You are on page 1of 46

Thursday,

January 13, 2005

Part II

Consumer Product
Safety Commission
16 CFR Parts 1633 and 1634
Standard for the Flammability (Open
Flame) of Mattresses and Mattress/
Foundation Sets; Standard To Address
Open Flame Ignition of Bedclothes;
Proposed Rules

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\13JAP2.SGM 13JAP2
2470 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

CONSUMER PRODUCT SAFETY Comments to OMB should be directed the fatalities that occur outside the room
COMMISSION to the Desk Officer for the Consumer where the fire originated and about half
Product Safety Commission, Office of of the fatalities that occur within the
16 CFR Part 1633 Information and Regulatory Affairs, room of origin. A mattress that reduces
OMB, Washington, DC 20503. The the likelihood of reaching flashover
Standard for the Flammability (Open Commission asks commenters to could significantly reduce deaths and
Flame) of Mattresses and Mattress/ provide copies of such comments to the injuries associated with bedroom fires.
Foundation Sets; Notice of Proposed Commission’s Office of the Secretary, [1&2] 2
Rulemaking with a caption or cover letter identifying The size of a fire is measured by its
the materials as comments submitted to rate of heat release. A heat release rate
AGENCY: Consumer Product Safety of approximately 1,000 kilowatts
OMB on the proposed collection of
Commission. (‘‘kW’’) leads to flashover in a typical
information requirements for the
ACTION: Notice of proposed rulemaking. proposed mattress standard. room. Tests of twin size mattresses of
FOR FURTHER INFORMATION CONTACT: traditional construction (complying
SUMMARY: The Commission is proposing with the existing mattress standard in
Margaret Neily, Directorate for
a flammability standard under the Engineering Sciences, Consumer 16 CFR 1632) without bedclothes have
authority of the Flammable Fabrics Act Product Safety Commission, measured peak heat release rates that
that would address open flame ignition Washington, DC 20207; telephone (301) exceeded 2,000 kW in less than 5
of mattresses and mattress and 504–7530. minutes. Tests of traditional king size
foundation sets (‘‘mattresses/sets’’). The mattresses measured nearly double that
SUPPLEMENTARY INFORMATION:
Commission currently has a peak rate of heat release. [2]
flammability standard that addresses A. Background Fire modeling and available test data
ignition of mattresses by cigarettes. On October 11, 2001, the Commission show that as a room fire grows, a layer
However, that standard does not address issued an advance notice of proposed of accumulating hot gases and smoke
mattress fires ignited by open flames. rulemaking (‘‘ANPR’’) concerning the thickens downward from the ceiling.
The proposed standard sets performance open flame ignition of mattresses/ For fires exceeding 600 kW, this layer
requirements based on research bedding. 66 FR 51886. The ANPR was typically descends to less than three feet
conducted by the National Institute of the result of the staff’s evaluation of fire from the floor. Heat release rates
Standards and Technology (‘‘NIST’’). data over the course of several years and exceeding 500 kW are generally
Mattresses/sets that comply with the petitions filed by Whitney Davis, considered to pose a serious threat of
proposed requirements will generate a director of the Children’s Coalition for incapacitation and of igniting nearby
smaller size fire, thus reducing the Fire-Safe Mattresses (‘‘CCFSM’’). items. [2]
possibility of flashover occurring. These The objective of the proposed
Although the Commission has an
improved mattresses should result in standard is to limit the size of mattress/
existing mattress flammability standard
significant reductions in deaths and bedding fires to below 1,000 kW for a
that addresses ignition by cigarettes, 16
injuries associated with mattress fires. period of time by reducing the heat
CFR Part 1632, no current Commission
Due to the interaction of mattresses and release from the bed, specifically the
standard directly addresses open flame
bedclothes discussed herein, elsewhere mattress and foundation, and by
ignition of mattresses. The most
in today’s Federal Register the reducing the likelihood that other
common open flame sources are
Commission is publishing an advance objects in the room will become
lighters, candles and matches. The
notice of proposed rulemaking to begin involved in the fire.
Commission is now issuing a notice of Research has shown that the mattress,
rulemaking on bedclothes. proposed rulemaking (‘‘NPR’’) foundation and bedclothes operate as a
DATES: Written comments in response to proposing a flammability standard to system in bedroom fires. Often the first
this document must be received by the address open flame ignition of item ignited is bedclothes, which then
Commission not later than March 29, mattresses.1 ignite the mattress. The gas burners
2005. Comments on elements of the Characteristics of mattress/bedding used in the proposed test method are
proposed rule that, if issued in final fires. A burning mattress generally designed to represent burning
form would constitute collection of provides the biggest fuel load in a bedclothes. Research has indicated that
information requirements under the typical bedroom fire. Once the mattress bedclothes themselves can contribute
Paperwork Reduction Act, may be filed ignites, the fire develops rapidly significantly to fires, even reaching heat
with the Office of Management and creating dangerous flashover conditions. release rates of up to 800 kW. [2&13]
Budget (‘‘OMB’’) and with the Flashover is the point at which the Because of the role of bedclothes in
Commission. Comments will be entire contents of a room are ignited mattress fires, the Commission is
received by OMB until March 14, 2005. simultaneously by radiant heat, making initiating a rulemaking on bedclothes
ADDRESSES: Comments should be filed conditions in the room untenable and through an ANPR that is published
by email to cpsc-os@cpsc.gov. safe exit from the room impossible. At elsewhere in today’s Federal Register.
Comments also may be filed by flashover, room temperatures typically The Commission received numerous
telefacsimile to (301)504–0127 or exceed 600–800 C (approximately 1100– comments on the mattress ANPR
mailed, preferably in five copies, to the 1470 F). In these conditions, carbon concerning the role of bedclothes and
Office of the Secretary, Consumer monoxide rapidly increases, and oxygen the need for a rule addressing them.
Product Safety Commission, is rapidly depleted. Mattress fires that These comments are discussed in
Washington, DC 20207–0001, or lead to flashover are responsible for section J of this document.
delivered to the Office of the Secretary, about two-thirds of all mattress
Consumer Product Safety Commission, fatalities. This accounts for nearly all of 2 Numbers in brackets refer to documents listed

Room 502, 4330 East-West Highway, at the end of this notice. They are available from
1 Commissioner Thomas H. Moore issued a the Commission’s Office of the Secretary (see
Bethesda, Maryland; telephone (301) statement, a copy of which is available from the ‘‘Addresses’’ section above) or from the
504–7530. Comments should be Commission’s Office of the Secretary or from the Commission’s Web site (http://www.cpsc.gov/
captioned ‘‘Mattress NPR.’’ Commission’s Web site, http://www.cpsc.gov. library/foia/foia.html).

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2471

NIST research. The industry’s Sleep Section 4 also sets forth the process upholstered furniture which does not
Products Safety Council (‘‘SPSC’’), an by which the Commission may issue a contain a mattress. However, the
affiliate of the International Sleep flammability standard. As required in Commission could decide to address
Products Association (‘‘ISPA’’), section 4(g), the Commission has issued mattress pads or other top of the bed
sponsored a research program at the an ANPR. 66 FR 51886. 15 U.S.C. items in its rulemaking on bedclothes.
National Institute of Standards and 1193(g). The Commission has reviewed Under the proposed standard, the
Technology (‘‘NIST’’) to better the comments submitted in response to mattress must be tested with its
understand mattress/bedding fires and the ANPR and now is issuing a notice corresponding foundation if the
establish the technological basis for of proposed rulemaking (‘‘NPR’’) mattress and foundation are offered for
future performance requirements of a containing the text of the proposed rule sale as a set. A foundation is a ticking
standard. NIST has conducted extensive along with alternatives the Commission covered structure used to support a
research, which has become the basis has considered and a preliminary mattress.
for California’s open flame mattress regulatory analysis. 15 U.S.C. 1193(i). According to ISPA, the top four
standard (Technical Bulletin or ‘‘TB’’ The Commission will consider producers of mattresses and foundations
603) and for the Commission’s proposed comments provided in response to the account for almost 60 percent of total
standard. NPR and decide whether to issue a final U.S. production. In 2001, there were 639
The NIST research showed that a full- rule along with a final regulatory establishments producing mattresses in
scale test is the most reliable method for analysis. 15 U.S.C. 1193(j). The the U.S. [10]
measuring fire performance of Commission cannot issue a final rule Mattresses and foundations are
mattresses/sets because they contain unless it makes certain findings and typically sold as sets. However, more
many materials in a complex includes these in the regulation. The mattresses are sold annually than
construction. Because the order of Commission must find: (1) If an foundations; some mattresses are sold as
materials, method of assembly, quantity applicable voluntary standard has been replacements for existing mattresses
of materials, and quality of construction, adopted and implemented, that (without a new foundation) or are for
among other factors, can affect fire compliance with the voluntary standard use in platform beds or other beds that
behavior, the complete product may is not likely to adequately reduce the do not require a foundation. ISPA
perform differently in a fire than the risk of injury, or compliance with the estimated that the total number of U.S.
individual components would. Based on voluntary standard is not likely to be conventional mattress shipments was
its research, NIST drafted a full-scale substantial; (2) that benefits expected 21.5 million in 2002, and is estimated
test method for mattresses that uses a from the regulation bear a reasonable to be 22.1 million in 2003 and 22.8
pair of gas burners to represent burning relationship to its costs; and (3) that the million in 2004. These estimates do not
bedclothes as the ignition source. Both regulation imposes the least include futons, crib mattresses, juvenile
the Commission’s proposed standard burdensome alternative that would mattresses, sleep sofa inserts, or hybrid
and California’s TB 603, use this test adequately reduce the risk of injury. 15 water mattresses. These ‘‘non-
method. [1&2] U.S.C. 1193(j)(2). In addition, the conventional’’ sleep surfaces are
Overview of the proposed standard. Commission must find that the standard estimated to comprise about 10 percent
With certain exceptions explained in (1) is needed to adequately protect the of total annual shipments of all sleep
section G below, the proposed standard public against the risk of the occurrence products. The value of mattress and
requires manufacturers to test of fire leading to death, injury or foundation shipments in 2002,
specimens of each of their mattress significant property damage, (2) is according to ISPA, was $3.26 and $1.51
prototypes (designs) before mattresses reasonable, technologically practicable, billion respectively. [10]
based on that prototype may be and appropriate, (3) is limited to fabrics, The expected useful life of mattresses
introduced into commerce. If a mattress related materials or products which can vary substantially, with more
and foundation are offered for sale as a present unreasonable risks, and (4) is expensive models generally
set, the mattress must be tested with the stated in objective terms. 15 U.S.C. experiencing the longest useful lives.
corresponding foundation. The 1193(b). Industry sources recommend
prototype specimens are tested using a replacement of mattresses after 10 to 12
pair of gas burners as the ignition C. The Product years of use, but do not specifically
source. The mattress and corresponding The proposed standard applies to estimate the average life expectancy. In
foundation, if any, must not exceed a mattresses and mattress and foundation the 2001 mattress ANPR, the
200 kW peak heat release rate at any sets (‘‘mattresses/sets’’). Mattress is Commission estimated the expected
time during the 30 minute test, and the defined as a resilient material, used useful life of a mattress at about 14
total energy released must be less than alone or in combination with other years. To estimate the number of
15 megajoules (‘‘MJ’’) for the first 10 materials, enclosed in a ticking and mattresses in use for analysis of the
minutes of the test. The proposed intended or promoted for sleeping upon. proposed rule, the Commission used
standard is discussed in greater detail in This includes adult mattresses, youth both a 10 year and 14 year average
section G of this document. mattresses, crib mattresses (including product life. Using CPSC’s Product
portable crib mattresses), bunk bed Population Model, the Commission
B. Statutory Authority mattresses, futons, flip chairs without a estimates the number of mattresses
This proceeding is conducted permanent back or arms, sleeper chairs, currently in use (i.e., in 2004) to be 233
pursuant to Section 4 of the Flammable and water beds and air mattresses if million mattresses using a ten-year
Fabrics Act (‘‘FFA’’), which authorizes they contain upholstery material average product life, and 302.6 million
the Commission to initiate proceedings between the ticking and the mattress mattresses using a fourteen-year average
for a flammability standard when it core. Mattresses used in items of product life. [8&10]
finds that such a standard is ‘‘needed to upholstered furniture such as According to industry sources, queen
protect the public against unreasonable convertible sofa bed mattresses are also size mattresses are the most commonly
risk of the occurrence of fire leading to included. Not included as mattresses used. In 2002, queen size mattresses
death or personal injury, or significant are: sleeping bags, mattress pads, or were used by 34 percent of U.S.
property damage.’’ 15 U.S.C. 1193(a). other items used on top of the bed, or consumers. Twin and twin XL are used

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2472 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

by 31.2 percent of U.S. consumers, Among the addressable casualties, NIST had previously estimated that a
followed by full and full XL (21 open flame fires accounted for about heat release rate that may cause
percent), king and California king (11 140 deaths (32 percent) and 1,050 flashover for an ordinary sized room is
percent), and all other sizes (2.6 injuries (49 percent) annually. Smoking about 1000 kW. Thus, a mattress that
percent). The average manufacturing fires accounted for 210 deaths (48 contributes more than 500 kW at the
price in 2002 was $152 for a mattress percent) and about 640 injuries (30 same time as bedclothes are
and $86 for a foundation. Thus, the percent) annually. Children younger contributing 400 kW could lead to
average manufacturing price of a than age 15 accounted for an estimated flashover. NIST conducted additional
mattress/foundation set was about $238 120 addressable deaths (27 percent) and tests concerning bedclothes for CPSC,
in 2002. Although there are no readily 500 addressable injuries (23 percent) which are discussed later in this
available data on average retail prices annually. Adults age 65 and older section. [2]
for mattress/foundation sets by size, accounted for an estimated 120 Other objects in the same room. Part
ISPA reports that sets selling under addressable deaths (27 percent) and 250 of the NIST study assessed the potential
$500 represent 40.7 percent of the addressable injuries (12 percent) of a bed fire to ignite other objects in the
market. Sets selling for between $500 annually. [3] same room. Other objects become
and $1000 represent 39.2 percent of the involved by either direct flame
E. NIST Research impingement or by fire generated
market. [10]
The top four manufacturers of Overview. NIST has conducted radiation. Although the location of
mattresses and foundations operate extensive research on mattress/bedding objects in a bedroom is highly variable,
about one-half of the 639 U.S. fires for SPSC and the Commission. their potential involvement is
establishments producing these SPSC sponsored several phases of significantly influenced by their shape
products. The remainder of the research at NIST to gain an and properties relating to ease of
establishments are operated by smaller understanding of the complex fire ignition. NIST concluded from this
firms. According to the Statistics of U.S. scenario involving mattresses and to research that further reducing the heat
Businesses Census Bureau data for 2001, develop an effective test method to release rate from the bed could reduce
there were 557 mattress firms operating evaluate a mattress’s performance when the potential for ignition of other objects
these 639 establishments. According to it is exposed to an open flame ignition and therefore reduce their contribution
the same data source, all but twelve source. The first phase of the research to the overall heat release rate. [2]
mattress firms had less than 500 program, known as Flammability Modeling. NIST used fire modeling to
employees. If one considers a firm with Assessment Methodology for Mattresses, explore the effect that heat and toxic
fewer than 500 employees to be a small involved four main objectives: (1) gases from bed fires can have
business, then 97.8 percent (557–12/ Evaluating the behavior of various throughout a home. Fire modeling is an
557) of all mattress firms are small combinations of bedclothes, (2) analytical tool that uses mathematical
businesses. [9&10] The potential impact characterizing the heat impact imposed calculations to predict real-world fire
of the proposed standard on these small on a mattress by bedclothes, (3) behavior. NIST used this modeling to
businesses is discussed in section M of developing burners to simulate burning corroborate test data exploring the
this document. bedclothes, and (4) testing the burners predicted levels of heat and toxic gases
on different mattress designs to ensure for the room of origin and outside the
D. Risk of Injury room of origin. The modeling suggested
their consistency. NIST’s findings,
Annual estimates of national fires and published in NISTIR 6498, established that untenable fire conditions would
fire losses involving ignition of a the basis for an appropriate test method occur within the room, with little
mattress or bedding are based on data and the next phase of the research difference between a small and large
from the U.S. Fire Administration’s program. [2] 3 room, at 10 minutes and 25 MJ. [2]
National Fire Incident Reporting System Phase 2 of the NIST research focused Gas burners’ correspondence to
(‘‘NFIRS’’) and the National Fire on (1) analyzing the hazard by bedclothes. In addition to the research
Protection Administration’s (‘‘NFPA’’) estimating the peak rate of heat release discussed above, NIST conducted
annual survey of fire departments. The from a mattress with an improved separate studies for CPSC. One series of
most recent national fire loss estimates design, (2) measuring a burning tests evaluated improved mattress
indicated that mattresses and bedding mattress’s ability to involve nearby designs and further supported the
were the first items to ignite in 19,400 items in the room, and (3) assessing (in correlation between full scale mattress
residential fires attended by the fire a limited way) bedclothes and their tests with the NIST gas burners and
service annually during 1995–1999. contribution to mattress fire hazards. actual bedclothes. The study, NISTIR
These fires resulted in 440 deaths, 2,230 This testing used mattresses with 7006–Flammability Test of Full-Scale
injuries and $273.9 million in property improved flammability properties while Mattresses: Gas Burners Versus Burning
loss each year. Of these, the staff the flammability properties of Bedclothes, found that mattress designs
considers an estimated 18,500 fires, 440 bedclothes remained unchanged. [2] showing good performance when tested
deaths, 2,160 injuries, and $259.5 The findings from Phase 2 are detailed with burners also exhibited significantly
million property loss annually to be in NIST Technical Note 1446, improved performance when tested
addressable by the proposed standard. Estimating Reduced Fire Risk Resulting with burning bedclothes. [2]
Addressable means the incidents were from an Improved Mattress Interaction between mattresses and
of a type that would be affected by the Flammability Standard. bedclothes. NIST’s work for CPSC also
proposed standard solely based on the Bedclothes. During phase 2, NIST reinforced observations from previous
characteristics of the fire cause (i.e., a conducted tests on twin and king size NIST research on the interaction
fire that ignited a mattress or that mattresses with corresponding size between the mattress and bedclothes.
ignited bedclothes which in turn ignited bedclothes. In some tests, the bedclothes NISTIR 7006. Tests on improved
the mattress). For example, an incident contributed up to 400 kW to the fire. mattress designs with burning
that involved burning bedclothes and bedclothes as the ignition source tend to
occurred in a laundry room would not 3 NIST publications can be found at NIST’s Web have two distinct heat release rate
be considered addressable. [3] site, (http://fire.nist.gov/bfrlpubs/). peaks. The first peak is predominantly

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2473

from the burning bedclothes, while the box spring sets and futons. TB 603 is crib mattresses, youth mattresses). It
second is predominantly from the expected to go into effect January 1, also refers to a glossary of terms where
mattress and foundation. In tests of good 2005 and applies to items manufactured these items are further defined.
performing mattress designs, NIST for sale in California. The California Specifically excluded from the
found the second peak (i.e., from the standard incorporates the same test definition of mattress are mattress pads,
mattress/foundation) to be comparable methodology as the Commission’s pillows and other top of the mattress
or lower than the first peak (i.e., from proposed standard. Both are based on items, upholstered furniture which does
the bedclothes) and to occur appreciably testing and research conducted at NIST. not contain a mattress, and juvenile or
later. [2] Both TB 603 and the Commission’s other product pads. Mattress pads and
Mattress size. In another study proposed standard require that other top of the bed items may be
conducted by NIST in 2004 for CPSC, mattresses not exceed a 200 kW peak addressed in the Commission’s
NIST found that a twin size mattress heat release rate during the 30 minute rulemaking on bedclothes.
made in a design that yields a very low test. However, the Commission’s Like the Commission’s existing
peak heat release rate (less than 50 kW) proposed standard requires that mattress standard, the proposed
tested with gas burners behaves mattresses not exceed a total heat standard allows an exemption for one-
essentially the same as a queen or king release of 15 MJ in the first ten minutes of-a-kind mattresses and foundations if
size mattress of the same design. of the test, while TB 603 sets test criteria they are manufactured to fulfill a
Mattress designs that yield a moderate of 25 MJ in the first 10 minutes of the physician’s written prescription or
heat release rate peak (greater than 100 test. manufactured in accordance with
kW, but less than 200 kW) tend to The Commission believes that the comparable medical therapeutic
behave the same for the first 30 minutes research NIST has conducted, discussed specifications.
in twin size and king size. After ignition above, establishes the most appropriate 3. Test Method
with the burners, the fire is localized basis for an open flame mattress
(i.e., its spread is limited) and is not The proposed standard uses the full
standard. Several comments on the
affected by the mattress size. [2] scale test method developed by NIST in
ANPR also expressed this view (see
NIST evaluated the same mattress the course of its research. Based on the
section J of this document).
designs and sizes with burning NIST work, the Commission believes
bedclothes. NIST found the mattress G. The Proposed Standard that a full scale test is necessary because
size to have an apparent effect during of the complexities of mattress
1. General
these tests due to the differences in the construction. Testing individual
size and fuel load of bedclothes. In tests The proposed standard sets forth components will not necessarily reveal
of ‘‘well performing’’ mattress designs performance requirements that each the likely fire performance of the
with burning bedclothes, the early heat mattress/set must meet before being complete mattress.
release rate peak when testing a king introduced into commerce. The test Under the proposed standard, the
size mattress was triple that when method is a full scale test based on the specimen (a mattress and corresponding
testing a twin size mattress. This was NIST research discussed above. The foundation if they are to be offered for
driven by the burning bedclothes. mattress specimen (a mattress or sale together as a set) is exposed to a
Mattress designs that showed a mattress and foundation set, usually in pair of T-shaped gas burners. The
moderate heat release rate peak when a twin size) is exposed to a pair of T specimen is to be no smaller than twin
tested with gas burners resulted in more shaped propane burners and allowed to size, unless the largest size mattress or
serious fires when tested with burning burn freely for a period of 30 minutes. set produced of that type is smaller than
bedclothes, especially in king size The burners were designed to represent twin size, in which case the largest size
mattresses. [2] burning bedclothes. Measurements are must be tested.
taken of the heat release rate from the The burners impose a specified local
F. Existing Open Flame Standards specimen and energy generated from the heat flux simultaneously to the top and
In the mattress ANPR the staff fire. The proposed standard establishes side of the mattress/set for a specified
reviewed 13 existing tests or standards two test criteria, both of which the period of time (70 seconds for the top
relevant to open flame hazards mattress/set must meet in order to burner and 50 seconds for the side
associated with mattresses/bedding. comply with the standard: (1) The peak burner). The burners were designed to
These included Technical Bulletin rate of heat release for the mattress/ represent the local heat flux imposed on
(‘‘TB’’) 129, TB 121, and TB 117 from foundation set must not exceed 200 kW a mattress by burning bedclothes. The
California, the Michigan Roll-up Test, at any time during the 30 minute test; heat flux and burner duration were
and Boston Fire Department (‘‘BFD’’) and (2) The total heat release must not derived from data obtained from
1X–11, as well as standards from ASTM exceed 15 MJ for the first 10 minutes of burning a wide range of bedding items.
International (formerly the American the test. As discussed above, NIST test results
Society for Testing and Materials) using the burners have been shown to
2. Scope
(ASTM E–1474 and ASTM E–1590), correlate with results obtained with
Underwriters Laboratories (UL 1895 and The proposed standard applies to bedclothes.
UL 2060), the National Fire Protection mattresses and mattress and foundation The proposed standard allows the test
Association (NFPA 264A and NFPA combinations sold as sets. Mattress is to be conducted either in an open
267) and the United Kingdom (British defined, as it is in the existing mattress calorimeter or test room configuration.
Standard (‘‘BS’’) 6807 and BS 5852). 66 standard at 16 CFR 1632, as ‘‘a resilient Tests have shown that either
FR 51886. material or combination of materials configuration is acceptable. Although
As directed by California Assembly enclosed by a ticking (used alone or in room effects (i.e., the size and
Bill 603, California’s Bureau of Home combination with other products) characteristics of the room) can be a
Furnishings and Thermal Insulation intended or promoted for sleeping factor in mattress flammability
(‘‘CBHF’’) adopted Technical Bulletin upon.’’ The proposed standard lists performance, test data show that room
603 (‘‘TB 603’’), an open flame fire several types of mattresses that are effects do not become an issue until a
standard for mattresses and mattress/ included in this definition (e.g., futons, fire reaches about 300 to 400 kW.

