[Federal Register Volume 66, Number 197 (Thursday, October 11, 2001)]
[Proposed Rules]
[Pages 51886-51891]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 01-25442]


=======================================================================
-----------------------------------------------------------------------

CONSUMER PRODUCT SAFETY COMMISSION

16 CFR Part 1633


Standard To Address Open Flame Ignition of Mattresses/Bedding; 
Advance Notice of Proposed Rulemaking

AGENCY: Consumer Product Safety Commission.

ACTION: Advance Notice of Proposed Rulemaking (ANPR).

-----------------------------------------------------------------------

SUMMARY: The Commission is considering issuing a flammability standard 
that would address open flame ignition of mattresses/ bedding. The 
Commission currently has a flammability standard that addresses 
ignition of mattresses by cigarettes. However, mattress/bedding fires 
ignited by small open flames are a significant problem not addressed by 
the existing standard. In 1995, the Commission staff began a project on 
mattress fires, and the information obtained from that research is 
reflected in the ANPR. This ANPR also addresses two subsequently-filed 
petitions from the Children's Coalition for Fire-Safe Mattresses 
(``CCFSM'') requesting that the Commission issue an open flame standard 
similar to the full-scale test set forth in California Technical 
Bulletin 129 or an open flame standard similar to the component test 
set forth in British Standard 5852. The Commission invites comments 
concerning the risk of injury identified in this notice, the regulatory 
alternatives being considered, and other possible alternatives. The 
Commission also invites submission of any existing standard or 
statement of intention to modify or develop a voluntary standard to 
address the flammability risk of mattress/bedding fires ignited by 
small open flames.

DATES: Comments and submissions must be received by December 10, 2001.

ADDRESSES: Comments should be mailed, preferably in five copies, to the 
Office of the Secretary, Consumer Product Safety Commission, 
Washington, DC 20207-0001, or delivered to the Office of the Secretary, 
Consumer Product Safety Commission, Room 502, 4330 East-West Highway, 
Bethesda, Maryland; telephone (301) 504-0800. Comments also may be 
filed by telefacsimile to (301)504-0127 or by email to [email protected]. Comments should be captioned ``Mattress ANPR.''

FOR FURTHER INFORMATION CONTACT: Margaret Neily, Directorate for 
Engineering Sciences, Consumer Product Safety Commission, Washington, 
DC 20207; telephone (301) 504-0508, extension 1293.

SUPPLEMENTARY INFORMATION

A. Background

    The Commission currently has a flammability standard for mattresses 
that addresses ignition by cigarettes. 16 CFR part 1632. Smoldering 
ignition of mattresses/bedding (usually caused by cigarettes) has 
declined since the standard took effect in 1973. However, the open 
flame ignition of mattresses/bedding continues to cause a significant 
number of deaths and injuries, especially to children. The most common 
open flame sources are lighters, candles and matches. The Commission 
staff has been evaluating data concerning such fires for several years 
to determine how best to address open flame ignition of mattresses/
bedding.
    In 1995, CPSC conducted a field investigation study to learn more 
about cigarette-ignited fires and open flame fires. The report, issued 
in 1997, showed that about 70% of the open flame fires involved child 
play and that 68% of the open flame deaths were to children playing 
with lighters, matches, or other open flame sources. The mattress was 
ignited directly by open flame in about 24% of the cases. However, 
bedding was the first item to ignite in about 60% of the cases. In the 
latter scenario, the fire had already developed to a considerable size 
before the mattress became involved. A similar study conducted by the 
National Association of State Fire Marshals (``NASFM'') in 1997 
confirmed these findings.
    A CPSC Chairman's Roundtable, conducted in February 1998, was 
intended to develop approaches to address these fires and fire deaths. 
The Roundtable concluded that technical studies were needed and that a 
public education effort should be considered. The industry's Sleep 
Products Safety Council (``SPSC''), an affiliate of the International 
Sleep Products Association (``ISPA''), sponsored a research program at 
the National Institute of Standards and Technology (``NIST'') to 
provide the technological basis for future performance requirements 
that could be included in a standard for mattresses and/or bedclothes. 
The mattress industry also began developing an expanded public 
education program in cooperation with other interested organizations.
    On March 28, 2000, Whitney Davis, director of the Children's 
Coalition for Fire-Safe Mattresses (``CCFSM'') submitted four petitions 
to the Commission concerning mattress

