Federal Motor Vehicle Safety Standards; Seat Belt Assemblies, 48883-48896 [05-16524]

Agencies

• National Highway Traffic Safety Administration
• DEPARTMENT OF TRANSPORTATION

[Federal Register Volume 70, Number 161 (Monday, August 22, 2005)]
[Rules and Regulations]
[Pages 48883-48896]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: 05-16524]


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

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Parts 571

[Docket No. NHTSA 2005-22052]
RIN 2127-AI38


Federal Motor Vehicle Safety Standards; Seat Belt Assemblies

AGENCY: National Highway Traffic Safety Administration (NHTSA), DOT.

ACTION: Final rule.

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

SUMMARY: This final rule amends the Federal motor vehicle safety 
standard (FMVSS) for seat belt assemblies to redefine the requirements 
and to establish a new test methodology for emergency-locking 
retractors. Specifically, this final rule establishes a new 
acceleration-time corridor, adds a figure illustrating the new 
acceleration-time corridor, provides a tolerance on angle measurements, 
and adopts the same instrumentation specifications currently found in 
other FMVSSs containing crash tests.

DATES: Effective Date: This final rule is effective October 21, 2005. 
The incorporation by reference of a certain publication listed in the 
regulation is approved by the Director of the Federal Register as of 
October 21, 2005.
    Compliance Date: Seat belt assemblies manufactured on or after 
February 22, 2007 must comply with this rule. Voluntary compliance is 
permitted prior to that date.
    Petitions for Reconsideration: If you wish to submit a petition for 
reconsideration of this rule, your petition must be received by October 
6, 2005.

ADDRESSES: Petitions for reconsideration should refer to the docket 
number above and be submitted to: Administrator, Room 5220, National 
Highway Traffic Safety Administration, 400 Seventh Street, SW., 
Washington, DC 20590.
    See the SUPPLEMENTARY INFORMATION portion of this document (Section 
VI; Rulemaking Analyses and Notice) for DOT's Privacy Act Statement 
regarding documents submitted to the agency's dockets.

FOR FURTHER INFORMATION CONTACT: For non-legal issues, you may call Mr. 
Christopher Wiacek, Office of Crashworthiness Standards (Telephone: 
202-366-4801) (Fax: 202-493-2290).
    For legal issues, you may call Mr. Eric Stas, Office of the Chief 
Counsel (Telephone: 202-366-2992) (Fax: 202-366-3820).
    You may send mail to these officials at National Highway Traffic 
Safety Administration, 400 Seventh Street, SW., Washington, DC 20590.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive Summary
II. Background
III. June 2004 Notice of Proposed Rulemaking (NPRM) and Public 
Comments
    A. The NPRM
    B. Summary of Public Comments on the NPRM
IV. The Final Rule and Response to Public Comments
    A. Summary of the Requirements
    B. Lead Time

[[Page 48884]]

    C. Response to Public Comments by Issue
    1. Acceleration-Time Corridor
    2. Data Acquisition
    3. Tolerances
    4. Request for Comments on Specific Issues
    5. Lead Time
    6. Other Issues
V. Benefits and Costs
VI. Rulemaking Analyses and Notices

I. Executive Summary

    In response to a petition for rulemaking, NHTSA published a notice 
of proposed rulemaking \1\ on June 3, 2004, which proposed to amend 
FMVSS No. 209, Seat Belt Assemblies, by redefining the requirements and 
establishing a new test methodology for emergency-locking retractors 
(ELRs). As noted above, the NPRM proposed to establish a new 
acceleration-time (A-T) corridor, to add a figure illustrating the new 
A-T corridor, to provide a tolerance on angle measurements, and to 
adopt the same instrumentation specifications currently found in other 
FMVSSs containing crash tests. The purpose of these proposed amendments 
was to clarify the test procedures for ELRs, while ensuring that those 
devices continue to perform their important safety function of locking 
up a seat belt in the event of a crash or emergency braking.
---------------------------------------------------------------------------

    \1\ 69 FR 31330 (June 3, 2004) (Docket No. NHTSA-2004-17980-1).
---------------------------------------------------------------------------

    After careful consideration of all available information, including 
public comments, the agency has decided to retain in this final rule 
the approach set forth in the NPRM, with minor technical modifications. 
All such modifications and the accompanying rationale are discussed 
fully in the balance of this document. The following points highlight 
the key changes to FMVSS No. 209 resulting from the final rule.
     The final rule modifies that portion of FMVSS No. 209's 
test procedures relevant to ELRs by adopting a new Figure 8, which 
provides a specified acceleration-time corridor for test pulses. The A-
T corridor includes an upper boundary onset rate of 375 g/sec and 
permits acceleration to peak at up to 0.8 g. The lower boundary of the 
A-T corridor allows for a minimum onset rate of 21.67 g/sec. The 
steady-state tolerance range is from 0.65 g to 0.72 g.
     During dynamic testing, the final rule requires each 
acceleration pulse to be recorded using an accelerometer having a full 
scale range of 10 g and to be processed according to the 
practices set forth in Society of Automotive Engineers (SAE) 
Recommended Practice J211-1 rev. December 2003, ``Instrumentation for 
Impact Test--Part 1--Electronic Instrumentation,'' Channel Frequency 
Class 60. (That SAE standard has been incorporated by reference into 
FMVSS No. 209.) Webbing displacement is required to be measured using a 
displacement transducer.
     Unless a range of angles is specified or a tolerance is 
otherwise explicitly provided, the final rule states that all angles 
and orientations of seat belt assemblies and components specified in 
the standard shall have a tolerance of 3 degrees.
    Manufacturers of seat belt assemblies must comply with the 
requirements of the final rule commencing on February 22, 2007. 
Voluntary compliance is permitted prior to the mandatory compliance 
date.
    In terms of the impacts, the agency anticipates that this final 
rule will not result in substantial changes to the performance of ELRs 
and that current ELRs will continue to comply with FMVSS No. 209 
without change. Instead, the final rule clarifies the specifications in 
the standard's test procedures. Furthermore, we expect that this rule 
will result in only a minimal cost burden to vehicle manufacturers. 
Testing laboratories might need to purchase new equipment, but this 
one-time cost is likewise expected to be minimal on a cost-per-vehicle 
basis.

