[Federal Register Volume 66, Number 137 (Tuesday, July 17, 2001)]
[Notices]
[Pages 37253-37260]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 01-17801]


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

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

[Docket No. NHTSA-1999-6583]


Request for Comments and Notice of Public Workshop; NCAP Consumer 
Braking Initiative

AGENCY: National Highway Traffic Safety Administration (NHTSA), 
Department of Transportation.

ACTION: Request for comments; notice of public workshop.

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

SUMMARY: The National Highway Traffic Safety Administration (NHTSA) is 
holding a public workshop and soliciting comments on a draft test 
protocol to expand the New Car Assessment Program (NCAP) to provide 
brake performance information on new light vehicles to consumers. Since 
1979, NHTSA has been providing consumers with useful information on the 
frontal crash performance of motor vehicles through the NCAP. The NCAP 
program has been expanded over the past few years to include side 
impact crash performance and rollover resistance ratings. Focus groups 
have indicated that motor vehicle brake performance is a prime area for 
consumer information. To date, brake testing variability has been 
NHTSA's primary concern in the development of an effective brake system 
rating. Based on new findings from vehicle research, the agency 
believes that testing variability can be sufficiently minimized to make 
a NCAP braking program viable when vehicles equipped with 4-wheel 
antilock braking systems are tested.

DATES: Written comments: Written comments may be submitted to this 
agency and must be received on or before October 15, 2001.
    Public workshop: The public workshop will be held on September 26, 
2001, from 9 a.m. to 4 p.m. Those wishing to participate should contact 
Mr. Jeff Woods by September 24, 2001.

ADDRESSES: Written comments: Comments must refer to the Docket and 
Notice numbers cited at the beginning of this Notice and be submitted 
to: Docket Management, Room PL-401, 400 Seventh Street, SW., 
Washington, DC 20590. The Docket Section is open on weekdays from 10 
a.m. to 5 p.m. Alternatively, you may submit your comments 
electronically by logging onto Docket Management System web site at 
http://dms.dot.gov. Click on ``Help & Information'' or ``Help/Info'' to 
view instructions for filing your comments electronically. Regardless 
of how you submit your comments, you should mention the docket number 
of this document.
    Public workshop: The public workshop will be held at the Nassif 
Building, 400 Seventh St., SW., Washington, DC 20590; room number to be 
provided to participants prior to the meeting.

FOR FURTHER INFORMATION CONTACT: Mr. Jeff Woods, Office of Safety 
Performance Standards, NPS-22, National Highway Traffic Safety 
Administration, 400 Seventh Street SW., Washington, DC 20590. 
Telephone: (202) 366-6206; Fax: (202) 366-4329.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Background
II. Vehicle Research
III. Agency Plan
IV. Draft Test Protocol
V. Implementation
VI. Request for Comment--Questions
VIII. Public Workshop

I. Background

    Since 1979, NHTSA has been providing consumers with valuable safety 
information on frontal crash performance of motor vehicles through the 
New Car Assessment Program (NCAP). NCAP is perhaps one of the most 
recognized motor vehicle consumer information programs in the U.S. and 
has been expanded to provide data on motor vehicle side impact 
performance. Other countries have joined in NHTSA's effort to give the 
public meaningful comparative information about the safety of different 
vehicles. At this time, Australia, Japan, and Europe have NCAP programs 
in place.
    However, no crash avoidance performance information has ever been 
made available from the U.S. NCAP vehicles. As a result, NHTSA has 
explored the possibility of providing crash avoidance consumer 
information through non-destructive testing of NCAP vehicles before 
they are crash tested. The agency believes that providing brake 
performance information to consumers would give consumers important and 
meaningful safety information and help motivate vehicle manufacturers 
to continue to improve the brake performance of light vehicles. Good 
braking performance can be a key factor in crash avoidance.
    Japan initiated its NCAP braking program in 1995 and has been 
providing braking performance information to its consumers since that 
time. The Japanese NCAP braking program provides stopping distances on 
dry and wet road surfaces from a vehicle speed of 100 km/h (62 mph) and 
indicates whether the vehicle remained in the test lane throughout the 
stop. This information is provided to the public together with the NCAP 
crash testing information.
    In August 1996, NHTSA released the results of a 4000-person 
national survey conducted in 1995 under the National Performance 
Review. Among the key findings was that 75.7% of drivers ranked safety 
as very important in affecting their purchase of a new vehicle.

[[Page 37254]]

    In the 1980's, NHTSA considered publishing comparative vehicle 
stopping distance data provided by manufacturers under the 
subsequently-rescinded consumer regulation on that subject. However, 
one of the drawbacks with those data was that many manufacturers, 
including Chrysler, Ford, and General Motors, were simply providing the 
stopping distance required for all of their models under FMVSS No. 105, 
rather than the actual stopping distance. This factor contributed to 
the agency's decision to rescind that consumer information regulation 
in 1985.
    NHTSA's chief technical concerns with developing a brake system 
performance rating have focused primarily on issues of variability. The 
three primary sources of variability are: vehicle-to-vehicle 
variability, test driver variability, and test conditions (test 
surface, etc.). In 1997, the agency initiated a vehicle research 
program to evaluate how best to minimize test driver and test surface 
variability expected from NCAP brake testing. We did not address the 
issue of vehicle-to-vehicle variability since it is a function of the 
vehicle manufacturing process and therefore would not be minimized by 
the test methodology. The two reports from the vehicle research 
conducted in 1998 and 1999 are summarized below and can be accessed 
through the Docket Management System web site at http://dms.dot.gov in 
Docket Nos. NHTSA-1999-6583-1 and NHTSA-1999-6583-2.