[[Page 51887]]

flammability. The petitions proposed four options: (1) An open flame 
standard similar to the full-scale test set forth in California 
Technical Bulletin 129; (2) an open flame standard similar to the 
component test set forth in British Standard 5852; (3) a warning label 
for mattresses warning of polyurethane foam fire hazards, and (4) a 
permanent, fire-proof mattress identification tag. The petitions are 
discussed in greater detail in section G.
    The Commission is considering a flammability standard that would 
address mattress fires ignited by small open flames. To be effective 
the standard must reflect the actual use of mattresses. Mattresses 
generally are not used alone, but are covered by bedding or bedclothes, 
such as sheets, blankets and comforters. The presence of these 
materials significantly affects the character of the fire. In some 
incidents the small open flame may ignite the mattress directly. But it 
is more common that the smaller flame source initially ignites the 
bedding, and these materials serve as a larger ignition source for the 
mattress. Thus, an effective standard must consider the interplay 
between the mattress and the bedding.

B. The Product

    According to the International Sleep Products Association 
(``ISPA''), 1999 sales of mattresses were an estimated $2.8 billion. 
ISPA represents about 725 wholesalers, retailers, and manufacturers of 
conventional mattresses and foundations, accounting for over 80% of 
total U.S. sales of these products.
    The expected useful life of mattresses is about 14 years. Based on 
estimates from ISPA and previous staff studies, the Commission's 
Product Population Model estimates about 240 million mattresses may 
have been in use in residential, commercial, and institutional 
applications at the end of 1999.
    The top four producers operate about one-half of the 800 production 
facilities in the U.S. and account for over 50% of the total U.S. 
production of mattresses. The remainder of the production facilities 
are operated by smaller manufacturers that tend to be family-owned 
firms supplying mattresses and foundations to a regional market. While 
renovated mattresses account for as much as 25% of those in use in some 
parts of the country, the total extent of such renovated mattress use 
is unknown. Mattresses produced for institutional and commercial use 
are available to consumers by special order.

C. Risk of Injury

    In 1998, mattress or bedding items were first to ignite in about 
18,100 residential fires that resulted in 390 deaths, 2,160 injuries, 
and $208.3 million in property damage. Over the five-year period from 
1994 through 1998, children under age 15 represented over 75% of the 
deaths in fires ignited by candles, matches, and lighters, and incurred 
over one third of the injuries from these fires. The most common 
ignition sources for the incidents involving deaths of these children 
were candles, matches and lighters. Among victims 15 years of age and 
older, smoking materials were the most common ignition sources causing 
death. In 1998, smoking materials accounted for 5,300 fires, 230 
deaths, 660 injuries, and $61.3 million in property damage.
    Since mattress fires often involve the ignition source of burning 
bedding, initially ignited by a smaller source, a standard 
incorporating an ignition source representing burning bedding could 
address deaths and injuries from fires caused by smoking materials, 
traditional small open flame sources, as well as other heat sources. 
Because few materials can resist such a large ignition source, the 
typical approach of preventing ignition of a mattress through a product 
performance standard is not reasonable. However, limiting the fire 
intensity and preventing flashover in mattress/bedding fires could 
result in a reduction in the number of casualties due to such fires. 
Flashover occurs when a fire becomes so intense that all exposed 
surfaces ignite nearly simultaneously, and quickly spreads through the 
structure. While victims intimate with the ignition may still be at 
risk due to their direct contact with the burning mattresses and 
bedclothes, preventing flashover may reduce the number of casualties to 
a portion of the other victims inside as well as those located outside 
the room of fire origin.

