Cosco, Inc.; Denial of Petition For Determination of Inconsequential Noncompliance

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

[Docket No. 93-48; Notice 2]

Cosco, Inc.; Denial of Petition For Determination of

Inconsequential Noncompliance

Cosco, Inc. (Cosco) of Columbus, Indiana determined that some of

its child safety seats failed to comply with the flammability

requirements of 49 CFR 571.213, ``Child Restraint Systems,'' Federal

Motor Vehicle Safety Standard No. 213, and filed an appropriate report

pursuant to 49 CFR part 573. Cosco also petitioned to be exempted from

the notification and remedy requirements of the National Traffic and

Motor Vehicle safety Act (15 U.S.C. 1381 et seq.) on the basis that the

noncompliance is inconsequential as it relates to motor vehicle safety.

Notice of receipt of the petition was published on July 7, 1993,

and an opportunity afforded for comment (58 FR 36510). No comments were

received. This notice denies the petition.

Paragraph S5.7 of Standard No. 213 states that ``[e]ach material

used in a child restraint system shall conform to the requirements of

S4 of FMVSS No. 302 (571.302).'' Paragraph S4.3(a) of Standard No. 302

states that ``[w]hen tested in accordance with S5, material described

in S4.1 and S4.2 shall not burn, nor transmit a flame front across its

surface, at a rate of more than 4 inches per minute.''

Between November 1, 1989, and March 31, 1993, Cosco produced

133,897 add-on (as opposed to built-in) child restraint seats, with

shoulder harness straps which it has determined do not comply with the

flammability requirements of Standard No. 213. The principal

restraining mechanism on the noncompliant seats is a soft-shield

harness assembly. The soft-shield harness assembly consists of a

buckle, a soft molded urethane shield, and two straps, protruding to

the top of the shield, which go over the child's shoulders through

slots in the back of the child restraint and attach to a metal bar,

which in turn is attached to an adjustment strap. Indications of a

possible noncompliance came to light during testing of the seats by

NHTSA at Detroit Testing Laboratory in February 1993, and retesting at

U.S. Testing Laboratory. The harness straps burned at a rate of 4.3

inches per minute. This formed the basis of NHTSA investigation NCI

3269.

Cosco supported its petition for inconsequential noncompliance with

the arguments set forth below. Cosco also submitted photographs of the

noncompliant seats, photographs of the tests being conducted on the

seats, and test data. These materials were available for review in the

NHTSA docket during the comment period.

The company began its petition by agreeing

generally that requiring child restraints to meet the [FMVSS] 302

standard does further the purpose of the standard when considering

such child restraint components as vinyl or fabric pads or their

foam contents. Cosco also concedes that the applicability of the

standard to the harness systems of certain child restraints furthers

the purpose of the standard, such as five-point harness systems

which attach to, or pass through, the seating surface of the child

restraint where sources of ignition such as cigarettes or matches

could become entrapped.

Cosco's principal argument dealt with the improbability that the

restraints would ignite. In support of this, it submitted that:

[I]t is not physically possible for the harness straps of the

soft shield to ignite or burn unless the entire child restraint or

the automobile seat upon which it is installed is already burning.

The configuration and placement of the straps of the Cosco soft-

shield assembly are such that these straps cannot come into contact

with an independent source of ignition, such as a cigarette or

match, which would result in any burning of the harness strap.

These are the only two possible causes for the ignition of the

shoulder straps of Cosco soft-shield child restraints. The first is

fire already consuming the child restraint and/or the vehicle seat

upon which the child restraint is installed is on fire. It cannot be

seriously questioned that, in such an instance, the child would be

seriously or fatally burned from these sources of fire as opposed to

the shoulder straps of the child restraint contributing in any

degree to the child's injury. Cosco retained John E. Pless, M.D.,

Director of Forensic Pathology, Department of Pathology, Indiana

University, School of Medicine, to review this issue. Dr. Pless, one

of the leading forensic pathologists in the country and, through his

work with Riley Children's Hospital in Indianapolis, one of the most

experienced pediatric pathologists, concludes that the webbing of

the Cosco soft-shield child restraints would have no practical

importance on the effects of such a fire on a child. [Dr. Pless'

report and curriculum vitae are in the docket.] Dr. Pless'

conclusions are supported by tests performed by Cosco [photographs

of the tests are in the docket.] The tests establish that the

webbing does not ``ignite'' as that term is commonly understood. The

webbing burns in a fashion that can be more accurately described as

smoldering and generally extinguishes itself after a brief period of

time. It should be noted that the tests * * * do not reflect any

possible ignition of the child restraint harness straps if the child

restraint were occupied by a child. [Cosco believes t]here is simply

no way for a source of ignition, such as a lighter, to come into

contact with the strap * * * when the child restraint is occupied by

a child.