II. Vehicle Research

    The agency has conducted light vehicle brake testing in a variety 
of research programs, including the Light Vehicle ABS Research Program 
that is evaluating the effectiveness of ABS in reducing crashes. We 
believe that of the brake system performance measures evaluated during 
testing, the easiest for consumers to understand and use is probably 
stopping distance. Other measures of brake performance evaluated during 
research, such as brake efficiency, ABS efficiency and brake pedal 
gain, showed higher levels of variability, and are less intuitive 
concepts to communicate to consumers.
    Based on the agency's findings from prior light vehicle brake 
research, we have tentatively concluded that (a) stopping distance is 
the best measure of brake performance for consumer use; (b) variability 
exists between vehicles of the same model; (c) ABS generally improves 
stopping distance performance; (d) and low coefficient of friction 
surfaces, such as wet jennite, produce the most variability and would 
not be useful for consumer information.

Aberdeen Test Center

    The agency initiated additional testing at Aberdeen Test Center 
(Aberdeen) in 1998 to evaluate a simplified test protocol and the 
magnitude of driver and surface variability. The ten ABS-equipped 
vehicles selected for testing included 5 passenger cars, 2 minivans, 1 
full-size van, 1 Sport Utility Vehicle (SUV) and 1 pickup truck with 
rear-wheel-only ABS. One of the passenger cars was used as a control 
vehicle, and was tested throughout the duration of the testing period. 
Ten straight line stops were conducted on each test surface condition, 
including dry and wet asphalt, from a vehicle speed of 100 km/h (62 
mph), with the vehicle in the loaded and unloaded conditions. The 
agency used ten stops to ensure that any variability in brake 
performance from stop to stop could be well identified.
    The results of the stopping distance tests showed that the five 
passenger cars were the best performers with an average stopping 
distance of 46.3 m (152 ft) on the dry asphalt and 51.2 m (168 ft) on 
the wet asphalt road surface. The three vans were mid-performers with 
dry road stops averaging 50.3 m (165 ft) and wet road stops averaging 
52.7 m (173 ft). The average stopping distance of the SUV and the 
pickup truck was 56.4 m (185 ft) on the dry asphalt surface and 62.2 m 
(204 ft) on the wet asphalt surface, although the pickup truck had 
longer stops and more variability since it was equipped with rear-
wheel-only ABS. The test results were also analyzed to provide a 
standard deviation and a 95th percentile stopping distance value for 
the ten stops. The 95th percentile stopping distance provides a measure 
of brake performance based on the average stopping distance and the 
variability of the data set, and represents the distance within which 
the vehicle would stop 95 percent of the time. Vehicles with high 
variability will have 95th percentile stopping distances significantly 
longer than the reported average.
    A comparison of the standard deviation for the ten stops for each 
test vehicle shows the low variability that was achieved by each 
vehicle grouping. The standard deviation, which is a measure of the 
variability of the data set, indicates a low variability for the stops 
conducted on the passenger cars and vans, and a somewhat higher 
variability for the sport utility vehicle on the dry road surface. The 
pickup truck had a higher degree of variability as well. The standard 
deviation for the passenger cars tested on the dry surface in a lightly 
loaded condition had a range of 0.43-0.98 m (1.4-3.2 ft), and on the 
wet surface in a lightly loaded condition ranged from 0.55-1.88 m (1.8-
6.0 ft). Similarly, for the vans tested, the standard deviation for the 
dry, lightly loaded condition, and for the wet, lightly loaded 
condition ranged from 0.40-0.95 m (1.3-3.1 ft) and 0.27-1.01 m (0.9-3.3 
ft), respectively. The SUV had a standard deviation of 2.47 m (8.1 ft) 
on the dry test surface and 0.82 m (2.7 ft) on the wet test surface, 
both in the lightly loaded condition. The pickup truck, which was 
equipped with a rear-wheel-only ABS had a larger standard deviation 
mainly because of the driver modulation that was required to prevent 
front wheel lockup and achieve the best stop.
    The brake pedal application force for the stopping distance tests 
was targeted at 500 Newtons (112 pounds). However, even though the peak 
pedal forces were up to three times higher than target forces, this did 
not affect the stopping distance results. Since the vehicles were all 
ABS-equipped, once the ABS activated the stopping distance performance 
seemed impervious to brake pedal force, except for the pickup truck, 
which required test driver brake pedal modulation to prevent front 
wheel lockup. An analysis of the test data showed that even though the 
test drivers were able to achieve pedal forces as high as 1730 N (390 
lbs) in some of the test runs, such high pedal forces did not improve 
the stopping distance performance of the vehicle. For example, on the 
Pontiac Grand Am, which was used as the control vehicle, the shortest 
stop (42.4 m [139 ft]) was achieved with 1050 N (237 lbs) of pedal 
force, whereas the longest stop (45.7 m [150 ft]) was achieved with a 
higher pedal force of 1370 N (309 lbs). The parameter that seemed the 
most relevant to consistent and shorter stops was the brake application 
rate. The results show that consistency could be achieved using a brake 
application rate of greater than 445 N (100 lbs) of pedal force in 0.2 
seconds or less.
    The test surfaces used for this testing were dry asphalt and wet 
asphalt. These are typical of the road surfaces that most drivers 
experience and would therefore provide useful information for 
consumers. The peak friction coefficient (PFC) measurement for the dry 
asphalt test surface ranged from 0.89 to 0.95 during the testing 
period, and for the wet asphalt surface 0.85 to 0.88. The ambient 
temperatures during the testing period ranged from 7  deg.C to 22 
deg.C (45  deg.F to 71  deg.F). Although these moderate temperatures 
did not show any

