Federal Motor Vehicle Safety Standards; Occupant Crash Protection

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Parts 552, 571, 585 and 595

[Docket No. NHTSA 00-7013; Notice 1]

RIN 2127-AG70

Federal Motor Vehicle Safety Standards; Occupant Crash Protection

AGENCY:

National Highway Traffic Safety Administration (NHTSA), DOT.

ACTION:

Final rule; interim final rule.

SUMMARY:

This rule amends our occupant crash protection standard to require that future air bags be designed to create less risk of serious air bag-induced injuries than current air bags, particularly for small women and young children; and provide improved frontal crash protection for all occupants, by means that include advanced air bag technology. To achieve these goals, it adds a wide variety of new requirements, test procedures, and injury criteria, using an assortment of new dummies. It replaces the sled test with a rigid barrier crash test for assessing the protection of unbelted occupants.

The issuance of this rule completes the implementation of our 1996 comprehensive plan for reducing air bag risks. It is also required by the Transportation Equity Act for the 21st Century (TEA 21), which was enacted in 1998.

This rule will ensure that advanced air bag technologies are installed across the full spectrum of future fleets of motor vehicles. As a result, the air bags in those vehicles will be even more effective than the current redesigned air bags in saving lives. At the same time, those air bags will be much less likely than those redesigned air bags to cause deaths or serious injuries.

The provisions of this rule, particularly the maximum test speed for the unbelted rigid barrier test, reflect the uncertainty associated with simultaneously achieving the twin goals of TEA 21. This uncertainty leads us to take an approach that best assures improved air bag protection for occupants of all sizes, without compromising efforts to reduce the risks of injury to vulnerable occupants, including children and short women seated very close to air bags and out-of-position occupants. Such an approach is one that involves the least uncertainty for the occupants who have been most at risk. As long as the manufacturers improve the already substantial overall level of real world protection provided by current redesigned air bags, the uncertainty associated with the challenge of simultaneously achieving the twin goals of TEA 21 is best resolved at this point in favor of minimizing risk. This is especially true in the early stages of the introduction of advanced air bag technologies.

In light of that uncertainty, we are selecting the lower of two proposed speeds as the maximum test speed for the unbelted rigid barrier crash test and issuing that part of this rule as an interim final rule. To resolve that uncertainty, we are planning a multi-year effort to obtain additional data. We will issue a final decision regarding the maximum test speed after giving notice and seeking public comment. If we were to increase the speed, we would provide leadtime commensurate with the extent of that increase.

DATES:

Effective Date:

The amendments made in this rule are effective June 12, 2000.

The incorporation by reference of the publications listed in the rule is approved by the Director of the Federal Register as of June 12, 2000.

Petitions:

Petitions for reconsideration must be received by June 26, 2000.

ADDRESSES:

Petitions for reconsideration should refer to the docket and notice number of this document and be submitted to: Administrator, National Highway Traffic Safety Administration, 400 Seventh Street, SW, Washington, DC 20590.

In light of our decision to issue the maximum test speed for the unbelted rigid barrier test as an interim final rule, we are keeping the docket for this document open to receive public input. Persons making submissions to the docket should refer to the docket and notice number of this document. As we obtain and analyze data, we will place the results in that docket.

FOR FURTHER INFORMATION CONTACT:

For non-legal issues, you may contact Clarke Harper, Chief, Light Duty Vehicle Division, NPS-11. Telephone: (202) 366-2264. Fax: (202) 366-4329. E-mail: Charper@NHTSA.dot.gov.

For legal issues, you may contact Edward Glancy or Rebecca MacPherson, Office of Chief Counsel, NCC-20. Telephone: (202) 366-2992. Fax: (202) 366-3820.

You may send mail to these officials at the National Highway Traffic Safety Administration, 400 Seventh St., S.W., Washington, D.C., 20590.

SUPPLEMENTARY INFORMATION:

For information about air bags and related rulemakings:

Visit the home page of the NHTSA web site at http://www.nhtsa.dot.gov and select “Air Bags” under “Popular Information” on the left hand side of the screen. On the next screen, select “Advanced Air Bags.”

You may also view the materials in the docket for this rulemaking on the Internet. To do this, do the following:

(1) Go to the Docket Management System (DMS) Web page of the Department of Transportation (http://dms.dot.gov/).

(2) On that page, click on “search.”

(3) On the next page (http://dms.dot.gov/search/), type in the four-digit docket number shown at the beginning of this document. Example: The docket number for the Supplemental Proposal in this rulemaking is “NHTSA 99-6407.” If you want to view the materials filed for that notice, you would type “6407.” (For this rule, you would type “7013.”) After typing the docket number, click on “search.”

(4) On the next page, which contains docket summary information for the docket you selected, click on the desired comments. You may download the comments and other materials.

Note to readers:

As an aid to readers who are outside the engineering community, we have provided a glossary that briefly explains the key technical terms used in this preamble. In the case of the term, “fixed barrier crash test,” we have supplemented the explanation with illustrations. That glossary appears in Appendix A at the end of the preamble, before the regulatory text. Readers may find it helpful to review that glossary before reading the rest of this document.

Table of Contents

I. Safety Problems

A. Frontal Crashes and the “Second Collision”—The Leading Cause of Occupant Deaths

B. Preventing or Mitigating the Effects of the Second Collision Using Seat Belts and Air Bags

C. Air Bag Risks and Fatalities

D. Causes of Air Bag Fatalities

II. The Rule, Its Rationale, and Its Implementation

A. Key Provisions of the Rule

B. Other Provisions of the Rule

C. Future Rulemaking Plans

D. Monitoring of Implementation and Field Experience; Research and Technology Assessment

III. Our Proposals for Advanced Air Bags

A. Our Initial Proposal (September 1998)

B. Our Supplemental Proposal (November 1999)

IV. Public Comments on the Supplemental Proposal

V. Diagrams of the Rule Requirements

VI. Improving the Protection of Unbelted Occupants in Serious Crashes

A. Summary of Proposed Requirements

B. Type of Test

C. Agency Decision to Establish Maximum Test Speed at 40 km/h (25 mph)

1. The Supplemental Proposal

2. Summary of Comments

3. Decision on Maximum Test Speed

D. Other Issues

1. Location of 5th Percentile Adult Female Dummy

2. Minimum Test Speed

VII. Improving the Protection of Belted Occupants in Serious Crashes

A. Belted Rigid Barrier Crash Test

B. Belted Offset Deformable Barrier Crash Test

VIII. Minimizing the Risk of Injuries and Deaths Caused by Air Bags

A. Safety of Infants

1. Option 1: Feature (

e.g.,

Weight or Size Sensor) That Suppresses the Air Bag When an Infant Is Present

2. Option 2: Low-Risk Deployment for Infants in Rear-Facing Child Safety Seats

B. Safety of Young Children

1. Option 1: Feature (

e.g.,

Weight or Size Sensor) That Suppresses the Air Bag When a Child Is Present

2. Option 2: Low-Risk Deployment for Young Children

3. Option 3: Feature That Suppresses the Air Bag When a Child Is Out-of-Position

C. Safety of Teenage and Adult Drivers

1. Option 1: Low-Risk Deployment for Drivers

2. Option 2: Feature That Suppresses the Air Bag When a Driver Is Out-of-Position

IX. Injury Criteria

A. Head Injury Criteria

B. Neck Injury Criteria

C. Thoracic Criteria

D. Other Criteria

X. Lead Time and Effective Date

A. Large Manufacturers

B. Limited Line, Small, Multi-Stage Manufacturers and Alterers

1. Limited Line Manufacturers

2. Small Manufacturers

3. Multi-Stage Manufacturers and Alterers

XI. Availability of Original Equipment and Retrofit Manual On-Off Switches

XII. Warning Labels, Consumer Information, and Telltale Devices

A. Warning Labels and Consumer Information

B. Telltale Devices

XIII. Miscellaneous Issues

A. Child Restraints Used for Testing Suppression and Low-Risk Deployment Features

B. Dummy Positioning for Static Suppression and Low-Risk Deployment Tests

C. Due Care Provision

D. Selection of Compliance Options

E. Credits for Early Compliance

F. Choice Between Complying with Existing and/or New Injury Criteria and Test Requirements

G. Time Periods for Measuring Injury Criteria During Tests

H. Cruise Controls

I. Rescue Operations

J. Hybrid III Dummy Neck

K. Seating Procedure for 5th Percentile Adult Female Dummy

L. Deletion of Tests Between the Initial and the Supplemental Proposals

M. Consideration of Unintended Consequences

N. Reporting Requirements

O. Use of Children and Adults for Testing Static Suppression Systems

P. Small Business Concerns

Q. Other Issues

1. Ability to Comment Effectively on the Supplemental Proposal

2. Resubmittal of Petition for Rulemaking by Donald Friedman and Carl Nash

XIV. Benefits and Costs

XV. Rulemaking Analyses and Notices

Appendix A Glossary

Appendix B Evolution of the Air Bag Provisions in Standard No. 208

Appendix C Chronology of DOT and NHTSA Responses to Air Bag Risks and Fatalities

Appendix D Installation of Advanced Technologies in Current Production Motor Vehicles

Regulatory Text

I. Safety Problems

A. Frontal Crashes and the “Second Collision”—The Leading Cause of Occupant Deaths

Frontal crashes are the most significant cause of motor vehicle fatalities. More than

2/3

's of the people killed in frontal crashes are unbelted. Young people,

i.e.,

those in their teens and twenties, account for about 40 percent of the unbelted deaths.

The frontal crash of a vehicle involves two collisions. The first collision occurs when the vehicle strikes another vehicle or an object such as a tree. The second collision is the human collision with the vehicle interior.

When a vehicle collides with an object, a front seat occupant who is not wearing a seat belt becomes a projectile and keeps moving forward at speeds up to the vehicle's pre-crash speed. If that unbelted occupant is not protected by an air bag, the head or chest of the occupant usually slams into the steering wheel, dashboard, roof pillars or windshield. In the absence of an air bag, even belted occupants, particularly belted drivers, are likely to strike the vehicle interior with their head and neck or chest in a serious crash.

B. Preventing or Mitigating the Effects of the Second Collision Using Seat Belts and Air Bags

To prevent or mitigate the effects of the second collision, Standard No. 208 requires that vehicles be equipped with seat belts and frontal air bags.

1

Seats belts are estimated to save 9,500 lives in America each year. Research has found that lap/shoulder belts, when used properly, reduce the risk of fatal injury to front seat passenger car occupants by 45 percent and the risk of moderate-to-critical injury by 50 percent. For light truck occupants, seat belts reduce the risk of fatal injury by 60 percent and moderate-to-critical injury by 65 percent.

1

For a history of NHTSA's rulemaking concerning air bags, see Appendix B, “Evolution of the Air Bag Provisions in Standard No. 208.”

Air bags are also highly effective in reducing fatalities from frontal crashes. Between 1986 and March 1, 2000, air bags have saved an estimated 5,303 front seat occupants (4,496 drivers (85 percent) and 807 right front passengers (15 percent)).

2

Of the 5,303 people, 72 percent were unbelted and 28 percent belted. If observed seat belt use rates were to increase to 85 percent, the goal for 2000 set by DOT in 1997, the distribution of lives saved would change from 72 percent unbelted/28 percent belted to 60 percent unbelted and 40 percent belted.

2

These estimates are based on comparisons of the frequency of front seat occupant deaths in vehicles without air bags and in vehicles with air bags. These life savings occurred predominantly in moderate and high speed crashes; i.e., those with a velocity change (delta V) above 20 mph.

The number of lives saved annually by air bags is continuing to increase as the percentage of air bag-equipped vehicles on the road increases. We estimate that air bags will save more than 3,200 lives annually in passenger cars and light trucks when all light vehicles on the road are equipped with driver and passenger air bags. This estimate is based on an anticipated fleet of vehicles meeting all of the requirements in this rule and on 1997 seat belt use rates (66.9 percent, according to State-reported surveys). However, if observed seat belt use rates were to reach 85 percent, the annual savings of lives due to air bags would be reduced to approximately 2,400.

C. Air Bag Risks and Fatalities

As the numbers above indicate, the attempt through seat belts and air bags to substitute a survivable event for an unsurvivable one or to substitute a less injurious event for a more injurious one is not always successful. While air bags are saving an increasing number of people in moderate and high speed crashes, they have occasionally caused fatalities, especially to unrestrained, out-of-position children, in relatively low speed crashes. As of April 1, 2000, NHTSA's Special Crash Investigation (SCI) program had confirmed a total of 158 fatalities induced by the deployment of an air bag. Of that total, 92 were children, 60 were drivers, and 6 were adult passengers. An additional 38 fatalities were under investigation by SCI on that date, but they had not been confirmed as having been induced by air bags.

Changes have already occurred that are reducing the number of persons killed by air bags. Some changes are behavioral. As a result of public education programs, improved labeling and media coverage, the public is much more aware of the dangers air bags pose to children in the front seat and to drivers sitting too close to the air bag and is taking steps to reduce those dangers. For example, more children are being put in the back seat. More short-statured drivers are moving back from the steering wheel.

Other changes are technological. First, as NHTSA noted in its report, “Air Bag Technology in Light Passenger Vehicles” (December 1999), the air bag outputs (

i.e.,

pressure rise rate and the peak pressure) were reduced significantly in many MY 1998 and later motor vehicles in comparison to the earlier vehicles.

3

Hence, the sled test option successfully expedited the depowering of existing air bags. While there are many means by which air bag aggressiveness can be reduced, reducing air bag outputs is a quick means of accomplishing this goal. The agency's analyses also show that, between MY 1997 and MY 1998, 50 to 60 percent of the vehicles in the fleet covered by the 1997 IR lowered the output of the driver-side air bag, while about 40 to 50 percent of the vehicles in that fleet lowered the output for the passenger side. Comparison of the data for MY 1997 and MY 1998 vehicles shows that, on average, the pressure rise rate in MY 1998 vehicles decreased about 22 percent for the driver air bag and 14 percent for the passenger air bags.

3

The report indicates that some vehicle manufacturers had already depowered some air bags prior to the March 1997 rule.

The data provided by the manufacturers also show that they have made significant changes in the design of their air bag systems other than the air bag pressure rise rate and peak pressure in their air bag designs, some over a period of many years.

4

Thus, depowering is not the only technological option for reducing risk. One change is the recessing of driver air bags so that the module is located farther away from the plane of the steering wheel, and thus farther from the driver. Although this feature was not common in the early 1990s, it is found in almost half of the MY 1997 and MY 1998 vehicles in the responses to the 1997 IR. Similarly, the air bag mounting location on the passenger side has also shown significant changes. Other features, such as cover tear patterns, tear pressure, fold patterns and the number and type of tethers, have changed in recent years, all of which may have collectively contributed to the reduced aggressiveness of air bags.

4

Again, these changes began before the March 1997 rule, but have accelerated since then.

To assess the impact of the redesigned air bags on the numbers of air bag-induced fatalities, we used the available SCI data. We compared the rate per million registered vehicles of air bag-induced fatalities for the first 27 months that MY 1998 redesigned vehicles were on the road with the rate of air bag-induced fatalities for the first 27 months that MY 1996-97 vehicles were on the road. We took this approach in an effort to ensure that the amount of exposure was comparable for both groups of vehicles. We found that the air bag-induced fatality rate for all MY 1998 vehicles is 66 percent less than the fatality rate for MY 1996-97 vehicles (0.48 for MY 1998 versus an average of 1.43 for MY 1996-97).

Part of this reduction is the result of changes in vehicle design and part is the result of changes in behavior;

i.e.,

using seat belts more frequently, moving children into the back seat, and moving the driver's seat further back. We found evidence of behavioral changes by examining the front seat and rear seat distributions of all child passengers (age 0 to 12) in passenger cars, survivors plus fatalities, in the Fatal Analysis Reporting System (FARS) from 1995 through mid-1999. In cars with passenger air bags, the percentage of toddlers and infants riding in the back seat increased from about 70 percent in 1995 to about 90 percent in 1999.