The other possible source of ignition of the harness strap would

be from a localized heat source, such as a match or cigarette. It is

critical in the analysis of whether such a risk exists to examine

the configuration and placement of the straps of the Cosco soft-

shield child restraints. [T]hese straps only contact the child who

is occupying the child restraint at the mid-chest level, or higher

on the child's body. The straps are essentially vertical as they

leave the shield. Cosco conducted tests attempting to ignite the

harness strap with a burning cigarette. [Photographs of this test

are in the docket.] Simply stated, a lighted cigarette cannot ignite

the harness strap. Cosco conducted these tests under controlled

conditions which, frankly, seemed inconceivable to occur in the

actual use of child restraints.

For example, in order to come into contact, for any length of

time, with the child restraint harness strap, a lighted cigarette

would have to be balanced at the point where the strap emerges from

the molded shield. This is so unlikely as to be virtually

inconceivable. Dr. Pless also commented on this possibility and

indicated that such a localized heat source is ``not within the

realm of practical consideration.'' Cosco believes that any

practical examination of these issues concludes that the risk of the

ignition of the harnesses of Cosco soft-shield child restraints

could not, under any conceivable set of circumstances, result in

injury or death to the occupant of the child restraint. The

noncompliance of Cosco soft-shield child restraints is therefore

inconsequential as it relates to motor vehicle safety as set forth

in FMVSS 302.

NHTSA has given careful consideration to this argument, and

disagrees with Cosco's assertion that the straps cannot become ignited.

With respect to the photographs Cosco submitted of tests conducted on

the straps with a cigarette lighter, Cosco believes that they show that

the strap webbing smolders and then extinguishes itself without

igniting. NHTSA interprets the photographs as illustrating that the

straps support a flame which could injure a child restrained in the

child seat.

In issuing Standard No. 302 in 1971 (36 FR 289), the agency cited

matches, cigarettes or short circuits in interior wiring as examples of

sources for fires occurring in the interior of vehicles. The agency

believes that there are situations where the straps could become

ignited. One example is children in the back seat of a car, playing

with matches, a cigarette lighter, or other ignition source near a

child restrained in a Cosco seat.

In point of act, had the tests been conducted under real life

circumstances, the results could have been worse. Webbing samples are

tested horizontally, but webbing is worn vertically. If a fire begins

at the bottom of webbing, it will travel upward at a faster rate than

it would in a horizontal placement.

NHTSA considered Cosco's argument that, if fire is ``already

consuming the child restraint and/or the vehicle seat upon which the

child restraint is installed,'' the child would be injured from these

sources as opposed to the shoulder straps of the restraint. This

argument cannot seriously be presented as ground for granting an

inconsequentiality petition. If a vehicle fire is of such intensity

that it is destroying a child restraint or vehicle seat that is

certified as complying with Standard No. 302, then it will destroy the

shoulder strap as well whether or not it complies. NHTSA is concerned

with fires of less intensity, where it is critical that interior

components (the child seat as well as the components specified in

Standard No. 301) do not ignite, or if they do, that they burn at a

slow enough rate that there will be time to remove the child from the

occupant compartment.

Cosco also argued that there was no real-world indication of a

safety threat. It said that:

Cosco has never received a report of the burning of a soft-

shield harness strap. Cosco is unaware of any study that indicates

that the burning of a child restraint harness has caused any injury

or death. All occupant protection studies which Cosco has reviewed

indicate an almost infinitesimal risk of injury or death by vehicle

fires in total, at least in collisions. Cosco is unaware of any data

on fires of the interior of vehicles unrelated to collisions.

In NHTSA's view, the fact that Cosco has not received any reports

is not a sufficient basis on which to grant its petition. The present

lack of such reports does not necessarily diminish the future potential

of such incidents.