[[Page 37255]]

correlation with the stopping distance performance of the vehicles 
tested, the agency believes that testing in moderate ambient 
temperatures in the Fall and Spring might yield more consistent results 
for testing conducted in northerly parts of the U.S.
    The conclusions we tentatively reached from the Aberdeen research 
were that driver and surface variability can be minimized to make the 
NCAP brake performance program a viable one. Driver variability could 
be minimized by testing only ABS-equipped vehicles, by using straight 
line stops, and by specifying a minimum application rate for the brake 
pedal force. Surface variability could be minimized by specifying high 
coefficient of friction dry and wet test surfaces, and by specifying an 
ambient temperature or surface temperature range for testing.

Round-Robin Testing

    The agency initiated a round-robin test in September 1999 to 
further evaluate the effects of surface variability on braking 
performance. The objective was to determine the impact that surface 
variability has on stopping distance performance by analyzing and 
comparing the stopping distance performance of the same vehicles tested 
at different facilities using the same test protocol. The agency also 
wanted to determine if different test drivers could obtain similar 
results. Four vehicles (a passenger car, a SUV, a minivan, and a pickup 
truck) were tested at three different test sites and again at the first 
test site. The PFC of the test surface was measured at each test 
facility during the vehicle testing.
    As was the case for the earlier testing (with the exception of the 
pickup truck) all of the vehicles were equipped with four-wheel 
antilock braking systems. By using only ABS-equipped vehicles, the 
driver is able to make a rapid, hard brake pedal application resulting 
in ABS activation and control of the brake forces at the wheels to 
prevent wheel lockup and optimize stopping distance performance.
    Four rounds of testing were conducted at three test facilities--at 
Aberdeen, MGA Research in Madison, Wisconsin, and the Transportation 
Research Center (TRC) in East Liberty, Ohio. The first and fourth 
rounds of testing were both conducted at Aberdeen. Pavement friction 
was measured at each facility using a skid trailer, and meteorological 
measurements including air and road surface temperatures and wind speed 
were monitored during the testing. Test surface slope and grade 
measurements were recorded.
    PFC measurements taken at Aberdeen indicated that during the first 
round of testing, the dry PFC was 0.94 and the wet PFC was 0.93. PFC 
measurements during the fourth round of testing at Aberdeen were higher 
for the dry pavement, at 0.95 and 1.00 for pre- and post-test 
measurements, respectively. PFC measurements for the wet surface at 
Aberdeen for the fourth round were 0.91 and 0.90 for pre- and post-test 
measurements, respectively.
    The PFC measurements from TRC for the dry surface were 0.91 and 
0.94 for the pre- and post-test measurements, respectively, and for the 
wet surface were 0.84 and 0.83 for pre- and post-test, respectively. 
The difference in PFC between Aberdeen (higher PFC) and TRC (lower PFC) 
resulted in stopping distances of 6 to 15 feet longer at TRC than the 
fourth round Aberdeen stopping distances. The PFC measurements at MGA 
were 0.99 and 0.95 for pre- and post-test dry pavement, respectively, 
and 0.97 and 0.96 for pre- and post-test wet pavement, respectively. 
The MGA test surface had several pavement repair strips, which affected 
the vehicle stopping results with larger standard deviations for each 
series of stops on each vehicle, compared to the results at Aberdeen 
and TRC.
    The results of this Round Robin testing indicate that specifying 
the test surface in terms of PFC will be of primary importance for the 
NCAP braking program since the PFC value does affect the stopping 
distance results. The results also indicate that conducting the brake 
testing for all NCAP vehicles at the same test facility would reduce 
the surface variability and result in more consistent stopping 
distances for all tested vehicles.

III. Agency Plan

1. ABS-Equipped Vehicles

    The test vehicles used during the research program were all ABS-
equipped so as to minimize the effects of driver variability due to 
driver skill. A vehicle's ABS senses impending wheel lockup and 
automatically modulates the brake to provide the shortest stop for the 
given road surface condition. This automatic modulation performed by 
the ABS maintains the braking force close to the level just short of 
wheel lockup. For the Phase I testing conducted at Aberdeen, the 
control vehicle tested using different brake application rates showed 
very little change in its stopping distance performance even though the 
brake pedal application force ranged from 472 N to 1721 N (106 lbs to 
387 lbs). In essence, once the ABS activates, increasing the brake 
pedal force has no impact on the stopping distance performance.
    The agency has no immediate plans to conduct brake testing on 
vehicles not equipped with 4-wheel ABS for the NCAP program. The 
concern associated with testing vehicles with rear-wheel-only systems 
or without ABS is that it would increase the influence of driver 
variability since a driver would be required to modulate the brakes 
manually to achieve the no-wheel-lock requirement. Driver brake pedal 
modulation introduces more variability from stop to stop and results in 
larger deviations between test runs. The agency notes that in recent 
NCAP braking tests conducted in Japan, all vehicles were equipped with 
ABS although the information provided does not indicate if these were 
all vehicles with 4-wheel ABS.