II. Background

    The seat belt emergency-locking retractor is a device that was 
first developed in the 1960's for the purpose of maintaining occupant 
position during rapid vehicle deceleration. Since its inception, the 
ELR's locking sensitivity has been an important issue because of the 
need to assure that the retractor would lock very early during a 
collision or emergency braking, but not be so sensitive as to cause 
``nuisance'' locking during routine driving.
    Based upon the limited knowledge and technology available at that 
time, the SAE Motor Vehicle Seat Belt Committee (MVSBC) developed 
Recommended Practice SAE J-4b, Motor Vehicle Seat Belt Assemblies, and 
subsequently, SAE J-4c, Motor Vehicle Seat Belt Assemblies. These 
Recommended Practices provided performance requirements, laboratory 
test procedures, and minimal design requirements for seat belt 
assemblies for use in motor vehicles, in order to minimize the risk of 
bodily harm in an impact. In promulgating FMVSS No. 209, NHTSA 
ultimately adopted SAE J-4c, although the test methodologies for ELRs 
developed by SAE were not clearly defined. As a result, the test 
methodology, instrumentation, and measurements for assessing 
conformance were not explicitly described in S4.3(j) and S5.2(j) of 
FMVSS No. 209. This situation has not changed appreciably since 
adoption of our safety standard in a final rule published in the 
Federal Register on February 3, 1967.\2\
---------------------------------------------------------------------------

    \2\ 32 FR 2408, 2415 (Feb. 3, 1967).
---------------------------------------------------------------------------

    Currently, there are two modes of ELR sensors in production: (1) 
webbing withdrawal-sensitive ELRs and (2) vehicle acceleration-
sensitive ELRs. The latter mode of a retractor responds directly to a 
0.7 g acceleration pulse, and lock-up usually occurs within a short 
period of time. The former mode of a retractor responds to the webbing 
withdrawal speed, which slowly builds up from zero to the threshold 
(i.e., lock-up) speed, when the assembly is subjected to the 0.7 g 
acceleration pulse. As a result, a longer time period may be required 
for the webbing-sensitive type of retractor to respond.
    Despite the two different basic ELR designs, FMVSS No. 209 has a 
unified set of requirements for compliance testing. Specifically, under 
S4.3(j)(1) of FMVSS No. 209, an emergency-locking retractor of a Type 1 
or Type 2 seat belt assembly,\3\ when tested in accordance with 
S5.2(j), ``shall lock before the webbing extends 25 mm when the 
retractor is subject to an acceleration of 7 m/s2 (0.7 g).'' 
Paragraph S5.2(j) of the standard states in relevant part that ``[t]he 
retractor shall be subject to an acceleration of 7 m/s2 (0.7 
g) within a period of 50 milliseconds (ms), while the webbing is at 75 
percent extension[.]''
---------------------------------------------------------------------------

    \3\ Under S3 of FMVSS No. 209, a ``Type 1 seat belt assembly'' 
is defined as ``a lap belt for pelvic restraint,'' and a ``Type 2 
seat belt assembly'' is defined as ``a combination of pelvic and 
upper torso restraints.''
---------------------------------------------------------------------------

    In addition, FMVSS No. 209 establishes a sensitivity threshold for 
ELRs to prevent ``nuisance'' locking during routine driving. Under 
S4.3(j)(2), an ELR sensitive to vehicle acceleration must not lock up 
when the retractor is rotated in any direction to any angle 15 degrees 
or less. Under S4.3(j)(3), an ELR sensitive to webbing withdrawal must 
not lock up before the webbing extends 51 millimeters (mm) when the 
retractor is subject to an acceleration of 0.3 g or less.
    Based upon FMVSS No. 209, the agency developed a laboratory test 
procedure for its compliance laboratories to follow, which provides 
more detail concerning test set up. The most recent version, TP-209-
05,\4\ was issued on January 17, 2003. In relevant

[[Page 48885]]

part, that laboratory test procedure specifies the use of a 0.72 g 
acceleration pulse, which is intended to ensure that the retractor will 
be subject to at least 0.7 g during testing, as required by the 
standard. This test pulse accounts for calibration and accuracy ranges 
of the test equipment.
---------------------------------------------------------------------------

    \4\ See https://www.nhtsa.dot.gov/staticfiles/DOT/NHTSA/
Vehicle%20Safety/Test%20Procedures/Associated%20Files/TP-209-05.pdf.
---------------------------------------------------------------------------

    In order to gain a better understanding of the seat belt emergency-
locking retractor test procedures and performance requirements, the 
Automotive Occupant Restraints Council (AORC \5\) wrote a letter to 
NHTSA requesting an interpretation of S4.3(j) and S5.2(j) of FMVSS No. 
209. The AORC stated that neither the SAE Committee nor NHTSA addressed 
the onset rate range and the deceleration tolerance for ELRs when their 
respective standards were developed or since that time. The AORC stated 
its belief that the intent of both the SAE Committee and NHTSA at the 
time FMVSS No. 209 was adopted was to mimic a hard braking deceleration 
pulse in which the 0.7 g level should be achieved with a sharp onset 
rate, followed by steady-state deceleration. NHTSA responded through an 
interpretation letter to Mr. Steven Fredin dated February 4, 2000.\6\ 
However, the AORC did not agree with the position expressed in the 
interpretation letter and subsequently submitted a petition for 
rulemaking on June 2, 2000.\7\
---------------------------------------------------------------------------

    \5\ The Automotive Occupant Restraints Council is an industry 
association of 49 suppliers of occupant restraints, components/
materials, and services to the automobile industry.
    \6\ In the February 4, 2000 letter of interpretation, the agency 
stated:
    Nothing in the standard purports to require a consistent 
acceleration (or a constant rate of increase of acceleration), to 
establish a specific period during which the acceleration must be 
maintained, or to prohibit any ``decay'' after the 0.7 g level is 
reached. Therefore, each retractor must be able to meet the locking 
requirements of the standard regardless of the rate of acceleration, 
or the extent of any subsequent ``decay.''
    See https://www.nhtsa.dot.gov/cars/rules/interps/files/
aorc3.ogm.html.
    \7\ Docket No. NHTSA-2127-2000-7073-12.
---------------------------------------------------------------------------

    The AORC petition requested that NHTSA amend paragraphs S4.3(j) and 
S5.2(j) of FMVSS No. 209 to specify: (A) a rate of onset; \8\ (B) an 
acceleration pulse duration; (C) an acceleration tolerance level, and 
(D) a subsequent acceleration decay.\9\ In addition, the AORC requested 
that NHTSA apply the same instrumentation specifications to those 
provisions as are used in other FMVSSs with dynamic performance 
requirements.
---------------------------------------------------------------------------

    \8\ ``Onset rate'' is defined as the rate (in g/sec) at which 
the seat belt retractor is initially accelerated from time zero.
    \9\ ``Acceleration decay'' is defined as the rate (in g/sec) at 
which the retractor acceleration is returned to zero.
---------------------------------------------------------------------------

    The AORC argued that it is necessary to amend the standard because 
many acceleration pulses conform to S4.3(j) and S5.2(j) in theory, but 
those pulses would cause retractors, currently compliant under FMVSS 
No. 209, to fail the locking requirements within the 25 mm webbing 
payout. Furthermore, the AORC asserted that NHTSA's interpretation 
letter permits testing methodologies that no known ELR could possibly 
meet. The petition provided several example pulses that, according to 
the AORC, would conform to the criteria in the interpretation letter, 
but would not be sufficient to consistently lock a production 
retractor.
    In suggesting a means of addressing these concerns, the AORC 
petitioned that S5.2(j) should include a specific A-T corridor, with 
maximum and minimum acceleration onset rates matching those specified 
in the Economic Commission for Europe Regulation No. 16, Uniform 
Provisions Concerning the Approval of: Safety Belts and Restraint 
Systems for Occupants of Power-Driven Vehicles and Vehicles Equipped 
with Safety Belts (ECE R16). The AORC also stated that the acceleration 
and the webbing displacement recording techniques should conform to SAE 
Recommended Practice J211-1 rev. March 1995, ``Instrumentation for 
Impact Test--Part 1--Electronic Instrumentation'' (SAE J211-1, rev. 
Mar. 1995). In addition, the AORC petition stated that the safety 
standard should require that the signals should be filtered with an SAE 
Class 60 filter, and that the accelerometer should be an 
instrumentation-grade, high-accuracy, 10 g device. The AORC 
contended that the addition of an A-T corridor and specification of the 
test methodology and instrumentation, in a manner consistent with its 
petition, would create needed objectivity and fully clarify the 
standard in this area.
    NHTSA granted the AORC's petition to clarify the relevant 
provisions of FMVSS No. 209.

