United States        Air and Radiation        EPA420-R-00-010
          Environmental Protection                 July 2000
          Agency
vxEPA    Regulatory Impact Analysis:
          Control of Emissions of Air
          Pollution from Highway
          Heavy-Duty Engines
                                 > Printed on Recycled Paper

-------
                                              EPA420-R-00-010
                                                      July 2000
of
    Assessment and Standards Division
  Office of Transportation and Air Quality
   U.S. Environmental Protection Agency

-------
                             Table of Contents
CHAPTER 1:  INTRODUCTION  	1
      I.  Summary of the Regulatory Impact Analysis 	1
             A.  Chapter 2—Health and Welfare Concerns	1
             B.  Chapter 3—Technological Feasibility of HD Diesel and Otto-cycle Standards
                    	1
             C.  Chapter 4—Economic Impact of HD Diesel Standards	1
             D.  Chapter 5—Economic Impact of HD Otto-cycle Standards 	1
             E.  Chapter 6—Environmental Impact of HD Diesel Standard	2
             F.  Chapter 7—Environmental Impact of HD Otto-cycle Standards	2
             G.  Chapter 8—Cost-effectiveness for HD Diesel and Otto-cycle Requirements
                    	2

CHAPTER 2:  HEALTH AND WELFARE CONCERNS	3
      I.  Health and Welfare Concerns  	3
             A.  Health and Welfare Effects from NMHC and NOx	3
             B.  Paniculate Matter	4
      II. Need for Ozone Control	6
             A.  Standards for 2004 HD Diesels Are a Key Part of State Air Pollution Control
                   Plans	6
             B.  New Standards for 2005 HD Gasoline Engines and Vehicles Are Important for
                   States in Meeting Their Air Quality Goals  	7
             C.     HD Diesel and Gasoline Engines Contribute to Total NOx and VOC
                   Emissions	13
      IE. Particulate Matter  	15
             A.  Current and Future Compliance with the PM10 NAAQS	15
                   B.    HD Diesel and Gasoline Engines Contribute to Secondary
                         Formation of Particulate Matter	18
      IV. Air Toxics from HD Engines and Vehicles 	18
      Chapter 2 References	20

CHAPTER 3:  TECHNOLOGICAL FEASIBILITY OF HD DIESEL AND OTTO-CYCLE
      STANDARDS	21
      I.  Overview	21
      II. Diesel Engine Technologies  	21
             A.  HD Diesel Technology Overview 	21
             B.  Exhaust Gas Re-circulation (EGR)  	22
             C.  Fuel Injection Rate-shaping	25
             D.  Exhaust After-treatment	26
             E.  Diesel Fuel Composition 	36

-------
             F. Performance of Projected Emission Control Technologies over Typical In-Use
                   Conditions  	44
             G.  Summary and Conclusions regarding HD Diesel Technologies 	45
      IE. HD Otto-cycle Engine & Vehicle Technologies	47
             A.  Base Engine Improvements	48
             B.  Improvements in Fuel Atomization	53
             C.  Improvements to Exhaust Aftertreatment Systems  	54
             D.  Improvements in Engine Calibration Techniques  	57
             E.  Advanced Technology 	58
             F. Technologies In-use On Current Otto-cycle HD Engines	59
             G.  Chassis-based standards	62
             H.  Engine-based Standards	63
      IV. On-board Diagnostics for HD Diesel and Otto-cycle Engines	66
      Chapter 3 References	69

CHAPTER 4:  ECONOMIC IMPACT OF HD DIESEL STANDARDS  	76
      I.  Methodology 	76
      II. Technologies for Meeting the New Standards	77
             A.  Primary Technologies  	80
             B.  Operating Costs  	84
             C.  Secondary Technologies	85
      IE. Summary of Costs	86
      IV. Aggregate Costs to Society	87
      Chapter 4 References	91

CHAPTER 5: ECONOMIC IMPACTS OF HD OTTO-CYCLE STANDARDS  	93
      I.  Methodology for Estimating Costs  	93
      II. Technology Packages for Compliance with the Regulations	93
      in. Technology Costs  	96
             A.  Improved Catalysts	96
             B.  Exhaust Gas Recirculation (EGR)	102
             C.  Secondary Air Injection	102
             D.  On-board Diagnostics	103
             E. Exhaust Systems	104
             F. Electronic Control Module	105
             G.  Onboard Refueling Vapor Recovery  	105
      IV. Fixed Costs	106
             A.  R&D and Tooling Costs	106
             B.  Certification Costs 	106
             C.  In-use Testing Costs	106
      V.  Summary of Costs  	106
      VI. Aggregate Cost to Society	107
      Chapter 5 References	109

-------
CHAPTER 6:  ENVIRONMENTAL IMPACT OF
      HD DIESEL STANDARDS 	110
      I. Introduction	110
      II. Description of Calculation Method  	Ill
             A.  General Equation	Ill
             B.  Conversion Factors	112
             C. Vehicle Miles Traveled	112
      IE.  Total Nationwide Inventories  	116
             A.  Current Inventories	117
             B.  NOx Emission Projections and Impacts	117
             C.  NMHC Emission Projections and Impacts	120
      IV.  Per Vehicle Emission Impacts  	121
      V. Environmental Impacts  of Emission Reductions	122
             A.  Ozone Impacts	122
             B.  Air Toxics	122
             C.  Other Impacts of Emission Reductions 	123
      VI.  Summary	123

CHAPTER 7: ENVIRONMENTAL IMPACT OF THE HD OTTO-CYCLE STANDARDS
       	127
      I. Introduction	127
      II. Exhaust NOx and NMHC Emission Factors 	127
             A.  Baseline Emission Rates (Zero-Mile Levels and Deterioration Rates) .... 127
             B.  Conversion Factors	128
             C.  Control Emission Rates (Zero-Mile Levels and Deterioration Rates)	128
             D.  Emission Rate Adjustments	129
      IE.  Per-Vehicle Exhaust NOx and NMHC Emission Reductions  	130
             A.  Per Vehicle Emission Rates	130
             B.  Mileage Accumulation and Scrappage Rates	131
             C.  Per-vehicle Lifetime Emissions and Emission Reductions 	133
      IV.  HDGV Exhaust Inventory and Reductions	134
      V. ORVR Benefits 	137
      Chapter 7 References	139

CHAPTER 8:  COST-EFFECTIVENESS FOR HD DIESEL AND OTTO-CYCLE
      REQUIREMENTS	140
      I. Cost-Effectiveness of the Diesel Requirements	140
      II. Cost-Effectiveness of the Otto-cycle Requirements	143
             A.  Exhaust Emission Standards 	143
             B.  Refueling Emission Standards	145
      IE.  Other Benefits	147
      IV.  Cost-Effectiveness Sensitivity Analyses	147
      V. Comparison of Cost-Effectiveness with Other Mobile Source NOx Control Strategies
             	149

-------
     Chapter 8 References	151




APPENDIX A




     ANMJALIZED COST EFFECTIVENESS ANALYSIS 	 A-l

-------
                                                           Chapter 1:  Introduction
                   CHAPTER 1:   INTRODUCTION

      The Regulatory Impact Analysis (RIA) for this rule presents analysis and supporting data for
the provisions EPA is reviewing or promulgating for model year 2004 and later on-highway heavy-
duty diesel and otto-cycle engines and vehicles.  This chapter presents a brief summary of each
chapter contained in the (RIA) that follows.

I.  Summary of the Regulatory Impact Analysis

      A.  Chapter 2—Health and Welfare Concerns

      Chapter 2 provides an overview of the health and environmental effects associated with
ozone, particulate matter, and other harmful pollution associated with emissions from heavy-duty
engines and vehicles. Chapter 2 reviews some of EPA's key concerns at this time. The chapter also
provides national NOx and VOC emissions inventories and emissions trends, with specific emphasis
on the contribution from on-highway heavy-duty diesel and otto-cycle vehicles.

      B.    Chapter  3—Technological  Feasibility  of HD  Diesel and  Otto-cycle
Standards

      To achieve  the  standards reviewed  or promulgated in  this  rule,  heavy-duty engine
manufacturers will need to consider a combination of new and existing emission control devices.
Chapter 3  presents the technologies available and discusses their ability to reach the required
emission levels.  Chapter 3 is  divided into two major sub-chapters, the first dealing with HD diesel
technologies, and the second with HD otto-cycle technologies.

      C.  Chapter 4—Economic Impact of HD Diesel Standards

      Chapter 4 presents EPA's best assessment of the economic impacts which will result from
the HD diesel standards.  The assessment includes EPA's estimates of the technology packages
manufactures will use, as  well as the costs associated with new certification and  compliance
requirements. Costs are estimated on a per-vehicle basis, as well as an aggregate cost to society.
Chapter 4 also includes an analysis which indicates how sensitive the cost assessment is to some of
EPA's best estimates.

      D.  Chapter 5—Economic Impact of HD Otto-cycle Standards

      Chapter 5 presents EPA's best assessment of the economic impacts which will result from
the HD otto-cycle engine and vehicle standards. The assessment includes EPA's estimates of the
technology packages manufactures will use, as well as the costs associated with new certification
and compliance requirements. Costs are estimated on  a per-vehicle basis, as well as an aggregate
cost to society.

-------
Regulatory Impact Analysis
       E. Chapter 6—Environmental Impact of HD Diesel Standards

       Chapter 6 describes the expected environmental impacts of the HD diesel engine NMHC plus
NOx emissions standards described in the preamble for this rule. The modeling methodology and
assumptions used to estimate nationwide NOx and VOC emission inventories (i.e., tons of pollutant
per year) are described, and the estimated benefits are presented.  In addition, estimates of
nationwide PM inventories for HD diesel vehicles are presented.

       F. Chapter 7—Environmental Impact of HD Otto-cycle Standards

       Chapter 7 describes the expected environmental impacts of the exhaust and ORVR standards
for heavy-duty gasoline engines and vehicles described in the previous chapters.  Specifically, the
chapter includes a description of how heavy-duty gasoline vehicle emission factors were developed,
the per-vehicle exhaust emission reductions due to the standards over the life of heavy-duty gasoline
vehicles, the estimated exhaust NOx and NMHC emission inventories from heavy-duty gasoline
vehicles, and the exhaust emission benefits from the exhaust standards. The chapter also includes
a description of the emission benefits from the ORVR requirements for Class 2b heavy-duty gasoline
vehicles.

       G. Chapters—Cost-effectiveness for HD Diesel and Otto-cycle Requirements

       Chapter 8 presents EPA's estimated cost-effectiveness of the requirements for new heavy-
duty engines, including the new standards and related requirements, OBD, useful life, allowable
maintenance, in-use testing, and rebuild provisions. This analysis relies in part on cost information
from Chapters 4 and 5 and emissions information from Chapters 6 and 7 to estimate the cost-
effectiveness of the provisions in  terms of dollars per ton of total emission reductions.  Separate
analyses were performed for otto-cycle engines and diesel engines. Cost-effectiveness values are
presented on a per-vehicle basis using total costs and total NOx plus NMHC emission reductions
over the typical lifetime of a heavy-duty vehicle, discounted  at a rate of seven percent to the
beginning of the vehicle's life. Analyses of the fleet cost-effectiveness for 30 model years after the
new engine standards take effect are also presented.

-------
                                          Chapter 2: Health and Welfare Concerns
   CHAPTER  2:  HEALTH AND WELFARE  CONCERNS


I.  Health and Welfare Concerns

      A.  Health and Welfare Effects from NMHC and NOx

      Ground-level ozone, the main ingredient in smog, is formed by complex chemical reactions
of volatile organic compounds (VOC) and nitrogen oxides (NOx) in the presence of heat and
sunlight. Ozone forms readily in the lower atmosphere, usually during hot summer weather. VOCs
are emitted from a variety of sources, including motor vehicles, chemical plants, refineries, factories,
consumer and commercial products, and other industrial sources. VOCs also are emitted by natural
sources such as vegetation. NOx is emitted largely from motor vehicles, nonroad equipment, power
plants, and other sources of combustion.

      The science of ozone formation, transport, and  accumulation is complex.  Ground-level
ozone is produced and destroyed in a cyclical set of chemical reactions involving NOx, VOC, heat,
and sunlight." As a result, differences in NOx and VOC emissions and weather patterns contribute
to daily, seasonal, and yearly differences in ozone concentrations and differences from city to city.
Many of the chemical reactions that are part of the ozone-forming cycle are sensitive to temperature
and sunlight. When ambient temperatures and sunlight levels remain high for several days and the
air is  relatively stagnant, ozone and its precursors can build  up and produce more ozone than
typically would occur on a single high temperature day.  Further complicating matters, ozone also
can be transported into an area from pollution sources found hundreds of miles upwind, resulting in
elevated ozone levels even in areas with low VOC or NOx emissions.

      Based on a large number of recent studies, EPA has identified several  key health effects
caused when people are exposed to levels of ozone found today in many areas of the country.1'2
Unless noted otherwise, the studies discussed in the remainder of sections I(A) and I(B) of this
Chapter are from  references (1)  and (2), EPA Report EPA-452/R-96-007  and EPA Report
EPA/600/P-93/004aF  respectively.  Short-term  exposures (1-3 hours) to high ambient ozone
concentrations have been linked to increased hospital admissions and emergency room visits for
respiratory problems. For example, studies conducted in the northeastern U.S. and Canada show that
ozone air pollution is associated with 10-20 percent of all of the summertime respiratory-related
hospital admissions. Repeated exposure to ozone can make people more susceptible to respiratory
infection and lung inflammation and can aggravate preexisting respiratory diseases, such as asthma.
Exposure to ozone can  cause repeated inflammation of the lung,  impairment of lung  defense
mechanisms, and irreversible changes in lung structure, which could lead to premature aging of the
(a)     Carbon monoxide also participates in the production of ozone, albeit at a much slower rate than most VOC
       and NOx compounds.

-------
Regulatory Impact Analysis
lungs and/or chronic respiratory illnesses such as emphysema,  chronic bronchitis  and chronic
asthma.

       Children are most at risk from ozone exposure because they typically are active outside,
playing and exercising, during the summer when ozone levels are highest. For example, summer
camp studies in the eastern U.S. and southeastern Canada have reported significant reductions in
lung function in children who are active outdoors. Further, children are more at risk than adults from
ozone exposure because their respiratory systems are still developing. Adults who are outdoors and
moderately  active during the summer months, such  as construction workers and other outdoor
workers, also are among  those most at risk.  These individuals, as well as people with respiratory
illnesses such as asthma,  especially asthmatic children, can experience reduced lung function and
increased respiratory  symptoms, such as chest pain and cough, when exposed to ozone  during
periods of moderate exertion.

       Evidence also exists of a possible relationship between daily increases in ozone levels and
increases in daily mortality levels. While the magnitude of this relationship is still too uncertain to
allow for direct quantification, the full body of evidence indicates a likely  positive relationship
between ozone exposure  and premature mortality.

       In addition to the  effects  on human health, ozone is known to  adversely affect the
environment in many ways.  These effects include reduced yield for commodity crops, for fruits and
vegetables, and commercial forests; ecosystem and vegetation effects in such areas as National Parks
(Class I areas); damage to urban grass, flowers, shrubs, and trees; reduced yield in tree seedlings and
non-commercial forests; increased susceptibility of plants to pests; materials damage; and visibility.
Nitrogen oxides (NOx), a key precursor to ozone, also results in nitrogen deposition into sensitive
nitrogen-saturated coastal estuaries and ecosystems, causing increased growth of algae and other
plants.3 NOx also is a contributor to acid deposition, which can damage trees at high elevations and
increases the acidity of lakes and streams, which can severely damage aquatic life. Finally, NOx
emissions can contribute to  increased levels of particulate matter by changing into nitric acid in the
atmosphere and forming particulate nitrate.

       In addition to their  contribution to ozone levels, emissions of NMHC contain toxic air
pollutants that may have a significant effect on the public health, as discussed below.

       B.  Particulate Matter

       Particulate matter (PM) represents a broad class of chemically and physically diverse
substances that exist as discrete particles  (liquid droplets or solids) over a  wide range of sizes.
Human-generated sources of particles include a variety of stationary and mobile sources.  Particles
may be emitted directly to the atmosphere or may be formed by transformations of gaseous emissions
such as sulfur dioxide or nitrogen oxides.  The major chemical and physical properties of PM vary
greatly with time, region, meteorology, and source category, thus complicating the assessment of
health and welfare effects  as related to various indicators of particulate pollution.  At elevated

-------
                                             Chapter 2:  Health and Welfare Concerns
concentrations, particulate matter can adversely affect human health, visibility,  and materials.
Components of particulate matter (e.g., sulfuric or nitric acid) contribute to acid deposition.4

       Key EPA findings can be summarized as follows:

1.      Health risks posed by inhaled particles are affected both by the penetration  and deposition
       of particles in the various regions of the respiratory tract, and by the biological responses to
       these deposited materials.

2.      The risks of adverse effects associated with deposition of ambient particles in the thorax
       (tracheobronchial and alveolar regions of the respiratory tract) are markedly greater than for
       deposition in the extrathoracic (head) region. Maximum particle penetration to the thoracic
       regions occurs during oronasal or mouth breathing.

3.      The key health effects categories associated with PM include premature death; aggravation
       of respiratory and cardiovascular disease, as indicated by increased hospital admissions and
       emergency room visits, school absences, work loss days, and restricted activity days; changes
       in lung function and increased respiratory symptoms; changes to lung tissues and structure;
       and altered respiratory defense mechanisms. Most of these effects have been consistently
       associated with ambient PM concentrations,  which  have been used  as  a measure of
       population exposure, in a large number of community epidemiological studies. Additional
       information and insights on these effects are provided by studies of animal toxicology and
       controlled human exposures to various constituents of PM conducted at higher than ambient
       concentrations. Although mechanisms by which particles cause effects are not well known,
       there is general agreement that the cardio-respiratory system  is the major target of PM
       effects.