D. Statutory Provisions

    Section 4 of the Flammable Fabrics Act (``FFA'') authorizes the 
Commission to initiate proceedings for a flammability standard when it 
finds that such a standard is ``needed to protect the public against 
unreasonable risk of the occurrence of fire leading to death or 
personal injury, or significant property damage.'' 15 U.S.C. 1193(a). 
That section also sets forth the process by which the Commission can 
issue a flammability standard. The Commission first must issue an 
advance notice of proposed rulemaking (``ANPR'') which: (1) Identifies 
the fabric or product and the nature of the risk associated with the 
fabric or product; (2) summarizes the regulatory alternatives under 
consideration; (3) provides information about existing relevant 
standards and reasons why the Commission does not preliminarily believe 
that these standards are adequate; (4) invites interested persons to 
submit comments concerning the identified risk of injury, regulatory 
alternatives being considered, and other possible alternatives; (5) 
invites submission of an existing standard or portion of a standard as 
a proposed regulation; and (6) invites submission of a statement of 
intention to modify or develop a voluntary standard to address the risk 
of injury. 15 U.S.C. 1193(g).
    If, after reviewing comments and submissions responding to the 
ANPR, the Commission determines to continue the rulemaking proceeding, 
it will issue a notice of proposed rulemaking. This notice must contain 
the text of the proposed rule along with alternatives the Commission 
has considered and a preliminary regulatory analysis. 15 U.S.C. 
1193(i). Before issuing a final rule, the Commission must prepare a 
final regulatory analysis, and it must make certain findings concerning 
any relevant voluntary standard, the relationship of costs and benefits 
of the rule, and the burden imposed by the regulation. Id. 1193(j). The 
Commission also must provide an opportunity for interested persons to 
make an oral presentation before the Commission issues a final rule. 
Id. 1193(d).

E. Existing Open Flame Standards

    The staff has reviewed 13 existing tests or standards relevant to 
open flame hazards associated with mattresses/bedding. State and local 
government tests and standards include Technical Bulletin (``TB'') 129, 
TB 121, and TB 117 from California, the Michigan Roll-up Test, and 
Boston Fire Department (``BFD'') 1X-11 from Boston. The staff reviewed 
other standards from the American Society for Testing and Materials 
standards (ASTM E-1474 and ASTM E-1590), Underwriters Laboratories (UL 
1895 and UL 2060), the National Fire Protection Association (NFPA 264A 
and NFPA 267) and the United Kingdom (British Standard (``BS'') 6807 
and BS 5852).
    Several of these standards specify tests which are duplicates or 
modifications of each other. To simplify the discussion of these 
existing standards, tests are grouped in two broad categories: Full-
scale fire tests of mattresses (sometimes including bedding items) and 
small-scale component tests of mattress materials.

[[Page 51888]]