2. Transmission Selector Control

    The agency's draft test protocol includes testing each vehicle with 
its transmission selector control in gear. Federal motor vehicle safety 
standards require most stopping distance tests to be conducted with the 
transmission selector control in neutral so that the stopping distance 
performance of the vehicle would not be affected by engine braking. We 
believe that stopping distance data with vehicles tested in gear would 
produce more relevant consumer information since this condition is more 
representative of what a driver encounters during an emergency braking 
situation. Even though engine braking may help to shorten vehicle 
stopping distance, we believe that its relevance for consumer use 
outweighs any small adverse impact on establishing a valid comparison 
of the performance of service brake systems on light vehicles.

3. Brake Application Rate

    The test data indicate that the rate of brake pedal application is 
more important for consistent and short stopping distances, than the 
magnitude of the brake pedal force. An analysis of the results showed 
that data sets including stops where the pedal application rate was at 
least 222 N (50 lbs) in 0.2 seconds generally had a higher variability 
in stopping distance than the same data sets with the stops having an 
application rate of below 445 N (100 lbs) in 0.2 seconds removed. We 
concluded that slow pedal force rates may have delayed the activation 
of the ABS system and consequently,

[[Page 37256]]

increased the stopping distance variability of the data set. These 
results independently correlate closely to the Japan NCAP braking test 
procedure which specifies that the brake pedal force shall reach 500 N 
(112 lbs) in 0.25 seconds. Therefore, in the interest of harmonizing to 
a certain extent with the Japanese program, the agency proposes that 
the brake pedal application force of 500 N (112 lbs) be achieved within 
0.25 seconds. This brake pedal application rate is important for 
minimizing the variability caused by differences in the initial pedal 
force input and will ensure repeatable ABS activation.
    In addition to the initial application rate, we believe that it is 
important to specify a steady state pedal force for the remainder of 
the stop. After the initial ramp up in the brake force to achieve the 
445 N (100 lbs) in 0.25 seconds, test drivers achieved pedal forces as 
high as 1735 N (390 lbs) to ensure that the ABS remained activated for 
the duration of the stop. An analysis of the data showed that such high 
pedal forces are not necessary to ensure ABS activation throughout the 
stop, and that a steady-state pedal force of about 670 N (151 lbs) 
would be appropriate. We also have considered establishing an upper 
limit for the brake pedal force peak after the initial application rate 
is satisfied. However, since that upper limit value could vary based on 
the vehicle and test driver performance, we believe that it would be 
better to establish a time frame within which the steady state pedal 
force condition should be achieved. This would ensure a consistent time 
frame for achieving the steady state braking force applied by the test 
driver, and therefore, enhance the repeatability of the test protocol.
    As mentioned above, the Japanese NCAP brake test procedure 
specifies an initial application rate of 500 Newtons in 0.25 seconds. 
In addition, their procedure specifies a steady state application force 
of 500  30 Newtons, without specifying a time frame within 
which this force should be achieved. The difficulty in using this 
protocol is that our tests indicate that a peak pedal force in the 670 
to 900 N (150 to 202 lbs) range always occurs in order to achieve the 
rapid brake pedal application rate. The agency believes that to achieve 
the rapid application rate without exceeding a 530 N (119 lbs) limit, a 
special brake application device may be required in lieu of using a 
test driver.
    The agency, therefore, contemplates specifying that a steady state 
pedal force of 670  70 N (151  15.7 lbs), and 
that this pedal force be attained within the initial 0.75 seconds of 
the brake pedal application.

4. Test Surface Variability

    The coefficient of friction of the test surface plays a major role 
in the braking performance of a vehicle. The PFC is currently used as 
the measure of the surface friction in the agency's light vehicle brake 
standard and has a nominal value of 0.90 for dry pavement. Vehicle 
compliance testing on dry pavement by the agency is conducted on a 
surface 0.90 or higher. Dry and wet asphalt surfaces were used for the 
NCAP brake testing because they represent the type of road surfaces on 
which consumers typically drive. The PFC measurements recorded during 
the Phase I testing at Aberdeen ranged for 0.89 to 0.95 for the dry 
asphalt surface, and 0.85 to 0.88 for the wet asphalt surface. For the 
Phase II round robin testing, the PFC recorded at the three test sites 
ranged from 0.91 to 1.00 for the dry surfaces and 0.83 to 0.97 for the 
wet test surfaces. The stopping distance results for the test vehicles 
in both Phase I and Phase II show no correlation between small changes 
in PFC and corresponding changes in vehicle stopping distance. 
Therefore, we believe that for the NCAP brake program, a test surface 
friction range should be specified to accommodate small daily variances 
in PFC.
    Based on the agency's experience with PFC values and given the PFC 
values obtained during the NCAP brake testing, we contemplate that the 
PFC specification for the dry surface would be 0.90 to 0.95, and for 
the wet surface 0.80 to 0.85.