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2474 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

Because the proposed standard limits to the fire. The Commission believes However, as discussed above, after 30
the peak rate of heat release to 200 kW, that the types of ticking (i.e, the minutes, test variability increases, costs
room effects should not be an issue in outermost fabric or material that covers increase, and substantially fewer
the test. Preliminary analysis of data the mattress) currently used on technologically feasible design
from the inter-laboratory study mattresses can continue to be used with approaches are available to meet the
(discussed in section I) does not suggest the 15 MJ/10 minute criteria. [2] This test. Most importantly, it is unclear from
any significant differences between tests will allow manufacturers considerable available data that much additional
based on either test configuration. The flexibility in their mattress designs benefit would accrue with a 60 minute
NIST test method allowed a third test because they should be able to change test.
configuration, essentially a smaller test tickings without affecting the mattresses
5. Prototype Testing
room than described in the proposed performance under the test method,
standard. However, in addition to safety except in the unusual case where the The proposed standard requires, with
concerns, using the burners in the ticking itself is part of the fire resistance certain exceptions, that mattress
smaller size room is awkward. Only one design. manufacturers must test specimens
laboratory in the country uses this California’s TB 603 prescribes a 25 MJ representative of their mattress/set
configuration. Therefore, the limit in the first 10 minutes of the test. prototype (design) before introducing a
Commission decided to propose only However, NIST research, supported by mattress/set into commerce. Mattresses
the two configurations. fire modeling, has shown that untenable then produced based on the prototype
fire conditions can occur in a room from mattress must be identical in all
4. Test Criteria a fire producing 25 MJ in the first 10 material aspects of their components,
The proposed standard establishes minutes of a test. This represents the materials, and method of construction to
two test criteria that the specimen must total contribution from all possibly the prototype. The term ‘‘manufacturer’’
meet to pass the test. The peak rate of involved items. That is, a fire that is defined as ‘‘an individual plant or
heat release must not exceed 200 kW at reaches a size of 25 MJ within 10 factory at which mattresses and/or
any time during the 30 minute test, and minutes could limit a person’s ability to mattress and foundation sets are
the total heat release must not exceed 15 escape the room. According to the manufactured or assembled.’’ The
MJ during the first 10 minutes of the mattress industry and available test definition includes importers. As in the
test. [2&8] data, there are numerous existing mattress standard (16 CFR part
Setting the peak rate of heat release technologically feasible approaches to 1632), this definition refers to the
limit at 200 kW (during the 30 minute mattress designs for meeting the establishment where the mattress is
test) ensures a less flammable mattress, proposed 15 MJ /first 10 minute limit. produced or assembled, not the
reducing the contribution from the [2] company. Thus, the plant or factory
mattress, while taking into account that The 30 minute test duration is related producing or assembling the mattress/
bedclothes and other room contents are to, but not equivalent to, the estimated set is required to conduct prototype
likely to contribute to the fire. time required to permit discovery of the testing. This is also true for importers.
Numerous technologically feasible fire and allow escape under typical fire However, there are three exceptions to
mattress designs are available that can scenarios. A mattress complying with this requirement.
meet the 200 kW criterion. Limiting the the proposed criteria under the 30 A manufacturer is allowed to sell a
peak rate of heat release represents a minute test is estimated to provide an mattress/set based on a prototype that
significant improvement in performance adequate time for discovery of and has not been tested if the prototype
compared to the 16 CFR part 1632 escape from the fire under certain differs from a qualified prototype (one
cigarette ignition standard for mattresses conditions or assuming the bedclothes that has been tested and meets the
and will have the most impact on do not contribute to the extent of posing criteria) only with respect to: (1) The
available escape time. A peak rate of a hazardous condition early in the fire. mattress/foundation size (e.g twin,
heat release lower than 200 kW could Compared to current scenarios, this is a queen, king etc.); (2) the ticking, unless
limit the mattress design approaches substantial increase in estimated escape the qualified ticking has characteristics
that would meet the standard, thus time. The effectiveness of the estimated that are designed to improve the
increasing costs. [2&8] We note that escape time is based on timely escape mattress’s test performance; and/or (3)
California’s TB 603 also prescribes a 200 from the potentially hazardous any other component, material or
kW peak rate of heat release. conditions. [2&3] method of construction, provided that
The proposed standard requires that Multiple test results indicate that a the manufacturer can show, on an
the total heat release in the first 10 large number of mattress designs (using objectively reasonable basis, that such
minutes of the test must not exceed 15 a range of fire retardant barrier change will not cause the prototype to
MJ. This early limit ensures that the technologies) can perform well in tests exceed the specified test criteria. The
mattress will have little involvement in with gas burners for 30 minutes. Many third numbered option allows a
the fire initially and provides a of the tested designs are able to meet the manufacturer to construct and test a
substantial increase in escape time by proposed test criteria for 30 minutes, ‘‘worst case’’ prototype and rely on it to
slowing the rate of fire growth and but perform erratically after 30 minutes. cover a range of related designs without
severity. The mattress’s initial The number of failures, test variability, having to perform additional testing. If
performance is important because if the and performance unreliability increases a manufacturer chooses to take this
mattress becomes significantly involved after 30 minutes. A substantial range of approach, he/she must maintain records
in the early stages of the fire, this will technologically feasible and viable documenting that the change(s) will not
greatly limit the time a person has to solutions and design choices exist that cause the prototype to exceed the test
escape. [2] meet the proposed test criteria for 30 criteria (see § 1633.11(b)(4) of the
The proposed 15 MJ limit in the first minutes. [2] We note that California’s proposed rule).
10 minutes takes into account that TB 603 also includes a 30 minute test When conducting prototype testing,
bedclothes, and possibly other items, duration. the manufacturer must test a minimum
will be burning during this initial The Commission considered of three specimens of the prototype in
period and will contribute significantly proposing a 60 minute test duration. accordance with the test method

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2475

described, and all of the mattresses/sets manufacturer can show (on an 9. Other Requirements: Labeling, One of
must meet both of the test criteria objectively reasonable basis) will not a Kind Exemption, and Policy on
discussed above. If any one prototype cause the prototype to exceed the Renovation of Mattresses
specimen that the manufacturer tests specified test criteria. Under the proposed standard, each
fails the specified criteria, the prototype mattress/set must bear a permanent
is not qualified (even if the 7. Quality Assurance Requirements
label stating the name and location of
manufacturer chooses to test more than the manufacturer, the month and year of
Research and testing indicates that
three specimens).
small variations in construction of a manufacture, the model identification,
The Commission believes that three
mattress/set (e.g. missed stitching prototype identification number, and a
specimens is the appropriate minimum
number for testing. Numerous research around the side of the mattress) can certification that the mattress complies
studies have typically used replicates of affect the fire performance of a mattress. with the standard. By placing the
three for tests using the developed gas Therefore, the proposed standard certification on the mattress, the
burners. This is also the number contains strict requirements for quality manufacturer is attesting that the
industry has generally used as it has assurance. Each manufacturer must specific mattress would comply with
researched and developed options for implement a quality assurance program the test criteria if tested.
meeting the requirements of California’s to ensure that the mattresses/sets it The proposed standard allows an
TB 603. Preliminary analysis of the exemption for a one-of-a-kind mattress/
produces are identical in all material
inter-laboratory study also indicates that set if it is manufactured in response to
respects to the prototype on which they
three replicates are appropriate to a physician’s written prescription or
are based. This means that at a
accurately characterize mattress manufactured in accordance with
minimum, manufacturers must: (1) Have comparable medical therapeutic
performance. [2] Moreover, because controls in place on components and
small changes in mattresses’ specifications.
materials to ensure that they are Subpart C of the proposed standard
construction or components can affect identical to those used in the prototype;
their flammability, testing more than restates the policy clarification on
(2) designate a production lot that is renovation of mattresses that is in
one mattress will provide a better
represented by the prototype; and (3) Subpart C of the existing mattress
indication of their performance. [1]
inspect mattresses/sets produced for standard (16 CFR Part 1632). The policy
6. Pooling sale. The Commission is not requiring statement informs the public that
The proposed standard allows for one manufacturers to conduct testing of mattresses renovated for sale are
or more manufacturers to rely on a given production mattresses. However, the considered by the Commission to be
prototype. Under this approach, one Commission recognizes the value of newly manufactured for purposes of the
manufacturer would conduct (or cause such testing as part of a quality requirements of the proposed standard.
to be conducted) the full prototype assurance program. Therefore, the H. Effectiveness Evaluation
testing required (testing three prototype Commission encourages manufacturers
specimens), obtaining passing results, to conduct random testing of mattresses/ To determine the potential
and the other manufacturer(s) may then sets that are produced for sale. effectiveness of the proposed standard,
produce mattresses/sets represented by CPSC staff conducted an effectiveness
that prototype so long as they conduct 8. Recordkeeping evaluation, focusing primarily on
one confirming test on a specimen they reduction of deaths and injuries. The
The proposed standard requires staff’s analysis is explained in detail in
produce. If the mattress/set fails the
manufacturers to maintain certain the memorandum ‘‘Residential Fires
confirming test, the manufacturer must
take corrective measures, and then records to document compliance with Involving Mattresses and Bedding.’’ [3]
perform a new confirmation test that the standard. This includes records The evaluation was based primarily on
must meet the test criteria. If a concerning prototype testing, pooling review of CPSC investigation reports
confirmation test specimen fails to meet and confirmation testing, and quality that provided details of the occupants’
the test criteria, the manufacturer of that assurance procedures and any situations and actions during the fire.
specimen must also notify the associated testing. The required records Staff reviewers identified criteria that
manufacturer of the pooled prototype must be maintained for as long as affected the occupants’ ability to escape
about the test failure. Pooling may be mattresses/sets based on the prototype the fires they had experienced. The staff
used by two or more plants within the are in production and must be retained used these criteria to estimate
same firm or by two or more for three years thereafter. percentage reductions in deaths and
independent firms. As discussed in the The purpose of these recordkeeping injuries expected to occur under the
initial regulatory flexibility analysis, requirements is to enable manufacturers much less severe fire conditions
pooling should reduce testing costs for anticipated with improved designs of
to keep track of materials, construction
smaller companies. Once they have mattresses that would comply with the
methods and testing. Thus, if a
conducted a successful confirmation proposed standard. The staff then
test, pooling firms can produce manufacturer produced a mattress/set
applied these estimated reductions to
mattresses based on a pooled prototype that failed to meet the test criteria, he/ national estimates of mattress/bedding
and may continue to do as long as any she should be able to use the records to fire deaths and injuries to estimate
changes to the mattresses based on the determine the prototype on which the numbers of deaths and injuries that
pooled prototype are limited to the three failing mattress was based, as well as could be prevented with the proposed
discussed above: (1) Size of the the components and method of standard. [3]
mattress/foundation; (2) the ticking, construction that were used. This As stated in section D of this
unless the qualified ticking has information would help the document, the most recent national fire
characteristics that are designed to manufacturer correct the problem that loss estimates indicated that mattresses
improve the mattress’s test performance, caused the mattress to fail the test and bedding were the first items to
and/or (3) any component, material or criteria. ignite in 19,400 residential fires
method of construction that the attended by the fire service annually

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2476 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

during 1995—1999. These fires resulted proposed requirements if the test ended The proposed standard is designed to
in 440 deaths, 2,230 injuries and $273.9 at 30 minutes, but appeared to perform address the identified hazard of
million in property loss each year. Of erratically after 30 minutes. [2] flashover resulting from open flame
these, the staff considers an estimated The preliminary analysis, supported ignition of mattresses, usually from
18,500 fires, 440 deaths, 2,160 injuries, by earlier data, suggests that significant burning bedclothes. Under the proposed
and $259.5 million property loss variability exists among currently standard, mattresses and mattress/sets
annually to be addressable by the available mattress designs. Although are exposed to gas burners, simulating
proposed standard (i.e., of the type that products appear to be moving toward burning bedclothes. Mattresses are
the proposed standard could affect consistency, manufacturers clearly need required to meet two performance
based on the characteristics of the fire). to control components, materials, and criteria that minimize the possibility of
[3] methods of construction. Thus, quality or delay flashover for a period of time.
Overall, CPSC staff estimates that the assurance measures, as required in the Mattresses must not exceed 200 kW
proposed standard may be expected to proposed rule, are important. [2] peak heat release rate during the 30
prevent 80 to 86 percent of the deaths The inter-lab study was only recently minute test, and the total heat released
and 86 to 92 percent of the injuries completed, and the discussion above is must be less than 15 MJ for the first 10
presently occurring in addressable based on the staff’s preliminary analysis minutes of the test.
mattress/bedding fires attended by the of the results. A final report on the inter- 2. Comment. Most commenters
fire service. Applying these percentage lab study is expected by the end of 2004 endorsed the direction of the mattress
reductions to 1998–2002 estimates of and will be available to the public. flammability test development research
addressable mattress/bedding fire J. Response to Comments On the ANPR underway at NIST and encouraged the
losses, staff estimates potential CPSC to issue a technologically
reductions of 310 to 330 deaths and On October 11, 2001, the Commission practicable, reasonable standard. More
1,660 to 1,780 injuries annually in fires published an ANPR in the Federal recent commenters suggest California
attended by the fire service when all Register. 66 FR 51886. During the TB 603 be adopted as the federal
existing mattresses have been replaced comment period, the Commission standard.
with mattresses meeting the new received sixteen written comments from Response. CPSC agrees with the
standard. There may also be reductions businesses, associations and interested technical approach suggested by the
in property damage resulting from the parties representing various segments of NIST research. A majority of the
proposed standard, but data are not the mattress and bedding industries. commenters agreed that preventing
sufficient for the staff to quantify this After the close of the comment period, flashover from mattress fires would
impact. [3] the Commission received a number of appropriately address the risk and that
additional comments, including one a full scale test with an ignition source
I. Inter-Laboratory Study from the California Bureau of Home comparable to burning bedclothes could
An inter-laboratory study was Furnishings and Thermal Insulation achieve that objective. They strongly
conducted with the support of the urging the Commission to adopt supported the NIST approach and
SPSC, NIST, and participating California’s TB 603 as a federal discouraged the adoption of any existing
laboratories to explore the sensitivity, standard. Significant issues raised by all standards.
repeatability, and reproducibility of the of these comments are discussed below. Before California’s adoption of TB
NIST test method. All of the [14&15] 603, one commenter suggested using a
participating labs conducted multiple modification of the small-scale British
tests of eight different mattress designs. Mattress Comments test, BS 5852, for smoldering and
The mattress designs varied critical 1. Comment. Commenters agree that flaming ignition of upholstered
elements (e.g., the barrier—sheet or the hazards associated with mattress furniture seating composites. However,
high-loft, the type of mattress—single or fires appear to be clearly identified. All a full-scale rather than small-scale test
double-sided) and the style of mattress of the commenters support the need for is generally considered the most reliable
(e.g., tight or pillow top). [2] an open flame standard for mattresses method for measuring performance of a
Preliminary analysis of the data does and initiation of federal rulemaking. product that contains many materials in
not suggest either unreasonable Response. CPSC agrees that mattress a complex construction, such as a
sensitivities (i.e., significantly different and bedding fires continue to be one of mattress. NIST research confirmed that
test results when minor variations in the major contributors to residential fire a full-scale test of the mattress was
test procedure are made) or practical deaths and civilian injuries among needed to measure its performance
limitations in the test protocol. The products within CPSC’s jurisdiction. when exposed to burning bedclothes or
preliminary analysis suggests that some The most recent national fire loss the representative set of gas burners.
mattress designs perform more estimates indicate that mattresses and NIST’s comprehensive, scientifically
consistently than others. The type of bedding were the first items to ignite in based research program was designed to
barrier appears to have a significant 19,400 residential fires attended by the address the open-flame ignition of
impact on the performance and fire service annually during 1995— mattresses and bedclothes under
repeatability of performance of all 1999. These fires resulted in an controlled conditions closely
mattress designs tested. However, the estimated 440 deaths, 2,230 injuries, resembling those of real-life fire
uniformity of other components and the and $273.9 million property loss scenarios. The program focused on
manufacturing process can also affect annually. In these fires, the bedclothes understanding the dynamics of fires
the variability in fire performance. [2] are most frequently ignited by a small involving mattress and bedclothing
The inter-lab tests also appear to open flame source. The burning bedding assemblies and on developing an
confirm earlier observations that then creates a large open-flame source appropriate and technologically
mattresses constructed with currently igniting the mattress and creating practicable methodology to effectively
available barrier technologies are able to dangerous flashover conditions, the measure the hazard.
limit the fire severity for a substantial point when the entire room and its NIST subsequently prepared a test
but not indefinite time. Most of the contents are ignited simultaneously by method which the state of California
tested mattress designs could meet the radiant heat. incorporated into their TB 603,

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2477

‘‘Requirements and Test Procedure for reproducibility of the test protocol. applications in many consumer
Resistance of a Mattress/Box Spring Set While the final report is not yet products. There are also flame resistant
to a Large Open-Flame’’ in 2004. The available, preliminary analysis of the (FR) materials that may be used for
proposed standard is also based on the data does not suggest either mattress barriers that have other
test method developed by NIST. unreasonable sensitivities or practical consumer product applications. For
Research on mattress and bedclothes limitations in the test protocol. example, melamine resins, which can be
fires conducted by NIST for CPSC and The Commission’s proposed standard used in FR barriers, are also used in
the industry provides the basis for the is similar to California’s TB 603. The many laminated counter tops.
test criteria specified in the proposed proposed standard and TB 603 use the Based on available data, the
standard. Manufacturers and suppliers same test method and limit the peak rate Commission believes that there are
have demonstrated that mattress designs of heat release of a mattress or mattress/ available options for meeting the
complying with these performance foundation to 200 kW. TB 603 also standard without posing an
criteria and suitable for the residential limits the size of the fire produced in unacceptable health risk to consumers
market can be produced. the first 10 minutes of the test to 25 MJ. or significantly affecting the
3. Comment. One commenter According to NIST research, untenable environment. Moreover, as described in
requested the exclusion of certain fire conditions could occur in a room section N of this preamble, even if a
product categories, such as mattresses from a fire of this size. Unlike TB 603, method used by some manufacturers to
used for therapeutic reasons and in the staff’s draft proposed standard meet the standard were suspected of
healthcare environments, from an open requires that the mattress contribute no posing an unacceptable risk, there
flame standard. more than 15 MJ to the early fire would be regulatory and other
Response. The proposed standard scenario. This ensures that the mattress mechanisms that can be used to control
includes all mattresses, including those will have little involvement in the fire that particular method. The staff is
used in or as part of upholstered for the specified period of time. This planning to conduct migration and
furniture items. ‘‘One-of-a-kind’’ lower limit partially compensates for exposure studies on various FR
mattresses and foundations are defined the contribution of an uncertain chemicals that could be used to meet
as physician prescribed mattresses to be combination of burning bedclothes on the standard.
used in connection with the treatment the bed, helping to preserve tenable The commenter suggested labeling of
or management of a named individual’s conditions for egress. chemically treated components as a
physical illness or injury. These 5. Comment. Two commenters possible requirement of the standard, to
products may be exempted from testing recognize the sophistication and inform consumers of the materials used.
under the proposed standard in complexity of the test method used in The Commission questions whether
accordance with the rules established by California TB 603 and potentially in a such information would be of practical
the Commission. The proposed standard federal standard. They suggest that value to consumers. Simply stating that
requires them to be permanently labeled CPSC explore laboratory accreditation a mattress component has been
with a warning statement indicating that programs to insure test labs are properly chemically treated does not indicate to
the mattress and foundation have not qualified to conduct this complex test. the consumer whether the mattress
been tested under the standard and may Response. The interlaboratory study poses any health risk or not. The
be subject to a large fire if exposed to may identify laboratory practices, proposed standard requires
an open flame. equipment, and other related factors manufacturers to maintain records
4. Comment. In October 2003 the that must be controlled to ensure specifying details of all materials,
California Bureau of Home Furnishings consistent and accurate test results. The including flame retardant treatments
(CBHF) urged the Commission to adopt report and findings of the study will be applied and inherently flame resistant
their new standard, TB 603. available to the public; and appropriate materials, used in each mattress design
Subsequently, a number of commenters guidance can be provided to interested (prototype). This will allow
expressed written support for adopting laboratories. While accrediting test identification of relevant mattresses and
the TB 603 test methodology and laboratories is not a CPSC function, the mattress/sets if an unacceptable risk is
performance criteria. Commission supports industry and identified.
CBHF claimed that harmonization of commercial laboratory development of 7. Comment. Another commenter
California and federal standards would such a program. recommended test provisions in the
avoid a number of potential problems. 6. Comment. A commenter expressed standard that address the long term
They noted potential problems such as concerns about environmental impact durability of the flame retardant
possible federal preemption and and consumer sensitivity to flame chemicals used in mattresses to ensure
negative impacts on interstate retardants that may be used in they continue to meet the fire
commerce. Since TB 603 is a newly mattresses, whether topically applied or performance requirements.
developed methodology, CBHF integrated into fibers. The commenter Response. It is expected that most
suggested that an inter-laboratory study recommends requiring a label that manufacturers will use some kind of
be conducted before a potential discloses the use of flame retardants in flame resistant barrier material to
adoption of TB 603 by CPSC. They the mattress and provides a source of protect the mattress components with
noted that data obtained from an inter- more specific information. the greatest combustible fuel load from
laboratory study would verify the Response. Mattress fire performance exposure to an open flame. Flame
credibility of the test method. can be improved by incorporating fire resistant barriers for mattresses may take
Response. An inter-laboratory study retardant chemicals into component several forms, including ticking fabrics,
was conducted with the support of materials or by using materials that are woven and non-woven interlinings, and
SPSC, NIST, CBHF, and other inherently fire resistant. Flame retardant battings. It is likely that these barriers
participating laboratories to collect chemicals are already widely used in will be made with an inherently flame
additional data and confirm the test other applications. More than one resistant fiber (e.g., para-aramid or
protocol developed by NIST. A number billion pounds of different flame fiberglass) or by treatment with flame
of laboratories participated in the study retardant chemicals are currently used retardant chemicals, many of which are
to evaluate sensitivity, repeatability, and annually in the United States, including incorporated within the fiber, foam, or

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2478 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

other material. At this point in the objective of the standard can be met in the occupancy. Even with more time,
development of technologies that may with the performance criteria specified: in such circumstances, the fire
be used to meet TB 603 or the proposed maximum 200 kW peak heat release rate continues to progress, and the chances
standard, the staff has seen no evidence during the 30-minute test and maximum of rescue are unpredictable.
that suggests that changes in these 15 MJ total heat release in the first 10 Based on the preliminary regulatory
materials over time will occur or affect minutes of the test. Laboratory tests of analysis, the expected benefits of the
fire performance. currently marketed crib mattresses of proposed standard, incorporating a 30-
8. Comment. One commenter which the Commission is aware show minute test, are greater than the costs.
expressed concerns about the unacceptable performance in one or The regulatory analysis also considered
potentially severe economic impact of a both of these fire performance measures. alternatives to the proposed standard,
federal regulation, similar to TB 603, on Like full-size mattresses, these crib including a 60-minute test; neither this
small businesses. mattresses would also need to be nor the other alternatives was shown to
Response. The Commission improved to meet the requirements of increase expected net benefits.
acknowledges that the cost of testing, the proposed standard. 11. Comment. A few commenters
record keeping, and quality control/ 10. Comment. One commenter expressed the need to maintain
quality assurance programs could be suggested that a 60-minute test duration protection from the threat from cigarette
disproportionately higher for small is needed in the standard to allow for ignitions while considering an open
businesses. While these costs are fire and rescue workers to respond and flame standard.
estimated to be a little over one dollar help occupants escape. Response. The standard that
per mattress per year for average-sized The commenter notes that the longer
addresses cigarette ignition resistance,
establishments, they could be test time will allow emergency
the Standard for the Flammability of
substantially higher for some small responders to assist vulnerable citizens
Mattresses and Mattress Pads, codified
mattress producers. The proposed to escape fires involving mattresses and
as 16 CFR part 1632, remains in effect
standard, however, allows bedding. They report that response
manufacturers to pool their prototype times can vary widely among local unless it is modified or revoked by the
qualification and testing, and thus these circumstances, from approximately 16 Commission in a separate rulemaking
costs can be mitigated. Moreover, if minutes to an hour or more. proceeding. If such a rulemaking
manufacturers produce mattress/set Response. To estimate the proposed occurred, the Commission would
constructions for longer than a year or standard’s potential effectiveness, the thoroughly evaluate the need for
use a worst-case prototype to represent staff reviewed in-depth investigations maintaining both an open flame
other mattress constructions, these costs that provided detailed information standard and the standard for cigarette
will be lower. It is also expected that about fires that ignited mattresses and ignition resistance.
some barrier suppliers or independent bedding, details of the occupants’ Bedclothes Comments
laboratories would be willing to do the situation, and occupants’ actions during
testing and quality control/assurance the fire. Most investigations also Comment. Most of the commenters
programs for small producers in included documentation from the fire refer to the impact of burning
exchange for a small charge. Therefore, department that attended the fire. The bedclothes on mattress/bedding fires
the proposed standard is expected to in-depth investigations involved fires and express opinions on the potential
minimize the impact on small occurring during 1999–2004, and scope of an open flame mattress
businesses, while maintaining the included a total of 195 deaths and 205 standard. Some commenters urge the
benefits resulting from the standard. injuries. In some of these cases, even Commission to limit the scope of a
The Commission is requesting with traditional mattresses and bedding, standard to mattresses while opposing
comments from small businesses on the other members of the household present commenters recommend that either the
expected economic impact of the at the time of the fire and emergency scope be expanded to incorporate
requirements of the proposed standard responders arriving within as little as 5 bedclothes or bedclothes should have
and the proposed effective date of 12 minutes were able to rescue victims. ignition standards of their own.
months after publication of the final With improved mattresses, those Commenters in support of regulating
rule in the Federal Register. complying with the 30-minute test bedclothes believe that studying the
9. Comment. One commenter reported specified in the proposed standard, the impact of burning bedclothes is
that some juvenile or crib mattresses, fire growth is slowed considerably and appropriate and would assist in the
while meeting the 200 kW peak rate of flashover conditions are delayed, development of better performing, safer
heat release requirement, produce large making successful rescue efforts of products. They note that bedclothes
amounts of flaming droplets that have family members and emergency contribute to the intensity and spread of
the potential for spreading flames responders more likely. The the original ignition source often
beyond the mattress. TB 603 does not Commission estimates that 310 to 330 involved in mattress fires. Therefore,
address these flaming droplets. deaths and 1,660 to 1,780 injuries burning bedclothes become a significant
Response. The objective of the resulting from mattress and bedding ignition source to the mattress and
proposed standard is to reduce the size fires could be prevented annually by the impact the burning characteristics of the
of mattress/bedding fires and, thereby proposed standard. A maximum mattress and foundation. They further
reduce the likelihood of or delay the additional 80 deaths and 280 injuries, note that bedclothes alone have been
development of flashover conditions in considered addressable by the draft shown to generate a fire large enough to
the room. Based on research conducted standard, might be further reduced with pose a hazard and can alone be the
by NIST, performance criteria were a 60-minute test. However, actual cause of ignition to nearby items.
developed to limit the size of the reductions would likely be much lower. According to these commenters,
mattress fire and reduce the likelihood This is because those considered likely improving the flammability of certain
of it involving other objects in the room. to die or be injured in conditions bedding items, such as filled items, is
The Commission believes that, while associated with a proposed 30 minute economically feasible. One commenter
the proposed standard may be less test are those incapable of acting on claims that mattress fires cannot be
effective in isolated circumstances, the their own and with no potential rescuer adequately addressed without also