D. Causes of Air Bag Fatalities

Several factors are common to air bag-induced fatalities. First, they involve air bags that do not meet the suppression or low risk deployment requirements of this rule. Second, the occupants are generally very close to an air bag module when the air bag begins to deploy during a crash.

5

The one fact that is common to all persons who died is

not

their height, weight, gender, or age. Instead, it is the fact that they were very close to an air bag when it started to deploy. For some people,

e.g.,

infants in rear-facing infant seats, this occurred because they were initially sitting very close to the air bag. For the other occupants, this typically occurred because they were not restrained by seat belts or child safety seats and moved forward during pre-crash braking.

5

Vehicle speed is not a causative factor. Most of the crashes involving fatalities that have been confirmed as air bag-induced occurred at relatively low speeds. If the passenger air bag had not deployed in those low speed crashes, the people would probably not have been killed or seriously injured.

Closeness is a problem because, in order for an air bag to cushion an occupant's head, neck, chest and abdomen and keep the occupant from hitting the steering wheel, windshield or instrument panel, the air bag must move into place quickly. The force of a deploying air bag is greatest as the air bag begins to inflate. If occupants are very close to or in contact with the cover of an air bag that does not meet the low risk deployment requirements of this rule, they can be hit with enough force to cause serious injury or death when the air bag begins to inflate. This can be caused either by the cover as the air bag breaks out of the module (known as the “punch-out” effect) or by the unfolding and inflating air bag as it first conforms to the contours of the occupant and then moves rapidly into its fully-inflated shape (known as the “membrane” effect).

6

6

In general, a driver can avoid any serious air bag risks by sitting at least 10 inches away from the air bag (measured from the breastbone to the center of the air bag cover) and by wearing a lap/shoulder seat belt. Teenage and adult passengers can avoid this risk by moving their seat as rearward as possible and wearing their seat belts. Even in a vehicle that does not have any air bags, children should ride in the rear seat whenever possible, since that is a significantly safer location.

In all of the 92 SCI confirmed fatalities involving children, the children were very close to the instrument panel when the air bag deployed. Because of their proximity, the children sustained fatal head or neck injuries from the deploying passenger air bag.

Eighteen fatally-injured infants were close to the air bag because they were in rear-facing infant seats installed directly in front of a passenger air bag. A rear-facing infant seat which is installed in the front seat of a vehicle with a passenger air bag will almost always position the infant's head very close to the passenger air bag. Several other infants were being held in the lap of a passenger.

All but a few of the 74 fatally-injured older children were not using any type of restraint.

7

Of those who were restrained, most were not correctly restrained. The non-use or improper use of occupant restraints allowed the vast majority of these children to move forward during pre-impact braking

8

before the actual crash. As a result, they

were very close to the air bag when it deployed.

7

NHTSA notes that almost all of the 68 fatally-injured children were 1-7 years old.

8

Pre-impact braking was a factor in a very high percentage of crashes resulting in the deaths of the older children.

As in the case of the children fatally injured by air bags, the key factor regarding the confirmed deaths of adults has been their closeness to the air bag when it deployed. The most common factor that allowed them to become very close to the air bag was the failure to use seat belts. Only 18 of the 60 drivers are known to have been properly restrained by lap and shoulder belts at the time of the crash.

II. The Rule, Its Rationale, and Its Implementation

A. Key Provisions of the Rule

Early Agency Efforts to Reduce Air Bag Risks

Since the early 1990s, NHTSA has been taking steps to induce changes in behavior and technology to reduce the risk of such deaths and serious injuries to children and small adult drivers, especially when they are out-of-position.

9

We focused our initial efforts to reduce air bag risks on a public education campaign to alert the public about the dangers of air bags to children in general and to infants in particular. We urged parents to place their children always in the back seat and to ensure that they were always properly restrained. We required informative, text-only, warning labels to be placed in new motor vehicles and on child restraints.

9

Even before the 1990's, the issue of air bag-induced risks to out-of-position occupants was addressed by the agency in its rulemaking and research related to air bags. For a history of those earlier activities, see Appendix B at the end of this preamble.

1996 Comprehensive Plan for Addressing Air Bag Risks

To address the problems that arose with the air bags installed in many motor vehicles, the agency announced a comprehensive plan in November 1996. The plan set forth an array of immediate, interim and long-term measures. The immediate and interim measures focused on behavioral changes and relatively modest technological changes. The long-term measures focused on more significant technological changes,

i.e.,

advanced air bag technologies. The immediate steps included expanding efforts to persuade parents to place their children in the rear seat; requiring new labels with eye-catching graphics and colors and strong, clear warning messages; extending the period of time for permitting the installation of original equipment on-off switches in new vehicles which either lacked a rear seat or had a rear seat too small to permit the installation of a child restraint system; and permitting the installation of retrofit on-off switches in vehicles-in-use to protect people in at-risk groups. Because of the lead time needed to develop and install advanced air bag technologies, NHTSA announced plans to propose an interim measure to accelerate manufacturer efforts to redesign their air bags. In the long term, the agency said that it would conduct rulemaking to require the installation of advanced air bags.

Contribution of Behavioral Changes and the Current, Redesigned Air Bags to Reducing Air Bag Risks

To implement the interim phase of the comprehensive plan and speed the redesigning and recertifying of air bags to reduce the risks to out-of-position occupants, we amended Standard No. 208, Occupant Crash Protection, 49 CFR 571.208, to establish a temporary option under which vehicle manufacturers could certify their vehicles based on a 48 km/h (30 mph) unbelted sled test using a 50th percentile adult male dummy, instead of the 48 km/h (30 mph) unbelted rigid barrier crash test using that dummy. 62 F.R. 12960; March 19, 1997.

Available data indicate that the redesigned air bags, together with behavioral changes, such as placing more children in the back seat, have reduced the risks from air bags for the at-risk populations. Although these real-world data reflect only about two years of field experience with redesigned air bags, they preliminarily indicate that the redesigned air bags in model year (MY) 1998 and 1999 vehicles provide the same level of frontal crash protection as that provided by earlier air bags.

While the redesigned air bags in current motor vehicles have contributed to the reduction in the risk of air bag-induced injuries, they can still cause death or serious injury to unrestrained occupants. We selected the provisions adopted in this rule to ensure that future air bags provide more frontal crash protection, and reduce risk further, than either the current redesigned air bags or air bags that would have been minimally compliant with the sled test.

Transportation Equity Act for the 21st Century

The Transportation Equity Act for the 21st Century (TEA 21), enacted by Congress in June 1998, requires us to issue a rule amending Federal Motor Vehicle Safety Standard No. 208, Occupant Crash Protection:

* * * to

improve occupant protection

for occupants of different sizes, belted and unbelted, under Federal Motor Vehicle Safety Standard No. 208,

while minimizing the risk

to infants, children, and other occupants from injuries and deaths caused by air bags,

by means that include advanced air bags.

(Emphasis added.)

TEA 21 specifies that its twin goals are to be accomplished by means that include advanced air bag technologies. Although these technologies are beginning to be incorporated in some new vehicles, many aspects of those technologies are still undergoing development and refinement today.

The rule is required to be consistent with section 30111 of Title 49. Section 30111 requires that, among other things, Federal motor vehicle safety standards be practicable, meet the need for motor vehicle safety, and be stated in objective terms.

Under TEA 21, we were to issue the rule by September 1, 1999, unless we determined that the rule could not be issued by that date. The many issues in this rulemaking led us to make such a determination. We notified Congress of this determination in a letter dated August 3, 1999. Therefore, under TEA 21, we were required to issue the rule by March 1, 2000.

TEA 21 addresses various other issues, including the effective date and phase-in for the requirements adopted in this rule, as well as the opportunity to earn phase-in credits through early compliance. A complete discussion of TEA 21's provisions is included in the 1998 notice of proposed rulemaking (NPRM). See 63 F.R. 49958 at 49961; September 18, 1998.

The Gathering of Information and Soliciting of Comments for This Rulemaking To Reduce Air Bag Risks Further

Since 1996, the agency has been carefully laying the groundwork for completing the implementation of its comprehensive plan by issuing this rule. We have made extensive efforts to gather information and solicit public comments that would help us identify and adopt a sensible, effective array of requirements for increasing protection and minimizing risk. In February 1997, we held a public technical workshop on advanced air bag technologies. In December 1997, we sent an Information Request (IR) to the vehicle manufacturers to obtain detailed information concerning their changes in air bag design during the 1990s. In April 1998, Jet Propulsion Laboratories completed, at NHTSA's request, a report titled “Advanced Air Bag Technology Assessment.” In mid-1998, Congress made the judgment that advanced air bags should be required. It enacted TEA

21 mandating that we amend our occupant protection standard to require vehicle manufacturers to improve the protection provided by air bags and to reduce the risks associated with them by means that include advanced air bag technologies. Although TEA 21 required only that we seek public comment once on our proposals before taking final action, we asked for public comment twice. We issued a notice of proposed rulemaking (NPRM) in September 1998, and a supplemental notice of proposed rulemaking (SNPRM) in November 1999. To help us thoroughly explore the issues, we proposed or discussed in those two notices a variety of alternatives and posed a wide-ranging array of questions. Based on the information we received in response to the 1997 IR, we completed a report titled “Air Bag Technology in Light Passenger Vehicles” in December 1999.

Changes to Our Initial Proposals in Response to Information and Comments

We carefully considered the information we gathered and the comments we received on the 1998 NPRM and appropriately adjusted our proposals in the 1999 SNPRM to respond to those materials. For example, based on the public comments on the NPRM regarding the type and number of tests needed to meet the risk minimization goals of TEA 21, we significantly reduced the number of those tests when we issued the SNPRM.

Further, there was a substantial shift between the NPRM and SNPRM in the issues that needed to be resolved in determining which test should be specified to promote the improvements required by TEA 21 in the ability of vehicles to protect unbelted people in moderate to high speed crashes,

i.e.,

those that are potentially fatal. In the NPRM, the primary issue was whether we should (1) retain the unbelted sled testing option, or (2) delete that option, leaving the existing 48 km/h (30 mph) unbelted rigid barrier crash testing provision as the sole basis for certification compliance with Standard No. 208's requirements regarding the protection of unbelted occupants.

10

10

As explained in greater detail below, this rule transforms unbelted rigid barrier testing under Standard No. 208 through the adoption of new and more stringent injury criteria, a new small adult female dummy seated far forward of where the existing mid-sized adult male dummy is placed in compliance testing, a new belted offset test to ensure that sensors work properly in crashes that are not similar to a crash into a rigid barrier and the establishment of complementary risk reduction tests that will affect how manufacturers comply with the unbelted rigid barrier test in the future.

In the SNPRM, the primary issue regarding unbelted testing was what type of unbelted crash test should be specified, and at what top speed. We proposed several alternatives. One alternative was to test unbelted protection in an improved unbelted rigid barrier crash test with a top speed within the range of 40 to 48 km/h (25 to 30 mph). We said that if we issued a rule setting the maximum speed at 40 km/h (25 mph), we might also increase the maximum speed of the belted rigid barrier test from the current 48 km/h to 56 km/h (30 to 35 mph). Another alternative was to test unbelted protection in an unbelted offset deformable barrier test with a maximum speed to be established in the rule within the range of 48 to 56 km/h (30 to 35 mph).

In addition, we sought comment on other possibilities. One was to issue “a final rule temporarily reducing the maximum speed for the unbelted rigid barrier test to 40 km/h (25 mph) (or some other speed,

e.g.,

44 km/h (27.5 mph))” and then returning “it to 48 km/h (30 mph) after an appropriate period of time,

e.g.,

after the TEA 21 phase-in.” Another was “to temporarily permit relaxed injury criteria performance limits (e.g., 72 g chest acceleration limit instead of 60 g chest acceleration limit) in unbelted rigid barrier tests between 25 mph and 30 mph.”

Finally, we proposed in the SNPRM that the agency would not test at a speed of less than 29 km/h (18 mph) under the unbelted rigid barrier test alternative, and that the agency would not test at a speed of less than 35 km/h (22 mph) under the unbelted offset deformable barrier test alternative. This was not only a departure from the proposal in the NPRM, but also from prior agency practice. In the NPRM, we had proposed to test at any speed up to the maximum test speed. One reason for this change was that we wanted to be sure that the standard did not push deployment thresholds downward,

i.e.,

cause air bags to be deployed at lower speeds than are appropriate for maximum occupant protection.

Public Comments on the SNPRM

The commenters on the SNPRM, including vehicle manufacturers, air bag manufacturers, insurance companies, public interest groups, academia, and the National Transportation Safety Board (NTSB), generally agreed with most aspects of that document. For example, the commenters agreed with the agency's proposals to reduce air bag-induced risks by specifying that driver air bags deploy in a low-risk manner in low speed crashes and the passenger air bags either deploy in that manner or turn off in the presence of young children.

With respect to our proposals for improving occupant protection, most commenters supported replacing the unbelted sled test with an unbelted rigid barrier crash test. The vehicle manufacturers, which had opposed a rigid barrier test in their comments on the NPRM, agreed to a return to such a test.

However, while there was a convergence of opinion as to the type of unbelted test, there was a sharp difference of opinion among the commenters on the SNPRM regarding the maximum speed for the unbelted rigid barrier crash test. Several safety advocacy and consumer groups urged that the maximum speed be kept at 48 km/h (30 mph). The vehicle manufacturers, air bag suppliers, an insurance industry safety organization, and several other organizations, believing that a maximum test speed of 48 km/h (30 mph) could make significant repowering necessary, urged that the maximum speed be set at 40 km/h (25 mph). They urged further that the speed be maintained at that level pending analysis of field experience with the air bags installed in motor vehicles during that period. For similar reasons, the NTSB also urged a maximum test speed of 40 km/h (25 mph).

There were also significant differences of opinion regarding our proposals about the provision providing a due care defense against findings of noncompliance with the air bag requirements of Standard No. 208 and about the wording of the statements regarding air bag-induced risks on the proposed vehicle labels.

We note that a substantial number of comments were submitted to the docket for the SNPRM after the comment closing date. In preparing this rule, we have considered all comments placed in the docket on or before April 28, 2000.

The Development of a Data-Driven Rule

Before we made decisions on which provisions should be included in this rule to improve air bag performance as required by TEA 21, we carefully considered the available information and the public comments, the underlying safety problems, the performance of air bag systems in current motor vehicles, the ability (including lead time needs) of vehicle manufacturers to achieve better performance in future motor vehicles, the air bag technology (including advanced air bag technology) currently available or being developed, the cost of compliance, and other factors. Because

the comments on the SNPRM focused on the alternatives for improving the protection provided by air bags, we were particularly careful in considering the comments concerning the costs, benefits and risks associated with each of those alternatives.

The requirements in today's rule for improving protection and minimizing risk are challenging and will push the vehicle manufacturers to make needed safety improvements in air bag performance. Our decisions regarding the selection of those requirements was based on available test data and analysis, and our informed judgment about the best way of implementing the requirements of TEA 21.

The Principal Provisions of the Rule

The rule will improve protection and minimize risk by requiring new tests and injury criteria and specifying the use of an entire family of test dummies: the existing dummy representing 50th percentile adult males, and new dummies representing 5th percentile adult females, six-year old children, three-year old children, and one-year old infants. With the addition of those dummies, our occupant crash protection standard will more fully reflect the range in sizes of vehicle occupants. As noted above, most aspects of this rule are supported by most commenters on this rulemaking, including vehicle manufacturers, air bag manufacturers, insurance companies, public interest groups, academia, and the NTSB.

