# Final Regulations for Revisions to the Federal Test Procedure for Emissions From Motor Vehicles

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A96-24485

## Record

- **Collection:** Federal Register
- **Document type:** Rule
- **Published:** October 22, 1996
- **Citation:** 61 FR 54852

## Text

SUMMARY: This rulemaking revises the tailpipe emission portions of the
Federal Test Procedure (FTP) for light-duty vehicles (LDVs) and light-
duty trucks (LDTs). The primary new element of the rulemaking is a
Supplemental Federal Test Procedure (SFTP) designed to address
shortcomings with the current FTP in the representation of aggressive
(high speed and/or high acceleration) driving behavior, rapid speed
fluctuations, driving behavior following startup, and use of air
conditioning. An element of the rulemaking that also affects the
preexisting ``conventional'' FTP is a new set of requirements designed
to more accurately reflect real road forces on the test dynamometer.
The Agency is also finalizing new emissions standards for the new
control areas with a specified phase-in period for these standards.
These regulations are expected to reduce emissions from LDVs and LDTs
by two percent for non-methane hydrocarbons (NMHC), 11 percent for
carbon monoxide (CO), and nine percent for oxides of nitrogen
(NOX).

EFFECTIVE DATE: This rule becomes effective on December 23, 1996,
except for Secs. 86.000-7,86.000-8, 86.000-9, 86.001-9, 86.004-9,
86.000-21, 86.001-21, 86.000-23, 86.001-23, 86.000-24, 86.001-24,
86.000-25, 86.001-25, 86.000-26, 86.001-26, 86.000-28, 86.001-28,
86.004-28, 86.108-00, 86.129-00, 86.159-00, 86.160-00, 86.161-00,
86.162-00, 86.162-03, and 86.163-03 which contain information
collection requirements that have not been approved by the Office of
Management and Budget (OMB). EPA will publish a document in the Federal
Register announcing the effective date of those sections. The
incorporation by reference of certain publications listed in the
regulations is approved by the Director of the Federal Register as of
December 23, 1996.

ADDRESSES: Materials relevant to this final rulemaking have been placed
in Docket No. A-92-64. The docket is located at the Air Docket Section,
U.S. Environmental Protection Agency, 401 M Street, SW, Room M-1500,
Waterside Mall, Washington, DC 20460 (phone 202/260-7548; Fax 202/260-
4400), and may be inspected weekdays between 8:00 a.m. and 5:30 p.m. A
reasonable fee may be charged by EPA for copying docket materials.

FOR FURTHER INFORMATION CONTACT: John German, Vehicle Programs and
Compliance Division, U.S. Environmental Protection Agency, National
Vehicle and Fuel Emissions Laboratory, 2565 Plymouth Road, Ann Arbor,
Michigan, 48105. Telephone (313) 668-4214.

SUPPLEMENTARY INFORMATION:

Regulated Entities

Entities potentially regulated by this action are those which
manufacture and sell motor vehicles in the United States. Regulated
categories and entities include:

------------------------------------------------------------------------
Category Examples of regulated entities
------------------------------------------------------------------------
Industry............................ New motor vehicle manufacturers.
------------------------------------------------------------------------

This table is not intended to be exhaustive, but rather provides a
guide for readers regarding entities likely to be regulated by this
action. This table lists the types of entities that EPA is now aware
could potentially be regulated by this action. Other types of entities
not listed in the table could also be regulated. To determine whether
your product is regulated by this action, you should carefully examine
the applicability criteria in Sec. 86.094-1 of title 40 of the Code of
Federal Regulations. If you have questions regarding the applicability
of this action to a particular product, consult the person listed in
the preceding FOR FURTHER INFORMATION CONTACT section.

Electronic Availability

The Preamble, Regulations, Response to Comments, and Regulatory
Impact Analysis (RIA) are available electronically from the EPA
Internet site and via dial-up modem on the Technology Transfer Network
(TTN), which is an electronic bulletin board system (BBS) operated by
EPA's Office of Air Quality Planning and Standards. Both services are
free of charge, except for your existing cost of Internet connectivity
or the cost of the phone call to TTN. Users are able to access and
download files on their first call using a personal computer per the
following information. The official Federal Register version is made
available on the day of publication on the primary Internet sites
listed below. The EPA Office of Mobile Sources also publishes these
notices on the secondary Internet sites listed below and on TTN.
Internet:
World Wide Web:
http://www.epa.gov/docs/fedrgstr/EPA-AIR/
or http://www.epa.gov/OMSWWW/
Gopher:
gopher.epa.gov Follow menus: Rules: EnviroSubset:Air
or gopher.epa.gov Follow menus: Offices:Air:OMS
FTP:
ftp.epa.gov Directory: pub/gopher/fedrgstr/EPA-AIR/
or ftp.epa.gov Directory: pub/gopher/OMS/
TTN BBS:
919-541-5742 (1,200-14,400 bps, no parity, eight data bits, one stop
bit) Off-line: Mondays from 8:00-12:00 Noon ET
Voice helpline: 919-541-5384

A user who has not called TTN previously will first be required to
answer some basic informational questions for registration purposes.
After completing the registration process, proceed through the
following menu choices from the Top Menu to access information on this
rulemaking.

GATEWAY TO TTN TECHNICAL AREAS (Bulletin Boards)
OMS--Mobile Sources Information
Rulemaking & Reporting
Light Duty
File area #1 FTP Review

At this point, the system will list all available files in the
chosen category in reverse chronological order with brief descriptions.
To download a file, select a transfer protocol that is supported by the
terminal software on your own computer, then set your own software to
receive the file using that same protocol.
If unfamiliar with handling compressed (i.e. ZIP'ed) files, go to
the TTN top menu, System Utilities (Command: 1) for information and the
necessary program to download in order to unZIP the files of interest
after downloading to your computer. After getting the files you want
onto your computer, you can quit the TTN BBS with the oodbye
command.
Please note that due to differences between the software used to
develop the document and the software into which the document may be
downloaded, changes in format, page length, etc. may occur.

Table of Contents

I. Introduction
II. Description of the Action
III. Statutory Authority
IV. Public Participation

[[Page 54853]]

A. Legal Requirements
B. SFTP--General
C. Aggressive Driving Cycle (USO6) Requirements
D. Intermediate Soak
E. Air Conditioning
F. Final Standards and Leadtime
G. Technical and Enforcement Issues
H. Regulatory Impact Statement
I. Cost and Benefit Estimates
V. Economic, Environmental, and Cost-Benefit Impacts
A. Environmental Impact
B. Economic Impact
C. Cost-Effectiveness
VI. Administrative Requirements
A. Administrative Designation
B. Unfunded Mandates Act
C. Paperwork Reduction Act
D. Regulatory Flexibility Act
E. Submission to Congress and the General Accounting Office
VII. Judicial Review

I. Introduction

Automobiles are among the largest producers of hydrocarbons (HC),
carbon monoxide (CO), and oxides of nitrogen (NOX), all of which
have documented adverse impacts on public health. This final rule
revises the test procedures used to measure emissions of CO, NOX,
HC, and particulate matter (PM) from MY2000 and later light-duty
vehicles (LDVs) and light-duty trucks (LDTs). It does this by adding
supplemental testing segments to cover driving conditions not
represented in the current procedure, referred to as the ``Federal Test
Procedure'' or ``FTP.''
These supplemental procedures were prompted by section 206(h) of
the Clean Air Act (CAA, or ``The Act''), as amended in 1990, which
reads,

``Within 18 months after the enactment of the Clean Air Act
Amendments of 1990, the Administrator shall review and revise as
necessary the regulations under subsection (a) and (b) of this
section regarding the testing of motor vehicles and motor vehicle
engines to insure that vehicles are tested under circumstances which
reflect the actual current driving conditions under which motor
vehicles are used, including conditions related to fuel,
temperature, acceleration, and altitude.''

EPA's FTP Review project team found that existing information was
clearly inadequate for evaluating the need for revisions to the FTP.
Consequently, a number of new data gathering and analytical efforts
were undertaken. EPA resources were greatly supplemented by cooperative
investments from other sources, including the American Automobile
Manufacturers Association (AAMA), the Association of International
Automobile Manufacturers (AIAM), and the California Air Resources Board
(CARB). These studies provided EPA with unprecedented data on which to
base its comparative review of the FTP.
The Agency published a Notice of Proposed Rulemaking (NPRM) on this
topic on February 7, 1995.1 The preamble to that proposed rule
contains substantial information relevant to the matters discussed
throughout this Notice. The reader is referred to that document for
additional background information and discussion of various issues.
---------------------------------------------------------------------------

\1\ 60 FR 7404
---------------------------------------------------------------------------

In the NPRM, the Agency proposed several additions and revisions to
the tailpipe emission portions of the FTP. The primary new element was
a Supplemental Federal Test Procedure (SFTP) designed to address
shortcomings with the current FTP. The SFTP consisted of three
elements: (1) A new test cycle, US06, designed to address
representation of aggressive (high speed and/or high acceleration)
driving behavior and rapid speed fluctuations, (2) testing of emissions
during actual air conditioning operation, and (3) testing of emissions
after intermediate-duration periods where the engine is turned off.
Another new cycle, SC01, was developed to represent start driving
behavior and rapid speed fluctuations and was proposed to be run after
a 60 minute soak with full air conditioning simulation.
A composite method was proposed to weigh results from each of the
new control areas with bag 1 of the FTP. With this composite approach,
non-methane hydrocarbons (NMHC) and CO SFTP standards were set at the
FTP standard level, while NOX SFTP standards were set 15 percent
above the FTP standard level. The SFTP standards were proposed to be
phased in at 40 percent of a manufacturers fleet for MY1998, 80 percent
for MY1999, and 100 percent for MY2000, with a provision that small
volume manufacturers did not have to comply until MY2000. A new set of
requirements designed to more accurately reflect real road forces on
the test dynamometer was also proposed.
A public hearing was held on April 19 and 20, 1995, in Ann Arbor,
Michigan, at which the Agency took comment on the NPRM. The comment
period initially remained open until May 22, 1995, but was extended to
July 19, 1995 when it became apparent that additional time was needed
to gather and analyze data. Additional comments, data, and analyses
were received after the close of the comment period, which the Agency
has considered in this final rule because the information helped the
Agency develop appropriate test procedures, cost estimates, and
leadtime.
As a result of the comments and significant new data submitted, the
Agency reanalyzed the proposed emission standards when developing the
Final Rule. The proposed US06 standards in the NPRM were largely based
upon available test data on vehicles designed to meet Tier 0 emission
standards. Subsequently, the vehicle manufacturers conducted testing on
32 Tier 1 vehicles over the FTP and US06 cycles and submitted this data
to EPA (this data set is commonly referred to as the ``US06 phase II''
test program). Manufacturers provided the EPA and the docket with this
new data in their comments. The US06 design targets in the Final Rule
are based in part on this new data set, as it is much more
representative of vehicles meeting the ``Tier 1'' emissions standards
than the data available for the NPRM. Similarly, the air conditioning
requirements proposed in the NPRM were based upon vehicles tested with
low mileage catalysts, which are less appropriate for directly setting
useful life emission standards. The vehicle manufacturers conducted
three additional air conditioning test programs subsequent to the NPRM.
The first, commonly referred to as ``ACR2'' (for phase 2 of testing at
General Motor's AC-Rochester environmental chamber), was erroneously
conducted with inappropriate humidity levels. The manufacturers
retested six vehicles from ACR2 in another test program, referred to as
``ACR3,'' which also included testing on two air conditioning
simulations. Finally, four vehicles from ACR3 were retested at
Chrysler's environmental chamber, both for correlation purposes and to
evaluate a third air conditioning simulation. This data is referred to
as ``ACC3.''
These regulations extend emission control comparable to that for
the FTP across in-use driving behavior and conditions that
significantly impact in-use emissions. Additional control is not
required because the main focus of this rule is to update and correct
the test procedure and to control previously unregulated areas to the
level of stringency of the existing requirements. Proper incorporation
of the full range of in-use driving conditions and behavior will allow
EPA to assess feasible increases in stringency when evaluating future
standards.
The next two sections of this preamble provide a description of
this final rule action and the consideration of public comment. The
final sections of the preamble describe the economic and environmental
impact, and cost

[[Page 54854]]

effectiveness, of the rule and address certain administrative
requirements.