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2479

considering the flammability of Ignitions of Mattresses,’’ addresses this A typical innerspring mattress
bedclothes. requirement. [8] construction might include ticking;
In support of limiting the scope to binding tape fabric; quilt cushioning
mattresses and not regulating 1. Introduction
with one or more separate layers; quilt
bedclothes, some commenters identify There were an estimated 18,900 fires backing fabric; thread; cushioning with
bedding items as an uncontrolled where the first item ignited was one or more separate layers; flanging;
variable. They claim that there is no mattress/bedding in 1998 (the last year spring insulator pad; spring unit; and
way to predict the type of bedclothes for which detailed data comparable to side (border) panels. Options for
that may be involved in an incident at previous years are available). These fires meeting the standard include the use of
any given time; the number and type of caused an estimated 2,260 civilian one or a combination of the following:
items used by consumers is indefinable injuries, 410 deaths, and $255.4 million fire resistant ticking; chemically treated
and consumers select items based on in property losses. As discussed or otherwise fire resistant filling
season, fashion, and climate. In elsewhere in this document, NIST has products; or a fire blocking barrier
addition, according to these conducted extensive research and (either a sheet style barrier, sometimes
commenters, there is no objective developed a test methodology to test called a fabric barrier, or a high-loft
method to determine if consumers open flame ignition of mattresses. barrier, sometimes called a fiber barrier).
would use regulated bedclothes; there is California Technical Bulletin (TB) The fire blocking barrier is placed either
little data to suggest that regulating 603, which is based on the use of NIST directly between the exterior cover
some selected items will have an impact test burners designed to mimic the local fabric of the product and the first layer
on the hazard; and flammability thermal insult (heat flux levels and of cushioning materials, or beneath one
performance should not be based on duration) imposed by burning or more ‘‘sacrificial’’ layers that can
what consumers may or may not use as bedclothes, is scheduled to become burn without reaching the proposed
bedclothes. These commenters also state effective in California January 1, 2005. heat release constraints.
that most U.S. textile manufacturers TB 603 requires all mattress/foundation There are already over twenty
already voluntarily test for small open sets, mattresses intended to be used different vendors of fire resistant
flame ignition of bedclothes using without a foundation, and futons to materials associated with the
ASTM voluntary test methods. They meet the following pass/fail criteria: (1) production of mattresses, including
assert that the additional burden and The peak heat release rate (‘‘PHRR’’) barriers, ticking, foam, tape, and thread.
expense of any regulation on bedclothes does not exceed 200 kW during the 30 These materials include chemically
would be substantial and could not be minute test, and (2) the total heat release treated cotton, rayon, and/or polyester,
justified. does not exceed 25 mega joules (MJ) in melamine, modacrylic, fiberglass,
Response. The Commission notes that the first 10 minutes of the test. aramid (Kevlar’’), or some combination
bedclothes substantially contribute to Large mattress manufacturers may of them. The cost of using sheet barriers
the complexity and magnitude of the eventually produce TB 603-compliant is higher than using high-loft barriers,
mattress fire hazard. In laboratory tests mattresses for sale nationwide, because since sheet barriers are thin and
peak heat release rates as high as 800 of legal liability and production therefore could not be substituted for an
kW were observed from some larger logistics. In the short-run, however, existing foam or cushioning layer. There
bedclothes items. This presents a clear some manufacturers may limit their sale is also concern that some sheet barriers,
risk of flashover; and this heat release of TB 603-complying mattresses to unlike high-loft barriers, may reduce the
rate is much higher than that allowed California. Sealy’s president and CEO comfort of the sleeping surface.
for a mattress/set in the proposed said that ‘‘[they] plan to be ready by the To qualify a prototype, three
standard. The extent to which end of this year [2004] if a national mattresses/sets must be tested and must
bedclothes can be modified in a manner retailer wants the same product’’ with pass the test requirements. To obtain a
that is technologically practicable and fire resistant technology, but will not passing result, each mattress/set must
economically feasible is unclear at this convert all production by January 2005 pass a 30 minute test, where the PHRR
time. However, reducing the (Furniture Today, March, 10, 2004). does not exceed 200 kW and the total
contribution of certain high fuel load Smaller producers are more likely to heat release does not exceed 15 MJ in
bedding items to a mattress/bedding fire wait until they have a better idea of the first 10 minutes of the test. If any of
is desirable. The Commission is issuing enforcement efforts in California, or the sets fail, the problem must be
an ANPR for a bedclothes flammability until a federal standard is adopted. The corrected, the prototype must be
standard. The Commission believes that mattress industry, represented by ISPA, retested and pass the test (in triplicate).
such a standard could increase the supports the development of a Manufacturers may sell any mattress/set
likelihood that mattress/bedding fire mandatory federal standard (Furniture based on a qualified prototype.
losses are effectively reduced. Today, May, 10, 2004). A Federal Manufacturers may also sell a mattress/
standard would eliminate the set based on a prototype that has not
K. Preliminary Regulatory Analysis been tested if that prototype differs from
uncertainty that may result from having
The Commission has preliminarily different flammability standards for a qualified prototype only with respect
determined to issue a rule establishing different states. to (1) mattress/foundation size; (2)
a flammability standard addressing the ticking, unless the ticking of the
open flame ignition of mattresses. 2. The Proposed Standard: Scope and qualified prototype has characteristics
Section 4(i) of the FFA requires that the Testing Provisions designed to improve performance on the
Commission prepare a preliminary The proposed standard will apply to burn test; and/or (3) any component,
regulatory analysis for this action and all mattresses, where the term mattress material, or method of construction that
that it be published with the proposed means a ticking (i.e., an outer layer of the manufacturer can demonstrate, on
rule. 15 U.S.C. 1193(i). The following fabric) filled with a resilient material an objectively reasonable basis, will not
discussion, extracted from the staff’s used alone or in combination with other cause the prototype to exceed the test
memorandum titled ‘‘Preliminary products intended or promoted for criteria specified above.
Regulatory Analysis of a Draft Proposed sleeping upon. This definition is If one or more establishments (plants
Standard to Address Open-Flame discussed further in section G.2. above. within the same firm) or independent

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2480 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

firms choose to ‘‘pool’’ prototypes, then Manufacturers of bedclothes may also 4. Characteristics of Mattresses Used in
each pooling plant or firm is required to be affected by the proposed standard. U.S. Households
test one mattress/set for confirmation Sales of bedclothes may increase or
testing. If that set fails, then the plant or decrease based on whether consumers The total number of U.S. conventional
firm will need to test another mattress/ view bedclothes as complements or mattress shipments was 21.5 million in
set after correcting its production to substitutes for a new mattress/set 2002 and is estimated to be 22.1 in 2003
make sure that it is identical to the (complements are goods generally and 22.8 in 2004. Mattress shipments
original prototype. consumed together, substitutes have grown at an average rate of three
A pooling firm may sell other generally substitute for each other). For percent over the period 1981 to 2004.
mattresses that have not been tested by example, if people tend to buy all parts Unconventional mattresses (including
the pooling firm if they differ from the of a new bed (mattress, foundation, and futons; crib mattresses; juvenile
pooled prototype only with respect to bedclothes consisting of a comforter, mattresses; sleep sofa inserts; and
(1) mattress/set size; (2) ticking, unless pillows, and sheets) at the same time, hybrid water mattresses) are estimated
the ticking of the qualified prototype then an increase in the quantity of to be about ten percent of the total
has characteristics designed to improve mattresses sold would cause an increase market. This yields an estimated total
performance on the burn test; and/or (3) in sales of bedclothes. If, alternatively, number of mattresses produced
any component, material, or method of people tend to have a fixed budget from domestically of 25.3 million in 2004.
construction that the manufacturer can which to buy all mattresses and bedding The value of mattress and foundation
demonstrate, on an objectively items, then an increase in the quantity shipments in 2002, according to ISPA,
reasonable basis, will not cause the of mattresses sold would lead to a was $3.26 and $1.51 billion
prototype to exceed the test criteria decrease in sales of bedclothes. Also, if respectively.
specified above. the decision to buy a new mattress (or The CPSC Product Population Model
mattress/set) involves buying a mattress (PPM) estimate of the number of
3. Products and Industries Potentially that is much thicker than the one
Affected mattresses in use in different years is
currently in use, then consumers will based on available annual sales data and
According to ISPA, the mattress most likely buy new sheets (and an estimate of the average product life
producers’ trade organization, the top possibly matching pillowcases and of a mattress. Industry representatives
four producers of mattresses account for other bedclothes items) to fit the new assert that the average consumer
almost sixty percent of total U.S. thicker mattress. replaces a mattress/set after ten years. A
production. In total, there are 639 If the cost increase is relatively small 1996 CPSC market study estimated the
establishments (as of 2001) that produce or there is no resulting increase in the average expected life of a mattress to be
mattresses in the U.S., using the U.S. price of a mattress/set, then the demand 14 years. The PPM estimates the number
Department of Commerce NAICS (North for bedclothes will only be affected if of (conventional and non-conventional)
American Industry Classification consumers place a higher value on the mattresses in use in 2004 to be 233
System) Code 33791 for mattresses. The safer mattress and replace their current million mattresses, using a 10-year
top four producers account for about mattress sooner than they would have average product life, and 302.6 million
half of the number of all these with no standard in place. An increased mattresses, using a 14-year average
establishments. The number of demand for the safer (and thicker, if the product life. These two numbers are
establishments has been declining over current mattress is relatively old) later used to estimate the pre-standard
time due to mergers and buy-outs. Total mattress will likely result in an baseline risk and the expected benefits
employment in the industry, using the increased demand for sheets that fit the of the proposed standard.
NAICS Code 33791, was 25,500 workers newer mattresses. This effect, however,
in 2001. is not directly resulting from the This analysis focuses principally on
adoption of the proposed standard since queen-size mattresses because they are
The mattress manufacturing industry
the thickness of the mattress need not be the most commonly used. In 2002
has three key supplying industries:
spring and wire product manufacturing, increased by the presence of either type queen-size mattresses were used by 34
broad-woven fabric mills, and foam of barrier. It is the result of the increased percent of U.S. consumers. Following
products manufacturing. Depending on utility some consumers may derive from the queen-size are the sizes: Twin and
the type of fire resistant barrier chosen the safer mattress and the consequent Twin XL (31.2 percent), Full and Full
by different manufacturers, the demand increase in demand for bedclothes. The XL (21 percent), King and California
for foam padding for mattresses might increased demand for safer mattresses King (11 percent), and all other (2.6
decline if it were replaced by the high- would most probably lead to an increase percent). ISPA data reflect that the
loft barrier in the construction of the in sales and employment in the spring average size of a mattress is increasing.
mattress and foundation. This would be and wire products, broad-woven fabric, The average manufacturing price in
offset by an increase in the demand for and foam products industries, as well as 2002 was $152 for a mattress of average
the high-loft barrier. If sheet barriers in the mattress and bedclothes size and $86 for a foundation of average
were chosen by some mattress industries. size. Hence the average manufacturing
producers, then sales of, and Other producers that could price of a mattress/set was about $238
employment by, the sheet barrier potentially be affected, if the price in 2002.
suppliers would increase. Since the change associated with producing There are no readily available data on
sheet barriers would not replace other compliant mattresses is significant, are average retail prices for mattress/
inputs, there would most likely be no those of other substitute products, like foundation sets by size. ISPA, however,
offsetting effect on other industries. airbeds, waterbeds, * * * etc. that reports that mattress/foundation sets
Fiberglass, melamine, and aramid contain no upholstered material and selling for under $500 represent 40.7
producers may also be affected to the would, therefore, not be covered by the percent of the market. Mattress/
extent that they are used to produce fire proposed standard. Their sales may foundation sets selling for between $500
resistant materials used in mattress increase as a proportion of total bedding and $1000 represent 39.2 percent of the
production. products. market.

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2481

5. Trends in Mattress/Bedding staff study of fire investigations from reports were initiated from death
Residential Fires, Deaths, Injuries, and 1999–2004. Risk reductions are then certificates with follow-up
Property Losses calculated on a per-mattress-in-use basis documentation from the fire
Open-Flame Ignition. The staff based on estimates of the number of department. This resulted in a total of
estimates average annual mattress/ mattresses in use. The monetary value 195 deaths and 205 injuries in the
bedding fires from open-flame ignitions of expected benefits per mattress is investigations to be evaluated. The
(including candles, matches and derived using current (i.e., 2004) distribution of mattress ignition sources
lighters) to have been 8,367 and 6,367 estimates for the value of a statistical was not representative of all fires
life and the average cost of a mattress involving mattresses and thus the data
over the 1993–95 and 1996–98 periods
fire injury. To derive the monetary value were weighted to match the NFIRS-
respectively. This represents a reduction
of expected benefits over the life of a based national fire data distributions.
of 23.9 percent. The resulting average Evaluations of the fire incidents by
mattress, the expected annual benefits
mattress/bedding deaths, injuries, and CPSC staff reviewers used the results of
are discounted (using a three percent
property losses from open-flame NIST testing (Ohlemiller, 2004;
discount rate), and then summed over
ignitions have decreased by 28.2 Ohlemiller and Gann, 2003; Ohlemiller
the expected life of the mattress. The
percent, 22.1 percent, and 5.6 percent and Gann, 2002) conducted to assess the
analysis considers mattress lives of 10
respectively, over the 1993 to 1998 hazard produced from burning
and 14 years.
period. When adjusted for inflation, the The potential benefits of the proposed mattresses and bedclothes. Specifically,
decrease in the value of property losses standard consist of the reduction in the evaluations were based on the
becomes 37.7 percent. deaths, injuries, and property damage expectation that occupants in bed when
Smoking Material Ignition. The staff that would result. Since the objective of the fire ignited but able to escape the
estimates average annual mattress/ the proposed standard is to reduce the burning bedclothes in the first three to
bedding fires from smoking material likelihood of flashover or increase the five minutes faced a minimal hazard.
ignition (including cigarettes, cigars, time before flashover occurs, and not to Occupants in direct contact with
and pipes) to have been 7,733 and 6,067 reduce fires, changes in property losses burning bedclothes for a longer period
over the 1993–95 and 1996–98 periods associated with the proposed standard (5 to 10 minutes) would be subject to
respectively. This represents a reduction are hard to quantify. Property losses are potentially hazardous levels of heat
of 21.6 percent over the 1993 to 1998 expected to decline but the extent of the release. If the burning bedclothes did
period. Average annual deaths, injuries, decline cannot be quantified. not ignite other non-bedding items or
and property losses due to mattress/ Consequently, for purposes of this produce flashover at this time, heat
bedding smoking material ignitions analysis, no reduction in property losses release would subside temporarily and
have decreased by 4.7 percent, 19.7 is assumed. That is, all expected then begin to increase as the
percent, and 9.7 percent, respectively, benefits from the proposed standard are involvement of the mattress increased.
over the same period. When adjusted for in the form of prevented deaths and These conditions would allow
inflation, the decrease in the value of injuries. This underestimates net occupants 10 to 15 minutes to escape
property losses becomes 40.4 percent. benefits, since there will likely be some the room of origin before the situation
Other Ignition Sources. The staff benefits from reduced property losses. in the room would become untenable.
estimates average annual mattress/ The proposed standard is expected to Since the proposed standard is expected
bedding fires from other ignition reduce the likelihood of flashover to slow the rate of fire spread and hence
sources (including sparks, embers, or resulting from fires started by smoking increase escape time, assuming that
flames escaping from fueled equipment, materials or other ignition sources, as bedclothes do not contribute enough
arcs or sparks from electric equipment, well as those started by open-flame heat to pose a hazardous condition, it
small torches, hot embers, and ignition. Fires, injuries, and property was assumed that no deaths would
fireworks, heat escaping from fueled losses resulting from smoking material occur among people who were outside
equipment, molten material, short ignition and other ignitions, and deaths the room of origin at the time of
circuit arc, and heat overloaded from smoking material ignition are ignition, unless they entered the room
equipment) to have been 8,633 and lower for the 1996–98 period than the later or were incapable of exiting on
7,767 over the 1993–95 and 1996–98 1993–95 period. (Deaths from other their own. The analysis focused on
periods respectively. This represents a ignition sources are more than 50 reduction of deaths and injuries because
reduction of 10 percent over the 1993 to percent higher). Any additional the proposed standard is designed to
1998 period. Average annual injuries reduction in these figures due to the limit fire intensity and spread rather
and inflation-adjusted property losses proposed open-flame ignition standard than prevent ignition.
have decreased by 13.8 percent and 38.7 will translate into societal benefits, as Each investigation was evaluated by
percent respectively. Average annual will be discussed in the benefit-cost CPSC staff reviewers to identify the
deaths increased by 51.7 percent (from analysis (Section 8). features related to the occurrence of a
97 to 147). This increase offsets the Estimates of the effectiveness of the death or injury once the fire was ignited.
decrease in deaths resulting from open- proposed standard are based on a CPSC These included casualty age, casualty
flame and smoking material ignition staff evaluation of in-depth investigation location when the fire started (at the
fires. The annual average number of reports of fires (including details of the point of ignition, in the room of origin
deaths from all ignition sources occupants’ situations and actions during but not at the point of ignition, or
remained unchanged over the period, the fire) occurring in 1999–2004 in outside the room of origin), whether the
equal to 510. which a mattress or bedding was the casualty was asleep, or suffered from
first item to ignite, the fire was of the additional conditions likely to increase
6. Expected Benefits of the Proposed type considered addressable by the the time needed to escape, whether the
Standard proposed standard, and a civilian death casualty engaged in fighting the fire, and
The expected benefits of the proposed or injury resulted. Most of the whether a rescuer was present. All of
standard are estimated as reductions in investigations also included these conditions were used to determine
the baseline risk of death and injury documentation from the fire department a range for the likelihood that each
from all mattress fires, based on a CPSC that attended the fire. Some incident individual death or injury would have

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2482 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

been prevented had the draft proposed mattress are 1.02 to 1.09 deaths and 5.49 7. Expected Costs of the Proposed
standard been in effect. Percentage to 5.88 injuries per million mattresses. Standard
reductions of deaths (injuries) within Annual risk reductions resulting from This section presents the expected
subcategories of heat source and age the proposed standard are used to resource costs associated with the
group were applied to equivalent derive the monetary benefits from proposed standard. Resource costs are
subcategories of the national estimates reduced deaths and injuries. The costs that reflect the use of a resource
based on the NFIRS and NFPA data for that would have been available for other
estimated reduction in the risk of death
1995–1999. The estimated reductions uses had it not been used in conjunction
is multiplied by the value of a statistical
per category were summed and the with the production of mattresses
life (and divided by a million) to derive
overall percentage reductions were compliant with the proposed standard.
calculated as the percent of addressable a first-year monetary estimate for the
range of benefits from lives saved per These costs include material and labor
deaths (or injuries) that would have costs; testing costs; costs to wholesalers,
been prevented if the likelihood of mattress. Based on the existing
literature, a value of a statistical life of distributors, and retailers; costs of
flashover were reduced in the first 30 producers’ information collection and
minutes and victims had 10 to 15 five million dollars is assumed (Viscusi,
record keeping; costs of quality control/
minutes of escape time. 1993). The estimated reduction in the
quality assurance programs; and
The staff indicates that the proposed risk of injury is similarly used to derive compliance and enforcement costs. The
standard is expected to reduce all the range of first-year monetary benefits effect on retail prices will be discussed
addressable deaths from mattress/ from injuries prevented. The benefits in Section 8.
bedding fires by 80 to 86 percent and from preventing an injury (the cost of an Material and Labor Costs. To comply
reduce all addressable injuries from injury) in 2004 are estimated to average with the proposed standard, the
mattress/bedding fires by 86 to 92 about $179,300, based on Miller et. al. construction of most mattress/sets will
percent. The results vary only slightly (1993). The first-year benefits associated include a barrier technology with
by source of ignition. These estimated with preventing deaths and injuries improved fire performance. This barrier
effectiveness percentages result in the equal $7.93 to $8.45 for an estimated may be thick (high-loft) or thin (sheet).
prevention of an estimated 310 to 330 mattress life of 10 years and $6.11 to High-loft barriers are generally used to
deaths and 1660 to 1780 injuries $6.51 for an estimated mattress life of 14 replace some of the existing non-woven
annually, for the 1998–2002 period. years. fiber, foam, and/or batting material,
The staff’s analysis presents the Lifetime benefits are derived by leading to a smaller increase in costs
estimated annual deaths and injuries than sheet barriers, which constitute an
projecting annual benefits for the life of
that are expected to be prevented by the addition to production materials (and
the mattress and summing the
proposed standard, based on average costs).
discounted (at a rate of 3 percent) According to several barrier producers
figures for 1998–2002. For purposes of stream of annual benefits (measured in
this analysis, it is assumed that the and mattress manufacturers, the price of
constant dollars). The number of a high-loft barrier that would make a
annual deaths and injuries prevented by mattresses in use is projected to grow at
the proposed standard equal the average mattress comply with the proposed
a rate of zero to three percent, based on standard, defined to have a width of 88
annual deaths and injuries prevented for
the average growth rate for the 1981– to 92 inches, is $3.00 to $5.00 per linear
the 1998–2002 period. The analysis is
2002 period. Since the number of deaths yard. The high-loft barrier replaces the
conducted as if the standard went into
and injuries are implicitly assumed to currently-used polyester batting, which
effect in 2004. All dollar estimates are
remain constant over time, a positive costs an average of $0.50 to $1.70 per
based on constant 2004 dollars. A
growth rate of mattresses in use implies linear yard. Hence, the net increase in
discount rate of 3 percent and average
expected lives of a mattress of 10 and a declining risk over time. The lower the cost attributed to the use of the high-
14 years are also assumed. end of the ranges for estimated (10 and loft barrier is $1.30 to $4.50 per linear
14 years) lifetime benefits correspond to yard, which translates to a net increase
The estimated ranges of deaths and
injuries prevented are calculated by a 3 percent projected growth rate and in barrier-related manufacturing costs of
applying the range of percent reductions the lower end of the effectiveness $7.80 to $27.00 for a queen-size
to average addressable deaths and ranges. The upper end of the ranges for mattress/set.4 The queen-size is used for
injuries for the period 1998–2002. estimated (10 and 14 years) lifetime all the cost estimates, because it is the
Staff’s analysis also presents the risk benefits correspond to a zero percent mode size, used by 34 percent of
reduction in deaths and injuries that projected growth rate and the upper end consumers in 2002.
would result from the proposed of the effectiveness ranges. For an According to several barrier producers
standard (per million mattresses). Based expected mattress life of 10 years, the and mattress manufacturers, the price of
on the estimated number of mattresses resulting expected lifetime benefits of a sheet barrier that would make a
in use (described in Section 4) and an saved lives associated with the mattress comply with the proposed
average expected life of 10 years, the proposed standard equal $51.70 to standard is $4.00 to $6.00 per linear
annual reduction in the risk of death $62.22 per mattress. The corresponding yard. Because of its different texture, the
equals 1.33 deaths per million benefits of prevented injuries equal sheet barrier would generally not
mattresses (310 deaths divided by the $9.93 to $12.03. Hence, for an expected replace any of the materials being used
estimated 233 million mattresses in use mattress life of 10 years, the expected in the construction of the mattress/set.5
in 2004) to 1.42 per million mattresses total lifetime benefits of a compliant
4 This calculation is based on the assumption that
(330 deaths / 233 million mattresses). mattress equal $61.66 to $74.25. For an a queen-size mattress/set requires six linear yards
The estimated reduction in the risk of expected mattress life of 14 years, total of the barrier mateiral to be used in the two (top
injury, similarly calculated, equals 7.12 benefits equal $59.88 to $75.71 per and bottom) panels of the mattress and the side
to 7.64 injuries per million mattresses mattress. The sensitivity analysis panels of both the mattress and foundation. Some
producers are able to use less than six linear yards,
for an estimated 10-year life of a section below examines how the results which reduces their cost per queen mattress/set.
mattress. The estimated risk reductions might change when a discount rate of 5 The only exception to this might involve using

for an estimated 14-year life of a seven percent is used. a sheet barrier in the side panel of the mattress and