The rule will be phased in during two stages. The first stage phase-in requires vehicles to be certified as passing the unbelted test requirements for both the 5th percentile adult female and 50th percentile adult male dummies in a 40 km/h (25 mph) rigid barrier crash, and belted test requirements for the same two dummies in a rigid barrier crash with a maximum test speed of 48 km/h (30 mph). In addition, the first stage requires vehicles to include technologies that will minimize risk for young children and small adults.

The second stage phase-in requires vehicles to be certified as passing the belted test requirements for the 50th percentile adult male dummy at 56 km/h (35 mph). This requirement will ensure improved protection for belted occupants.

Risk Minimization Provisions Implemented During First Stage Phase-in

During the first stage phase-in, from September 1, 2003 to August 31, 2006, increasing percentages of motor vehicles will be required to meet requirements for minimizing air bag risks, primarily by either automatically turning off the air bag in the presence of young children or deploying the air bag in a manner much less likely to cause serious or fatal injury to out-of-position occupants. If they so wish, manufacturers may choose to use a combination of those two approaches.

Manufacturers that decide to turn off the passenger air bag will use weight sensors and/or other means of detecting the presence of young children. To test the ability of those means to detect the presence of children, the rule specifies that child dummies be placed in child seats that are, in turn, placed on the passenger seat. It also specifies tests that are conducted with unrestrained child dummies sitting, kneeling, standing, or lying on the passenger seat.

The ability of air bags to deploy in a low risk manner will be tested using child dummies on the passenger side and the small adult female dummy on the driver side. For manufacturers that decide to design their passenger air bags to deploy in a low risk manner, the rule specifies that unbelted child dummies be placed against the instrument panel. This location was selected because pre-crash braking can cause unrestrained children to move forward into or near that position before the air bag deploys. The air bag is then deployed. The ability of driver air bags to deploy in a low risk manner will be tested by placing the 5th percentile adult female dummy against the steering wheel and then deploying the air bag.

Protection Improvement Provisions Implemented During First Stage Phase-in

In addition, the vehicle manufacturers will be required to meet a rigid barrier crash test with both unbelted 5th percentile adult female dummies and unbelted 50th percentile adult male dummies. The unbelted rigid barrier test replicates what happens to motor vehicles and their occupants in real world crashes better than the current sled test does. The maximum test speed for unbelted dummy testing will be 40 km/h (25 mph).

Our decision to set the maximum test speed for unbelted dummy testing at 40 km/h (25 mph) is being issued as an interim final rule. We conclude that is the appropriate test speed for at least the TEA 21 implementation period (MY2004-2007). That speed will provide vehicle manufacturers with the flexibility they need during that period to meet the technological challenges involved in simultaneously improving protection and minimizing risk. To achieve those twin goals, the manufacturers will have to comply with the wide variety of new requirements using an array of new dummies during this near-term time frame.

However, we draw no final conclusion about the appropriateness of that test speed in the longer run. At this time, we cannot assess whether the uncertainty about the manufacturers' ability to improve protection further and minimize risk simultaneously will persist beyond the TEA 21 implementation period. In addition, while we believe that it is unlikely that a 40 km/h (25 mph) maximum test speed will lead to a reduction in high speed protection, we cannot rule out that possibility. If manufacturers were to engage in significant depowering, it could result in lesser crash performance for teenage and adult occupants. On the other hand, even if current levels of real world protection were only maintained, rather than improved, the marginal benefits of a 48 km/h (30 mph) unbelted maximum test speed would be significantly diminished or eliminated.

To help resolve these issues and concerns, we are planning a multi-year effort to obtain additional data. The activities comprising that effort are described in the section below entitled, “Monitoring of Implementation and Field Experience; Research and Technology Assessment.” Based on the results of those information gathering and analysis efforts, we will make a final decision regarding the maximum test speed for unbelted dummy testing in the long run, after providing opportunity for informed public comment.

There are still other additions to Standard No. 208. To ensure that vehicle manufacturers upgrade their crash sensing and software systems as necessary to prevent late air bag deployments in crashes with soft pulses, vehicles will be required to meet an up-to-40 km/h (25 mph) offset deformable barrier test using belted 5th percentile adult female dummies. A late air bag deployment would allow enough time for an unrestrained occupant to move forward into the steering wheel or instrument panel during a crash before the air bag deploys. Thus, the occupant would be in contact with or very close to the air bag module when the air bag deploys, creating a risk of severe or fatal injury. In addition, the 5th percentile female dummy is added to the 48 km/h (30 mph) belted rigid barrier test.

Provision Implemented During Second Stage Phase-in

During the second stage phase-in, from September 1, 2007 to August 31, 2010, the maximum test speed for the belted rigid barrier test will increase

from 48 km/h (30 mph) to 56 km/h (35 mph) in tests with the 50th percentile adult male dummy only. As in the case of the first-stage requirements, this second-stage requirement will be phased in for increasing percentages of motor vehicles. We did not include the 5th percentile adult female dummy in this requirement because we have sparse information on the practicability of such a requirement. As noted below, we will initiate testing to examine this issue and anticipate proposing increasing the test speed for belted tests using the 5th percentile adult female dummy to 56 km/h (35 mph), beginning at the same time that the belted test must be met at that speed using the 50th percentile adult male.

Schedule for Implementation

We have changed the date on which the implementation of this rule begins from September 1, 2002, as proposed in the SNPRM, to September 1, 2003. This gives vehicle manufacturers as much lead time as TEA 21 allows for the first stage phase-in. TEA 21 does not permit a later starting date. This change will give the manufacturers a lead time of more than 3 years for vehicles produced during the first year (Model Year (MY) 2004) of that phase-in and more than 6 years for vehicles produced during MY 2007, the first MY in which vehicle manufacturers will be required to manufacture all of their vehicles in compliance with the first stage requirements without the aid of credits.

We changed the starting date for the first stage in part because of the breadth of the challenges that the vehicle manufacturers will be required to meet during that stage. They will need to certify their vehicles to an unbelted barrier test instead of a sled test. Moreover, they will need to meet this test for the new 5th percentile adult female dummy seated all the way forward as well as for the existing 50th percentile adult male dummy seated in the mid-track position. They will also need to meet a new belted offset deformable barrier test using the 5th percentile adult female dummy and a belted rigid barrier test for both 50th percentile adult male dummies and 5th percentile female dummies. For all of these tests, they will need to meet new injury criteria performance limits. Finally, the vehicle manufacturers will need to certify their vehicles to an array of test requirements to minimize the risk to infants, children, and other occupants from injuries and deaths caused by air bags using the 5th percentile adult female dummy and the child dummies. The starting date of September 1, 2003 will give the manufacturers additional time to gain experience with the new dummies, final specifications for which have only recently been established.

Further, the longer lead time for the first stage phase-in will also promote technological innovation regarding ways of minimizing risks. It will give vehicle manufacturers more time to complete development and testing of the advanced technologies they plan to use. Further, we are aware that suppliers are continuing work on additional technologies. The additional time will enable the manufacturers to explore further using some of these additional technologies.

Rationales for Risk Minimization Requirements

The agency drafted the risk minimization requirements to give vehicle manufacturers a broad choice among those advanced air bag technologies that can be used either to turn air bags off in appropriate circumstances or cause air bags to deploy in a low risk manner.

11

Thus, the vehicle manufacturers will have the freedom to choose from a variety of available technological solutions or to innovate by developing new ones if they so desire.

11

The rule also establishes very general performance requirements for dynamic automatic suppression systems (DASS) and a special expedited petitioning and rulemaking process for considering procedures for testing advanced air bag systems incorporating a DASS. In response to comments, modifications have been made to address concerns about confidentiality and timing.

We estimate that if advanced air bag technologies (suppression and low risk deployment) are 100 percent reliable, they could have eliminated 95 percent of the known air bag fatalities that have occurred to date in low speed crashes. For example, weight sensors can be installed in the passenger seat so that the passenger air bag is turned off when children, from infants up to the typical 6-year-old, are present. The use of weight sensors for that purpose should essentially eliminate the risk of air bag-induced fatal injuries for children in that size and age range. Based on available data, it does not appear that turning air bags off for those young children would result in the loss of any benefits. There is an element of uncertainty about the level of reliability and effectiveness of the suppression for children from 0 to 6 years old and low risk deployment designs that will be actually installed in vehicles. We also note that we do not currently have a dummy suitable for assessing the effectiveness of suppression and low risk deployment for children ages 7-12. (See the section below entitled, “Future Rulemaking Plans.”) Our decision concerning the maximum test speed for the unbelted rigid barrier test reflects, in part, these uncertainties and limitations.

The availability of advanced air bag technologies for minimizing risks is not just a theoretical possibility. Vehicle manufacturers are very actively working on completing their development and testing of weight sensor systems so that they will be ready for installation for the passenger air bags in their motor vehicles. Installation could begin as early as the next model year.

12

Means of reducing risk for drivers, including dual-stage air bags coupled with sensors for driver seat belt use and driver seat position, are already being installed in some vehicles. For a description of advanced technologies and a partial listing of current models equipped with one or more types of those technologies, see Appendix D, “Advanced Technologies for Improving Air Bags.”

12

The MY 2001 Ford Windstar will, according to a report in the April 24, 2000 edition of Automotive News, be equipped with an advanced air bag system “designed to prevent the deployment of the front passenger airbag when sensors determine the passenger's weight is less than 45 pounds.”

Rationales for Protection Improvement Requirements

Replacing the Unbelted Sled Test With the Unbelted Rigid Barrier Crash Test

The agency has decided to delete the sled test option and retain the unbelted rigid barrier crash test provision for the reasons explained in the NPRM and SNPRM. Among those reasons is that a crash test replicates how vehicle structures and air bag systems work together in real world crashes. A sled test cannot do that because while the vehicle is quickly decelerated in such a test, it never crashes into anything. As a result, the sled test cannot take into account the ability of a motor vehicle's structure to manage crash energy. Further, the sled test uses a generic crash pulse instead of the individual crash pulse of the particular vehicle being tested, and deploys all air bags at a fixed time during the event rather than having that decision made by the crash sensing system of the vehicle.

Selection of 40 km/h (25 mph) as Top Speed for Unbelted Rigid Barrier Test

In developing today's rule, we gave serious consideration to specifying 40 km/h (25 mph) as the maximum speed for the unbelted rigid barrier test for an initial period (so that vehicle manufacturers could focus during that period on risk minimization) and then phasing-in a 48 km/h (30 mph) unbelted test speed in the 2008 through 2010 model years. Our initial inclination to

increase the maximum test speed to 48 km/h (30 mph) during the second of two phase-ins reflected several considerations. First, our testing of vehicles with redesigned air bags indicated that the vast majority of current vehicles can meet that test using 50th percentile adult male dummies. Those test results were not rebutted by any significant test data provided by the motor vehicle manufacturers or others. Therefore, the record did not support the argument by the motor vehicle manufacturers and others that a 48 km/h (30 mph) test using a 50th percentile adult male dummy would require any kind of general “repowering” of air bags.

13

Second, we concluded that air bags could be designed to meet a 48 km/h (30 mph) test with both 5th percentile adult female dummies and 50th percentile adult male dummies without increasing risks to out-of-position occupants.

14

Third, we believed that a specific requirement to return eventually to a 48 km/h (30 mph) test should be adopted to ensure that vehicle manufacturers did not engage in significant additional depowering of air bags, or make them substantially smaller, which would reduce their protectiveness to occupants in high speed crashes.

15

We believed then that there could be an economic incentive to install air bags that were minimally compliant with a 40 km/h (25 mph) test.

13

Based on NHTSA crash tests (including tests of some vehicles with stiff crash pulses) of a variety of types and sizes of vehicles in the current new vehicle fleet, we concluded that the air bags in the vast majority of current vehicles would pass a 48 km/h (30 mph) crash test using a 50th percentile adult male dummy with much the same compliance margins as before. Thus, we concluded that more power was not needed for those vehicles to pass that test with that dummy. Further, the fact that some current vehicles with stiff pulses meet the injury criteria suggested that vehicle manufacturers might not have to repower air bag systems in other stiff pulse vehicles in order to meet the criteria.

More fundamentally, we reasoned that air bags that currently have enough power to meet the injury criteria using the 78 kg (172 lb.) 50th percentile adult male dummy would not need more power to meet the criteria using the much lighter 50 kg (110 lb.) 5th percentile adult female dummy. Our tests indicate that the primary problem for the small female dummy in those current vehicles that do not satisfy the injury criteria in a 48 km/h (30 mph) test with the 5th percentile adult female dummy is that the dummy experiences too much force in the neck or chest area. We said further that the solution for the smaller, lighter female dummy would not be to put additional power into the bag. Not only would that step be unnecessary to protect the 50th percentile adult male dummy, but also it would be likely to exacerbate the problems with the small female dummy. We concluded that the solution would be to redesign the air bag system-—using recessed air bags and new sensors, multiple inflation levels, fold patterns, bias flaps, etc.-—to assure compliance for both the 5th percentile adult female dummy and 50th adult male dummy.

14

We noted that tests of some current production vehicles demonstrate that they incorporate the designs and technologies necessary to enable them to comply at 48 km/h (30 mph) on both the driver and passenger sides with both dummies. These technologies include improved air bag folding, bias flaps, and internal baffles/tethers. We believed that manufacturers could add other measures such as dual-stage inflators, seat position sensors, recessed air bag modules, and better energy-absorbing steering columns. If any vehicle manufacturer wished to do so, it could also develop and provide chambered, dual-stage air bags that are designed to fill fully only an inner chamber of the air bag, instead of the entire air bag, when the driver seat is near the full forward adjustment position on the seat track.

15

We noted that the difference between a 40 km/h (25 mph) and a 48 km/h (30 mph) crash is significant. The significance does not lie in the 20 percent increase in speed, but in the 44 percent increase in crash energy. It is because of that increase in crash energy that the risk of serious or fatal injury is significantly higher at 48 km/h (30 mph) than at 40 km/h (25 mph). Further, a maximum test speed of 48 km/h (30 mph) represents a higher percentage of the crashes that produce serious or fatal occupant injuries. As a result, if air bags were designed only to minimally meet a 40 km/h (25 mph) crash test, many occupants, particularly larger occupants, would not be adequately protected in higher speed crashes. We estimated that 248 to 413 lives could be lost annually if manufacturers did only the minimum required of them by a 40 km/h (25 mph) crash test requirement.

After further examination of the issues and the information before us, and an assessment of the areas of uncertainty about simultaneously improving protection and minimizing risk, we have concluded that the adoption of a 48 km/h (30 mph) unbelted requirement would not be in the best overall interest of safety. We have decided instead to set the maximum test speed for the unbelted rigid barrier crash tests at 40 km/h (25 mph) as an interim final rule. We selected that test speed based on several factors.

First, particularly given the risks that the first generation of air bags posed to out-of-position children and small adult females, and the reaction of the public to those risks, it is very important that advanced air bags be properly designed from the very beginning. We note that air bags, by their nature, present a potential for safety trade-offs not presented by other safety features. That is, while air bags dissipate crash energy for most occupants when they interact with them when fully inflated, the energy released during deployment could be injurious to out-of-position occupants in their interaction with an inflating air bag. This interaction of an occupant and a deploying air bag can be a source of serious injury or death. In contrast, other safety features typically just dissipate energy when occupants interact with components in the vehicle interior in crashes; they do not add energy. Because of this potential for death and injury, we want to be cautious in how far and how fast vehicle manufacturers are required to advance the state of advanced air bag technologies in their vehicles.

Since a significant percentage of current vehicles can already satisfy the new unbelted barrier crash test at 40 km/h (25 mph) with both the 5th percentile adult female dummy and the 50th percentile adult male dummy, we conclude that setting the maximum speed at that level will help vehicle manufacturers to focus their resources and compliance efforts during the first stage on meeting the risk reduction requirements. We want to continue the progress already made in using technological means for reducing air bag-induced risks.

