ASTM E108 provides several different test methods to measure the relative fire characteristics of roof coverings when exposed to simulated fire sources that originate outside of a building. This standard measures and describes the response of materials, products, or assemblies when heat and flames are applied in a controlled manner. It does not account for all the factors that are required for fire hazard or fire risk assessment.
Manufacturers of roof coverings or components used in roof deck assemblies use ASTM E108 to demonstrate compliance with the fire classification requirements of the International Building Code and other specifications. Outside of a few exceptions, roof coverings used in all Types of Construction need a fire classification.
Classes of ASTM E108
ASTM E108 provides three classes of fire exposure: Class A, Class B, and Class C. Roof coverings in all classes must not slip from their position nor present a flying brand hazard. The classes are differentiated as follows:
- Class A – tests for roof coverings that are expected to be effective against severe fire exposure and/or afford a high degree of fire protection to the roof deck
- Class B – tests for roof coverings that are expected to be effective against moderate fire exposure and/or afford a moderate degree of fire protection to the roof deck
- Class C – tests for roof coverings that are effective against light fire exposure and/or afford a light degree of fire protection to the roof deck
ASTM E108 Test Methods
ASTM E108 includes five test methods for roof coverings to determine the assembly performance in response to different flame and fire conditions.
Intermittent flame exposure test
The intermittent flame exposure test is required when a roof covering is not restricted for use on noncombustible decks. Steady airflow is applied to the test deck and maintained throughout the test. Flames are applied to the test deck according to the desired Class (see chart). The airflow remains active until all evidence of flame, glow, and smoke has disappeared from the exposure surface and the material being tested or until a failure occurs.
The test results include the following observations:
- The appearance of sustained flaming on the underside of the test deck
- The production of flaming or glowing brands or displacement of portions of the test sample; and
- Exposure or falling away of portions of the roof deck
- To achieve a specified Class, the test requires a minimum of two roof decks to pass the performance requirements when tested in sequence.
Spread of flame test
The spread of flame test is required when a roof covering is not restricted to be used on noncombustible decks. Tests for Classes A and B expose the roof deck to a continuous application of the gas flame and air current for ten minutes or until the flame permanently recedes from the maximum spread point, whichever is shorter. Class C exposes the roof deck to the same conditions for four minutes instead of ten minutes.
The test results include the following observations:
- The distance to which flaming of the material has spread
- Production of flaming or glowing brands; and
- Displacement of portions of the test sample
- To achieve a specified Class, the test requires a minimum of two roof decks to pass the performance requirements when tested in sequence.
Burning brand test
The burning brand test is required when a roof covering is not restricted for use on noncombustible decks. The burning brand test is the most detailed of the ASTM E108 test methods, so we will only cover the basics of this test method.
Build the brands for Classes A and B from dry Doulas fir lumber; Class C brands must be built from dry, non-resinous white pine lumber. The brands in all tests must be free of knots and pitch pockets. Conditioning of the brands is conducted in an oven at 105 to 120°F (41 to 49°C) for a minimum of 24 hours.
Before adding the brands to the roof deck, the brands are ignited by a flame temperature of 1630 ± 50°F (888 ± 28°C). Each Class provides detailed, different instructions for how long each face of the brands are held under ignition conditions.
The test ends when the brand is completely consumed and all evidence of flame, glow, and smoke has disappeared from the exposed surface and underside of the roof deck or until failure occurs.
The test results include the following observations:
- The appearance of sustained flaming on the underside of the roof deck
- Production of flaming or glowing brands of roof covering material
- Displacement of the test sample; and
- Exposure or falling away of portions of the roof deck
To achieve a specified Class, the test requires a minimum of four roof decks (Class A) or two roof decks (Class B or C) to pass the performance requirements when tested in sequence
Flying brand test
The flying brand test is required when there is a possibility that the roof covering will break into pieces of flaming particles that may ignite on the floor. Like the spread of flame test, Classes A and B decks receive a continuous gas flame for ten minutes, while Class C decks receive the flame for four minutes.
To determine if flying brands will develop, the air current is maintained at a 12-mph flow until all evidence of flame, glow, and smoke has disappeared from the exposed surface. When treated wood shakes are included in the test sample, the air current is increased to 18 ± 0.75 mph. To achieve a specified Class, the test requires a minimum of two roof decks to pass the performance requirements when tested in sequence
Rain test
A rain test is required whenever the fire performance characteristics of the roof covering material or the roof deck assembly have the potential to be negatively impacted by water. When the rain test is required, prepare six roof decks, and expose them to an environmental cycle detailed below. After exposure, conduct the intermittent flame, burning brand, and flying brand tests. Each test uses two of the six prepared and exposed roof decks.
Expose the test decks to twelve one-week cycles, which consist of 96 hours of water exposure followed by 72 hours of drying at 140°F (60°C).
The rain test allows for an alternative testing cycle. As an alternative, expose two sets of six test decks to seven days of water exposure, two days of draining, and five days of curing at 140°F (60°C). Each cycle will be repeated seven times, except the seventh water exposure is reduced to six days. Control the temperature during the final drying cycle to keep the moisture content of the deck’s wood members between 8 to 12%. If the deck includes plywood, the plywood’s moisture content cannot exceed 8%.
Like many test standards related to fire performance, ASTM E108 provides a complex and intricate framework to test and evaluate assembly and product performance. With so many requirements for each test, manufacturers and builders often find navigating the intricacies of the standard to be difficult, confusing, and time-consuming. To prevent excessive, wasteful testing or delays in code and permit approvals, these groups typically find an experienced, accredited third-party agency best serves them.
The best third-party partners will discuss the manufacturer or builder’s needs, build an inclusive test plan that considers all assembly and component options, conduct their test(s), and certify the results through a Design Listing.
About ICC NTA, LLC: As an accredited third-party agency and part of the International Code Council (ICC) Family of Solutions, ICC NTA provides code evaluation, inspection, engineering, plan review, and product testing services, as well as independent quality and standards compliance verification for building product manufacturers. With offices, testing labs, and training facilities in Nappanee, Indiana and Bryan, Texas, ICC NTA serves residential and commercial builders, code officials, manufacturers, and suppliers throughout the building industry.
Sources: ASTM E108-20a. ASTM, 2020.