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2483

This translates to $24.00 to $36.00 for a (so that they can compete with other the first year of production will be
queen-size mattress/set. The large barriers available for sale). Hence the reduced. Pooling testing results across
difference in the net cost of the two total materials and labor costs will most establishments and/or firms will further
barrier types suggests that if a barrier’s likely be closer to the lower end of the reduce the average cost of testing per
fire performance is not a function of its estimated range. mattress. On an annual basis testing
type, most manufacturers will use high- Costs of Prototype and Confirmation costs will be further reduced because
loft barriers, the less costly alternative. Testing. Each mattress/set prototype is prototypes need only be tested in the
A large mattress manufacturer also required to be tested in triplicate for year they are first developed.
indicated that mattresses produced with prototype qualification. According to Cost of Information Collection and
sheet barriers in the top panel of the industry representatives, the cost of Record Keeping. In addition to
mattress (as opposed to the side panels) testing per twin-size mattress/set may be prototype testing, the proposed standard
may be less comfortable. about $500: the sum of the average cost will require detailed documentation of
In addition to the increase in material of the materials and shipping ($100) and all tests performed and their results
costs due to the use of a barrier, costs the cost of the use of the lab ($400). including video or pictures; prototype
will increase due to the use of fire- Hence, the cost of testing three or production identification number;
resistant (FR) thread for tape stitching. mattresses/sets for prototype date and time of test; and name and
According to several thread producers, qualification equals $1500. location of testing facility; test room
the cost of FR thread is $0.41 to $0.60 Additionally, if some mattress/set conditions; and test data for as long as
per queen-size mattress/set. Given that prototypes do not pass the first time, the prototype is in production and for
the cost of nylon (non-FR) thread is then the cost will be higher, because three years after its production ceases.
about $0.10 per queen-size mattress/set, additional tests will be done after action Manufacturers are also required to keep
the net increase in costs per queen-size is taken to improve the resistance of the records of a unique identification
mattress/set due to the use of FR thread prototype. If 10 percent of mattresses are number for the qualified prototype and
is $0.31 to $0.50. retested, then the average cost of testing a list of the unique identification
Costs may also increase due to a prototype would be 10 percent higher, numbers of each prototype based on the
slightly reduced labor productivity. or $1650. This cost is assumed to be qualified prototype and a description of
Based on industry estimates of an incurred no more than once per the materials substituted and/or the size
average of two labor hours for the establishment for each prototype. It is change. Moreover, they are required to
production of a queen-size mattress/set, expected that a qualified prototype will document the name and supplier of
and a 10 percent reduction in labor be used to represent a mattress each material used in construction of a
productivity and an industry average construction (e.g., single-sided pillow prototype and keep physical samples of
hourly wage rate of $11.50, the cost top) with all other prototypes using the the material. Additionally, they are
increase due to reduced labor same construction (with different size required to identify the details of the
productivity is about $2.30. and different ticking materials) being application of any fire retardant
The increase in the materials and based on the qualified prototype. If treatments and/or inherently fire
labor costs of a mattress, is thus equal companies pool their prototype resistant fibers employed relative to
to $10.41 ($7.80 barrier cost + $0.31 definitions across different mattress components. This
thread cost + $2.30 labor cost) to $29.80 establishments or different companies, documentation is in addition to
($27 barrier cost + $0.50 thread cost + testing costs would be smaller as all but documentation already conducted by
$2.30 labor cost) for a high-loft barrier one of the firms/establishments mattress manufacturers in their efforts
and $26.61 ($24.00 barrier cost + $0.31 producing to the specification of a to meet the cigarette standard. Detailed
thread cost + $2.30 labor cost) to $38.80 pooled prototype will burn one mattress testing documentation will be done by
($36 barrier cost + $0.50 thread cost + (for the confirmation test) instead of the test lab and is included in the
$2.30 additional labor cost) for a sheet three (for the prototype test). The estimated cost of testing. Based on CPSC
barrier.6 Various types of high-loft and probability of a mattress failing a Office of Compliance staff estimates, all
sheet barriers are widely available for confirmation test is small. Therefore, it requirements of the proposed standard
sale and therefore it is expected that is expected that the average cost of are expected to cost an establishment
those whose prices are at the upper end testing per mattress will be lower for about 110 minutes, or 1.3 hours, per
of the range will either not be produced firms and/or establishments that pool qualified prototype. Assuming that
(because mattress manufacturers will their results than for those that do not. every establishment will produce 20
not buy them) or their prices will drop If manufacturers test every mattress different qualified prototypes, the
construction (e.g., single-sided pillow increase in record keeping costs is about
foundation. Because the existence of cushioning top, double-sided pillow-top, tight-top, $935 (110 minutes × 20 qualified
along the side of the mattress and foundation would
probably not be noticed or missed by consumers,
euro-top, * * * etc.), which is prototypes × $25.50 in average civilian
substitution of the sheet barrier for the material estimated, based on conversations with workers’ compensation per hour) per
currently being used in the side panel may be manufacturers, to average about twenty establishment per year. (Note that
implemented to reduce the cost of using the sheet per manufacturer, for every pooling among establishments or using
barrier. The side panel is small, relative to the size establishment in a given year, then their
of the entire surface area of a mattress/set, and its
a qualified prototype for longer than one
possibly different construction is therefore not average testing cost per mattress would year will reduce this estimate.) This
included in the cost calculation. This leads to a approximately equal 92.5 cents ($1650 × translates to an average cost of 2.6 cents
slight over-estimation of the cost of the sheet barrier 20 constructions × 639 establishments/ per mattress for an average
and consequently the relative cost of using a sheet 22.8 million conventional mattresses)
instead of a high-loft barrier.
establishment, with average output of
6 Some producers are also using an FR mattress per mattress for the first year of 35,681 conventional mattresses.
edge binding tape, which costs an average of $2.52 production. If manufacturers use a Cost of Quality Control/Quality
per mattress, while a non-FR tape costs an average qualified prototype of the least fire- Assurance Programs. To ensure that all
of $1.68. This makes the net increase in costs, due resistant mattress/set construction mattresses are produced to the
to using FR edge binding tape, equal to $0.84. This
cost is not added to the total production costs,
(‘‘worst case’’) to represent other prototype specification across all
because it is not required for the mattress to pass mattress/set constructions, then the factories and over the years for which a
the burn test. average cost of testing per mattress for production line exists, mattress

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2484 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

manufacturers will need a thorough complying with the proposed standard discretion of the investigator and,
well-documented quality control/ will not be bigger than a similar therefore an accurate assumption about
assurance program. The top 12 mattress mattress produced before the standard the number of samples collected and
producers (with a market share of becomes effective, storage and sent for a burn test cannot be made. If,
almost 80 percent) have existing quality transportation costs are not expected to based on inspection, samples from 10
control programs which could be increase. Inventory financing costs will percent of all inspected establishments
modified to fit the new standard with increase by the average cost of were to be collected and sent to a lab for
minimal additional costs. Smaller borrowing money, applied to the a burn test, and if samples representing
producers, whose quality control wholesale price of a mattress over the 5 qualified prototypes are taken from
programs are less detailed or non- average inventory holding time period. each of these establishments, then the
existent, will incur some incremental Since most mattress producers use just- total cost of CPSC testing will be
costs as a result of the proposed in-time production and have small $157,500 (5 qualified prototypes ×
standard. These incremental costs will inventories, this additional cost will $1,500 (the cost of testing up to 3
be small for each manufacturer and less probably not exceed ten percent of the mattresses for each qualified prototype)
when measured per mattress. (See the increase in production cost (which is × 21 (10 percent of 213 inspected
section on impact of the proposed the sum of material, labor, testing, establishments)). These assumptions
standard on small businesses for a record keeping, and quality assurance about frequency of testing yield an
description of their cost of quality costs). A ten percent mark-up is, expected cost of testing per
control and quality assurance therefore, being used to measure the establishment of $246.48 ($157,500/
programs.) cost to wholesalers, distributors, and 639).
Additionally, although the proposed retailers. This yields a resource cost to Therefore the expected total CPSC
standard does not require production wholesalers, distributors, and retailers wage and testing costs associated with
testing, it encourages random equal to $1.15 to $3.98 per mattress/set. the proposed standard per
production testing to assure Retail prices may increase by more than establishment per year equal $1,145.59
manufacturers that their mattresses the ten percent mark-up. Section 8 ($899.11 + $246.48). With an average
continue to meet the requirements of the discusses the impact of the proposed production of 35,681 mattresses per
rule, as a possible component of the standard on retail prices of mattresses. establishment (22.8 million mattresses
quality control/quality assurance Costs of Compliance and divided by 639 establishments), the
program. Assuming that an average of 3 Enforcement. Compliance and average CPSC wage and testing costs
mattress/foundation constructions will enforcement costs refer to the costs equal 3.2 cents per mattress ($1,145.59/
be tested per establishment per year incurred by CPSC to ensure that 35,681). These costs are expected to
yields an estimated cost of production manufacturers are complying with the decrease over time as manufacturers
testing of about $1500. Based on this proposed standard. Based on past learn the requirements of the proposed
assumption, the estimated cost of testing experience with the existing mattress standard.
mattress/foundation sets for quality standard, the estimated CPSC inspection Total Resource Costs. Therefore total
assurance purposes, therefore, equals time spent per location (establishment) resource costs (including material costs,
4.2 cents per mattress ($1500/35,681) for equals 33 hours for inspection and 6 labor costs, costs of prototype and
an average establishment. hours for sample collection. This yields confirmation testing, paperwork
The labor needed to meet the quality a cost per inspection of about $1,664.52 collection and record keeping costs,
assurance measures required by the (39 hours × $42.68, the average wage costs of quality control/quality
standard is estimated by CPSC Office of rate for CPSC inspectors). Additionally, assurance programs, production testing
Compliance staff to be 224 minutes per compliance officers spend an average of costs, costs to wholesalers, distributors,
establishment per prototype per year. 20 hours per case, making their cost and retailers, and costs of compliance
Assuming that every establishment will equal to $1,032.80 (20 hours × $51.64, and enforcement) are estimated to range
produce 20 qualified prototypes, the the average hourly wage rate for from $12.63 to $43.86 per mattress. This
increase in labor costs associated with compliance officers). This yields an range includes both the high-loft and
quality assurance requirements of the average compliance and enforcement sheet barriers. The section on the impact
proposed standard is about $1904 (224 total labor cost of $2,697.32 per of the proposed standard on small
minutes × 20 qualified prototypes × inspected establishment per year. businesses and other small entities
$25.50 average civilian workers’ It should be noted that the expected discusses how costs of testing and
compensation per hour) per cost per establishment, if less than one quality control/quality assurance
establishment per year. (Note that hundred percent of establishments are programs may differ for small
pooling among establishments or using inspected every year, equals the cost per businesses and strategies that small
a qualified prototype for longer than one inspected establishment times the manufacturers might adopt to reduce
year will reduce this estimate.) This probability that a given establishment these costs.
yields an average cost of 5.3 cents per will be inspected. Though the Projected Future Costs. It is possible
mattress for an average establishment, probability that a given establishment that costs associated with the standard
with average output of 35,681 will be inspected in a given year is not will decline over time. A supplier of fire
mattresses. Hence total costs of quality known, assuming that a third of all resistant barriers predicts that the price
assurance/quality control programs may establishments will be inspected (i.e., of the barriers will decline by 40 percent
average about 9.5 cents (4.2 + 5.3) per about 213 establishments) yields a in the next two years, due to decreased
conventional mattress per year. compliance and enforcement total uncertainty and increased competition.
Costs to Wholesalers, Distributors, expected labor cost of $899.11 (They have already dropped
and Retailers. An added cost of the ($2,697.32 × (1⁄3)) per establishment per significantly since TB603 was
proposed standard is the increase in year. proposed.) The increase in labor costs
costs to wholesalers, distributors, and In addition to labor costs, CPSC will due to decreased productivity is
retailers in the form of additional incur testing costs. It should be noted expected to be temporary and be
storage, transportation, and inventory that the decision to collect samples after reduced when workers get more training
financing costs. Since a mattress an inspection visit is made at the and/or the older machines get replaced

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2485

with newer machines that are more 8. Benefits and Costs of the Proposed from 10 to 14 years, while using the 3
capable of handling the FR thread and Standard percent discount rate, expands the
material used in fire resistant barriers. This section compares benefits and positive range of net benefits, it does not
Moreover, as noted above, prototype costs of the proposed standard, presents affect the conclusion regarding net
testing costs are expected to decline a sensitivity analysis, and highlights the benefits. A further increase of the
after the first year of the standard. impact of the proposed standard on expected life of a mattress similarly
The proposed standard references an retail prices, small businesses, children, would not affect the estimate of net
effective date of twelve months and the environment. The sensitivity benefits. For example, using the Product
following publication of a final rule. analysis examines the effect of changing Population Model estimate of the
The costs reported here are based on the some of the assumptions used earlier. number of mattresses in use based on an
assumption that supplier companies The analysis shows that net benefits expected mattress life of 18 years (equal
will be able to maintain existing continue to be positive under a to 367.1 million mattresses) yields net
capacity. If federal standards for reasonable range of assumptions about benefits of $14.42 to $64.49 per
bedclothes and upholstered furniture the death and injury effectiveness of the mattress, using a discount rate of 3
were mandated at the same time and proposed standard, the reduction in percent.
input producers were not given enough injuries resulting from the proposed Net benefits are also positive using
time to increase their capacity, input standard, the value of a statistical life discount rates of 3 and 7 percent. Using
prices would rise in the short-run estimate, the discount rate, or the a 3 percent discount rate, net benefits
because of increased demand for the FR expected mattress life. per mattress equal $17.79 to $61.62 for
material used by all three industries. The expected aggregate lifetime an average life of 10 years and $16.01
Unquantifiable Costs. A mattress benefits associated with one year’s to $63.08 for an average life of 14 years.
manufacturer indicated that in response production of mattresses (25.3 million Using a 7 percent discount rate, net
to an FR mattress standard, the number units) using a discount rate of three benefits per mattress equal $9.36 to
of models/styles produced may be cut percent and an expected 10-year $50.88 for an average life of 10 years
by half. If this response is typical, then mattress life are $1.56 to $1.88 billion and $5.15 to $48.26 for an average life
there may be a reduction in consumers’ ($61.66 to $74.25 per mattress × 25.3 of 14 years. Assuming a larger discount
utility, because of the reduction in million mattresses). The corresponding rate reduces net benefits, because future
mattress types that they would have to expected aggregate costs of the proposed benefits reaped over the life of the
choose from. Others indicate that there standard are $0.32 to $1.11 billion mattress contribute less to total benefits.
will be an aversion to producing double- ($12.63 to $43.86 times 25.3 million). Net benefits are based on an estimated
sided mattresses, because it would be The resulting net aggregate benefits value of a statistical life equal to $5
harder for them to pass the burn test. equal $0.45 to $1.56 billion ($17.79 to million. Changing the estimate used for
Double-sided mattresses possibly have a $61.62 times 25.3 million). For a the value of a statistical life does not
longer expected life than single-sided mattress life of 14 years (and a 3 percent have a major impact on the results. For
ones. To the extent that consumers discount rate), aggregate lifetime example, if $3 million, the lower bound
prefer double-sided mattresses to single- benefits, costs, and net benefits of the estimate in Viscusi (1993), is used as an
sided mattresses, the shift away from proposed standard associated with one estimate of the value of a statistical life,
producing double-sided mattresses year of production are $1.52 to $1.92, net benefits become -$2.90 to $36.73 per
imposes a non-monetary cost. Though $0.32 to $1.11, and $0.41 to $1.60 mattress (using a 3 percent discount rate
unquantifiable, this reduction in choices billion respectively. The expected and an estimated mattress life of 10
of construction type and design is an benefits of the proposed standard will years).7 Alternatively, a $7 million
added cost to consumers of the accrue for a long period of time and estimate, the higher bound estimate in
proposed standard. discounted net benefits will, therefore, Viscusi (1993), yields net benefits equal
Another unquantifiable cost is the be much greater than net benefits to $38.48 to $86.51 per mattress (using
possible increase in liability insurance associated with only the mattress a 3 percent discount rate and an
faced by mattress manufacturers. production in the first year the standard estimated mattress life of 10 years).
Because the draft proposed standard becomes effective. Changing the estimate used for the
measures the performance of the entire Sensitivity Analysis. The previous cost of injury will have minimal impact
mattress when exposed to fire, and not analysis compares benefits and costs of on the results, because the share of
its individual components, liability will the proposed standard using expected benefits from reduced injuries is 16
be shared by input suppliers and mattress lives of 10 and 14 years, a percent of total benefits. Hence, even if
mattress manufacturers. Industry discount rate of 3 percent, an expected there were no reduction in injuries from
representatives expect that effectiveness rate of the proposed the proposed standard, the net benefits
manufacturers’ liability insurance will standard of 80 to 86 percent of deaths would be $7.86 to $49.59 per mattress
increase to reflect the additional and 86 to 92 percent of injuries, an (using a mattress life of 10 years and a
possibility of litigation. This increase, estimated value of a statistical life of 5 3 percent discount rate).
however, cannot be quantified because million dollars, and an estimated cost of The analysis assumes that the
of the novelty of this performance test. injury of $179,300. This section effectiveness of the proposed standard
Compliance of more mattress firms with examines the effect of changing any of ranges from 80 to 86 percent for deaths
the California TB 603 standard may these assumptions on the expected net and 86 to 92 percent for injuries. The
enable us to estimate the additional benefits of the proposed standard. 7 The range for net benefits was derived by
liability insurance. Notice that any Comparing expected benefits and subtracting the upper end of the cost range from the
increase in liability insurance faced by costs of the proposed standard, it is lower end of the benefits range to get the lower end
FR input suppliers will be included in clear that net benefits are expected to be of the range for net benefits and subtracting the
the price charged for the FR inputs and positive (i.e., expected total benefits lower end of the cost range from the higher end of
the benefits range to get the higher end of the range
does not add to the total increase in exceed expected costs) for an average for net benefits. Because of this method, both ends
resource cost that is expected to result mattress life of 10 or 14 years. Though of the range for net benefits are a very unlikely
from the proposed standard. increasing the expected mattress life occurrence.

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2486 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

net benefits will remain positive, with a curve is unaffected, the equilibrium Impact on Small Businesses and
lower effectiveness rate. For example, price will reflect the price elasticity of Other Small Entities. The increase in
assuming an effectiveness rate of demand (i.e. the sensitivity of the material and labor costs to meet the
preventing death of only 55 percent change in the quantity demanded to the proposed standard is not likely to be
yields net benefits of $1.86 to $39.84 per change in price) as well as the shift in dependent on a firm’s size and will
mattress and aggregate net benefits of 50 supply. In the short-run, consumers therefore not disproportionately affect
million to 1.01 billion dollars from all have a relatively elastic demand curve, small businesses. The cost imposed
mattresses produced the first year the because they can always postpone the disproportionately (per unit produced)
proposed standard is mandated (using a purchase of a durable good, and on small businesses will be the cost of
mattress life of 10 years, a 3 percent therefore the increase in the equilibrium testing, information collection and
discount rate, and the same price is expected to be much lower than record keeping and quality control/
effectiveness for injuries as used in the the increase in the supply price (what quality assurance programs. While these
baseline analysis). Also, assuming a producers would want to sell the same costs are estimated to be a little over one
smaller number of deaths and injuries number of mattress/sets for). Because of dollar per mattress per year for average-
before the proposed standard is the relatively high elasticity of demand, sized establishments, they could be
mandated (a smaller baseline risk) sales are likely to decrease in the short- substantially higher for small mattress
would still result in positive net run. In the long-run, the demand curve manufacturers. The proposed rule
benefits. A 25 percent reduction in is less elastic, and therefore the includes measures that these
baseline death and injury risks yields equilibrium price and quantity (sales) manufacturers can use to minimize the
net benefits of $2.38 to $43.06 per will be higher than the short-run price testing burden. Furthermore, firms with
mattress and aggregate net benefits of and quantity. more than one establishment, or
$60 million to $1.09 billion from all Given the availability of mattresses different firms, may be able to reduce
mattresses produced the first year the whose retail prices will not increase and these costs by pooling their testing and
mattress standard is mandated (using a the competitive nature of the industry, quality control programs over all
mattress life of 10 years, a 3 percent it is possible that, on average, prices establishments or firms.
discount rate, and the estimated will rise by about twice the costs Use of pooling across establishments
effectiveness measures used in the associated with the standard (i.e., retail and firms would ameliorate the impact
baseline analysis). price mark-up will average about twice of the proposed standard on small
Impact on Retail Prices. One of the the increase in manufacturing costs). businesses. By getting together across
top four mattress manufacturers in the Under this assumption, consumers different states and regions, small
industry has re-merchandised its would pay an additional $22.91 ($11.46 manufacturers who do not share a
product lines to lower the costs of other × 2) to $79.69 ($39.85 × 2) per mattress/ common market (and therefore do not
materials so that total costs (and prices) set (compared to the price they would compete with each other) can resemble
are the same as they were before the have paid for a current mattress that a large producer in their testing and
production of mattresses that comply quality control/quality assurance efforts
does not comply with the proposed
with TB603. Other manufacturers have and therefore reduce their costs per
standard.8 Assuming that the demand
indicated that they will have to increase mattress. It is also expected that some
curve for mattresses is unaffected by the
their price which, according to some barrier suppliers would be willing to do
draft proposed standard, some
manufacturers and based on reported the testing and quality control/
consumers will choose not to purchase
traditional industry mark-ups, might assurance programs for small
(or at least delay the purchase of) a new
translate to an increase in the retail manufacturers in exchange for a small
mattress/set. These consumers who
price to consumers that could reach charge, which will be similar to the
delay or choose not to purchase a new
approximately four-fold the increase in average cost per mattress for large
set will not be getting the value (or
manufacturer’s costs. Hence the average businesses, because the volume of
increase in the price at which mattress benefits) that they would have gained output will be large.
manufacturers are willing to sell their from purchasing a new set. This loss, Impact on the Environment. The
products (supply price) will be though difficult to quantify, is extraction, processing, refinement, and
anywhere between the price of a similar sometimes measured as a loss in conversion of raw materials to meet the
mattress without FR material and that consumer surplus (McCloskey, 1982). proposed standard involve energy
price plus four times the increase in the It is unlikely, however, that the post- consumption, labor, and the use of
costs of production. Given the presence standard demand curve for mattresses potentially toxic chemicals. Most
of at least one company that will not will be the same as the current demand. manufacturing has some impact on the
increase the price, it is unlikely that the Early 2004 market observations indicate environment, and manufacturing fire
new average price will be close to the consumer and retail enthusiasm about resistant mattresses is no exception.
higher end of the range because of the fire resistant mattresses already Because the proposed standard is a
competition for market share among available for sale (Furniture Today, performance standard, it does not
manufacturers. April 26th, 2004.) If this enthusiasm restrict manufacturers’ choice of fire
The market (equilibrium) price is generally reflects consumers’ resistant materials and methods that
determined by the intersection of preferences, then the demand for could be used in the production of
consumers’ willingness to buy and mattresses may increase. This would mattresses. There appear to be several
producers’ willingness to sell the tend to offset any reduction in mattress economically viable options to meet the
product at different prices. The value sales and possible losses in consumer standard that, based on available
the equilibrium price will take (relative surplus. information, do not impose health risks
to the price before the introduction of to consumers or significantly affect the
fire resistant mattress/sets) will be 8 These cost figures include labor and material
environment. (See discussion at Section
affected by the change in the demand costs; testing costs; record-keeping costs; and N of this preamble.)
quality assurance program costs. They do not
and supply curves for fire resistant include the costs to wholesalers, distributors, and
Impact on Children. Deaths and
mattress/sets and their relative retailers or compliance costs because they are not injuries among children constitute a
elasticities. Assuming that the demand incurred by the manufacturers. substantial proportion of mattress-