III. June 2004 Notice of Proposed Rulemaking (NPRM) and Public Comments

A. The NPRM

    As noted above, on June 3, 2004, NHTSA published an NPRM, which 
proposed to address the issues raised by the AORC in its petition for 
rulemaking. The NPRM provided a 60-day public comment period, which was 
subsequently extended.\10\ In general, the NPRM proposed to redefine 
certain requirements of FMVSS No. 209 to establish a new test 
methodology for emergency-locking retractors. To accommodate the time 
needed for vehicle manufacturers and testing laboratories to 
reconfigure their testing equipment in conformity with the proposed 
amendments, NHTSA proposed that the final rule would provide lead time 
of one year. The following discussion highlights the key provisions of 
the proposal.
---------------------------------------------------------------------------

    \10\ The NPRM provided a public comment period through August 2, 
2004. However, in a letter dated July 14, 2004, the AORC petitioned 
for a 60-day extension of the comment period in order to provide 
time for the gathering of additional technical information in 
response to the NPRM's proposed provisions (Docket No. NHTSA-2004-
17980-4). On August 4, 2004, the agency published a notice in the 
Federal Register to extend the public comment period from August 2, 
2004 to October 1, 2004, to allow the industry additional time to 
generate data relevant to the proposal (69 FR 47075) (Docket No. 
NHTSA-2004-17980-5).
---------------------------------------------------------------------------

Rate of Onset
    The agency proposed a new acceleration corridor with an increased 
maximum onset rate, which represents a modified version of the A-T 
corridor suggested by the AORC in its petition. The proposed corridor 
was sufficiently wide as to allow a range of onset rates to be tested 
that were preliminarily determined to be more representative of real 
world crashes and emergency braking events. The NPRM proposed a maximum 
onset rate of 375 g/sec and a minimum onset rate of 16.25 g/sec, which 
would accommodate purely linear pulses during the first 50 ms interval.
    Although the agency found that the onset rate for various crash 
test pulses varied greatly (from over 1,000 g/sec for crash pulses to 2 
g/sec for emergency braking pulses), the agency tentatively decided 
that its proposed maximum onset rate would capture pulses that 
historically have been used for ensuring a minimum level of safety 
performance for the ELR in vehicle seat belts along with a wide range 
of acceleration pulses (including those used by the agency's compliance 
testing laboratories). As a result, the agency tentatively concluded 
that the proposed A-T corridor would permit the generation of 
repeatable and reproducible acceleration pulses and that the proposed 
onset rate corridor should eliminate the potentially problematic 
``theoretical'' test pulses cited by the AORC, while at the same time 
maintaining the integrity of FMVSS No. 209.
Acceleration Pulse Duration
    The NPRM did not propose a minimum time duration for the test 
pulse, as had been requested by the

[[Page 48886]]

AORC in its petition. The agency reasoned that once the onset rate of 
the acceleration pulse is given, the pulse duration that is required to 
produce a 25 mm webbing payout is implicitly determined. Therefore, a 
pulse time duration specification is not necessary.
Acceleration Tolerance Level
    Based upon current compliance test data, the agency proposed that 
an initial peak above 0.7 g should be allowed within the first 50 ms 
time period of the test pulse. The proposed A-T corridor would have an 
upper bound of 0.8 g from 2 ms to 50 ms to allow the initial peak to 
exceed 0.7 g prior to reaching a ``steady-state'' response. For the 
remainder of the A-T corridor (i.e., from 50 ms to the end of the 
test), the A-T corridor would be specified at 0.7 g with a +0.02/-0.05 
g tolerance boundary (i.e., a tolerance range between 0.72 g and 0.65 
g), which is consistent with NHTSA's current compliance test procedures 
and test data. As discussed in the NPRM, the agency expected that the 
proposed A-T corridor would simulate the worst-case test condition, 
similar to those observed in laboratory hard (emergency) braking tests, 
while recognizing that acceleration may peak before reaching a 
``steady-state'' condition.
Subsequent Acceleration Decay
    In the NPRM, the agency stated that the proposal addresses the 
AORC's concerns about rapid acceleration decay after the initial peak, 
even though we did not include a specification for acceleration decay 
(i.e., pulse shape and duration). The NPRM stated that the lower 
boundary of the proposed A-T corridor would prevent the use of 
acceleration pulses that have early, rapid acceleration decay. 
Furthermore, after either a lock-up occurs or the webbing payout 
reaches 25 mm, the test is officially over. The acceleration pulse 
after this point does not affect the test results and is no longer a 
concern to test accuracy (i.e., after this point, it is permissible for 
the pulse to cross the lower boundary of the corridor).
Test Procedures and Measurement Specification
    In agreement with the AORC petition, the NPRM proposed that the 
acceleration specifications under FMVSS No. 209 be recorded and 
processed according to the practices specified in SAE J211-1, rev. 
March 1995. Specifically, the proposal stated that the instrumentation 
used to record the A-T history and the webbing payout would be in 
conformance with the instrumentation requirements of SAE J211-1, rev. 
March 1995, that the electronic signals would be filtered with an SAE 
Class 60 filter, and that the accelerometer used for retractor testing 
would be an instrumentation-grade, high-accuracy, 10 g 
device. The proposed instrumentation requirements were the same as 
those currently specified in other FMVSSs with a dynamic performance 
component.
    As part of the proposed test procedures, the NPRM specified use of 
a displacement transducer to directly measure and record webbing 
displacement, thereby eliminating the uncertainty inherent in indirect 
measurement techniques (e.g., numerical integration of accelerometer 
data). In addition, the NPRM's proposed test procedures included a 
tolerance of 3 degrees for all angles and orientations of 
the seat belt assemblies and component, unless a range of angles is 
otherwise specified.
``Nuisance'' Locking
    In order to address the issue of ``nuisance locking,'' the NPRM 
proposed to amend S4.3(j)(2) of FMVSS No. 209's test procedures to 
require retractors sensitive to webbing withdrawal to be subjected to 
an acceleration of 0.3 g occurring within a period of the first 50 ms 
and sustaining an acceleration no greater than 0.3 g throughout the 
test, while the webbing is at 75 percent extension.
Request for Comments on Specific Questions
    In addition to the matters discussed above, the NPRM requested 
responses to several questions regarding the ability of current ELRs to 
comply with the proposed A-T corridor, methods used by the industry to 
determine when ELR lock-up occurs, and potential modifications to the 
proposal (e.g., narrowing the A-T corridor).