4.      Based on a qualitative assessment of the epidemiological evidence of effects associated with
       PM for populations that appear to be at  greatest risk with respect to  particular health
       endpoints, the EPA has concluded the following with respect to sensitive populations:

       a.      Individuals with respiratory disease (e.g., chronic obstructive pulmonary disease,
              acute bronchitis)  and  cardiovascular disease (e.g., ischemic heart  disease) are at
              greater risk of premature mortality  and hospitalization due to exposure to ambient
              PM.

       b.      Individuals with infectious respiratory disease (e.g., pneumonia) are at greater risk
              of premature mortality and morbidity (e.g., hospitalization, aggravation of respiratory
              symptoms) due to exposure to ambient PM. Also, exposure to PM may increase
              individuals' susceptibility to respiratory infections.

       c.      Elderly individuals are also at greater risk of premature mortality and  hospitalization
              for cardiopulmonary problems due to exposure to ambient PM.

-------
Regulatory Impact Analysis
       d.      Children are at greater risk of increased respiratory symptoms and decreased lung
              function due to exposure to ambient PM.

       e.      Asthmatic individuals are at risk of exacerbation  of symptoms associated with
              asthma, and increased need for medical attention, due to exposure to PM.

5.     There are fundamental physical and chemical differences between fine and coarse fraction
       particles and it is reasonable to expect that differences may exist between the two subclasses
       of PM10 in both  the nature of potential effects and the relative concentrations required to
       produce such effects.  The specific components of PM that could be of concern to health
       include components typically within  the fine fraction (e.g., acid aerosols, sulfates, nitrates,
       transition metals, diesel particles, and ultra fine particles), and other components typically
       within the coarse fraction (e.g.,  silica and resuspended dust). While components of both
       fractions can produce health effects, in general, the fine fraction appears to contain more of
       the  reactive  substances  potentially linked  to the kinds of effects  observed in  the
       epidemiological  studies. The fine fraction also contains the  largest number of particles and
       a much larger aggregate surface area than the coarse fraction which enables the fine fraction
       to have a substantially greater potential for absorption and deposition in the thoracic region,
       as well as for dissolution or absorption of pollutant gases.

       With respect to welfare or secondary effects, fine particles have been clearly associated with
the impairment of visibility over urban areas and large multi-state regions. Fine particles, or major
constituents thereof, also are implicated in materials damage, soiling and acid deposition.  Coarse
fraction particles contribute to soiling and materials damage.

       Paniculate pollution is a problem affecting localities, both urban and non-urban, in all regions
of the United States.  Manmade emissions that contribute to airborne particulate matter result
principally from stationary point sources (fuel  combustion and industrial  processes), industrial
process fugitive particulate emission sources, non-industrial fugitive sources (roadway dust from
paved and unpaved roads, wind erosion from crop land, etc.), and transportation sources.  In addition
to manmade emissions,  consideration must  also be given to natural emissions including dust,  sea
spray, volcanic emissions, biogenic emanation (e.g., pollen from plants), and emissions from wild
fires when assessing particulate pollution and devising control strategies.

II. Need for Ozone Control

       A. Standards for 2004 HD Diesels Are a  Key Part of State Air Pollution Control
Plans

       Since we published the final rule establishing the 2004 HD diesel emission standards in 1997,
those states with State Implementation Plans  (SIPs) have  considered the projected emission
reductions from these engines to be an important component of their overall SIPs.  The NOx and
NMHC nationwide emission reductions that will result from these standards beginning in the 2004
model year will help states to attain the ozone NAAQS.  These states have  incorporated  the

-------
                                           Chapter 2:  Health and Welfare Concerns
beneficial effects of the 2004 HD diesel standards into their air quality modeling and they continue
to count on the emission reductions from this program to meet their air quality goals.

       B. New Standards for 2005 HD Gasoline Engines and Vehicles Are Important
for States in Meeting Their Air Quality Goals

         Today,  many states are finding it difficult to  show how they can meet or maintain
compliance with  the current National Ambient Air Quality Standard (NAAQS) for ozone by the
deadlines established in the Clean Air Act. In December, 1999,92 million people (1990 population)
lived in 32 metropolitan areas designated nonattainment under the 1-hour ozone NAAQS.5

       There is a very clear risk that there will be elevated levels of ground-level ozone above the
1-hour NAAQS during the time period when the heavy-duty gasoline vehicle standards of this
rulemaking will take effect.  The reductions in oxides of nitrogen (NOx) and volatile organic
compounds (VOCs) projected from the  proposed new standards will benefit public health and
welfare by reducing ozone levels. This assessment is based upon our recent and extensive ozone air
quality modeling  and analysis performed for the Tier 2/Gasoline Sulfur rulemaking, which predicts
that a significant  number of areas across the nation are at  risk of failing to meet the 1-hour ozone
NAAQS  even with Tier 2 and  other controls currently in place.  Because ozone concentrations
causing violations of the 1-hour ozone standard are well established to endanger public health and
welfare, we conclude that today's new standards for 2005 and later gasoline heavy-duty vehicles are
warranted.

             1.  Projected Air Quality Problems Remain After Tier 2/Gasoline Sulfur
             Program Is in  Place

       In conjunction with our Tier 2/Gasoline Sulfur rulemaking efforts, we performed ozone air
quality modeling for nearly the  entire eastern U.S. covering metropolitan areas from Texas to the
Northeast, and for a western U.S. modeling domain.  As a  part of this modeling, we considered the
air quality of these areas after the Tier 2/Gasoline Sulfur program is implemented.  This modeling
predicted that without further emission reductions, a significant number of areas now experiencing
ozone exceedances across the nation are at risk of failing to meet the 1-hour ozone NAAQS in 2004
and beyond, even with the Tier 2/Gasoline Sulfur program and other current controls in place.

       The general pattern that  the ozone modeling shows is a broad reduction between 1996 and
2007 in the geographic extent  of ozone concentrations  above the 1-hour NAAQS,  and in  the
frequency and severity of exceedances. In the absence of additional controls beyond those that will
be achieved by current control  programs — including the Tier 2/Gasoline Sulfur program - we
expect there  will be  a slight  decrease  below 2007 ozone concentrations and  frequencies of
exceedances in 2030. However,  the general trends and modeling results show that many of the areas
we modeled may  have exceedances continuously throughout the period from 2007 to 2030 without
further reductions in emissions.  Others may briefly attain and then return to nonattainment by 2030
or earlier. Although for practical reasons we limited our modeling of ozone concentrations to 1996,
2007, and 2030,  we expect that concentrations between  2007 and 2030 will generally track the

-------
Regulatory Impact Analysis
national emissions trend, showing a period of improvement after 2007 followed by a reversal of the
trend and deterioration back towards the 2007 levels.  Because individual areas' emissions trends
differ, we expect that the air quality of individual areas will also vary from this general pattern.

       We believe that there is a risk that future air quality in each of these areas would exceed the
ozone standard during the time period when this rule will take effect. This belief is based on three
factors: (1) recent exceedances in 1995-1997 or 1996-1998, (2) predicted exceedances in 2007 or
2030 after accounting for reductions from Tier 2 and other local or regional controls currently in
place or required, and (3) our assessment of the magnitude of recent exceedances, the variability of
meteorological conditions,  transport from areas with  later attainment dates, and other variables
involved in predicting  future attainment such as  the potential for some areas to experience
unexpectedly high economic growth rates,  growth in vehicle miles traveled, varying population
growth from area to area, and differences in vehicle  choice.

       Based on the Tier 2 modeling analyses and information from recently-submitted SIPs, we
have determined that over 71 million people (1996 population) in 21 metropolitan areas are likely
to be exposed to unhealthy levels of ground level ozone at some point in time between 2004 and
2030 without significant additional controls. These 21 areas are those that currently violate the 1-
hour ozone NAAQS and are predicted by the best ozone modeling we have available to exceed the
1-hour ozone standard without significant new controls.  This analysis accounts for the expected
benefits from the Tier 2 program and other control programs already in place.b It does not include
additional control measures that states would need to implement to meet their requirements under
the recently proposed SIP findings.  Table 2.1 lists these 21 areas.
(b) Air quality modeling shows that improvements in ozone levels can be expected to occur throughout the country
because of the Tier 2/Gasoline Sulfur program. EPA found that the program significantly lowers the model-
predicted number of exceedances of the ozone standard by one tenth in 2007, and by almost one-third in 2030 (Tier
2/Gasoline Sulfur Final RIA, Docket A-97-10, Document Number V-B-1).

-------
                                          Chapter 2: Health and Welfare Concerns
                                      Table 2.1
  Twenty-one Metropolitan Areas Likely to Exceed the 1-Hour Ozone NAAQS in 2004 or
                  Thereafter Without Additional Emission Reductions
Metropolitan Area
   1996
Population
 (millions)
Baltimore-Hagerstown, MD PSMA                                             2.6
Barnstable-Yarmouth, MA MSAa                                               0.2
Baton Rouge, LA MSA                                                        0.6
Beaumont-Port Arthur, TX MSA                                               0.4
Birmingham, AL MSA                                                        0.9
Boston-Worcester-Lawrence, MA-NH-ME-CT CMSAa                            5.6
Chaiiotte-Gastonia-Rock Hill, NC-SC MSAa                                     1.3
Dallas-Fort Worth, TX CMSA                                                 4.6
Houma, LA MSAa                                                           0.2
Houston-Galveston-Brazoria, TX CMSA                                        4.3
Huntington-Ashland, WV-KY-OH MSAa                                        0.3
Indianapolis, IN MSAa                                                        1.5
Los Angeles-Riverside-San Bernardino CA CMSA                                15.5
Louisville, KY-IN MSA                                                       1.0
Memphis, TN-AR-MS MSAa                                                   1.1
Nashville, TN MSAa                                                          1.1
New York-Northern New Jersey-Long Island, NY-NJ-CT-PA CMSA                19.9
Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD CMSA                     6.0
Providence-Fall River-Warwick, RI-MA MSAa                                   1.1
Richmond-Petersburg, VA MSAa                                               0.9
St. Louis, MO-IL MSA                                                        2.5
                                                 TOTAL POPULATION      71.5
a. The 1-hour ozone NAAQS does not currently apply, but we have proposed to re-instate it.

       There are 14 additional metropolitan areas, with another 35 million people in 1996, for which
the available ozone modeling and other evidence is less clear regarding the need for additional
reductions.  Table 2.2 lists the areas we put in this second category.  Our own ozone modeling
predicted these 14 areas to need further reductions to avoid exceedences during the period when the
standards are effective.

-------
Regulatory Impact Analysis
                                         Table 2.2
Fourteen Metropolitan Areas With Some Risk of Exceeding the 1-Hour Ozone Standard in
               2004 or Thereafter Without Additional Emission Reductions.
Metropolitan Area
Atlanta, GA MSA
Benton Harbor, MI MSAa
Biloxi-Gulfport-Pascagoula, MS MSAa
Chicago-Gary-Kenosha, IL-IN-WI CMSA
Cincinnati-Hamilton, OH-KY-IN CMSAC
Cleveland-Akron, OH CMSAa
Detroit-Ann Arbor-Flint, MI CMSAa
Grand Rapids-Muskegon-Holland, MI MSAa
Greater Connecticut, CT-RI CMSAb
Milwaukee-Racine, WI CMSA
New Orleans, LA MS Aa
Pensacola, FL MSAa
Tampa, FL MSAa
Washington, DC MSA
1996 Population
(millions)
3.5
0.2
0.3
8.6
1.9
2.9
5.3
1.0
2.4
1.6
0.3
0.4
2.2
4.6
TOTAL POPULATION 35.3
a. The 1-hour ozone NAAQS does not currently apply, but we have proposed to re-instate it.
b. For the New London-Norwich portions of this area, the 1 -hour ozone NAAQS does not currently apply, but we have
proposed to re-instate it.
c. Based on more recent air quality monitoring data not considered in the Tier 2 analysis, and on ten year emissions
projections, we expect to re-designate Cincinnati-Hamilton to attainment soon.

       For all of these areas, recent air quality monitoring data indicate that exceedances may occur
in 2007 or 2030. Eight areas have recent exceedances, but local ozone modeling and other evidence
indicates attainment in 2007.c  Based on this evidence, we have kept these areas separate from the
previous set of 21 areas.  However, we still consider there to be some risk of future exceedances for
these eight areas.

       For the other six  of the 14  areas/ the air quality monitoring data shows current attainment
but with less than a 10 percent margin below the NAAQS. This suggests that these areas may remain
without exceedances for  some time, but that there is still a risk of future exceedance of the NAAQS
due, for example, to meteorological conditions that may be more severe in the future.
(c) Atlanta, Benton Harbor, Chicago, Cincinnati, Grand Rapids, Greater Connecticut, Milwaukee, and Washington,
DC. (The Cincinnati-Hamilton area did not have exceedances in the 1996-1998 period).

(d) Biloxi, Cleveland, Detroit, New Orleans, Pensacola, and Tampa.

                                                                                          10

-------
                                           Chapter 2: Health and Welfare Concerns
       There is significant risk that at least some of these 35 areas will violate the NAAQS in 2004
or thereafter without additional reductions.  We consider the situation in these areas to support our
belief that, overall, additional reductions are needed.

             2.   Today's Program Will Help  Areas Meet  Their Attainment and
       Maintenance Requirements

       The HD gasoline vehicle standards finalized today, and the HD diesel standards reviewed
today, will help all of the areas discussed above to either meet their attainment deadlines, to maintain
attainment in the future, or both. The new program will be very important to each of the areas with
deadlines in 2005 and later that will require (or may require) additional emission reductions (2005
is the year that new gasoline HD vehicles will begin to enter the fleet). As Table 2.3 shows, there
are  10  such areas with almost 66 million people. The following table lists these areas and their
expected attainment dates:
                                                                                    11

-------
Regulatory Impact Analysis
Table 2.3
Areas with 2005, 2007, or 2010 Attainment Deadlines
Metropolitan Area
Baltimore
Philadelphia
Greater Connecticut
(Hartford and other
MSAs)
New York City, NY-
NJ-CT
Houston, TX
Chicago, IL-IN
Milwaukee, WI
Dallas, TX
Beaumont-Port
Arthur, TX
Los Angeles (South
Coast Air Basin), CA

Attainment
Deadline
2005
2005
2007 (requested
extension)
2007
2007
2007
2007
2007 (requested
extension)
2007 (requested
extension)
2010

Modeling
Predictions
VOC Shortfall
NOx and VOC
Shortfall
Contingent on New
York Attainment
VOC and NOx
Shortfall
NOx Shortfall
Regional modeling to
analyze existence of
shortfall is underway
Regional modeling to
analyze existence of
shortfall is underway
Local modeling
shows nonattainment
in 2007
Local modeling
shows nonattainment
in 2007
Approved SIP with
commitments for
unspecified
additional controls

Population
2.6
6.0
2.4
19.9
4.3
8.6
1.6
4.6
0.4
15.5
65.9
       All of the areas in Table 2.3 with 2005 or later attainment deadlines will be able to take credit
in their attainment demonstrations (or  in  revisions  to their  demonstrations) for the expected
reductions for the preexisting standards forHD diesel engines and for today's new standards for HD
gasoline  engines and vehicles.   (EPA  has not approved  deadline extensions  for Dallas and
                                                                                     12

-------
                                            Chapter 2: Health and Welfare Concerns
Beaumont/Port Arthur at this time; if their requested extensions (to 2007) are approved, these areas,
too, could take credit for today's program).  The ability to take credit for the new HD gasoline
vehicle standards will be especially important for the several areas with emission "shortfalls" (i.e.,
those for which we have made our proposal to approve their attainment demonstrations contingent
on their adoption of new measures for further emission reductions).

       In addition to helping 8 areas from Table 2.3 meet their attainment deadlines (plus Dallas and
Beaumont/Port Arthur if they receive a deadline extension to 2005 or later), the new program will
help these and all other areas with current or potential future ozone problems to maintain their
attainment into the future. This includes at least the 37 areas we expressed concern about earlier.
In effect, the emission reductions of this program will reduce the risk that these areas that today are
in or approaching attainment will fall back into nonattainment as they face economic development
and growth in vehicle travel.

             3.  The  Program Will Help States Avoid  More  Costly Measures

       In general, the task of states to reach and/or maintain attainment will be  easier and the
economic impact on their industries and citizens will be lighter if, as a result of today' s new gasoline
HD vehicle standards,  they are able to forego other, less cost effective programs.  Following
implementation of the Regional Ozone Transport Rule, states will have already adopted emission
reduction requirements for nearly all large sources of VOC andNOx for which cost-effective control
technologies are known and for which they have authority to control.   Those that remain in
nonattainment therefore will have to consider their remaining alternatives.

       Thus, the emission reductions from the standards we are finalizing today will help states that
would otherwise need to seek controls for the first time from the sources that have not yet been
controlled — mostly smaller sources including area sources that are closely related to the activities
of individuals and small businesses. The emission reductions from today's standards will also help
states prevent or  delay deeper reductions from large  and  small sources that have previously
implemented emission controls.