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2487

related fire losses, and of the potential resource costs to manufacturers of a net benefits may not justify the large
benefits of the proposed standard. A queen mattress/foundation set by $15.06 increase in retail price that would result
CPSC staff report, based on a field to $50.65 compared to non-complying from a stricter standard.
investigation study in 1995 to learn products (i.e., those not conforming to A bedding official estimated that such
more about cigarette-ignited fires and the proposed standard.) 9 Adding the price increases may result in reduction
open-flame fires, found that 70 percent costs to wholesalers, distributors, and in sales of 25 percent or more
of open-flame fires involved child play retailers, the costs of testing, quality (Furniture/Today, July 21, 2004). The
and that child play was involved in 83 control/assurance programs, record- larger increase in prices (compared to
percent of the 150 deaths of children keeping, and CPSC compliance efforts, the less strict test requirements) and the
less than five years of age. A National yields a total resource cost of the stricter resulting reduction in sales could drive
Association of State Fire Marshals 1997 standard (150 kW and 60 minutes) of some of the smaller manufacturers out
study also indicated that 66 percent of $16.59 to $55.74 (costs to manufacturers of business. (The stricter standard is
the small open-flame ignitions were {$15.06 to $50.65} + cost to wholesalers, more likely to require replacing some
reportedly started by children under the distributors, and retailers, equal to 10% existing machines to accommodate the
age of 15 (21 percent by children under of costs to manufacturers {$1.51 to denser barrier material, which would be
5). $5.06} + 3.2 cents CPSC compliance disproportionately more costly for
For virtually all of the fires started by costs) per mattress. This represents a smaller firms whose machinery is older
children less than 15 years of age, the marginal increase in costs of $3.96 and less sophisticated.) Since mattresses
ignition was not witnessed by an adult ($16.59¥$12.63) to $11.88 are durable goods, one would expect a
(Boudreault and Smith, 1997). Reducing ($55.74¥$43.86) over the costs larger drop in sales in the short-run, as
the likelihood of flashover in the first 30 associated with the proposed consumers choose to keep their old
minutes of the fire may therefore benefit standard.10 mattresses longer than before. This
children disproportionately, as it allows Such increase in costs would likely would make the reduction in sales more
enough time for adults to detect the fire result in consumers facing higher pronounced in the short-run, increasing
and save young children in close mattress prices. Based on traditional the likelihood that some firms may exit
proximity to the fire. Also children industry mark-ups, the new price may the market. Moreover, if a large number
between 5 and 9 who sometimes do not reflect a two- to four-fold increase over of consumers choose to extend the life
cooperate with adults and run away the increase in production costs, of their mattresses for a longer time
from adults to other parts of the depending on the relative elasticity of period, it will take longer to achieve the
occupancy will have enough time to be demand and supply for mattresses. This benefits expected to be associated with
found and rescued by an adult. yields a total increase in the average the safer mattresses.
The Epidemiology staff’s price of a queen mattress/set of $30.11 Alternative Total Heat Released in the
memorandum shows that, based on (2 × $15.06) to $202.58 (4 × $50.65). First Part of the Test. TB 603 limits the
national fire estimates for the years Potential benefits of the stricter standard total heat released during the first 10
1995–1999, children younger than 15 could be higher than the proposed minutes of the test to 25 MJ. The
accounted for 27 percent of addressable standard, but the extent is uncertain. proposed standard’s stricter limit (15 MJ
deaths and 23 percent of addressable Given an effectiveness rate of greater in the first 10 minutes) reduces the
injuries. They also indicate that the than 80 percent of the proposed expected size of the initial fire and
proposed standard would reduce deaths standard, the additional benefits of hence allows consumers a greater
and injuries to children ages 5 and stricter test requirements are limited. chance to escape the fire and get out of
younger by 85 to 92 percent and 80 to Assuming that the stricter standard the room, even if the room never
87 percent respectively. Deaths and could eliminate 50 percent of the reaches flashover. The effectiveness
injuries to children ages 5 to 14 were remaining deaths and injuries (i.e., it rates presented in the analysis are based
estimated to be reduced by 94 to 97 could save 39 additional lives and on the stricter criterion. Using the TB
percent and 88 to 94 percent prevent 136 additional injuries), then an 603 criterion (25 MJ in the first 10
respectively. This represents a total of additional benefit of about $7.66 per minutes) would likely reduce estimated
100 to 110 deaths of children less than mattress would be expected. This benefits (the estimated reductions in
15 years of age per year for the 1995– additional benefit, however, would deaths and injuries), without having any
1999 period. It also represents 410 to come with additional costs (discussed significant effect on costs. According to
440 injuries to children less than 15 above) and therefore may reduce net several producers, mattresses that use
years of age for the same period. benefits. Moreover, a small increase in existing barrier technology release total
heat that is far below the 25 MJ
9. Alternatives to the Proposed Standard 9 The lower end of the range is based on barrier requirement of TB 603. Therefore, using
Alternative Maximum Peak Heat price of one supplier, whose capacity is expected the TB 603 criterion for the total heat
Release Rate (PHRR) and Test Duration. to meet 25 to 30 percent of the whole market
released would not change costs but
demand in the short run. The next cheapest
The initial California TB 603 proposal alternative costs $24 for the barrier material alone. could potentially reduce the benefits
required the duration of the test to last 10 These cost estimates (and the resulting and, hence, the net benefits of the
60 minutes with a maximum PHRR of marginal increase) should be viewed as proposed standard.
150kW. After receiving comments on approximate since no extensive tests of the barriers Moreover, because of the small fuel
have been conducted for 60 minutes, as most
this proposal, the California Bureau of manufacturers are focused on meeting the less strict load of ticking materials currently being
Home Furnishings and Thermal requirements. Input suppliers generally do not used, the lower total heat release
Insulation changed the criterion to a assemble and test large numbers of mattresses, and requirement allows the production of
may therefore underestimate reduced labor mattress/sets based on a prototype that
maximum of 200 kW PHRR in the first productivity and/or reduced output per machine
30 minutes, the requirement for both the (compared to a maximum PHRR of 200 kW for a has not been tested so long as it differs
federal proposed standard and the 30-minute test) due to handling the thicker denser from the qualified prototype only with
current TB 603. barrier. A number of mattress producers estimate respect to ticking and the ticking
that to meet the stricter standard, manufacturing
Increasing the duration of the test and costs would increase (over those of non-compliant
material is not part of the fire resistance
reducing the PHRR would, according to mattresses) by $50 to $70 for a queen-sized set solution. Requiring a test for every
several input suppliers, increase the (Furniture/Today, July 21, 2004). prototype with a different ticking was

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2488 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

rejected because of the magnitude of the earlier effective date may result in Labeling Requirements. The
burden it would impose on small higher input costs to manufacturers. Commission could require labeling on
manufacturers who do not produce large More importantly, it is expected that mattresses to warn consumers in lieu of
numbers of any one prototype and who smaller manufacturers will be a standard. Labeling is not considered
would have been disproportionately disproportionately affected, as they are an effective option for reducing the risk
adversely affected by these more likely to wait to invest in of fires. Since mattress labels are usually
requirements. development efforts until the covered by bedclothes and may not be
Alternative Testing Requirements. technology is developed by larger firms, seen by the mattress users, labeling
The proposed standard requires or until the proposed standard becomes mattresses is likely to be an ineffective
prototype testing (of three mattress/sets) effective. A later effective date (longer means of warning consumers. Moreover,
before a manufacturer starts production than twelve months) could reduce fires started by children who cannot
of a given mattress design and a expected net benefits as more fires, read or do not change the bed sheets
confirmatory test of one mattress if more deaths, and injuries associated with will not be reduced by a labeling
than one establishment or firm are mattresses would occur between the requirement. Hence, while labeling
pooling their results. Manufacturers date of publication in the Federal costs are probably negligible, labels are
may sell a mattress/set based on a Register and the date the standard unlikely to reduce mattress fires.
prototype that has not been tested if that becomes effective. The staff is unaware Labeling of chemically treated
prototype differs from the qualified of evidence that small manufacturers components has been suggested as a
prototype only with respect to: (1) would be negatively impacted by a possible requirement of the draft
Mattress/foundation size; (2) ticking, twelve months period relative to a standard, to inform consumers of the
unless the ticking of the qualified longer period, such as eighteen or materials used. The costs of such
prototype has characteristics designed twenty-four months. The Commission is labeling would also be negligible, since
to improve the performance on the burn requesting comments from small existing mattresses have labels and
test; and/or (3) any component, businesses on the expected economic producers could probably add a
material, or method of construction that impact of the effective date and other description of the chemical treatment (if
the manufacturer can demonstrate on an requirements of the proposed rule (see any) to the existing label. Labeling of
objectively reasonable basis will not section M of this document below). chemically treated components could
cause the prototype to exceed the test Taking No Action or Relying on a provide small unquantifiable benefits to
criteria. Though production testing (i.e., Voluntary Standard. If the Commission consumers as it would provide some
random burning of mattress/sets to chose to take no action, California may additional information. However,
ensure that all production units meet attempt to enforce its standard despite because a label would only provide the
the standard) is encouraged by the the Commission General Counsel’s name of any chemical treatment without
proposed standard under quality position on preemption. Larger any information about whether the
assurance program requirements, it is producers are already moving to comply treatment has any potential health
not required. The individual with California’s standard. They also effects, it would be of little practical use
manufacturer’s decision on whether to want to avoid product liability claims for the consumer. Information on the
conduct production testing (and if so, at associated with selling mattresses with use of chemically treated components
what frequency) will clearly depend on different fire resistance in other areas of is, however, required as part of the
the efficacy of his/her quality assurance/ the country. Three of the largest four record keeping requirements of the
control efforts. producers plan to meet TB 603 standard.
As an alternative, the proposed nationwide by the end of 2005. Some
Federal standard could, like TB 603, L. Paperwork Reduction Act
small California manufacturers may
omit testing or prototype definition have a smaller incentive to meet TB 603 The proposed standard will require
requirements. Without testing, however, than a Federal standard. Small manufacturers (including importers) of
it might be difficult for manufacturers to manufacturers who do not sell in mattresses/sets to perform testing and
know whether their mattresses will California may similarly have no maintain records of their testing and
comply with the standard. incentive to sell mattresses that meet TB quality assurance efforts. For this
Alternatively, the standard could 603 requirements in other parts of the reason, the rule proposed below
require production testing with a country. Hence, expected aggregate net contains ‘‘collection of information
specified frequency. This specification, benefits associated with the draft requirements,’’ as that term is used in
however, could result in unnecessary proposed standard are higher than the the Paperwork Reduction Act, 44 U.S.C.
costs if they are not justified given the net benefits that might result under 3501–3520. Therefore, the proposed rule
quality control measures generally California TB 603 even if it could be is being submitted to the Office of
undertaken by manufacturers in the enforced in the face of preemption Management and Budget (‘‘OMB’’) in
absence of the proposed standard. concerns. accordance with 44 U.S.C. 3507(d) and
Requiring more tests per establishment, No effort has been undertaken to implementing regulations codified at 5
prototype, or enterprise will increase develop a voluntary standard. CFR 1320.11. The estimated costs of
the estimated costs per mattress and Furthermore, industry representatives these requirements are discussed below.
could reduce net benefits. support a mandatory standard to level Costs of Prototype and Confirmation
Alternative Effective Date. The the playing field among domestic Testing. According to industry
proposed effective date is twelve producers (large and small) and representatives, the cost of testing per
months from the date of publication of importers. If a voluntary standard were twin-size mattress/set may be about
the final rule in the Federal Register. developed, the economic burden would $500: the sum of the average cost of the
Given the length of time needed to fall primarily on the larger firms (who materials and shipping ($100) and the
ensure the availability of inputs for the would likely be the first to comply), cost of the use of the lab ($400). Hence,
production of barrier materials, their market shares could be reduced the cost for testing three specimens as
availability of barriers for mattress and benefits to consumers (in terms of required by the proposed rule
producers, and a sufficient volume of reduced deaths and injuries) would mattresses equals $1500. This cost is
inventories at retailers’ showrooms, an likely decline accordingly. assumed to be incurred no more than

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2489

once per establishment for each producers, whose quality control this document (see section G.2. above).
prototype. If manufacturers test every program is less detailed or non-existent, Options that manufacturers may choose
mattress construction (e.g., single-sided will incur some incremental costs as a in order to meet the proposed standard
pillow top, double-sided pillow-top, result of the proposed standard. These include one or a combination of the
tight-top, euro-top, * * * etc.), which is incremental costs will be small for each following: fire resistant ticking;
estimated, based on conversations with manufacturer and less when measured chemically treated or otherwise fire
manufacturers, to average about 20 per per mattress. (See the section on impact resistant filling products; or a fire
manufacturer, for every establishment in of the proposed standard on small blocking barrier (either a sheet style
a given year, then the estimated businesses for a description of their cost barrier or a high-loft barrier, sometimes
industry testing cost per establishment of quality control and quality assurance called a fiber barrier).
per year would approximately equal programs.) For each qualified prototype, three
$30,000. The total number of Additionally, the proposed standard mattresses/sets must be tested and must
establishments producing conventional encourages random production testing pass the test requirements. To obtain a
mattresses in 2001 was 639. Using an to assure manufacturers that their passing result, each mattress/set must
estimated number of 739 producers of mattresses continue to meet the pass a 30 minute test, where the PHRR
both conventional and unconventional requirements of the rule. Assuming that does not exceed 200 kW and the total
mattresses, the annual cost to all an average of 3 mattress/set heat release does not exceed 15 MJ in
(conventional and unconventional) constructions will be tested per the first 10 minutes of the test. A failure
mattress producers is $22.17 million. establishment per year yields an of any of the sets would require that the
Cost of information collection and estimated cost of production testing of problem be corrected and the prototype
recordkeeping. In addition to prototype about $1500. The labor needed to meet be retested and pass the test (in
testing, the proposed standard will the quality assurance measures required triplicate). Manufacturers may sell any
require detailed documentation of by the standard is estimated by CPSC mattress/foundation set based on a
prototype identification and testing Office of Compliance staff to be 224 qualified prototype. Manufacturers may
records, model and prototype minutes per establishment per qualified also sell a mattress/set based on a
specifications, inputs used, name and prototype per year. Assuming that every prototype that has not been tested if that
location of suppliers, and confirmation establishment will produce twenty prototype differs from a qualified
test record, if establishments choose to different qualified prototypes, the prototype only with respect to (1)
pool a prototype. This documentation is increase in labor costs associated with mattress/foundation size; (2) ticking,
in addition to documentation already quality assurance requirements of the unless the ticking of the qualified
conducted by mattress manufacturers in draft proposed standard is about $1904 prototype has characteristics designed
their efforts to meet the cigarette (224 minutes × 20 qualified prototypes to improve performance on the burn
standard. Detailed testing × $25.50 average civilian workers’ test; and/or (3) any component,
documentation will be done by the test compensation per hour) per material, or method of construction that
lab and is included in the estimated cost establishment per year. Hence total the manufacturer can demonstrate,
of testing. Based on CPSC Office of costs of quality assurance/quality based on an objectively reasonable
Compliance staff estimates, all control programs may average about basis, will not cause the prototype to
requirements of the proposed standard $3,404 ($1500+1904) per establishment exceed the test criteria specified above.
are expected to cost an establishment If one or more establishments (plants
per year. This translates to an annual
about 110 minutes, or 1.3 hours, per within the same firm) or independent
cost to all (conventional and
qualified prototype. Assuming that firms choose to ‘‘pool’’ prototypes, then
unconventional) mattress producers of
every establishment will produce 20 each pooling plant or firm is required to
$2,515,556 ($3,404 × 739).
different qualified prototypes, the conduct a confirmation test for one
increase in record keeping costs is about M. Initial Regulatory Flexibility mattress/set it produces locally. If that
$935 (110 minutes × 20 prototypes × Analysis set fails, then its producer cannot sell
$25.50 in average civilian workers’ mattresses based on that prototype
1. Introduction unless it successfully tests another
compensation per hour) per
establishment per year. (Note that The Regulatory Flexibility Act mattress/set after correcting its
pooling among establishments or using (‘‘RFA’’) generally requires that agencies production to make sure that it is
a prototype qualification for longer than review proposed rules for their potential identical to the original prototype. A
one year will reduce this estimate.) This economic impact on small entities, pooling firm may sell other mattresses
translates to an annual cost to all including small businesses. Section 603 that have not been tested by the pooling
(conventional and unconventional) of the RFA calls for agencies to prepare firm if they differ from the pooled
mattress producers of $690,965 ($935 × and make available for public comment prototype only with respect to (1)
739). an initial regulatory flexibility analysis mattress/foundation size; (2) ticking,
Cost of quality control/quality describing the impact of the proposed unless the ticking of the qualified
assurance programs. To ensure that all rule on small entities and identifying prototype has characteristics designed
mattresses are produced to the impact-reducing alternatives. to improve performance on the burn
prototype specification across all Accordingly, staff prepared an initial test; and/or (3) any component,
factories and over the years for which a regulatory flexibility analysis for the material, or method of construction that
production line exists, mattress mattress proposed rule. 9 A summary of the manufacturer can demonstrate,
manufacturers will need a thorough that analysis follows. based on an objectively reasonable
well-documented quality control/ basis, will not cause the prototype to
assurance program. The top 12 mattress 2. Impact on Small Businesses and exceed the test criteria specified above.
producers (with a market share of Other Small Entities Manufacturers are required to keep
almost 80 percent) have a existing Summary of proposed requirements. records of all tests performed and their
quality control programs which could The proposed standard will apply to results, including video or pictures,
be modified to fit the new standard with all mattresses and mattress and prototype identification number, date
minimal additional costs. Smaller foundation sets, as discussed earlier in and time of test, name and location or

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2490 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

testing facility for as long as the been tested extensively and is known to 3. Alternatives and Their Possible Effect
prototype is in production and for three meet the standard. The price smaller on Small Businesses
years after its production ceases. firms pay to cover the development and Alternatives considered by the
Manufacturers are also required to keep testing costs borne by the supplier will Commission are discussed in the
records of a unique identification not disproportionately impact them, Preliminary Regulatory Analysis section
number for the qualified prototype and because it is not measured relative to of this preamble (Section K). As
a list of the unique identification their small output, but relative to the discussed therein, increasing the
numbers of all other prototypes based supplier’s output. Other smaller firms duration of the test and reducing the
on the qualified prototype, together with may combine their development efforts PHRR would increase costs without
a description of the material substituted to be able to benefit from dividing the necessarily increasing benefits. This
and/or size change. Moreover, they are costs over a larger number of firms. would necessitate an increase in costs.
required to document the name and Finally, small mattress producers who Staff estimates the marginal increase in
supplier of each material used in
do not assemble the mattress panels, but costs to be $3.96 to $11.88 over the costs
construction and keep physical samples
buy them from a panel supplier are of the proposed standard. Although a
of the material. Additionally, they are
effectively acting as a large producer by stricter standard might increase benefits,
required to identify the details of the
combining all their output. This is any increase is likely to be small since
application of any flame retardant
because the panel supplier will be the proposed standard has an
treatments and/or inherently flame
responsible for including a barrier in the effectiveness rate of 80 percent.
retardant fibers employed relative to
panel assembly and will pass that cost An increase in costs would likely
mattress components. Finally, they are
on to the mattress producers, again not result in an increase in mattress prices.
required to have an adequate quality
disproportionately impacting the small A bedding official estimated that such
assurance program in place.
Impact on small businesses. The producers who buy the already price increases may result in reduction
proposed standard covers manufacturers assembled panels. in sales of 25% or more (Furniture/
and importers of mattresses. There were Today, July 21, 2004). The larger
The cost imposed disproportionately increase in prices (compared to the less
557 mattress firms and 639 mattress (per unit produced) on small businesses
establishments in 2001, according to the strict test) and the resulting reduction in
will be the cost of testing, information sales could drive some of the smaller
Statistics of U.S. businesses, Census collection and record keeping, and
Bureau data. All but the largest twelve producers out of business. (A stricter
quality control/quality assurance standard would be more likely to
firms had less than 500 employees. The
programs. While the regulatory analysis require replacing some existing
U.S. Small Business Administration’s
estimates these costs to be a little over machines, to accommodate the denser
Office of Advocacy defines a small
business as one that is independently one dollar per mattress per year for barrier material, which would be
owned and operated and not dominant average-sized establishments, they disproportionately more costly for
in its fields. A definition that is used could be substantially higher for small smaller firms, whose machinery is older
frequently and is less subject to mattress producers. If manufacturers use and less sophisticated.) Since mattresses
interpretation is a firm with fewer than a prototype qualification to produce are durable goods, one would expect a
500 employees. The latter definition mattress/set constructions for longer larger drop in sales in the short-run than
classifies 97.8 percent ((557 ¥ 12)/557) than a year, or if they use a worst-case in the long-run, as consumers choose to
of all mattress firms as small businesses. prototype to represent other mattress keep their old mattresses longer than
Average employment per firm for the constructions, these costs will be lower. before. This would make the reduction
whole industry is 45.8 employees. Furthermore, firms with more than one in sales more pronounced in the short-
Average employment for the 1–4 establishment may be able to reduce run, increasing the likelihood that some
employees per enterprise group, which these costs by pooling their testing and firms may exit the market. Moreover, if
represents 22.98 percent of all firms, is quality control programs over all a large number of consumers choose to
2.4 employees. Average employment for establishments. Small independent extend the life of their mattresses for a
the fewer than twenty employees per firms could also pool their testing to longer time period, it will take longer to
enterprise group, which represents reduce their costs per mattress. achieve the benefits expected to be
61.22 percent of all firms, is 6.2 associated with the safer mattresses.
Use of pooling across establishments As discussed in the preliminary
employees. Hence more than half of
and firms would ameliorate the impact regulatory analysis, the Commission
mattress firms have fewer than twenty
of the proposed standard on small also considered a different criterion for
employees.
In addition to domestic producers, businesses. By getting together across the total heat released during the first 10
importers will be affected by the different states and regions, small minutes of the test, i.e., 25 MJ instead
proposed standard. Imported mattresses manufacturers who do not share a of the 15 MJ the Commission is
represent less than two percent of total common market (and therefore do not proposing. Using the 25 MJ criterion
U.S. shipments. compete with each other) can resemble would likely reduce estimated benefits
The increase in material and labor a large manufacturer in their testing and (the estimated reductions in deaths and
costs of the proposed standard quality control/quality assurance efforts injuries), without having any significant
(estimated to be $10.41 to $38.80 per and therefore reduce their costs per effect on costs. According to several
mattress) is not likely to be dependent mattress. It is also expected that some producers, mattresses that use existing
on a firm’s size and will therefore not barrier suppliers would be willing to do barrier technology release total heat that
adversely affect small businesses. Larger the testing and quality control/ is far below the 25 MJ level.
firms are bearing all the capital assurance programs for small Moreover, because of the small fuel
investment costs of research and manufacturers in exchange for a small load of ticking materials currently being
development, sharing some of these charge, which will be similar to the used, the lower total heat release
costs with input suppliers. Most smaller average cost per mattress for large requirement allows the production of
firms are waiting to buy from the businesses, because the volume of mattress/sets based on a prototype that
suppliers a barrier solution, which has output will be large. has not been tested so long as it differs

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2491

from the qualified prototype only with Another possible option would be to • Twelve month effective date
respect to ticking and the ticking is not require labeling on mattresses to warn Also, it would be most useful to
part of the fire resistance solution. consumers in lieu of a standard. receive comments on ways in which the
Requiring a test for every prototype with However, as discussed in the proposal could be modified to reduce
a different ticking was rejected because Preliminary Regulatory Analysis, any costs or burdens for small entities,
of the magnitude of the burden it would labeling is not likely to be effective at and whether and how technological
impose on small producers who do not reducing mattress-related deaths and developments could reduce the costs for
produce large numbers of any one injuries. small entities of complying with the
prototype and would have been A final possible option might be to set rule.
adversely affected by these a later effective date for small
N. Environmental Considerations
requirements. companies. If needed, this could allow
The Commission also considered smaller companies more time to prepare General. Usually, CPSC rules
alternative testing requirements. The for the standard since they are less establishing performance requirements
proposed standard requires prototype likely to be currently preparing for are considered to ‘‘have little or no
California’s similar TB 603. However, potential for affecting the human
testing (of three specimens of mattress/
the Commission has no evidence at this environment,’’ and environmental
sets) before a manufacturer starts
time that a split effective date is assessments are not usually prepared for
production of a given mattress design
necessary. these rules (see 16 CFR 1021.5 (c)(1)).
and a confirmatory test of one mattress
However, in order to meet this standard,
if a firm is producing a mattress based 4. Conclusion many manufacturers will need to
on a prototype produced by another
Almost all mattress firms would be change some materials that they use to
manufacturer in a pooling arrangement.
considered small businesses, using the manufacture mattresses: either using
Although production testing (i.e.
Small Business Administration more inherently flame resistant
burning mattress/sets to ensure that
definition. Material and labor costs for materials or incorporating flame
production units meet the standard) is retardant (FR) chemicals into their
all firms are expected to initially
encouraged by the proposed standard products. Therefore, the Commission
increase on average by $10–$39 dollars
under quality assurance program concluded that a more thorough
per mattress set produced. These cost
requirements, it is not required. The consideration of the potential for
increases are expected to be borne
individual manufacturer’s decision on environmental impacts is warranted.
equally by all firms and hence do not
the need for and frequency of The staff’s analysis contained in the
have an adverse impact on the smaller
production testing will clearly depend memorandum ‘‘Preliminary
mattress producers. These costs are
on the efficacy of its quality assurance/ Environmental Assessment of a Draft
expected to decline in the future due to
control efforts. Proposed Open-Flame Ignition
improved technology of producing fire
As an alternative, the Federal Resistance Standard for Mattresses,’’ [7]
retardant materials and increase
standard could, like TB 603, omit concludes that since the proposed
competition among input suppliers.
testing requirements. However, without Although testing and recordkeeping standard states performance
testing, it might be difficult for requirements may have a requirements, manufacturers will have
manufacturers to know whether their disproportionate impact on small several options for meeting the
mattresses will comply with the manufacturers, the proposed standard requirements of the proposed standard.
standard. Alternatively, the standard allows manufacturers to pool test Although there are still some unsettled
could require production testing with a results, to vary their tickings without questions, there appear to be numerous
specified frequency. This specification, new prototype testing (unless the promising methods that manufacturers
however, could result in unnecessary ticking had characteristics designed to could use without posing an
costs if they are not justified given the improve performance on the specified unacceptable health risk to consumers
quality control measures generally mattress test), and to make other or significantly affecting the
undertaken by producers in the absence changes in their prototype without new environment. Moreover, even if a
of the proposed standard. Requiring prototype testing if the change does not chemical used by some manufacturers
more tests per establishment, prototype, negatively effect the mattress’s ability to were shown to pose an unacceptable
or enterprise will increase the estimated meet the test criteria. These options risk to human health or the
costs per mattress and could reduce net should minimize burdens on small environment, there would be various
benefits. businesses. regulatory and other mechanisms that
The Commission also could have The Commission requests comments could be used to remove the chemical
chosen to take no action. In this on any or all of the provisions in the from applications where it poses a risk.
situation, the larger producers would proposed rule with regard to : (1) The Possible approaches to meet the
probably follow TB 603 for all their impact of the provisions (including any proposed standard. The standard does
mattresses, not just those sold in benefits and costs), if any, on small not prescribe the means that
California, in order to avoid product entities and (2) what alternatives, if any, manufacturers must use to meet the
liability claims. Some small California the Commission should consider, as standard. The staff expects, however,
manufacturers may decline to meet TB well as the costs and benefits of those that most manufacturers will use some
603 on the basis that it is preempted by alternatives to small entities in light of kind of flame resistant barrier to protect
the existing federal standard. Small the above analysis. The Commission is the mattress components with the
manufacturers who do not sell in particularly interested in information greatest combustible fuel loads from the
California may have no incentive to with regard to the impact of the flames. These barriers may be fabric,
meet TB 603 requirements throughout following aspects of the proposed rule: batting, or other materials that are either
the country. Hence, expected aggregate • Prototype and confirmation testing inherently flame resistant or that have
net benefits associated with the draft requirements been treated with flame retardant
proposed standard are higher than the • Quality Control/quality assurance chemicals. [6&7]
net benefits that might result under program requirements Because manufacturers are now
California TB 603. • Recordkeeping requirements evaluating their alternatives, the staff