Important aspects of the standards are briefly summarized below.
    Full-scale Tests: A full-scale test is generally considered the 
most reliable in measuring product performance, especially when the 
product contains multiple materials in a complex construction such as a 
mattress or mattress/bedding combination. Nine of the tests reviewed 
are full-scale burn tests of mattresses that can produce large fires. 
There are only about twelve laboratories in the United States that have 
test facilities capable of safely conducting these tests and properly 
controlling emissions produced. These tests are costly, ranging from 
$2,000-5,000 per test; and CPSC does not have this type of facility.
    TB 129, TB 121, BFD IX-11, ASTM E-1590, NFPA 267, UL 1895, and UL 
2060 use gas burners simulating a newspaper fire in a wastebasket, 
newsprint in a metal container, or burning bedding as the ignition 
source. The mattress is sometimes tested in combination with a 
foundation and bedding. Bedclothes are generally optional and 
unspecified (chosen by the tester). The ignition sources are applied to 
the side or underneath the mattress. The acceptance criteria, when 
specified, are intended to minimize the size/intensity of the fire and 
related hazards rather than prevent ignition. The standards limit the 
peak rate of heat release and/or total heat release, maximum 
temperature above the mattress, carbon monoxide concentration, and mass 
loss.
    BS 6807, a voluntary British standard, provides multiple ignition 
source options for a full-scale test of a mattress or mattress/
foundation combination. The top or underside of the mattress is 
exposed, depending on the specific ignition source. Ignition/non 
ignition is determined from the exposure to a cigarette, butane flame, 
wood crib, or bedclothes chosen by the tester.
    The Michigan Roll-up Test was designed to test jail pads that had 
been rolled up and intentionally ignited by inmates. The pad or 
mattress is rolled and tied, stuffed with newsprint, leaned against a 
bed frame, and ignited. No test criteria are specified.
    Small-scale tests: The staff reviewed four smaller scale standards, 
all of which are used for evaluating mattress components rather than 
the full mattress. One serious drawback of component tests is their 
inability to accurately predict the real life performance of the full 
product, a complex combination of mattress, foundation and bedclothes.
    TB 117 is mandatory in California for polyurethane foam used in 
mattresses. The test requires the average flame spread time of 5 inch 
specimens to be 10 seconds or more.
    ASTM E-1474 and NFPA 264A measure the heat release rate of a small 
specimen of a mattress component material exposed to 35 kilowatts per 
square meter (kW/m\2\) from the burner of a Cone Calorimeter.
    BS 5852 is a British standard, mandatory for mattress filling 
materials (typically foam) used in single-filling mattresses. A 
horizontal/vertical crevice of foam covered with a standard flame-
resistant (FR) polyester fabric is exposed to an ignition source. 
Options include a cigarette, butane flames, and wood cribs of varying 
sizes with increasing thermal outputs. Maximum smoldering/flaming time 
and mass loss are specified.
    Several of these standards, small and large scale, may ultimately 
offer the best choices for a test method, test conditions, magnitude 
and nature of the ignition source, technical rationale, acceptance 
criteria, and so forth. However, more data are necessary to determine 
the most appropriate test. As a group, these standards lack clear links 
to the specific hazard of ignition from burning bedding materials 
typical of residential fire incidents, which is especially important 
for establishing effective acceptance criteria. A better understanding 
of the fire scenario, the magnitude of the hazard to be addressed, the 
contribution of burning bedding, and the effectiveness of product 
changes is needed. With this information, preparation of a reasonable, 
effective performance standard to reduce deaths and injuries is 
possible; and mattress materials and constructions suitable for the 
residential mattress market can be developed.

F. Technical Research and Test Development

    From the CPSC and ISPA/NASFM studies of mattress fire incidents and 
the roundtable discussions, it became clear that a better understanding 
of the problem, desired performance objectives, and technical means to 
meet the objectives were needed. As discussed above, existing standards 
and tests were inadequate and new technical research was needed to 
support and develop an effective test method and standard. In 1998, in 
consultation with CPSC staff, SPSC began sponsoring the necessary 
research at NIST to define and measure the hazard from open flame 
ignition of mattresses from burning bedding. The first phase of the 
research was completed in June 2000, and work on Phase 2 has begun and 
is scheduled for completion later in 2001. CPSC is sponsoring NIST to 
develop a complementary, smaller scale test method to address practical 
issues of enforcement and product development. The small-scale test 
method development will continue into 2002. These programs are 
summarized below.
1. Phase One
    The Flammability Assessment Methodology for Mattresses-Phase 1, 
involved four main objectives: (1) Initial evaluation of bedding 
products, (2) characterization of heat impact on a mattress, (3) design 
of gas burners, and (4) tests of mattresses/bedclothes with burners.
    Because the bedclothes are most likely to be the item first ignited 
and serve as a magnifier for the original, small open flame source, 
NIST characterized the fire behavior of bedclothes typically used in 
residential settings. Tests of twelve combinations of bedclothes 
(sheets, pillows, comforters, and blankets) produced peak heat release 
rates that ranged from 50 kW to about 200 kW; all substantially higher 
than a match or lighter. Peak heat release rate is basically a measure 
of the intensity of the fire produced by these items.
    NIST measured the heat impact imposed on the surface of a mattress 
by six bedding combinations covering a range of performance, from 
moderate to most intense ignition threat. Measurements of heat flux, 
duration and affected location were taken. Distinctly different burning 
conditions existed on the top and side of the mattress, the top being 
more severe.
    NIST then designed two gas burners to consistently simulate the 
typical heat impact imposed on a mattress top and side by burning 
bedding products. This is necessary for providing controlled and 
reproducible test results. The heat flux of the top surface burner is 
65 kW/m\2\ with a duration time of either 45 seconds or 70 seconds. The 
heat flux of the side surface burner is 50 kW/m\2\ with a duration time 
of either 25 seconds or 50 seconds. These measurements were used to 
establish appropriate burner intensities and exposure times when 
applied to the mattress.
    The burners were tested on five different types of mattresses to 
ensure their ability to produce results that correlated with actual 
tests of burning bedding. One mattress represented current residential 
technology. The other four mattresses were constructed with different 
types of potentially fire resisting components, including barrier 
fabrics, modified fibers, and treated foams. Correlation was good 
except for one mattress construction that exhibited internal over-
pressurization with the