B. Summary of Public Comments on the NPRM

    NHTSA received six comments on the June 3, 2004 NPRM from a variety 
of interested parties including an industry association (the AORC), 
suppliers (Renfroe Engineering, Inc.; TK Holdings, Inc.), a vehicle 
manufacturer (Ford Motor Company (Ford)), a public interest group 
(Public Citizen), and an individual (Dr. Ave Ziv). All of these 
comments may be found in Docket No. NHTSA-2004-17980.
    The commenters generally supported the proposal but suggested a 
number of modifications to the proposed requirements, including ones 
related to the A-T corridor, the data acquisition methodology and 
related equipment, tolerances, requirements for dual-sensing 
retractors, and lead time. The following discussion summarizes the main 
issues raised by these public comments and the positions expressed on 
these topics. A more complete discussion of the public comments is 
provided under Section IV.C, which provides an explanation of the 
agency rationale for the requirements of the final rule and addresses 
related public comments by issue.
    At least one commenter acknowledged that existing ELRs would 
continue to comply with FMVSS No. 209 if the proposed A-T corridor were 
to be adopted, although another commenter (Ford) argued that the 
corridor is overly broad and, therefore, not objective. Overall, 
however, commenters recommended adoption of the A-T corridor with 
certain modifications. For example, one commenter recommended 
redefining the lower corridor, because of concerns that a lower onset 
rate could result in nuisance locking, and providing a longer locking 
distance. In terms of the upper portion of the corridor, at least one 
commenter supported the proposed upper boundary; however, another 
commenter argued that the high maximum onset rate is unrealistic in 
light of the more limited capabilities of existing test equipment, and 
it recommended a new upper corridor with a maximum onset rate of 150 g/
sec.
    One commenter sought modifications to the range of the A-T corridor 
after 50 ms, such that 0.7 g is at the center of the upper and lower 
limits of the corridor. Commenters generally agreed with the proposal 
to allow acceleration decay outside of the proposed corridor after the 
compliance test is completed.
    There were several comments pertaining to the proposed data 
acquisition requirements, including the following points. There was 
support for the use of an SAE Class 60 filter. Commenters also 
supported use of SAE Recommended Practice J211-1, although there was a 
recommendation to use a more recent December 2003 version of that 
standard, which provides a more detailed test methodology. One 
commenter recommended use of a 20 g accelerometer, rather 
than the 10 g accelerometer proposed in the NPRM.
    Regarding the angle tolerances of 3 degrees proposed in 
the NPRM, commenters generally supported such a tolerance for most 
applications, unless a range is specified. However, commenters 
requested a tighter tolerance of 0.5 degrees for angles and 
orientations specifically addressed in the proposal, in order to 
prevent the

[[Page 48887]]

need to redesign currently compliant ELRs to account for such 
tolerance.
    Commenters also raised some issues not covered by the NPRM, such as 
requiring a seat belt assembly with dual-sensing retractors to comply 
with the standard for both designs, including the no-lock test at low 
accelerations. Another commenter requested specification of a defined 
A-T corridor for the no-lock requirement for accelerations no greater 
than 0.3 g.
    Regarding lead time, commenters that addressed this issue requested 
that lead time be extended to 18 months, from the 12 months proposed in 
the NPRM, in order to provide companies with additional time to 
purchase and install new equipment, if necessary, to ensure compliance 
with the amended standard.

IV. The Final Rule and Response to Public Comments

A. Summary of the Requirements

    After careful consideration of the public comments, in this final 
rule amending FMVSS No. 209, we are adopting the approach set forth in 
the June 2004 NPRM, with certain modifications. In general, this rule 
redefines the requirements and establishes a new test methodology for 
emergency-locking retractors. The standard is intended to be 
technology-neutral, so as to permit compliance with any available ELR 
technology that meets the standard's performance requirements.
    The following points highlight the key change resulting from the 
final rule.
     The final rule modifies that portion of FMVSS No. 209's 
test procedures relevant to ELRs by adopting a new Figure 8 which 
provides a specified acceleration-time corridor for test pulses. The A-
T corridor includes an upper boundary onset rate of 375 g/sec and 
permits acceleration to peak at up to 0.8 g. The lower boundary of the 
A-T corridor allows for a minimum onset rate of 21.67 g/sec. The 
steady-state tolerance range is from 0.65 g to 0.72 g.
     During dynamic testing, the final rule requires each 
acceleration pulse to be recorded using an accelerometer having a full 
scale range of 10 g and to be processed according to the 
practices set forth in SAE Recommended Practice J211-1 rev. December 
2003, ``Instrumentation for Impact Test--Part 1--Electronic 
Instrumentation,'' Channel Frequency Class 60. (That SAE standard has 
been incorporated by reference into FMVSS No. 209.) Webbing 
displacement is required to be measured using a displacement 
transducer.
     Unless a range of angles is specified or a tolerance is 
otherwise explicitly provided, the final rule states that all angles 
and orientations of seat belt assemblies and components specified in 
the standard shall have a tolerance of 3 degrees.

B. Lead Time

    Consistent with the request of commenters, the agency has decided 
to provide 18 months of lead time for manufacturers to meet the 
requirements of the amended standard. Accordingly, compliance with the 
requirements of the final rule commences for seat belt assemblies 
manufactured on or after February 22, 2007. Voluntary compliance is 
permitted prior to the mandatory compliance date.