While the manufacturers' resources for dealing with air bags, as well as all the other engineering issues associated with future motor vehicles, are extensive, there are limits to how much can be done at any one time. We need to consider the variety and complexity of changes in air bag testing and technology that will be required by this rule. As we noted above in the discussion of the implementation schedule, the array of new requirements that the manufacturers will have to meet in the first stage is challenging. We are requiring the use of a new test dummy (the 5th percentile adult female) in high speed tests, adding a new test (offset belted), adding new neck injury criteria, and making existing injury criteria more stringent (chest deflection). We are also adding an entire new series of risk minimization tests, which will require manufacturers to install air bag suppression systems or low-risk deployment systems, or both.

We are particularly concerned about the difficulties of trying to meet the unbelted rigid barrier test at 48 km/h (30 mph) with both adult dummies while simultaneously trying to reduce the risks of air bag-induced injuries and deaths. As noted above, the unbelted rigid barrier crash test specified by this rule for the future is an improved test that differs fundamentally from the unbelted rigid barrier test that Standard No. 208 has specified in the past. In the past, the Standard specified only that test and the belted rigid barrier test, and used only one dummy, the 50th percentile adult male dummy. The injury criteria for the unbelted rigid barrier crash test did not evaluate the potential for neck injuries and allowed even greater chest deflection. The Standard specified no other requirement (such as one for out-of-position testing) that had the effect of making it more difficult to achieve compliance with the unbelted rigid barrier test.

In the future, however, greater efforts will be needed to comply with that test because manufacturers will be required to meet a greater variety of requirements.

The unbelted rigid barrier test will have to be met using new or more stringent injury criteria with the new 5th percentile adult female dummy as well as the existing 50th percentile adult male dummy. The necessity of meeting those criteria with the 5th percentile adult female dummy placed in vehicle seats that have been moved all the way forward will add considerably to the challenge of meeting that test. For both belted and unbelted tests, we are adopting improved injury criteria to assure greater protection by air bags in high speed crashes. More specifically, we are changing the way in which the risk of head injuries is measured, adding a new neck injury measure that accounts for the combination of flexion, extension, tension, and compression, and reducing the amount of allowable chest deflection.

Further, efforts to comply with the unbelted rigid barrier test will be affected by the simultaneous need to comply with the risk reduction requirements. In the future, air bags will be explicitly required to be tested for their potential to harm vulnerable occupants as well as offer protection in high-speed crashes.

While advanced air bag technologies will facilitate simultaneously achieving the goals of improving protection and minimizing risk, we cannot forecast the pace of development of those technologies. Setting the maximum speed at 40 km/h (25 mph) will give vehicle manufacturers greater flexibility to choose among and gain experience with advanced air bag technologies. It will also give NHTSA a chance to gather data about the performance of vehicles using advanced air bag technologies. We want the installation of advanced air bag technologies by the vehicle manufacturers across the full spectrum of their fleets to be done correctly—the first time.

Accordingly, we believe that it is preferable to take an approach that best assures improved air bag performance for occupants of all sizes, without compromising efforts to minimize the risks of injury to vulnerable occupants, including children and short women seated very close to air bags, and out-of-position occupants. Such an approach is one that involves the least uncertainty for the occupants who have been most at risk. In other words, as long as the manufacturers improve the already substantial overall level of air bag protection provided by current redesigned air bags, the uncertainty involved in meeting the challenge to improve high-speed protection and minimize risk simultaneously is best resolved at this point in favor of minimizing risk. This is especially true in the early stages of the introduction of advanced air bag technologies. Compared with a 48 km/h (30 mph) unbelted rigid barrier test, a 40 km/h (25 mph) unbelted rigid barrier test presents less chance of inadvertently increasing risks to out-of-position occupants.

Second, while we believe that it should eventually be possible for vehicles to provide protection for both small females and mid-sized males in a 48 km/h (30 mph) unbelted test without compromising efforts to minimize the risks of serious air bag-induced injuries, there are unresolved issues. Our laboratory tests and knowledge of advanced technologies do not tell us how or when developments might reach that point. They also do not provide us with a full picture of the real world consequences of adopting that test speed. Thus, this type of information, by itself, is not necessarily sufficient to enable us to determine whether adopting that speed is worthwhile, much less needed, from a safety standpoint.

We assessed the relative merits of maximum test speeds of 40 km/h (25 mph) and 48 km/h (30 mph) in light of the initial advanced air bag systems that manufacturers will introduce over the next several years. Based on that assessment, we are concerned that the need for vehicle manufacturers to take steps to enable them to certify to a 48 km/h (30 mph) unbelted test could create difficulties in improving protection and minimizing risks for the wide range of occupants and crashes in the real world. A good example of how these potential problems might occur relates to how a vehicle manufacturer might use a dual-stage air bag to meet the goals of this rulemaking.

One strategy for meeting an unbelted 48 km/h (30 mph) barrier requirement for both 5th percentile adult females and 50th percentile adult males would be to use the first stage inflation level for the 5th percentile adult female and the second stage inflation level for the 50th percentile adult male. However, under that strategy, the need to certify to the 48 km/h (30 mph) barrier test for the 5th percentile adult female dummy would require a relatively faster inflation in the first stage. Because that dummy will be placed in a vehicle seat moved all the way forward, the air bag will have to deploy especially quickly to provide protection. The use of a relatively faster first stage would conflict with the strategy of using as benign a first stage inflation level as possible in lower speed crashes to reduce risks to out-of-position occupants. Alternatively, the vehicle manufacturer could use the second stage inflation level for both the 5th percentile adult female and 50th percentile adult male dummies. While this strategy might be a good one for passing a rigid barrier test, in which the dummy does not move forward much before deployment, it might not be a good strategy for high speed real world crashes in which small adult females, who already sit close to the air bag, and unrestrained children move considerably closer as a result of pre-crash braking.

While we believe that dual-stage inflators represent a significant improvement over single level inflators, it is important to recognize that they have limitations. Some of these limitations could be overcome by inflators with more than two stages. However, this would add greater complexity, including additional gray zones. While these and other more advanced technologies, such as chambering and real time occupant position sensing, may become available in the future, we want to be cautious about the possibility of inducing manufacturers to install more advanced technologies before those technologies are fully ready. For example, vehicle manufacturers should gain real world experience with dual-stage inflators before they adopt inflators with additional stages. Also, in areas in which there is uncertainty as to what strategies might be best for safety, such as the specific performance characteristics for dual-level inflators, we want to be careful about adopting requirements that might be inappropriately design restrictive in making it difficult for vehicle manufacturers to design their air bags so that they perform well both in rigid barrier tests and in the wide range of real world crashes.

Third, we are also aware that the vehicle manufacturers need design flexibility to address issues regarding performance in real world crashes not directly replicated by Standard No. 208's tests.

As we have discussed on many occasions, one of the greatest limitations of non-advanced air bags is that they typically deploy in the same manner regardless of such factors as crash severity or occupant size, weight or position. In other words, they are non-adjusting, one-size-fits-all air bags. One of the principal strategies for improving

air bag performance is to provide different levels of deployment for different situations.

The most basic redesigned advanced air bags would allow different types of deployment to suit different crash situations. These air bags would have a dual-level inflator instead of a single-level inflator. Some vehicles already have such inflators. With two levels of inflation, the vehicle manufacturer can design the air bag system so that the level of inflation is dependent on such factors as crash severity, size and weight of the occupant, and position of the occupant. For example, the high level of inflation might be selected for high speed crashes and the low level of inflation for low speed crashes. Of course, the ability to select an inflation level based on these various factors would depend on the existence of sensors that provide relevant information about the above factors.

Successful implementation of air bags designed to vary their performance in response to sensed differences in crash severity or other conditions presents a challenge to the manufacturers in that these air bags have “gray” or transition zones,

i.e.,

ranges of conditions in which the air bag changes from one level of performance to another. At very low speeds, there will be uncertainty within a gray zone about whether the air bag will deploy or not deploy and at higher speeds, there will be uncertainty about which level of performance will be triggered. For example, there will be a gray zone of crash severity in which there is uncertainty whether a dual-stage air bag will deploy only its lowest powered stage or both stages.

Particularly given the importance we place on vehicle manufacturers “getting it right” the first time with advanced air bags, we believe it is appropriate for them to initially introduce relatively simple advanced systems, such as ones incorporating dual-level inflators and sensors that provide basic information about one or more of the factors identified above. While we believe that more complex systems, incorporating such features as several levels of inflation, chambering (e.g., creating, in effect, a small bag inside a larger bag) and real time occupant position sensing, offer promise of even greater benefits, there are significant uncertainties regarding the feasibility and thus availability of such systems, particularly the dynamic position sensing systems.

Fourth, a 40 km/h (25 mph) maximum test speed gives vehicle manufacturers more flexibility to address the greater compliance problems associated with vehicles, e.g., SUVs, with particularly stiff crash pulses. Since unbelted occupants moving forward in frontal crashes of these vehicles will have to be engaged more quickly than in vehicles with softer crash pulses, the task of designing air bag systems in stiff pulse vehicles is significantly more challenging. Our test experience with LTVs using the 5th percentile adult female dummy is very limited. We have conducted only three 48 km/h (30 mph) unbelted rigid barrier tests and only one 40 km/h (25 mph) unbelted rigid barrier test of LTVs with that dummy. Particularly given this limited test experience, we believe there are uncertainties with respect to the ability of manufacturers to meet a 48 km/h (30 mph) rigid barrier test requirement for both 50th percentile adult male dummies and 5th percentile adult female dummies.

Fifth, we believe that it is unlikely that vehicle manufacturers will significantly depower their air bags and minimally comply with the 40 km/h (25 mph) test. Our Final Economic Analysis concludes that there would not be any significant cost savings in installing air bags that were minimally compliant with a 40 km/h (25 mph) test. Vehicle manufacturers have not depowered their air bags so much that they minimally comply with the sled test. In fact, their current redesigned air bags significantly exceed the level of performance needed to meet not only the sled test, but also a 40 km/h (25 mph) rigid barrier crash test with the 50th percentile adult male dummy. As discussed above, the real world data to date for vehicles certified to the sled test, while preliminary, indicates that there has not been a loss of frontal crash protection compared to pre-MY 1998 vehicles.

16

If this result continues, future bags will greatly exceed the minimum performance requirements of the 40 km/h (25 mph) unbelted barrier test. Indeed, the vehicle manufacturers have indicated that they would not engage in significant, widespread additional depowering if a 40 km/h (25 mph) test were adopted. They argue that their need to perform well in NHTSA's 56 km/h (35 mph) belted NCAP tests limits, as a practical matter, any inclination that might theoretically otherwise exist to depower their air bags further. NHTSA notes that this rule increases the influence of 56 km/h (35 mph) belted testing by making passing such testing with 50th percentile adult male dummies mandatory. Thus, NHTSA believes that it is not risking a substantial loss of benefits by establishing an unbelted barrier test of 40 km/h (25 mph).

16

To obtain a fuller understanding of these results, and the role played by the sled test and other provisions of Standard No. 208 in obtaining them, the agency conducted tests and examined information obtained from the vehicle manufacturers. The vehicle manufacturers did not depower all models. There was a wide range in the power of pre-MY 1998 air bags. As to those models that they did depower, they did not depower their air bags as much as they said they could or as much as we anticipated they might when we prepared the economic analysis accompanying our 1997 final rule that adopted the sled test option. Instead, as NHTSA tests have shown, the manufacturers typically chose levels of power that still enabled them to pass the pre-existing 48 km/h (30 mph) unbelted crash test with a 50th percentile adult male dummy. Further, these tests have revealed that vehicles with redesigned air bags pass that test with that dummy by roughly the same margin of compliance as earlier vehicles did. (It should be emphasized, as we note below in the section entitled, “Selection of 40 km/h (25 mph) as Top Speed for Unbelted Rigid Barrier Test,” that meeting the unbelted rigid barrier crash test at 48km/h (30 mph) with that dummy and a 5th percentile adult female dummy is significantly more challenging than meeting it with 50th percentile adult male dummy alone. It is still more challenging to meet that test with both dummies and minimize risk simultaneously. Thus, the ability to meet the 48 km/h (30 mph) unbelted crash test with a 50th percentile adult male dummy isn't, by itself, predictive of a vehicle manufacturer's ability to meet that test with both dummies, and the other requirements added by this rule.)

As to the differences between the anticipated amount of depowering and the amount of depowering actually performed and as to the performance of the current redesigned air bags, we also note that, as discussed below, depowering is not the only way of reducing the aggressiveness of air bags. There are other design changes that were made by some manufacturers.

Sixth, our decision to replace the 48 km/h (30 mph) generic sled test with the 40 km/h (25 mph) unbelted rigid barrier test requires a significantly higher level of safety. The agency estimates that the sled test is roughly equivalent to a 35.5 km/h (22 mph) rigid barrier perpendicular (0 degree) crash. During the 1997 rulemaking, we looked at the relative safety consequences of an air bag designed to just meet the performance requirements associated with a 48 km/h (30 mph) generic sled test. The agency estimated the fatality impacts of designing a vehicle to minimally meet the performance requirements imposed by the current 48 km/h (30 mph) generic sled test and compared these to the fatality impacts of designing a vehicle to just meet the 40 km/h (25 mph) unbelted rigid barrier test. If these different design tasks did not have any impact on air bag size, air bags designed to the 40 km/h (25 mph) unbelted rigid barrier test could save 64 to 144 more lives than air bags designed to the generic sled test (assumed to be 35.5 km/h (22 mph)). If, on the other hand, air bags designed to the generic sled test would be smaller and provide no benefit in partial frontal impacts, because the 40 km/h (25 mph) unbelted rigid barrier test includes an up to 30

degree oblique test while the generic sled test has no angular component, 282 to 308 more lives (this range includes the 64 to 144 estimates mentioned earlier) could be saved by air bags designed to the 40 km/h (25 mph) unbelted rigid barrier test with the oblique test than lives saved by air bags designed to just comply with the generic sled test.

Increasing Belted Test Speed to 56 km/h (35 mph) for 50th Percentile Male Dummy.

In the SNPRM, we asked for comment on whether we should increase the speed for the belted test using the 50th percentile adult male dummy from 48 km/h to 56 km/h (30 mph to 35 mph) if we adopted 40 km/h (25 mph) as the maximum test speed for the unbelted rigid barrier test. This rule adopts that provision. It will be phased-in for increasing percentages of each manufacturer's fleet beginning in the 2008 model year. We did not propose including the 5th percentile adult female dummy in this requirement because we had sparse information on the practicability of such a requirement. NHTSA will initiate testing to examine this issue and anticipates proposing increasing the test speed for belted tests using the 5th percentile adult female dummy to 56 km/h (35 mph), beginning at the same time that the 50th percentile adult male is required to be used in belted testing at that speed.

NHTSA notes that Standard No. 208 previously specified the same maximum test speed for both belted and unbelted rigid barrier testing. The practical consequence of specifying the same test speed for both types of testing was to make unbelted testing the primary determinant of air bag designs. The reason for this is that, at the same test speed, the unbelted test is more difficult to pass than the belted test. Consequently, air bag designers typically focused their attention on performance in the 48 km/h (30 mph) unbelted test. After they optimized performance attributes for that test, they conducted belted tests to ensure that there were not any anomalies. Nothing in the Standard required, or had the effect of requiring, designers to optimize air bag performance for belted occupants.