       C.    HD Diesel and Gasoline  Engines Contribute to Total  NOx and VOC
             Emissions

       HD engines and vehicles are major contributors to nationwide emissions of NOx and they
are moderate contributors to nationwide  emissions of VOC (estimates of a geographic area's
emissions are called "emission inventories").  Table 2.4 summarizes EPA's current estimates for
national NOx and VOC contributions from maj or mobile source categories. (See Chapter 6 for more
information about how we developed these values.)
                                                                                     13

-------
Regulatory Impact Analysis
                                       Table 2.4
          2000 National NOx and VOC Emissions, (thousand short tons per year)
Emission Source
Light-Duty Vehicles
Heavy-Duty Vehicles
Nonroad Engines and Vehicles
Other (Stationary Point and Area Sources)
Total Nationwide Emissions
NOx
4,420
3,759
5,343
10,656
24,178
NOx
%
18%
15%
22%
44%

VOC
4,098
355
2,485
9,567
16,505
VOC %
25%
2%
15%
58%

       Table 2.4 indicates that HD gasoline and diesel vehicles currently represent about 15 percent
of national NOx emissions and two percent of national VOC emissions.  Moreover, the local heavy-
duty vehicle NOx contributions are higher than the national average in many important urban areas.
Table 2.5 shows projections of local contributions to NOx from HD vehicles in 2007 in several
metropolitan areas with ozone concerns. In each of these cases, the local contributions  of NOx are
greater than the projected national contribution — in several cases, significantly higher.

                                       Table 2.5
            2007 Heavy-Duty Vehicle Contribution to Urban NOx Inventories
Metropolitan Statistical Area
National
Atlanta
Dallas
Charlotte
Washington
Los Angeles
New York
Philadelphia
Cleveland
Contribution to
Total NOx
72%
18%
17%
15%
15%
15%
14%
13%
13%
Contribution to
Mobile Source
NOx
23%
28%
21%
27%
29%
19%
23%
23%
23%
       Chapters 6 and 7 of this RIA also present updated emission inventory modeling for HD
vehicles in future years.  The results show that without additional HD NOx control beyond the 1998
standards, national NOx emissions from HD vehicles would decline for the next few years but that
this trend would reverse around 2006.  After that, without additional emission controls, NOx
emissions from the HD vehicle fleet would again increase as a result of future growth in the HD
                                                                                     14

-------
                                             Chapter 2:  Health and Welfare Concerns
vehicle market.  A similar trend is seen for national NMHC emissions from HD vehicles — we
project that NMHC emissions will decrease until around 2009, after which growth in numbers of
vehicles will offset emission reductions and NMHC emissions from HD vehicles will increase.6

III.  Participate Matter

       A. Current and Future Compliance with the PM10 NAAQS

       Compliance with the current PM10 standard continues to be a problem. This section reviews
our most recent analyses regarding PM10 air quality, discussed in detail in the Regulatory Impact
Analysis for the Tier 2/Gasoline Sulfur final rule.6 The most recent PM10 monitoring data indicates
that 15 counties designated PM10 nonattainment counties, with a population of 8.6 million in 1996,
violated the PM10 NAAQS in the period 1996-1998.  The areas that are violating do so because of
exceedances of the 24-hour PM10 NAAQS.  No areas had monitored violations of the annual
standard in this period. Table 2.6 lists these 15 counties. The table also indicates the classification
for each area, the status of our review of the SIPs, and population for each area in 1996.
(e)  The emission inventory modeling we performed for this rule includes the excess emissions that occurred as a
result of certain HD diesel engines manufactured between 1988 and 1998. These engines were at issue in the
"consent decrees" involving certain HD diesel engine manufacturers, as discussed in Section I.C. of the preamble
for this final rule.
                                                                                       15

-------
Regulatory Impact Analysis
                                       Table 2.6
           PM10 Nonattainment Areas Violating the PM10 NAAQS in 1996-1998
Area
Clark Co., NV
El Paso, TX
Gila, AZ
Imperial Co., CA
Inyo Co., CA
Kern Co., CA
Mono Co., CA
Kings Co., CA
Maricopa Co., AZ
Power Co., ID
Riverside Co., CA
San Bernardino Co., CA
Santa Cruz Co., AZ
Tulare Co., CA
Walla Walla Co., WA

Classification
Serious
Moderate
Moderate
Moderate
Moderate
Serious
Moderate
Serious
Serious
Moderate
Serious
Serious
Moderate
Serious
Moderate

SIP Approved?
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
Yes
TOTAL POPULATION
1996
Population
(millions)
0.93
0.67
0.05
0.14
0.02
0.62
0.01
0.11
2.61
0.01
1.41
1.59
0.04
0.35
0.05
8.61
       Using a PM10 modeling approach conducted for the Tier 2 rulemaking, we have concluded
that the 8 areas shown in Table 2.7 have a high risk of failing to attain and maintain the PM10
NAAQS without further emission reductions.  These areas have a population of nearly 8 million.
Included in the group are the counties that are part of the Los Angeles, Phoenix, and Las Vegas
metropolitan areas, where traffic from heavy-duty vehicles is substantial.  These 8 areas would
clearly benefit from the reduction in emissions which would occur from the new standards for heavy-
duty vehicles.
                                                                                    16

-------
                                            Chapter 2:  Health and Welfare Concerns
                                       Table 2.7
                Areas With High Risk of Failing To Attain and Maintain
                 the PMio NAAQS Without Further Emission Reductions
                             Area                                  1996 Population
                            	(millions)
 Clark Co., NV                                                           0.93
 Imperial Co., CA                                                        0.14
 Kern Co., CA                                                           0.62
 Kings Co., CA                                                           0.11
 Maricopa Co., AZ                                                        2.61
 Riverside Co., C A                                                        1.41
 San Bernardino Co., CA                                                  1.59
 Tulare Co., CA                                                          0.35
 TOTAL POPULATION                                                    7.76
       Table 2.7 is limited to designated PM10 nonattainment areas which both had monitored
violations of the PM10 NAAQS in 1996-1998 and are predicted to be in nonattainment in 2030 in our
PM10 air quality modeling.  This gives us high confidence that these areas require further emission
reductions to attain and maintain, but does not fully consider the possibility that there are other areas
which are now meeting the PM10 NAAQS which have at least a significant probability of requiring
further reductions to continue to maintain it.

       In our Tier 2 analysis, we created a second category of areas with a risk of failing to attain
the PM10 NAAQS in the future that did not rise to the level of risk attributed to those areas listed in
Table 2.7. The Tier 2 air quality modeling predicted that even considering the emission reductions
from the Tier 2/Gasoline Sulfur  program and  other controls already in place, there would be
violations in 2030 of the annual average PM10 NAAQS in five additional counties. We chose these
five counties because they registered, in either 1997 or 1998, single-year annual average monitored
PM10 levels of at least  90 percent of the NAAQS, but did not exceed the formal definition of the
NAAQS over the three-year period ending in 1998.   Table  2.8 shows these areas. They have a
combined population of almost  17 million, and a broad geographic spread.
                                                                                     17

-------
Regulatory Impact Analysis
                                       Table 2.8
                  Five areas with a significant risk of failing to maintain
                the PM10 NAAQS without further reductions in emissions

New York Co., NY
Cuyahoga Co., OH
Harris, Co., TX
San Diego Co., CA
Los Angeles Co., CA

Area





TOTAL POPULATION
1996 Population
(millions)
1.33
1.39
3.10
2.67
8.11
16.6
       Unlike the situation for ozone, for which precursor emissions are generally declining over
the next 10 years or so before beginning to increase, we estimate that emissions of PM10 will rise
steadily unless new controls are implemented.  The small margin of attainment which these areas
currently enjoy will likely erode; the PM air quality modeling suggests that it will be reversed. We
therefore consider these areas to each individually have a significant risk of failing to maintain the
NAAQS without further emission reductions. There is a substantial risk that at least some of them
would fail to maintain attainment without further emission reductions. The emission reductions
from the new standards for heavy-duty vehicles will help to keep them in attainment.

       B.    HD Diesel and Gasoline Engines Contribute to Secondary Formation of
             Particulate Matter

       Because we are not changing the paniculate matter emission standards for HD vehicles in
this rule, the effect of this rule on PM results primarily from reductions in NOx emissions and in turn
reductions  in the secondary formation  of nitrate particles in the atmosphere.   Most available
modeling of PM emissions, however, focuses only on direct (primary)  emissions of PM.

       We have not attempted to quantify the contribution of HD vehicles to the secondary nitrate
particles formed from the large NOx emissions of these vehicles in this final rule. We are convinced
that this contribution is substantial, especially in regions of the country where ammonia levels in the
air are relatively high (NOx reacts with ammonia to form ammonium nitrate particles). Similarly,
we believe that the very significant NOx reductions from HD diesel and gasoline vehicles that will
result from the 2004 standards will also result in important reductions in the HD contribution to
nitrate PM.

IV.  Air Toxics from  HD Engines and Vehicles

       In addition to contributing to the health and welfare problems associated with exceedances
of the National Ambient Air Quality Standards for ozone and PM10, emissions from HD diesel and
gasoline vehicles include a number of air pollutants that increase the risk of cancer or have other

                                                                                     18

-------
                                            Chapter 2:  Health and Welfare Concerns
negative health  effects.   These air  pollutants include benzene, formaldehyde, acetaldehyde,
1,3-butadiene, and diesel particulate matter. For several of these pollutants, motor vehicle emissions
are believed to account for a significant proportion of total nation-wide emissions.  All of these
compounds are products of combustion; benzene is also found in non-exhaust emissions from
gasoline-fueled vehicles. The reductions in hydrocarbon emissions from HD vehicles resulting from
today's program will further reduce the potential cancer risk and other health risks from these air
toxics (other than diesel PM) because many of these pollutants are themselves VOCs.

       Diesel engine particulate matter  is  also  a  potential concern because of  its possible
carcinogenic and mutagenic effects on people. However, because today's program does not include
more stringent standards for emissions of diesel PM, this action will not make a large difference in
any health effects from direct diesel PM.

       We are addressing the issues raised by air toxics from motor vehicles and their fuels in a
separate rulemaking that we recently proposed and was signed by the Administrator on July 14,2000
under section 202(1)(2) of the Act. That rulemaking process will address the emissions of hazardous
air pollutants from motor vehicles and fuels,  and the appropriate level of control of hazardous air
pollutants from these sources.
                                                                                      19

-------
Regulatory Impact Analysis
Chapter 2 References

1.     U.S. EPA, 1996, Review of National Ambient Air Quality Standards for Ozone,
      Assessment of Scientific and Technical Information, OAQPS Staff Paper, EPA-452/R-
      96-007.

2.     U.S. EPA, 1996, Air Quality Criteria for Ozone and Related Photochemical Oxidants,
      EPA/600/P-93/004aF.

3.     Vitousek, Pert M., John Aber, Robert W. Howarth, Gene E. Likens, et al. 1997. Human
      Alteration of the Global Nitrogen Cycle: Causes and Consequences.  Issues in Ecology.
      Published by Ecological Society of America, Number 1, Spring 1997.

4.     U.S. EPA, 1996, Air Quality Criteria for Particulate Matter, EPA/600/P-95/001aF.

5.     Memorandum to the Docket, Drew Kodjak, EPA, January 12, 2000 (found in the docket
      for this rule as well). Information on ozone nonattainment areas and population as of
      December 13, 1999.

6.     Regulatory Impact Analysis for the Tier 2/Gasoline Sulfur final rule, available in Docket
      A-97-10 (document number V-B-1) and through the Office of Transportation and Air
      Quality Tier 2 web  page at www.epa.gov/otaq/tr2home.htm.
                                                                                   20

-------
                                              Chapter 3:  Technological Feasibility
  CHAPTER 3:   TECHNOLOGICAL FEASIBILITY OF HD

           DIESEL AND OTTO-CYCLE  STANDARDS

I. Overview

      This chapter provides a technical discussion on emission related control technologies for
lower emissions from HD diesel and otto-cycle engines and vehicles. The chapter is divided into
two sub-chapters, the first dedicated to diesel controls, the second to otto-cycle controls.  In addition,
the final section discusses on-board diagnostics for HD, both  diesel and otto-cycle.

II. Diesel Engine Technologies

      A. HD Diesel Technology Overview

      This sub-chapter presents an assessment of emission control strategies that EPA expects will
be available for diesel engine manufacturers to use to meet the emission standards contained in this
final rule. In addition, we present a review of the technologies examined which provided the basis
on which we reviewed the HD 2004 NMHC+NOx standards  which were established in 1997.  To
meet the 1998 emissions standards for heavy-duty diesel engines, manufacturers have implemented
high-pressure fuel injection systems with retarded injection strategies, waste-gated turbo-chargers,
air-to-air after-coolers, advanced combustion chamber designs, and electronic controls. EPA expects
that incremental improvements will occur with respect to these strategies, but EPA does not expect
that improvements in these strategies alone will achieve the levels required by the new standards.
To meet these levels, EPA expects that, in addition to the aforementioned strategies, manufacturers
will utilize exhaust gas re-circulation (EGR), fuel injection rate shaping, and possibly exhaust after-
treatment. It is important to note that each of these technologies should be able to achieve emission
reductions over a broad range of in-use operating conditions. Of these, EGR is expected to achieve
most of the necessary reductions.  As is discussed in more detail below, EGR has been  shown to
reduce NOx emissions by up to 90 percent under laboratory conditions. Because these future
emission control strategies will rely on electronic controls for adequate performance, EPA expects
that the best available on-board diagnostics will be implemented to ensure that  these strategies
remain effective in-use. Furthermore, although changes in diesel fuel composition might be required
to enable certain emerging aftertreatment technologies, no change in diesel fuel composition will
be required to meet the new standards.  In addition, the current status of technologies which EPA
does not believe will be necessary for the requirements contained in this final rule, but which could
provide additional emission reductions beyond the final rule requirements are discussed (these
technologies include NOx absorber catalysts, urea-based SCR systems, and PM traps).

      This sub-chapter is divided into five sections: EGR, fuel inj ection rate shaping, exhaust after-
treatment, fuel composition, and new test cycles.  Several sections also discuss  strategy-enabling
technologies such as variable geometry turbo-chargers (VGT) for driving EGR, or common rail fuel
systems for performing fuel injection rate shaping.
                                                                                 21

-------
Regulatory Impact Analysis
       The RIA for the  1997 HD rulemaking  contains additional information  regarding the
effectiveness of several of the technologies discussed here, primarily cooled EGR systems. The
conclusions in the 1997 rulemaking regarding the effectiveness of cooled EGR for the reduction of
NOx emissions from HD diesel engines continue to be relevant, but the analysis which led EPA to
the conclusion that cooled EGR would be the principle technology for meeting the standards will not
be repeated here. The reader should refer to Chapter 4 of the 1997 RIA for additional discussion of
EGR  systems beyond what is covered in this RIA.  In addition, a discussion  of the potential
incremental improvements from control strategies already being used to meet the 1998 standards can
be found in the RIA of the 1997 final rule.

       B. Exhaust Gas Re-circulation (EGR)

       EGR is the re-circulation of exhaust gas from a point in an engine's exhaust system to a point
in its intake system. EGR is used to decrease nitric oxide (NO) emissions, the primary species in
diesel oxides of nitrogen (NOx). EGR dilutes intake air with combustion products, namely carbon
dioxide (CO2) and  water vapor.  These  diluents decrease the adiabatic stoichiometric flame
temperature  for  a  given mass of  fuel and  oxygen burned.1   This decrease in temperature
exponentially decreases the oxidation rate of dissociated nitrogen (N) to nitric oxide (NO).2 EGR
also decreases the overall mole fraction of oxygen, which proportionally decreases the oxidation rate
of N to NO.3 Finally, the specific heats of CO2 (above 532° K) and water vapor are greater than that
of air; therefore they absorb more heat per increase in temperature than air, thus lowering the peak
temperature  for a given release of heat. This last effect on NO  formation, however,   is small
compared to the first two.4

       EGR is very effective at decreasing NOx.  Laboratory studies have shown that EGR can
reduce NOx  emissions by up to 90 percent at light load and up to 60 percent at full load near rated
speed.5  These  studies and others have shown similar reductions at other speeds and loads.6
However, because EGR decreases the overall rate of combustion in the cylinder,  EGR tends to
increase  particulate matter (PM) emissions  and brake  specific  fuel  consumption (BSFC).
Furthermore, if EGR is not cooled before it is introduced  to the intake system, it will reduce the
density of the intake charge, and thus decrease the volumetric efficiency of the engine, which will
decrease maximum power and increase BSFC. Hot EGR also offsets EGR's beneficial effect on
combustion temperature because hot EGR increases the initial temperature of the air charge. Finally,
EGR without additional boost air decreases the excess air ratio.  This can result  in incomplete
combustion during some modes of operation and an increase in PM emissions. Through proper EGR
system design, however, researchers have demonstrated that these undesirable effects of EGR can
be minimized so that the 2004 emission standards can be met.7

       From a design perspective, there have been several challenges to EGR's feasibility, all of
which have been addressed.  First, a sufficient positive pressure difference must exist between the
point in the exhaust system where the exhaust gas is extracted and the point in the intake system
where it is introduced.  Second, under most conditions, EGR should be cooled for best performance,
which raises corrosion, fouling and design issues.  Third, the rate of EGR must be  controlled
accurately, and the control system must respond quickly to changes in engine operation.8

                                                                                    22

-------
                                                Chapter 3:  Technological Feasibility
       The positive pressure difference required to drive EGR may be achieved a number of ways.
Extracting the exhaust gas downstream of the exhaust turbine and introducing it to the inlet of the
intake compressor is called Low-Pressure-Loop (LPL) EGR.  LPL EGR possesses the advantage of
having a sufficient pressure differential to drive EGR over a wide engine operating range, but LPL
EGR may cause durability problems with the intake compressor and after-cooler.9'10  However,
through confidential  discussions with  engine manufacturers,  EPA has  learned  that  some
manufacturers may have overcome these durability issues at least for light and medium heavy-duty
engine applications.

