Estimating Reduced Fire Risk Resulting From an Improved Mattress Flammability Standard.
Estimating Reduced Fire Risk Resulting From an Improved
Mattress Flammability Standard.
(1346 K)
Ohlemiller, T. J.; Gann, R. G.
NIST TN 1446; 80 p. August 2002.
Available from:
Government Printing Office, Washington,
DC 20401-0003.
Telephone: 202-512-1800.
Fax:
202-512-2250.
Website: http://www.bookstore.gpo.gov
Keywords:
mattresses; fire risk; flammability; standards;
ignition; cone calorimeters; heat release rate; heat
flux; heat transport; smoke transport; thermal ignition
Abstract:
This study addresses the hazards posed by bed fires of
varied sizes in an effort to relate potential fire size
reduction to decreased risk of bed fire fatalities. For
this purpose, a "bed" refers to a mattress, a foundation
and bedclothes. There are three hazards: (I) the
potential for a bed fire, by itself, to cause flashover
in a bedroom; (2) the probability that a bed fire will
ignite additional objects in the same room as a result
of the fire's radiated heat; possibly leading to
flashover and (3) the heat and toxic gas threat, in the
room of fire origin and beyond, due to the bed fire
alone. To address the first two issues, twin and
king-size beds of three designs (always using the same
bedclothes) were burned in duplicate both under an open
hood and in a room. The three designs (termed MI, M3 and
M5, consistent with previous usage) produced widely
different peak heat !'j. release rates; open hood peaks
ranged from 160 kW (M3 twin fire) to 3850 kW (MI king
fire). . The radiant heat flux distribution around the
fires was measured using up to two arrays of five flux
gages each. These radiant flux reach data were
translated into piloted ignition reach outward from the
edge of the burning bed by means of ignitability results
for seven materials. IN These materials are viewed as
surrogates for the surfaces of other potentially
significant combustible objects such as chairs or drapes
which, if ignited by the bed fire, could push the room
to a flashover condition. The results were then cast
into a form that should be related to the probability of
ignition in the overall population of bedrooms, i.e.,
the fraction of the room area (beyond the bed) that is
at risk of piloted ignition of a second object by a bed
fire. This fraction, is strongly influenced by the size
of the bed fire and of the size of the room. Both MI and
M5 \ fires are inferred to present substantial threats
of second object ignition. The threat for an M3 fire is
localized to the bed periphery but could be
problematical in small bedrooms. To address the third
hazard, heat and toxic gases, CF AST was run using the
measured heat release rate curves from the room tests as
input. This program predicts the spread of hot, toxic
smoke in the room of fire origin and beyond. Here the
context was a single story house consisting of four
monitored spaces. The MI and M5 fires presented a
substantial lethal threat due to heat exposure in the
room of fire origin and some spaces beyond. The M3 fire
(consisting mainly of the bedclothes with little
contribution from the mattress/foundation) was
substantially less threatening but not innocuous,
especially with a king size bed. Minimizing the
occurrence of flashover and its attendant fatalities
would require limiting the heat release rate peak from a
mattress/foundation to less than that of an M5 fire.
Localizing the fire to the bed would save significantly
more lives but would require limiting the
mattress/foundation heat release rate to near zero as
with the M3 design. Even this case presents some threat
of heat lethality away from the bed (in addition to the
localized threat) since the bedclothes set used here
yielded a 400 kW fire atop a king-size M3 mattress.
Overall the study suggests that beds with fire
performance similar to the M3 design would achieve very
significant reduction in fire risk. Since the bedclothes
contributed the bulk of the heat release seen with the
M3 beds, further reduction in fire fatalities would
probably have to address the bedclothes flammability as
well.
Building and Fire Research Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899