Today's rule changes that. By specifying a maximum test speed for belted testing that is significantly higher that the maximum test speed for unbelted testing, Standard No. 208 will oblige occupant protection designers to focus separately on evaluating protection in both belted and unbelted testing as significant design factors, instead of having one type of testing serve simply as a check on the other. This is a major step forward for improving occupant protection for belted occupants. This step is in keeping with the agency's ongoing efforts in its Buckle Up America campaigns. It assures enhanced protection, especially for those 70 percent of occupants who currently wear their belts, and may help persuade those who do not wear their belts to do so.

B. Other Provisions of the Rule

Facilitation of low risk deployment technologies.

In the Supplemental Notice of Proposed Rulemaking (SNPRM), we proposed that the low risk deployment requirements would have to be met for inflation levels at which air bags would deploy in rigid barrier crash tests at speeds up to 29 km/h (18 mph). 64 FR 60556; November 5, 1999. We also proposed that the injury criteria for the unbelted rigid barrier crash test would have to be met within the range between a minimum speed of 29 km/h (18 mph) and the maximum speed, inclusive. Some vehicle manufacturers responded that being required to test under the low risk deployment option for the inflation level (or levels) at which their air bags would deploy in crashes below 29 km/h (18 mph), combined with being required to protect unbelted dummies in crashes at 29 km/h (18 mph) and above, would limit design flexibility and discourage development of low risk deployment air bag systems. The manufacturers claimed that it is difficult with current sensors to design dual-stage air bags that could both meet the low risk deployment requirements

and

the barrier crash test injury criteria, particularly given the gray zone in which either a low level or high level deployment may occur.

To avoid inadvertently discouraging the development of low risk deployment technologies, we have decided that air bags with multiple inflation levels must meet the injury criteria for the low risk deployment tests for the inflation levels at which the air bags would be deployed in crashes of 26 km/h (16 mph) or below (with unbelted 5th percentile adult female dummies at both seating positions), instead of crashes of 29 km/h (18 mph) or below. However, if these air bags do not deploy at all in crashes of 26 km/h (16 mph) or below, the injury criteria must be met using the lowest level of inflation. We have also decided to raise the lower end of the range of speeds at which the unbelted rigid barrier crash test is conducted from 29 km/h (18 mph) to 32 km/h (20 mph). Together, these two changes are intended to facilitate use of the low risk deployment option by providing flexibility for the transition of dual-stage air bag systems from low level deployments designed to protect occupants in low speed crashes and not to injure out-of-position occupants in high level deployments designed to protect occupants from injuries in severe crashes.

Elimination of unneeded tests.

In developing this rule, as in developing the SNPRM, we looked for opportunities to reduce the number and types of test configurations necessary to assure that future air bags minimize the risk of air bag-induced injuries.

17

We have made several further reductions.

17

As noted above, when we issued the SNPRM, we reduced the number of proposed dynamic and static tests, especially those relating to the proposed requirements for reducing the risks of air bags. We reduced, from 14 to nine, N the number of proposed dynamic crash tests that would be applicable to all vehicles. We originally proposed that vehicles equipped with static air bag suppression systems (

e.g.,

weight sensors and pattern sensors) be subject to being tested with any child restraint manufactured over a ten-year period. This would have created the possibility of testing with any one of several hundred different models of child restraints. Recognizing that, we solicited comments to aid us in identifying a much more limited number of specific models that would be representative of the array of available child restraints. Based on the public comments, we proposed to require that vehicles be able to meet the applicable requirements when tested with any one of a far more limited number of child restraints representing a cross-section of the restraints currently on the market. We also significantly reduced the number of positions in which test dummies or child restraints could be placed for testing a static suppression system. This was accomplished largely by eliminating positions that were substantially similar to other positions.

We have dropped several test conditions for testing features (

e.g.,

weight or size sensor) that suppress the air bag when an infant or young child is present. We eliminated some test conditions because we concluded that they were inappropriate for testing this type of feature. The test conditions we dropped for this reason included an unrestrained RFCSS tipped forward onto the dashboard and the 3 year old and 6 year old dummies placed in the low risk deployment positions,

i.e.,

against the instrument panel.

The basic concept behind weight sensors or other features that suppress the air bag when an infant or young child is present is to automatically suppress the air bag unless weight or some other factor indicates that an older child or adult is present. In testing such a device, we believe it is appropriate to test for a variety of positions an infant or young child might likely be placed in by a parent or caregiver or that might likely be assumed by the child. The conditions we are dropping do not fall within this category, but are instead ones that might occur dynamically as a

result of pre-crash braking. However, since the air bag would already be automatically suppressed by this type of device in such a situation, we do not believe it necessary to test for these conditions.

We also proposed testing unrestrained rear-facing child seating systems (RFCSS) at any angle plus or minus 45 degrees from the vehicle seat's longitudinal plane. Because of difficulties in setting up the test and the unlikelihood that parents would place a RFCSS in an angled position, we have revised this test procedure to specify placement only at zero degrees of the longitudinal plane.

As proposed in the SNPRM, we have dropped the requirement for conducting oblique angle tests on vehicles using belted 5th percentile adult female dummies. We have adopted the proposal because we believe that if a vehicle can pass the perpendicular test with 5th percentile adult female dummies and the oblique tests with unbelted 50th percentile adult male dummies, it will also pass the oblique test using 5th percentile adult female dummies. Additionally, we have dropped the belted oblique angled tests for the belted 50th percentile adult male dummy. Given the unbelted oblique tests, we believe that the belted oblique angled tests are unnecessary.

New, more stringent injury criteria.

In the rule, we have added a neck injury criterion and adopted a more stringent limit on chest deflection. The injury criteria are very similar to the ones we proposed in the SNPRM. The Nij approach to the risk of neck injury was generally accepted by the vehicle manufacturers, although they requested some modifications. We have made those modifications.

Due care provision.

In the SNPRM, we proposed to maintain the “due care” provision for the existing crash test requirements and apply it to the new ones as well. However, we did not propose to apply the provision to test requirements that do not involve crashes, based on our belief that these tests are not affected by the variability associated with dynamically-induced dummy movement and/or vehicle deformation.

In this rule, we have decided against extending the due care provision to new crash tests, although it will still be available for vehicles that are not certified to the advanced air bag requirements. Our testing has indicated that manufacturers can easily meet the new injury criteria with 50th percentile adult male dummies in a 40 km/h (25 mph) unbelted test with existing air bag systems and should be able to make what ever improvements are needed to do so with 5th percentile adult female dummies without major uncertainties before they are required to certify any vehicle as meeting the advanced air bag requirements of this rule. Based on our experience with Standard No. 208 compliance activities, we do not believe there is an intrinsic need for a “due care provision.” Further, as we explained in the earlier notices in this rulemaking proceeding, the inclusion of such a provision in a safety standard does not fit very well with the overall statutory scheme.

Extended availability of air bag on-off switches.

As proposed, we have decided to sunset the provisions which allow original equipment (OE) and retrofit on-off switches under specified circumstances. However, instead of sunsetting those provisions at the end of the TEA 21 phase-in period, as we proposed in the SNPRM, we are sunsetting them on September 1, 2012, two years after the end of the second phase-in. In response to a wide consensus among commenters, we have concluded that extending their availability to that date is desirable to ensure that consumers have had a chance to gain substantial experience with advanced air bag systems. This should ensure that confidence in those systems is strong enough by the sunset date to remove any desire for a manual on-off switch in vehicles produced with an advanced air bag.

Labels with strong warning messages.

We have decided to adopt a new permanent sun visor label for vehicles certified as meeting the requirements of this rule. We proposed to alter the wording of the label to reflect the lower risk that will be associated with advanced air bags. However, all commenters, including the safety groups which supported a higher maximum test speed for the unbelted rigid barrier test, objected. They noted that while advanced air bags will significantly reduce the risk of death or serious injury, they will not eliminate all risk. Accordingly, we have decided that the new label should have warnings similar to those on the current label. The label will also have new graphics. In addition, we have adopted a new temporary label that states that the vehicle meets the new requirements for advanced air bags. Like the new permanent label, the new temporary label will have warnings similar to those on the current temporary one.

C. Future Rulemaking Plans

Final decision on maximum test speed for unbelted rigid barrier test.

As noted above, we are planning a multi-year effort to obtain additional data to help resolve the issues and concerns relating to the maximum test speed for the unbelted rigid barrier test in the long run. Those activities are described in the section below entitled, “Monitoring of Implementation and Field Experience; Research and Technology Assessment.” Based on the results of those information gathering and analysis efforts, we will make a final decision regarding the maximum test speed for unbelted dummy testing in the long run, after providing an opportunity for informed public comment.

New rulemaking proposals.

NHTSA plans to issue several proposals for further improvements in frontal occupant crash protection. One proposal would be to increase the maximum speed for the belted rigid barrier test using the 5th percentile adult female from 48 km/h to 56 km/h (30 to 35 mph). That proposal would bring the top speed for belted testing with the 5th percentile adult female dummies into line with the top speed for belted testing with the 50th percentile adult male dummies adopted in this rule. To provide data to support that proposal, we plan to initiate testing with the 5th percentile adult female in 56 km/h (35 mph) belted tests. We anticipate that if this proposal were adopted as a final rule, implementation would begin during the second stage phase-in established by today's rule. Because 56 km/h (35 mph) is the same speed at which we currently conduct our New Car Assessment Program (NCAP) frontal crash tests using belted 50th percentile adult male dummies, we will ask also for public comments on what adjustments, if any, we should make to the frontal NCAP test program.

Another proposal would be to adopt a high speed belted offset deformable barrier test. The addition of this test to Standard No. 208 would lead to improved vehicle structure, improved occupant compartment integrity and thus reduced injuries due to intrusion. This would benefit both belted and unbelted occupants. We submitted a first status report on this initiative to Congress in April 1997, and will submit a second one this spring. We expect to issue the proposal later this year.

NHTSA is also developing proposals for adding additional test dummies to Part 572 of Title 49 CFR. The two dummies that are furthest along in their development are a dummy representing a 10-year-old child and a dummy representing a 95th percentile adult male.

D. Monitoring of Implementation and Field Experience; Research and Technology Assessment

To promote the achievement of the goals of this rule and to obtain additional data that will aid us in making a final decision about the maximum test speed that should be specified for the unbelted rigid barrier test, we are planning a multi-year effort to obtain additional data.

18

This effort will include a variety of activities. We will continue to gather and evaluate real-world crash data to monitor the effectiveness of redesigned and advanced air bags in protecting various groups and subgroups of occupants and in preventing air bag-induced deaths and injuries. We are going to continue our research program, including conducting unbelted barrier tests of current vehicles at various speeds, including 48 km/h (30 mph), and analyzing those test results. In that way, we can assess how well the manufacturers simultaneously preserve and improve protection for all occupants, belted and unbelted, and minimize risk. Further, we need to continue our research and testing regarding advanced air bag technologies to gain an understanding of the safety performance implications of various features of air bag design. In addition, we will prepare an annual “compliance margins” report to assess the extent to which vehicle manufacturers exceed the 40 km/h (25 mph) test requirement.

18

NHTSA would welcome the help of interested persons in gathering data useful in achieving these purposes. The agency notes that the Alliance of Automobile Manufacturers has offered to gather information on how people die in high speed crashes.

III. Our Proposals for Advanced Air Bags

A. Our Initial Proposal (September 1998)

Pursuant to TEA 21, on September 18, 1998, we published in the

Federal Register

(63 FR 49958) a notice of proposed rulemaking (NPRM) to upgrade Standard No. 208 to require vehicles to be equipped with advanced air bags that meet new, more rigorous performance requirements. The NPRM proposed to require advanced air bags in some new passenger cars and light trucks beginning September 1, 2002, and in all new cars and light trucks beginning September 1, 2005.

We proposed several new performance requirements to ensure that the advanced air bags do not pose unreasonable risks to out-of-position occupants. The NPRM gave options for complying with those requirements so that vehicle manufacturers would be free to choose from a variety of effective technological solutions and to develop new ones if they so desire. With this flexibility, they could use either technologies that modulate or otherwise control air bag deployment so deploying air bags do not cause serious injuries or technologies that prevent air bag deployment if children or out-of-position occupants are present, or both.

To ensure that the new air bags are designed to avoid causing injury to a broad array of occupants, we proposed test requirements using a family of dummies, including ones representing 12-month-old, 3-year-old and 6-year-old children, and 5th percentile adult females, as well as tests representing 50th percentile adult males. We noted that many of the proposed test procedures were new, and specifically requested comments about their suitability for measuring the performance of the various advanced systems under development.

We also proposed requirements to ensure that the new air bags are designed to protect an array of belted and unbelted occupants, including teenagers and small adults. The standard's current crash test requirements specify the use of 50th percentile adult male dummies only. We proposed also to specify the use of 5th percentile adult female dummies in crash tests. The weight and size of these dummies are representative of not only small women, but also many teenagers. By testing with both the 50th percentile adult male dummy and the 5th percentile adult female dummy, we can address the risks faced by most of the entire adult female population and much of the adult male population.

19

19

A 95th percentile adult female, on average, weighs 199 lb and stands 5′7″ tall. The 50th percentile adult male dummy weighs 171 lb and stands 5′9″ tall.

In addition to the existing rigid barrier test, representing a relatively “stiff” or “hard” pulse crash when conducted perpendicularly, and a more moderate pulse crash when conducted obliquely, we proposed to add a deformable barrier crash test, representing a relatively “soft” pulse crash. This proposed new soft pulse crash test requirement was intended to ensure that air bag systems are designed so that they do not deploy too late. Some current air bags deploy relatively late in certain types of crashes, such as pole impacts. If an air bag deploys too late, normally seated occupants may move too close to the air bag before it starts to inflate. In such a situation, the air bag is less likely to protect the occupant and more likely to pose a risk to the occupant. We proposed to use belted 5th percentile adult female dummies in this test because small adults sit farther forward than larger adults and thus represent a greater challenge for restraint system design.

We also proposed to phase out the unbelted sled test option as we phased in requirements for advanced air bags. We acknowledged that the sled test option has been an expedient and useful temporary measure that enabled the manufacturers to speed up the redesigning all of their air bags to reduce risks. The sled test also helped to ensure that protection would continue to be provided by air bags in high-speed crashes. Nevertheless, we stated that sled testing was not a fully satisfactory means of assessing the extent of occupant protection that a vehicle and its air bag together will afford occupants in the real world and thus was not suitable in the long run.

Finally, we proposed new and/or upgraded injury criteria for each of the proposed new test requirements, and also proposed to upgrade some of the injury criteria for the standard's existing test requirements.

B. Our Supplemental Proposal (November 1999)

We received comments on the September 1998 NPRM from a wide range of interested persons including vehicle manufacturers, air bag manufacturers, insurance companies, public interest groups, academia, and government agencies. Commenters expressed widely differing views as to how to accomplish the goals mandated by TEA 21— improving the benefits of air bags, while minimizing risks from air bags.

On November 5, 1999, in response to the public comments on our 1998 NPRM and to other new information we obtained after issuing that proposal, we published the SNPRM (64 FR 60556), which updated and refined the amendments under consideration in this rulemaking.

In the SNPRM, we reiterated the goals set for us by Congress in TEA 21,

i.e.,

to improve occupant protection for occupants of different sizes, belted and unbelted, while minimizing the risk to infants, children, and other occupants from injuries and deaths caused by air bags. Further, we emphasized the need to ensure that the needed improvements in occupant protection were made in accordance with the statutory implementation schedule.

In developing the SNPRM, we sought to reduce the number of proposed tests to the extent possible without

significantly affecting the benefits of the NPRM. We were persuaded by the commenters that reducing the amount of testing was important, given the costs to manufacturers (and ultimately consumers) associated with certifying vehicles to such a large number of new test requirements. At the same time, we wanted to be sure that the final rule would include sufficient tests to ensure that air bags would meet the goals of TEA 21. Given the continued debate over what requirements should be relied upon to ensure protection to unbelted occupants, we also wanted to be sure that we received and considered public comments on the various alternative approaches reflecting the more recent views and information available to us.