[[Page 51889]]

ignited bedding. Internal over-pressurization occurs when a flammable 
gas mixture builds up within the mattress causing rupturing of the 
mattress seams and allowing fire to penetrate the interior. Mattresses 
with this behavior should be avoided or designed to resist rupturing 
during a fire.
    The research conducted during Phase 1 provided extremely useful 
information regarding fires involving mattresses and the interaction 
with bedclothes. Burning bedclothes by themselves were shown to produce 
large fires, reaching heat release rates up to 200 kW. A 200 kW fire is 
a much larger fire than a match, candle or lighter ignition source but 
not large enough to create flashover conditions. Mattresses without 
bedclothes, however, were shown to produce fires large enough to cause 
room flashover, adding to the complexity of the hazard. The gas burners 
appear to successfully simulate most burning bedding conditions and 
show how mattress materials and construction techniques can improve 
mattress fire behavior.
2. Phase Two
    Phase 2 of the NIST/SPSC research will determine the ability of 
small-scale mattresses to predict burning behavior of twin size and 
larger bedding systems. Phase 2 will also provide an analytical basis 
for estimating the performance characteristics of the mattress needed 
to address and reduce the hazard.
    Most available fire test data relate to twin size mattresses. To 
understand the effects of mattress size, it will be necessary to obtain 
data on larger size mattresses. The research will evaluate the effects 
of scale from king size to a 2'  x  2' mini-mattress, a size commonly 
used by manufacturers as a selling tool. If the heat release rate 
behavior or other measure (e.g. weight or mass loss) seen in smaller 
mattresses correlates with that of larger size mattresses, the 
feasibility of conducting safe, convenient mattress tests and producing 
fire safe products increases. Additional tests will evaluate how the 
lateral dimensions of mattresses affect fire intensity and how 
different size mattresses affect a specified room environment.
    Several factors will be considered in order to estimate the peak 
rate of heat release from a mattress that would substantially reduce 
the fire hazard. These include: (1) The effect of bed size and room 
size on fire size, (2) the proximity of other furnishings around the 
bed fire and the ignition threat of surrounding objects, and (3) the 
location of persons with respect to the location of fire origin. Three 
tiers of hazard for victims of mattress/bedding fires have been 
identified using National Fire Incident Reporting System (NFIRS) data: 
(1) Outside the room of origin, (2) within the room of origin but not 
in contact with mattress fire and, (3) direct contact with mattress 
fire. Through analysis of the various tests, NIST will explore the 
relationship between fire size and the number of fatalities and 
determine what reduction in bed fire intensity will significantly 
reduce fatalities based on the three hazard tiers.
    Phase 2 has been expanded to include tests of bedclothes (quilts, 
comforters, pillows) constructed with a variety of flame-resistant 
filling and cover materials to assess the effect of material changes on 
the flammability behavior.
3. Small-scale Screening Test
    To be conducted concurrently with Phase 2, CPSC (with funding 
support from the U.S. Fire Administration) has contracted with NIST to 
develop a bench scale screening test to be used as a surrogate for 
full-scale tests of mattresses exposed to burning bedding or equivalent 
gas burners. Although the most reliable measures of mattress 
performance are full-scale tests, they are expensive and require 
specialized facilities. A bench scale test could be used by CPSC for 
compliance screening and by manufacturers for screening designs/
materials. A similar concept is used in the mattress standard (16 CFR 
part 1632) for substitution of tickings and materials used at the tape 
edge. Test specimens will be from actual production mattresses. Based 
on the performance of a variety of materials, designs, and 
constructions, the test will be designed to be more stringent than the 
full-scale test to avoid problems (such as approving a mattress 
construction that fails the full-scale test and must be recalled 
later).