II. Description of the Action

Today's action deals primarily with four areas of driving behavior
that are not adequately represented in the current test procedure:
aggressive driving behavior (such as high acceleration rates and high
speeds); rapid speed fluctuations (microtransient driving behavior);
start driving behavior; and actual air conditioner (A/C) operation. The
Agency is finalizing new requirements for these areas. These
requirements shall be included in a supplemental federal test procedure
(SFTP) that will be required in addition to the existing FTP
requirements. Adjustments are included to accommodate certain vehicle
types, transmission types, and performance categories where the
additions are not representative of in-use driving.
These additions to the tailpipe emission portions of the FTP apply
to all LDVs and LDTs certifying with gasoline and LDVs and LDT1s
certifying with diesel motor fuel \2\. These additions do not apply to
vehicles certifying with alternative fuels, although they do apply to
flexible fuel vehicles and dual fuel vehicles. The changes apply to
testing conducted during certification, Selective Enforcement Audits
(SEA), and in-use enforcement (recall). The standards apply for full
useful life under section 202 of the Clean Air Act. The warranty
provisions under section 207 of the Clean Air Act also apply to this
rulemaking. However, EPA is not requiring that the standards
promulgated today be met at high altitude.
---------------------------------------------------------------------------

\2\ Light-duty trucks are divided into two classes based on
weight, each of which is further subdivided into two classes, also
based on weight. Light light-duty trucks (LLDT) are those with a
gross vehicle weight rating (GVWR) up to 6000 lbs. A light-duty
truck 1 (LDT1) falls in this GVWR range and has a loaded vehicle
weight (LVW) of no more than 3750 lbs; a light-duty truck (LDT2)
falls in the same GVWR range but has an LVW greater than 3750 lbs.
Heavy light-duty trucks (HLDT) are those with a GVWR greater than
6000 lbs but not greater than 8500 lbs, which are broken into light-
duty trucks 3 (LDT3), those with an adjusted loaded vehicle weight
(ALVW) up to 5750 lbs, and light-duty trucks 4 (LDT4), which are
those with a ALVW greater than 5750 lbs. See 40 CFR 86.094-2 for
definitions of LDT categories and vehicle weight terms.
---------------------------------------------------------------------------

The requirements of this rule are phased-in, applying to 40 percent
of each manufacturer's separate production (or at the manufacturer's
option, combined production) of LDVs and light LDTs (LDT1s and LDT2s)
for MY2000, 80 percent in MY2001, and 100 percent in MY2002. The
requirements apply to 40 percent of each manufacturer's production of
heavy LDTs (LDT3s and LDT4s) in MY2002, 80 percent in MY2003, and 100
percent in MY2004. Small volume manufacturers would not have to comply
until MY2002 for LDVs and light LDTs, and MY2004 for heavy LDTs. All of
the rule's requirements would apply during this phase-in period. The
Agency recognizes that this phase-in schedule could create an
additional burden for auto manufacturers if the National Low Emission
Vehicle (National LEV) Program goes into effect as proposed with a
MY2001 implementation nationwide (60 FR 53734, October 10, 1995). The
Agency intends to address this issue by proposing language in an
upcoming National LEV rulemaking that, contingent upon a National LEV
program that is ``in effect,'' would harmonize the above phase-in
schedule with the MY2001 nationwide implementation of National LEV. EPA
expects such action would also harmonize with CARB's planned SFTP
requirements for LEVs.
The new SFTP addresses various conditions under which vehicles are
actually driven and used that are not in the FTP. The SFTP includes two
new single-bag emission test driving cycles: (1) the US06, to represent
aggressive and microtransient driving, and (2) the SC03, to represent
driving immediately following vehicle startup and microtransient
driving.
The US06 is run with the vehicle in the hot stabilized condition;
that is, with the vehicle fully warmed up such that the engine and
catalytic converter have reached typical operating temperatures. The
SC03 follows a 10-minute soak and is run with vehicle air conditioning
(A/C) in operation or with proper simulation of air conditioning
operation. The cycles of the SFTP can be run as a sequence to save on
preconditioning and setup time; however, separate runs of the cycles
are permissible with the appropriate soak or preconditioning steps
appended.
High-volume exhaust flow for heavier vehicles run on the US06 will
dictate the use on some vehicles of a larger capacity constant volume
sampler (CVS) than is needed for current FTP testing. The A/C
simulation is not required for this test cycle. Appropriate shift
schedules for manual transmission vehicles are to be determined by the
manufacturer and submitted to EPA for approval.
Hot stabilized condition is achieved by including several
preconditioning options as part of the formal procedure immediately
prior to the US06 Cycle. If the vehicle has undergone a soak of 2 hours
or less, the preconditioning may be a 505 Cycle, the 866 Cycle, the
highway cycle, a US06, or the SC03.\3\ Following longer soaks, the
final preconditioning cycle is an LA4.\4\ For manufacturers who have
concerns about fuel effects on adaptive memory systems, the rule allows
manufacturers and, upon manufacturer request, requires EPA to run the
vehicle over the US06 Cycle on the certification test fuel before
entering the formal test procedure.
---------------------------------------------------------------------------

\3\ 505 refers to the driving cycle that consists of the first
505 seconds (seconds 1 to 505) of the EPA Urban Dynamometer Driving
Schedule, 866 refers to last 866 seconds (seconds 505 to 1372) of
the EPA Urban Dynamometer Driving Schedule. SCO3 refers to the
driving cycle run during air conditioning operation test
requirement.
\4\ LA4 is the name commonly given to the Urban Dynamometer
Driving Schedule.
---------------------------------------------------------------------------

The rule includes adjustments to the US06 test cycle for low-
performance LDVs and LDTs. These adjustments reflect the actual
operation of low performance vehicles in use and are designed to
minimize problems with high engine and catalyst temperatures. The
adjustments are applied dynamically by the dynamometer for any vehicle
after it has been at wide open throttle for 8 seconds (only the lowest
performance vehicles constituting a small portion of the fleet remain
at WOT for 8 seconds over any part of the US06 cycle). Load adjustments
will be made only during the five most aggressive portions of the US06
Cycle. In addition, for US06 Cycle testing of Heavy Light-Duty Trucks
(HLDTs), the truck is to be ballasted to curb weight plus 300 lbs with
the dynamometer inertia weight determined from this same basis, while
FTP testing remains at Adjusted Loaded Vehicle Weight.
The required elements for the SC03 include the preconditioning,
soak period, test cycle, and air conditioning requirements. Prior to
the 10-minute soak period, the vehicle is to be preconditioned to allow
engine and catalyst temperatures to stabilize at typical warmed-up
operating temperatures. The Agency believes that running the vehicle
over EPA's Urban Dynamometer Driving Schedule (LA4) is adequate to
achieve engine and catalyst stabilization regardless of the time period
for which the vehicle was not operational prior to preconditioning.
However, in the event the vehicle was shut off for less than two hours
prior to preconditioning, any of a 505, 866, or SC03 cycle is adequate
for preconditioning the vehicle.
Immediately following the preconditioning cycle, the vehicle's

[[Page 54855]]

engine is turned off for a 10-minute soak period with cooling fans
directed at the vehicle. The vehicle may be removed from the
dynamometer, provided the vehicle is not subjected to unrepresentative
cooling of the engine or catalyst. Following the soak period, the
vehicle will be run over the SC03 cycle using a full environmental
chamber, with vehicle A/C on, for proper representation of start
driving, microtransient driving, and air conditioning operation.
Procedures in a standard test cell that simulate actual air
conditioning effects will be allowed as a option to using full
environmental chambers. The Agency is allowing these conditions as a
cost-effective surrogate for testing in a fully controlled
environmental chamber set to simulate ozone-exceedance conditions of
ambient temperature, humidity, solar load, and pavement temperature.
For MY2000 through MY2002, either the AC1 simulation or the AC2
simulation may be used, as discussed in section IV.E.2.5 Starting
with MY2003, only simulations that can demonstrate correlation with the
use of a full environmental chamber will be allowed. The use of a fully
controlled environmental chamber is permitted at any time.
---------------------------------------------------------------------------

\5\ During the development of these simulations, the AC1 and AC2
methods were informally referred to as the Nissan-II and Toyota
simulations, respectively. The Agency has chosen to apply formal
names to these procedures for regulatory purposes.
---------------------------------------------------------------------------

Manufacturers who choose to use an air conditioning simulation
beginning with MY2003 must submit a description of the simulation
procedure, data supporting the correlation between the simulation and
the full environmental chamber, and any vehicle specific parameters to
EPA in advance. In general, EPA will conditionally approve any
procedure, provided that the procedure can be run by EPA for SEA and
in-use enforcement testing and available data, including past
correlation testing, does not indicate a correlation problem. EPA may
require the manufacturer to demonstrate emission correlation between
the simulation and the full environmental chamber on up to five
vehicles per model year (one for small volume manufacturers). The
vehicles will be selected by EPA and two additional vehicles may be
selected by EPA to demonstrate emission correlation for every vehicle
that fails the correlation criteria.
If a vehicle is selected for correlation demonstration, the
demonstration is accepted if any of the following steps are met:

1: The NOX emissions from the first simulation test are at
least 85 percent of the NOX emissions from the first test in a
full environmental chamber and the fuel consumed is at least 95
percent of the fuel consumed in the full environmental chamber.
These allowances are due to the inherent test to test emission
variability, which is particularly large for NOX emissions (see
section IV.E.2 and the Response to Comments for further discussion).
2: Either the simulation test or the full environmental chamber
test is rerun, at the manufacturers option, and, using the
replacement test, the NOX emissions from the simulation are at
least 85 percent of the NOX emissions from the full
environmental chamber and the fuel consumed is at least 95 percent
of the fuel consumed in the full environmental chamber.
3: Either the simulation test or the full environmental chamber
test, whichever was not rerun in step 2 above, is rerun and the
average of the two simulation tests are at least 85 percent of the
average of the two full environmental tests for NOX and at
least 95 percent of the fuel consumed in the full environmental
chamber.

If a spot check is failed, the Adminstrator will allow up to 60
days for the manufacturer to supply additional data. If that data prove
to the satisfaction of the Administrator that the simulation produces
results that correlate sufficiently with the environmental test
chamber, the Administrator may allow the continued use of the
simulation.
If a correlation is not passed, no further air conditioning testing
will be accepted with the simulation until the manufacturer submits an
engineering evaluation of the cause of the improper simulation and the
extent of the vehicles affected. This evaluation is subject to review
and approval by EPA. For vehicles determined to be represented by an
improper simulation, the manufacturer will be given an opportunity to
demonstrate that the simulation can be corrected. While there are no
direct penalties for failing a correlation demonstration, all future
emission testing on the affected vehicles, including SEA and in-use
enforcement, will be conducted using the corrected simulation or a full
environmental chamber.
The results from each manufacturers correlation demonstrations will
also be tracked over time. The manufacturer is expected to target the
simulation to at least 100 percent of the emissions from the full
environmental chamber. If, over time, the emissions from the
simulations are found to be statistically lower than the full
environmental chamber, further use of simulations by that manufacturer
will not be allowed until the causes of the offset are identified and
corrected.
With the exception of changes prompted by use of new dynamometers
and a change in the wording of driving instructions on following the
speed trace, there are no changes in the final rule to the driving
cycle of the preexisting conventional FTP. Similarly, EPA is retaining
unchanged the method of calculating compliance with the existing FTP.
EPA is finalizing a ``composite'' compliance calculation for
NMHC+NOX that weighs results from the conventional FTP with
results from the SFTP. In the composite SFTP calculation, emissions
from the FTP are weighted at 35 percent, emissions from the SC03 at 37
percent, and US06 emissions at 28 percent. If an engine family or
vehicle configuration is not available with air conditioning, the air
conditioning test is not run and emissions from the FTP are weighted at
72 percent and US06 emissions at 28 percent (note that the air
conditioning test is required for any vehicle available with air
conditioning, even if the installation rate is projected to be less
than 33 percent). For gasoline vehicles, the standards for the SFTP
composite NMHC+NOX emissions are the same as the combined NMHC and
NOX standards applicable under the conventional FTP.
Unlike NMHC+NOX, a composite CO standard was not set based
upon the weighted average of the individual CO standards over the
various cycles. Due to the additional allowance in the US06 CO standard
for commanded enrichment, discussed below, the final rule sets separate
CO standards for the US06 and SC03 testing cycles. A composite CO
standard is allowed, at the manufacturers' option, which is set at the
level of the CO standard applicable under the conventional FTP.
Standards for light-duty diesel vehicles and light-duty diesel
trucks in the LDT1 category are different than those for gasoline-
powered vehicles in those categories. The supplemental FTP for diesel
LDVs and LDT1s does not include the SC03 cycle, because sufficient test
data was not available at this time to create an appropriate air
conditioning standard for these diesel vehicles. In addition, the
NMHC+NOX standard is higher for diesel LDVs and LDT1s because of
the inherently higher NOX emissions associated with diesel
engines. This is similar to EPA's treatment of conventional FTP Tier I
standards for diesel LDVs and LDT1s, which are less stringent for
NOX emissions. Diesel LDVs and LDT1s will have to comply with the
same US06

[[Page 54856]]

standards (or optional composite standards) for CO as gasoline-fueled
LDVs and LDT1s. The composite SFTP NMHC+NOX and CO standards will
be weighted at 72 percent for the conventional FTP cycle and 28 percent
for the US06 cycle. At this time, due to the absence of relevant test
data on which to base a decision, no supplemental standards are being
promulgated for light-duty diesel truck classes LDT2, LDT3 and LDT4,
and no supplemental standards or test procedures are being promulgated
for diesel particulate emissions.