5. Surface Temperature

    The agency believes that ambient and test surface temperatures have 
an impact on vehicle stopping distance performance. However, an 
analysis of the temperature effects was not possible since the 
temperature changes were not sufficiently large to draw any conclusions 
or establish any correlation between ambient and/or surface temperature 
and stopping distance. The vehicle testing in both Phase I and Phase II 
was conducted in the Fall with the ambient temperatures ranging from 
2 deg.C to 24 deg.C (35 deg.F to 76 deg.F) and within the ambient 
temperature range specified in FMVSS No. 135 (0 deg.C and 40 deg.C).
    We believe that conducting vehicle testing in moderate ambient 
temperatures, as those experienced in the northern continental U.S. 
during the Fall or Spring, would provide more repeatable stopping 
distance results, compared with testing at ambient temperature extremes 
during the Winter or Summer. The Japanese NCAP brake test procedure 
specifies a surface temperature range and makes no reference to ambient 
temperature. The surface temperature they specify include 25  deg.C-45 
deg.C (77  deg.F-113  deg.F) for the dry surface and 22  deg.C-32 
deg.C (72  deg.F-90  deg.F) for the wet surface. In the interest of 
developing a test procedure that is similar to the Japanese procedure, 
we are contemplating specifying a surface temperature range, instead of 
an ambient temperature range. We believe that variances in the surface 
temperature would have a more direct impact on the PFC and stopping 
distance performance of tested vehicles, and by specifying surface 
temperatures for NCAP brake testing the surface variability would be 
minimized.

6. Number of Stops

    The stopping distance performance requirements specified in the 
agency's brake standards generally require the best of six stops for 
specific test conditions, which considers that the test driver needs 
several attempts in order to achieve his best stop. However, the agency 
believes that ten stops would allow for a better determination of the 
stopping distance value we convey to consumers. Since the NCAP braking 
program is for consumer information, as opposed to for vehicle 
compliance, and since it is necessary to convey a stopping distance 
value that the consumer is likely to achieve in an emergency braking 
situation, the agency believes that ten stops would be more appropriate 
than six stops for the NCAP braking test procedure. Furthermore, even 
though ABS-equipped vehicles reduce driver variability compared with 
non-ABS vehicles, there still exists some small variability in the 
performance of ABS that could be minimized by requiring more stops.

7. Presentation of Data

    The goal of the NCAP braking program is to provide accurate, 
unbiased brake performance information that is useful and informative 
to the consumer. Two measures of stopping distance performance that 
could be used to inform the consumer of a vehicle's braking performance 
are average stopping distance and the 95th percentile stopping 
distance. The average stopping distance represents a mean of the 
vehicle's brake performance over the ten stops performed during the 
testing, with all stops included in the calculated average. The 95th 
percentile stopping distance provides a measure of brake performance 
based on the average stopping distance and the variability of the data 
set, and informs the consumer of the distance within which the vehicle

[[Page 37257]]

should stop 95 percent of the time. For the Aberdeen testing, the 95th 
percentile stopping distance is equal to: (10-stop average) + (1.645 
x  Standard Deviation). Vehicles with high stop-to-stop variability 
will have 95th percentile stopping distances significantly higher than 
the reported average, while those with small deviations between 
individual stopping distances will have values closer to the reported 
average.
    The agency believes that presenting the information in the form of 
the 95th percentile stopping distance would be more beneficial to 
consumers since this stopping distance value is based on the average 
stopping distance and the variability experienced in the ten stops. We 
believe that providing the average stopping distance value would not, 
by itself, indicate the variability from stop to stop; hence a 
comparison of two vehicles with similar averages but with different 
stop-to-stop variability could be misleading to the consumer in 
conveying the performance that he is most likely to achieve for that 
vehicle. Given that a consumer has one opportunity to obtain a best 
stop in an emergency braking situation, the 95th percentile stopping 
distance represents the stopping distance that he/she is likely to 
achieve in such a situation, provided that the brake pedal application 
rate and force, the road surface friction and load conditions are 
similar to those used during the NCAP braking test. The agency, 
therefore, contemplates that the 95th percentile stopping distance 
value would be presented as the brake performance measure for the NCAP 
brake program.
    The test lane width is specified at 3.5 meters (11.5 ft) for the 
NCAP brake testing and is the same as specified in Standard No. 135. 
NHTSA believes that vehicle stability and stopping distance are both 
important for achieving good braking performance, and that we should 
indicate, along with the stopping distance data, whether the vehicle 
stayed within the lane throughout the stop. Japan currently indicates 
in their data whether the vehicle deviated from the test lane during 
the stopping distance test, and a review of results reported in their 
consumer information shows that none of the vehicles tested that were 
equipped with 4-wheel ABS deviated from the lane during the braking 
test.

IV. Proposed Test Protocol

    The test conditions that the agency included in the draft test 
protocol are based on the conditions specified in FMVSS No. 135, Light 
vehicle brake systems, with a few modifications.

Definitions

    Gross vehicle weight rating or GVWR means the value specified by 
the manufacturer as the loaded weight of a single vehicle.
    Initial brake temperature or IBT means the average temperature of 
the service brakes on the hottest axle of the vehicle 0.32 km (0.2 
miles) before any brake application.
    Lightly loaded vehicle weight or LLVW means unloaded vehicle weight 
plus the weight of a mass of 180 kg (396 pounds), including driver and 
instrumentation. The unloaded vehicle weight includes all fluid 
reservoirs filled to maximum capacity, but without cargo and 
accessories that are ordinarily removed from the vehicle when they are 
not in use.
    Peak friction coefficient or PFC means the ratio of the maximum 
value of braking test wheel longitudinal force to the simultaneous 
vertical force occurring prior to wheel lockup, as the braking torque 
is progressively increased.
    Stopping distance means the distance traveled by a vehicle from the 
point of application of force to the brake control to the point at 
which the vehicle reaches a full stop.