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2492 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

does not know the methods that each workers involved in manufacturing analysis concludes that there are FR
manufacturer will use to meet the protective apparel, carpets, and chemicals and flame resistant materials
proposed standard. Therefore, the staff’s transportation upholstery may already that, based on currently available data,
analysis attempts to provide some be exposed to these chemicals as are the are not expected to pose unacceptable
context for considering the consumers of the products. Some of risks to the environment and that are
environmental impacts of the standard. these FR chemicals and materials may widely used in other applications. [7]
More definitive conclusions should be already be used in mattress and bedding In accordance with the National
possible as more information applications. For example, boric acid is Environmental Policy Act (‘‘NEPA’’),
concerning the methods that already used to treat cotton batting in the Executive Director of CPSC has
manufacturers will use to meet the mattresses and futons. [6&7] issued a Finding of No Significant
standard becomes available. Possible regulatory protections. Some Impact (‘‘FONSI’’) for the proposed
How the proposed standard could chemicals that have been used for their mattress standard. The FONSI is based
affect the environment. About 25 fire resistant properties have been on the staff’s Environmental
million mattresses are sold annually, determined to have unacceptable Assessment, which has been
and most will probably require some adverse impacts on health and the summarized above. The FONSI
changes in materials used or environment in some applications (e.g., concludes that there will be no
construction to meet the standard. TRIS (2,3,-dibromopropyl) phosphate, significant impacts on the quality of the
These changes, such as the pentabromodiphenyl oxide (‘‘PBDPO’’) human environment as a result of the
incorporation of a flame resistant barrier and octabromodiphenyl oxide proposed mattress flammability
or other materials, will increase the (‘‘BDPO’’)). [7] standard. The Commission requests
manufacture of fire resistant materials or The U.S. Environmental Protection comments on both the Environmental
FR chemicals. This could mean Agency (EPA) has the authority to Assessment and the FONSI.11
increased exposure to such chemicals regulate the use of toxic chemicals
under the Toxic Substances Control Act O. Executive Order 12988
for workers and consumers.
Additionally, at the end of their useful (TSCA) (15 U.S.C. 2601 et seq.). EPA According to Executive Order 12988
lives, the mattresses/sets will be also monitors and promotes research (February 5, 1996), agencies must state
disposed of. Potential environmental into potential toxic or environmental the preemptive effect, if any, of new
impacts will vary depending on the effects of chemicals which it believes regulations.
method the manufacturer used to meet could pose environmental risks. With The FFA provides that, generally, if
the standard and the potential for the regard to flame retardants, the EPA is the Commission issues a flammability
particular FR chemicals used to persist developing a significant new use rule standard for a fabric, related material or
in the environment. [7] (SNUR), under section 5(a)(2) of TSCA, product under the FFA, ‘‘no State or
FR chemicals widely in use, but new which is expected to cover the use of political subdivision of a State may
applications possible. Many FR several flame retardants in residential establish or continue in effect a
chemicals are widely used. In the U.S., upholstered furniture. A SNUR would flammability standard or other
the consumption of flame retardant require chemical manufacturers and regulation for such fabric, related
chemicals is estimated to be over 1 importers to report scientific data to the material or product if the standard or
billion pounds annually and is EPA so that EPA may determine other regulation is designed to protect
increasing. This includes various fire whether controls on the use of the against the same risk of the occurrence
retardant chemicals based on bromine, chemical may be warranted. There is of fire with respect to which the
antimony, chlorine, phosphorous, expected to be some overlap between standard or other regulation under this
nitrogen, and boron. Additionally, there the flame retardants that will be covered Act is in effect unless the State or
are some fibers where the FR chemical by the SNUR for use in upholstered political subdivision standard or other
is incorporated into the polymer of the furniture and flame retardants that can regulation is identical to the Federal
fiber itself or that are inherently fire be used in mattresses. standard other regulation.’’ 15 U.S.C.
resistant. These include some Decabromodiphenyl oxide (DBDPO), for 1203(a). Upon application to the
modacrylic, melamine, and para-aramid example, can be used as a backcoating Commission, a State or political
fibers. in upholstery fabric or on fire resistant subdivision of a State may be exempted
Because the chemicals and materials barriers for mattresses. Additional from this preemptive effect if
that would be used to meet a mattress activities by EPA, The National compliance with the State or political
standard are already being used in other Toxicology Program (NTP) of the subdivision requirement would not
applications, the manufacture of these Department of Health and Human cause the fabric, related material or
materials will not create new impacts, Services (DHHS), and the Occupational product to be in violation of any FFA
though it could intensify effects that are Safety and Health Administration standard or regulation, and the State or
already occurring. A mattress (OSHA) can provide information about political subdivision’s standard (1)
flammability standard could result in any adverse health effects of FR provides a significantly higher degree of
some FR chemicals or flame resistant chemicals and take actions to limit their protection from the risk of occurrence of
materials being used in applications use if necessary. [6&7] Work by the fire than the FFA standard and (2) does
where they have not been used before. National Research Council of the not unduly burden interstate commerce.
This would result in some new National Academy of Sciences on Id. 1203(c)(1). In addition, the Federal
exposure patterns for these materials. selected flame-retardant chemicals is an government, or a State or local
For example, workers in mattress additional source of information on government, may establish and continue
factories could be exposed to the these FR chemicals. [16] in effect a non-identical flammability
chemicals as could the ultimate Conclusion. The staff’s environmental standard or other regulation for the
consumers. However, these new analysis examines some of the methods
11 Both of these documents are available from the
exposure patterns may be similar to manufacturers might use to meet the
Commission’s Office of the Secretary (see
ones that are already occurring since proposed standard and discusses what ‘‘Addresses’’ section above) or from the
these chemicals are widely used in is known about their potential toxicity Commission’s web site (http://www.cpsc.gov/
other applications. For example, and possible environmental impact. The library/foia/foia.html).

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2493

Federal, State or local government’s should allow sufficient time for aggregate lifetime net benefits for all
own use if it provides a higher degree manufacturers to develop products for mattresses produced in the first year of
of protection than the FFA standard. Id. nationwide markets that will meet the the standard of $450 to $1,560 million
1203(b). Thus, with the exceptions proposed requirements. The from the standard. Thus, the
noted above, the proposed open flame Commission requests comments, Commission preliminarily finds that the
standard for mattresses would preempt especially from small businesses on the standard is needed to adequately protect
non-identical state or local mattress proposed effective date and the impact the public from the unreasonable risk of
flammability standards designed to it would have on them. the occurrence of fire.
protect against the same risk of the The standard is reasonable,
Q. Proposed Findings technologically practicable, and
occurrence of fire.
The issue of preemption has been Section 1193(a) and (j)(2) of the FFA appropriate. Through extensive research
raised with regard to the proposed require the Commission to make certain and testing, NIST developed a test
standard because of California’s TB 603. findings when it issues a flammability method to assess the flammability of
In a letter to the Bureau Chief of standard. The Commission must find mattresses ignited by an open flame.
California’s Bureau of Home that the standard: (1) Is needed to The test method represents the typical
Furnishings and Thermal Insulation, adequately protect the public against the scenario of burning bedclothes igniting
dated April 9, 2003, the Commission’s risk of the occurrence of fire leading to a mattress. Based on NIST’s testing, the
General Counsel has taken the position death, injury or significant property standard establishes criteria that will
that CPSC’s existing federal Standard for damage; (2) is reasonable, reduce the fire intensity of a burning
the Flammability of Mattresses (16 CFR technologically practicable, and mattress, allowing more time for
1632) preempts California’s TB 603. appropriate; (3) is limited to fabrics, occupants to escape before flashover
That conclusion was based on related materials or products which occurs. NIST testing has also
legislative history and CPSC’s General present unreasonable risks; and (4) is demonstrated that mattresses can be
Counsel Advisory Opinion 289 (Dec. 8, stated in objective terms. Id. 1193(b). In constructed with available materials and
1983) indicating that if federal and state addition, the Commission must find construction that will meet the test
requirements are both designed to that: (1) If an applicable voluntary criteria. Therefore, the Commission
address the same risk (i.e., the standard has been adopted and finds that the standard is reasonable,
occurrence of fire), the federal standard implemented, that compliance with the technologically practicable, and
will have preemptive effect even if the voluntary standard is not likely to appropriate.
two standards use different ignition adequately reduce the risk of injury, or The standard is limited to fabrics,
sources. compliance with the voluntary standard related materials, and products that
Legislative history of the FFA’s is not likely to be substantial; (2) that present an unreasonable risk. The
preemption provision states: benefits expected from the regulation standard applies to mattresses and
bear a reasonable relationship to its mattress and foundation sets. It is a
[A] State standard designed to protect costs; and (3) that the regulation performance standard. Thus, it neither
against the risk of injury from a fabric imposes the least burdensome requires nor restricts the use of
catching on fire would be preempted by a
requirement that would prevent or particular fabrics, related materials or
Federal flammability standard covering the
same fabric even though the Federal adequately reduce the risk of injury. The products. Manufacturers may choose the
flammability standard called for tests using last three findings must be included in materials and methods of construction
matches and the State standard called for the regulation. Id. 1193(j)(2). These that they believe will best suit their
tests using cigarettes. When an item is findings are discussed below. business and result in mattresses that
covered by a Federal flammability standard, The standard is needed to adequately can meet the specified test criteria. As
* * * a different State or local flammability protect the public against unreasonable discussed above, the Commission
requirement applicable to the same item will risk of the occurrence of fire. National concludes that current mattresses
be preempted since both are designed to fire loss estimates indicate that present an unreasonable risk. Therefore,
protect against the same risk, that is the mattresses and bedding were the first the Commission finds that the standard
occurrence of or injury from fire. items to ignite in 19,400 residential fires is limited to fabrics, related materials,
H.R. Rep. No. 1022, 94th Cong., 2d Sess. attended by the fire service annually and products that present an
29 (1976). The Commission believes that during 1995–1999. These fires resulted unreasonable risk.
this legislative history indicates that the in 440 deaths, 2,230 injuries and $273.9 Voluntary standards. The
proposed standard would preempt non- million in property loss each year. Of Commission is not aware of any
identical state requirements addressing these, the staff considers an estimated voluntary standard in existence that
the flammability of mattresses. 18,500 fires, 440 deaths, 2,160 injuries, adequately and appropriately addresses
and $259.5 million property loss the specific risk of injury addressed by
P. Effective Date
annually to be addressable by the this standard. Thus, no findings
The Commission proposes that the proposed standard. The Commission concerning compliance with and
rule would become effective one year estimates that the standard will prevent adequacy of voluntary standards are
from publication of a final rule in the 80 to 86 percent of deaths and 86 to 92 necessary.
Federal Register and would apply to percent of the injuries occurring with Relationship of Benefits to Costs. The
mattresses entering the chain of these addressable mattress/bedding Commission estimates that the total
distribution on or after that date. The fires. Thus, the Commission estimates lifetime benefits of a mattress complying
Commission is aware that many that when all mattresses have been with this standard will range from $62
mattress manufacturers are modifying replaced by ones that comply with the to $74 per mattress (based on a 10 year
their products to comply with standard, 310 to 330 deaths and 1,660 mattress life and 3% discount rate). The
California’s TB 603 which prescribes to 1,780 injuries will be avoided Commission estimates that total
requirements that are similar to this annually as a result of the standard. resource costs of the standard will range
proposed rule and will become effective The regulatory analysis explains that from $13 to $44 per mattress. This
January 1, 2005. Thus, the Commission the Commission estimates lifetime net yields net benefits of $18 to $62 per
believes that a one-year effective date benefits of $18 to $62 per mattress or mattress. The Commission estimates

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2494 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

that aggregate lifetime benefits 1633.2 Definitions. subject to the requirements of the
associated with all mattresses produced 1633.3 General requirements. standard.
the first year the standard becomes 1633.4 Prototype testing requirements. (2) One-of-a-kind mattresses and
1633.5 Prototype pooling and confirmation foundations may be exempted from
effective range from $1,560 to $1,880
testing requirements.
million, and that aggregate resource 1633.6 Quality assurance requirements.
testing under this standard in
costs associated with these mattresses 1633.7 Mattress test procedure. accordance with § 1633.13(c).
range from $320 to $1,110 million, 1633.8 Findings. (c) Applicability. The requirements of
yielding net benefits of about $450 to 1633.9 Glossary of terms. this part 1633 shall apply to each
$1,560 million. Therefore, the ‘‘manufacturer’’ (as that term is defined
Subpart B—Rules and Regulations in § 1633.2(i)) of mattresses and/or
Commission finds that the benefits from
the regulation bear a reasonable 1633.10 Definitions. mattress and foundation sets which are
1633.11 Records. manufactured for sale in commerce.
relationship to its costs.
1633.12 Labeling.
Least burdensome requirement. The 1633.13 Tests for guaranty purposes, § 1633.2 Definitions.
Commission considered the following compliance with this section, and ‘‘one
alternatives: alternative maximum peak In addition to the definitions given in
of a kind’’ exemption.
heat release rate and test duration, section 2 of the Flammable Fabrics Act
alternative total heat released in the first Subpart C—Interpretations and Policies as amended (15 U.S.C. 1191), the
10 minutes of the test, mandatory 1633.14 Policy clarification on renovation following definitions apply for purposes
production testing, a longer effective of mattresses. of this part 1633.
Figure 1 to Part 1633—Test Assembly, (a) Mattress means a resilient material
date, taking no action, relying on a
Shown in Furniture Calorimeter or combination of materials enclosed by
voluntary standard, and requiring (Configuration A) a ticking (used alone or in combination
labeling alone. As discussed in the Figure 2 to Part 1633—Test Arrangement in with other products) intended or
preamble above and the regulatory 3.05m × 3.66m (10 ft × 12 ft) Room promoted for sleeping upon.
analysis, these alternatives are expected (Configuration B) (1) This term includes, but is not
to increase costs without increasing Figure 3 to Part 1633—Details of Horizontal
limited to, adult mattresses, youth
benefits, or significantly reduce the Burner Head
Figure 4 to Part 1633—Details of Vertical mattresses, crib mattresses (including
benefits expected from the rule. portable crib mattresses), bunk bed
Therefore, the Commission finds that Burner Head
Figure 5 to Part 1633—Details of Burner mattresses, futons, flip chairs without a
the standard imposes the least permanent back or arms, sleeper chairs,
Stand-off
burdensome requirement that would Figure 6 to Part 1633—Burner Assembly and water beds or air mattresses if they
adequately reduce the risk. Showing Arms and Pivots (Shoulder contain upholstery material between the
R. Conclusion Screws), in Relation to, Portable Frame ticking and the mattress core. Mattresses
Allowing Burner Height Adjustment used in or as part of upholstered
For the reasons stated in this Figure 7 to Part 1633—Elements of Propane
preamble, the Commission preliminarily furniture are also included; examples
Flow Control for Each Burner
finds that an open flame flammability Figure 8 to Part 1633—Jig for Setting are convertible sofa bed mattresses,
standard for mattresses and mattress Mattresses and Foundation Sides in corner group mattresses, day bed
and foundation sets is needed to Same Plane mattresses, roll-away bed mattresses,
adequately protect the public against the Figure 9 to Part 1633—Burner Placements on high risers, and trundle bed mattresses.
unreasonable risk of the occurrence of
Mattress/Foundation See § 1633.9 Glossary of terms, for
Figure 10 to Part 1633—Jig for Setting definitions of these items.
fire leading to death, injury, and Burners at Proper Distances from (2) This term excludes mattress pads,
significant property damage. The Mattress/Foundation mattress toppers (items with resilient
Commission also preliminarily finds Figure 11 to Part 1633—Diagrams for filling, with or without ticking, intended
that the standard is reasonable, Glossary of Terms
Appendix A to Part 1633—Calibration of
to be used with or on top of a mattress),
technologically practicable, and
Propane Flowmeters sleeping bags, pillows, liquid and
appropriate. The Commission further
Appendix B to Part 1633—Burner Operation gaseous filled tickings, such as water
finds that the standard is limited to the
Sequence beds and air mattresses that contain no
fabrics, related materials and products
Authority: 15 U.S.C. 1193, 1194.
upholstery material between the ticking
which present such unreasonable risks.
and the mattress core, upholstered
List of Subjects in 16 CFR Part 1633 Subpart A—The Standard furniture which does not contain a
Consumer protection, Flammable mattress, and juvenile product pads
§ 1633.1 Purpose, scope, and applicability. such as car bed pads, carriage pads,
materials, Labeling, Mattresses and
mattress pads, Records, Textiles, (a) Purpose. This Part 1633 establishes basket pads, infant carrier and lounge
Warranties. flammability requirements that all pads, dressing table pads, stroller pads,
For the reasons stated in the mattress and mattress and foundation crib bumpers, and playpen pads. See
preamble, the Commission proposes to sets must meet before sale or § 1633.9 Glossary of terms, for
amend Title 16 of the Code of Federal introduction into commerce. The definitions of these items.
Regulations by adding a new part 1633 purpose of the standard is to reduce (b) Foundation means a ticking
to read as follows: deaths and injuries associated with covered structure used to support a
mattress fires by limiting the size of the mattress or sleep surface. The structure
PART 1633—STANDARD FOR THE fire generated by a mattress or mattress may include constructed frames, foam,
FLAMMABILITY (OPEN-FLAME) OF and foundation set during a thirty box springs, or other materials, used
MATTRESSES and MATTRESS AND minute test. alone or in combination.
FOUNDATION SETS (b) Scope. (1) All mattresses and all (c) Ticking means the outermost layer
mattress and foundation sets, as defined of fabric or related material of a mattress
Subpart A—The Standard in § 1633.2(a) and § 1633.2(b), of any or foundation. It does not include any
Sec. size, manufactured or imported after other layers of fabric or related materials
1633.1 Purpose, scope and applicability. [the effective date of this standard] are quilted together with, or otherwise

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2495

attached to, the outermost layer of fabric mattress, such as springs, foam, water (b) Notwithstanding the requirements
or related material. bladder, air bladder, or resilient filling. of paragraph (a) of this section, a
(d) Upholstery material means all manufacturer may sell or introduce into
material, either loose or attached, § 1633.3 General requirements. commerce a mattress or mattress and
between the mattress ticking and the (a) Summary of test method. The test foundation set based on a prototype that
core of a mattress, if a core is present. method set forth in § 1633.7 measures has not been tested according to
(e) Edge seam means the seam or the flammability (fire test response § 1633.3(b) if that prototype differs from
border edge of a mattress or foundation characteristics) of a mattress specimen a qualified prototype only with respect
that joins the top and/or bottom with by exposing the specimen to a specified to:
the side panels. flaming ignition source and allowing it (1) Mattress/foundation size (e.g.,
(f) Tape edge means an edge seam to burn freely under well-ventilated, twin, queen, king);
made by using binding tape to encase controlled environmental conditions. (2) Ticking, unless the ticking of the
and finish raw edges. The flaming ignition source shall be a qualified prototype has characteristics
(g) Binding tape means a fabric strip pair of propane burners. These burners (such as chemical treatment or special
used in the construction of some edge impose differing fluxes for differing fiber composition) designed to improve
seams. times on the top and sides of the performance on the test prescribed in
(h) Seam thread means the thread specimen. During and after this this part; and/or
used to form stitches in construction exposure, measurements shall be made (3) The manufacturer can
features, seams, and tape edges. of the time-dependent heat release rate demonstrate, on an objectively
(i) Manufacturer means an individual from the specimen, quantifying the reasonable basis, that a change in any
plant or factory at which mattresses energy generated by the fire. The rate of component, material, or method of
and/or mattress and foundation sets are heat release must be measured by means construction will not cause the
manufactured or assembled. For of oxygen consumption calorimetry. prototype to exceed the test criteria
purposes of this Part 1633, an importer (b) Test criteria. When testing the specified in § 1633.3(b).
is considered a manufacturer. mattress or mattress and foundation set (c) All tests must be conducted on
(j) Prototype means a specific design in accordance with the test procedure specimens that are no smaller than a
of mattress and corresponding set forth in § 1633.7, the specimen shall twin size, unless the largest size
foundation, if any, which, except as comply with both of the following mattress or mattress and foundation set
permitted by § 1633.4(b), is the same in criteria: produced is smaller than a twin size, in
all material respects as, and serves as a (1) The peak rate of heat release shall which case the largest size must be
model for, production units intended to not exceed 200 kilowatts (‘‘kW’’) at any tested.
be introduced into commerce. time within the 30 minute test; and (d)(1) If each of the three specimens
(k) Prototype pooling means a (2) The total heat release shall not meets both the criteria specified in
cooperative arrangement whereby one exceed 15 megajoules (‘‘MJ’’) for the first § 1633.3(b), the prototype shall be
or more manufacturers may rely on a 10 minutes of the test. In the interest of qualified. If any one (1) specimen fails
prototype produced by a different safety, the test operator should to meet the test criteria of § 1633.3(b),
manufacturer. discontinue the test and record a failure the prototype is not qualified.
(l) Production lot means any quantity if a fire develops to such a size as to (2) Any manufacturer may produce
of finished mattresses or mattress and require suppression for the safety of the mattresses and foundations, if any, for
foundation sets that are produced in a facility. sale in reliance on prototype tests
production interval defined by the (c) Testing of mattress and performed before [the effective date of
manufacturer, and are intended to corresponding foundation. Mattresses to this Standard], provided that such tests
replicate a specific prototype that be offered for sale with a foundation were conducted in accordance with all
complies with this part 1633. shall be tested with that foundation. requirements of this section and
(m) Confirmation test means a pre- Mattresses to be offered for sale without § 1633.7 and yielded passing results
market test conducted by a a foundation shall be tested alone. according to the test criteria of
manufacturer that is relying on a pooled (d) Compliance with this standard. § 1633.3(b).
prototype produced by another Each mattress or mattress and
manufacturer. A confirmation test must foundation set sold or introduced into § 1633.5 Prototype pooling and
be conducted in accordance with the commerce after [the effective date of this confirmation testing requirements.
procedures set forth in § 1633.7 to standard] shall meet the test criteria (a) Prototype pooling. One or more
confirm that the manufacturer can specified in paragraph (b) of this section manufacturers may rely on a prototype
produce a mattress and corresponding and otherwise comply with all produced by another manufacturer
foundation, if any, that is identical to applicable requirements of this part provided that:
the prototype in all material respects. 1633. (1) The prototype meets the
(n) Specimen means a mattress and requirements of § 1633.4; and
corresponding foundation, if any, tested § 1633.4 Prototype testing requirements. (2) The mattresses or mattress and
under this part. (a) Except as otherwise provided in foundation sets being produced based
(o) Twin size means any mattress with paragraph (b) of this section, each on the prototype have components,
the dimensions 38 inches (in) (96.5 manufacturer shall cause three materials, and methods of construction
centimeters (cm)) x 74.5 in. (189.2 cm), specimens of each prototype to be tested that are identical in all material respects
all dimensions may vary by ± 1⁄2 in. (± according to § 1633.7 and obtain passing to the prototype except as otherwise
1.3 cm) test results according to § 1633.3(b) permitted by § 1633.4(b).
(p) Qualified prototype means a before selling or introducing into (b) Confirmation testing. Any
prototype that has been tested in commerce any mattress or mattress and manufacturer (‘‘Manufacturer B’’)
accordance with § 1633.4(a) and meets foundation set based on that prototype, producing mattresses or mattress and
the criteria stated in § 1633.3(b). unless the manufacturer complies with foundation sets in reliance on a
(q) Core means the main support the prototype pooling and confirmation prototype produced by another
system that may be present in a testing requirements in § 1633.5. manufacturer (‘‘Manufacturer A’’) shall