       Another way of performing EGR is by extracting the exhaust gas from the exhaust manifold
and routing it to the intake manifold.  This minimizes the durability issues associated with the LPL
method by introducing the EGR after the compressor and after-cooler. This is called High-Pressure-
Loop (HPL)  EGR. HPL  EGR short-circuits the compressor  and after-cooler, but the required
pressure differential is difficult to achieve at high load, and particularly in heavy-duty  engine
applications.11  To improve the  pressure  differential to  enable HPL EGR, researchers have
investigated enabling technologies such as exhaust back-pressure valves, variable geometry turbo-
chargers (VGTs), and full-flow and bypass intake Venturis.  In three different studies investigators
positioned exhaust back-pressure valves downstream of the exhaust turbine to drive HPL EGR.
Researchers reported significant NOx reductions, but the turbo-chargers extracted less energy in
these configurations, and the re-circulated exhaust displaced fresh air without any increase in charge
air pressure.12  Unacceptable increases in PM  and BSFC resulted due to decreased excess air
ratios.13'14'15

       Two recent studies concluded that turbo-charger nozzle geometry must vary in order to drive
EGR without unacceptable decreases in excess air ratios,16'17 and a third  study investigated the
application of a by-pass venturi to draw exhaust  gas into the intake system.18 A variable geometry
turbocharger (VGT) is a turbocharger that has adjustable turbine inlet nozzle vanes.  Closing these
vanes decreases the nozzle area, whereby exhaust back pressure is increased to drive EGR, while
simultaneously, the turbine and compressor work are increased, as well as the compressor pressure
ratio. VGTs have been demonstrated to drive EGR without significantly decreasing excess air ratios.
In fact, under some operating conditions researchers achieved simultaneous decreases in NOx, PM
and BSFC by driving HPL EGR with a VGT.19  One study combined a VGT with a full-flow EGR
venturi that was positioned within the intake system just upstream of the intake manifold. On a 12
liter 315 kW heavy-duty diesel, the venturi increased EGR suction pressure by up to 20 kPa with an
intake pressure recovery of 60% downstream of the venturi .20  Because the venturi restricted airflow,
it caused decreased excess air ratios which resulted in increased PM and  BSFC. However, the
venturi can significantly extend the range of EGR flow, and it might improve the durability of a VGT
by allowing the VGT to operate at lower back pressures and temperatures.21 Another variation of
the venturi concept that does not employ a VGT is the bypass venturi. In this system EGR is
introduced into a venturi positioned in an intake duct that flows parallel to another intake duct in
which there is a controllable butterfly valve. By closing the butterfly valve in the one duct, more
intake flow is forced through the venturi's duct, which causes more EGR to be drawn into the intake
flow.22 Results from this configuration indicated that about 30% reductions in NOx were achievable
                                                                                     23

-------
Regulatory Impact Analysis
with no significant increase in BSFC or PM over a wide range of operating conditions. Further
decreases in NOx were achievable with some increase in PM and only a slight increase in BSFC.23

       Another important enabling technology for EGR is effective and durable EGR coolers. As
mentioned  previously, cooled EGR is desirable under most operating conditions to maximize
volumetric efficiency and to lower intake charge temperatures.  Studies have indicated that the issues
concerning EGR coolers - namely, corrosion, fouling, and compact design - have been resolved.

       Corrosion is an issue because current on-highway diesel fuel contains up to 0.04% fuel sulfur
(S) by weight, which forms corrosive sulfuric acid (H2SO4) in diesel exhaust. During combustion
S is oxidized 97-99%24 to sulfur dioxide (SO2) and trace amounts of sulfate (SO3).  SO3 also forms
in the exhaust manifold as equilibrium thermodynamics begin to favor its formation below -730 C.
Reaction kinetics limit SO3's formation rate, however.25 In diesel exhaust SO3 immediately reacts
with water  vapor to form aqueous sulfuric acid (-73% H2SO4 by wt.),26 and this acid begins to
condense from about 80  to 145 C ,27'28 depending upon engine operating conditions. Although the
acid's concentration is strong, the acid at this point only accounts for -0.5% of  the fuel sulfur.
However, once the exhaust cools below the water vapor dew point (-30 to 80 C),  SO2, which
accounts for nearly all of the fuel sulfur, will begin to react significantly with condensed water to
form H2SO4.29 For this reason, EPA expects that EGR coolers will utilize engine coolant, which is
thermostatically controlled typically between 80-90 C. This will help to prevent EGR cooling below
the exhaust's water vapor dew point.  Because -0.5% of the <0.04% S in the fuel may condense as
strong sulfuric acid and because additional H2SO4 may form during cold engine operating conditions
(start-up, idle, cool-down,  winter conditions), stainless steels with special corrosion resistance to
sulfuric acid have been selected to resolve the corrosion issue.30'31

       EGR cooler fouling can be minimized if the cross-sectional area of the exhaust channel can
be designed sufficiently  large. This  is generally problematic because this design leads to a large
EGR cooler. However, one heat exchanger manufacturer has implemented a heat exchanger channel
design that  simultaneously minimizes fouling while increasing heat transfer, thereby reducing the
EGR cooler size. The design implements winglets that are vortex-generators arranged in pairs in the
gas channel. They are opened in a V-shaped configuration in the direction of flow.32 These winglets
increase turbulence, which increases heat transfer by reducing the thermal  boundary layer in the
channel. They also decrease fouling  by forcing particles and vapors back toward the center of the
tube.  This stable, turbulent action counters thermophoretic deposition, condensation, and diffusion
due to a concentration  gradient.33 Experimental results indicate that cooler fouling stabilized after
100 hrs, and that in the end, fouling decreased cooler efficiency by only 15%.

       Controlling EGR flow rate is a crucial aspect for successful EGR system design. EPA
expects manufacturers to  make full use of an engine's electronic control system to measure
parameters  that should be used to control EGR rate.  Many  of these parameters,  such  as engine
speed, fuel rate, manifold pressures, temperatures and flow rates, are already being measured. EPA
expects individual manufacturers to match their control parameters to their unique EGR systems.
Sufficient control for transient response may be achieved by a number of methods. As mentioned
above, some researchers have demonstrated the use of VGTs and bypass Venturis with continuously

                                                                                     24

-------
                                                Chapter 3:  Technological Feasibility
variable valves in the intake system to achieve EGR control. For transient response, however, quick
and temporary EGR shut-off seems to be the best method for maintaining adequate torque response
without a sharp increase in transient PM emissions.34 For this reason EPA expects that EGR systems
will  have valves  positioned in the EGR loop to achieve  fast response for  transient  engine
operation.35'36'37 Durable EGR valves have been demonstrated by various manufacturers.38  One
valve manufacturer indicated that their EGR valve design will incorporate a fast acting (<50 ms)
electrically actuated rotary solenoid, which operates an airfoil-shaped valve plate. The manufacturer
expects to have the valve in production within the 2002 time frame. 39

       Because researchers and manufacturers have demonstrated that EGR strategy can result in
significant NOx reductions without unacceptable effects on PM emissions, BSFC, or performance,
and because manufacturers have demonstrated enabling technologies such as VGTs, Venturis, EGR
coolers, and control valves for complete EGR system implementation, EPA expects EGR  to be a
primary strategy for achieving the emission standards in this rule.

       C. Fuel Injection Rate-shaping

       Another key emission control  strategy that  EPA  expects  heavy-duty  diesel  engine
manufacturers to use to meet the new emission standards is fuel injection rate shaping. Fuel
injection rate shaping refers to precisely controlling the rate of fuel injected into the cylinder on a
crank-angle by crank-angle resolution. Specific rate-shaping methods include pilot injection where
a pilot quantity of fuel, typically less than 2% of the total fuel charge,40 is injected at some crank
angle before the main injection event. Split fuel injection refers to splitting, more or less evenly, the
main injection into two or more  separate injections. Other methods include ramping the main
inj ection event so that it resembles a triangular profile, rather than a square-shaped profile. Effective
injection rate-shaping systems modulate the  fuel injection  timing, pressure, rate, and duration
independent of engine speed and load.  This characteristic of the fuel system  implies that it is
mechanically de-coupled from the engine. Timing is then  achieved, presumably, by electronic
control.

       Injection rate shaping has been shown to simultaneously reduce NOx by 20 percent and PM
by 50 percent under some conditions.41 It has also been shown to reduce BSFC by up to 10 percent
without increasing NOx emissions.42 However, it can also lead to increases in smoke emissions and
may not be as effective on low-NOx engines equipped with EGR.

       Fuel injection rate shaping is used to control the rate of combustion within the cylinder. By
controlling the combustion rate, the rate of pressure and temperature rise is controlled. Therefore,
rate shaping controls NOx formation by one of the same  mechanisms as EGR; it is used to lower
peak combustion  temperatures.   Rate shaping can affect the time and temperature at  which
combustion ends, therefore, it can also lower PM emissions by enhancing the mechanisms of
in-cylinder soot oxidation.43

       Several manufacturers and fuel system suppliers have demonstrated fuel injection systems
that can achieve effective rate shaping. The three most common systems are the  common rail; the

                                                                                     25

-------
Regulatory Impact Analysis
mechanically actuated electronically controlled unit inj ector (MEUI); and the hydraulically actuated,
electronically controlled unit injector (HEUI). The common rail system consists of a high-pressure
(-25,000 psi.) fuel pump that pressurizes a pressure-regulated fuel header, or rail, that is connected
to each fuel injector.  The fuel injectors are  actuated by individual electronically  controlled
solenoids.44'45 A variation of the common rail system eliminates the individual solenoids by utilizing
a distributor sub-system.46'47 The MEUI system has low-pressure fuel (~ 60 psi.)48 delivered to its
inj ectors. The MEUI inj ectors pressurize the fuel when an overhead cam actuates them. By passing
the  pressurized fuel via an electronically controlled spill-valve controls the injection rate.49 The
HEUI  system  is  similar except that a high-pressure  hydraulic/accumulator system is used to
pressurize the fuel.  One advantage of the HEUI system over the MEUI is that it is not limited in
injection timing, pressure  or rate by a cam lobe profile. However, a HEUI system tends to have
lower peak  injection pressures  versus a MEUI; 25,000 psi vs. 30,000 psi.50 Other  rate shaping
systems may utilize spool  valve acceleration and fuel-hammering in the injection line, fuel tube
geometry, or dual springs at the  injector needle to perform rate shaping.51'52

       Several  studies have  suggested rate-shaping  methods to  achieve emissions benefits.
Researchers have reported decreased NOx and PM emissions at intermediate speeds  and loads by
optimizing reduced-rate pilot injection with a high-pressure main injection,53'54'55 and one report
suggested a fuel injection strategy at high loads. At intermediate loads, burnt pilot fuel is used as
a torch to decrease ignition delay of the main injection event. This lowers peak flame temperatures
and, thus, NOx formation.  At high loads the ignition delay is not as significant, but a very early pilot
event (>20°  before top-dead center) can be used to distribute low-temperature burnt  gas in the
cylinder, similar  to EGR.  This method can be optimized to decrease NOx, PM, and BSFC
simultaneously.56 Other reports have suggested ramped main inj ection at high loads and high speeds
to decrease NOx,  square main injection at peak torque to decrease PM, and split injection at idle to
decrease volatile PM (i.e. white  smoke).57

       EPA expects manufacturers to utilize fuel injection rate shaping in combination with EGR
and 1998 engine technologies to meet the new emission standards.  EPA believes that the fuel
injection rate shaping strategy  is  technologically feasible because fuel injection rate shaping is used
to a limited  extent today to meet 1998 emissions standards and has been shown in  testing to be
reliable and  effective.58

       D. Exhaust After-treatment

       As described in the introduction section,  engine manufacturers have been very  successful in
developing a mix of technologies to lower PM and NOx concurrently while continuing to improve
fuel economy and engine durability.  Although EPA is not finalizing a reduction in  the highway
heavy-duty engine PM standard  beyond the level of 0.10 g/bhp-hr (0.05 g/bhp-hr for urban buses),
PM control will continue to be very important. PM will remain a primary consideration along with
fuel economy and engine durability in the development of engines with lower NOx emissions. As
discussed above, HC emissions control has not been a primary focus for diesel engines due to their
relatively low HC emissions levels. With a NOx plus NMHC standard, HC emissions  levels would
                                                                                      26

-------
                                                 Chapter 3:  Technological Feasibility
become a greater consideration in the packaging of technologies to meet overall emission targets.

       Exhaust aftertreatment technologies for PM and NOx control are discussed in this section.
An extensive description of aftertreatment technologies was  presented in the  1997  rulemaking
package for the 2004 standards, including the final Regulatory Impact Analysis document.  The
reader is directed to the final RIA from the 1997 rulemaking for a discussion  of aftertreatment
technologies as of the 1997 time frame. The following discussion will include information which
has become available since the 1997 rulemaking, and will not repeat what was in that final RIA.

              1.  Particulate Matter Control

       Two aftertreatment technologies have received the most attention for paniculate control, the
flow-through oxidation catalyst and the particulate trap.  The oxidation catalyst provides relatively
moderate overall PM reductions by oxidizing a portion of the particulate as the  exhaust passes
through it.  Oxidation catalysts are relatively inexpensive and are now  being used by engine
manufacturers on some engines to meet the current 0.10 g/bhp-hr PM standard (0.05 for urban
buses).

       Particulate traps capture a very high percentage of the particulate and hold it until the PM can
be removed.  Removing the PM from the trap,  termed trap regeneration, is  accomplished by
oxidizing (i.e., burning) the PM. Because diesel exhaust almost never reaches the high temperatures
needed to ignite the PM, oxidation requires either an external heat source or a catalyst material to
lower the oxidation temperature of the PM.  Particulate traps have not gained wide  acceptance and
use due to several concerns that have not yet been overcome, including high cost, system complexity,
fuel economy penalty, and trap durability. Also, engine manufacturers have not needed the very high
level of PM control provided by traps to meet current standards.  However, research on traps has
been on-going, and some recent iterations look promising.

                    (a) Diesel Oxidation Catalysts

       As mentioned above, engine manufacturers have started to use diesel oxidation catalysts
(DOC) in cases where engines have needed help meeting the particulate standards. For the 1994
model year, about 30 percent of engine families certified were equipped with oxidation catalysts
(with the exception of urban buses, all of these were either light or medium HDDE's).  Another 30
percent of the engine families were certified to PM levels above the 0.10 g/bhp-hr standard through
the averaging, banking and trading program.  As these families are redesigned or  retired, the
percentage  of engine families equipped with oxidation catalysts may change.  Recent sales data on
oxidation catalyst for HD from the Manufacturers of Emission Control Association shows  a
continual decrease in the number of DOC's being sold in the U.S. (See Figure 3-1 below).
                                                                                     27

-------
Regulatory Impact Analysis
             Figure 3-1 - U.S. Sales Figures for HD Oxidation Catalysts

           DIESEL CATALYST EQUIPPED
                              VEHICLES

               History (94-97) and Forecast (98-99)
                       Light, Medium, and Heavy Duty
                       1994  1995   1996   1997   1998  1999
      Flow-through oxidation catalysts oxidize both gaseous hydrocarbons and the portion of PM
known as the soluble organic fraction (SOF). The SOF consists of hydrocarbons adsorbed to the
carbonaceous solid particles and may also include hydrocarbons that have condensed into droplets
of liquid.59 The carbon portion of the PM remains essentially unaffected by the catalyst. In recent
years, SOF has been reduced through new piston ring designs for oil control and fuel injection and
combustion chamber modifications for more complete combustion of the fuel. The amount of SOF
varies widely among engines but SOF often makes up 30 to 60 percent of the total  mass  of PM.
Catalyst efficiency for SOF varies with  exhaust temperature in the range of about 50 percent
conversion at 150°Ctomore than 90 percentabove350°C.60 Typically, exhaust temperatures during
the FID-FTP fluctuate between 100°C and 400°C.  The reduction in total particulate mass provided
by catalysts is relatively modest both because the efficiency is low at low exhaust temperatures and
because catalysts oxidize only the SOF and not the carbon portion of the PM.

      Improvements in catalyst technology have been hindered to some degree by sulfur contained
in diesel fuel.  Especially at higher exhaust temperatures, catalysts oxidize sulfur dioxide to form
sulfates, which contribute to total PM emissions.  Catalyst manufacturers have been  successful at
developing catalyst formulations that minimize sulfate formation.61 Catalyst manufacturers have also
compromised in the placement of the catalyst such that the exhaust is warm enough to achieve the
needed SOF reduction but not so warm as to cause substantial sulfate formation.62 Manufacturers
have noted that fuel with sulfur concentrations lower than 0.05 weight percent would permit the use
of more active, higher efficiency oxidation catalysts. Recent published reports show that for modern
HD diesel engines, palladium based oxidation catalysts can achieve an approximate 30% reduction
                                                                                  28

-------
                                                 Chapter 3: Technological Feasibility
in PM under steady-state (European 13-mode) operation using current U.S. diesel fuel, and these
formulations show good durability.63

       A recent test program sponsored by the Manufacturers of Emission Controls (MECA),
included the testing of several oxidation catalysts on a modern HD diesel engine certified to the 1998
U. S. HD standards. The results of this report showed up to a 29% reduction in PM over the transient
FTP, and PM reductions ranging between 0 and 67% on a series of 13 steady-state modes, with one
high load mode showing a slight (15%) increase in PM due to sulfate formation, these results were
all using a typical D2 diesel fuel used in the U.S. today (sulfur content approx. 350ppm)64. This
project also reported an additional 13 percent reduction in PM from the use of low sulfur diesel fuel
(54ppm).

        Oxidation catalyst development and use is likely to continue.  Future improvements in
oxidation catalysts will likely provide marginal improvements in overall PM reductions and such
refinements may prove to be valuable to engine manufacturers.