NHTSA has, in fact, received a report which may have some relevance

in this matter. The complaint was made to the agency's Auto Safety

Hotline on August 30, 1993, reporting the burning of the belt of a

Cosco child seat while placed for three hours in a vehicle parked in

sunlight. According to the report, the claimant ``[n]oticed burning

fumes and found burn marks [brownish discoloration] on the harness

straps of the child restraint where the straps had been in contact with

the top edge of the restraint's plastic shell as they lay across it.''

This incident raises concern that the burned fiber of the strap may

have weakened the strength of the harness so that it might not provide

the needed safety protection for a child occupant during a crash.

Cosco's final major argument was the owners might not respond to a

future campaign of a more serious nature, if they were notified of one

that concerned only a technical noncompliance. It argued:

That child passenger safety advocates, child restraint

manufacturers, and the Agency are aware of the negative impact of

recalls resulting from technical noncompliance or defects that do

not, as a practical matter, have true safety consequences. The most

important negative effects of such recalls are:

1. That the public, because of the number and frequency of such

recalls, pays no attention to recalls that actually affect, in a

practical way, child passenger safety; and

2. That the public, upon seeing the number of recalls, concludes

that child restraints currently available are unsafe and therefore

decline to use them. The Agency is aware and, in fact, has publicly

advised consumers to use child restraints that have defects or

noncompliances that have that have resulted in recalls until such

child restraints can be corrected. [An example of this advisement is

contained in the docket.] This is in recognition of the fact that

technical noncompliances or relatively insignificant safety defects

do not compromise the overall effectiveness of child restraints.

The statement that consumers ignore recalls because of their number

and frequency is unsubstantiated. Further, NHTSA views it equally

unlikely that, because of campaigns, consumers would conclude that

child restraints are unsafe and decline to use them. Indeed, the

opposite is more likely the case. Responses to safety notifications

depend on factors including the type of noncompliance or defect, the

type and extent of the notification campaign, media coverage, and the

efforts of manufacturers.

Future campaigns are more likely to be effective than past ones.

Standard No. 213 has been amended to provide for the registration of

child restraints. The purpose of the program is to increase the

effectiveness of campaigns to recall child seats. It requires

manufacturers to take steps that will increase their ability to inform

owners of particular child restraints about problems in these

restraints and that encourage owners to register their child seats. And

the agency does not agree that the noncompliance is ``technical'' in

nature.

NHTSA also notes that the petition failed to acknowledge that the

agency tested and retested the harness webbing in March 1993, and

encountered a more serious test failure than the 4.3 inch burn rate

presented in the petition and which reflected NHTSA's original tests.

In the second series of tests, NHTSA found burn rate test and retest

failures of 5.4 and 5.2 inches respectively. Thus the test and retest

failures uncovered by NHTSA average 4.85 inches and 4.75 inches

respectively, a margin of failure of 20%, and hardly inconsequential or

of a ``technical'' nature in the agency's opinion.

Finally, NHTSA believes flammability requirements for child

restraints should be stringently adhered to for the following reasons.

The test requirement of not more than 4 inches a minute was justified

by the need ``to prevent injury to occupants from rapidly spreading

interior fires, to allow sufficient time for the driver to stop the

vehicle, and, if necessary, for occupants to leave it before injury

occurs'' (36 FR 10817). This is even more critical in the case of child

restraints as a small child is typically not capable of exiting a

vehicle without assistance. Therefore, some additional time is required

for another person to remove the child. Moreover, the child most often

is in the rear seat and the adult is in front, also requiring

additional time to reach the child. Finally, because webbing rests

against the child's body, noncompliant webbing has a great potential

for injuring the child if ignited.

For the reasons discussed above, the agency has concluded that the

petitioner has not met its burden of persuasion that the noncompliance

herein described is inconsequential as it relates to safety, and its

petition is denied.

(15 U.S.C. 1417; delegations of authority at 49 CFR 1.50 and 49 CFR

501.8)

Issued on March 22, 1994.

Barry Felrice,

Associate Administrator for Rulemaking.

[FR Doc. 94-7158 Filed 3-25-94; 8:45 am]

BILLING CODE 4910-59-M