ASTM E84 and Class A: what the flame-spread and smoke numbers actually mean

Nearly every "Class A" claim on a commercial wallcovering traces to one test method: ASTM E84, published in parallel as UL 723 and NFPA 255, and known in the field as the Steiner tunnel test. A sample is mounted along the ceiling of a 25-foot horizontal tunnel, a gas flame is applied at one end, and the apparatus records how far and how fast flame travels along the specimen and how much smoke it produces. The method does not ask whether a material eventually burns — most do — but how it behaves as a surface finish in the early minutes of a fire.

Two numbers come out. The Flame Spread Index (FSI) is scaled against two anchors that the method fixes: inorganic reinforced cement board at 0 and select red-oak flooring at 100. The Smoke Developed Index (SDI) is derived the same way from the light obscured in the tunnel's exhaust. Both are dimensionless indices, which is why a data sheet lists a bare "FSI 15 / SDI 25" with no units.

The class letters are simply bands on the FSI scale, with a single smoke cap across all of them. Under the IBC's interior-finish tables, Class A (Class I) is FSI 0–25, Class B (Class II) is 26–75, and Class C (Class III) is 76–200, and each class also caps SDI at 450 or less. Class A therefore states two conditions at once: flame spread no worse than 25, and smoke no worse than 450. Products with a very low flame number sometimes fail Class A on smoke alone, which is a known trap with heavily filled or coated goods.

For wallcovering specifically, the detail that governs everything is that E84 rates the assembly as tested — the wallcovering on its intended substrate, with its intended adhesive. A vinyl that earns Class A over Type X gypsum board with a named adhesive has not thereby earned it over wood, over another wallcovering, or with a substituted paste. The report states the test assembly, and the rating only follows a field installation that matches it. "The product is Class A" is a weaker statement than "the product is Class A over gypsum with adhesive X, per the attached E84 report."

E84 has boundaries a specifier has to respect. It is a surface-burning test of a single finish; it says nothing about a wall assembly's fire-resistance rating in hours, which is a separate test — ASTM E119 — measuring how long a wall holds fire back. The roughly ten-minute tunnel run also does not model a fully developed room fire, and for exactly that reason the IBC (Section 803) lets certain finishes, notably textile wall coverings and expanded vinyl, qualify instead through the NFPA 286 room-corner burn test rather than E84. The method is the right instrument for selecting most interior finishes under the code, but a specifier who cites it as proof of a rated wall assembly has cited the wrong test.