Table 1.--Composite NMHC+NOX Emissions Standards
--------------------------------------------------------------------------------------------------------------------------------------------------------
Intermediate
useful life Full useful life
Type GVWR LVW ALVW standards standards
NMHC+NOX (g/mi) NMHC+NOX (g/mi)
--------------------------------------------------------------------------------------------------------------------------------------------------------
LDV............................................................. All All All 0.65 0.91
LDV-diesel...................................................... All All All 1.48 2.07
LDT1............................................................ 0-6000 0-3750 All 0.65 0.91
LDT1-diesel..................................................... 0-6000 0-3750 All 1.48 2.07
LDT2............................................................ 0-6000 3751-5750 All 1.02 1.37
LDT3............................................................ >6000 All 3751-5750 1.02 1.44
LDT4............................................................ >6000 All >5750 1.49 2.09
--------------------------------------------------------------------------------------------------------------------------------------------------------

Table 2.--CO Emission Standards
--------------------------------------------------------------------------------------------------------------------------------------------------------
Intermediate useful life standards Full useful life standards (g/mi)
(g/mi) -----------------------------------
Type GVWR LVW ALVW ------------------------------------
Composite A/C US06 Composite
A/C US06 (option) (option)
--------------------------------------------------------------------------------------------------------------------------------------------------------
LDV......................................... All All All 3.0 9.0 3.4 3.7 11.1 4.2
LDV-dies.................................... All All All NA 9.0 3.4 NA 11.1 4.2
LDT1........................................ 0-6000 0-3750 All 3.0 9.0 3.4 3.7 11.1 4.2
LDT1-dies................................... 0-6000 0-3750 All NA 9.0 3.4 NA 11.1 4.2
LDT2........................................ 0-6000 3751-5750 All 3.9 11.6 4.4 4.9 14.6 5.5
LDT3........................................ >6000 All 3751-5750 3.9 11.6 4.4 5.6 16.9 6.4
LDT4........................................ >6000 All >5750 4.4 13.2 5.0 6.4 19.3 7.3
--------------------------------------------------------------------------------------------------------------------------------------------------------

The CO standards for the US06 cycle have been set at levels that
allow limited amounts of commanded enrichment, i.e., the air/fuel ratio
is deliberately set richer than necessary for complete combustion of
the fuel. Commanded enrichment is needed to reduce the peak engine and
catalyst temperatures experienced under very high engine loads, which
are generated during certain short periods of high acceleration on the
US06 cycle. If the standards for the US06 cycle did not allow for any
commanded enrichment, there could be a danger of excessive heat that
can cause severe damage to the engine or catalyst. However, commanded
enrichment also causes a sharp increase in the amount of CO emitted
during the enrichment period. The CO increase is directly proportional
to the amount of additional fuel. To ensure that excessive amounts of
enrichments and, hence, excessive CO emissions, do not occur during
commanded enrichment, this Final Rule includes a minimum air/fuel ratio
requirement. The air to fuel ratio shall not be richer at any time than
the leanest air to fuel mixture required to obtain maximum torque at a
given speed and load, termed the lean best torque, plus a tolerance of
6 percent of the lean best torque fuel consumption. Manufacturers may
request additional enrichment, based upon the need to protect the
engine or emissions control hardware.
As indicated above, 35 percent of the new composite SFTP standards
for NMHC+NOX are comprised of the standards from the conventional
FTP. Currently, those conventional FTP standards are the Tier 1
standards promulgated under CAA sections 202 (g) and (h). However, for
vehicles certified under any future National Low Emission Vehicle
(National LEV) Program, the appropriate levels for the conventional FTP
portion of the composite SFTP emissions standards will be the ``on
cycle'' National LEV standards appropriate for such vehicles. As the
composite approach is not mandated for CO, this adjustment would have
no impact on the stand-alone CO standards for US06 and air
conditioning, although a similar adjustment would apply if a
manufacturer opted to use the composite CO standard. The formula for
the new SFTP composite for NMHC+NOX would be:

New SFTP standard = Old SFTP standard--[0.35 * (Tier 1 FTP
standard--New FTP standard)], where all standard references are
based upon NMHC+NOX and the result is rounded to the nearest
two decimal places.

The new US06 cycle requires significantly higher power absorption
capacity, due to the higher power requirements of this aggressive
driving cycle. Dynamometer improvements are needed to properly conduct
this test. The dynamometer improvements also allow better
representation of actual road load forces on all test cycles. Thus,
each test cycle, including the conventional FTP, is to be run on a
system providing accurate replication of real road load forces at the
interface between drive tires and the dynamometer over the full speed
range. While EPA intends to use a 48-inch single-roll dynamometer with
electronic control of power absorption to meet these requirements for
both the new SFTP and current FTP testing, any system will be allowed
that yields equivalent or superior test results. The appropriate
dynamometer load to match actual road load shall be determined for each
vehicle. The EPA shall conduct confirmatory testing using a 48-inch
single-roll dynamometer and manufacturers' test results must correlate
with the EPA test results.
Dynamometers simulate vehicle weight with inertia forces.
Currently,

[[Page 54857]]

this simulation of vehicle weight is capped at 5500 pounds equivalent
test weight (ETW) due to dynamometer limitations. The existing 5500 ETW
cap is removed concurrently with phase-in of the new dynamometer
requirements.
The current 10 percent increase in dynamometer load to simulate the
average nationwide, year-around air conditioning effects during FTP
testing is deleted, as this effect cannot be accurately duplicated on
the improved dynamometer simulation and it did a poor job of estimating
actual average air conditioning loads. The emissions impacts of air
conditioning are being addressed in this Final Rule. Adjustments to the
dynamometer load for fuel economy purposes will be addressed as part of
subsequent rulemaking on test procedure adjustments.
The improved road load simulation and the removal of the 5500 ETW
cap for all test cycles are implemented concurrently with the SFTP
requirements. Thus, any engine family that is included in the SFTP
phase-in must also comply with the improved road load simulation and
the removal of the 5500 ETW cap, although use of the pre-existing
dynamometer requirements is allowed for Part 600 fuel economy testing
for phase-in years 2000 and 2001. In addition, the improved road load
simulation and the removal of the 5500 ETW cap apply to engine families
not covered by the SFTP standard (alternative fuel vehicles and diesel
LDT2s, LDT3s, and LDT4s), effective MY2002 for LDVs and LLDTs and
MY2004 for HLDTs. Manufacturers may elect to use improved road load
simulations on engine families prior to their inclusion in the SFTP
phase-in, at their option.
Regulatory language regarding throttle and pedal movement while the
vehicle is driven on the dynamometer is also revised. The current
requirement to drive with ``minimum'' accelerator pedal movement is
replaced with a requirement to drive the vehicle with appropriate
accelerator pedal movement necessary to achieve the speed versus time
relationship prescribed by the driving schedule. Both smoothing of
speed variations and excessive accelerator pedal perturbations are to
be avoided.
Note that this rule does not address heavy-duty engines or test
requirements with respect to fuel and ambient temperature conditions.
These aspects of the FTP were explicitly excluded from consideration in
this rule, as discussed in the proposed rule and its support documents.
The Agency did not receive any comments on these issues.

III. Statutory Authority

The promulgation of these regulations is authorized by sections
202, 206, 208, and 301 of the Clean Air Act (CAA or the Act) as amended
by the Clean Air Act Amendments of 1990 (42 U.S.C. 7521, 7525, 7542,
and 7601). Section 206(h) of the Act requires EPA to ``review and
revise as necessary * * * the testing of motor vehicles and motor
vehicle engines to insure that vehicles are tested under circumstances
which reflect the actual current driving conditions under which motor
vehicles are used, including conditions relating to fuel, temperature,
acceleration, and altitude.'' Congress mandated that EPA exercise its
authority under section 206(a) of the Act, giving broad authority to
determine appropriate test procedures, consistent with the broad
direction of section 206(h), to determine appropriate changes to
reflect real world conditions.
Although the text of the statute and the legislative history do not
provide explicit criteria or intent for this review, EPA believes the
primary concern of Congress is having test procedures for motor
vehicles and motor vehicle engines reflect in-use conditions in order
to obtain better in-use emission control. This flows from the basic
purpose of test procedures--to measure compliance with the emission
standards--and from standards designed to obtain in-use emission
reductions. Therefore, EPA made this the primary concern and objective.

IV. Public Participation

A number of interested parties commented on EPA's February 7, 1995
NPRM. The comments include written submittals to the rulemaking docket
and those presented at the April 19 and 20, 1995 public hearing held in
Ann Arbor, Michigan. The Agency has fully considered these comments in
developing today's final rule.
The following section presents a brief synopsis of the comments
received on the NPRM and the EPA responses to those comments. A
separate and more detailed Response to Comments has been prepared and
is available in the public docket and electronically (as described in
SUPPLEMENTARY INFORMATION) for review. The interested reader is
referred to that document for a more complete discussion of the
comments and EPA's response, including some of the comments which,
though evaluated in the Response to Comments, are not presented here.
Issues that are discussed only in the Response to Comments include:

--Adjustments for LDTs over 6000 lbs GVWR and for low performance
vehicles
--General Criteria for setting US06 standards
--Determination of LDT2/LDT3/LDT4 and full-useful life standards
--Two-second timer requirement on high performance vehicles
--Equivalent test weight for electric dynamometers
--Road-load determination
--Dynamometer coefficient adjustments for ambient temperature
--Equivalent test weight cap
--Defeat device policy
--US06 shift schedules for manual transmission vehicles

A. Legal Requirements

1. Impact on Stringency of Tier 1 Emission Standard and Consistency
with Section 202(b)(1)(C)
Summary of Proposal. In the Proposal, EPA noted that the proposed
regulations were authorized by sections 202, 206, 208, and 301 of the
Act, including section 206(h), which requires EPA to:

``* * * review and revise as necessary the regulations under
subsection (a) and (b) of this section regarding the testing of
motor vehicles and motor vehicle engines to insure that vehicles are
tested under circumstances which reflect the actual current driving
conditions under which motor vehicles are used, including conditions
relating to fuel, temperature, acceleration, and altitude.''

The Support Document to the Proposal noted that section 206(h) is
silent on the impact that test procedure changes should have on
emission standards, and does not limit or restrict EPA's authority to
establish emission standards. The Support Document also noted that the
proposed emission standards for the supplemental portion of the FTP do
not violate section 202(b)(1)(C)'s prohibition on modification of the
numerical emission standards specified in 202 (g) and (h) (i.e. the
Tier 1 exhaust standards) prior to MY2004, as the standards proposed
were new standards that were in addition to, not alternative to, the
existing Tier 1 standards.
Finally, the Support Document noted that section 202(b)(1)(C)
restricts EPA's ability to relax the Tier 1 numerical emission
standards in order to account for changes in test procedure. EPA has
dual requirements to revise the test procedures used to measure
compliance with Tier 1 and to not revise the Tier 1 numerical standards
prior to MY2004.
Summary of Comments. AAMA/AIAM argued that the EPA's proposal would
effectively increase the stringency of the existing emission standards
and that the 1990 amendments to the CAA do not give the EPA such
authority. It is their