C. Response to Public Comments by Issue

    As noted previously, public comments on the June 2004 NPRM to amend 
FMVSS No. 209 raised a variety of issues with the NPRM's proposed 
requirements. Each of these topics will be discussed in turn, in order 
to explain how these comments impacted the agency's determinations in 
terms of setting requirements for this final rule.
1. Acceleration-Time Corridor
    The NPRM proposed an A-T corridor with a maximum onset rate of 375 
g/sec, a minimum onset rate of 16.25 g/sec, and a width sufficient to 
accommodate acceleration test pulses preliminarily determined to be 
representative of real world crashes and emergency braking events. The 
proposal also provided an acceleration tolerance that would permit the 
pulse to attain an upper bound peak of 0.8 g within the first 48 ms 
corridor (i.e., between 2 ms and 50 ms) prior to reaching a steady-
state response. For the remainder of the A-T corridor, the NPRM 
proposed 0.7 g with a +0.02/-0.05 tolerance boundary. (See Figure 8 of 
the NPRM.) The agency did not deem it necessary to specify a minimum 
time duration for the acceleration pulse or a specification for 
acceleration decay (i.e., pulse shape and duration).
    A number of commenters raised concerns about the proposed A-T 
corridor, including the AORC, TK Holdings, Ford, and Dr. Ziv. The AORC 
commented that the NPRM's expansion of the A-T corridor beyond the 
boundaries originally recommended in its petition for rulemaking is 
unnecessary. Specifically, the AORC objected to the NPRM's proposed 
lower onset rate, because the AORC believes that static friction in the 
ELR, coupled with the low onset rate, could result in nuisance locking 
during routine driving. To address its concern, the AORC developed a 
new lower A-T corridor as part of its comment submission, which 
reflects a compromise between the AORC's original suggested boundary 
and the one proposed in the NPRM. (TK Holdings supported such a 
compromise approach in its comments.)
    The AORC further commented that if a lower onset rate were to be 
adopted, a longer locking distance would be required. To illustrate its 
point, the AORC argued that with an onset rate of 13 g/sec, the ELR 
would have 21.5 mm of payout available to lock up once it reached 0.7 
g, as compared to 25 mm of payout being available for an ELR 
experiencing a nearly instantaneous rise to 0.7 g.
    Regarding the upper boundary of the proposed A-T corridor, 
commenters expressed divergent viewpoints. TK Holdings concurred with 
the upper boundary presented in the NPRM. However, the AORC objected to 
the high onset rate (i.e., 375 g/sec). Although the AORC acknowledged 
that high onset rates do occur during high-speed barrier crashes, it 
argued that these tests serve the purpose of demonstrating performance 
under these conditions, so no component-level test is necessary. In 
addition, the AORC argued that it does not know of any commercially-
available, component-level test equipment that can reliably conduct a 
test with an onset rate above 200 g/sec. As an alternative, the AORC 
developed and submitted a new upper corridor, which: (1) Adopts the 
agency's upper corridor limit of 0.8 g; (2) modifies the limit along 
the ``sustain'' portion at the end of the test to 0.75 g (i.e., the 
portion of the A-T corridor in which the steady-state response should 
have been achieved), and (3) provides a maximum onset rate of 150 g/
sec.
    TK Holdings expressed concern about the range of the corridor after 
50 ms, arguing that the boundary should be controlled such that 0.7 g 
is at the center of the upper and lower limit of the corridor. 
Accordingly, TK Holdings recommended a range of 0.7 g 0.05 
g for the corridor after 50 ms.
    The AORC and TK Holdings agreed with the agency's proposal to allow 
acceleration decay outside the proposed corridor after the compliance 
test is complete.
    Ford commented that the NPRM's proposed A-T corridor is not 
objective because it is overly broad and that other concerns about test 
objectivity have not been adequately addressed. For example, Ford 
expressed concern that an agency contracting laboratory could choose an 
audit test pulse that is substantially different from the pulse

[[Page 48888]]

selected by the manufacturer. The company requested that the agency 
demonstrate a safety need for test pulses that are both more severe and 
less severe than those within the A-T corridor originally recommended 
by the AORC. Ford stated that if the agency does identify a safety need 
for the augmented regions of the A-T plot, that there should be 
additional, objectively-defined corridors to assess ELR compliance.
    In his comments, Dr. Ziv sought clarification as to whether a 
retractor must meet the requirements for any acceleration pulse within 
the proposed corridor, or at least one acceleration pulse within the 
corridor.
    In response to these comments, the agency has decided to modify the 
lower boundary of the A-T corridor in the manner suggested by the AORC 
in its latest submission. NHTSA's intention in proposing the lower 
boundary in the NPRM was to ensure that it encompassed current test 
pulses, particularly those with slower onset rates. Although the AORC 
did not provide any data to demonstrate the nature and extent of this 
``nuisance locking'' problem, we believe that the AORC's proposed new 
lower boundary would address the concern of potential ``nuisance 
locking,'' while maintaining inclusion of all current test pulses. In 
addition, we believe that the new lower A-T corridor should minimize 
the variation in onset rates, while maintaining the repeatability and 
reproducibility of the test procedures.
    Regarding comments on the upper corridor boundary, the agency has 
decided to adopt, as part of this final rule, the same upper corridor 
boundary that was presented in the NPRM. High onset rates do occur in 
crashes, and even though current equipment cannot generate pulses of 
that magnitude, technological developments may permit generation of 
such pulses in the future. The agency believes that a high onset rate 
limit is not detrimental to current ELR performance or vehicle safety. 
Instead, we believe that it is advantageous for manufacturers to reach 
0.7 g in the shortest time period possible, because that would make the 
maximum amount of webbing payout available to achieve compliance. In 
addition, we believe that the specificity in the final rule's data 
acquisition methodology (discussed below) will prevent the generation 
of unreliable test pulses with overly-high onset rates.
    Although the maximum onset rate recommended by the AORC would 
(barely) encompass current test pulses, we do not believe that the AORC 
has demonstrated a need for its recommended change. In addition, the 
AORC did not provide evidence to demonstrate a compliance problem with 
its test pulses to meet a steady-state tolerance between 0.65 g and 
0.72 g, as would justify its request to change the upper limit on the 
``sustain'' portion of the boundary to 0.75 g; all test pulses included 
in the AORC's comments fell within the proposed tolerance, and the 
pulses generated by the agency during compliance testing similarly fell 
within that range.
    In response to the AORC's comment regarding adoption of a longer 
locking distance, we have decided that such an amendment is not 
necessary for this new lower corridor. We believe that the test pulses, 
arising under the final rule, would provide sufficient onset rates to 
adequately permit enough webbing payout to comply with the standard.
    We do not agree with Ford's opinion that the proposed A-T is overly 
broad and, therefore, not objective. NHTSA did not have an issue with 
performance of the existing test pulses used for compliance purposes. 
We found that those acceleration pulses have proven repeatable, 
reproducible, and indicative of pulse experience in the real world. The 
proposed A-T corridor was developed to ensure inclusion of these 
pulses, and in contrast to Ford's characterization, the proposed A-T 
corridor actually narrows the range of potential test pulses and 
addresses potential problems arising from the need to certify to 
theoretical pulses that might not exist in real world events. We 
believe that the proposed test corridor (further narrowed in the final 
rule through adoption of the AORC's newly suggested lower boundary) is 
objective because it clearly delineates which pulses are valid for the 
test procedure, thereby helping to meet the safety need of ensuring 
proper ELR lock-up. Furthermore, Ford did not state the criteria it 
believes necessary to define a corridor narrow enough to be objective. 
We would also note that, by definition, a corridor will accommodate 
more than one pulse; therefore, there will always be the possibility 
that the agency will choose to test a different pulse than the 
manufacturer.
    In response to Dr. Ziv's comment, we would clarify that the ELR 
must meet the standard's requirements for any and all acceleration 
pulses that could be generated within the A-T corridor. Otherwise, 
proper functioning of the ELR could be limited to a highly targeted 
subset of the conceivable test pulses than would otherwise occur in 
actual crash events.
2. Data Acquisition
    The NPRM proposed that the acceleration specifications under FMVSS 
No. 209 be recorded and processed according to the practices specified 
in SAE J211-1, rev. March 1995. It also proposed to require electronic 
signals to be filtered with an SAE Class 60 filter and use of an 
instrumentation-grade, high-accuracy 10 g accelerometer. 
The proposal also called for use of a displacement transducer to 
measure webbing displacement. (See S5.2(j)(3) of the NPRM.)
    While generally supporting the aspect of the agency's proposal that 
would require proper filtering, TK Holdings recommended that, as part 
of the final rule, NHTSA require use of a 20 g full-scale 
accelerometer because of the potential for damage to a 10 g 
accelerometer during testing.
    Both the AORC and Ford supported specification of the SAE Class 60 
filter. However, they commented that NHTSA should further define the 
accelerometer type and that hardware/digital filters should be added in 
order to ensure objective test results. The AORC stated that in order 
to ensure meaningful comparisons, the data acquisition process must 
include identical sample rate, accelerometer sizing/type, and 
filtering. Accordingly, the AORC recommended adoption of a newer 
version of SAE J211-1 (December 2003), which was issued since the time 
of its initial petition, because the AORC believes that the updated 
versions of the SAE standard provides a more detailed data acquisition 
methodology; the AORC's view is that this change would help preclude 
the use of erroneous test conditions and facilitate correlation of data 
between test laboratories.
    On another matter related to data acquisition, the AORC commented 
that the preamble of the NPRM discussed ``direct measurement of webbing 
displacement,'' but that related language was not incorporated into the 
proposed regulatory text. The AORC concurred with NHTSA that indirect 
measurement of webbing displacement by means of numeric integration 
could impart a degree of uncertainty to the results. The AORC suggested 
that it is unnecessary to accept such uncertainty, because all modern 
acceleration sleds utilized by the restraints industry and independent 
test laboratories use high-precision and high-accuracy linear 
displacement transducers. By nature of these instruments, the AORC 
argued that no interpretation or filtering is necessary. According to 
the AORC, test laboratories use one of two designs to measure

