facts on flammability

Slide 7 of 37

Any test for ignition and burning performance of a piece of upholstered furniture must take into account the many variables involved. There is no way to cover all the possible variations in a single test and this limitation is something we are trying to communicate to fire safety officials and regulators.

Some of the variables include filling material such as foams of different types and grades, synthetic fiber, natural materials such as garnetted cotton, sisal, feathers and hair. And combinations of these materials sometimes including layers of different grades of foam. Fabrics are key to fire performance and fabric variables include: weight, weave and denier. Porosity and construction are significant variables including the presence of welt cords and the possibility of raised fibers as found in flocks and piles. Fabric backing plays a role as do fiber treatments for soil and stain resistance and waterproofing.

The composition of the fabric is key. Fiber content can vary greatly including: cellulosics such as cotton, rayon and linen and protein-based fibers like wool and leather. There are synthetics in the thermoplastics family like polyester, nylon, polypropylene and PVC, and other synthetics such as acrylics and modacrylics and countless blends and mixtures.

Inner liners which are layers of material between the fabric and filling material include fiberglass, Nomex and carbon fibers. Inner liners impact ignition and burning characteristics.

Furniture construction is important including size of the piece, thickness of the seat and back, positioning of the seat to the back, the presence of full sides, flanges, skirts and other trim.

The ignition source is also critical including whether smoldering or open flame, the size of the flame, length of contact, configuration of the flame, and location of the ignition point in relation to the upholstered item. Room conditions, ventilation, and the presence or absence of secondary ignition sources also impact fire performance.

It is therefore apparent why a single test protocol with the goal of covering all these variables likely to be found in an actual fire remains an elusive target.

<-BACK NEXT->