What STC measures
Sound Transmission Class is a single number derived from transmission loss measured in sixteen one-third-octave bands from 125 Hz to 4000 Hz. A reference contour is fitted to the measured curve under rules set out in ASTM E413, and the resulting classification is the STC. Higher is better. The measurement itself is normally made to ASTM E90 in a laboratory.
The rating exists to make comparison possible, and it does that job well. What it is not is a prediction of how much quieter a room will be, because that depends on the source, the background level in the receiving room and every path other than the one that was measured.
| STC | What you experience through the wall |
|---|---|
| 25 | Normal speech clearly understood |
| 30 | Loud speech understood, normal speech audible |
| 35 | Loud speech audible but not intelligible |
| 40 | Loud speech a faint murmur; a reasonable minimum between dwellings |
| 45 | Loud speech barely audible; typical hotel guest room target |
| 50 | Loud speech inaudible; music and bass still transmit |
| 60 and above | Most sources inaudible; requires a double-leaf construction |
STC belongs to an assembly, not to a material
This is where most specification errors begin. A wall does not inherit the STC of the best-performing product in it. STC is measured on a complete assembly mounted in a test aperture: studs, boards, cavity, insulation, fixings and seals. Change any of those and the number changes.
A material can be tested on its own, and doing so is legitimate for comparing barrier products against each other. BlastBlock BB830 is STC 32 at 3 mm and STC 19 at 1 mm, according to the manufacturer’s datasheet, reference BB830/20/v2. Those are the sheet by itself. What the same sheet contributes inside a wall is a different question, with a different answer.
| Assembly | STC |
|---|---|
| 3 mm MLV alone, material only | 32 |
| 12 mm plasterboard, steel studs, 12 mm plasterboard, no MLV | 33 |
| As above, with 3 mm MLV added within the wall | 51 |
| 3 mm MLV plus 50 mm mineral wool in the cavity | 53 |
| 3 mm MLV to both faces, 50 mm mineral wool | 56 |
| Double stud, separate frames, 50 mm cavity, mineral wool | 68 |
Note that the bare partition and the sheet on its own are both around STC 32 to 33, and combining them reaches STC 51. That is the point: the system is not the sum of its parts, and no product datasheet can tell you what your wall will do.
Laboratory versus field
A laboratory test isolates the specimen. The two chambers are structurally separated so that sound can only travel through the sample, and the sample is installed under ideal conditions by people who do this for a living. A building is not like that.
Field measurement to ASTM E336 produces ASTC, or NIC where the result is not normalised. As a working rule, expect a field result 5 to 10 dB below the laboratory figure for the same construction. On poorly sealed work the gap is far wider. The causes are consistent and none of them are the material:
- Flanking. Sound travels around the partition through the floor slab, the facade, a continuous ceiling void or a shared duct.
- Penetrations. Sockets, switches, pipework, sprinkler drops and cable trays.
- Perimeter sealing. A head detail left open to the deck, or a base track never caulked.
- Doors. A partition at STC 50 with an unsealed hollow-core door in it is not an STC 50 partition.
- Workmanship. Boards butted with gaps, insulation compressed, screws through resilient channels into the studs behind.
How to read a manufacturer’s number
Published acoustic figures are marketing documents as often as they are engineering documents. Six questions separate the two.
- Who measured it? A named laboratory with a report number and a date is evidence. An unattributed number in a brochure is a claim. Ours are manufacturer data, and we say so in the sentence rather than in a footnote.
- Is it a material or an assembly? If it is an assembly, is every layer described, including stud type, cavity depth, insulation and fixings? If not, you cannot reproduce it and it does not belong in a specification.
- Is it STC, Rw, or transmission loss at one frequency? A 45 dB transmission loss quoted at 4000 Hz is not an STC 45 wall. Conflating the two is the most common trick in the category.
- At what thickness and, more importantly, at what areal density? Mass is what blocks sound. Two 3 mm sheets at different densities are not comparable, and that comparison is exactly where a cheaper import wins on paper.
- Laboratory or field? If it does not say, assume laboratory and deduct 5 to 10 dB.
- Is the tolerance published? Calendered sheet varies by roll. A specification quoting density to two significant figures with no tolerance is quoting one specimen.
STC and the ratings it is confused with
| Rating | Measures | Where you meet it |
|---|---|---|
| STC | Airborne sound through an assembly, single number | North American practice, ASTM E413 |
| Rw | The same idea under ISO rules, slightly different weighting | European and much Asian specification, ISO 717-1 |
| ASTC / NIC | Airborne performance measured in the finished building | Commissioning and dispute resolution, ASTM E336 |
| IIC | Impact noise through a floor, such as footfall | Floor and ceiling build-ups, ASTM E492 |
| NRC / α | Sound absorbed by a surface, not transmitted through it | Reverberation control, ASTM C423 |
STC and Rw are close but not equal, and for the same construction they usually land within a couple of points of each other. Do not convert between them arithmetically in a specification. Ask for the rating the project is written against.
Where STC misleads you
- It stops at 125 Hz. Everything below that — bass, plant, traffic rumble — is outside the rating, so two walls with identical STC can perform very differently against music.
- It says nothing about impact noise. Footfall is IIC, and a high STC floor can still be unacceptable to the unit below.
- It describes the assembly tested, not the one built. Substitutions during construction routinely cost 5 dB or more.
- A door or a glazed screen sets the ceiling for the whole partition. Rate the weakest element, not the average.
- It is a laboratory value unless it explicitly says otherwise.
Using this on a real project
If you are specifying, state the required performance as a field value with a tolerance, describe the assembly in full, and name who verifies it and how. If you are buying material, ask for the areal density before the thickness, ask whether the number is laboratory or field, and ask who issued it. Those three questions eliminate most of what goes wrong.
IAC Acoustics (Thailand) publishes the BlastBlock datasheet and the independent fire test report in full, without a registration form, and labels laboratory values as laboratory values. Where a project needs an assembly figure rather than a material figure, that is an acoustic design question and it should be answered against the actual construction.