                    (b)  Particulate Trap

       The promise of particulate reductions of greater than 90 percent and the 1994 and later PM
standard of 0.10 g/bhp-hr prompted the development of paniculate trap technology in the late 1980s.
Particulate trap filters that capture a high percentage of the PM in the exhaust stream were
developed.  These initial particulate trap filters  needed to be regenerated (cleaned) after a period of
time because the filters  eventually began to fill up,  creating unacceptable back pressure on  the
engine.  Engine manufacturers have been able  to meet the 1994  particulate standards with engine
modifications and using oxidation catalysts where necessary and no trap-equipped engines were
certified for the 1994 model year.

       Several companies  and universities are developing a new generation of trap technologies
which have the potential to be simpler, more reliable, and less expensive than previous systems. The
majority of research and development is focused on devising new methods for trap regeneration. A
number of active and passive trap regeneration methods are in various stages of development and
testing.  The  1997 RIA  discusses  both  active  and passive trap  regeneration, however, the most
promising areas of improvement since that time have been in the area of passive systems, and only
those systems will be discussed here.

       Many regeneration techniques being researched involve using catalyst materials that lower
the PM oxidation temperature to the range normally experienced in diesel exhaust. The addition of
a catalyst often provides HC reductions as well.  Such systems are often called passive regeneration
systems because they do not require some action to take place for regeneration at regular intervals,
such as heating the PM or blowing the PM out  of the trap. Instead, regeneration occurs somewhat
continuously depending on the exhaust gas temperature. Catalysts both in the form of coatings and
fuel additives are being developed. Johnson-Matthey has developed a system that places a catalyst
at the inlet facing of the trap filter such that the exhaust flows though the catalyst before entering the
filter. The catalyst will oxidize sulfur and Johnson-Matthey is requiring the use of fuel with a sulfur

                                                                                      29

-------
Regulatory Impact Analysis
level much lower than EPA specifications.  One recent study utilizing this type of trap reported large
reductions in both mass based PM and HC on a modern, direct inj ection, turbo-charged, intercooled,
6.8 liter HD engine, but the system requires ultra-low sulfur fuel, less than lOppm.65

       As discussed in the 1997 RIA, several companies have explored the use of fuel additives
which assist in the regeneration process by lowering the PM ignition temperature. For example, fuel
additives including a cerium-oxide additive has been developed by Rhodia Chimie (formerly Rhone-
Poulenc) and a copper-oxide additive has been developed by Lubrizol Corporation.

       A recent test program sponsored by the Manufacturers of Emission Controls Association
(MECA), included the testing of two PM filter technologies tested in a laboratory on a modern HD
diesel engine certified to the 1998 U.S.  HD standards.66  One filter employed a catalytic coating
applied directly to the filter element (system A), the second filter technology utilizes a catalyst
element placed  directly upstream of the filter element (system B).  System A was tested on D-2
diesel fuel with current sulfur levels (368ppm),  while System B requires low sulfur fuel, and was
tested with a low sulfur  (54ppm) diesel fuel.  System A was tested over the transient U.S. FTP,
System B was tested on both the U.S. FTP, as well as a series of 13 steady-state modes. Table 3-1
contains a summary of the FTP results.

                                        Table 3-1
           PM trap testing results from MECA test program, U.S. HD FTP test cycle

System A - tested w/ fuel
sulfur level = 368ppm
System B - tested w/ fuel
sulfur level = 54ppm
Engine Baseline (g/bhp-hr)
0.073
-0.06
Results w/ trap system
installed (g/bhp-hr)
0.022
0.008
       Emission results on the 13-steady-state test cycle from the low sulfur fuel with System B
showed reductions ranging between approximately 20 and 70 percent, with the exception of one high
power mode, where PM increased approximately 30 percent. These emission results indicate that
PM traps applied to a 1998 technology HD diesel engine can provide large reductions in PM with
current fuel sulfur levels, and even lower PM levels may be achievable with the use of low sulfur
fuel. Durability information was not collected in this test program.

       Catalyst materials bring down the temperatures needed for PM oxidation, but still may be
challenged to reach the very low exhaust temperatures of diesel engines, which have been further
reduced by the use of air-to-air aftercooling.  For systems using catalysts, it will be necessary to
optimize the system for the specific engine application under real world operating conditions.  If the
temperature remains lower than the PM ignition temperature for long periods of time, say during idle
and low load conditions, the PM will continue to accumulate in the trap. When ignition temperature
                                                                                      30

-------
                                                Chapter 3:  Technological Feasibility
is reached, there may be too much PM in the trap, causing overheating and trap filter damage. It may
be necessary to have a back-up active regeneration system in some cases, but these back-up systems
would likely be expensive.

       Filter development is also focused on reducing the amount of exhaust back pressure and
associated fuel economy loss caused by the trap. Additionally, there are problems with ash in the
exhaust stream, which the trap captures along with the particulate matter. The ash does not oxidize
during trap regeneration and over time builds up within the trap; eventually, the filter must be
cleaned or replaced.   If traps begin to play a larger role as an emission  control technology,
improvements to engine oil (e.g. use of ashless oils) may increase the amount of time  a trap can
perform before ash build-up becomes a maintenance issue.

       In  the long term, traps may be among the mix of technologies  considered by  engine
manufacturers in meeting future standards, if a durable system with consistent regeneration and a
reasonable cost becomes available.  Issues such as regeneration, ash accumulation, and sulfur
tolerance have yet to be resolved by the 2004 time frame.

             2.  Oxides of Nitrogen Control

       The  1997 RIA  contains  a  description  of the major developments in NOx  control
aftertreatment devices which have been investigated in recent years,  including lean-NOx high
temperature and low temperature catalysts, NOx absorber catalysts, and urea-based SCR systems.
Additional development work has occurred in all of these  areas since the fmalization of the 1997
rulemaking. The discussion below will not repeat what was contained in the final RIA for the 1997
rule, however, much of that information continues to be relevant and the reader should refer to the
final RIA for the 1997 rule for additional information.

       In general, the issues associated with lean NOx catalysts, NOx absorber catalysts, and urea-
based SCR systems are similar today as they were in 1997.  These three systems continue to be the
focus of intensive research  because of the benefits they may someday offer.  The technical
difficulties discussed in 1997 continue to exist, though some progress has been made.

       Lean NOx catalysts continue to offer limited NOx reduction capability when considered
across the entire temperature operating range encountered by HD diesel engines, while peak
reduction capabilities may approach 60 percent under limited operating range, overall reductions on
the U. S. HD FTP continue to be modest, between 20 and 30 percent.  Lean NOx absorber catalysts
have shown a potential for much higher levels of NOx reduction, perhaps as high as 80 or 90 percent.
However, at today's on-highway diesel fuel sulfur levels, catalysts activity can be severely impacted
in a matter of hours.  Urea-based SCR systems have shown the potential for high levels of NOx
reduction from diesel engines, however, the technical issues such as urea refueling, tampering, and
ammonium slip  remain to be solved.  Finally, if the above issues can be solved for these
aftertreatment technologies, issues such as in-use durability, fuel economy impact, cost, and  and
cost-effectiveness will also need to be examined.
                                                                                     31

-------
Regulatory Impact Analysis
       The discussion below on each of these technologies discusses in more detail some of the
promise offered by NOx aftertreatment.

                    (a)  NOx Storage Catalysts

       NOx storage catalysts (also referred to as NOx absorber catalysts) are probably the best
example of a diesel emissions  control capable  of large reductions (>25%) reductions in NOx
emissions, but only if diesel fuel sulfur levels are considerably reduced. A generalized schematic
of their operation is included in Figure 3-2.  This catalyst system employs a high-platinum (Pt)
content catalyst for oxidation of nitric oxide (NO) to nitrogen dioxide (NO2)f The NO2 is then
stored, using one of a number of barium compounds, as barium nitrate.  For approximately two-
second durations every two minutes, diesel fuel is either sprayed into the exhaust, or fuel is injected
into the cylinder after combustion to provide the necessary hydrocarbons to remove the NOx from
the storage components. The NOx is then reduced over a standard three-way catalytic converter.
The average NOx reduction potential for this technology over the light-duty Federal Test Procedure
(FTP) is 50 to 75%, with a  fuel consumption penalty of approximately 3 to 5%.67  Figure 3-3
compares the NOx reducing  capabilities of a NOx storage catalyst system to two other lean-NOx
catalyst systems (one of which is sulfur tolerant).

       Unfortunately, the  chemistry for sulfate storage in such systems is similar to the desired
nitrate storage. Sulfur dioxide from combustion of fuel sulfur compounds is oxidized to SO3 by the
platinum catalyst, and stored as barium sulfate.  Purging sulfate from  the storage  components
requires significantly longer  periods of fuel-rich  conditions and significantly higher temperatures
(600 to 700 °C). The extended periods of high exhaust temperatures necessary for sulfate purging
from the storage components of the catalyst would be difficult to achieve, even for many heavy duty
diesel applications. Extended high temperature operation would also have a detrimental impact on
the useful life of the NOx  storage  components of the system.  Creation of the necessary fuel-rich
environment would pose  a  significant fuel consumption penalty, and would increase PM and
hydrocarbon emissions levels.
(f) In the absence of an oxidation catalyst, total oxides of nitrogen (NOx) in diesel exhaust is primarily NO
(typically >80%) with lesser amounts of NO2.
                                                                                      32

-------
                                                Chapter 3: Technological Feasibility
                                      Figure 3-3
           NOx-storage catalyst operation under oxidizing and reducing conditions.
                           Oxidizing (Lean) Conditions
             NO
                         Pt Catalyst
            NO,
 '.' •  Tirap •
'(Nt>2 stored as*
barium -nitrate)
            HC
                             Reducing (Rich) Conditions
                                HC,
NO.
  HLO, CO2, N.
            Approximately 2 seconds of stored nitrate regeneration needed for
            every 2 minutes of operation.
       Without sulfate purging, fuel sulfur levels of 350 ppm result in near complete deactivation
of NOx storage within 20 hours of operation. NOx storage catalysts are clearly not a viable NOx
exhaust aftertreatment control at current diesel fuel sulfur levels.  Diesel engines employing NOx
storage catalyst systems will probably be limited to the use of diesel fuels with less than 30 to 50
ppm sulfur68. Even at such fairly low sulfur levels, additional development of catalyst components
that reduce sulfur poisoning of the NOx storage components and less frequent, lower temperature
sulfate purging cycles may still be needed.

                    (b) Lean-NOx Catalysts

       Various types of active (requiring a post-combustion fuel injection event) and passive (no
post-injection) lean-NOx catalysts are in production or are under investigation for reduction of NOx
emissions in lean exhaust environments such as those present in diesel exhaust. Lean-NOx catalysts
typically reduce NOx efficiently over a fairly narrow range of catalyst temperatures. There are both
"high" and "low" temperature varieties of lean-NOx catalysts. Low temperature, platinum-based
lean-NOx catalysts using zeolites for support, catalyst promotion, and adsorption of NOx and HC,
would be typical of a lean-NOx catalyst technology for light-duty diesel vehicles with catalyst
temperatures  primarily in the 200  to 300 °C range.   High-temperature  lean-NOx catalyst
formulations  are under  investigation  primarily for highly-loaded,  heavy-duty diesel  engine
applications. High-temperature lean-NOx catalysts are primarily base metal  catalysts that are only
effective at exhaust temperatures exceeding 300  °C.
                                                                                    33

-------
Regulatory Impact Analysis
       A number of new common rail fuel injection systems are capable of injecting fuel after
combustion to provide additional hydrocarbons for use as a NOx reductant with active lean-NOx
catalysts. One example is the introduction of an active lean-NOx catalyst system for a European
light-duty diesel application69. Although active Pt-zeolite catalyst systems have higher NOx removal
efficiencies than similar passive catalyst systems, NOx removal efficiencies are still only in the range
of 15 to 35 % on average, and significantly below that of NOx storage catalyst systems (Figure 3-3).
It is more likely that low-temperature systems like the Pt-zeolite lean-NOx catalyst systems will be
used for incremental NOx  reduction  for light-duty applications  in combination  with other
technologies, such as cooled EGR.

       An  approximately 25% reduction in catalyst NOx efficiency due to adsorption of sulfur
compounds has been reported after 40,000 miles of roadway aging in a light-duty application at a
nominal 500 ppm fuel sulfur limit70. Sulfate PM emissions (primarily sulfuric acid),  rather than
sulfur poisoning, will probably be  a more pressing issue with  respect to fuel sulfur content8.
Conversion efficiencies for fuel sulfur to sulfuric acid of up to 20% are possible with Pt-zeolite lean-
NOx catalysts71.

       High-temperature base metal catalysts reduce NOx emissions by up to 30 % over the heavy-
duty FTP cycle. One such catalyst is the Cu ZSM5 catalyst72.  Similar to low temperature systems,
they may be used for incremental NOx reduction in combination with cooled EGR for heavy-duty
diesel engine applications, however, in-use durability issues remain. It is not clear whether or not
long term exposure to SO2 poses a significant problem for this technology.
(g)  Direct PM emissions from diesel engines primarily consist of 3 constituents: elemental carbon (soot), organics,
and sulfates.

                                                                                       34

-------
                                               Chapter 3:  Technological Feasibility
Figure 3-3: A comparison of the NOx reduction efficiency over a range of temperature conditions
for the sulfur-intolerant NOx storage  catalyst system (system for lean Gasoline Direct Injection
engine application shown73), the more sulfur-tolerant, active Pt-zeolite catalyst system and a high-
temperature base-metal (Cu-ZSM5) catalyst system.74 Although peak NOx reductions efficiencies
for various types of non-storage lean-NOx catalysts (similar to the Pt-Zeolite catalyst shown here)
approach 50-60%, average reductions are 15 to 30% over various light- and heavy-duty vehicle and
engine certification cycles.	
           100%"
         a
         o
            80%+   Active
                    Pt-Zeolite
                    Lean NOx
         X
         O
60%
            40%--
            20%--
               0
                             Pt/Rh/Ba
                           NOx-Storage
                             Catalyst
                     Catalyst
                                              Hi-Temp.
                                              Base-Metal
                                              Lean-NOx
                                           ''"•-..Catalyst
                      200         300        400          500
                          Catalyst Inlet Temperature (°C)
                                                                                  35

-------
Regulatory Impact Analysis
                    (c) Selective Catalytic Reduction

       Selective catalytic reduction (SCR) for NOx control is currently available for stationary diesel
engines, and prototype systems have been developed for mobile light- and heavy-diesel applications.
SCR uses ammonia as a reducing agent for NOx over a catalyst composed of precious metals, base
metals, and zeolites. The ammonia is supplied by introducing a urea/water mixture into the exhaust
upstream of the catalyst.  The urea/water mixture is typically stored in a separate tank that must be
periodically replenished. NOx reductions of 70% to 90% are possible using such systems.75 These
systems appear to be tolerant of current U.S. on-highway diesel fuel sulfur levels.

       Control of the quantity of urea injection into the exhaust, particularly during  transient
operation, is an important issue with SCR systems. Injection of too large of a quantity of urea leads
to a condition of "ammonia slip", whereby excess ammonia formation can lead to both direct
ammonia emissions and oxidation of ammonia to produce (rather than reduce) NOx. There are also
a number of potential hurdles to overcome with respect to a major emission control system that
requires frequent  replenishing in order to function. This  raises issues related to supply, quality
control, tampering, and the possibility of running the urea tank dry.  There is currently no wide-
spread distribution system in the U.S. for supplying the necessary water/urea mixtures for diesel
vehicles and trucks.

       E.  Diesel Fuel Composition

              1.  Introduction

       The purpose of this  section is to assess the  current understanding of the  role diesel  fuel
quality plays in the ability of diesel engines to meet the emission standards in this rule. The effects
of fuel formulation on exhaust emission formation as well as engine durability are examined.

       It has long been realized that diesel engine technology alone is not the only mechanism to
lower emissions, diesel fuel quality also plays an important role in emission formation as well as
engine performance. In addition, diesel fuel quality can play a  role in the effectiveness of certain
emission control technologies, and in some cases can be considered a technology enabler, i.e., some
emission control devices may not function because of certain diesel fuel properties, such as sulfur
content.

       In EPA's 1997 final rulemaking for the 2004 standards, we stated that we believed the 2004
standards were technologically feasible thru diesel engine technology modifications alone, without
changes to  diesel  fuel quality (see 62 Federal Register, 54700,  Oct. 21, 1997). However, we  also
stated that  this issue would be revisited in  the 1999 technology review rulemaking,  "EPA will
evaluate in light of any new information whether diesel fuel  improvements are needed for the
standards to be appropriate for 2004"  (see 62 Federal Register, 54700,  Oct. 21,  1997). In section
2 below we review the new information which has become available since the 1997 rulemaking thru
a study performed by the Heavy-duty Engine Working Group and durability information supplied

                                                                                      36

-------
                                                Chapter 3: Technological Feasibility
by manufacturers. Section 3 below addresses issues regarding the effect of diesel fuel sulfur levels
on emission control system and engine durability for the technology neccessary for HD diesel
engines to comply with the standards in this rule.