[[Page 48889]]

webbing payout: (1) A pinch roller mechanism that acts directly on the 
webbing, with a transducer at the roller to measure webbing movement, 
or (2) a displacement transducer on a sled carriage that moves in a 
linear direction. The AORC suggested that NHTSA should add this 
information to the Laboratory Test Procedure for FMVSS No. 209.
    In response to these comments, NHTSA has decided to make certain 
modifications in the final rule. We concur with the commenters that, 
with the development of the A-T corridor, the test procedures should be 
specific enough to ensure repeatability and reproducibility and that a 
more detailed data acquisition methodology would help preclude variance 
among testing laboratories and would improve test objectivity and 
enforceability. To this end, we have decided to adopt the AORC's 
recommendation to utilize SAE J211-1 (Dec. 2003 version), which we are 
incorporating by reference in FMVSS No. 209.
    We also agree with the AORC that filtering is not necessary for 
data related to webbing payout, in light of the direct measurement 
equipment utilized by the industry. The agency's compliance test 
laboratories currently utilize high-precision and high-accuracy 
displacement transducers to directly measure webbing payout, thereby 
eliminating the need for numeric integration and data filtering. 
Accordingly, we have eliminated the statement in S5.2(j)(3) of the NPRM 
which had provided, ``The displacement data shall be processed at 
Channel Frequency Class 60.''
    However, we have decided not to adopt TK Holdings' recommendation 
that we adopt a 20 g full-scale accelerometer, because we 
do not believe that such device is necessary for the present 
application. The commenter did not provide any supporting data to 
demonstrate that current 10 g accelerometers are at a high 
risk for damage, and the agency is unaware of any accelerometer 
failures at its compliance test laboratories due to an overshoot in the 
acceleration pulse. Furthermore, we are concerned that the precision of 
the pulse up to 0.7 g would be diminished by switching to an 
accelerometer with a larger range. Accordingly, we have decided to 
retain the requirement for use of a 10 g accelerometer.
3. Tolerances
    The NPRM proposed to require a tolerance of 3 degrees 
for all angles and orientation of the seat belt assemblies and 
components, unless otherwise specified. (See S5.4 of the NPRM.)
    On the issue of tolerances, the AORC, TK Holdings, and Ford all 
concurred that NPRM's proposed angle tolerances should not apply to 
requirements where a range of angles is specified. However, these 
commenters argued that the proposed tolerance of 3 degrees 
is inappropriate for certain provisions of the standard, because it 
would necessitate a more sensitive ELR design, in order to compensate 
for mounting error during testing. The commenters stated that ELR 
designs with increased sensitivity are likely to be more nuisance-
prone. For this reason, the AORC and TK Holdings recommended a 
tolerance level of 0.5 degrees for the angles and 
orientations specifically addressed in the NPRM.
    We agree with the commenters that a tolerance level of 3 degrees for certain angle and orientation requirements might 
drive nuisance-prone ELR designs. Excessive tolerance, beyond the 
minimum level that is consistent with the ability of the test 
equipment, could introduce more error into the test procedure, thereby 
forcing unwanted compensation in the design of the ELR. Accordingly, we 
have decided to modify the relevant provisions in S5.2(j)(2) of the 
final rule to explicitly provide a tolerance level of 0.5 
degrees for all angle and orientation requirements contained in that 
paragraph. The language of S5.4, ``Tolerance on angles,'' has also been 
modified to reflect this change.
4. Request for Comments on Specific Issues
    As noted above, the NPRM requested responses to several questions 
regarding the compliance of current ELRs to the proposed A-T corridor 
and methods that could be employed to accurately determine when ELR 
lock-up occurs. Each of the questions posed in the NPRM is repeated 
below, followed by the comments received on that issue, if any.
     The AORC suggested a corridor more narrowly defined at the 
beginning (i.e., a 0-4 ms window). Would a narrower corridor as 
suggested by the AORC be feasible? Would a narrower corridor more 
accurately specify the A-T onset?
    The AORC provided another suggested A-T corridor which was broader 
than the one it originally suggested. Specifically, the AORC extended 
the bottom portion of the corridor from 0-4 ms to 0-10 ms, in order to 
accommodate a potential lag in the initiation of the test pulse. 
However, the AORC's newly recommended corridor was narrower than the 
one proposed in the NPRM.
     Would any currently compliant emergency-locking retractor 
be unable to comply under the proposed corridor?
    TK Holdings responded by stating that all of its currently 
compliant ELR seat belt assemblies would comply with the A-T corridor 
proposed in the NPRM.
     Is 50 ms at the beginning of the time period sufficient to 
allow for an initial peak above 0.7 g limit?
    In response to this question, TK Holdings stated that 50 ms 
provides sufficient time to reach 0.7 g.
    The agency notes that in this final rule, we have modified the 
lower boundary of the A-T corridor such that the initial peak must be 
obtained within 40 ms. However, we do not believe that this 
modification will impact any existing compliant ELR because agency data 
show that current acceleration pulses reach 0.7 g well before 40 ms.
     ELR lock-up occurs when rotation of the ELR gear assembly 
stops. The methods employed by test laboratories to determine ELR lock-
up are indirect methods rather than direct measurement of the ELR gear. 
In general, an ELR lock-up occurrence is determined by the observation 
of a sudden change in sled acceleration-time curve. Thus, the exact 
time of lock-up is subject to test laboratory's interpretation of this 
event. We are requesting input on methods that can be employed in our 
test procedures to accurately determine when ELR lock-up occurs. Your 
response should include the following:
    (a) The type of sensing device and/or test equipment to be employed 
for detecting lock-up.
    (b) Any procedures for performing a lock-up test. Please provide 
technical support.
    (c) Any criteria used to evaluate the lock-up condition. Please 
provide technical support.
    The AORC and TK Holdings both responded to this question by 
suggesting the use of a threshold load, which they stated is consistent 
with current industry practice. According to the commenters, a typical 
set-up includes a belt load sensor in the webbing path between the 
fixed webbing end and the retractor. They stated that the standard 
industry practice is to use a 35 Newton (N) 10 N belt load 
to indicate that a lock-up has occurred. However, the AORC argued that 
an additional 3-5 mm of allowable webbing payout is necessary to 
account for the additional webbing travel between the actual lock-up 
time and the time it takes to achieve a 35 N load on the webbing.