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2496 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

cause to be tested in accordance with forth in § 1633.3(b), the manufacturer exiting from the room is caught by a
§ 1633.7 at least one (1) specimen shall cease production and distribution hood system instrumented for heat
produced by Manufacturer B of each in commerce of such mattresses and/or release rate measurements. The room
prototype of Manufacturer A upon mattress and foundation sets until shall have no openings permitting air
which said Manufacturer B is relying. corrective action is taken. infiltration other than a doorway
The tested specimen must meet the (2) Corrective actions. A manufacturer opening 0.97 m ± 6.4 mm by 2.03 m ±
criteria under § 1633.3(b) before must take corrective action when any 6.4 mm (38 in by 80 in) located as
Manufacturer B may sell or introduce mattress or mattress and foundation set indicated in Figure 2 of this part and
any mattresses or mattress and is manufactured or imported for sale other small openings as necessary to
foundation sets based on the pooled fails to meet the flammability test make measurements. Construct the test
prototype. criteria set forth in § 1633.3(b). room of wood or metal studs and line
(c) Confirmation test failure. (1) If the it with fire-rated wallboard or calcium
§ 1633.7 Mattress test procedure.
confirmation test specimen fails to meet silicate board. Position an exhaust hood
the criteria of § 1633.3(b), the (a) Apparatus and test materials (1) outside of the doorway so as to collect
manufacturer thereof shall not sell any Calorimetry. The rate of heat release all of the combustion gases. There shall
mattress or mattress and foundation set must be measured by means of oxygen be no obstructions in the air supply to
based on the same prototype until that consumption calorimetry. The the set-up.
manufacturer takes corrective measures, calibration should follow generally (2) Location of test specimen. The
tests a new specimen, and the new accepted practices for calibration. The location of the test specimen is shown
specimen meets the criteria of calorimetry system shall be calibrated at in Figure 2 of this part. The angled
a minimum of two (2) calibration points, placement is intended to minimize the
§ 1633.3(b).
(2) If a confirmation test specimen at 75 kW and 200 kW. interaction of flames on the side
(2) Testroom. The testroom must have surfaces of the test specimen with the
fails to meet the criteria of § 1633.3(b),
either Test Configuration A or B. room walls. One corner of the test
the manufacturer thereof must notify the (i) Test Configuration A. (an open
manufacturer of the prototype of the test specimen shall be 13 centimeters (cm)
calorimeter (or furniture calorimeter)). to 17 cm from the wall and the other
failure. In this configuration, the specimen to be corner shall be 25 cm to 30 cm from the
§ 1633.6 Quality assurance requirements. tested is placed under the center of an wall. The test room shall contain no
(a) Quality assurance. Each open furniture calorimeter. Figure 1 of other furnishings or combustible
manufacturer shall implement a quality this part shows the test assembly atop materials except for the test specimen.
assurance program to ensure that a bedframe and catch surface. The (3) Bed frame. For twin size
mattresses and mattress and foundation specimen shall be placed under an open mattresses, the specimen shall be placed
sets manufactured for sale are identical hood which captures the entire smoke on top of a welded bed frame (1.90 m
in all material respects to the prototype plume and is instrumented for heat by 0.99 m by 115 mm high; 75 in by 39
on which they are based. At a minimum release rate measurements. The area in by 4.5 in high) made from 38 mm (1.5
these procedures shall include: surrounding the test specimen in an in) steel angle. The frame shall be
(1) Controls, including incoming open calorimeter layout shall be completely open under the foundation
inspection procedures, of all mattress sufficiently large that there are no heat except for two crosspieces, 25 mm wide
and mattress and foundation set re-radiation effects from any nearby (1 in) at the 1⁄3 length points. If testing
components and materials to ensure that materials or objects. The air flow to the a size other than twin, the relationship
they are identical in all material test specimen should be symmetrical of the mattress to the frame shall be
respects to those used in the prototype; from all sides. The air flow to the comparable to that specified in this
(2) Designation of a production lot calorimeter hood shall be sufficient to paragraph.
that is represented by the prototype; and ensure that the entire fire plume is (4) Catch pan. The bed frame feet
(3) Inspection of mattresses and captured, even at peak burning. Skirts shall rest on a surface of either calcium
mattress and foundation sets produced may be placed on the hood periphery to silicate board or fiber cement board, 13
for sale sufficient to demonstrate that help assure this plume capture, if mm (0.5 in) thick, 2.11 m by 1.19 m (83
they are identical to the prototype in all necessary, though they must not be of in by 47 in). The board serves as a catch
material respects. such an excessive length as to cause the surface for any flaming melt/drip
(b) Production testing. Manufacturers incoming flow to disturb the burning material falling from the bed assembly
are encouraged to conduct, as part of the process. Skirts must also not heat up to and may be the location of a pool fire
quality assurance program, random the point that they contribute significant that consumes such materials. This
testing of mattresses and mattress and re-radiation to the test specimen. The air surface must be cleaned between tests to
foundation sets being produced for sale supply to the hood shall be sufficient avoid build-up of combustible residues.
according to the requirements of that the fire is not in any way limited Lining this surface with aluminum foil
§§ 1633.3 and 1633.7. or affected by the available air supply. to facilitate cleaning is not
(c) Failure of mattresses produced for The fire plume should not enter the recommended since this might increase
sale to meet flammability standard. (1) hood exhaust duct. Brief (seconds) fire intensity via reflected radiation.
Sale of mattresses and foundations. If flickers of flame that occupy only a (5) Ignition source. (i) General. The
any test performed for quality assurance minor fraction of the hood exhaust duct ignition source shall consist of two T-
yields results which indicate that any inlet cross-section are not a problem shaped burners as shown in Figures 3
mattress or mattress and foundation set since they do not signify appreciable and 4 of this part. One burner impinges
of a production lot does not meet the suppression of flames. flames on the top surface of the
criteria of § 1633.3(b), or if a (ii) Test Configuration B. The test mattress. The second burner impinges
manufacturer obtains test results or room shall have dimensions 3.05 meters flames on the side of the mattress and
other evidence that a component or (m) ± 25 millimeters (mm) by 3.66 m ± on the side of the foundation. Each of
material or construction/assembly 25 mm by 2.44 m ± 25 mm (10 feet (ft) the burners shall be constructed from
process used could negatively affect the by 12 ft by 8 ft) high. The specimen is stainless steel tubing (12.7 mm diameter
test performance of the mattress as set placed within the burn room. All smoke with 0.89 ± 0.5 mm wall thickness; 0.50

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2497

in diameter with 0.035 ± 0.002 in wall). in OD, about 3.2 in long) of copper (viii) Flow control system. Each
Each burner shall incorporate a stand- tubing shall be placed in the inlet gas burner shall have a flow control system
off foot to set its distance from the test line just before the burner to facilitate of the type shown in Figure 7 of this
specimen surface (Figure 5 of this part). making the burner nominally parallel to part. Propane gas from a source such as
Both burners shall be mounted with a the test specimen surface (by a a bottle is reduced in pressure to
mechanical pivot point but the side procedure described in this paragraph). approximately 70 kilopascals (‘‘kPa’’)
burner is locked in place to prevent The copper tube on the top surface (20 pounds per square inch gage
movement about this pivot in normal burner must be protected from excessive (‘‘psig’’)) and fed to the system shown
usage. The top burner, however, is free heat and surface oxidation by wrapping in Figure 8 of this part. The gas flow to
to rotate about its pivot during a burner it with a suitable layer of high the burner is delivered in a square-wave
exposure and is lightly weighted so as temperature insulation. Both copper manner (constant flow with rapid onset
to exert a downward force on the tubes are to be bent by hand in the and termination) by means of the
mattress top through its stand-off foot so burner alignment process. They must be solenoid valve upstream of the
that the burner follows a receding top replaced if they become work-hardened flowmeter. An interval timer (accurate
surface on the test specimen (Figure 6 or crimped in any way. The gas inlet to ± 0.2 s) determines the burner flame
of this part). The combination of burner lines (12.7 mm OD stainless steel duration. The pilot light assures that the
stand-off distance and propane gas flow tubing; 0.50 in) serve as arms leading burner will ignite when the solenoid
rate to the burners determines the heat back to the pivot points and beyond, as valve opens 2. The gas flow shall be set
flux they impose on the surface of the shown in Figure 6 of this part. The using a rotameter type of flowmeter,
test specimen so that both of these length to the pivot for the top burner with a 150 mm scale, calibrated for
parameters are tightly controlled. shall be approximately 1000 mm (40 in). propane. When calibrating the
(ii) Top surface burner. The T head of (v) Frame. Figure 6 shows the frame flowmeter, take into account that the
the top surface burner (horizontal that holds the burners and their pivots, flow resistance of the burner holes
burner, Figure 3 of this part) shall be which are adjustable vertically in causes a finite pressure increase in the
305 ± 2 mm (12 ± 0.08 in) long with gas height. All adjustments (burner height, flowmeter above ambient. (If a
tight plugs in each end. Each side of the burner arm length from the pivot point, calibration at one atmosphere is
T shall contain 17 holes equally spaced counterweight positions along the provided by the manufacturer, the
over a 135 mm length (8.5 mm ± 0.1 mm burner arm) are facilitated by the use of flowmeter reading, at the internal
apart; 0.333 ± 0.005 in). The holes on knobs or thumbscrews as the set screws. pressure existing in the meter, required
each side shall begin 8.5 mm (0.33 in) The three point footprint of the burner to get the flow rates listed in this
from the centerline of the burner head. frame, with the two forward points on paragraph must be corrected, typically
The holes shall be drilled with a #56 wheels, facilitates burner movement and by the square root of the absolute
drill and are to be 1.17 mm to 1.22 mm burner stability when stationary. pressure ratio. This calls for measuring
(0.046 in to 0.048 in) in diameter. The (vi) Arms. The metal arms attached to the actual pressure in the flow meters
holes shall be pointed 5° out of the the burners shall be attached to a when set near the correct flow values.
plane of the Figure. This broadens the separate gas control console by flexible, A value roughly in the range of 1 kPa
width of the heat flux profile imposed reinforced plastic tubing.1 The gas to 3 kPa—5 in to 15 in of water—can be
on the surface of the test specimen. control console is mounted separately expected.) Useful guidelines for
(iii) Side surface burner. The T head so as to facilitate its safe placement calibration are provided in Appendix A
of the side surface burner (vertical outside of the test room throughout the of this part.
burner) shall be constructed similarly to test procedure. The propane gas lines
the top surface burner, as shown in (ix) Gas flow rate. Use propane gas
running between the console and the with a known net heat of combustion of
Figure 4 of this part, except that its burner assembly must be anchored on
overall length shall be 254 ± 2 mm (10 46.5 ± 0.5 MJ/kg (nominally 99% to
the assembly before running to the 100% propane). Each burner has a
± 0.08 in). Each side of the burner head burner inlet arms. A 1.5 m ± 25 mm (58
shall contain 14 holes spaced evenly specific propane gas flow rate set with
in ± 1 in) length of flexible, reinforced its respective, calibrated flowmeter. The
over a 110 mm length (8.5 mm ± 0.1 mm tubing between the anchor point and the
apart; 0.333 ± 0.005 in). The holes shall gas flow rate to the top burner is 12.9
end of each burner inlet allows free liters per minute (‘‘L/min’’) ± 0.1 L/min
be drilled with a #56 drill and are to be movement of the top burner about its
1.17 mm to 1.22 mm (0.046 in to 0.048 at a pressure of 101 ± 5 kPa (standard
pivot point. The top burner arm shall atmospheric pressure) and a
in) in diameter. The holes shall be have a pair of moveable cylindrical
pointed 5° out of the plane of the Figure. temperature of 22 ± 3 °C. The gas flow
counterweights that are used, as rate to the side burner is 6.6 ± 0.05 L/
(iv) Burner stand-off. The burner
described below, to adjust the min at a pressure of 101 ± 5 kPa
stand-off on each burner shall consist of
downward force on the stand-off foot. (standard atmospheric pressure) and a
a collar fixed by a set screw onto the (vii) Burner head. Each burner head
inlet tube of the burner head (Figure 5 temperature of 22 ± 3 °C. For the
shall have a separate pilot light flowmeters supplied with the burner
of this part). The collar shall hold a 3 consisting of a 3 mm OD (1⁄8in OD)
mm diameter stainless steel rod having assembly, the black float setting for the
copper tube with an independently- top burner is expected to be in the 85
a 12.7 mm by 51 mm by (2–2.5 mm) controlled supply of propane gas. The
thick (0.5 in by 2 in by (0.08–0.10) in mm to 95 mm range. For the side
tube terminates within 10 mm of the burner, the expected range for the black
thick) stainless steel pad welded on its
center of the burner head. Care must be float is 115 m to 125 mm. The total heat
end with its face (and long axis) parallel
taken to set the pilot flame size small release rate of the burners is 27 kW.
to the T head of the burner. The foot pad
enough so as not to heat the test
shall be displaced about 10 mm to 12
specimen before the timed burner
mm from the longitudinal centerline of 2 If the side burner, or more commonly one half
exposure is begun. of the side burner, fails to ignite quickly, adjust the
the burner head so that it does not rest
position of the igniter, bearing in mind that propane
on the test specimen in an area of peak 1 Fiber-reinforced plastic tubing (6 mm ID by 9.5 is heavier than air. The best burner behavior test
heat flux. A short section (9.5 mm outer mm OD; 1⁄4 inch ID by 3⁄4 inch OD) made of PVC assessment is done against an inert surface (to
diameter (‘‘OD’’), about 80 mm long; 3⁄8 should be used. spread the gas as it would during an actual test).

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2498 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

(b) Conditioning. Remove the (e) Location of the gas burners. Place The location shall be within 30 cm (1
specimens from any packaging prior to the burner heads so that they are within ft) of the longitudinal center of the
conditioning. Specimens shall be 300 mm (1 ft) of the mid-length of the mattress. The intended location of the
conditioned in air at a temperature mattress. The general layout for the stand-off foot of the top burner shall not
greater than 18 °C (65 °F) and a relative room configuration is shown in Figure be in a dimple or crease caused by the
humidity less than 55 percent for at 2 of this part. For a quilted mattress top quilting of the mattress top. Press the
least 48 continuous hours prior to test. the stand-off foot pad must alight on a platen laterally inward from the edge of
Specimens shall be supported in a high, flat area between dimples or the mattress so that its side makes
manner to permit free movement of air quilting thread runs. The same is to be contact with either the top and bottom
around them during conditioning. true for the side burner if that surface is tape edge or the vertical side of the
(c) Test preparation. (1) General. quilted. If a specimen design presents a mattress.4 Use a 20 cm (8 in) strip of
Horizontal air flow at a distance of 0.5 conflict in placement such that both duct tape (platen to mattress top) to
m (20 in) on all sides of the test burners cannot be placed between local hold the platen firmly inward in this
specimen at the mattress top height depressions in the surface, the top position.
shall be ≤ 0.5 m/s. If there is any visual burner shall be placed at the highest flat (ii) With both burner arms horizontal
evidence that the burner flames are surface. (pinned in this position), fully retract
being shifted around during their (f) Burner set-up. The burners shall be the stand-off feet of both burners and, if
exposure durations, the burner regions placed in relation to the mattress and necessary, the pilot tubes as well 5.
must be enclosed on two or more sides foundation surfaces in the manner (Neither is to protrude past the front
by at least a triple layer of screen wire. shown in Figure 9 of this part, i.e., at the face of the burner tubes at this point.)
The screen(s) for the top burner shall sit nominal spacings shown there and with Move the burner assembly forward
on the mattress top but must be far the burner tubes nominally parallel 3 to (perpendicular to the mattress) until the
enough away (typically 30 cm or more) the mattress surfaces on which they vertical burner lightly contacts the sheet
so as not to interfere or interact with impinge. Since the heat flux levels seen metal platen. Adjust the height of the
flame spread during the burner by the test specimen surfaces depend on vertical burner on its vertical support
exposure. The screen for the side burner burner spacing, as well as gas flow rate, column so as to center the tube on the
will require a separate support from care must be taken with the set-up crevice between the mattress and the
below. All screens shall be removed at process. foundation. (This holds also for pillow
the end of the 70 second exposure (g) Burner alignment procedure. (1) top mattress tops, i.e., ignore the crevice
interval. Preparation. Complete the following between the pillow top and the main
(2) Specimen. Remove the test before starting the alignment procedure: body of the mattress.) 6 Adjust the
specimen from the conditioning room (i) Check that the pivot point for the height of the horizontal burner until it
immediately before it is to be tested. Be mattress top burner feed tube and the sits lightly on top of the sheet metal
sure the bed frame is approximately two metal plates around it are clean and platen. Its burner arm should then be
centered on the catch surface. Place the well-lubricated so as to allow smooth, horizontal.
specimen on the bed frame. Carefully free movement. (iii) Move the horizontal burner in/out
center them on the bed frame and on (ii) Set the two burners such that the (loosen the thumb screw near the pivot
each other. The mattress shall be 5° out-of-plane angling of the flame jets point) until the outer end of the burner
centered on top of the foundation (see makes the jets on the two burners point tube is 13 mm to 19 mm (1⁄2 in to 3⁄4 in)
Figure 1 of this part). However, in order slightly toward each other. from the corner bend in the platen (this
to keep the heat flux exposure the same (iii) Check the burner stand-off feet for is facilitated by putting a pair of lines
for the sides of the two components, if straightness and perpendicularity on the top of the platen 13 mm and 19
the mattress is 1 cm to 2 cm narrower between foot pad and support rod and mm from the bend and parallel to it).
than the foundation, the mattress shall to see that they are clean of residue from Tighten the thumb screw.
be shifted so that the side to be exposed a previous test. (iv) Make the horizontal burner
is in the same plane as the foundations. (iv) Have at hand the following items parallel to the top of the platen (within
Refer to Figure 8 of this part. A product to assist in burner set-up: the jig, shown 3 mm, 1⁄8 in over the burner tube length)
having an intended sleep surface on in Figure 10 of this part, for setting the by bending the copper tube section
only one side shall be tested with the stand-off feet at their proper distances
sleeping side up so that the sleeping from the front of the burner tube; a 3 4 Mattresses having a convex side are treated

surface is exposed to the propane mm thick piece of flat stock (any separately since the platen cannot be placed in the
burner. material) to assist in checking the above manner. Use the platen only to set the top
(d) Burner flow rate/flow timer parallelness of the burners to the burner parallelness. Set the in/out distance of the
top burner to the specification in the paragraph
confirmation. Just prior to moving the mattress surfaces; and a 24 gage (g)(2)(iii). Set the side burner so that it is
burner adjacent to the test specimen, stainless steel sheet metal platen that is approximately (visually) parallel to the flat side
briefly ignite each burner at the same 30 mm (12 in) wide, 610 mm (24 in) surface of the foundation below the mattress/
time, and check that the propane flow long and has a sharp, precise 90° bend foundation crevice once its foot is in contact with
the materials in the crevice area. The burner will
to that burner is set at the appropriate 355 mm (14 in) from one 30 mm wide not be vertical in this case. If the foundation side
level on its flowmeter to provide the end. is also non-flat, set the side burner vertical (± 3 mm,
flows listed in paragraph (a)(5)(ix) of (2) Alignment. (i) Place the burner as above) using a bubble level as a reference. The
this section. Check that the timers for assembly adjacent to the test specimen. side surface convexities will then bring the bowed
out sections of the specimen closer to the burner
the burner exposures are set to 70 Place the sheet metal platen on the tube than the stand-off foot.
seconds for the top burner and 50 mattress with the shorter side on top. 5 The pilot tubes can normally be left with their
seconds for the side burner. For a new ends just behind the plane of the front of the burner
burner assembly, check the accuracy of 3 The top burner will tend to be tangential to the tube. This way they will not interfere with
the gas flow timers against a stop watch mattress surface at the burner mid-length; this positioning of the tube but their flame will readily
orientation will not necessarily be parallel to the ignite the burner tubes.
at these standard time settings. Set pilot overall average mattress surface orientation nor will 6 For tests of the mattress alone, set the side
flows to a level that will not cause them it necessarily be horizontal. This is a result of the burner mid-height equal to the lower tape edge of
to impinge on sample surfaces. shape of the mattress top surface. the mattress.

VerDate jul<14>2003 20:07 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2499

appropriately. Note: After the platen is burner. The outer end of the burner tube spans, start the video lights and video
removed in paragraph (g)(2)(vii), the should extend at least 6 mm to 12 mm camera and data logging systems two
burner tube may not be horizontal; this (1⁄4 in to 1⁄2 in) out beyond the minutes before burner ignition (or, if not
is normal. For mattress/foundation uppermost corner/edge of the mattress using video, take a picture of the setup).
combinations having nominally flat, so that the burner flames will hit the (4) Start the burner exposure by
vertical sides, the similar adjustment for tape edge. (For a pillow top mattress, activating power to the burner timers.
the vertical burner is intended to make this means the outer edge of the pillow Also start a 30 minute timer of the test
that burner parallel to the sides and top portion and the distance may then duration. If not using video, one photo
vertical. Variations in the shape of be greater than 6 mm to 12 mm.) If this must be taken within the first 45
mattresses and foundations can cause is not the case, move the burner seconds of starting the burners.
the platen section on the side to be non- assembly (perpendicular to the mattress (5) When the burners go out (after 70
flat and/or non-vertical. If the platen is side)—not the horizontal burner alone— seconds for the longer exposure),
flat and vertical, make the vertical until it is. Finally, move the vertical carefully lift the top burner tube away
burner parallel to the side of the platen burner tube until its stand-off foot just from the specimen surface, producing as
(± 3 mm) by bending its copper tube touches the side of the mattress and/or little as possible disturbance to the
section as needed. If not, make the side the foundation. (Use the set screw near specimen. Remove the burner assembly
burner parallel to the mattress/ the vertical burner pivot.) from the specimen area to facilitate the
foundation sides by the best visual (ix) Make sure all thumbscrews are video camera view of the full side of the
estimate after the platen has been adequately tightened. Care must be specimen. In the case of the room-based
removed. taken, once this set-up is achieved, to configurations, remove the burner
(v) Move the burner assembly avoid bumping the burner assembly or assembly from the room to protect it.
perpendicularly back away from the disturbing the flexible lines that bring Remove all screens.
mattress about 30 cm (1 ft). Set the two propane to it. (i) Video recording/photographs.
stand-off feet to their respective (x) If there is any indication of flow Place a video or still frame camera so as
distances using the jig designed for this to have (when the lens is zoomed out)
disturbances in the test facility which
purpose. Install the jig fully onto the just slightly more than a full-length
cause the burner flames or pilot flames
burner tube (on the same side of the view of the side of the test specimen
to move around, place screens around
tube as the stand-off foot), with its side being ignited, including a view of the
the burners so as to minimize these
edges parallel to the burner feed arm, at flame impingement area while the
disturbances 9. These screens (and any
burner assembly is present. The view
about the position where one end of the holders) must be far enough away from
must also include the catch pan so that
foot will be. Loosen the set screw and the burners (about 30 cm or more for the
it is clear whether any melt pool fire in
slide the foot out to the point where it top, less for the side) so that they do not
this pan participates significantly in the
is flush with the bottom end of the jig. interact with the flames growing on the
growth of fire on the test specimen. The
Tighten the set screw. Make sure the specimen surfaces. For the top surface camera shall include a measure of
long axis of the foot is parallel to the burner, at least a triple layer of window elapsed time to the nearest 1 second for
burner tube. It is essential to use the screen approximately 30 cm high sitting video and 1 minute for still frame
correct side of the spacer jig with each vertically on the mattress top (Figure 9 within its recorded field of view
burner. Double check this. The jig must of this part) has proved satisfactory. For (preferably built-in to the camera). For
be clearly marked. the side burner at least a triple layer of the room-based configuration, the
(vi) Set the downward force of the screen approximately 15 cm wide, required full-length view of the sample
horizontal burner. Remove the retainer formed into a square-bottom U-shape may require an appropriately placed
pin near the pivot. While holding the and held from below the burner has window, sealed with heat resistant
burner feed arm horizontal using a proved satisfactory. Individual glass, in one of the room walls. Place the
spring scale 7 hooked onto the laboratories will have to experiment camera at a height just sufficient to give
thumbscrew holding the stand-off foot, with the best arrangement for a view of the top of the specimen while
move the small and/or large weights on suppressing flow disturbances in their remaining under any smoke layer that
the feed tube appropriately so that the facility. may develop in the room. The specimen
spring scale reads 170 g to 225 g (6 oz (xi) Proceed with the test (see Test shall be brightly lit so that the image
to 8 oz). Procedure in paragraph (h) of this
(vii) Remove the sheet metal platen does not lose detail to over-exposed
section and Appendix B of this part). flames. This will require a pair or more
(and tape holding it). (h) Running the test. (1) Charge the
(viii) Hold the horizontal burner up of 1 kW photo flood lights illuminating
hose line to be used for fire suppression the viewed side of the specimen. The
while sliding the burner assembly with water.
forward until its stand-off foot just lights may need to shine into the room
(2) Ignite the pilot lights on both from the outside via sealed windows.
touches the mattress and/or the burners and make sure they are small
foundation 8, then release the horizontal (j) Cessation of test. (1) The heat
enough as to not heat the test specimen release rate shall be recorded and video/
surfaces significantly. photographs taken until either 30
7 An acceptable spring scale has a calibrated
(3) With the calorimetry system fully minutes has elapsed since the start of
spring mounted within a holder and hooks on each
end. operational, after instrument zeroes and the burner exposure or a fire develops
8 The foot should depress the surface it first
of such size as to require suppression
contacts by no more than 1 mm to 2 mm. This is burners too close to the vertical mattress/foundation
best seen up close, not from the rear of the burner sides, it will be necessary to use the spacer jig for the safety of the facility.
assembly. However, if a protruding tape edge is the (rather than the stand-off foot) above or below this (2) Note the time and nature of any
first item contacted, compress it until the foot is in crevice to set the proper burner spacing. Compress unusual behavior that is not fully within
the plane of the mattress/foundation vertical sides. the mattress/foundation surface 1 mm to 2 mm the view of the video camera. This is
The intent here is that the burner be spaced a fixed when using the jig for this purpose.
most easily done by narration to a
distance from the vertical mattress/foundation 9 The goal here is to keep the burner flames

sides, not from an incidental protrusion. Similarly, impinging on a fixed area of the specimen surface camcorder.
if there is a wide crevice in this area which would rather than wandering back and forth over a larger (3) Run the heat release rate system
allow the foot to move inward and thereby place the area. and datalogger until the fire has been