[[Page 54858]]

contention that the authority granted under section 202(a) of the act
is expressly limited by 202 (b) and (g). They also reasoned that the
Agency may propose an SFTP and supplemental standards that may require
recalibration or adjustments, but cannot require such standards or
procedures which require the installation of additional equipment or
substantial alterations to existing vehicles.
AAMA/AIAM claimed that the authority granted in section 206(h) must
be consistent with other provisions in the Act, i.e., EPA may not
increase the stringency of the Tier I standards. AAMA/AIAM averred that
section 206(h) did not provide the Agency with any new authority to
revise the emission standards either directly or indirectly through
revisions to the FTP. They also presented a related argument that
section 206(h) does not provide the agency additional discretion to
revise the Tier I standards. While not specifying how the Agency should
revise the test procedures, the AAMA/AIAM suggested that Congress
expected the Agency to exercise its 206(a) authority, as directed in
206(h) within the limits of 202(a) and 202(b)(1)(C).
Two other commenters, Volvo and Manufacturers of Emission Controls
Association (MECA), also stated that the revised test procedures should
not effectively increase the stringency of the current Tier 1 standards
or future standards.
By contrast, both National Resources Defense Council (NRDC) and
Northeast States for Co-ordinated Air Use Management (NESCAUM) quoted
section 206(h) and interpreted the section as indicating that Congress
was concerned with a large gap between the real world emissions and
emissions measured during the existing test procedure. NRDC and NESCAUM
believe that Congress wanted the EPA to revise the test procedure to be
representative of actual driving conditions. The comments note that
Congress explicitly prohibit EPA from revising the Tier 1 standards
prior to 2004.
The comments stated, in the context of EPA's supplemental
standards, that Congress did not indicate that the EPA was to develop
any new emission standards. Both commenters went on to cite section
202(b)(1)(c) as evidence that Congress ``unequivocally prohibited EPA
from modifying those numerical standards.''
Both NRDC and NESCAUM expressed their dismay that the EPA was
proposing supplemental procedures while leaving essentially unchanged
the current FTP. Both commenters also believed that the emission
standards associated with the supplemental tests were more lenient than
existing standards for the FTP, and thus, the EPA's proposal was
inconsistent with Congressional intent.
Response to Comments. EPA reaffirms that its actions under section
206(h) and 202(a) to strengthen the test procedure and adopt related
standards are not prohibited by section 202(b)(1)(C). EPA disagrees
with the comments of AAMA/AIAM regarding their claims that section
202(b)(1)(C) limits EPA actions under section 206(h). On the contrary,
the requirements of section 206(h) and 202(b)(1)(C) are separate
requirements that create two different duties for EPA. EPA's actions
under section 206(h), strengthening the test procedure, are not
prohibited by section 202(b)(1)(C).
The provisions of section 206(h) and sections 202(g) and (b)(1)(C)
are designed to address two different concerns of Congress. The
legislative history shows that Congress' intent in adding section
206(h) was for EPA to increase the scope of the test to make it more
representative, as well as to increase the overall in-use emissions
control resulting from the test.
Congress added section 202(b)(1)(C) to keep the new Tier 1
``numerical emission standards'' stable. However, Congress specifically
restricted the language of section 202(b)(1)(C) to refer only to
``numerical emission standards.'' Thus, it is clear on the face of the
statute that the language of section 202(b)(1)(C) does not apply to
revisions of the test procedure. Congress could have included language
that prevented EPA from revising its regulations in any way to make the
Tier 1 standards more stringent. Congress also could have limited the
scope of section 206(h) by stating that any actions revising the test
procedure would have to be accompanied by a revision of the numerical
emission standards to account for changes in the stringency of the
standards resulting from such test revisions.
Congress made absolutely clear that EPA was to revise its test
procedure to make it more representative and EPA was not to revise the
numerical Tier 1 exhaust standards prior to MY2004. It is AAMA/AIAM who
wish to avoid the clear intent of Congress by requesting that EPA
either not revise its test procedures as Congress required or that EPA
revise the Tier 1 standards prior to MY2004, which Congress clearly
forbid.
Regarding AAMA/AIAM's claim that section 206(h) is limited to test
revisions that require only ``minimal'' changes to vehicles (``minimal
changes'' could include recalibration of existing emission control
equipment, but could not require installation of additional equipment
or substantial alteration of existing vehicles), absolutely nothing in
section 206 or 202 indicates any such limitation on EPA's authority
under section 206.
Finally, EPA has not failed to recognize that there is an
interconnection between numerical emission standards and the procedures
that test for compliance with such standards. EPA is merely noting that
the prohibitions in section 202(b)(1)(C) are directed specifically
towards the former, not the latter, and that section 206(h)'s mandate
specifically requires that EPA revise the latter to ensure that the
test for compliance with such standards, including the Tier 1
standards, are consistent with the actual conditions under which the
vehicles are used.
Regarding the comments of NRDC and NESCAUM, EPA agrees that
Congress specifically intended that the Tier 1 standards not be revised
prior to 2004. Moreover, EPA agrees that Congress was worried about the
gap between emissions as measured by the FTP and real world emissions
and that Congress intended EPA to revise the test procedure to
eliminate that gap. However, EPA does not agree that Congress intended
to prevent EPA from promulgating supplemental standards in order to
effectuate the requirements of section 206(h). Congress provided no
prohibition on EPA promulgating supplemental standards under section
202(a). In fact, EPA has clear authority to promulgate such standards
and was given broad authority by Congress to revise appropriate
regulations under section 206(h). Moreover, section 202(b)(1)(C) merely
prevents EPA from changing the specific standards of sections 202 (g)
and (h). It does not prevent EPA from promulgating supplementary
standards relevant to procedures that were not in existence and
emissions that were not regulated prior to the promulgation of these
regulations. The standards promulgated today are in addition to, not
instead of, Tier 1 standards. In the long term EPA believes it makes
sense to consolidate all the test requirements into a revised FTP
because replacing the FTP would simplify the test procedure.
Nevertheless, to avoid jeopardizing work on more stringent emission
standards and to avoid delaying implementation of this rule, EPA
believes it is better to incorporate consolidation of the FTP with
future consideration of tighter federal standards.

[[Page 54859]]

2. High altitude
Summary of Proposal. The Agency did not propose to supplement by
further regulation the altitude testing flexibility in current law. EPA
stated that it believed any emission controls required for aggressive
driving would also be effective during high altitude driving. However,
the EPA reaffirmed its authority to perform vehicle testing at any
altitude.
Summary of Comments. AAMA/AIAM, Ford and Suzuki comments were
against high altitude testing on the SFTP. They noted that EPA did not
consider the issue of high altitude compliance in the NPRM and that EPA
had no basis or technical support for requiring an SFTP standard at all
altitudes. They also commented that significant redesign to all
vehicles would be necessary to comply at high altitude. AAMA/AIAM also
argued that the clause in section 206(h) only requires EPA to review
and revise the test procedures ``as necessary'' and does not require
that the new requirements apply at all altitudes. Finally, AAMA/AIAM
commented that the Agency had not complied with section 202(a) (1) and
(2), given the absence of data for high altitude.
Response to Comments. The Agency acknowledges comments that EPA did
not have any data on the SFTP requirements at high altitude. The EPA
reviewed the data submitted by AAMA/AIAM and member companies on
vehicles tested at high altitude. The data clearly show the dramatic
impact high altitude has on wide-open throttle (WOT) time during the
aggressive driving cycle. As discussed in the context of the CO
standard, EPA has concluded that control of WOT emissions should be
limited to 2 to 4 seconds due to the durability impact of elevated
engine and catalyst temperatures. Testing at high altitude would go
well beyond the level of WOT control which EPA feels is appropriate. In
addition, the lower performance levels at high altitude may affect
driving behavior. As the Agency does not have any data on driving
behavior at high-altitude, it is not known whether or not the US06
cycle is representative of high-altitude driving.
For all elements of the SFTP, the emission control attained by
compliance at low altitude would also be achieved at high altitudes.
Given that low-altitude emission control will also be effective at high
altitude and the lack of data on driving behavior and emissions at high
altitude, the EPA will not extend the SFTP requirements to high
altitude testing at this time.
3. Motor Vehicle Information and Cost Savings Act
Summary of Proposal. The EPA did not explicitly discuss fuel
economy impacts in the NPRM.
Summary of Comments. AAMA/AIAM commented that the EPA did not
address the issue of fuel economy decreases in the proposal. The
comments requested that EPA issue fuel economy test procedure
adjustments as soon as possible and to work with NHTSA to assure
similar adjustments for light-duty trucks. AAMA/AIAM argued that the
Motor Vehicle and Information Cost Savings Act required the EPA to give
adjustments for measuring fuel economy whenever it modified the test
procedures for measuring fuel economy.
AAMA/AIAM also commented on the timing of the test procedure
adjustments. Citing the Preamble to the CAFE adjustment rule published
as 50 Fed. Reg. 27183 (1985), they stated that the EPA must make test
procedure adjustments at the same time that it promulgates the final
regulations on the FTP changes. AAMA/AIAM concluded that, to comply
with its legal obligations, the EPA should do the following: delay
finalizing proposed rule until fuel economy test procedure adjustments
are developed, issue a notice of proposed rulemaking on the final test
procedures with sufficient information so the EPA and industry can
carry out a comprehensive test program, and issue final changes to the
test procedures at the same time as the fuel economy test procedure
adjustments.
Response to Comments. EPA agrees that, to the extent changes in the
portion of FTP also used to measure fuel economy have an effect on the
fuel economy test that is run in conjunction with the FTP, then EPA
must issue adjustment factors to ensure comparability with the fuel
economy test procedures used in 1975. EPA will promulgate any
adjustments to the fuel economy calculations through notice and comment
rulemaking. EPA will address the substantive issues raised by AAMA in
that rulemaking.
Regarding the timing of promulgation of the FTP revisions and the
rulemaking for CAFE calculation adjustments, EPA disagrees with AAMA/
AIAM's suggestion that EPA should delay promulgating final regulations
revising the FTP until it makes a final determination regarding CAFE
calculations. EPA was required by Congress to promulgate its FTP
revisions by March 15, 1992. These regulations are well overdue. EPA is
under court order to promulgate these regulations by August 15, 1996.
Therefore, EPA cannot fail to promulgate these regulations by that
date.
Nor does EPA believe that either the Motor Vehicle and Information
Cost Savings Act or EPA's rules require that EPA delay its FTP
revisions until the rulemaking regarding CAFE calculations is complete.
The preamble language in the 1985 rulemaking cited by AAMA/AIAM
expresses EPA's intentions, the actual rules do not require the result
sought by AAMA/AIAM. In any case, this preamble language cannot control
the timing of rulemaking that is mandated by more recent statutory
obligations. Moreover, given the changes that have occurred as a result
of comment on the proposal to revise the FTP, the calculations and
procedures necessary to begin a rulemaking to determine CAFE
adjustments resulting from today's rule could not easily have been
initiated until its final regulations were relatively certain. EPA
does, however, recognize the manufacturers' need for sufficient
leadtime once the Agency makes a final determination of CAFE
calculation adjustments, if any. Thus, for only Part 600 fuel economy
testing for phase-in years 2000 and 2001, the manufacturers may use the
pre-existing dynamometer requirements for their entire fleet.
EPA notes that these final regulations delay implementation of the
FTP revisions until MY2000. EPA also notes that the July 1, 1985
rulemaking cited by AAMA/AIAM instituted retroactive changes to the
CAFE calculations for all manufacturers.

B. SFTP--General

1. Margin for Variability (Headroom)
Summary of Proposal. To account for various sources of vehicle and
test variability, vehicles are designed to meet emissions targets below
the standard. The NPRM proposed a composite standard that would
preserve the FTP cold start/hot stabilized driving mix, such that the
current FTP compliance headroom would be implicitly preserved. The
proposal stated that if data were submitted to help establish
appropriate in-use margins, EPA would reevaluate this compliance
structure.
Summary of Comments. No comments were received that disagreed with
the NPRM proposal to use the same headroom factor for off-cycle
standards as has been used historically for the

[[Page 54860]]

FTP.6 AAMA/AIAM presented substantial amounts of in-use data on
FTP emissions that support an historical headroom factor of two. The
data also indicate that hot, stabilized emissions from bags two and
three of the FTP are more variable than bag one.
---------------------------------------------------------------------------

\6\ ``Compliance Margin/Headroom, Compliance Standards vs. In-
Use Emissions,'' Attachment V to a letter from Gerald A. Esper,
AAMA, and Gregory J. Dana, AIAM, to U.S. EPA, January 30, 1995.
Available in the public docket for review.
---------------------------------------------------------------------------