The most significant differences between the NPRM and the SNPRM can be summarized as follows:

•

Two alternative unbelted test procedures.

While we proposed one unbelted test procedure in the NPRM, an up-to-48 km/h (30 mph) rigid barrier test using the 50th percentile adult male dummy and the 5th percentile adult female dummy, we proposed and sought comments on two alternative unbelted test procedures in the SNPRM.

The first alternative was an unbelted rigid barrier test whose injury criteria would have to be met within the range of a minimum speed of 29 km/h (18 mph) and a maximum speed to be established between 40 to 48 km/h (25 to 30 mph), inclusive. Within this alternative was the potential for a phase-in sequence in which the maximum speed would initially be set at 40 km/h (25 mph) to provide vehicle manufacturers additional flexibility when they are introducing advanced air bags during the phase-in. Under this phase-in sequence, the final rule could provide that the maximum speed would return to 48 km/h (30 mph) after some period of time. We also sought comment on setting the maximum speed at 48 km/h (30 mph) but temporarily permitting relaxed injury criteria performance limits (

e.g.

72 g chest acceleration limit instead of 60 g chest acceleration limit) in rigid barrier crashes between 40 km/h (25 mph) and 48 km/h (30 mph).

The second alternative was an unbelted offset deformable barrier test within the range of a minimum speed of 35 km/h (22 mph) and a maximum speed to be established within the range of 48 to 56 km/h (30 to 35 mph). The latter alternative was developed in response to a recommendation made by IIHS in its comment on the NPRM.

20

20

We noted that IIHS's views had changed since making that recommendation. Its changed views were discussed in the SNPRM.

We proposed the 29 and 35 km/h (18 and 22 mph) lower ends of the ranges of test speeds for the two alternatives because we wanted to be sure that the standard would not inadvertently create incentives to push deployment thresholds downward;

i.e.,

cause air bags to be deployed at lower speeds.

•

Possible higher speed belted rigid barrier test.

We stated that if we reduced the maximum speed of the unbelted rigid barrier test to 40 km/h (25 mph), we might also increase the maximum speed of the belted rigid barrier test from the current 48 km/h to 56 km/h (30 to 35 mph) and use both 5th percentile adult female and 50th percentile adult male dummies.

•

Reduced number of tests.

In the SNPRM, we significantly reduced the total number of proposed tests as compared to the NPRM. In a number of situations, we tentatively concluded that a proposed test could be deleted because the performance we sought to secure by means of that test would largely be assured by one or more of the other tests.

•

Reduced offset testing.

The proposed up-to-40 km/h (25 mph) offset crash test using belted 5th percentile adult female dummies would be conducted only with the driver side of the vehicle engaged, instead of both testing with the driver side engaged and separately testing with the passenger side engaged.

•

Ensuring that certain static suppression systems can detect real children and adults.

For our proposed static test requirements for systems which suppress air bags in the presence of infants and children (

e.g.,

weight sensors), we proposed a new option which would permit manufacturers to certify to requirements referencing actual children, instead of 3-year-old and 6-year-old child dummies, in a stationary vehicle to test the suppression systems. (This option would not apply to systems designed to suppress the air bags only when an infant is present.) Adult human beings could also be used in the place of 5th percentile adult female dummies for the portions of those static test requirements which make sure that the air bag is activated for adults. Steps would be taken to ensure the safety of all subjects used for these tests,

e.g.,

by turning off the air bags.

•

Reduced number of child restraints used for testing suppression systems.

Instead of requiring manufacturers to assure compliance of a vehicle in tests using any child restraint which was manufactured for sale in the United States any time during a specified period prior to the manufacture of the vehicle, we proposed to require them only to assure compliance using each child restraint on a relatively short list of specified child restraint models. Those models would be chosen to be representative of the array of available child restraints. The list would be updated from time to time to reflect changes in the types of available child restraints.

•

Modified requirements for systems that suppress the air bag for out-of-position occupants.

We significantly modified the proposed requirements for systems that suppress the air bag when an occupant is out of position during a crash. In the NPRM, we proposed a single test procedure for all types of such suppression systems. However, we were persuaded by the commenters that the proposed test procedure was not appropriate for many of the systems that are currently under development. Because we did not have sufficient information or prototype hardware to develop a new test procedure, and because no single test procedure may be appropriate for the broad spectrum of suppression technologies currently being developed, we proposed a provision that would permit manufacturers or others to petition the agency to establish technology-specific test procedures under an expedited rulemaking process.

•

No full scale dynamic out-of-position test requirements.

We eliminated from this rulemaking the proposed option for full scale dynamic out-of-position test requirements (the option which included pre-impact braking as part of the test procedure). We were persuaded by the commenters that the proposed test procedure was not workable at this time. Moreover, we concluded that this option was unnecessary at this time, since other options were available for the range of effective technologies we understand to be currently under development.

In developing the SNPRM, we carefully considered all of the comments we received in response to the NPRM. Moreover, because the SNPRM differed significantly in many aspects from the NPRM, we explained that we did not contemplate any further consideration of the comments on the NPRM in developing the final rule. We stated that if any persons believe that we did not adequately consider particular issues raised in comments on the NPRM, they should raise those

issues again in commenting on the SNPRM.

Accordingly, in developing today's final rule, we have focused our consideration on the comments submitted in response to the SNPRM.

IV. Public Comments on the Supplemental Proposal

As in the case of the NPRM, we received comments on the November 1999 SNPRM from a wide range of interested persons including vehicle manufacturers, air bag manufacturers, insurance companies, public interest groups, and government agencies. In this section, we provide a general summary of those comments. A more detailed description of the comments is provided below in the sections which address the issues raised by commenters, and in the Final Economic Assessment and three separate technical papers which are being placed in the public docket.

Improving the Protection of Unbelted Occupants in Serious Crashes

Nearly all commenters supported the unbelted rigid barrier test over the unbelted offset deformable barrier test.

Vehicle manufacturers stated that the rigid barrier test is practicable and repeatable and does not entail the variability associated with deformable barriers and the kinematics of an offset test. They also stated that the European barrier used in the offset test is not appropriate for testing larger SUVs and light trucks.

The Center for Auto Safety (CAS) stated that the unbelted offset test holds promise as a supplemental test, but is not yet suitable for inclusion in Standard No. 208. That organization stated that there are currently insufficient data to allow for a comprehensive analysis of the consequences that would accompany the adoption of the offset test.

Some other commenters also argued that an unbelted offset test offers promise for the future, either as a replacement for the rigid barrier test or as a supplemental test.

While a near-consensus of commenters supported adoption of an unbelted rigid barrier test, there was sharp disagreement over the maximum speed for that test. The vast majority of commenters, including all auto companies and all air bag suppliers, the Insurance Institute for Highway Safety (IIHS), and the National Transportation Safety Board (NTSB) supported a maximum speed of 40 km/h (25 mph). Safety groups including Public Citizen, CAS, Consumers Union, and Parents for Safer Air Bags (Parents) supported returning to 48 km/h (30 mph).

The primary arguments made by those commenters supporting a maximum speed of 40 km/h (25 mph) can be summarized as follows:

• Current redesigned air bags work well.

• There has been no loss in benefits.

• There is no reason to believe that manufacturers would reduce air bag effectiveness in the future under a 40 km/h (25 mph) maximum test speed.

• A 40 km/h (25 mph) test speed allows flexibility to design air bags for all occupants.

• A return to a 48 km/h (30 mph) test speed would require a return to overly aggressive air bags.

• Aggressive air bags cause deaths in high speed crashes as well as low speed crashes.

• A 48 km/h (30 mph) test speed could result in disbenefits in low speed crashes.

• There are significant technological challenges in meeting a 48 km/h (30 mph) requirement for both the 50th percentile adult male dummy and the 5th percentile adult female dummy.

• Advanced technologies are not currently available that address aggressivity and practicability problems.

The primary arguments made by those commenters supporting a maximum test speed of 48 km/h (30 mph) can be summarized as follows:

• A maximum test speed of 48 km/h (30 mph) will result in higher benefits than a test speed of 40 km/h (25 mph).

• Half of all fatalities in frontal crashes occur at a delta V above 48 km/h (30 mph); a maximum test speed of 48 km/h (30 mph) represents significantly more potentially fatal crashes than a test speed of 40 km/h (25 mph).

• In NHTSA tests, almost all vehicles with redesigned air bags passed the 48 km/h (30 mph) rigid barrier test with the 50th percentile adult male dummy, implying that a return to a 48 km/h (30 mph) test speed would not require a return to overly aggressive air bags.

• Advanced technologies can be used to enable all vehicles to meet requirements for high speed protection and risk reduction.

• There is no justification to reduce the test speed to 40 km/h (25 mph).

• A 40 km/h (25 mph) test speed would not encourage use of advanced technologies.

• A 40 km/h (25 mph) test speed would be inconsistent with the TEA 21 requirement to improve protection for unbelted occupants.

• The increase of the belted test speed to 56 km/h (35 mph) would not recover lives lost as a result of reducing the unbelted test speed to 40 km/h (25 mph).

While maximum speed was the most controversial issue concerning the unbelted test, commenters raised other issues as well. Some vehicle manufacturers objected to the proposal to test over a range of speeds from 29 km/h (18 mph) to the highest speed. They argued that being required to meet test requirements to ensure protection beginning at 29 km/h (18 mph), combined with the proposal to test under the low risk deployment option for inflation level (or levels) that would be deployed in crashes below 29 km/h (18 mph), would limit design flexibility and discourage development of low risk deployment air bag systems.

Another significant issue addressed by commenters concerned the seating procedure for the 5th percentile adult female dummy. Vehicle manufacturers objected to the proposal to test with the seat in the full forward position. They argued that occupants, including small females, rarely if ever sit in that position. They also argued that adoption of this position could result in consequences such as smaller, less protective air bags, and reduced ingress/egress space for rear passengers.

Several safety advocacy groups argued in favor of testing with the seat in the full forward position. They argued that some occupants sit in that position and that it is necessary to test in the “worst case” condition.

Improving the Protection of Belted Occupants in Serious Crashes

Commenters supported our proposal to add the 5th percentile adult female dummy to the existing 48 km/h (30 mph) belted rigid barrier test.

Most supporters of a 40 km/h (25 mph) unbelted rigid barrier test, including most vehicle manufacturers, also supported increasing the maximum speed of the belted rigid barrier test to 56 km/h (35 mph). However, these commenters urged that the 56 km/h (35 mph) belted rigid barrier test be phased in after the TEA 21 phase-in period. They also urged that the higher speed test initially be conducted only with the 50th percentile adult male dummy, and that a separate rulemaking be initiated to consider whether the 5th percentile adult female dummy should be tested at that speed.

Most commenters also supported our proposal to add the up-to-40 km/h (25 mph) offset deformable barrier test using belted 5th percentile adult female dummies. Some of these commenters, however, urged that an out-of-position test for the passenger side be developed as an alternative to the test.

DaimlerChrysler opposed adoption of this test, arguing that the European barrier used in the test is not appropriate for testing heavier vehicles such as SUVs and light trucks.

Some commenters expressed concerns that our proposal would result in there being too many crash tests in Standard No. 208, and requested that we reconsider whether all of the proposed tests are needed.

Minimizing the Risk of Injuries and Deaths Caused by Air Bags

Commenters supported the basic approach of our proposed requirements to minimizing the risk of injuries and deaths caused by air bags, including providing a variety of testing options that account for the kinds of effective technological solutions that are under development.

Vehicle manufacturers argued that some of the test conditions specified for the proposed static suppression tests, including the range of seat back angles and seat track positions, would make the tests impracticable.

Some commenters emphasized that we need to allow manufacturers to use both suppression and low risk technologies. As noted earlier, some commenters argued that adjustments need to be made in both the unbelted rigid barrier test requirements and in the requirements for the low risk deployment option to avoid limiting use of the low risk deployment option.

Commenters were generally supportive of our proposal to permit manufacturers to certify to requirements referencing human beings in a stationary vehicle to test suppression systems, so long as steps are taken to ensure the safety of all subjects used for testing.

Other Issues

Commenters generally supported the proposed injury criteria and associated performance limits, although vehicle manufacturers recommended some changes.

We received numerous comments raising specific technical issues concerning how dummies are to be positioned for the various tests.

Commenters generally argued that current provisions allowing manual on-off switches for air bags under certain circumstances should remain in effect for a longer period of time, and a number of commenters argued that existing warning labels should not be weakened or eliminated at this time.

There was also significant differences of opinion regarding our proposals about the provision providing a due care defense against findings of noncompliance with the air bag requirements of Standard No. 208.

Several commenters raised concerns about possible unforeseen consequences resulting from the use of advanced air bag technologies.

We received several comments expressing concern about the potential impacts of this rulemaking on small businesses.

V. Diagrams of the Final Rule Requirements

After carefully considering the comments, we have decided to issue a final rule along the lines of the SNPRM. The key differences between the SNPRM and the final rule are discussed earlier and will not be repeated here. The test requirements to improve occupant protection for different size occupants, belted and unbelted, and to minimize risks to infants, children, and other occupants from injuries and deaths caused by air bags, are shown in Figures 1 and 2 below.

ER12MY00.000

ER12MY00.001

BILLING CODE 4910-59-P

VI. Improving the Protection of Unbelted Occupants in Serious Crashes

A. Summary of Proposed Requirements

In the SNPRM, we proposed to phase out the unbelted sled test option as the requirements for advanced air bags are phased in. As explained below, sled tests have inherent limitations as compared to crash tests in measuring occupant protection.

We explained that, unlike a full scale vehicle crash test, a sled test cannot measure the actual protection an occupant will receive in a crash. We noted that while the current sled test measures some performance attributes of the air bag, it cannot measure the performance provided by the vehicle structure in combination with the air bags or even the full air bag system by itself. We also noted that the sled test does not evaluate the actual timing of air bag deployment (

e.g.,

crash sensors), does not replicate the actual crash pulse of a particular vehicle model, does not measure the potential for harm from vehicle components that are pushed back into the occupant compartment during a crash, and does not measure how a vehicle performs in angle crashes.

The purpose of the sled test option was to make it easier for vehicle manufacturers to make quick changes to their air bags to reduce risks to out-of-position occupants. Vehicle manufacturers could not immediately incorporate advanced technologies in their vehicles, and the sled test facilitated the process of quickly certifying large numbers of vehicles with redesigned air bags to Standard No. 208. We believe the sled test has been useful as a short-term measure. Over the longer time frame, however, we believe that a better test is needed to ensure the protection of unbelted occupants.

To replace the sled test, we proposed two alternative unbelted crash test procedures: an unbelted rigid barrier test and an unbelted offset deformable barrier test. We proposed that the unbelted rigid barrier test be conducted perpendicular and up to ± 30 degrees oblique to perpendicular with 50th percentile adult male dummies, but perpendicular only in tests with 5th percentile adult female dummies. The injury criteria would have to be met within the range of a minimum speed of 29 km/h (18 mph) and a maximum speed to be established within the range of 40 to 48 km/h (25 to 30 mph). This alternative was based on the unbelted crash test that has been part of Standard No. 208 for many years but which has, as a practical matter, been temporarily superseded by the sled test option since March 1997. The barrier test represents a vehicle striking a vehicle of the same size, weight and structure head on at the same speed.

We indicated that within this first alternative, the potential existed for a phase-in sequence in which the maximum speed would temporarily be set at 40 km/h (25 mph) to provide vehicle manufacturers additional flexibility when they are introducing advanced air bags during the TEA 21 phase-in. Under this approach, the final rule could provide that a maximum speed of 48 km/h (30 mph) would apply after that period. We also indicated that if we were to reduce the maximum speed to 40 km/h (25 mph), we might also increase the maximum speed of the belted rigid barrier test from the current 48 km/h to 56 km/h (30 to 35 mph).