             2.  Heavy-duty Engine Working Group

                    (a) Background

       In anticipation of the need for new information regarding the influence of diesel fuel quality
on future emission technologies and achievable levels, in December of 1995 a new Working Group
called the  Heavy-duty Engine Working Group  (HDEWG) was formed under the Mobile Source
Technical  Advisory Subcommittee of the Clean Air Act Advisory Committee.  The HDEWG
consists of approximately 30 members,  including representatives from EPA, heavy-duty engine
OEMs, the oil industry, state air quality agencies, private consultants and members of academic
institutions. The HDEWG formed a steering committee which consisted of representatives from
EPA, Cummins, Caterpillar, Navistar, Ford, British Petroleum, Equilon, Mobil Oil, Phillips, the
Engine Manufacturers Association, the American Petroleum Institute, and the National Petroleum
Refinery Association.  The HDEWG set as their research objective to contribute to EPA's 1999
technology review of emission standards for model year 2004 heavy-duty diesel engines by assessing
relative merits of achieving 2.5 g/bhp-hrHC+NOx level either through engine system modifications
alone, or a combination of engine system and fuel modifications.

       The HDEWG established a three phase process in order to meet their objective. In Phase 1,
the goal was to determine  whether the combined effects of diesel fuel properties on exhaust
emissions  of "black box", advanced prototype engines being developed by engine manufacturers
were large enough to warrant a Phase 2.  "Black box" engines are advanced engines being designed
by engine manufacturers to meet the 2004 standards, but the details of each black box engine would
not be shared with the HDEWG.  In addition, the HDEWG agreed to use one "transparent" engine
at an independent test facility, Southwest Research Institute (SwRI).  During Phase 1, testing was
to be performed  on the transparent engine at  SwRI,  as well as the black box engines at
manufacturers own testing facilities, to determine if the transparent engine was representative of the
black box  engines with respect to diesel fuel effects on NOx emissions.

       Phase 2 of the program, which would occur upon successful completion of Phase 1, would
be used to test a range of relevant fuel properties on the transparent engine at SwRI, in order to
determine  the effects of those fuel properties on emissions.  Finally, Phase 3 of the test program
would determine whether or not the results seen during Phase 2 on the transparent engine was in fact
representative of black box  engines, i.e., advanced prototype engines being developed by engine
manufacturers to meet the 2004 standards.  Phase 3 would be performed at engine manufacturer's
laboratories using a subset of the fuel matrix from Phase 2.

                    (b) Phase 1 of the HDEWG Test Program
                                                                                     37

-------
Regulatory Impact Analysis
       The Phase 1 test program consisted of two test phases; first, testing on three fuels by engine
manufacturers at their facilities of "black box" engines, i.e., advanced prototype engines being
designed to meet the 2004 HC+NOx standard, and second, testing on the same three fuels at SwRI
of the transparent engine. The purpose of Phase 1 was to determine first, whether or not changes in
relevant fuel properties had an important effect on NOx emissions for the black box engines which
would justify continuing to Phase 2, and second, whether or not the transparent engine behaved
similarly to the black box engines, and, thus, could be used for Phase 2 testing. Two reports are
available in the docket for this rulemaking which contain detailed information on the Phase 1 portion
of the program, the following discussion will summarize the results of Phase 1, the reader should see
the detailed reports  for more in depth information.76'77   Table  3-2 describes  the three  fuel
formulations used for Phase 1 testing.

                                       Table 3-2:
  Diesel Fuel Formulations used for Phase 1 Testing by the Heavy-duty Engine Working Group
Fuel Property
Density kg/m3
Cetane Number
Monoaromatics %
Polyaromatics %
Total Aromatics %
Baseline Fuel
856
45.9
26.6
9.1
35.7
Baseline Fuel w/
Cetane Enhancer
856
52.4
26.2
8.9
35.1
Naturally High Cetane,
Low Aromatic Fuel
823
56.9
15.5
4.5
20
       It should be noted that the HDEWG's  primary focus was on the effects of diesel fuel
properties on HC and NOx emissions, not on PM emissions, and therefore fuel sulfur level was not
investigated.  A significant amount of data exists on the effects of diesel fuel sulfur on engine
emissions, and in fact this data was summarized recently in an SAE paper published by members of
the HDEWG which will be summarized below. Based on the existing data on recent model year HD
engines, diesel fuel sulfur level does have a statistically significant effect on PM emissions, but no
statistically significant effect on HC, CO, or NOx emissions (on engines with no aftertreatment).
For this reason, and because of the focus on HC and NOx emissions, as well as the limitations of the
SwRI transparent engine discussed below, the HDEWG did not include fuel sulfur level as a variable
in Phase 1, 2 or 3 of their test program, nor were PM emissions measured in Phase 1 or 2.

       Engine manufacturers tested the three fuels shown in Table 3-2 on a total of six black box
engines. In addition, SwRI tested the transparent engine on the same three fuels. The test cycle used
by SwRI was the so-called AVL 8-mode test. This steady-state test cycle, with associated weighting
factors, has been shown in the past to correlate very well with NOx emissions measured over the
U.S. FTP. The transparent engine is representative of a modern, heavy-heavy duty diesel engine
which could be certified to 1998 U.S. emission standards in it's baseline condition. SwRI calibrated
                                                                                     38

-------
                                                Chapter 3: Technological Feasibility
the transparent engine on the baseline test fuel to a 2.7g/hp-hr HC+NOx level utilizing a prototype
low-pressure loop cooled EGR system, this was followed by testing on the two non-baseline fuels.
The cooled EGR system developed by SwRI was not capable of transient operation, and while the
AVL 8-mode does adequately predict transient U.S. FTP NOx, it does not accurately predict PM
emissions, therefore, PM emissions were not measured. Table 3-3 below summarizes the results of
the Phase 1 testing program.

                                       Table 3-3:
                        Summary of HDEWG Phase 1 Test Results

Test Engines
Six Black Box
Engines
SwRI Transparent
Engine
% Change in NOx Emissions

Naturally High Cetane, Low
Aromatic Fuel vs. Baseline Fuel
7.6 percent decrease
7.0 percent decrease
Baseline Fuel w/ Cetane
Enhancer vs. Baseline Fuel
2.4 percent increase
3. 4 percent increase
       The HDEWG concluded the following from the Phase 1 test results; the transparent engine
at SwRI responds to  fuel property changes similarly to the black box engines and therefore the
transparent engine is appropriate for the Phase 2 test program, and the magnitude of the fuel effects
on NOx emissions for the transparent engine and the black box engines was significant enough to
warrant the continuation of the program into the Phase 2 testing.

       In  addition to the test program  portion of Phase 1, several members of the  HDEWG
performed an extensive literature review of existing data on the effects of diesel fuel formulation on
emissions.  The result of this work was recently published by Society of Automotive Engineers,
paper number 982649,  "Fuel Quality Impact On Heavy-Duty Diesel Emissions:- A Literature
Review."   This paper reviewed publically  available data which looked at the following fuel
properties;  sulfur, cetane number, total aromatics, polyaromatics,  density, volatility (back-end
volatility as determined by T90/T95) and oxygenates. This paper reviewed published results which
include test data measured from both the U. S. HD transient FTP, as well as the European steady-state
13-mode ECE R49 test cycle. The literature search included engines of various levels of emission
control technology, in general the engines were designed to meet U. S. 1991 through 1998 standards,
or European 1993 through 1996 emission limits.  The authors divided the available engines into two
groups; "low emission emitting engines" and "high emission emitting engines." Low emission
engines were those engines with NOx emissions between approximately 3.5 and 5  g/hp-hr, and PM
emissions  approximately between .05 and .2 g/hp-hr.  High emission engines were those engines
with NOx  emissions between approximately 5.5 and 8 g/hp-hr, and PM emissions approximately
between .4 and .5 g/hp-hr. The paper offers an excellent overview of available information, and the
                                                                                     39

-------
Regulatory Impact Analysis
details of the paper will not be restated here.  A summary of the effects which were found on the
"low emission emitting engines" is summarized in Table 3-4 below.

                                      Table 3-4:
  Summary of Diesel Fuel Properties on Recent Model Year Heavy-duty Diesel Emissions from
                  "low emission emitting engines" from SAE paper 982649
Fuel
Modification
Reduced Sulfur
Increase
Cetane
Reduce Total
Aromatics
Reduce Density
Reduce
Polyaromatics
Reduce
T90/T95
HC
no effect
no effect
no effect
large increase in
HC
small decrease
inHC
very small
increase in HC
CO
no effect
no effect
no effect
small increase in
CO
no effect
very small
increase in CO
NOx
no effect
small decrease
in NOx
small decrease
in NOx
small decrease
in NOx
small decrease
in NOx
very small
decrease in NOx
PM

large effect for
moving from
.3% to .05%,
minimal effect
for reducing S
from 0.05%
no effect
no effect
no effect
no effect
no effect
       The authors noted that there was very little information available on the effect of increasing
oxygenates, and any conclusions would be very tentative, therefore, the summary of oxygenates is
not included here. It should be noted that the term "low emission emitting engines" employed by
the authors is well above the 2.5g/hp-hr HC+NOx level.

       Based on the results of the Phase 1 results for "black box" engines, the "transparent" engine,
and the literature review of available data, the HDEWG agreed to proceed to Phase 2.

                    (c) Phase 2 of the HDEWG Test Program

       The purpose of the Phase 2 component of the test program was to test a range of relevant fuel
properties on the transparent engine at SwRI in order to determine the effects of various fuel
properties on emissions. All testing during Phase 2 of the test program was done at SwRI on the
transparent engine.   The  parameters  investigated and the results of the Phase 2 testing are
                                                                                    40

-------
                                                Chapter 3:  Technological Feasibility
summarized in this section. A document containing detailed information on the Phase 2 test program
is available in the docket for this rulemaking, the following discussion will summarize the relevant
results of Phase 2, the reader should see the detailed report for more in depth information.78

      Based on the results of the Phase 1 testing, as well as the literature review performed under
Phase 1, the HDEWG selected four fuel properties for investigation under Phase 2: density, cetane
(natural and "boosted"11), monoaromatic content and polyaromatic content. As mentioned previously,
fuel sulfur level was not investigated. A test matrix was designed to decouple these fuel properties
from each other, in addition, fuel  blends were added to the matrix to evaluate density effects as a
function of engine injection timing and a direct comparison of natural and boosted cetane number.
The design matrix included two levels of density, monoaromatic hydrocarbons, polyaromatic
hydrocarbons, and three levels of cetane. The final matrix included eighteen test fuels, with density
varying from 830 to 860 kg/m3, cetane numbers from 42 to 48 to 53, monoaromatic content from 10
to 25 percent, and polyaromatic content from 2.5 to 10 percent.  For all emission testing, the AVL
8-mode test was utilized, and all emission tests were performed at least in duplicate. In addition to
the fuel property  effects, the effects of injection timing and EGR were evaluated.  The SwRI
prototype, low-pressure loop, cooled EGR system was manually controlled to set EGR rates in order
to approach an AVL 8-mode composite NOx level of 2.5g/hp-hr.

      The large quantity of test data generated by the test program was evaluated using statistical
techniques in order to develop exhaust emission and fuel consumption prediction models based on
the four fuel properties. All properties were evaluated using a significance level of 5 percent. The
HDEWG examined the dependence of emissions and fuel consumption on the four  parameters
(density, cetane, monoaromatic content and polyaromatic content).

      The following tables summarize the most important results of the Phase 2 test  program.
Table 3-5 summarizes the effects of individual fuel properties on predicted NOx, HC, and HC+NOx
emissions.  Table 3-6 summarizes the combined effects of fuel properties on predicted NOx, HC,
and HC+NOx emissions.  Table 3-6 contains a summary of percent changes in predicted results for
two fuels, a blend representative of current U.S. diesel fuel (based on national fuel surveys for 1994
and 1995, except for polyaromatic content, which was estimated by the HDEWG), and a "clean"
diesel fuel, i.e., a fuel low in density, high in cetane, and low in both monoaromatics and
polyaromatics.
(h)  Boosted cetane is achieved by the addition of a fuel additive, in this case ethylhexyl nitrate

                                                                                     41

-------
Regulatory Impact Analysis
                                      Table 3-5:
Effects of Individual Fuel Properties on Predicted Emissions from Phase 2 Testing of the Heavy-
duty Engine Working Group Project (Reference values for NOx, HC, and HC+NOx of 2.57 g/hp-
  hr, 0.13 g/bhp-hr, and 2.7 g/bhp-hr respectively were used. Negative percentages represent a
           decrease in emissions with the corresponding decrease in fuel property)
Pollutant
% NOx Change
@ 2.57 g/bhp-hr
% HC Change
@ 0.1 3 g/bhp-hr
% HC+NOx
Change @ 2.70
g/bhp-hr
Density
860 -> 830
kg/m3
-4.8
Not Significant
-4.3
Cetane Number
52-M2
-1.3
14.3
Not Significant
Monoaromatics
25 -> 10 %
-3.8
-7.8
-4.3
Polyaromatics
10 -H.5 %
-2.2
-9.2
-2.3
                                      Table 3-6:
 Combined Effects Fuel Properties on Predicted Emissions from Phase 2 Testing of the Heavy-
duty Engine Working Group Project (Reference values for NOx, HC, and HC+NOx of 2.57 g/hp-
  hr, 0.13 g/bhp-hr, and 2.7 g/bhp-hr respectively were used. Negative percentages represent a
                                 decrease in emissions)








Average U.S.
Diesel Fuel
"Light", High
Cetane, Low
Aromatic Fuel
Fuel Property
Density
kg/m3






845


830
Cetane
Number






45


52
Mono-
aromatics
%





25


10
Poly-
aromatic
s%





9


2.5
Predicted Emission Change
% Change
in NOx
vs.
"Light" at
2.57g/bh
p-hr level





-7.2
% Change
in HC vs.
"Light" at
0.13
g/bhp-hr
level





OC Q
-Zj.o
% Change
in
HC+NOx
vs.
"Light" at
2.70g/bh
p-hr level




-8.4
       The test data was also analyzed to look at the effect of the prototype low pressure loop,
cooled EGR system on measured emissions and on measured fuel consumption (not predicted).
                                                                                    42

-------
                                                Chapter 3:  Technological Feasibility
Duplicate emission tests performed on each of seven test fuels with the EGR system on and off were
examined. The results indicated EGR had a strong, statistically significant effect onNOx emissions,
no effect on HC emissions, and a strong effect on HC+NOx emissions.  The EGR system used
reduced NOx emissions between 35.9 and 37.2 percent, and HC+NOx emissions by  34.2 to
35.3percent.  The EGR system had no statistically  significant impact on brake-specific fuel
consumption.

                    (d) Phase 3  of the HDEWG Test Program

       Phase 3 of the test program has not been completed. The purpose of the Phase 3 program
will be to  determine whether or not the Phase 2 results seen  on the transparent  engine are
representative of "black box" engines, i.e., advanced, prototype HD diesels being  developed by
manufacturers to meet the 2004  standards.  The Phase 3 testing will  occur at individual engine
manufacturers facilities, and will utilize full U.S. FTP transient emission testing, and will include
PM measurement. The Phase 3 program is scheduled to be competed in mid-1999.

                    (e) EPA Assessment of HDEWG Data

       The most significant data for this rulemaking activity generated up to this point in time by
the HDEWG is presented in Tables  3-5 and 3-6. The data in Table 3-5 indicates that for engines
utilizing advanced fuel injection and a cooled EGR  system operating at emissions levels near the
2004 standards, the effects of relatively large changes in individual fuel properties is statistically
significant but rather small, and for cetane number not statistically significant. A large decrease in
fuel density (from 860 to 830 kg/m3 ) or in monoaromatic content (from 25 to 10  percent) is
predicted to result in  a 4.3 percent decrease in HC+NOx emissions, and a large decrease in
polyaromatics content (from 10 to 2.5 percent) is predicted to result in a 2.3 percent decrease in
HC+NOx emissions.

       The data in Table 3-6  indicates  the potential  impacts on HC+NOx emissions from the
combined effects of significantly changing diesel fuel formulation from today's currently available
U. S. on-highway diesel fuel. The results predict that a combined, relatively large decrease in density,
large increase in cetane, and large decrease in both monoaromatic content and polyaromatic content
would result in a 8.4 percent decrease in HC+NOx emissions.

             3.     Fuel Sulfur  Impact on Engine Durability

                    (a) Condensate Issues

       Cooled EGR poses several design issues, one of those being corrosion from EGR condensate.
This condensate is composed of two major components, water and sulfuric acid. The water is a
normal byproduct of combustion and the sulfuric acid (H2SO4) is formed primarily from sulfur in
the fuel.  The rate of acid condensation is proportional to the concentration of sulfur in the fuel.
Current on-highway requirements limit diesel fuel sulfur to 500 ppm or less.  Manufacturers have
proposed at least 30 ppm maximum  sulfur fuel to minimize sulfur induced corrosion.

                                                                                    43

-------
Regulatory Impact Analysis
       The EGR cooler, intake plumbing, intake manifold, cylinder kit (piston rings and cylinder
liner), and engine oil will be exposed to this condensate.  The EGR cooler will be the most critical
component from a corrosion standpoint. It will be cooling raw exhaust, which is more likely to
condense than the diluted exhaust found in the engine intake system. Corrosion of the EGR system
and intake charge plumbing can lead to contamination of the intake charge. Particles from the walls
of the intake plumbing can be released by the corrosion process and carried by the intake charge into
the cylinders. Once there, the particles act to abrasively wear the cylinder kit causing loss of oil
control. Corrosion induced pitting on the cylinder liner from the sulfuric acid entrained in the EGR
could also be an issue.

       The engine oil will also be impacted by the fuel sulfur and EGR.  The sulfuric acid can get
into the engine oil via the blow-by or via deposition on the cylinder liner.  The result will be
accelerated depletion of the oil PH control package.

                    (b) Corrosion  Resistance

       Most of the EGR induced corrosion issues will be dealt with through careful material and
bonding process selection. Stainless steels with higher nickel or cobalt content may be necessary
to provide the required EGR cooler life.79 Bonding methods used in the construction of these coolers
are also available to reduce corrosion. Along with corrosion resistant materials, the EGR can also
be controlled to minimize condensation under adverse conditions, such as cold start.  This attention
to material selection and the level of EGR cooling will minimize the condensation impact on engine
durability.