Mercedes-Benz commented that if the EPA were to promulgate SFTP
standards for diesel vehicles, that they be diesel-only NMHC+NOx
standards with sufficient headroom. They did not elaborate as to what
they considered sufficient headroom.
Response to Comments. Headroom is necessary to account for
variability in emissions due to normal production tolerances, variation
between prototype and production parts, test-to-test variability, and
variability in lab correlation. Not only does historical data indicate
that manufacturers currently use a headroom factor of two for the FTP,
but the new cycles being promulgated are hot, stabilized tests and,
thus, may share the higher variability of the bag two and bag three
emissions from the FTP. Based upon these factors, EPA concurs with
AAMA/AIAM's assessment that a headroom factor of two is appropriate for
the SFTP.
In examining the most recent diesel LDV certification data, it
became apparent that the historical headroom factor of two for gasoline
vehicles did not apply to diesel LDV for NOx. For the diesel
LDV's, the Tier 1 NOx standard is 1.0 g/mi. Certification emission
data indicates that diesel LDV's NOx emissions average 0.82 g/mi
This results in a headroom factor of 1.22. Therefore, a headroom factor
of 1.22 will be used for setting SFTP standards for diesel LDVs and
LDT1s.
2. NMHC+NOx Standards
Summary of Proposal. The NPRM proposed separate NOx and NMHC
standards for the supplemental test requirements. The NPRM stated that
the Agency was also considering the alternative of establishing a
single standard for NMHC+NOx, instead of separate standards, and
invited comment on the cost and emission impacts of this alternative.
Summary of Comments. CARB supported setting a combined
NMHC+NOx standard for high speed/acceleration compliance on US06,
stating that they had committed to proposing the setting of an
NMHC+NOx standard for US06 in response to an October 1994 proposal
by the automotive industry. However, CARB does not believe it would be
appropriate to employ an NMHC+NOx standard for air conditioning
standards. CARB recommended setting separate standards for NMHC, CO,
and NOx emissions for A/C-on operation, because the range of
engine loads encountered with the A/C on is similar to the standard FTP
and the evidence suggests that little or no increment to current NMHC
or CO standards is necessary for A/C-on operation.
AAMA/AIAM recommended the use of NMHC+NOx standards for all of
the supplemental test requirements. All of AAMA/AIAM's standard
analyses were presented in terms of NMHC+NOx. AAMA/AIAM also
stated as a general rule that there are tradeoffs in catalyst
efficiency between NMHC/CO and NOx.
NRDC stated that a combined NMHC+NOX standard would be in
direct contradiction of the Congressionally established standards,
which set separate limits for specific pollutants, and for the same
reasons that EPA can't relax the standards, it can't combine them.
Response to Comments. EPA's analyses of the second-by-second
emission data from the US06 testing program clearly indicate that
catalyst conversion efficiency is very sensitive to air/fuel ratio.
Air/fuel shifts less than 1 percent lean of stoichiometry can cause
dramatic reductions in NOX conversion efficiency. While NMHC
conversion efficiency is not as sensitive to short air/fuel shifts as
NOX conversion efficiency, consistent operation about 1 percent
rich of stoichiometry can cause dramatic reductions in NMHC conversion
efficiency. Thus, there is only a very narrow range of air/fuel ratio
in which the catalyst will convert both NMHC and NOX at the levels
required to meet the individual design targets in this rule for NMHC
and NOX.
Unfortunately, the oxygen sensors which are used as the basis for
air/fuel control are not 100 percent accurate and normal variation
occurs in production. Thus, some production vehicles will run slightly
richer than designed and some slightly leaner due to the normal
variation. This is not a major problem for compliance with the current
FTP emission standards, as about 70 percent of the NMHC emissions over
the entire cycle are generated during the cold start, as well as about
30 percent of the NOX emissions, and cold start emissions are
largely unaffected by minor changes in air/fuel ratio. However, the
variation in air/fuel ratio is a much larger problem for both the US06
and air conditioning requirements in this rule, as they are conducted
in hot, stabilized conditions.
An NMHC+NOX standard minimizes the risk of failing the
supplemental requirements in this rulemaking simply due to production
variation in oxygen sensor output. In addition, the NMHC+NOX
standard should have no negative impact on overall in-use ozone
precursor emissions, as any substantial increase in either NMHC or
NOX must be offset by a decrease in the other to avoid failing the
standards. As there should be no negative emission impact and it allows
the manufacturers increased flexibility in meeting the standards, the
Agency is adopting NMHC+NOX standards in the Final Rule.
Adoption of NMHC+NOX standards is consistent with AAMA/AIAM's
comments about the tradeoffs between NMHC/CO and NOX and their
recommendations to use NMHC+NOX standards. It is also consistent
with CARB's position on US06 standards. It is not consistent with
CARB's position on air conditioning standards. While EPA understands
CARB's reasons for not using NMHC+NOX standards for air
conditioning, EPA believes they are less important than giving
flexibility to account for production variation in air/fuel ratio. In
addition, CARB's position would make any composite of US06 and air
conditioning standards impossible, which is inconsistent with EPA's
position on composite standards (see below).
Regarding the comments of NRDC against a combined NMHC+NOX
standard, NRDC's comments were based upon the same legal basis as their
argument that EPA can't relax the standards by setting emission levels
different from the Tier 1 standards. As discussed in section I.A., EPA
does not agree that Congress intended to prevent EPA from promulgating
supplemental standards in order to effectuate the requirements of
section 206(h). Section 202(b)(1)(C) merely prevents EPA from changing
the specific standards of sections 202 (g) and (h). It does not prevent
EPA from promulgating supplementary standards relevant to procedures
that were not in existence and emissions that were not regulated prior
to the promulgation of these regulations. As the standards promulgated
today are in addition to, not instead of, Tier 1 standards, there is no
prohibition against a combined NMHC+NOX standard.

[[Page 54861]]

C. Aggressive Driving Cycle (US06) Requirements

1. Use of US06 Cycle for Aggressive Driving Standard
Summary of Proposal. The EPA proposed the US06 driving cycle and
corresponding emission standards for the control of emissions resulting
from aggressive driving. The US06 driving cycle was originally
developed with extensive coordination with CARB and the vehicle
manufacturers. The US06 driving cycle is ten minutes in duration and
has a maximum speed of 80.3 mph.
Summary of Comments. NESCAUM and MECA indicated general support for
the US06 cycle to account for the aggressive driving behavior of
today's drivers. NESCAUM did, however, express concern that the data
EPA used may not be representative of regional-scale driving, which
they felt was more heavily influenced by high speed driving and hard,
high-speed acceleration.
AAMA/AIAM and Specialty Equipment Manufacturers Association (SEMA)
raised a number of concerns about the US06 cycle. AAMA/AIAM stated that
the US06 is a very poor compliance cycle for significant NOX
reductions, because EPA designed a cycle concentrating on controlling
enrichment. AAMA/AIAM also stated that the EPA incorrectly claimed US06
represents driving done by all vehicles, claiming that it represents
only the single vehicle that generated the cycle, that most vehicle
classes aren't represented, and that the cycle is clearly not
representative for those vehicles that cannot follow it.
SEMA also commented that the US06 cycle contains non-representative
conditions. Specifically, SEMA noted concern that maximum speed on US06
was 15 mph over the legal speed limit, which only represents infrequent
and illegal activity. They also felt that EPA incorrectly implied that
the fraction of vehicle time spent outside the envelope of the LA4
speed and accelerations (13 percent) was only the higher speed and
accelerations. SEMA also had comments regarding their power statistics
that are addressed in the Response to Comments document.
Response to Comments. EPA is finalizing the US06 driving cycle as
proposed. The agency believes that, as a control cycle, the US06
adequately represents the range of in-use operation and provides for
the necessary emission control of such operation.
In developing the US06, the EPA sought to create a cycle that was
comprised of segments of in-use driving and would control emissions
under driving conditions not represented by the FTP. The US06 cycle is
made up of portions of EPA's inventory cycle (REP05) and the California
Air Resources cycle ARB02, and is representative of driving behavior
outside of the traditional FTP for most vehicles. EPA agrees that the
US06 cycle, unadjusted, is not appropriate for all vehicles classes;
EPA therefore proposed and is finalizing cycle adjustments for certain
cases, as summarized in the Summary of Proposal, above, and discussed
in the Response to Comments.
The Agency disagrees with AAMA/AIAM's comment that a cycle segment
can only represent the vehicle that generated the segment in use. The
underlying cycle generation methodology used by the EPA selected
representative segments of actual in-use driving data from a very large
database to match the distribution of in-use speeds and accelerations.
Thus, the segments were selected as the best representation of the
entire data set.
The EPA also disagrees with AAMA/AIAM's comment that the US06 is a
poor NOx control cycle. The US06 cycle was not designed for
control of enrichment but, rather, to control emissions during high
load and high speed operation. It should also be noted that the
relationship between US06 and REP05 emissions, with and without
enrichment, is more stable for NOx than for either NMHC or CO.
This indicates that US06 does a good job of correlating with the
NOx emission levels on REP05, the high speed/acceleration emission
inventory cycle.
EPA disagrees with SEMA's characterization that EPA included
outliers in the in-use driving behavior database. First, the raw
driving behavior data went through a quality control process to remove
any suspect data before inclusion into the final database. Second, the
Baltimore/Spokane database contains nearly 7 million seconds of driving
behavior data, and thus one-tenth of one percent represents nearly 7000
seconds of real in-use driving behavior. As with any dataset, the data
will be distributed across a range of values. It is not appropriate to
assume that data in the tails of the distribution should be treated as
outliers, especially when working with a dataset as large as the in-use
driving behavior dataset.
The Agency believes that it is appropriate to include speeds above
65 mph, since EPA believes it was Congress' intent for EPA to
characterize actual current driving conditions, without constraining
the characterization to behavior within the legal speed limits.
2. US06 CO Standards and Durability Impact Considerations
Summary of Proposal. The implicit US06 CO standard proposed by EPA
in the NPRM for Tier I LDV and LDT1 vehicles was 3.4 g/mi. Due to the
extremely high CO emissions emitted during commanded enrichment, the
3.4 g/mi CO standard proposed in the NPRM would have completely
eliminated commanded enrichment over the US06 cycle. Comments were
specifically requested on the need to allow some commanded enrichment
events during the US06 cycle to avoid elevated catalyst temperature
levels from in-use operation that would lead to catalyst deterioration.
Summary of Comments. AAMA/AIAM had a number of comments on the
potential impacts of the proposed rules on catalyst durability. They
commented that, first, EPA's proposed standards seek to eliminate all
enrichment without regard for impact on durability. Second, EPA glossed
over the impact of completely eliminating commanded enrichment on
increasing catalyst temperature, since in-use catalyst temperatures can
easily exceed those experienced over the US06 cycle if in-use WOT
events are preceded by higher loads or the WOT events occur at higher
speeds. Third, catalyst deterioration is not on-off; a long period of
time at 850 deg.C can produce the same deterioration as a short period
of time at 900 deg.C. Fourth, the catalyst temperature data used in
the analyses were from Tier 0 vehicles without close-coupled catalysts.
Fifth, if it is true, as EPA stated, that extended WOT in-use driving
situations will be infrequent and not of much consequence on catalyst
temperature, then the same can be said about the need to control
emissions during these situations. CO emissions from WOT events over 2
seconds have an extremely small impact on fleet-average CO emissions
and air quality. Finally, all vehicles should be allowed to use
enrichment after two seconds of WOT. A two second limit will keep
NOx increases down and the increase in catalyst temperature to
manageable limits for Tier I vehicles.
A number of comments from individual manufacturers and from SEMA
echoed AAMA/AIAM's catalyst durability concerns. Honda stated that the
maximum catalyst temperature they could tolerate was 900 deg.C and
that the CO standard would need to be less stringent to protect
catalysts from overheating on US06. SEMA stated that EPA's imposition
of a timer and/or

[[Page 54862]]

elimination of commanded enrichment will further aggravate the tendency
for vehicles, particularly high performance vehicles, to experience
excess catalyst and engine/component temperatures. Both GM and Suzuki
stated that extended stoichiometric control results in excess
temperature in warm-up catalysts.
Ford stated that, if longer WOT times are dictated, then the CO
standard should be raised commensurately to allow commanded enrichment
to cool the catalysts.
MECA did not support concerns about catalyst durability, stating
that catalyst formations exist which are capable of withstanding
temperatures in excess of 900 deg.C.
CARB, in an April 10, 1996 memo 7, stated that they were
revising their position on the control of commanded enrichment and now
supported allowing limited amounts of commanded enrichment. CARB
recommended establishing a US06 CO standard, without a WOT enrichment
delay criterion, based on both stoichiometric non-WOT operation and
four seconds of WOT enrichment delay on lower performance vehicles.
---------------------------------------------------------------------------

\7\ Memorandum from Robert H. Cross, Assistant Chief, Mobile
Source Division, CARB, to Margo Oge, Director, Office of Mobile
Sources, EPA, ``Reference No. TF-96-008'', April 10, 1996. Available
from EPA Air Docket A-92-64.
---------------------------------------------------------------------------

Response to Comments. EPA shares the concerns expressed by most
commenters about impacts of stoichiometric control during WOT on
catalyst deterioration. EPA and CARB spent considerable time evaluating
three approaches to limit the duration of WOT stoichiometric control to
periods that would not be likely to cause catalyst deterioration (i.e.
2-4 seconds, based upon EPA analyses and manufacturer comments):

1. Dynamically adjust the load during the test whenever a
vehicle had stayed at WOT for two seconds, so that the vehicle can
continue to follow the trace without having to stay at WOT.
2. Raise the CO standard and extend the two-second timer
criteria for high-performance vehicles in the NPRM to all vehicles.
3. Raise the CO standard to a level that would allow enrichment
on most vehicles after, at most, two seconds of WOT operation and no
more than four seconds of operation on any vehicle.

