPlan's sound insulation calculator includes 52 wall assemblies with both STC and Rw ratings, plus full one-third octave band transmission loss data. You can compare assemblies side by side, check compliance against IBC, DIN 4109, NCC, and other building codes, and see exactly where each assembly's frequency-dependent weaknesses lie.
Model your partition, check it against your applicable building code, and get a clear answer about whether your wall is high enough to hold back the flood.