       Engine oil reformulation studies have already begun to set a new  standard for engines with
cooled EGR. Improved TEN control additive packages will be part of this standard along with
increased oil soot tolerance capability. These improvements should allow the oil to perform at least
as well as current (non-EGR) oils.

       F. Performance  of Projected Emission Control Technologies over Typical In-
       Use Conditions

       The technologies discussed in this chapter, cooled EGR, advanced fuel inj ection systems with
rate shaping, and variable-nozzle turbochargers, combined with electronic control systems, are all
applicable for in-use operation, under both steady-state and transient operation. The Agency expects
that this technology package can achieve the emission reductions necessary to comply with the 2004
standards under a large variety of operating conditions, not simply the test cycle contained in the
Federal transient FTP. Many of the published reports in the past several years have looked at the
application of these technologies not only under transient operation, but also under steady-state test
cycle conditions. Some of these reports included data for the  steady-state cycles used in Japan and
Europe, which are similar to the EPA cycle for the steady-state MAEL requirements. As indicated
in Table 3-7, NOx and PM performance in these steady-state conditions are at or near the standards
in this  rulemaking.  In addition, the test results included in the 1997 RIA indicate NOx and PM
performance at  or near the standards using both transient and steady-state tests.  The Agency sees

                                                                                      44

-------
                                                Chapter 3: Technological Feasibility
no reason why the technologies discussed previously would not function sufficiently well under a
wide range of operating conditions, and thus we expect them to be designed to provide comparable
levels of emission control under a wide range of operating conditions in the timeframe require by
this rule.  As discussed previously, cooled EGR alone has been demonstrated under laboratory
conditions to provide NOx reductions up to 90 percent at light load conditions and up to 60 percent
near rated speed.   These  conclusions are fully consistent with the expectation  in the 1997
rulemaking.

       Following the October 29, 1999 Federal Register publication of the proposal for this final
rule the Agency met with a number of HD diesel engine manufacturers to discuss the supplemental
test procedure requirements.  During this time we received a substantial amount of confidential
business information from a number of FID diesel  manufacturers who are designing engines to
comply with the NTE requirements under the  1998 FID  diesel consent decrees.  We have
summarized this information into a memorandum which is available for review in the public docket
for this rule.80 This memorandum shows that a number of manufacturers have made significant
progress towards meeting the 2007  supplemental requirements.   The  principle  issue which
manufacturers are now in the process of addressing is the ability to drive and control cooled EGR
under high load conditions  for highly  rated versions  of their engines.  Manufacturers have
encountered a number of technical issues, depending on the engine design path they have chosen.
Some manufacturers are encountering the physical limitations of current turbocharger designs, while
others are addressing complex control strategies which in some cases push the limits of currently
available sensors and actuators. Engine manufacturers are at different stages of product development
for achieving NTE limits. Based on the information provided to the Agency, those manufacturers
who have progressed the furthest have narrowed the remaining technical issues substantially. These
manufacturers have narrowed the focus of their efforts to the ability to drive and control EGR for
the highly rated versions of their engines while maintaining engine performance, under the high load
regions of the NTE,  and only under ambient conditions of high temperature and/or high altitude.
The  Agency  also spoke with a  major  turbocharger manufacturer regarding the performance
capabilities of turbocharger in the 2004 time frame and beyond.81 Based on the emission capabilities
of the  emission control technologies previously discussed in this RIA (cooled EGR, advanced
turbomachinery, and advanced fuel inj ection systems), the information the Agency has received from
manufacturers, and the summary of the confidential information provided by a number of engine
manufacturers, the Agency concludes the supplemental test procedure requirements (NTE and SSS)
are technically feasible by 2007.

       For these reasons, EPA believes the primary technologies discussed in this chapter will
provide the necessary NMHC+NOx and PM control  on the existing transient FTP by 2004, as well
as the supplemental test cycles, procedures and associated standards contained in this rule for 2007.
Thus, we do not expect these requirements to impose new hardware burdens.  Manufacturers are
expected to only need to conduct additional emission testing and perform recabilbration of their
engines to comply with these requirements.

       G. Summary and Conclusions regarding HD Diesel Technologies
                                                                                    45

-------
Regulatory Impact Analysis
       The Regulatory Impact Analysis document for the 1997 HD diesel FRM documents EPA's
analysis which lead to the conclusion that the Agency believed the 2004 HD NOx+NMHC standards
were technologically feasible. This RIA contains EPA's reassessment of the technological feasibility
of these standards, including a discussion of the role diesel fuel quality plays in the appropriateness
of the standards. Table 3-7 summarizes the emission performance results of several studies that were
recently conducted on heavy-duty diesel  engines, and which have been discussed earlier in this
chapter.  In the technological feasibility chapter of the 1997 RIA for this rule, a similar table is
presented for results up to 1997.

                                       Table 3-7:
            Summary of recently published data on 2004 capable control  strategies
Technology
VGT turbocharged, aftercooled, 4-
valve/cyl, high-pressure fuel
injection, HPL cooled EGR, with
full-flow venturi mixer82
VGT turbocharged, aftercooled,
high-pressure electronic fuel
injection, HPL cooled EGR, with
full-flow venturi mixer83
VGT turbocharged, aftercooled,
HPL cooled EGR84
waste-gate turbocharged, air-air
aftercooled, 4 valve/cyl, MEUI fuel
injection, HPL cooled EGR with
partial flow venturi mixer85
same as above, including reference
same as above, including reference
Test
Cycle
ECE R49
13 -mode
ECE R49
13 -mode
Japanese
13 -mode
Euro-3
ESC
Euro-3
ESC
Euro-3
ESC
NOx
2.24 g/hp-hr
1.80g/hp-hr
22% dec.
from no
EGR & VGT
3. 24 g/hp-hr
2.33 g/hp-hr
1.83 g/hp-hr
PM
0.08 g/hp-hr
0.08 g/hp-hr
No
significant
change
0.06 g/hp-hr
0.08 g/hp-hr
0.1 5 g/hp-hr
BSFC
No
significant
change
2.3% inc.
from no
EGR
1.5% dec.
from no
EGR & VGT
No
significant
change
0.9% inc.
from no
EGR
2.4% inc.
from no
EGR
       These results and the results indicated in the 1997 RIA show the types of emission values
which can be achieved from the combination of cooled EGR, advanced electronic controls, advanced
turbo-chargers, and high-pressure fuel injection systems with rate shaping capabilities. The results
above indicate that current technology can achieve NOx and PM results  at or near the 2004
                                                                                      46

-------
                                                Chapter 3:  Technological Feasibility
standards. Results referenced in the 1997 RIA include a study showing HC + NOx levels of 2.54
g/bhp-hr on the current transient cycle FTP. Based on the tests that have been conducted in the past
few years, EPA projects that manufacturers will continue to optimize fuel injection and EGR
strategies in the lead time available to them , and will be able to meet the NMHC+NOx emission
standards in this rule, while continuing to meet the PM standard, with little or no brake specific fuel
consumption penalty. Also, as discussed previously, the Agency has placed information into this
rulemaking record summaries  of information  and discussion with engine manufacturers and
component suppliers regarding  the feasibility  of  the  2007  supplemental requirements.  This
information shows manufacturers have made substantial progress towards meeting the supplemental
requirements  (NTE and supplemental steady-state requirements).   Considering the remaining
technical issues, the potential solution to these issues, and the significant lead time (more than six
years), the Agency concludes the supplemental  requirements contained in this final rule will be
feasible by model year 2007.

III.  HD Otto-cycle Engine & Vehicle Technologies

       The purpose of this sub-chapter is to further  expand upon the technical discussion that was
presented in the preamble.  HD otto-cycle vehicle and engine exhaust emissions can be reduced by
a number of  technologies, but the most potential  for improvement  exists in reductions to base
engine-out emissions, improvement in air-fuel ratio control, better fuel delivery and atomization, and
continued advances in exhaust aftertreatment.

       The following descriptions provide an overview of the latest technologies capable of reducing
exhaust emissions.  The descriptions will also discuss the state of  development and current
production usage of the various technologies.  It is  important to point out that the use of all of the
following technologies is not required to further  reduce emissions. The choices and combinations
of technologies will depend on several factors,  such as current engine-out emission  levels,
effectiveness of existing emission control systems, and individual manufacturer preferences. With
the exception of a few technologies, many of these technologies are used in some heavy-duty and
light-duty vehicles already in production.

       EPA used a number of references for the following discussion. EPA consulted an Energy
and Environmental Analysis, Inc. (EEA), study evaluating emission control technologies for light-
duty vehicles and light-duty trucks.86 EPA used as references, the State of California Air Resources
Board  (CARB) staff reports on "Low-Emission Vehicle and Zero-Emission Vehicle Program
Review," and "LEW published inNovember 1996 and September 1998 respectively.87'88 EPA also
used as a reference information from the Manufacturers of Emission Controls Association (MECA)
and vehicle manufacturers.

       While the EEA report focused on light-duty vehicles, the emissions controls for heavy-duty
vehicles would be very  similar.  Often technologies are first introduced on light-duty vehicles and
then later applied to heavier vehicles as needed. For example, most heavy-duty vehicles and engines
are now equipped with sequential fuel inj ection, three way catalyst systems with closed loop control,
and EGR.  The CARB medium-duty vehicle program applies to vehicles up to 14,000 pounds

                                                                                     47

-------
Regulatory Impact Analysis
GVWR and includes LEV and ULEV standards.  For heavy-duty vehicles and engine specifically,
EPA contracted Arcadis Geraghty and Miller to review technologies and perform cost analyses for
the standards that were proposed in the NPRM for this rulemaking.89

       A. Base Engine Improvements

       There are several design techniques that can be used for reducing engine-out emissions,
especially for HC and NOx. The main causes of excessive engine-out emissions are unburned HCs
and high combustion temperatures for NOx. Methods for reducing engine-out HC emissions include
the reduction of crevice volumes in the combustion chamber, reducing the combustion of lubricating
oil in the combustion chamber and developing leak-free exhaust systems. Leak-free exhaust systems
are listed under base engine improvements because any modifications or changes made to the
exhaust manifold can directly affect the design of the base engine.  Base engine control strategies
for reducing NOx include the use of "fast burn" combustion chamber designs, multiple valves with
variable-valve timing, and exhaust gas recirculation.

             1. Combustion Chamber Design

       Unburned fuel can be trapped momentarily in crevice volumes (i.e., the space between the
piston and cylinder wall) before being subsequently released.  Since trapped and re-released fuel can
increase engine-out HC, the reduction of crevice volumes is beneficial to emission performance.
One way to reduce crevice volumes is to design pistons with reduced top "land heights" (distance
between the top of the piston and the first piston ring). The reduction of crevice volume is especially
preferable for vehicles with larger displacement engines, since they typically produce greater levels
of engine-out HC than smaller displacement engines.

       Another cause of excess engine-out HC emissions is the combustion of lubricating oil that
leaks into the combustion chamber, since heavier hydrocarbons in oil do not oxidize as readily as
those in gasoline. Oil in the combustion chamber can also trap gaseous HC from the fuel and release
it later unburned. In addition, some components in lubricating oil can poison the catalyst and reduce
its effectiveness.  To reduce oil consumption,  vehicle manufacturers will  tighten tolerances and
improve surface finishes for cylinders and pistons, improve piston ring design and material, and
improve exhaust valve stem seals to prevent excessive leakage of lubricating oil into the combustion
chamber.

       As discussed above, engine-out NOx emissions result from high combustion temperatures.
Therefore, the main control strategies  for reducing engine-out  NOx are designed to lower
combustion temperature. The most promising  techniques for reducing combustion temperatures,
and thus engine-out NOx emissions, are  the combination of increasing the rate of combustion,
reducing spark advance, and adding a diluent to the air-fuel mixture,  typically via  exhaust gas
recirculation (EGR). The rate of combustion can be increased by  using "fast burn" combustion
chamber designs. A fast burn combustion rate provides improved thermal efficiency and a greater
tolerance for dilution from EGR resulting in better fuel economy and lower NOx emissions.  There
are numerous ways to design a fast burn combustion chamber.  However,  the most  common

                                                                                    48

-------
                                                 Chapter 3: Technological Feasibility
approach is to induce turbulence into the combustion chamber which increases the surface area of
the flame front and thereby increases the rate of combustion, and to locate the spark plug in the
center of the combustion chamber. Locating the spark plug in the center of the combustion chamber
promotes more thorough combustion and allows the ignition timing to be retarded, decreasing the
dwell time of hot gases in the combustion chamber and reducing NOx formation.  Many engine
designs induce turbulence into the combustion chamber by increasing the velocity of the incoming
air-fuel mixture and having it enter the chamber in a swirling motion (known as "swirl").

              2.  Improved EGR Design

       One of the most effective means of reducing engine-out NOx emissions is  exhaust gas
recirculation. By recirculating spent exhaust gases into the combustion chamber, the overall air-fuel
mixture is diluted, lowering peak combustion temperatures and reducing NOx. As discussed above,
the use of high swirl, high turbulence combustion chambers  can  allow the amount of EGR to be
increased from current levels of 15 to 17 percent to levels possibly as high as 20 to 251 percent,
resulting in a 15 to 20 percent reduction in engine-out NOx emissions.

       Many EGR systems in today's vehicles utilize a control valve that requires vacuum from the
intake manifold to regulate EGR flow. Under part-throttle operation where EGR is needed, engine
vacuum is sufficient to open the valve.  However, during throttle applications near or at wide-open
throttle, engine vacuum is too low to open the EGR valve. While EGR operation only during part-
throttle driving conditions has been sufficient to control NOx emissions for most vehicles in the past,
more stringent NOx standards and emphasis on controlling off-cycle emission levels  may require
more precise EGR control and additional EGR during heavy throttle operation to reduce NOx
emissions.  Many manufacturers now use electronic EGR in place of mechanical back-pressure
designs. By using electronic solenoids to open and close the  EGR valve, the flow of EGR can be
more precisely controlled.

              While most manufacturers agree that electronic EGR gives more precise control of
EGR flow rate, not all manufacturers are using it. Numerous LEV vehicles certified for the 1998
model year still use mechanical EGR systems, and in some cases,  no EGR at all. Nonetheless, the
use of EGR remains a very important tool in reducing engine-out NOx emissions, whether
mechanical or electronic.

              3.  Multiple Valves and Variable-Valve Timing

       Conventional engines have two valves per cylinder, one for intake of the air-fuel mixture and
the other for exhaust of the combusted mixture. The duration and lift (distance the valve head is
(i)  Some manufacturers have stated that EGR impacts the ability to control net air-fuel ratios tightly due to
dynamic changes in exhaust back pressure and temperature, and that the advantages of increasing EGR flow rates
are lost partly in losses in air-fuel ratio control even with electronic control of EGR. Higher EGR flow rates can be
tolerated by modern engines with more advanced combustion chambers, but EGR cooling may be necessary to
achieve higher EGR flow rates within acceptable detonation limits without significant loss of air-fuel control.

                                                                                      49

-------
Regulatory Impact Analysis
pushed away from its seat) of valve openings is constant regardless of engine speed.  As engine
speed increases, the aerodynamic resistance to pumping air in and out of the cylinder for intake and
exhaust also increases. By doubling the number of intake and exhaust valves, pumping losses are
reduced, improving the volumetric efficiency and useful power output.

       In addition to gains in breathing, the multiple-valve (typically 4-valve) design allows the
spark plug to be positioned closer to the center of the combustion chamber (as discussed above)
which decreases the distance the flame must travel inside the chamber. In addition, the two streams
of incoming gas can be used to achieve greater mixing of air and fuel, further increasing combustion
efficiency which lowers engine-out HC emissions.

       Even greater improvements to combustion efficiency can be realized by using valve timing
and lift control to take advantage of the 4-valve configuration. Conventional engines utilize fixed-
valve timing and lift across all engine speeds.  Typically the valve timing is set at a level that is a
compromise between low speed torque and high engine speed horsepower. At light engine loads it
would be desirable to close the intake valve earlier to reduce pumping losses. Variable valve timing
can enhance both low speed torque and high speed horsepower with no necessary compromise
between the two.  Variable valve timing can allow for increased swirl and intake charge velocity,
especially during low load operating conditions where sufficient swirl and turbulence tend to be
lacking. By providing a strong swirl formation in the combustion chamber, the air-fuel mixture can
mix  sufficiently,  resulting in a faster, more complete combustion, even under lean  air-fuel
conditions,  thereby reducing emissions.  Variable valve technology by itself may have somewhat
limited effect on reducing emissions. Several vehicle manufacturers estimate emission reductions
of 3%-10% for both, NMHC and NOx, but reductions could be increased when variable valve timing
is combined with optimized spark plug location and additional EGR.

       Multi-valve engines already exist in numerous federal and California certified vehicles and
are projected by CARB to become even more common.  CARB also projects that in order to meet
LEV and ULEV standards, more vehicles will have to make improvements to the induction system,
including the use of variable valve timing.

              4.  Leak-Free Exhaust System

       Leaks in the exhaust  system  can result in increased emissions, but not necessarily from
emissions escaping from the exhaust leak to the atmosphere.  With an exhaust system leak, ambient
air is  typically sucked into the exhaust system by the pressure difference created by the flowing
exhaust gases inside the exhaust pipe. The air that is sucked into the exhaust system is unmetered
and, therefore, unaccounted for in the fuel system's closed-loop feedback control, resulting in erratic
and/or overly rich fuel control. This results in increased emission levels and potentially poor drive
ability. In addition, an air leak can cause an oxidation environment to exist in a three-way catalyst
at low speeds that would hamper reduction of NOx and lead to increased NOx  emissions.