Despite the small loss of CO control on higher performance
vehicles, EPA has concluded that Option 3, raising the CO standard
without a two-second design criteria, is the most appropriate choice.
Option 3 avoids the potential NOX increase associated with the
frequent load reductions that would occur during testing for Option 1,
as well as the complexity of having a secondary timer criteria and some
increased potential for catalyst degradation for Option 2. The approach
in Option 3 is also consistent with that recommended by CARB. In
addition, the CO loss associated with WOT operation on high performance
vehicles is small, as about two-thirds of enrichment CO is generated at
part-throttle in use, plus most WOT operation occurs on lower
performance vehicles.
In setting the level of the CO standard for the US06 cycle, EPA's
primary criteria was to select a CO standard that most vehicles could
meet while eliminating enrichment for no more than two seconds at WOT.
However, setting the CO standard at a high enough level to allow low
performance vehicles to meet it while eliminating commanded enrichment
for only two seconds would allow higher performance vehicles to use
enrichment at part throttle. To prevent this and to reflect the much
higher proportion of time low performance vehicles spend at WOT in use,
a secondary criteria was added to allow the CO standard to be set at a
level that would require low performance vehicles to use stoichiometric
control at WOT for up to four seconds.
Based upon these criteria, total CO emissions over the US06 cycle
were calculated from a combination of the production and stoichiometric
calibration data. The data showed that a CO design target of 4.5 g/mi
meets the primary criteria that most vehicles meet the standard with no
more than two seconds of stoichiometric control at WOT and, with the
allowance of dynamic load adjustments for the lowest performance
vehicles, would allow all vehicles to meet the standard with no more
than four seconds of stoichiometric control at WOT.
Using the ``times two'' headroom previously determined to be
appropriate for off-cycle standards, the result is a 50,000 mile US06
CO standard of 9.0 g/mi for LDV and LDT1 vehicles. While this almost
triples the CO standard proposed in the NPRM, the impact on in-use CO
emissions is proportionally far less. This is because the US06 cycle
only represents 28 percent of all in-use operation and, even within
this window, overstates the amount of extended WOT operation compared
to in-use operation. (This overstatement is intentional in order to
insure control over the range of high load acceleration events which
are associated with the extended WOT operation.) 8 Most enrichment
CO emissions are generated during part-throttle and most in-use WOT
throttle operation does not last more than two seconds in duration.
Thus, even at 9.0 g/mi, about 80 percent of CO from commanded
enrichment will be controlled.
---------------------------------------------------------------------------

\8\ A discussion on the development of the US06 can be found in
the ``Final Technical Report on Aggressive Driving Behavior for the
Revised Federal Test Procedure Notice of Proposed Rulemaking,''
available in the public docket.
---------------------------------------------------------------------------

EPA believes that US06 is the preferable method for establishing
control of emissions from non-LA4 driving behavior. The US06 covers the
range of non-LA4 driving, while targeting severe, high emission events.
Because the driving modes generating the highest emissions differed
widely across vehicles, it is very important to include a variety of
high load events representing actual aggressive driving behavior. In
addition, the US06 cycle achieves the objectives of both EPA and CARB,
thus eliminating issues or costs associated with the respective
agencies having two different control. An important CARB objective is
to make sure outer bounds of in-use aggressive driving is represented
and controlled; this is achieved with the inclusion of the ARB02 high-
speed microtrip. A second, ARB02 high-speed microtrip was rejected due
to an extended, high-speed acceleration which might result in excessive
catalyst temperatures in vehicles which are controlling commanded
enrichment. Thus, the US06 provides for control of short-duration
commanded enrichment events associated with aggressive driving. As
discussed in the feasibility section which follows, the duration of
commanded enrichment control needs to be limited due to catalyst
temperature concerns. EPA's analysis of catalyst temperature data from
the manufacturer's test program concluded that the ARB02 high-speed
microtrip used in US06 provides for a reasonable duration of control.
The amount of CO control inherent in the CO standard is illustrated
by the average CO emissions generated on US06 by the Tier 1 vehicles in
the US06 phase II test program. LDV and LDT1 vehicles averaged 17.6 g/
mi with production calibrations. Compared to this baseline level,
raising the CO design target from the implicit level of 1.7 g/mi in the
NPRM to the Final Rule level of 4.5 g/mi reduces the CO benefit on the
US06 cycle from 15.9 g/mi to 13.1 g/mi, a reduction of only 18 percent.
The in-use emission impact will be less yet, as the US06 cycle
overstates the amount of WOT operation. While it may

[[Page 54863]]

seem as if raising the standard from 3.4 to 9.0 g/mi should have a
major impact on the stringency of the standard, given the severity of
the US06 cycle and the extremely high baseline emission levels,
analyses support that a standard of 9.0 g/mi will still achieve the
large majority of the potential CO emission benefits.
The CO standard needs to be at this level because of the extreme
sensitivity of CO emissions to commanded enrichment. Each second of
commanded enrichment generates 2-4 grams of CO, enough to add about
0.3-0.5 g/mi to the overall weighted US06 test results. Thus, raising
the standard from 3.4 to 9.0 g/mi, which raises the design target level
from 1.7 to 4.5 g/mi, is an allowance of only about 6-10 seconds of
enrichment on a cycle which over represents extended WOT operation.
The CO standards on US06 have been deliberately set at this level
to allow limited amounts of commanded enrichment, which is needed to
ensure excessive engine and catalyst temperatures do not occur. As CO
emissions are directly proportional to the amount of extra fuel, this
Final Rule includes a minimum air/fuel ratio requirement to ensure that
excessive amounts of enrichment and, hence, CO emissions, do not occur
during commanded enrichment. The air/fuel ratio shall not be richer
than the lean best torque, plus a tolerance of six percent of the lean
best torque fuel consumption. The six percent tolerance is included to
allow for normal variance in production torque characteristics, as well
as the impact of engine deposits on knock in use.
The CO standards for truck classes and for full-useful life
standards are calculated based upon the ratio of the FTP CO standards.
The full list of the CO standards was presented in the ``Description of
the Action'' section.
3. Performance Impacts of US06 CO Standards
Summary of Comments. In their comments AAMA/AIAM stated that they
felt EPA's proposed standards sought to eliminate all enrichment
without regard for impact on performance and in doing so EPA glossed
over the impact of completely eliminating commanded enrichment on
reducing engine power. AAMA/AIAM argued that EPA must either factor the
lost value of performance to consumers or factor in engine or drive
train modifications into it's analysis of emissions and fuel economy.
AAMA/AIAM also stated that EPA did not use proper statistical
techniques to distinguish variability from consistent trends in the WOT
time analysis used to claim minimal effects on performance, and AAMA/
AIAM alternatively proposed that a two second limit on WOT control
would keep the loss of power to manageable limits for Tier I vehicles.
Both GM and Suzuki stated that extended stoichiometric control at
WOT would result in elimination of small displacement engines.
SEMA expressed their belief that stoichiometric control at WOT
would create a safety concern for low-powered vehicles, as they could
be underpowered and thus less safe when merging onto highways or
climbing hills. SEMA also stated that the use of timers on high
performance vehicles will cause an in-use safety problem when
enrichment is invoked and extra power is suddenly introduced.
Response to Comments. EPA believes the revisions to the CO
standards render the comments on performance impact moot, for all
practical purposes. With the 9.0 g/mi CO standard, higher performance
vehicles will be able to use enrichment immediately at WOT, most
vehicles will need to delay enrichment for no more than two seconds,
and no vehicle should need to delay enrichment for more than four
seconds. As the manufacturers stated in their comments that a two
second limit on WOT control will keep the loss of power to manageable
limits for Tier 1 vehicles and proposed a method for such control that
would inherently require a three to four second timer, there should not
be a significant performance impact even on the lower performance
vehicles that would need a short period of WOT enrichment control.
EPA disagrees with SEMA's statements about potential safety
concerns on low-powered vehicles and the use of timers on high-
performance vehicles. Even if enrichment were eliminated for extended
periods of time, the performance reduction would be very small (3-5
percent) compared to the range of performance levels that already exist
in the vehicle fleet (which differ by a factor of 2-3). Similar logic
applies to the use of timers on high performance vehicles. The
introduction of enrichment after a period of stoichiometric operation
causes an increase in the power output of the engine of no more than
five percent. This impact is quite small compared to the engine output
increase as the engine increases in RPM from second to second and to
the sudden increase in power delivered by a turbocharger, which can be
in the range of a 50 percent power boost.
4. US06 NMHC+NOX Standard
Summary of Proposal. The NPRM proposed to hold US06 NOX
emissions to overall FTP emission levels and NMHC emissions to FTP bag
2 emission levels. For Tier I LDV and LDT1 vehicles, the FTP NOX
standard is 0.4 g/mi. While no standards exist for FTP bag 2 emissions,
the average FTP bag 2 emissions for Tier I LDV and LDT1 vehicles would
correspond to an NMHC standard of roughly 0.05 g/mi. Thus, the NPRM
implicitly proposed an US06 NMHC+NOX standard of about 0.45 g/mi
for LDV and LDT1 vehicles.
Summary of Comments. AAMA/AIAM submitted a proposal to set US06
standards by averaging all the Tier I LDV and LDT1 US06 stoichiometric
test results, multiplied by a factor of two to provide necessary
headroom. Based upon this methodology, they proposed US06 standards of
1.1 g/mi NMHC+NOX. AAMA/AIAM also stated that this emission level,
with appropriate load adjustments, should be feasible with only
recalibration for most vehicles.
AAMA/AIAM also submitted a number of comments questioning the data
analysis done by EPA to develop proposed NOX standards, and stated
that recalibration alone would be insufficient to meet EPA's proposed
standards and larger catalysts would be required.
Ford also commented that EPA's proposed standards could not be met
with only calibration changes and stated that catalyst systems would
have to be redesigned, including catalyst volume, precious metal
loading, and catalyst placement. Ford also expressed concern that
increasing EGR flow to reduce NOX over the US06 cycle could have
negative impacts on driveability, HC emissions, and fuel economy.
Response to Comments. Comments and new data provided by AAMA/AIAM
convinced EPA to revise the US06 standards based on new data for Tier 1
vehicles.
EPA expended considerable effort examining the impact of a wide
variety of factors on US06 NMHC+NOX emissions, including vehicle
and engine size, vehicle weight, performance, catalyst loadings and
size, exhaust flow, and eight different air/fuel parameters. The only
factor identified with a consistent, significant impact on US06
emissions was the bias of the air/fuel ratio (i.e., whether the vehicle
exhibited significant lean or rich bias during US06 operation). Of the
29 LDV, LDT1, and LDT2 Tier 1 vehicles tested over the US06 cycle, 14
were identified as having no significant air/fuel bias. Ten

[[Page 54864]]

vehicles were identified with a lean-bias to their air/fuel calibration
or with a shift in the air/fuel calibration from the production to
stoichiometric calibration; these vehicles generated NOX emissions
two to four times higher than the unbiased vehicles. The remaining five
vehicles with a rich bias all had significant increases in NMHC and CO
emissions, with erratic NOX impacts (i.e. some had relatively low
NOX emissions, but two had high NOX emissions).
The 14 vehicles with unbiased air/fuel calibrations covered a wide
range of manufacturers, size, weight, performance, and catalyst
loadings and size. Substantial work on identifying additional factors
causing differences in emissions and catalyst conversion efficiency
between these 14 vehicles again failed to reveal any other significant
influences. Given the lack of additional factors identified and the
reasonable representation of the whole fleet by the vehicles having
unbiased air/fuel calibrations, EPA established Tier 1 US06
NMHC+NOX design targets based on the simple average of the
vehicles identified as having unbiased air/fuel calibrations. The
intermediate useful life NMHC+NOX design target was calculated to
be 0.29 g/mi for LDVs and LDT1s.
The Agency believes that the great majority of vehicles can meet
the design target level simply with better attention to proper air/fuel
calibration. This conclusion is supported by the following factors:

1. Each vehicle identified as having a lean-bias or an erratic
stoichiometric calibration had NMHC+NOX levels over twice the
design target. The Agency believes that better air/fuel calibration
will reduce the emissions from all of the vehicles with lean-bias
and erratic calibrations to the level of the vehicles with good
calibrations.
2. The conclusion from the preceding paragraph is supported by
the emissions from the LDT1 and LDT2 trucks. All five of the LDT1s
tested had unbiased air/fuel control; four of the five meet the
design level even with the unoptimized stoichiometric calibrations
used for the test program. For the LDT2s, four of the six vehicles
tested had unbiased air/fuel control; all four of these vehicles
plus one vehicle with a rich air/fuel bias meet or come very close
to meeting the design target with the unoptimized stoichiometric
calibration used for the test program. While the stoichiometric
emissions were higher on the sixth vehicle, with the production
calibration this vehicle produced NMHC+NOX emissions right at
the design target level. Thus, it appears likely that all six of the
LDT2s can meet the design target level with little, if any,
modification. As these trucks constitute an extremely broad range of
weight, performance, and engine size, the Agency believes that LDVs
would be able to duplicate the emission performance of the trucks,
given similar air/fuel calibration strategies.
3. The US06 NOX design target is about 75 percent above the
current NOX emission level from hot, stabilized driving over
the FTP driving cycles. As engine-out NOX emissions are also
about 75 percent higher on the US06 compared to the FTP, the US06
design target can be met by maintaining the same NOX conversion
efficiency on US06 as the vehicle achieves during hot, stabilized
FTP operation. Analyses conducted by EPA indicate that equivalent
NOX conversion efficiency is a reasonable assumption.