We proposed that the second alternative procedure, the unbelted offset deformable barrier test, would be conducted using both 50th percentile adult male dummies and 5th percentile adult female dummies, with a minimum speed of 35 km/h (22 mph) and a maximum speed to be established within the range of 48 to 56 km/h (30 to 35 mph). This alternative was based on a type of crash test used by IIHS and by Europe, except that unbelted dummies would be used.

For both alternatives, we proposed to conduct the crash tests with 50th percentile adult male dummies with the seat in the middle seat track position. However, we proposed in the SNPRM to conduct tests using 5th percentile adult female dummies with both the driver and passenger seats in the full forward position. We tentatively selected this position because some small adults sit there and because we believe that air bags should protect those people.

We noted, however, that placement of the 5th percentile adult female dummy in the full forward position tests the occupant restraint system under a condition that may not generally occur in the real world. The University of Michigan Transportation Research Institute (UMTRI) conducted a study in which it concluded that even drivers who are approximately the same size as the 5th percentile adult female dummy generally do not sit in the full forward seat track position (Docket No. NHTSA-1998-4405-69). Also, while some short-statured drivers might need to move the driver's seat all the way forward to reach the controls, a passenger in the front passenger seat would be less likely to have a similar need. Another concern was whether, in order to meet tests for conditions that rarely occur in the real world, manufacturers might select air bag designs that offer reduced protection for conditions that are more common in the real world. Accordingly, we requested comments on whether testing the 5th percentile adult female dummy with the seat in something other than the full forward seat track position would adequately protect properly-seated individuals of all sizes while potentially allowing more design freedom.

B. Type of Test

Commenters that previously advocated retention of the sled test indicated a willingness to accept the unbelted rigid barrier test. No commenters disputed the inherent limitations of sled tests as compared to crash tests. Nearly all commenters supported the unbelted rigid barrier test over the unbelted offset deformable barrier test. However, as discussed in the next section, the commenters that had previously supported the sled test wanted the maximum speed of the unbelted barrier test reduced to 40 km/h (25 mph).

Vehicle manufacturers stated that the rigid barrier test is practicable and repeatable and does not entail the variability associated with deformable barriers and the kinematics of an offset test. They also stated that the European barrier used in the offset test is not appropriate for testing larger SUVs and light trucks. Several vehicle manufacturers, including GM, Honda and DaimlerChrysler, stated that a high speed unbelted offset test would pose problems for vehicle sensor systems.

CAS stated that the unbelted offset test holds promise as a supplemental test, but is not yet suitable for inclusion in Standard No. 208. That organization stated that there are currently insufficient data to allow for a comprehensive analysis of the consequences that would accompany the adoption of the offset test.

Several other commenters also argued that an unbelted offset test offers promise for the future, either as a replacement for the rigid barrier test or as a supplemental test. Ford stated that although not practicable during the TEA 21 phase-in period, it believes that a 48 km/h (30 mph) offset test potentially represents a better long-term approach for enhancing unbelted protection.

Parents stated that the final rule should include both the unbelted rigid barrier test and the unbelted offset test. That organization argued that the two tests provide distinct means of ensuring protection in very different circumstances, and that inclusion of both tests is necessary in order to ensure adequate protection for unbelted occupants.

After considering the comments, we have decided to adopt the unbelted rigid barrier test to ensure protection for unbelted occupants in serious crashes. This is the unbelted crash test included in Standard No. 208 for the past 30 years. We also use a belted rigid barrier test for Standard No. 208 and our New Car Assessment Program (NCAP). Detailed information about this type of test is presented in a paper prepared by our Office of Research and Development titled “Updated Review of Potential Test Procedures for FMVSS No. 208.” That paper was prepared to accompany our SNPRM.

21

21

One commenter, DaimlerChrysler, submitted a critique of that paper as part of its comments. We are placing in the docket an addendum to the paper which responds to that critique.

We note that we sought comment in the SNPRM on the unbelted offset test principally to ensure that we received the benefit of public comments on all of the various alternative approaches that are available at this time. In the NPRM, we indicated that while we believed the unbelted rigid barrier test was a good approach, we were also willing to consider alternative unbelted crash tests. The only alternative unbelted crash test advocated by a commenter that could realistically be implemented within the time frame of this rulemaking was the offset deformable barrier test.

However, the commenter that originally suggested consideration of the unbelted offset test, IIHS, withdrew its support before the SNPRM was published. No commenter on the SNPRM supported adopting the unbelted offset test instead of the unbelted rigid barrier test.

As to Parents' recommendation that we adopt both unbelted tests, we believe that adoption of the proposed unbelted high speed offset test would be inappropriate at this time. We have scant data on the repeatability of this test. Nearly all the offset testing to date has used belted dummies. As noted above and also discussed in the SNPRM, several manufacturers have raised concerns that the proposed high speed unbelted offset test would pose problems for vehicle sensor systems. See 64 FR 60579.

We also note that while we agree with Parents that the two high speed tests provide distinct means of ensuring protection in different circumstances, this does not mean that adoption of those particular two tests would be needed to ensure protection in those different circumstances. We believe that the combination of an unbelted rigid barrier test and

belted

offset tests can accomplish the same purpose.

As discussed in the SNPRM, the high speed unbelted rigid barrier test and the high speed unbelted offset test are significantly different, and each has potential advantages as compared to the other. The two principal advantages of an offset test are that it provides a more challenging test of vehicle crash sensors and of vehicle structure. However, these areas of performance are addressed by belted offset tests as well as unbelted offset tests.

As discussed later in this document, we are adopting an up to 40 km/h (25 mph) belted offset deformable barrier test as part of today's final rule. This test will help ensure improved sensing systems, which will benefit both belted and unbelted occupants. We are also separately pursuing our previously-announced plans to consider adding a high speed belted offset test to Standard No. 208. This test would help ensure improved vehicle structure and reduced intrusion injuries, again benefitting both belted and unbelted occupants. Because the combination of an unbelted rigid barrier test and belted offset tests (either being adopted today or currently being considered by the agency for rulemaking) can accomplish the same purpose as an unbelted offset test, we do not currently plan to consider further adopting an unbelted offset test.

C. Agency Decision to Establish Maximum Speed at 40 km/h (25 mph)

1. The Supplemental Proposal

In the SNPRM, we proposed that the maximum speed for the unbelted rigid barrier test be established within the range of 40 to 48 km/h (25 to 30 mph).

We stated that it was our intent to maximize, to the extent consistent with TEA 21, the protection that air bags offer in crashes potentially resulting in fatal injuries. Thus, we stated that it was our preference to establish such a test requirement at as high a severity as practicable. We stated that the 40 km/h (25 mph) lower end of the maximum test speed range was set forth for comment to ensure that commenters addressed a crash test recommended by AAM in late August 1999.

We also stated that the potential existed for a phase-in sequence in which the maximum speed would initially be set at 40 km/h (25 mph) to provide vehicle manufacturers additional flexibility when they are introducing advanced air bags during the phase-in. We explained that under this phase-in sequence, the final rule could provide that a maximum speed of 48 km/h (30 mph) would apply after a reasonable period of time.

We noted that, in commenting on the NPRM, the commenters opposing the 48 km/h (30 mph) unbelted barrier test had raised two primary issues. First, they argued that the test is not representative of typical crashes. Second, they argued that returning to this test would prevent continued use of “depowered” air bags and would require a return to “overly aggressive” air bags.

We addressed each of these issues in the SNPRM. As to whether the test is representative of typical crashes, we stated that because the purpose of Standard No. 208 is primarily to reduce serious and fatal injuries, we believed that the relevant question is how representative the test is of the crashes that produce those injuries. We presented data from the National Automotive Sampling System (NASS) for years 1993-1997 showing, among other things, that about 50 percent of fatalities in frontal crashes occur at delta Vs below 48 km/h (30 mph), and about 50 percent occur at delta Vs above 48 km/h (30 mph). Looking separately at unbelted and belted occupants, we noted that 51 percent of the fatalities involving unbelted occupants and 47 percent of the fatalities involving belted occupants occur in frontal crashes at delta Vs below 48 km/h (30 mph). We noted that the delta V in NASS represents the speed at which the vehicle would strike a rigid barrier to duplicate the amount of energy absorbed in the crash. Thus, about half of fatalities in frontal crashes occur in crashes that are more severe than a 48 km/h (30 mph) rigid barrier crash, and half of all frontal crash fatalities occur in crashes that are less severe than a 48 km/h (30 mph) rigid barrier crash.

Given that Standard No. 208's unbelted crash test requirements are intended to save lives, we stated that we disagree that 48 km/h (30 mph) rigid barrier crashes are unrepresentative of the kinds of crashes in which we are seeking to ensure protection. We also noted that because we were proposing to require vehicles to meet the unbelted test requirements for a range of speeds up to and including 48 km/h (30 mph), we were addressing protection for lower severity crashes as well as higher severity crashes.

As to the argument that returning to the unbelted 48 km/h (30 mph) rigid barrier test would prevent continued use of “depowered” air bags and require use of “overly aggressive” air bags, we stated that a key way of assessing the validity of the argument that a return to the 48 km/h (30 mph) barrier test would—at least in the absence of additional technological improvements—prevent continued use

of redesigned air bags was to test vehicles with those air bags in 48 km/h (30 mph) barrier tests and see how they perform. We noted that we had tested a total of 13 MY 1998-99 vehicles with redesigned air bags in a perpendicular rigid barrier crash test at 48 km/h (30 mph) with unbelted 50th percentile adult male driver and passenger dummies. The vehicles represented a wide range of vehicle types, sizes, and crash pulses.

22

22

In particular, the 13 vehicles included one sub-compact car, one compact car, four mid-size cars (representing high sales volume vehicles), one full-size car, two mid-size sport utility vehicles, one full-size sport utility vehicle, one pickup truck, one minivan, and one full-size van. The specific vehicles and their classes included a Saturn (sub-compact car), a Neon (compact car), an Intrepid, Camry, Taurus, and Accord (mid-size cars), an Acura RL (full-size car), an Explorer and Cherokee (mid-size SUVs), an Expedition (large SUV), a Tacoma (pickup truck), a Voyager (minivan), and an Econoline (full-size van).

We stated that 11 of the 13 vehicles passed the injury criteria performance limits proposed in the SNPRM. For the driver position, 12 of the 13 vehicles passed all the relevant injury criteria performance limits. In the one vehicle with a failure, the MY 1999 Acura RL, the driver dummy exceeded the femur load criteria. For the passenger position, 12 of the 13 vehicles also passed all of the relevant injury criteria performance limits. The MY 1998 Dodge Neon slightly exceeded the 60 g chest acceleration limit (with a value of 61.4 g). The other proposed injury criteria performance limits (

i.e.,

for HIC, chest deflection, and Nij) were easily met in all the tests; for most vehicles, there was a greater than 20 percent margin of compliance for both the driver and passenger seating positions.

Based on these test results, we stated that the tested vehicles with redesigned air bags, ranging widely in vehicle type and size, appeared to continue to meet Standard No. 208's 48 km/h (30 mph) unbelted rigid barrier test requirements for 50th percentile adult male dummies, many of them by wide margins.

We also noted that the relevant issue for this rulemaking is not whether some MY 1998-99 vehicles with redesigned, single-inflation level air bags would not meet a 48 km/h (30 mph) unbelted barrier test requirement. The more relevant issue is whether vehicles to be manufactured in MY 2003 and later would be able to comply with such a requirement, perhaps by means of currently available technologies not in many air bag systems as well as technologies still being or yet to be developed.

We explained that today's air bag systems are not advanced air bags and thus do not respond to factors such as crash severity, occupant weight and occupant location. By contrast, the incorporation of advanced technologies would make air bag systems responsive to those factors.

We also noted:

If a manufacturer decided to use a somewhat more powerful air bag to meet a 48 km/h (30 mph) unbelted rigid barrier test, or to provide protection in more severe crashes, the manufacturer could use advanced air bag technologies to provide less powerful levels of inflation in lower severity crashes, for smaller occupants, for belted occupants, and for occupants sitting with the seat in the full-forward position. Manufacturers could also reduce aggressivity of air bags by various means such as optimizing fold patterns, different cover designs, lighter fabrics, etc. Advanced technologies would also enable the manufacturer to suppress air bag deployment in appropriate circumstances, such as when children are present.

In our Preliminary Economic Assessment (PEA) accompanying the SNPRM, we estimated the benefits of an unbelted rigid barrier test with a maximum speed of 40 km/h (25 mph) vs. 48 km/h (30 mph). The PEA concluded that if the full fleet of vehicles' air bags were designed in the context of unbelted 40 km/h (25 mph) rigid barrier and oblique tests, an estimated 214 to 397 lives saved annually by pre-MY 1998 air bags might not be saved.

2. Summary of Comments

Commenters on the SNPRM nearly unanimously supported adoption of an unbelted rigid barrier test, but sharply disagreed over the maximum speed for that test. Safety advocacy groups, supported returning to 48 km/h (30 mph). Most commenters, however, including all auto companies and all air bag suppliers, IIHS, and NTSB supported a maximum speed of 40 km/h (25 mph).

Commenters supporting 40 km/h (25 mph).

Commenters supporting a maximum test speed of 40 km/h (25 mph) argued that there would not be a loss of benefits associated with a test at this speed, as compared to a 48 km/h (30 mph) standard.

AAM stated that the benefits of redesigned air bags will be maintained with a 40 km/h (25 mph) test. It argued that there is no reason to believe air bags designed to the sled test requirements have compromised protection, and that a 40 km/h (25 mph) barrier test is more severe than the sled test.

AAM also stated that a new 40 km/h (25 mph) test cannot simply be compared to the old 48 km/h (30 mph) test because the new test would include additional injury criteria and an additional dummy. It stated that the benefits of the other tests included in the final rule, such as the new belted offset test and the low speed risk reduction tests, should also be considered.

AAM argued that the analyses of benefits presented in the PEA are based on dummy readings from one dummy at one position in a single type of crash test in a single direction at a single speed. It stated that this approach is not comprehensive enough. AAM also argued that the strongest evidence that there are analytical limitations inherent in the agency's benefit analyses (past and present) is that past analyses predicting 1,250 lives lost from the adoption of the sled test that simply have not come true.

AAM stated that it had considered the level-of-benefit question from two different perspectives. The first involved the generation of benefit estimates using a MADYMO math model to develop a theoretical “optimum” design for both the 40 km/h (25 mph) and 48 km/h (30 mph) suite of tests. The performance of those designs was then modeled over a broad spectrum of real world crash configurations. Based upon an injury/fatality risk analysis of the model's output injury measures, relative benefits were calculated. The second perspective utilized an “opportunities matrix” approach to examine relative benefits by generating effectiveness estimates and applying these estimates to the spectrum of real world crash conditions.

According to AAM, both of these approaches yield the same conclusion—when considering air bag designs constrained by testing unbelted occupants at 40 km/h (25 mph) or 48 km/h (30 mph), the desired goal of reducing serious-to-fatal injuries in real world crashes is best served by requiring testing at 40 km/h (25 mph).

GM submitted an analysis which it said explains why a 25 mph rigid barrier test drives air bag designs that protect unbelted occupants in severe frontal crashes. Among other things, it said that ride down analysis shows that a 25 mph rigid barrier test requires more air bag restraint capacity than an unbelted offset deformable barrier impact at 40 mph.

Vehicle manufacturers stressed the argument that the agency should focus on the experience of redesigned air bags in MY 1998 and MY 1999 models. They argued that these redesigned air bags have provided real world benefits and that there is no evidence that more power is needed.