[[Page 48890]]

    NHTSA understands that there is currently more than one methodology 
in use for determining ELR lock-up. Some laboratories use the industry 
standard (i.e., a 35 N threshold), while others determine lock-up 
through observation of a sudden change in the A-T curve. In the final 
rule, we have decided not to specify a required method for determining 
ELR lock-up for the following reasons. First, the industry load 
threshold approach is also an indirect measurement of lock-up, and the 
agency does not have sufficient technical information to assess and 
adopt that approach. Furthermore, we have not heard of any problems 
associated with existing methods for determining ELR lock-up.
5. Lead Time
    The NPRM proposed to provide affected entities with lead time of 
one year from the time of publication of a final rule to meet the 
requirements of the amended standard.
    The AORC and TK Holdings requested that the lead time for 
compliance with the final rule's requirements be extended from 12 
months, as proposed, to 18 months. The commenters stated that such 
additional time is necessary to permit companies to purchase and to 
install new equipment, if necessary, to ensure compliance with the 
amended standard.
    NHTSA has decided to extend the compliance date with these 
amendments to FMVSS No. 209 to 18 months after the date of issuance of 
this final rule, as requested by the commenters. Because we do not 
anticipate that the changes contained in this final rule would have any 
significant impact upon the effectiveness or compliance of existing 
ELRs, we believe that it is appropriate to afford companies additional 
time to purchase and configure their equipment, if necessary, to comply 
with the amended standard.
6. Other Issues
    Commenters also raised a number of other sundry issues with the 
NPRM, as discussed below.
    The AORC commented that in the proposed regulatory text in 
S4.3(j)(2), the agency changed certain wording in that paragraph from 
``when the retractor is subjected to an acceleration'' to ``after the 
retractor is subjected to an acceleration.'' In its submission, the 
AORC argued that this wording change affects the meaning of that 
provision, and it requested that in the final rule, the agency revert 
to the original language.
    We have decided to adopt the recommendation of the AORC and 
reintroduce the phrase ``when the retractor is subjected to an 
acceleration'' at the appropriate place in the final rule. We agree 
that using the phrase ``after the retractor is subjected to an 
acceleration'' could be misinterpreted as permitting the retractor to 
lock up anytime after an acceleration pulse of 0.7 g, something that 
the agency clearly did not intend. We believe that this modification 
will correctly capture the relationship between acceleration and ELR 
lock-up.
    Renfroe Engineering commented that there is not any existing 
minimum acceleration requirement for webbing-sensitive retractors, so 
long as the assembly complies with the vehicle-sensitive test. It also 
argued that a range of 1-4 g is necessary to induce lock-up in webbing-
sensitive retractors (although the commenter provided no technical data 
in support of this position). Accordingly, Renfroe Engineering 
requested that FMVSS No. 209 be amended to require ELRs equipped with 
dual-sensitive retractors to comply with the standard for both designs.
    We believe that Renfroe's request is outside the scope of the 
present rulemaking. Furthermore, we believe that having two separate 
lock-up requirements for each assembly would introduce unnecessary 
duplicity into the standard, because compliance is based on whether or 
not the ELR locks up at the proper acceleration and webbing payout, 
regardless of the type of sensor used to accomplish this.
    In a similar vein, the AORC raised the issue of ``nuisance 
locking'' for multi-sensing ELRs. Specifically, the AORC expressed 
concern about multi-sensing ELRs for which only the vehicle-sensing 
capability is certified, thereby leaving the webbing-sensing mode 
unchecked. The AORC stated that the vehicle sensor might engage a lock-
up on a multi-sensing ELR when testing for a webbing-sensitive ``no 
lock'' by a 0.3 g acceleration of the retractor. To remedy this 
potential problem, the AORC suggested that the regulatory text be 
amended either by requiring webbing acceleration of 0.3 g for dual-
sensing retractors or by providing a related provision in the test 
procedures. In addition, the AORC stated that on the issue of the 
requirements for locking of a webbing-sensitive retractor, the webbing 
of the retractor should be accelerated, rather than the retractor 
itself.
    In a February 19, 1981 letter of interpretation to Mr. Frank 
Pepe,\11\ we stated that dual-sensitive ELRs should be treated as 
either a vehicle-sensitive retractor or a webbing-sensitive retractor 
for purposes of the standard. In that letter, the agency explained its 
intention to require use of either type of retractor. Accordingly, the 
agency decided to require manufacturers to elect one type of retractor 
for certification purposes and to conduct testing for only that type of 
retractor (while voluntarily permitting a different type of retractor). 
In that interpretation letter, we expressed our belief that this 
approach would eliminate the apparent conflict that had arisen in the 
compliance envelopes established in S4.3(j)(1) and (2), given the 
compliance tolerances built into these dual-sensitive systems. That 
approach also would not discourage manufacturers from providing the 
overlapping protection of a dual-sensitive ELR.
---------------------------------------------------------------------------

    \11\ See https://www.nhtsa.dot.gov/cars/rules/interps/gm/81/
nht81-1.14.html.
---------------------------------------------------------------------------

    As to the issue of whether the webbing or the retractor should be 
accelerated, the same letter of interpretation points out that 
paragraph S4.3(j)(2) specifically states that the retractor is to be 
accelerated, not the belt webbing, because there are inertial forces 
that react on the retractor during its acceleration that are not 
present when the webbing alone is accelerated. We believe that this 
reasoning remains valid, and it is reflected in the regulatory text of 
this final rule.
    The agency has not been receiving complaints regarding ``nuisance 
locking'' of multi-sensing ELRs, and we do not believe that this issue 
presents a safety concern in the present fleet. However, if the agency 
were presented with supporting data to document a genuine problem, we 
might reconsider our 1981 interpretation.
    In its comments, the AORC also argued as to the need for an A-T 
corridor for the no-lock requirement at an acceleration of no greater 
than 0.3 g, citing similar reasoning as contained in its petition for 
the corridor in the 0.7 g lock-up requirement. Specifically, the AORC 
recommended a corridor with only an upper boundary, with an initial 
onset rate of 150 g/sec and an upper limit sustained at 0.3 g.
    After carefully considering the AORC's comment on this issue, we do 
not believe that it is necessary to amend the standard to provide an A-
T corridor for the no-lock requirement because the existing 
specification is valid. In the existing standard, the requirement in 
S4.3(j)(2) states that the retractor shall