While NMHC+NOX standards were not promulgated for US06
separately, a US06 standard level of 0.58 g/mi for LDVs and LDT1s (the
0.29 g/mi design target multiplied by the headroom factor of two) was
used in the calculation of the NMHC+NOX composite standards
presented in the ``Description of the Action'' section, above. Further
description of how the composite standards were calculated can be found
in the ``Composite Standard'' section, below.

D. Intermediate Soak

Summary of Proposal. The Agency proposed to control tailpipe
emissions following soaks of intermediate duration (between 10 minutes
and 3 hours) by requiring that emissions on the SC01 cycle following a
60 minute soak not be greater than emissions over Bag 3 of the FTP. The
NPRM also stated that the decision to finalize the intermediate soak
requirement would be contingent on the cost effectiveness of the
requirement for vehicles complying with LEV and lower standards. The
Agency surmised that increased thermal insulation around the catalytic
substrate(s) would be used to meet this requirement.
Summary of Comments. All comments received from auto manufacturers
and manufacturer organizations, including AAMA/AIAM, GM, Honda, and
Land Rover, objected to the intermediate soak requirement on the basis
of the cost not justifying the benefits. These arguments were centered
on four major points: (1) The emissions benefit would be significantly
reduced as more advanced cold start technologies are implemented to
comply with lower emission standards, (2) the cost of implementing
EPA's primary control strategy, catalyst insulation, would be
prohibitive from an exhaust system packaging standpoint, (3) the use of
catalyst insulation would increase the thermal severity of the catalyst
environment, bringing greater risk of catalyst deterioration over the
life of the vehicle, and (4) the test facility implications of adding
an intermediate soak procedure would be significant.
Comments that supported the inclusion of the intermediate soak
requirement were submitted by the NESCAUM, the National Renewable
Energy Laboratory (NREL), and the MECA. NESCAUM and MECA supported the
intermediate soak requirement in the context of making the test
procedure representative of in-use driving per the intent of the Clean
Air Act Amendments of 1990. NREL recommended that the intermediate soak
period be extended to at least 2 hours to provide an improved
representation of in-use soak periods, with waivers available for
catalyst technology that is demonstrated to remain at high temperature
during such soaks. Comments supplied by NREL and MECA also provided
information on technology under development that would mitigate
intermediate soak emissions.
Response to Comments. Controlling intermediate soak emissions would
require hardware changes to keep the catalyst warm longer or to heat it
up faster. Possible techniques include catalyst insulation and catalyst
preheaters, but any technique will likely result in significant
redesign and retooling investments. For example, the most inexpensive
technique, as discussed in the NPRM, is likely to be catalyst
insulation. Even this option would require redesign of the catalyst
can, possibly including new can material, and development of a thicker,
insulated, catalyst mounting material. The overall size of the catalyst
would increase due to the insulating material, possibly to the point at
which it would not fit into current space, which would require redesign
of the vehicle floorpan. Finally, the catalyst insulation would
increase internal catalyst temperatures, potentially leading to higher
catalyst deterioration.
In the analysis conducted by EPA in support of the NPRM, all of the
redesign problems were considered manageable and cost effective for
Tier 1 vehicles, provided that the high up-front redesign and tooling
costs could be amortized over at least five years of production. This
differs from US06 and air conditioning control, which can be
predominantly accomplished without hardware changes and high retooling
costs. Because of the hardware investment to meet intermediate soak
requirements and the high potential for intermediate soak requirements
to be in effect on Tier 1 vehicles for only a couple of years before
being replaced by National LEV or Tier 2 requirements, it would likely
be a waste of

[[Page 54865]]

manufacturers' resources to establish intermediate soak requirements
only for Tier 1 vehicles. Thus, one of EPA's criteria in promulgating
intermediate soak requirements was whether or not they would continue
to be cost effective for LEV-like vehicles.
Unfortunately, the feasibility of intermediate soak requirements on
Tier 2 or NLEVs is much less certain. While catalyst temperature data
indicate that the increased catalyst temperature caused by catalyst
insulation is not likely to be a problem for Tier 1 vehicles, Tier 2 or
NLEVs are likely to move catalysts closer to the engine, increasing the
temperature concerns with catalyst insulation. EPA does not have
sufficient information on the impact of catalyst insulation on the
durability of Tier 2 or NLEVs catalysts, including their higher
baseline temperatures and improved catalyst formulations, to quantify
the extent of this concern.
Moving the catalysts closer to the engine will also reduce catalyst
light-off time, potentially reducing intermediate soak emissions even
without intermediate soak standards. Using new emission data provided
by AAMA/AIAM and CARB in their comments on vehicles certified to
emission standards lower than Tier 1, EPA assessed the potential
emission benefits of the intermediate soak requirement on Tier 2 or
NLEVs. This data indicated that the benefit on LEV vehicles would be
about 60 percent of that on Tier 1 vehicles, or about 0.04 g/mi
NMHC+NOX. Under the Agency's ``best-case'' cost scenario, this
would result in a cost per ton of NMHC+NOX reduced of
approximately $3100. Taking into account some uncertainties about the
need to revise floorpans on some vehicles, possible reduced benefit of
insulation, and possibly requiring insulation on multiple catalysts,
the upper bound estimate is approximately $13,000 per ton NMHC+NOX
reduced. These estimates include an estimate of the NOX increase
resulting from A/C operation over soaks based on data from a LEV
prototype vehicle.
Although the analysis of the LEV soak data indicates that there
would continue to be some emissions benefits from controlling soak
emissions, these data also indicate that intermediate soak emissions
are being reduced as a result of the technology to be used for
complying with Tier 2 or LEV standards, which target cold start
emission reductions. The Agency believes that adding a 1 to 2 hour soak
would add little value to the FTP for the purpose of controlling
emissions. As a result of the reduced benefit on LEV-like vehicles and
uncertainties regarding cost and feasibility of control discussed
above, the Agency has decided not to finalize the intermediate soak
requirement at this time.
However, because this action is based on emission levels from a
small sample of prototype vehicles as well as current technological
restrictions, the Agency is not ruling out the possibility of
promulgating this requirement at a later time. Intermediate soak
emissions will continue to contribute somewhat to the in-use inventory
even as LEV and ULEV technologies penetrate the in-use fleet. The
Agency will monitor the performance of production LEV and ULEV vehicles
over intermediate soaks to verify the conclusions from the prototype
analysis. At the same time, the Agency will encourage the development
of technologies that will allow for the control of intermediate soak
emissions in a manner that is cost effective and not detrimental to the
emission control system.

E. Air Conditioning

1. Test Cycle
Summary of the Proposal. The proposed SFTP included an air
conditioning simulation to be performed during the hot stabilized 866
cycle and the start control cycle (SC01). The standards implicitly
assumed that emissions over the SC01 cycle could be held to the same
level as emissions over the 505 cycle used for Bag 3 of the FTP.
Comments were specifically solicited on the possibility of
substituting the 505 component of the LA4 (The LA4 consists of a 505
cycle followed by an 866 cycle) for SC01 and on whether full air
conditioning simulation should be added to the US06 cycle. The Agency
also stated that it believes it may be appropriate to return to the
issue of cold start testing with air conditioning operation with
respect to future technologies and future test procedures and emission
standards; comments were also solicited on this issue.
Summary of Comments. NESCAUM, MECA, and CARB all supported the need
to account for air conditioning load over the cycles proposed. NESCAUM
and CARB also supported testing with actual air conditioning load over
cold start conditions (bag 1 of the FTP). MECA and CARB stated that air
conditioning load should also be accounted for during aggressive diving
(US06).
AAMA/AIAM stated that EPA has not demonstrated the feasibility of
its proposed standards for operation over the SC01 cycle. They were
especially critical of EPA's conclusion that the difference in
emissions between SC01 and the 505 were due to microtransient emission
response, which could be controlled with sequential multi-point fuel
injection and better calibrations. AAMA/AIAM stated that the data did
not justify using SC01 and recommended that the air conditioning test
procedure consist of the hot LA4 without a soak. AAMA/AIAM also stated
that cold start emissions related to air conditioning operation are
already addressed through the FTP and can only be improved by
increasing the overall stringency of the current Tier 1 standards.
Suzuki stated that the SC01 cycle is too aggressive in general and
too severe for small engines. They recommended that EPA consider a
unique schedule or cycle adjustment for small engines, due to the
disproportional load that air conditioning places on small engines.
Response to the Comments. As discussed in the NPRM, EPA recognized
that the proposed SC01 cycle needed revisions to better reflect the in-
use speed/acceleration distribution; the revised cycle is known as
SC03. The final A/C test requirement will consist of a 10 minute soak
and the SC03 cycle. Except for the revisions to SC01, EPA did not find
the arguments presented by the commenters sufficient to make additional
modifications.
EPA is concerned about emissions from microtransient driving
behavior. Many vehicles' emissions are sensitive to driving behavior,
and data indicate that small speed variations actually occur about 50
percent more frequently than on the LA-4 driving cycle. On the other
hand, there is some merit to AAMA/AIAM's arguments that factors other
than microtransients likely impact the difference in emissions seen on
the SC01 versus the 505 driving cycles. Thus, the standards have been
adjusted for the difference in emissions between the new cycle and the
505.
As indicated in the NPRM, an error was made in the generation of
the SC01 cycle. Proper matching of the in-use driving distribution
yielded a revised cycle, called SC03. Overall, the positive kinetic
energy (PKE) from accelerations on the SC03 cycle is about halfway
between the PKE of the 505 and the SC01 cycles. EPA calculated the
likely difference in emissions between the 505 and SC03 to be 48
percent of the difference in emissions observed between SC01 and the
505.
The adjustments made in SC03 address Suzuki's comment that the SC01
was too aggressive in general, although EPA disagrees that SC01 is too
severe for small engines. While it is true that air conditioning places
a