General Conditions

    Pavement friction dry. The road test surface produces a peak 
friction coefficient (PFC) of 0.90-0.95 when measured using an American 
Society for Testing and Materials (ASTM) E1136 standard reference test 
tire, in accordance with ASTM Method E1337-90, at a speed of 64.4 km/h 
(40 mph), without water delivery.
    Pavement friction wet. The road test surface produces a peak 
friction coefficient (PFC) of 0.80-0.85 when measured using an American 
Society for Testing and Materials (ASTM) E1136 standard reference test 
tire, in accordance with ASTM Method E1337-90, at a speed of 64.4 km/h 
(40 mph), with water delivery.
    Pavement temperature dry. The test temperature for the dry pavement 
is 35  deg.C  10  deg.C (95  deg.F  18  deg.F).
    Pavement temperature wet. The test temperature for the wet pavement 
is 27  deg.C  5  deg.C (81  deg.F  9  deg.F).
    Wet pavement condition. For wet surface testing, the test area 
shall be fully wet with standing water not deeper than 3 mm (\1/8\ 
inch). Water shall be applied to the test surface prior to each brake 
stop.
    Gradient. The test surface has no more than a 0.5% gradient in the 
direction of testing and no more than 1.5% gradient perpendicular to 
the direction of testing.
    Lane width. Tests are conducted on a test lane 3.5 m (11.5 ft) 
wide.

Vehicle Conditions

    Vehicle weight. The vehicle shall be tested at lightly loaded 
vehicle weight (LLVW).
    Tire inflation pressure. Tires are inflated to the pressure 
recommended by the vehicle manufacturer for the LLVW of the vehicle.

Instrumentation

    Brake temperature measurement. The brake temperature is measured by 
plug-type thermocouples installed in the approximate center of the 
facing length and width of the most heavily loaded shoe or disc pad, 
one per brake. A second thermocouple may be installed at the beginning 
of the test sequence if the lining wear is expected to reach a point 
causing the first thermocouple to contact the metal rubbing surface of 
a drum or rotor. For center-grooved shoes or pads, thermocouples are 
installed within 3 mm (0.12 in) to 6 mm (0.24 in) of the groove and as 
close to the center as possible.
    Vehicle speed and stopping distance measurement. The vehicle speed 
measurement is performed using a calibrated rolling fifth-wheel 
transducer with quadrature capability. Prior to testing, fifth-wheel 
calibration shall be performed with maximum error not exceeding 0.5 
percent of measured value as verified on a pre-measured 60-m (200-ft) 
test lane.
    Brake pedal effort measurement. The pedal effort measurement is 
performed with a calibrated transducer on the brake pedal. This 
transducer should not interfere with normal brake application.
    Brake pedal force indicator. An indication of the pedal force is to 
be located in view of the driver.
    Ambient temperature. The ambient temperature shall be measured 
continuously during stopping distance testing, using a calibrated 
thermometer.
    Anemometer. The wind speed and wind direction shall be measured 
continuously during stopping distance testing, using a calibrated 
anemometer located at the test site.
    Surface temperature. The road surface is measured at the test lane 
with a calibrated hand-held pyrometer, prior to each test run.

Procedural Conditions

    Brake control. All vehicle brake stops must be met solely by use of 
the service brake control.
    Test speed. The vehicle is tested at a speed of 100 km/h (62.1 
mph).
    Stopping distance. The braking performance of a vehicle is 
determined

[[Page 37258]]

by measuring the stopping distance from a given initial speed. The stop 
is initiated when the stop lamp circuit is closed.
    Vehicle position and attitude. (a) The vehicle is aligned in the 
center of the lane at the start of each brake application. Steering 
corrections are permitted during each stop.
    (b) Stops are made without any part of the vehicle leaving the lane 
and without rotation of the vehicle about its vertical axis of more 
than  15 deg. from the center line of the test lane at any 
time during any stop.
    Transmission selector control. All vehicle brake stops are made 
with the transmission selector in a control position recommended by the 
manufacturer for driving on a level surface at the applicable test 
speed. In initiating each test run, (a) Exceed the test speed by 6 to 
12 km/h (3.7 to 7.5 mph); (b) close the throttle and coast in gear; and 
(c) when the test speed is reached, apply the brakes; (d) to avoid 
engine stall, a manual transmission may be shifted to neutral (or the 
clutch disengaged) when the vehicle speed is below 30 km/h (18.6 mph).
    Initial brake temperature (IBT). If the lower limit of the IBT for 
the first stop in the test has not been reached, the brakes are heated 
to the IBT by making one or more brake applications from a speed of 50 
km/h (31.1 mph), at a deceleration not greater than 3 m/s2 
(9.8 fps2).

Required Test Data

    Test data to be collected includes:
     Vehicle speed.
     Stopping distance.
     Brake pedal application force.
     Brake lining temperatures.
     Road Surface temperature.
     Ambient temperature.
     Tire pressure.