G. The Petitions

    CCFSM's petitions (Petitions FP 00-1, FP 00-2, FP 00-3, and FP 00-
4) proposed four options to address open flame ignition of mattresses: 
(1) An open flame standard similar to the full-scale test set forth in 
California Technical Bulletin 129; (2) an open flame standard similar 
to the component test set forth in British Standard 5852; (3) a label 
warning of polyurethane foam hazards and (4) a permanent, fire-proof 
mattress identification tag. The petitioner also requested that the 
Commission impose fines and take other actions to enforce the existing 
mattress flammability standard against renovated mattresses. This 
request was not docketed as a petition because it concerned action that 
cannot be taken through rulemaking.
    The petitioner noted that the existing mattress flammability 
standard addresses deaths and injuries associated with cigarette 
ignition of mattresses, not small open flame ignition. The petitioner 
observed that the greater amount of polyurethane foam used in today's 
mattresses provides increased fuel for mattress fires. He argued there 
is a significant need for a standard that would address open flame 
ignition of mattresses.
    In one petition (FP 00-1) the petitioner requested that the 
Commission issue a standard based on a full-scale test like that in 
California TB 129, discussed above. In another petition (FP 00-2) the 
petitioner requested that the Commission issue a standard based on a 
component test like that in BS 5852, discussed above. The Commission 
has determined to grant both of these petitions requesting standards. 
The Commission also voted to deny the remaining two petitions. A label 
warning of the flammability of polyurethane foam may not accurately 
reflect the flammability of a finished mattress, particularly as it may 
actually be used with bedding. As for the requested fire-proof 
identification tag, although it might help identify a mattress after a 
fire, it would not affect a mattress's flammability performance.
    The Commission will consider both full-scale and component tests in 
the course of rulemaking to determine the most effective standard to 
address mattress fires ignited by small open flames. As explained 
above, the Commission staff is involved in extensive research that is 
examining the characteristics of mattress/bedding fires and evaluating 
all relevant tests that could form the basis for a standard.

H. Response to Comments on the Petitions

    On June 12, 2000, the Commission published a request in the Federal 
Register for public comments on these petitions. 65 FR 36890. Nine 
comments were submitted by a fire safety expert and various industry 
associations. Most of these comments were on the general issue of open 
flame ignition of mattresses rather than the specific petition 
recommendations. The major issues raised by the comments and responses 
to them are discussed below.

1. General Comments

    Comment: Some commenters were concerned that standard tests 
recommended by the petitions do not reflect real hazards typical of 
residential mattress fire scenarios. Some stated that

[[Page 51890]]

NIST's work examining mattresses and bedclothes is a preferable basis 
for a standard.
    CPSC Response: Real-life residential bedding fires involve a 
complex system of materials, typically a mattress and foundation with a 
collection of bedclothes which may include any number of sheets, 
blankets, comforters, pillows, quilts and decorative items. The 
bedclothes add to the complexity of the hazard. Often, the mattress is 
essentially exposed to burning bedding, a much larger ignition source 
than the flame from a match, candle or lighter that may have been the 
original source of ignition. Two of the petitions request test methods 
to address the hazard of residential mattress fires (FP00-1 and FP00-
2). The ability of the requested test methods to address real-life 
residential mattress fires is unclear at this time. An appropriate test 
method should effectively address the hazard as it exists in real-life 
fire scenarios, representing all materials present, the typical 
ignition source, and the point of ignition.
    The current study being conducted by NIST is a scientifically based 
research program designed to address the open flame ignition of 
mattresses and bedclothes under conditions that closely resemble real-
life residential fire scenarios. The study is focused on understanding 
the dynamics of fires involving mattresses and bedclothes assemblies 
and on developing appropriate and technologically practicable 
methodology that can effectively address the hazard.
    Comment: Some commenters stated that any new regulation should not 
compromise cigarette resistance. Commenters stated that any new 
regulation should provide a standard with a simple test that can be 
widely used. It should have the attributes of a good standard.
    CPSC Response: The Commission agrees that any new regulation 
regarding mattress flammability should be closely assessed for possible 
impacts on the benefits of the existing regulation. While full-scale 
mattress tests may provide the most definitive measures of mattress 
fire behavior, they are costly, dangerous, and cannot be widely 
conducted. A valid bench or small-scale test that is practical and cost 
effective is a necessary component of a performance standard when many 
tests are needed. A simple bench scale test would enable manufacturers 
to conduct some of their own testing, allowing them to proceed more 
easily with product and design innovation an address safety concerns 
regarding their facilities and employees. A bench scale test that uses 
products obtained at retail would also be useful for regulatory and 
compliance purposes. The Commission agrees that any new standard would 
need to be representative of the real-life fire hazard, and the 
methodology should be reasonable, technologically practicable and based 
on sound comprehensive research.