[[Page 54866]]

disproportional load on small engines, this is merely a reflection of
what actually occurs in use. In addition, the total mass flow through a
small engine is still lower than occurs with larger engines and
vehicles; thus, small engines should be able to comply with the
standards.
The 866 cycle was dropped in the final rule because inclusion of
the 866 cycle would greatly over-represent low speed, low acceleration
driving. Emission reductions achieved on the 866 with air conditioning
operation may not result in equivalent in-use emission reductions. As
the SC03 cycle was specifically developed to match the speed and
acceleration distribution of in-use driving, less the high speed and
acceleration driving represented by US06, the SC03 offers far more
assurance that emission reduction on the cycle will proportionally
reduce in-use emissions.
While EPA agrees in principal with comments from MECA and CARB that
air conditioning load should be included in aggressive driving (US06),
EPA believes that, in practical terms, adding air conditioning load to
the US06 cycle would be largely meaningless. The US06 cycle already
pushes virtually all vehicles into WOT; inclusion of air conditioning
load would simply expand the amount of time spent at WOT and increase
the overall engine-out NOX emissions proportionally to the extra
load. This increase would wind up being incorporated into higher
emission levels, without any real impact on the control of emissions
during air conditioning operation.
EPA also agrees in principal with comments from NESCAUM and CARB
that air conditioning operation during cold starts should be accounted
for. Unfortunately, as AAMA/AIAM points out in their comments, the
primary way to control the addition to emissions during cold starts
would be to shorten catalyst light-off time. The Agency believes that
requiring control of air conditioning-related emissions on a cold start
test is inappropriate at this time because of the lead time and cost
necessary to implement new catalyst technology. The Agency intends to
revisit this issue as part of the Tier 2 standards, when the air
conditioning impact can be assessed as part of the standard setting
process.
2. Air Conditioning Simulation
Summary of the Proposal. As an alternative to using a full
environmental chamber for air conditioning testing, the Agency proposed
a simulation procedure that could be conducted in a standard test cell.
The proposed simulation included a 95 deg.F 5 deg.F test
cell ambient temperature, front-end supplemental fan cooling, driver's
window down, and vehicle climate controls settings of maximum A/C,
interior air recirculation, high interior fan, and coldest temperature.
Testing in a full environmental chamber was proposed to also be
permitted, at the manufacturer's option.
Comments were also requested on two other simulations, bench
testing and a dynamometer simulation approach proposed by the vehicle
manufacturers, dubbed ``Nissan-II.''
Summary of the Comments. NESCAUM stated that EPA should rely on the
actual operation of the air conditioner with an environmental
simulation. They also expressly requested that EPA not lower the
maximum ambient temperature. Horiba also opposed using the dynamometer
to simulate the air conditioning load, stating that it would affect the
driveability of the vehicle on the dynamometer differently from highway
driving. Horiba suggested that the air conditioning be turned on for
the test, with the windows open and an auxiliary heat source if
necessary.
CARB advocated the use of full environmental chambers for air
conditioning testing, stating that its incremental cost would be less
than $3 per test and requesting that EPA also do a cost-effectiveness
analysis of using full environmental chambers. CARB was willing to
consider options for a ``short-cut procedure if sufficient correlation
with environmental chamber data can be demonstrated.''
AAMA/AIAM stated that correlation of the proposed simulation with
the full environmental chamber results was poor and that EPA's analysis
of the correlation was misleading. AAMA/AIAM also noted cost concerns
with performing the simulation, since facilities must be capable of
handling the increased cell temperature, humidity, and air flow.
Honda stated that a full environmental chamber would not be cost
effective, considering the cost of the technology needed to comply with
the air conditioning requirement. They strongly recommended that EPA
not only address air conditioning simulation technology, but also
consider facility cost and feasibility so that all manufacturers could
conduct SFTP tests without an additional heavy burden.
Response to the Comments. As neither CARB nor vehicle manufacturers
supported the air conditioning simulation as proposed, much work has
been done since the NPRM developing other air conditioning simulations.
None of the simulations, at this relatively early stage of development,
have yet demonstrated sufficient correlation to be used as a permanent
substitute for full environmental chambers. However, there is a strong
probability that further development could yield an effective air
conditioning simulation.
Meanwhile, EPA has spent considerable effort evaluating the cost of
using full environmental chambers, as well as the incremental savings
associated with an air conditioning simulation. While EPA estimates
that using full environmental chambers for all air conditioning testing
would cost a little more than estimated by CARB, $3.05 per vehicle, the
cost is still low enough to support CARB's conclusion that using full
environmental chambers is cost-effective. However, a workable
simulation would allow a significant cost reduction to manufacturers
and consumers, which would be worthwhile so long as it did not
significantly impact the air quality benefits.
The long range solution reached by EPA is to mandate the use of
full environmental chambers, with an option for using a simulation if
correlation can be demonstrated. To encourage proper development and
use of simulations, ``acceptance criteria'' have been developed. Before
a simulation procedure may be used by a manufacturer, the manufacturers
must agree to perform spot check verifications to demonstrate that the
simulation procedure satisfactorily correlates with the full
environmental chamber for each engine/vehicle combination covered. This
consists of verifying the correlation for up to five vehicles per
manufacturer (one for small volume manufacturers) of EPA's choice at
the time of certification. Five vehicles per manufacturer are specified
to allow EPA flexibility in targeting new A/C simulations and
manufacturers with poor track records; in other cases EPA will likely
specify only two vehicles per manufacturer. Due to the large
variability in emissions from test to test and lab to lab and EPA's
desire to avoid improperly failing good simulations, the simulation
tailpipe NOX emissions must be at least 85 percent of the full
environmental chamber NOX emissions. The fuel consumption, (a good
surrogate for overall load on the engine) in the simulation must be at
least 95 percent of the fuel consumption in the full environmental
chamber. Retests and reapplication of these thresholds are also
allowed, as described in the ``Description of the Action.'' If an
engine/vehicle fails, the manufacturer must remedy the air conditioning
load imposed during the simulation or use

[[Page 54867]]

full environmental chambers for future testing. Data must also be
supplied establishing how many other engine/vehicle combinations are
similar to the failing configuration. Any future data generated on
these engine/vehicle combinations, including in-use enforcement
testing, must use the corrected procedure. If any vehicle fails to meet
the tailpipe emission standards due to a corrected air conditioning
load, all applicable vehicles are subject to an emissions recall;
however, there would be no recall liability associated with the air
conditioning load correction itself. For every engine/vehicle
combination which fails this demonstration, EPA may require the
manufacturer to verify the correlation between the simulation and the
full environmental chamber for an additional two vehicles of EPA's
choice.
The results from each manufacturers correlation demonstrations will
also be tracked over time. The manufacturer is expected to target the
simulation to at least 100 percent of the emissions from the full
environmental chamber. If, over time, the emissions from the
simulations are found to be statistically lower than the full
environmental chamber, further use of simulations by that manufacturer
will not be allowed until the causes of the offset are identified and
corrected.
While these acceptance and verification procedures should encourage
development of accurate air conditioning simulations in the long run,
applying them immediately would create a leadtime problem. No
simulations have been developed yet that can meet the criteria and
building full environmental chambers is time consuming and expensive.
To avoid significant delays in implementing the air conditioning
requirements and to allow additional time to develop simulations, EPA
is allowing the use of the AC1 or the AC2 simulations used in the ACR3
and ACC3 testing programs without verification during the three-year
phase-in period.\9\ Starting with MY2003, any simulation procedure will
be subject to the quality audit verification test program discussed
above. Testing in a full environmental chamber will be acceptable at
any time.
---------------------------------------------------------------------------

\9\ During the development of these simulations, the AC1 and AC2
methods were referred to as the Nissan-II and Toyota simulations,
respectively. See Sec. 86.162-00 of today's final regulations for
details of these simulations.
---------------------------------------------------------------------------

The long term requirement for any simulation to correlate with
actual air conditioning operation in a full environmental chamber
should satisfy the concerns expressed by NESCAUM and CARB. The
requirement to correlate with a full environmental chamber also
addresses Horiba's opposition to using the dynamometer due to
inappropriate driveability impacts, as a procedure could not pass the
correlation criteria if this effect were to occur.
3. Air Conditioning Standards
Summary of the Proposal. The NPRM proposed that vehicles maintain
existing NMHC and CO emission levels with the air conditioning turned
on. The NPRM concluded that 25 percent of the NOX increase with
the air conditioning engaged was likely to be unavoidable without
increasing the stringency of the current NOX standard, but
proposed controlling the other 75 percent. In the proposed composite
standard, the allowable 25 percent NOX emission increase was
calculated to be equivalent to an adjustment factor of 1.15 applied to
the FTP NOX standard. The NPRM specifically requested comments on
the feasibility of the proposed levels of control and the technology
implications of controlling emissions to this level.
Summary of the Comments. NRDC opposed the 15 percent ``relaxing''
of NOX standards, stating that any revised standard requires a
reduction in emissions.
CARB was generally supportive, but commented that there was no data
on vehicles that were optimized for emissions with A/C on.
AAMA/AIAM commented that the proposed standards were not based on
available test data or ``sound engineering analysis.'' Specifically,
they stated that EPA performed no technical feasibility analysis for an
A/C NOX standard. They argued that their analyses indicated that
74 percent of the NOX increase was due to an increase in engine-
out emissions that was an inherent function of the additional load
placed on the engine by the air conditioner. AAMA/AIAM did acknowledge
that it may be possible to inexpensively eliminate much or most of the
loss in NOX conversion efficiency which occurred with the air
conditioner on, which their analyses indicate was 26 percent of the
total NOX increase.
AAMA/AIAM also claimed that EPA did not adequately explain the CO
increase with A/C on and that, in assessing NOX conversion
efficiencies, EPA ignored NMHC and CO levels. They also argued that
EPA's approach of turning the air conditioning compressor off for brief
periods of time at high load points actually produces very little
emission improvement, as EPA did not add back in any additional
compressor operation during other parts of the cycle and ignored the
impacts of this additional cycling on compressor durability or
efficiency. They claimed that EPA did not assess the feasibility of
reducing engine-out NOX emissions.
Response to the Comments. There is some validity to AAMA/AIAM's
criticisms that EPA did not adequately explain the CO increases with
the air conditioning on, ignored NMHC and CO levels when assessing
NOX emissions, did not add back in additional compressor operation
to compensate for turning off the compressor at high load points, and
did not adequately assess the feasibility of reducing engine-out
NOX emissions. In addition, subsequent to the publication of the
NPRM, EPA learned that the vehicles used in the NPRM to set standards
were tested with low mileage catalysts. Consequently, EPA and the
manufacturers agreed to conduct a new test program.
Unfortunately, examination of the available data indicates that
directly setting tailpipe air conditioning standards has some
significant problems:

1. The ACR1 data was tested with low-mileage catalysts,
2. Only four LDVs were tested in the ACR3/ACC3 test programs,
three of which were Fords,
3. One of the four LDVs was identified in the US06 analysis as
having a lean air/fuel bias and generating high NOX emissions
under higher loads,
4. Another of the four LDVs had extremely high variability in
tailpipe emissions from test to test, indicating an erratic emission
control system.

Fortunately, it is reasonable to assume that catalyst conversion
efficiency should not be significantly impacted by air conditioning
operation. AAMA/AIAM comments that air conditioning emission increases
due to loss in catalyst conversion efficiency can be relatively easily
controlled support this assumption. This equivalency in conversion
efficiency means that air conditioning design targets can be set by
calculating the engine-out ratio of emissions with the air conditioning
on to air conditioning off and applying this ratio to baseline tailpipe
emissions with the air conditioning off.
Baseline hot, stabilized tailpipe emissions exist from 22 LDVs and
LDT1s in the US06 test program. As these vehicles were chosen as a
representative cross-section of the new vehicle fleet, they provide
excellent baseline tailpipe emissions. The second step in the process
is to assess what portion of the observed engine-out emission increase
is unavoidable and what portion could be reduced with appropriate
emission control. As this analysis can be done on engine-out emissions,
EPA was able to assess the

[[Page 54868]]

performance of 12 cars and trucks in the ACR1 and ACR3/ACC3 test
programs, a much larger and much more representative data set than the
four cars (two of which have suspect emission controls) available to
set tailpipe emission standards directly.
Air conditioning operation increases the overall, average load on
the engine by about 25 percent. However, this increase in load has a
disproportionate impact on NOX formation, as very little NOX
is formed at low engine loads and the amount of EGR that can be
tolerated decreases as engine speeds and loads increase beyond a
relatively low level. As discussed more fully in the RTC, EPA has
concluded that the load imposed by current air conditioning systems
results in an unavoidable 50 percent increase in engine-out NOX
emissions. This NOX increase is inherent to the additional load
placed upon the engine and how this increased load impacts the peak
combustion temperature in the engine. The conclusion of an inherent 50
percent engine-out NOX increase is supported by the average
NOX increase on the Ford vehicles of 53 percent, as the Ford
vehicles had closed-loop electronic EGR systems and the EGR flow rates
were more carefully calibrated throughout the entire speed/load range
than the other vehicles (engine-out NOX on non-Ford vehicles in
the test programs increased by an average of 67 percent with the air
conditioning on). The only way to further reduce the emission increase
is to reduce overall emissions, such as with improved catalyst
formulations, or by reducing the load placed on the engine by the air
conditioning system.
In the case of NMHC, EPA's analyses indicate that the best
conclusion is still that reached in the NPRM, that HC emissions should
not be affected by air conditioning operation.
In the NPRM, EPA attributed the increase in CO emissions with the
air conditioning on to increased periods of brief commanded enrichment
and proposed that CO emissions not increase with the air conditioner
on. This assumption was challenged by the manufacturers in their
comments, stating that CO emissions should be proportional to the
overall load. While EPA continues to believe that the additional load
imposed by the air conditioner triggers brief periods of commanded
enrichment that will not occur once vehicles have been recalibrated to
comply with the high speed and acceleration requirements, EPA also
acknowledges that the mass flow through the engine is likely to have
some impact on engine-out CO emissions. As engine-out CO emissions in
both the ACR1 and ACR3 programs increased only moderately, the average
increase in engine-out CO emissions from the ACR1 an

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A96-24485. Public record. Not legal advice.