Road Test Procedures

    1. Burnish
    Vehicle conditions.
     Vehicle load is at GVWR.
     Transmission position. In gear.
    Test conditions.
     IBT: 65  deg.C to 100  deg.C (149  deg.F to 212  deg.F).
     Test speed: 80 km/h (49.7 mph).
     Pedal force: Adjust as necessary to maintain specified 
constant deceleration.
     Deceleration: Maintain a constant deceleration of 3.0 m/
s2 (9.8 fps2).
     Number of runs: 200 stops.
     Interval between runs: The interval from the start of one 
service brake application to the start of the next is either the time 
necessary to reduce the IBT to 100  deg.C (212  deg.F) or less, or the 
distance of 2 km (1.24 miles), whichever occurs first.
     Accelerate to 80 km/h (49.7 mph) after each stop and 
maintain that speed until making the next stop.
     After burnishing, adjust the brakes according to the 
manufacturers' recommendation.
2. Stopping distance test
    Vehicle conditions.
     Vehicle load is at LLVW.
     Transmission position. In gear.
    Environmental conditions.
     Wind speed not greater than 5 m/s.
    Test conditions.
     IBT: 65  deg.C to 100  deg.C (149  deg.F to 212  deg.F)
     Test speed: 100 km/h (62.1 mph).
     Pedal force: The brake pedal is to be applied so that the 
pedal force is at least 500 N (112 lbs.) in 0.25 seconds or less, and a 
steady state application force 670 N  70 N (151 
 15.7 lbs.) achieved within 0.75 seconds. The steady state 
application force is to be held constant until the vehicle comes to a 
complete stop.
     Number of runs: 10 stops.
     Test surface--dry: PFC of 0.90 to 0.95.
     Surface temperature--dry: 35  deg.C  10 
deg.C.
     Test surface--wet: PFC of 0.80 to 0.85.
     Water depth of 3 mm (\1/8\ inch) or less.
     Surface temperature--wet: 27  deg.C  5  deg.C
     For each stop, bring the vehicle to test speed and then 
stop the vehicle using the pedal force application method described in 
Pedal force section above.
3. Water Application Procedure
    For wet surface testing, water shall be applied using a water 
tanker truck that is equipped to distribute water evenly across the 
width of the test lane. Prior to wet surface testing, three passes 
shall be made with the water tanker truck traveling longitudinally 
along the test lane. The total length of the wetted area shall be at 
least 100 m (330 ft). Prior to each brake stop event, an additional 
pass shall be made with the water tanker truck along the test lane 
where the brake stops are to be performed. Water shall be distributed 
to fully wet the asphalt surface while keeping the water depth in any 
area of the test lane below 3 mm (\1/8\ inch).
4. Stopping Distance Normalization
    All stopping distance measurements shall be normalized in 
accordance with SAE J299 SEP93, Stopping Distance Test Procedure. 
Stopping distance corrections for initial speed errors greater than 
3.2 km/h (2 mph) are invalid due to inaccuracy.
[GRAPHIC] [TIFF OMITTED] TN17JY01.004

where:

Vd = desired initial vehicle stopping speed, km/h (mph)
Va = actual initial vehicle stopping speed, km/h (mph)
Sm = measured stopping distance, m (ft)
Sc = calculated stopping distance from Vd, m (ft)

V. Implementation

    The agency hopes to gather data to support the NCAP brake testing 
program beginning with model year 2001 vehicles. The data obtained from 
testing these MY 2001 vehicles would not be published as consumer 
information, but would be used to make any refinements to the test 
procedure and/or data presentation. Assuming no major issues are 
identified in the comments on this Notice or the data gathered from the 
2001 vehicles, the agency hopes to fully implement the NCAP brake 
program, in MY 2002, by testing and releasing the stopping distance 
information for the vehicles tested.

VI. Request for Comments--Questions

    The agency seeks comments about topics relating to the NCAP braking 
program and the draft test procedure that has been developed from 
vehicle research. For ease of reference, the questions posed are 
numbered consecutively. The agency requests that commenters identify 
each answer they give by the number of each question being answered.
    1. Based on the agency's existing brake performance requirements in 
the Federal motor vehicle safety standards and its experience with 
brake testing of light vehicles, stopping distance appears to be one of 
the best measures of brake performance. For the NCAP braking program, 
our desire is to provide consumers with a measure of brake performance 
that would be useful for comparing the capabilities of new vehicles. 
The agency requests comments about stopping distance and other measures 
of braking performance. Are other measures of brake performance more 
useful for consumer information? If so, please explain why.
    2. The agency seeks to minimize driver variability by testing 4-
wheel ABS-equipped vehicles only. If we were to expand the program to 
include non-ABS-equipped vehicles, how could we best minimize driver 
variability in testing non-ABS vehicles?

[[Page 37259]]