VerDate jul<14>2003 20:07 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2500 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

fully out for several minutes to allow Commission finds that an open flame (o) Infant carrier and lounge pad. Pad
the system zero to be recorded. standard for mattresses with the testing to cushion a baby in an infant carrier.
requirements and criteria that are (p) Mattress foundation. This is a
§ 1633.8 Findings. ticking covered structure used to
specified in the Commission rule is the
(a) General. In order to issue a least burdensome requirement that support a mattress or sleep surface. The
flammability standard under the FFA, would prevent or adequately reduce the structure may include constructed
the FFA requires the Commission to risk of injury for which the regulation is frames, foam, box springs or other
make certain findings and to include being promulgated. materials used alone or in combination.
these in the regulation, 15 U.S.C. (q) Murphy Bed. A style of sleep
1193(j)(2). These findings are discussed § 1633.9 Glossary of terms. system where the mattress and
in this section. (a) Absorbent pad. Pad used on top of foundation are fastened to the wall and
(b) Voluntary standards. No findings mattress. Designed to absorb moisture/ provide a means to retract or rotate the
concerning compliance with and body fluids thereby reducing skin bed assembly into the wall to release
adequacy of a voluntary standard are irritation, can be one time use. more floor area for other uses.
necessary because no relevant voluntary (b) Basket pad. Cushion for use in an (r) Pillow. Cloth bag filled with
standard addressing the risk of injury infant basket. resilient material such as feathers,
that is addressed by this regulation has (c) Bunk beds. A tier of beds, usually down, sponge rubber, urethane, or fiber
been adopted and implemented. two or three, in a high frame complete used as the support for the head of a
(c) Relationship of benefits to costs. with mattresses (see Figure 11 of this person.
The Commission estimates the potential part). (s) Playpen pad. Cushion used on the
total lifetime benefits of a mattress that (d) Car bed. Portable bed used to carry bottom of a playpen.
complies with this standard to range a baby in an automobile. (t) Portable crib. Smaller size than a
from $62 to $74 per mattress (based on (e) Carriage pad. Cushion to go into a conventional crib. Can usually be
a 10 year mattress life and a 3% baby carriage. converted into a playpen.
discount rate). The Commission (f) Chaise lounge. An upholstered (u) Quilted means stitched with
estimates total resource costs of the couch chair or a couch with a chair thread or by fusion through the ticking
standard to range from $13 to $44 per back. It has a permanent back rest, no and one or more layers of material.
mattress. This yields net benefits of $18 arms, and sleeps one (see Figure 11). (v) Roll-away-bed. Portable bed which
to $62 per mattress. The Commission (g) Convertible sofa. An upholstered has frame that folds with the mattress
estimates that aggregate lifetime benefits sofa that converts into an adult sized for compact storage.
associated with all mattresses produced bed. Mattress unfolds out and up from (w) Sleep lounge. Upholstered seating
the first year the standard becomes under the seat cushioning (see Figure section is mounted on a frame. May
effective range from $1,560 to $1,880 11). have bolster pillows along the wall as
million, and that aggregate resource (h) Corner groups. Two twin size backrests or may have attached
costs associated with these mattresses bedding sets on frames, usually headrests (see Figure 11).
range from $320 to $1,110 million, slipcovered, and abutted to a corner (x) Stroller pad. Cushion used in a
yielding net benefits of about $450 to table. They also usually have loose baby stroller.
$1,560 million. Accordingly, the bolsters slipcovered (see Figure 11). (y) Sofa bed. These are pieces in
Commission finds that the benefits from (i) Crib bumper. Padded cushion which the back of the sofa swings down
the regulation bear a reasonable which goes around three or four sides flat with the seat to form the sleeping
relationship to its costs. inside a crib to protect the baby. Can surface. All upholstered. Some sofa beds
(d) Least burdensome requirement. also be used in a playpen. have bedding boxes for storage of
The Commission considered the (j) Daybed. Daybed has foundation, bedding. There are two types: The one-
following alternatives: Alternative usually supported by coil or flat springs, piece, where the back and seat are
maximum peak heat release rate and test mounted between arms on which upholstered as a unit, supplying an
duration, alternative total heat released mattress is placed. It has permanent unbroken sleeping surface; and the two-
in the first 10 minutes of the test, arms, no backrest, and sleeps one (see piece, where back and seat are
mandatory production testing, a longer Figure 11). upholstered separately (see Figure 11).
effective date, taking no action, relying (k) Dressing table pad. Pad to cushion (z) Sofa lounge—(includes glideouts).
on a voluntary standard, and requiring a baby on top of a dressing table. Upholstered seating section is mounted
labeling alone (without any performance (l) Drop-arm loveseat. When side arms on springs and in a frame that permit it
requirements). The alternatives of taking are in vertical position, this piece is a to be pulled out for sleeping. Has
no action, relying on a voluntary loveseat. The adjustable arms can be upholstered backrest bedding box that is
standard (if one existed) requiring lowered to one of four positions for a hinged. Glideouts are single sleepers
labeling alone are unlikely to adequately chaise lounge effect or a single sleeper. with sloping seats and backrests. Seat
reduce the risk. Requiring a criterion of The vertical back support always pulls out from beneath back and evens
25 MJ total heat release during the first remains upright and stationary (see up to supply level sleeping surface (see
10 minutes of the test instead of 15 MJ Figure 11). Figure 11).
would likely reduce the estimated (m) Futon. A flexible mattress (aa) Studio couch. Consists of
benefits (deaths and injuries reduced) generally used on the floor that can be upholstered seating section on
without having much effect on costs. folded or rolled up for storage. It usually upholstered foundation. Many types
Both options of increasing the duration consists of resilient material covered by convert to twin beds (see Figure 11).
of the test from 30 minutes to 60 ticking. (bb) Studio divan. Twin size
minutes and decreasing the peak rate of (n) High riser. This is a frame of sofa upholstered seating section with
heat release from 200 kW to 150 kW seating height with two equal size foundation is mounted on metal bed
would likely increase costs significantly mattresses without a backrest. The frame. Has no arms or backrest, and
without substantial increase in benefits. frame slides out with the lower mattress sleeps one (see Figure 11).
Requiring production testing would also and rises to form a double or two single (cc) Trundle bed. A low bed which is
likely increase costs. Therefore, the beds (see Figure 11). rolled under a larger bed. In some lines,

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2501

the lower bed springs up to form a intervals, starting at 5 minutes and (d) Quality assurance records. In
double or two single beds as in a high ending at 30 minutes), with the addition to the records required by
riser (see Figure 11). prototype identification number or paragraph (a) of this section, the
(dd) Tufted means buttoned or laced production lot identification number of following quality assurance records
through the ticking and upholstery the mattress or mattress foundation set, shall be maintained:
material and/or core, or having the date and time of test, and name and (1) A written copy of the
ticking and loft material and/or core location of testing facility clearly manufacturer’s quality assurance
drawn together at intervals by any other displayed. procedures.
method which produces a series of (b) Prototype records. In addition to (2) Records of any production tests
depressions on the surface. the records specified in paragraph (a) of performed. Production test records must
(ee) Twin studio divan. Frames which this section, the following records be maintained and shall include in
glide out (but not up) and use seat related to prototype testing shall be addition to the requirements of
cushions, in addition to upholstered maintained: paragraph (a) of this section, an assigned
foundation to sleep two. Has neither (1) Unique identification number for production lot identification number
arms nor back rest (see Figure 11). the qualified prototype and a list of the and the identification number of the
(ff) Flip or sleeper chair. Chair that unique identification numbers of each prototype associated with the specimen
unfolds to be used for sleeping, prototype based on the qualified tested.
typically has several connecting fabric prototype. (3) For each prototype, the number of
covered, solid foam core segments. (2) A detailed description of all mattresses or mattress and foundation
materials, components, and methods of sets in each production lot based on that
Subpart B—Rules and Requirements construction for each prototype mattress prototype.
or prototype mattress and foundation (4) The duration of manufacture of the
§ 1633.10 Definitions. production lot, i.e., the start and end
set. Such description shall include at a
(a) Standard means the Standard for minimum, the specifications of all dates of production of that lot.
the Flammability (Open-Flame) of materials and components, name and (5) Component, material and assembly
Mattresses and Foundations (16 CFR location of each material and records. Every manufacturer conducting
part 1633, subpart A). component supplier, and a physical tests and/or technical evaluations of
(b) The definition of terms set forth in sample of each material and component components and materials and/or
§ 1633.2 of the standard shall also apply of the prototype. methods of construction must maintain
to this subpart. (3) A list of which models and detailed records of such tests and
production lots of mattresses or mattress evaluations.
§ 1633.11 Records. (e) Record retention requirements.
and foundation sets are represented by
(a) Test and manufacturing records— each prototype identification number. The records required under this section
General. Every manufacturer (including (4) Where a prototype is not required shall be maintained by the manufacturer
importers) or other person initially to be tested before sale, pursuant to (including importers) for as long as
introducing into commerce mattresses § 1633.4(b), the prototype identification mattresses/foundations based on the
or mattress and foundation sets subject number of the qualified prototype on prototype in question are in production
to the standard, irrespective of whether which the mattress to be offered for sale and shall be retained for 3 years
guarantees are issued relative thereto, is based, and, at a minimum, the thereafter. Records shall be available
shall maintain the following records: manufacturing specifications and a upon the request of Commission staff.
(1) Test results and details of each test description of the materials substituted § 1633.12 Labeling.
performed by or for that manufacturer and/or the size change, photographs or
(including failures), whether for (a) Each mattress or mattress/
physical specimens of the substituted foundation set subject to the standard
prototype, confirmation, or production, materials, and documentation based on
in accordance with § 1633.7. Details shall bear a permanent, conspicuous,
objectively reasonable criteria that the and legible label containing:
shall include: Location of test facility, change in any component, material, or
type of test room, test room conditions, (1) Name of the manufacturer;
method of construction will not cause (2) Location of the manufacturer,
prototype or production identification the prototype to exceed the test criteria including street address, city and state;
number, and test data including the specified in § 1633.3(b). (3) Month and year of manufacture;
peak rate of heat release, total heat (5) Identification, composition, and (4) Model identification;
release in first 10 minutes, a graphic details of the application of any flame (5) Prototype identification number
depiction of the peak rate of heat release retardant treatments and/or inherently for the mattress; and
and total heat release over time. These flame resistant fibers or other materials (6) A certification that the mattress
records shall include the name and employed in mattress components. complies with this standard.
signature of person conducting the test, (c) Pooling confirmation test records. (b) The information required on labels
the date of the test, and a certification With respect to pooling confirmation by this section shall be set forth
by the person overseeing the testing as testing, records shall be maintained to separately from any other information
to the test results and that the test was show: appearing on such label. Other
carried out in accordance with the (1) The prototype identification information, representations, or
Standard. For confirmation tests, the number assigned by the original disclosures, appearing on labels
identification number must be that of prototype manufacturer. required by this section or elsewhere on
the prototype tested. (2) Name and location of the the item, shall not interfere with,
(2) Video and/or a minimum of eight prototype manufacturer. minimize, detract from, or conflict with
photographs of the testing of each (3) Copy of prototype test records, and the required information.
mattress or mattress and foundation set, records required by paragraph (b)(2) of (c) No person, other than the ultimate
in accordance with § 1633.4 (one taken this section. consumer, shall remove or mutilate, or
before the test starts, one taken within (4) A list of models of mattresses, and/ cause or participate in the removal or
45 seconds of the start of the test, and or mattress and foundation sets, mutilation of, any label required by this
the remaining six taken at five minute represented by the prototype. section to be affixed to any item.

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2502 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

§ 1633.13 Tests for guaranty purposes, than 0.5 inch (1.27 cm) in height and all clarification will better protect the
compliance with this section, and one of a letters on the label shall be in a color public against the unreasonable risk of
kind exemption. which contrasts with the background of fires leading to death, personal injury or
(a) Tests for guaranty purposes. the label. The warning statement which significant property damage, and assure
Reasonable and representative tests for appears on the label must also be that purchasers of renovated mattresses
the purpose of issuing a guaranty under conspicuously displayed on the invoice receive the same protection under the
section 8 of the Flammable Fabrics Act, or other sales papers that accompany Flammable Fabrics Act as purchasers of
15 U.S.C. 1197, for mattresses or the mattress in commerce from the new mattresses.
mattress and foundation sets subject to manufacturer to the final point of sale (c) For purposes of this subpart,
the standard shall be the tests performed to a consumer. mattress renovation includes a wide
to show compliance with the standard. (2) The manufacturer of a mattress or range of operations. Replacing the
(b) Compliance with this section. No mattress and foundation set exempted ticking or batting, stripping a mattress to
person subject to the Flammable Fabrics from testing under this paragraph shall, its springs, rebuilding a mattress, or
Act shall manufacture for sale, import, in lieu of the records required to be kept replacing components with new or
distribute, or otherwise market or by §1633.10, retain a copy of the written recycled materials, are all part of the
handle any mattress or mattress and prescription or other comparable process of renovation. Any one, or any
foundation set which is not in written medical therapeutic combination of one or more, of these
compliance with the provisions under specification for such mattress during a steps in mattress renovation is
subpart B of this part. period of three years, measured from the considered to be mattress manufacture.
(c) ‘‘One of a kind’’ exemption for date of manufacture. (d) If the person who renovates the
physician prescribed mattresses. (1)(i) A (3) For purposes of this subpart the mattress intends to retain the renovated
mattress or mattress and foundation set term physician shall mean a physician, mattress for his or her own use, or if a
manufactured in accordance with a chiropractor or osteopath licensed or customer or a renovator merely hires the
physician’s written prescription or otherwise permitted to practice by any services of the renovator and intends to
manufactured in accordance with other State of the United States. take back the renovated mattress for his
comparable written medical therapeutic or her own use, ‘‘manufacture for sale’’
specification, to be used in connection Subpart C—Interpretations and has not occurred and such a renovated
with the treatment or management of a Policies mattress is not subject to the mattress
named individual’s physical illness or standard.
injury, shall be considered a ‘‘one of a § 1633.14 Policy clarification on
renovation of mattresses. (e) However, if a renovated mattress is
kind mattress’’ and shall be exempt
(a) Section 3 of the Flammable Fabrics sold or intended for sale, either by the
from testing under the standard
Act (15 U.S.C. 1192) prohibits, among renovator or the owner of the mattress
pursuant to § 1633.7 thereof: Provided,
other things, the ‘‘manufacture for sale’’ who hires the services of the renovator,
that the mattress or mattress and
of any product which fails to conform such a transaction is considered to be
foundation set bears a permanent,
to an applicable standard issued under ‘‘manufacture for sale’’.
conspicuous and legible label which
the Act. The standard for the (f) Accordingly, mattress renovation is
states:
Flammability (Open-Flame) of considered by the Commission to be
WARNING: This mattress or mattress and ‘‘manufacture for sale’’ and, therefore,
foundation set may be subject to a large fire Mattresses and Foundations in subpart
A of this part, issued pursuant to the subject to the open-flame Mattress
if exposed to an open flame. It was
manufactured in accordance with a Act, provides that, with certain Standard, when renovated mattresses
physician’s prescription and has not been exceptions, mattresses must be tested are sold or intended for sale by a
tested under the Federal Standard for the according to a prescribed method. The renovator or the customer of the
Flammability (Open-Flame) of Mattresses standard does not exempt renovation; renovator.
and Foundation Sets (16 CFR part 1633). nor does it specifically refer to (g) A renovator who believes that
(ii) Such labeling must be attached to renovation. certain mattresses are entitled to one-of-
the mattress or mattress and foundation (b) The purpose of this subpart is to a-kind exemption, may present relevant
set so as to remain on or affixed thereto inform the public that mattresses facts to the Commission and petition for
for the useful life of the mattress or renovated for sale are considered by the an exemption. Renovators are expected
mattress and foundation set. The label Commission to be mattresses to comply with all the testing
must be at least 40 square inches (250 manufactured for sale and, therefore, requirements of the open-flame Mattress
sq. cm) with no linear dimension less subject to the requirements of the open- Standard until an exemption is
than 5 inches (12.5 cm). The letters in flame Mattress Standard. The approved.
the word ‘‘WARNING’’ shall be no less Commission believes that this policy BILLING CODE 6355–01–P

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2503

EP13JA05.000</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
2504 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

EP13JA05.001</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00036 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2505

EP13JA05.002</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
2506 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

EP13JA05.003</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00038 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2507

EP13JA05.004</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00039 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
2508 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

EP13JA05.005</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00040 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2509

EP13JA05.006</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00041 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
2510 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

EP13JA05.007</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00042 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2511

EP13JA05.008</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00043 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
2512 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

EP13JA05.009</GPH>

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00044 Fmt 4701 Sfmt 4725 E:\FR\FM\13JAP2.SGM 13JAP2
Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules 2513

BILLING CODE 6355–01–C 3. The gas pressure downstream of the propane gas flow rate using the recorded time
Appendix A: Calibration of Propane rotameters that are installed in the control and number of rotations (total flow in that
Flowmeters box of the burner assembly is maintained time). Use the pressure and temperature
near atmospheric pressure (only a few readings to convert to standard conditions.
1. Once the assembly of the burner is millimeters of water above atmosphere). Repeat this measurement for two additional
completed and all the connecting points are Therefore, the best location to place the meter setting to allow for calibrating the
checked for gas leakage, the most critical task diaphragm test meter for gas flow calibration flowmeter throughout the range of interest.
is ensuring the exact flow rates of propane is right downstream of the control box. The Plot the flow versus meter reading, fit a best
into the top and side burners, as described pressure at the propane tank must be set at line (possibly quadratic) through these points
in the test protocol. The gas flow rates are to find the meter setting for a flow of 12.9
20 ± 0.5 pounds per square inch gage (psig).
specified at 12.9 Liters per minute (LPM) ± Calibration Procedure:
LPM at the above ‘‘standard’’ conditions.
0.1 LPM and 6.6 LPM ± 0.05 LPM for the top Install the diaphragm test meter (DTM)
Repeat this procedure for ‘‘Burner 2’’ using
and side burners (Burners 1 and 2), three meter readings to find the setting that
downstream of the control box in the line for gives a flow rate of 6.6 LPM at the standard
respectively, at a pressure of 101 ± 5 the top burner. Check all connecting points
kiloPascal (kPa) (standard atmospheric conditions. After completion of the
for gas leakage. Open the main valve on the calibration, re-set the timers to 70 and 50
pressure) and a temperature of 22 ± 3° propane tank and set a pressure of 20 ± 0.5
Centigrade (C). The rotameters that are seconds.
psig. Set the timers in the control box for 999
installed in the control box of the burner seconds (or the maximum range possible). Appendix B: Burner Operation Sequence
assembly need to be calibrated for accurate Record the barometric pressure. Turn the 1. Starting point: AC power on (red knob
measurement of these flow rates. ‘‘Burner 1’’ switch to ON and ignite the top out); propane pressure set to 20 psig at bottle;
2. The most practical and accurate method burner. Allow the gas to flow for 2–3 minutes timers set to 70 s (top burner) and 50 s (side
of measuring and calibrating the flow rate of until the DTM is stabilized. Record the burner); flowmeters pre-set to values that
gases (including propane) is use of a pressure and temperature in the DTM. Use a give the requisite propane gas flow rates to
diaphragm test meter (also called a dry test stopwatch to record at least one minute each burner. Pilot tubes set just behind front
meter). A diaphragm test meter functions worth of complete rotations while counting surface of burners; pilot flow valves set for
based on positive displacement of a fixed the number of rotations.1 Calculate the ca. 2 cm flames.
volume of gas per rotation and its reading is 2. Position burner on test specimen and
therefore independent of the type of the gas 1 With a diaphragm test meter well-sized to this remove sheet metal platen.
being used. The gas pressure and application, this should be more than five rotations.
temperature, however, can have significant A one liter per rotation meter will require 10 to 15 than the minimum of one minute recording time
EP13JA05.010</GPH>

impact on the measurement of flow rate. rotations for the flow measurements and greater specified here.

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00045 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2
2514 Federal Register / Vol. 70, No. 9 / Thursday, January 13, 2005 / Proposed Rules

3. Place screens around both burners. 10. Terrance R. Karels, EC, to Margaret L. to address small open flame ignition of
4. Open pilot ball valves one at a time and Neily, ES, ‘‘Updated Mattress Market bedclothes.
ignite pilots with hand-held flame; adjust Information,’’ October 5, 2004.
DATES: Comments and submissions
flame size if necessary being very careful to 11 Memorandum from Linda Smith EPI,
avoid a jet flame that could prematurely ‘‘Involvement of Bedclothes in Residential
must be received by March 14, 2005.
ignite the test specimen (Beware: after a long Mattress Fires,’’ May 2004. ADDRESSES: Comments should be
interval between tests the low pilot flow rate 12. Terrance R. Karels, EC, to Margaret L. mailed, preferably in five copies, to the
will require a long time to displace air in the Neily, ES, ‘‘Bedding Market Information,’’ Office of the Secretary, Consumer
line and achieve the steady-state flame size.) October 5, 2004. Product Safety Commission,
5. Open both burner ball valves. 13. Memorandum from Allyson Tenney, Washington, DC 20207–0001, or
6. Start test exposure by simultaneously ES, to Margaret Neily, Engineering Sciences, delivered to the Office of the Secretary,
turning on power to both timers (timers will ‘‘Bedclothes Flammability,’’ October 29, Consumer Product Safety Commission,
turn off burners at appropriate times). 2004.
Room 502, 4330 East-West Highway,
7. Check/adjust propane flow rates (DO 14. Memorandum from Martha A. Kosh,
THIS ESSENTIAL TASK IMMEDIATELY. OS, to ES, ‘‘Standard to Address Open Flame Bethesda, Maryland; telephone (301)
Experience shows the flow will not remain Ignition of Mattresses/Bedding; ANPR,’’ List 504–0800. Comments also may be filed
the same from test-to-test in spite of fixed of comments on CF 02–1, December 13, 2001. by telefacsimile to (301) 504–0127 or by
valve positions so adjustment is essential.) 15. Memorandum from Martha A. Kosh, email to cpsc-os@cpsc.gov. Comments
8. After burners are out: OS, to ES, ‘‘Standard to Address Open Flame should be captioned ‘‘Bedclothes
a. Lift top burner and back assembly away Ignition of Mattresses/Bedding; ANPR ANPR.’’
from specimen. (Revised),’’ List of comments on CF 02–1,
FOR FURTHER INFORMATION CONTACT:
b. Turn off power to both timers. September 27, 2004.
c. Remove screens. 16. National Research Council (2000). Margaret Neily, Directorate for
d. Turn off pilots at their ball valves. Toxicological Risks of Selected Flame- Engineering Sciences, Consumer
Retardant Chemicals (Washington, DC, Product Safety Commission,
Dated: December 22, 2004.
National Academy Press). Washington, DC 20207; telephone (301)
Todd Stevenson, 504–0508, extension 1293.
Secretary, Consumer Product Safety [FR Doc. 05–416 Filed 1–12–05; 8:45 am]
SUPPLEMENTARY INFORMATION:
Commission. BILLING CODE 6355–01–P

List of Relevant Documents


A. Background
1. Briefing memorandum from Margaret An existing flammability standard for
CONSUMER PRODUCT SAFETY mattresses addresses ignition of
Neily, Project Manager, Directorate for COMMISSION
Engineering Sciences, to the Commission, mattresses and mattress pads by
‘‘Notice of Proposed Rulemaking for Mattress 16 CFR Part 1634 cigarettes. 16 CFR Part 1632. On October
Flammability (Open Flame) and Options for 11, 2001, the Commission published an
Addressing Bedclothes Involvement in Standard To Address Open Flame advance notice of proposed rulemaking
Mattress/Bedding Fires,’’ November 1, 2004. (‘‘ANPR’’) addressing open flame
Ignition of Bedclothes; Advance Notice
2. Memorandum from Allyson Tenney, ES, ignition of mattresses. 66 FR 51886.
to Margaret Neily, Engineering Sciences, of Proposed Rulemaking
‘‘Background and Technical Rationale for
That ANPR was the result of several
AGENCY: Consumer Product Safety years of evaluation by Commission staff
Draft Proposed Standard for Open Flame
Mattress Flammability,’’ October 29, 2004. Commission. and petitions on mattress flammability
3. Memorandum from Linda Smith and ACTION: Advance Notice of proposed submitted by Whitney Davis, Director of
David Miller, EPI, ‘‘Residential Fires rulemaking. the Children’s Coalition for Fire-safe
Involving Mattresses and Bedding,’’ October Mattresses. As explained in the ANPR,
2004. SUMMARY: The Commission is the Sleep Products Safety Council
4. Memorandum from Carolyn Meiers, considering issuing a flammability (‘‘SPSC’’), an affiliate of the
ESHF, to Margaret Neily, Project Manager, standard that would address open flame International Sleep Products
‘‘Criteria for Judging Effectiveness of ignition of bedclothes. (Commissioner
Proposed Mattress Standard,’’ October 14,
Association (‘‘ISPA’’), sponsored a
2004.
Thomas H. Moore issued a statement, a research program at the National
5. Memorandum from Carolyn Meiers, copy of which is available from the Institute of Standards and Technology
ESHF, to Margaret Neily, Project Manager, Commission’s Office of the Secretary or (‘‘NIST’’). The NIST research program
‘‘Human Behavior in Fire,’’ October 7, 2004. from the Commission’s Web site, has provided a great deal of technical
6. Memorandum from Treye Thomas and http://www.cpsc.gov.) Elsewhere in information about mattress fires,
Patricia Brundage, HS, ‘‘Qualitative today’s Federal Register, the including the role of bedclothes in such
Assessment of Potential Risk from the Use of Commission is proposing a flammability fires.
Flame Retardant Chemicals in Mattresses,’’ standard that addresses open flame As noted in the mattress ANPR,
October 25, 2004. ignition of mattresses/foundations. mattresses generally are not used alone,
7. Memorandum from Robert Franklin, EC,
to Margaret L. Neily, ES, ‘‘Preliminary
Research indicates that in mattress fires but are covered by bedding or
Environmental Assessment of a Draft the mattress and bedclothes operate bedclothes, whose presence
Proposed Open-Flame Ignition Resistance together as a system. Thus, the significantly affects the character of the
Standard for Mattresses,’’ October 21, 2004. Commission believes that a fire. In most incidents a small open
8. Memorandum from Soumaya Tohamy, flammability standard for bedclothes in flame initially ignites the bedding, and
EC, to Margaret Neily, Project Manager, addition to one for mattresses may be these materials serve as a larger ignition
‘‘Preliminary Regulatory Analysis of a Draft appropriate. The Commission invites source for the mattress. Because few
Proposed Standard to Address Open Flame comments concerning the risk of injury materials can resist such a large ignition
Ignitions of Mattresses,’’ October 27, 2004. identified in this notice, the regulatory source, the typical approach of
9. Memorandum from Soumaya Tohamy,
EC, to Margaret Neily, Project Manager,
alternatives being considered, and other preventing ignition of a mattress
‘‘Initial Regulatory Flexibility Analysis of a possible alternatives. The Commission through a product performance standard
Draft Proposed Standard to Address Open also invites submission of any existing may not be fully adequate for an open
Flame Ignitions of Mattresses,’’ October 27, standard or statement of intention to flame mattress standard. Therefore, the
2004. modify or develop a voluntary standard Commission has taken the approach in

VerDate jul<14>2003 17:56 Jan 12, 2005 Jkt 205001 PO 00000 Frm 00046 Fmt 4701 Sfmt 4702 E:\FR\FM\13JAP2.SGM 13JAP2