Toyota stated that NHTSA's concern that manufacturers will substantially decrease power in future air bags compared to current systems is unfounded. It presented data comparing velocity vs. time traces for the sled test and the 40 km/h (25 mph) test for both an SUV and a sedan, and noted that the 40 km/h (25 mph) test pulses were more severe. Toyota argued that, in order to manage this level of energy, the air bags for these vehicles cannot be depowered further than the current levels, and that there is no reason to believe that air bags designed to the 40 km/h (25 mph) rigid barrier test will perform worse in high speed collisions than those designed to the sled pulse.

IIHS stated that it does not agree that a high-speed barrier test using unbelted dummies will necessarily lead to improved protection for any occupants, belted or unbelted. That organization stated that it disagreed with what it characterized as the agency's claim that, unless it returns to the 48 km/h (30 mph) barrier test, air bags will offer inadequate protection to many unbelted occupants, especially large people in more severe frontal crashes. That organization stated that in a number of studies of air bag performance in moderate to severe frontal crashes, it has shown that drivers are not dying because air bags offer too little protection; rather, drivers are dying because of overwhelming intrusion that no air bag design can overcome, ejection of occupants, or because of injury from the air bag itself.

IIHS argued that these observations call attention to what it believes are two errors in the agency's logic for returning to a 48 km/h (30 mph) test. First, that commenter argued that if air bags are not powerful enough, there should be some real world cases in which the energy of the deploying bags was inadequate to protect individuals in otherwise survivable frontal crashes. IIHS stated that it is not aware of any such case. It also stated that the agency's concern that air bags certified to the unbelted generic sled pulse would be less effective in frontal crashes has no foundation in real world crash data.

Second, IIHS argued that the agency has failed to appreciate that serious and fatal injuries from deploying air bags are happening not only in low speed crashes, but also in the high speed crashes in which air bags are supposed to be most effective. That commenter stated that a recent update (including 1996 data) of its analyses of driver fatalities in air bag-equipped cars indicates air bags were the most likely source of the fatal injuries in about 15 percent of frontal crash deaths. IIHS argued that the agency must account for these deaths, as well as those more easily documented in low speed crashes, before it can justify a return to the 48 km/h (30 mph) unbelted barrier test.

IIHS also addressed the agency's concern that, without a “severe crash test” for unbelted occupants, manufacturers may reduce air bag inflation energy, or the size of air bags, thereby compromising their effectiveness. IIHS argued that such changes are constrained by other non-regulatory crash tests to which the manufacturers are subject. That organization stated that NCAP requires that air bags be reasonably deep in order to prevent dummies' heads from striking through the bags, and that offset crash testing by it and others worldwide means manufacturers will continue to install air bags with sufficient radial size to keep occupants squarely behind their air bags, even under conditions of sharp vehicle rotation.

NADA argued that the agency's proposed advanced air bag performance criteria fail to account for reasonably projected increases in safety belt and child restraint usage or for the real-life incremental benefits attributable to “depowered” air bags. NADA stated that it is reasonable to assume that by MY 2003, proper driver and passenger (including children) seat belt usage and child restraint usage rates will exceed 80 percent, and that by MY 2006, these rates should exceed 90 percent.

Vehicle manufacturers also argued that it is difficult or impossible to comply with the 48 km/h (30 mph) rigid barrier test for both the 50th percentile adult male dummies and the 5th percentile adult female dummies. They also argued that it may not be possible to satisfy both the 48 km/h (30 mph) unbelted rigid barrier test for both dummies and the low risk deployment tests.

AAM stated that while the agency has claimed that most vehicles with redesigned air bags continue to meet the unbelted 48 km/h (30 mph) barrier test, very little testing has been done with these same vehicles at 48 km/h (30 mph) with 5th percentile adult female dummies. AAM stated that the little testing that has been done produced a 50 percent failure rate. That organization stated that this testing illustrates the design tensions that the industry has been emphasizing. According to that organization, these tensions result from technology constraints which presently discern limited information about occupant size and location, crash sensors with limited predictive capability and air bags with only two power levels.

According to AAM, it is especially challenging to balance occupant protection for both the 5th percentile adult female and the 50th percentile adult male dummies and assure compliance with the barrier test. As an example, AAM cited the agency's test of the Toyota Tacoma, which resulted in an Nij of 2.65 for the 5th female passenger dummy, nearly three times the allowable injury reference value. According to AAM, the air bag size and fill needed to assure compliance with the chest injury limits with 50th percentile adult male dummies at 48 km/h (30 mph) results in noncompliant neck and thorax injury reference values for 5th percentile adult female dummies seated closer to the air bag. Conversely, according to AAM, if the air bag is sized for the unbelted 5th percentile female dummy at 48 km/h (30 mph), there is insufficient restraint of the unbelted 50th male dummy. AAM argued that testing at 40 km/h (25 mph) allows the restraint engineer to design the air bag to provide reasonable occupant protection for a broader range of occupant sizes.

GM made arguments similar to those of AAM. It argued that the unbelted 48 km/h (30 mph) barrier test using the 50th percentile adult male dummy determines the restraint energy, drives the depth of the air bag, and requires a deeper air bag that has more potential to injure a 5th percentile adult female. It argued that the unbelted 48 km/h (30 mph) barrier test using the 5th percentile adult female would require a shallower air bag that would not assure compliance for an unbelted 50th percentile adult male. According to GM, a 40 km/h (25 mph) test would permit air bag depth to be optimized for both the 5th percentile adult female and 50th percentile adult male dummies.

Ford stated that testing of the MY 2000 Taurus using 5th percentile adult female and 50th percentile adult male dummies demonstrates the difficulties of balancing requirements with a 48 km/h (30 mph) test even for vehicles equipped with advanced technologies. That company noted that the MY 2000 Taurus has dual-level inflators and other advanced technologies.

GM argued that there is no technology or combination of technologies existing today that could satisfy both the 48 km/h (30 mph) unbelted rigid barrier test and the low risk deployment tests. Honda stated that it had concerns about being able to meet the rigid barrier test for the 50th percentile adult male dummy and also meet the low risk

deployment test for out-of-position occupants.

Commenters supporting a maximum speed of 40 km/h (25 mph) also argued that a 48 km/h (30 mph) maximum speed would require a return to overly aggressive air bags.

AAM stated that field evidence suggests that the current depowered air bags offer a high level of occupant protection in the real world while enhancing protection for at-risk groups. That organization stated that a return to 48 km/h (30 mph) unbelted testing would require increasing air bag inflator outputs in some vehicles, serving to increase the risk of harm to certain groups.

GM stated that it strongly recommends that “depowered” air bags continue to be the highest force level inflation boundary necessary to comply with Standard No. 208. It argued that given the positive indications from the field on the effects of depowering, and the continued positive indications in engineering laboratory testing, it would be a serious setback to motor vehicle safety should the agency send Standard No. 208 backwards by mandating a 48 km/h (30 mph) unbelted rigid barrier test.

Toyota stated that it believes a return to 48 km/h (30 mph) unbelted barrier testing would require an increase in air bag power in many models. That company stated that, given the lack of evidence that higher powered air bags are necessary, it strongly believes that reinstating this requirement would serve only to increase risk to at-risk groups, including out-of-position children and small statured adults.

DaimlerChrysler argued that a return to the unbelted 48 km/h (30 mph) barrier test would necessitate an increase in air bag inflator power, all things being equal. That commenter stated that staged inflators can reduce, but not eliminate, the risk to smaller and out-of-position occupants in lower speed deployments. DaimlerChrysler asserted that to assure compliance, it would expect the power level of the staged deployment necessary to meet the requirements of an unbelted 48 km/h (30 mph) impact to be comparable to the pre-depowering level.

IIHS stated that while NHTSA crash tests indicate that some vehicles may meet the unbelted 48 km/h (30 mph) test without adding more energy, it believes the agency must recognize that this may not be possible in all, or even most, cases. That organization stated that when compliance becomes difficult, it will be far too easy for manufacturers to meet the 48 km/h (30 mph) test by increasing air bag inflation energy (or the second stage of the air bag).

NTSB stated that it is concerned that the 48 km/h (30 mph) unbelted barrier test could result in a return to higher energy air bags.

Recognizing the significant disagreement among commenters concerning whether there should be a return to the 48 km/h (30 mph) test, a broad range of commenters supporting a 40 km/h (25 mph) test argued that the solution should be for the agency to adopt a 40 km/h (25 mph) test in the current rulemaking, and defer any future consideration of a 48 km/h (30 mph) test. As part of this process, they recommended that NHTSA expedite a focused examination of frontal crashes with fatalities to determine, for vehicles with depowered air bags and the latest generation of advanced air bags, how people are dying in these crashes. A 48 km/h (30 mph) test would be considered further if scientific evidence indicated that the 40 km/h (25 mph) test resulted in inadequate protection. Supporters of this approach included NTSB, IIHS, AAM, the National Safety Council, the American Trauma Society, and the National Association of Governors' Highway Safety Representatives. AAM stated that it was committing to provide additional resources for a major real-world data gathering program to provide a greater factual basis for future air bag rulemakings.

Commenters supporting 48 km/h (30 mph).

Safety advocacy groups supporting a maximum test speed of 48 km/h (30 mph) argued that it would result in higher life-saving benefits than a 40 km/h (25 mph) speed.

These commenters emphasized that half of all fatalities in frontal crashes occur at delta Vs above 48 km/h (30 mph). Parents argued that a 48 km/h (30 mph) test speed is very typical of potentially fatal crashes since it is in the middle of the crash speeds that cause fatalities. That commenter also argued that air bag systems certified as meeting the injury criteria at the higher speeds proposed in the rule will have greater efficacy in severe frontal collisions than would air bags certified as complying at some lesser speed.

CAS stated that the 5 mph difference between 40 km/h (25 mph) and 48 km/h (30 mph) is substantial. It stated that a 48 km/h (30 mph) barrier crash is 40 percent more severe than a 40 km/h (25 mph) crash. It also stated that NHTSA data show that almost 20 percent of occupant fatalities in frontal crashes occur between 40 km/h (25 mph) and 48 km/h (30 mph) delta V.

Public Citizen stated that real world driving conditions require the return to a 48 km/h (30 mph) test. That organization stated that these conditions include higher speed limits, as well as the prevalence of vastly increased numbers of SUVs and LTVs designed with stiff front ends. Public Citizen stated that the stiffness of these vehicles, as well as other factors including higher mass, transmit increased forces to passenger cars in crashes.

Public Citizen also argued that over the past 30 years, Americans have used the 48 km/h (30 mph) rigid barrier test as the litmus test for a vehicle's crashworthiness. It noted that other motor vehicle safety standards are based on a 48 km/h (30 mph) test. Public Citizen stated that if the 48 km/h (30 mph) test were dropped, the public would view the decision as a step backward.

Public Citizen stated that one indicator of the inadequacy of a 40 km/h (25 mph) test is a statement by GM in the 1980's that it could pass an unbelted 40 km/h (25 mph) test with “friendly interiors” and no air bag at all.

CAS also stated that a 40 km/h (25 mph) unbelted test, even if coupled with a 56 km/h (35 mph) belted test, is but a slight variation of GM's proposal to Secretary Dole in 1984 for a 40 km/h (25 mph) unbelted and 48 km/h (30 mph) belted standard. CAS argued that if a car with friendly interiors could meet a 40 km/h (25 mph) barrier test in 1984 without an air bag, as GM suggested then that it could, then the addition of a cosmetic air bag would enable a vehicle to meet Standard No. 208 today, even with its revised injury criteria.

These commenters also cited the agency's estimates in the PEA that a 40 km/h (25 mph) test speed could result in 214 to 397 fewer lives saved each year.

These safety advocacy groups also argued that there is no justification to reduce the longstanding 48 km/h (30 mph) test speed and that such a reduction would be inconsistent with the TEA 21 requirement to improve protection of occupants of different sizes, belted and unbelted.

CAS argued that reducing the unbelted test speed to 40 km/h (25 mph) would decrease the level of protection for unbelted occupants who are involved in moderate to high speed collisions. According to that commenter, Congress cannot possibly have envisioned a backward step as an improvement to safety when it mandated that the advanced air bag rulemaking take place.

Public Citizen stated that the whole point of upgrading Standard No. 208 is

to ensure that automakers make better air bag restraint systems and that the standard should reflect as much as possible the protection needed in real world crashes.

According to Public Citizen, a 48 km/h (30 mph) unbelted barrier test would force manufacturers to incorporate more advanced technology. Public Citizen argued that without the additional challenge of the 48 km/h (30 mph) unbelted test, the automakers would have little motivation to move forward technologically in the future.

These commenters strongly disagreed with the arguments of the industry and some others that a 48 km/h (30 mph) standard would require overly aggressive air bags or not be possible to meet for both 50th percentile adult male dummies and 5th percentile adult female dummies.

Parents stated that the industry's rationale for a 40 km/h (25 mph) maximum speed is that the traditional 48 km/h (30 mph) speed compels production of air bag systems that are necessarily and unavoidably dangerous for small occupants in lower speed collisions. That organization stated that it strongly disagrees with this position. According to Parents, this position ignores the outstanding safety record of many well designed air bag systems that have complied with the 48 km/h (30 mph) requirement over the years. Parents also stated that this argument does not take into account advanced air bag technologies, the technologies that the advanced air bag rule is supposed to foster.

Parents also argued that the SNPRM rebutted the industry's argument that adoption of a 48 km/h (30 mph) test speed would necessarily require vehicle manufacturers to revert to excessive deployment forces found in many systems prior to sled testing. Parents stated that the agency pointed out that virtually all of the depowered air bag systems it tested still passed the 48 km/h (30 mph) test. That organization also stated that compliance margins were fairly wide and typically as wide as margins used by industry in complying with the 48 km/h (30 mph) test. Parents stated that for systems that don't meet the 48 km/h (30 mph) test, development of advanced technologies would allow these vehicles to also meet the test.

Consumers Union argued that the agency's testing of 13 vehicles with redesigned air bags leads it to conclude that even before the comprehensive redesign in air bag systems contemplated in this rulemaking, a wide variety of vehicles with depowered air bags already can pass the 48 km/h (30 mph) unbelted test. That organization stated that, contrary to the industry argument, air bags in many varieties of vehicles apparently do not need to be repowered or made “overly aggressive” in order to pass the 48 km/h (30 mph) test.

Consumers Union also stated that in NHTSA tests, two of four vehicles tested, the MY 1999 Saturn and MY 1998 Taurus, passed all the injury criteria for the driver and passenger using unbelted 5th percentile adult female and 50th percentile adult male dummies in 48 km/h (30 mph) rigid barrier tests. That organization argued that if these vehicles can pass these tests even before they have been redesigned to meet a revised Standard No. 208, other vehicles can be engineered to do so as well. These tests were also cited by other commenters supporting a 48 km/h (30 mph) standard.

Public Citizen argued that any trade-offs between meeting requirements for the 5th percentile adult female and 50th percentile adult male dummies can be overcome with the right combination of new technologies. Public Citizen cited dual or multi-level inflators, innovative folding patterns and bag shapes, lighter weight fabrics, tethers, pedal extenders, moving modules, deep dish steering wheels, collapsible steering columns, knee bolsters, stitching that keeps bags narrow to protect in low-level inflation and separates to protect occupants in higher impact crashes, top mounted vertically deploying air bags, chambered air bags (in effect, a smaller bag inside a larger one), and occupant position sensors that adjust deployment level or suppress deployment altogether.

Public Citizen also stated that the new test requirements, including static and dynamic tests using infant, child and small adult size dummies, already address the manufacturers' concerns regarding the “excessive” power of air bags in low severity crashes.

Public Citizen expressed concern about the suggestion of some commenters that more data be collected before any decision is made to return to a 48 km/h (30 mph) test. It argued that this was an excuse to delay a safety standard and that there is plenty of real worl

This text is long and has been trimmed here. Open the source document for the complete record.