    3. During vehicle research, the agency found that the brake pedal 
application rate is an important parameter in achieving consistent and 
short stopping distances because it reduces the driver variability for 
the brake application. Based on the agency's independent research, an 
application rate of 445 Newtons (100 lbs.) in 0.2 seconds was derived, 
which is almost identical to the brake application rate of 500 Newtons 
(112 lbs.) in 0.25 seconds specified by the Japanese for their NCAP 
brake testing and that we are now considering to use as well. Are these 
brake pedal application rates achievable for all light vehicles, 
including full-size sport utility vehicles, pickup trucks and vans? Are 
there any concerns about NHTSA using the same brake application rate 
specified by Japan?
    4. After the initial brake application rate is achieved, the agency 
believes that it is important to establish additional criteria for the 
steady state brake application force and the time to attain that force. 
We have specified in the draft test protocol 670  70 
Newtons (151  15.7 lbs.) in 0.75 seconds as the steady 
state force. How appropriate is this force and the specified time frame 
for achieving consistent stopping distance performance? Should a peak 
value be established in addition to the steady state force or as an 
alternative?
    5. Straight line stops are specified for the draft NCAP procedure 
so as to minimize braking performance variability due to driver skill. 
We have also considered braking-in-a-curve, lane-change and other 
maneuvers where a steering maneuver is combined with braking, and 
concluded that straight line stops might be the most useful for 
consumer information. What are your views on the various maneuvers that 
could be used for NCAP braking? Which maneuvers do you consider to be 
best for consumer information, and why?
    6. The agency seeks to minimize surface variability by specifying 
high coefficient of friction dry and wet surfaces. We specify in the 
draft test protocol a dry surface with a PFC of 0.90-0.95 and a wet 
surface with a PFC of 0.80-0.85. Are these PFC ranges appropriate for 
dry and wet asphalt surfaces? Would a smaller range ensure less 
variability in vehicle braking performance? Is such a range realistic 
given the variability in PFC readings from day to day and from week to 
week? What range would you recommend given your experience with test 
surface variability? How often should the PFC for the surface be 
measured during NCAP brake testing?
    7. The agency believes that stopping distance testing in extreme 
ambient temperatures is likely to produce greater performance 
variability than testing in milder ambient temperatures, primarily 
because the surface temperature impact on the PFC of the surface. Japan 
specifies a surface temperature range for its NCAP brake testing, with 
the dry surface temperature between 25 deg.C and 45 deg.C and for the 
wet surface and the wet surface temperature between 22 deg.C and 
32 deg.C. The agency seeks comments on whether such a surface 
temperature range is appropriate for brake testing on a dry surface and 
on a wet surface, and whether the range should be changed? Would 
specifying an ambient similar to the range specified in FMVSS No. 135, 
Light vehicle brake systems, 0 deg.C-40 deg.C, be adequate for NCAP 
brake testing? Would the PFC specification without any temperature 
requirements be adequate? Please support any recommendations for a 
different surface temperature range with data showing its impact on 
vehicle braking performance.
    8. The agency has specifies in the draft test protocol that ten 
(10) stops be made for each test condition. Given that test driver and 
surface variability can be minimized but not eliminated during brake 
testing, we believe that 10 stops would provide a large enough sample 
with which to calculate an average or a 95th percentile value. The 
agency seeks comments on the number of stops that would be considered 
sufficient for providing an average or 95th percentile stopping 
distance value to consumers.
    9. Given that the 95th percentile stopping distance is based on a 
calculated average and the standard deviation among the number of 
stops, would it be considered more appropriate for consumer use? What 
are the pro's and con's of providing the 95th percentile stopping 
distance? Is it important to convey the variability between stops as 
part of the stopping distance information? Why or why not?
    10. The agency contemplates testing the vehicles in the lightly 
loaded condition only since most consumers seek braking performance 
information in that condition. The lightly loaded condition is defined 
as the unloaded vehicle weight plus 180 kg. for driver and 
instrumentation. For the research program, we also tested the vehicles 
at their gross vehicle weight rating (GVWR) and found that, as 
expected, stopping distances were longer than for the lightly loaded 
condition. Do you believe that stopping distance tests should also be 
conducted at GVWR? Why would this information be useful to consumers? 
We note that data shows that the vast majority of consumers operate 
their vehicles lightly loaded.
    11. The agency specifies in the draft test protocol testing 
vehicles with the transmission selector in gear since this is the 
transmission position that most consumers' vehicles are in when faced 
with an emergency braking situation. Testing with the transmission in 
neutral provides a stopping distance performance that does not include 
the effects due to engine braking and is more appropriate for vehicle 
compliance testing. Which method do you believe would provide relevant 
or useful information to consumers? Should the stopping distance value 
exclude the effects of engine braking? Why or why not?
    12. The water depth specified is 3mm or less for the wet pavement 
test. What alternative method can be used to describe a wetted surface 
while ensuring that no puddles or excessive standing water is present? 
What measurement method should be specified for measuring water depth?

VII. Public Workshop

    All interested persons and organizations are invited to attend the 
workshop. To assist interested parties in preparing for the September 
26, 2001 workshop, this agency has developed a preliminary agenda, 
shown below, of introductory presentations and of topics for discussion 
at the meeting. Requests for this agency to consider additional topics 
should be addressed to Mr. Jeff Woods at the address or numbers given 
above.

A. Purpose

    NHTSA is holding a workshop to facilitate an exchange of ideas 
among all participants. The purpose of the workshop is to present and 
discuss the test protocol that has been developed for the NCAP braking 
program. The agency hopes that this workshop will provide opportunities 
for improving and refining the test protocol and other areas of the 
program. We plan to consider the information and the views presented at 
the workshop and in the subsequent written comments in developing the 
final braking test protocol.

B. Procedures

    The agency intends to conduct the workshop informally. The 
Associate Administrator for Safety Performance Standards will preside 
at the workshop. Any person planning to participate in the workshop 
should contact Mr. Jeff Woods at the address and telephone number 
provided at the beginning of this notice, no later than September 24. 
2001.

[[Page 37260]]

C. Agenda

i. Opening remarks
ii. NHTSA Presentation--NCAP braking program
iii. Presentations by organizations and the public
iv. Open discussion

    Issued on: July 12, 2001.
Stephen R. Kratzke,
Associate Administrator for Safety Performance Standards.
[FR Doc. 01-17801 Filed 7-16-01; 8:45 am]
BILLING CODE 4910-59-P