2. Petition FP 00-1 Suggesting California TB 129

    Comment: One commenter noted that TB 129 provides a direct measure 
of the danger posed by the mattress tested and is excellent for 
assessing product performance. Another commenter, however, observed 
that the type of ignition source and point of ignition used in the test 
are not appropriate for residences. Two commenters noted that TB 129 
tests are expensive and can only be conducted by a fire test laboratory 
with large-scale heat release measurement capabilities.
    CPSC Response: TB 129 was developed to address hazards associated 
with ignition of mattresses in public institions. It is not clear that 
TB 129 provides a test method that is a true and direct measure of the 
danger posed by a typical residential mattress fire. The CPSC staff has 
concerns about the lack of bedclothes and mattress foundations in the 
test, the intensity of the specified ignition source, and the required 
side ignition point. It is also true that full-scale open flame 
mattress tests, like TB 129, require specialized fire test facilities 
and are dangerous and costly to conduct.

3. Petition FP 00-2 Suggesting BS 5852

    Comment: One commenter stated that British Standard 5852 has been 
effective in significantly reducing deaths and injuries from upholstery 
fires.
    CPSC Response: Limited data are available for assessing the 
effectiveness of BS 5852 in reducing deaths and injuries, particularly 
for assessing losses from mattress fires. The UK Department of Trade 
and Technology's report evaluting benefits of the 1988 regulations 
states that the data on mattresses is less clear than the data for 
upholstered furniture. Mattress regulations require the filling 
materials to meet the regulations for polyurethane foams, but do not 
specify fire resistant requirements for mattress fabric coverings or 
tickings. Moreover, the report did not consider variables such as a 
decrease in smoking, increase in consumer awareness, increased use of 
smoke alarms, and increase in use of FR products.
    Comment: One commenter reported on full-scale tests of UK 
mattresses which, mostly ignited by a match, show reduced fire 
intensity. It is not necessary to ensure resistance to burning bedding 
because the British experience using complying foams has been good and 
complying foams do not cause big fires with larger ignition sources.
    CPSC response: Full-scale tests of British mattresses composed of 
treated foam components may exhibit a resistance to small open flames, 
such as matches, lighters and candles when compared to mattresses 
composed of untreated foam. Recent tests, however, show that British 
mattresses are clearly inadequate when presented with the intense 
flames and higher heat fluxes typically caused by burning bedding. 
Several full-scale tests of British mattresses were included in the 
mattress flammability study conducted at the NIST. While the British 
mattresses may take several minutes to reach their peak rate of heat 
release, the peak rate of heat release observed for the mattresses 
alone (without bed clothing) was significantly above the level 
necessary to cause flashover. Testing of mattresses complying with 
British regulations with bed clothing resulted in an even higher peak 
rate of heat release, clearly showing that bedding continues to be a 
major contributor to the fire hazard causing serious flaming of the 
foam.
    Comment: Commenters indicated that BS 5852 is easy to run and 
relatively inexpensive. However, it is a composite test, does not 
assess heat release and does not account for bed clothing in the 
residential fire scenario.
    CPSC response: The Commission agrees that BS 5852 is a relatively 
inexpensive and easy to run test method, but at the same time, is 
concerned that the test does not measure heat release rates or account 
for the more severe ignition source from burning bedding.
    Comment: One commenter suggested that a simple test, like BS 5852, 
that can be used very widely is the most appropriate for a national 
regulation.
    CPSC Response: The Commission agrees that an easy-to-run test is 
appropriate. It is unclear, however, if the most appropriate test is BS 
5852.