[[Page 48891]]

not lock before the webbing payout extends to the minimum limit of 51 
mm when the retractor is subjected to an acceleration no greater than 
0.3 g, which is to occur within the first 50 ms and is to be sustained 
throughout the test. The agency believes that this requirement 
implicitly provides the appropriate boundary for the acceleration pulse 
(with a range specified at 0.3 g or less), so there is not any need to 
explicitly define an acceleration tolerance corridor for the no-lock 
requirement. We likewise do not believe that it is necessary to limit 
the onset rate limit to 150 g/sec. If the acceleration pulse meets the 
existing requirements of the hardware and data acquisition methodology, 
a no-lock corridor should not be necessary. Furthermore, even if we did 
agree with the AORC's suggestion in this regard, it would not be 
appropriate to make this change immediately in the final rule without 
the opportunity for public comment, because the issue of a no-lock 
corridor was not raised in either the AORC's original petition or the 
NPRM.
    Public Citizen submitted its report titled, ``Rolling Over on 
Safety: The Hidden Failures of Belts in Rollover Crashes,'' which 
documents what that organization perceives to be inadequacies in 
current safety belt design and performance during rollover events. 
Although rollover crashes are a topic of significant concern for the 
agency, our assessment is that the Public Citizen report does not 
directly address the specific issues in this rulemaking because of the 
different nature of rollover sensors and seat belt technology such as 
pretensioners.

V. Benefits and Costs

    In preparing its June 3, 2004 proposal, NHTSA did not estimate 
benefits for this rulemaking because we anticipated that it would not 
result in substantial changes to the performance of emergency-locking 
retractors. This assessment has not changed at the final rule stage. 
These amendments to FMVSS No. 209 more directly affect the test 
procedure specifications and are intended only to clarify the test 
specifications.
    NHTSA anticipates only a minimal cost burden to vehicle 
manufacturers from this final rule. Testing laboratories might have to 
develop new specifications for the instrumentation used to generate the 
acceleration pulses and may be required to obtain the specified 
accelerometer. However, NHTSA anticipates that only a small number of 
businesses will need to purchase new equipment, since the 
specifications were requested by the AORC in its petition. The members 
of the AORC constitute the majority of seat belt suppliers in the U.S. 
Those who would have to purchase new equipment may do so for a one-
time, minimal cost to the test laboratory. Furthermore, it is 
anticipated that all current ELRs will continue to comply with FMVSS 
No. 209 without change under the final rule's amendments.

VI. Rulemaking Analyses and Notices

A. Vehicle Safety Act

    Under 49 U.S.C. Chapter 301, Motor Vehicle Safety (49 U.S.C. 30101 
et seq.), the Secretary of Transportation is responsible for 
prescribing motor vehicle safety standards that are practicable, meet 
the need for motor vehicle safety, and are stated in objective 
terms.\12\ These motor vehicle safety standards set a minimum standard 
for motor vehicle or motor vehicle equipment performance.\13\ When 
prescribing such standards, the Secretary must consider all relevant, 
available motor vehicle safety information.\14\ The Secretary also must 
consider whether a proposed standard is reasonable, practicable, and 
appropriate for the type of motor vehicle or motor vehicle equipment 
for which it is prescribed and the extent to which the standard will 
further the statutory purpose of reducing traffic accidents and 
associated deaths.\15\ The responsibility for promulgation of Federal 
motor vehicle safety standards has been delegated to NHTSA.\16\
---------------------------------------------------------------------------

    \12\ 49 U.S.C. 30111(a).
    \13\ 49 U.S.C. 30102(a)(9).
    \14\ 49 U.S.C. 30111(b).
    \15\ Id.
    \16\ 49 U.S.C. 105 and 322; delegation of authority at 49 CFR 
1.50.
---------------------------------------------------------------------------

    In developing this final rule to further clarify the test 
procedures of FMVSS No. 209, Seat Belt Assemblies, the agency carefully 
considered the statutory requirements of 49 U.S.C. Chapter 301.
    First, this final rule arose from a petition for rulemaking brought 
by the industry association for seat belt assembly manufacturers, which 
recommended changes for amending the standard to more clearly define 
requirements and to establish a new test methodology for emergency-
locking retractors. This final rule is preceded by an NPRM, which 
facilitated the efforts of the agency to obtain and consider relevant 
motor vehicle safety information, as well as public comments. Further, 
in preparing this document, the agency carefully evaluated available 
research, testing results, and other information related to various ELR 
technologies. In sum, this document reflects our consideration of all 
relevant, available motor vehicle safety information.
    Second, to ensure that the requirements for ELRs are practicable, 
the agency considered the form and functionality of currently compliant 
ELRs, consistent with our safety objectives and the statutory 
requirements. We note that ELRs are already required on light vehicles, 
and we believe that it will be practicable to adopt the new 
requirements and test methodology of this final rule without 
necessitating redesigns on the part of ELR manufacturers. We expect 
that vehicle manufacturers will continue to have a number of 
technological choices available for meeting the requirements of the 
FMVSS No. 209 for ELRs. In sum, we believe that this final rule is 
practicable and will provide greater clarity in terms of the test 
procedures for ELRs.
    Third, the regulatory text following this preamble is stated in 
objective terms in order to specify precisely what performance is 
required and how performance will be tested to ensure compliance with 
the standard. Specifically, the final rule sets forth performance 
requirements for operation of the ELR, including the circumstances 
under which the ELR must lock. The final rule also includes revised 
test requirements for ELRs, including establishment of a new 
acceleration-time corridor, provision of a tolerance for angle 
measurements, and adoption of the same instrumentation specifications 
currently found in other FMVSSs containing crash tests. The standard's 
test procedures carefully delineate how testing will be conducted. 
Thus, the agency believes that this test procedure is sufficiently 
objective and would not result in any uncertainty as to whether a given 
vehicle satisfies the requirements of FMVSS No. 209.
    Fourth, we believe that this final rule will meet the need for 
motor vehicle safety because the standard will better define the 
acceleration pulse that will be utilized in testing ELRs, mechanisms 
which serve the critical function of ensuring that seat belts are 
properly locked up in the event of sudden deceleration or a crash.
    Finally, we believe that this final rule is reasonable and 
appropriate for motor vehicles subject to the applicable requirements. 
As discussed elsewhere in this notice, the agency is addressing the 
petitioner's concern that to better define the ELR requirements and 
test procedures, actions which we do not expect will increase the 
present stringency of the standard or cause

[[Page 48892]]

compliance problems for existing ELRs. Accordingly, we believe that 
this final rule is appropriate for the seat belt assemblies in covered 
vehicles that are subject to these provisions of FMVSS No. 209 because 
it furthers the agency's objective of preventing deaths and serious 
injuries by ensuring that ELRs in seat belts function properly.

B. Executive Order 12866 and DOT Regulatory Policies and Procedures

    Executive Order 12866, ``Regulatory Planning and Review'' (58 FR 
51735, October 4, 1993), provides for making determinations whether a 
regulatory action is ``significant'' and therefore subject to OMB 
review and to the requirements of the Executive Order. The Order 
defines a ``significant regulatory action'' as one that is likely to 
result in a rule that may:
    (1) Have an annual eff