4. Petition FP 00-3, Mattress Combustability Warning Labels

    Comment: One commenter noted that Sleep Product Safety Council 
product labels have been used on finished mattresses since 1989. The 
commenter stated that the petition suggests a label that is extreme and 
does not represent the performance of the finished product in a real 
life fire situation.

[[Page 51891]]

    CPSC Response: The Commission agrees that the label recommended by 
the petition does not represent the hazard presented by the finished 
product in a real life fire situation. Polyurethane foam is just one of 
many components used to construct a mattress. Since it is unclear what 
relation the fire behavior of an individual component has to the likely 
fire performance of a completed product, the Commission agrees that the 
suggested warning is not appropriate for the final mattress product.

5. Petition FP 00-4, Fire-proof Mattress Identification Tags

    Comment: One commenter argues that an ID tag would have no impact 
on the propensity of a mattress to ignite or the intensity of the 
resulting fire.
    CPSC response: Petition FP 00-4 requests that all mattresses have 
an identification tag designed to survive a fire permanently attached 
to the innerspring unit. The Commission agrees that such a tag is 
unlikely to have any impact on reducing mattress fires or the 
propensity of a mattress to ignite when exposed to an open flame. Such 
a tag is not visible to consumers to influence their behavior, and the 
tag has no influence on the mattress's ability to resist ignition or 
its performance once ignited. An ID tag could be desirable for 
identifying mattresses involved in fires to improve the utility of 
collected fire data and support further regulatory actions. However, 
the tag cannot be justified in terms of directly reducing death or 
injury from fires.

I. Invitations to Comment

    In accordance with section 4(g) of the FFA, the Commission invites 
comments on this notice. Specifically, the Commission invites the 
following types of comments.
    1. Comments concerning the risk of injury identified in this 
notice, the regulatory alternatives discussed above, and other 
alternatives to address the risk of injury;
    2. An existing standard or portion of a standard as a proposed 
rule;
    3. A statement of intention to modify or develop a voluntary 
standard to address the risk of injury identified in the notice along 
with a description of a plan to modify or develop the standard.
    In addition, the Commission is interested in obtaining further 
information about the following issues.
    1. Materials that could improve mattress performance in open flame 
tests.
    2. Any adverse consequences that an open flame standard might have 
on cigarette ignition of mattresses/bedding.
    3. The appropriate scope of the standard, that is, particular items 
that should be included or excluded.

    Dated: October 4, 2001.
Todd Stevenson,
Consumer Product Safety Commission.

List of Relevant Documents

    1. Briefing memorandum from Margaret Neily, Project Manager, 
Directorate for Engineering Sciences, to the Commission, ``Options 
to Address Open Flame Ignition of Mattress/Bedding and Petitions 
from the Children's Coalition for Fire Safe Mattresses,'' August 16, 
2001.
    2. Memorandum from Signe Hiser, EPHA, to Margaret Neily, 
Engineering Sciences, Residential Fires in Mattresses and Bedding 
1980 ``1998,'' June 11, 2001.
    3. Memorandum from Terrance R. Karels, EC, to Margaret L. Neily, 
ES, ``Mattress Petitions,'' June 15, 2001.
    4. Memorandum from Allyson Tenney, ES, to Margaret Neily, 
Project Manager, ``Current Research Program to Evaluate Open flame 
Mattress Flammability,'' April 25, 2001.
    5. Memorandum from Allyson Tenney, ES, to Margaret Neily, 
Project Manager, ``Response to Comments Received on Petitions FP 00-
1 through FP 00-4, Requesting Standards, Labeling and Identification 
Tags for Mattresses,'' April 25, 2001.
    6. Memorandum from Carolyn Meiers, ESHF, to Margaret Neily, 
Project Manager, ``Petition to Provide Rulemaking Regarding Mattress 
Combustibility Warning Labels,'' March 16, 2001.
[FR Doc. 01-25442 Filed 10-10-01; 8:45 am]
BILLING CODE 6355-01-P