       Some vehicles currently use leak-free exhaust systems today.  These systems consist of an
improved exhaust manifold/exhaust pipe interface plus a corrosion-free flexible coupling inserted

                                                                                     50

-------
                                                 Chapter 3: Technological Feasibility
between the exhaust manifold flange and the catalyst to reduce stress and the tendency for leakage
to occur at the joint.  In addition, improvements to the welding process for catalytic  converter
canning could ensure less air leakage into the converter and provide reduced emissions. CARB and
MECA project that vehicle manufacturers will continue to incorporate leak-free exhaust systems as
emission standards become more stringent.

              5.  Improvements in Air-Fuel Ratio Control

       Modern three-way catalysts require the air-fuel ratio (A/F) to be as close to stoichiometric
operation (the amount of air and fuel just sufficient for nearly complete combustion) as possible.
This is because three-way catalysts simultaneously oxidize HC and CO, and reduce NOx.  Since HC
and CO are oxidized during A/F operation slightly lean of stoichiometry, while NOx is reduced
during operation slightly rich of stoichiometry, there exists a very small A/F window of operation
around stoichiometry where catalyst conversion efficiency is maximized for all three pollutants (i.e.,
less than 1% deviation in A/F or roughly ±0.15). Contemporary vehicles have been able to maintain
stoichiometric, or very close to it, operation by using closed-loop feedback fuel control systems.  At
the heart of these systems has been a single heated exhaust gas oxygen (HEGO) sensor. The FIEGO
sensor continuously  switches between rich and lean readings. By maintaining an equal number of
rich readings with lean readings over a given period,  the fuel control system is able to maintain
stoichiometry. While this fuel control system is capable of maintaining the A/F with the required
accuracy under steady-state operating conditions, the system accuracy is challenged during transient
operation where rapidly changing throttle conditions occur.  Also, as the sensor ages, its accuracy
decreases.

                    (a) Dual Oxygen Sensors

       Many vehicle manufacturers  have placed a second FIEGO sensor(s) downstream of one or
more catalysts in the exhaust system as a method for monitoring the catalyst effectiveness of the
federally and California mandated on-board diagnostic (OBDII) system.  In addition to monitoring
the effectiveness of the catalyst, the downstream sensors can also be used to monitor the primary
control sensor and adjust for deterioration, thereby maintaining  precise A/F  control  at higher
mileages.   Should the front  primary HEGO  sensor, which operates  in a  higher temperature
environment, begin to exhibit slow response or drift from its  calibration point, the secondary
downstream sensor can be relied upon for modifying the fuel system controls to compensate for the
aging effects. By placing the second sensor further downstream from the hot engine exhaust, where
it is also less susceptible to poisoning, the rear sensor is less susceptible to aging over the life of the
vehicle. As a result, the use of a dual  oxygen sensor fuel control system can ensure more robust and
precise fuel control,  resulting in lower emissions.

       Currently, all vehicle manufacturers use a dual oxygen sensor system for monitoring the
catalyst as part of the OBD n system.  As discussed above, most manufacturers also utilize the
secondary HEGO sensor for trim (i.e., adjustments to) of the fuel control system. It is anticipated
that all manufacturers will soon use the secondary sensor for fuel trim.
                                                                                      51

-------
Regulatory Impact Analysis
                    (b)  Universal Oxygen Sensors

       The universal exhaust gas oxygen (UEGO) sensor, also called a "linear oxygen sensor", could
replace conventional HEGO sensors. Conventional HEGO sensors only determine if an engine's A/F
is richer or leaner than stoichiometric, providing no indication of what the magnitude of the A/F
actually is. In contrast, UEGO's are capable of recognizing both the direction and magnitude of A/F
transients since the voltage output of the UEGO is "proportional" with changing A/F (i.e., each
voltage value corresponds to a certain A/F). Therefore, proportional A/F control is possible with the
use of UEGO sensors, facilitating faster response of the  fuel feedback control  system and tighter
control of A/F.

       Although some manufacturers are currently using UEGO  sensors, discussions with various
manufacturers suggest that some manufacturers are of mixed opinion as to the future applicability
of UEGO sensors. Because of their high cost, manufacturers claim that it may be cheaper to improve
FIEGO technology rather than utilize UEGO sensors. An example of this is the use of a "planar"
design for FIEGO sensors.  Planar HEGO sensors  (also known as "fast light-off HEGO sensors)
have a thimble design that is considerably lighter than conventional designs. The main benefits are
shorter heat-up time and faster sensor response.

                    (c)  Individual Cylinder A/F Control

       Another method for tightening fuel  control is to control the A/F in each individual cylinder.
Current fuel control systems control the A/F  for the entire engine or a bank of cylinders. By
controlling A/F for the entire engine or a bank of cylinders, any necessary adjustments made to fuel
delivery for the engine are applied to all cylinders simultaneously, regardless of whether all cylinders
need the that amount of fuel delivered. For example, there is usually some deviation in A/F between
cylinders. If a particular cylinder is rich, but the "bulk" A/F indication for the engine is lean, the fuel
control system will simultaneously increase the amount of fuel  delivered to all of the cylinders,
including the rich cylinder.  Thus, the rich cylinder becomes even richer having a potentially negative
effect on the net A/F.

       Individual cylinder  A/F control helps diminish variation among individual cylinders.  This
is accomplished by modeling the behavior of the exhaust gases in the exhaust manifold and using
sophisticated software algorithms to predict individual cylinder A/F. Individual cylinder A/F control
requires use of an UEGO  sensor in lieu  of the traditional HEGO sensor, and requires a  more
powerful engine control  computer.

                    (d)  Adaptive Fuel Control Systems

       The fuel control  systems of virtually all current vehicles incorporate a feature known as
"adaptive memory" or "adaptive block learn."  Adaptive fuel control systems automatically adjust
the amount of fuel delivered to compensate for component  tolerances, component wear, varying
environmental conditions,  varying fuel compositions, etc.,  to more closely maintain proper fuel
control under various operating conditions.

                                                                                      52

-------
                                                 Chapter 3:  Technological Feasibility
       For most fuel control systems in use today, the adaption process affects only steady-state
operation conditions (i.e., constant or  slowly changing throttle conditions).   Because transient
operating conditions have always provided a challenge to maintaining precise fuel control, the use
of adaptive fuel control for transient operation would be extremely valuable.  Accurate fuel control
during transient driving conditions  has traditionally been difficult because  of inaccuracies in
predicting the air and fuel flow under rapidly changing throttle conditions. Air and fuel dynamics
within the intake manifold (fuel evaporation and air flow behavior), and the time delay between
measurement of air flow and the injection of the calculated fuel mass, result in temporarily lean A/F
during transient operation.  Variation in fuel properties, particularly distillation characteristics, also
increases the difficulty in predicting A/F during transients. These can all lead to poor drive ability
and an increase in NOx emissions.

              6. Electronic Throttle Control Systems

       As mentioned above, the time delay between the air mass measurement and the calculated
fuel delivery presents one of the primary difficulties in maintaining accurate  fuel control and good
drive ability during transient driving conditions. With the conventional mechanical throttle system
(a metal linkage connected from the accelerator pedal to the throttle blade in the throttle body), quick
throttle openings can result in a lean A/F spike in the combustion chamber.  Although algorithms
can be developed to model air and fuel flow dynamics to compensate for these time delay effects,
the use of an electronic throttle control system, known as "drive-by-wire" or "throttle-by-wire," may
better synchronize the air and fuel flow to achieve proper fueling during transients (e.g., the  driver
moves the throttle, but the fuel delivery is momentarily delayed to match the inertial lag  of the
increased airflow).

       While this technology is currently used in several vehicle models, it is considered expensive
and those vehicles equipped with the feature are expensive higher end vehicles.  Because of its high
cost, it is not anticipated that drive-by-wire technology will become commonplace in the near future.

       B.  Improvements in Fuel Atomization

       In addition to maintaining a stoichiometric A/F ratio, it is also importantthat a homogeneous
air-fuel mixture be delivered at the proper time and that the mixture is finely atomized to provide
the  best  combustion characteristics and lowest emissions.   Poorly prepared air-fuel mixtures,
especially after a cold start and during the warm-up phase of the engine, result in significantly higher
emissions of unburned HC since combustion of the mixture is less complete. By providing better
fuel atomization, more efficient combustion can be attained, which should  aid in improving fuel
economy and reducing emissions. Sequential multi-point fuel inj ection and air-assisted fuel inj ectors
are examples of the most promising technologies available for improving fuel atomization.

              1. Sequential Multi-Point

       Typically, conventional multi-point fuel inj ection systems inj ect fuel into the intake manifold
by injector pairs. This means that rather than injecting fuel into each individual cylinder, a pair of

                                                                                      53

-------
Regulatory Impact Analysis
injectors (or even a whole bank of injectors) fires simultaneously into several cylinders. Since only
one of the cylinders is actually ready for fuel at the moment of injection, the other cylinder(s) gets
too much or too little fuel.  With this less than  optimum fuel injection timing, fuel puddling and
intake manifold wall wetting can occur, both of which can hinder complete combustion. Sequential
injection, on the other hand, delivers a more precise amount of fuel that is required by each cylinder
to each cylinder at the appropriate time. Because of the emission reductions and other performance
benefits "timed" fuel injection offers, sequential  fuel injection systems are very common on today's
vehicles and are expected to be incorporated in all vehicles soon.

              2.  Air-Assisted Fuel Injectors

       Another method to further homogenize the air-fuel mixture is through the use of air-assisted
fuel injection.  By injecting high pressure air into the fuel injector, and subsequently, the fuel spray,
greater atomization of the fuel droplets can occur. Since achieving good fuel atomization is difficult
when the air flow into the engine is low, air-assisted fuel injection can be particularly beneficial in
reducing emissions at low engine speeds.  In addition, industry studies have shown that the short
burst of additional  fuel needed for responsive, smooth  transient maneuvers  can be reduced
significantly with air-assisted fuel injection due to a decrease in wall wetting in the intake manifold.

       C.  Improvements to Exhaust Aftertreatment Systems

       Over the last five  years or  so, there have  been tremendous  advancements in exhaust
aftertreatment  systems.  Catalyst manufacturers  are progressively moving to palladium (Pd) as the
main precious metal in automotive catalyst applications. Improvements to catalyst thermal stability
and washcoat technologies, the design of higher cell densities, and the use of two-layer washcoat
applications are just some of the advancements  made to catalyst technology. There has also been
much development  in HC and NOx absorber technology.   The  advancements to exhaust
aftertreatment systems are probably the single most important area of emission control development.

              1. Catalysts

       As previously mentioned, significant changes in catalyst formulation, size and design have
been made in recent years and additional advances in these areas are still possible. Palladium (Pd)
is likely to continue as the precious metal of choice for close-coupled applications and will start to
see more use in underfloor applications. Palladium catalysts, however, are less resistant to poisoning
by oil-and fuel-based additives than conventional platinum/rhodium (Pt/Rh) catalysts.  Based on
current certification trends and information from vehicle manufacturers and catalyst suppliers, it is
expected that  Pd-only  and Pd/Rh catalysts will be used in  the close-coupled locations while
conventional or tri-metal (Pd/Pt/Rh) catalysts will continue to be used in underfloor applications.
Some manufacturers  have suggested that they will use Pd/Rh in lieu of tri-metal or conventional
Pt/Rh catalysts for underfloor applications.  As palladium technology continues to improve, it may
be possible for a single close-coupled catalyst to replace both catalysts.  If fact, at least one vehicle
manufacturer currently uses a single Pd-only catalyst for one of their models.  According to MECA,
new Pd-based catalysts are now capable of withstanding exposure to temperatures as high as 1100 ° C

                                                                                      54

-------
                                                 Chapter 3:  Technological Feasibility
and, as a result, can be moved very close to the exhaust manifold to enhance catalyst light-off
performance.

       In  addition to reliance on Pd and tri-metal applications, catalyst  manufacturers have
developed "multi-layered" washcoat technologies. Automotive catalysts consist of a cylindrical or
oval shaped substrate, typically made of ceramic or metal. The substrate is made up of hundreds of
very small, but long cells configured in a shape similar to a honey-comb.  The substrate is coated
with a substance containing precious metals, rare earth metals, and base-metal oxides, that is known
as the catalyst washcoat.  Typical washcoat formulations consist of precious metals which either
oxidize or reduce pollutants, base-metal  oxides, such as alumina, which provide the surface area
support for the precious metals to adhere to, and base components (rare earth metals) such as
lanthanum, ceria, and zirconia, which act as  promoters, stabilizers, and encourage storage  and
reduction of oxygen.  Conventional catalysts have had a single layer of washcoat and precious
metals applied to the catalyst substrate. Multi-layered washcoats use a combination of washcoat and
precious metals on different layers. The washcoat can be applied to the substrate such that one layer
can be applied on top of another.  The use of multi-layered washcoat technology allows precious
metals that have adverse reactions together to be separated such that catalyst durability and emission
reduction performance are significantly  enhanced.  For example, Pd and Rh can have adverse
reactions when combined together in a single washcoat formulation.   A multi-layer washcoat
architecture that uses Pd and Rh could have the Pd on the bottom layer and the Rh on the top layer
or vice versa.  Figure  3-4 illustrates the  impact coating  architecture  (multi-layered washcoat
technology) can have on emission performance.

Figure 3-4. Impact of Coating Architecture on HC and NOx Emissions
          SAE 960802: 1.8 liter 4 cyl; 100 h aged; Pd/Rh=5/l @ 50 g/cu ft.
                                                 • Single layer Pd/Rh
                                                 [~1 Two layer - Pd top
                                                 n Two layer - Pd bottom
                     THC
NOx
       Manufacturers have also been developing catalysts with substrates which utilize thinner walls
in order to design higher cell density, low thermal mass catalysts for close-coupled applications
                                                                                       55

-------
Regulatory Impact Analysis
(improves mass transfer at high engine loads and increase catalyst surface area). The greater the
number of cells there are the more surface area that exists for washcoat components and precious
metals to adhere to, resulting in more precious metal  sites available for oxidizing and reducing
pollutants. Cell densities of 600 cells per square inch (cpsi) have already been commercialized, and
research on 900 cpsi catalysts has been progressing. Typical cell densities for conventional catalysts
are 400 cpsi.

       The largest source for HC continues to be from  cold start operation where the combination
of rich A/F operation and the ineffectiveness of a still relatively cool catalyst result in excess HC
emissions. One of the most effective strategies for controlling cold start HC emissions is to reduce
the time it takes to increase the operating temperature of the catalyst immediately following engine
start-up.  The effectiveness or efficiency of the  catalyst increases as  the catalyst temperature
increases.  One common strategy is to move the catalyst closer to the exhaust manifold where the
exhaust temperature is greater (e.g., a close-coupled catalyst).  In addition to locating the catalyst
closer to the engine, retarding the spark timing, which causes combustion to occur late in the power
stroke allowing more heat to escape into the exhaust manifold during the exhaust stroke, increased
idle speed. Increased idle speed leads to a greater amount of combustion per unit time and thus to
a greater quantity of heat for heating the exhaust manifold, headpipe, and catalyst. Another strategy
is to use an electrically-heated catalyst (EHC).  The EHC consists of a small electrically heated
catalyst placed directly in front of a conventional catalyst. Both substrates are located in a single can
or container.  The EHC is powered by the  alternator, or solely from the vehicle's battery, or from a
combination of the alternator and battery.  The EHC is capable of heating up almost immediately,
assisting the catalyst that directly follows it to also heat up and obtain light-off temperature (e.g., the
catalyst temperature where catalyst efficiency is 50 percent) quickly. Manufacturers have indicated
that EHC's will probably only be necessary for a limited number of LEV/ULEV engine families,
mostly larger displacement V-8's where cold start emissions are difficult to control.

              2.  Adsorbers/Traps

       Other potential exhaust aftertreatment systems that are used in conjunction with a catalyst
or catalysts, are the HC and NOx adsorbers/traps. Hydrocarbon adsorbers are designed to trap HC
while the  catalyst is cold and unable to  sufficiently convert the HC.  They accomplish this by
utilizing an adsorbing material which holds onto the HC. Once the catalyst is warmed up, the
trapped HC are released  from the adsorption material  and directed  to the fully functioning
downstream three-way catalyst.  There are three principal methods for incorporating the adsorber
into the exhaust system.  The first is to coat the adsorber directly on the catalyst substrate.  The
advantage is that there are no changes to  the exhaust system required, but the desorption process
cannot be easily controlled and usually occurs before the catalyst has reached light-off temperature.
The second method locates the adsorber in another exhaust pipe parallel with the main exhaust pipe,
but in front of the catalyst and includes a series of valves that route the exhaust through the adsorber
in the first few seconds after cold start, switching exhaust flow through the catalyst thereafter. Under
this system, mechanisms to purge the trap  are also required. The third method places the trap at the
end of the exhaust system, in another exhaust pipe parallel to the muffler, because of the low thermal
                                                                                       56

-------
                                                Chapter 3:  Technological Feasibility
tolerance of adsorber material. Again a purging mechanism is required to purge the adsorbed HC
back into the catalyst, but adsorber overheating is avoided.

       NOx adsorbers have been researched, but according to MECA, are generally recognized as
a control for NOx resulting from reduced EGR. They are typically used for lean-burn applications
and are not applicable to engines that attempt to maintain stoichiometry all the time.

             3. Secondary Air Injection

       Secondary injection of air  into exhaust ports after cold start (e.g., the first 40-60 seconds)
when the engine is operating rich,  coupled with spark retard, can promote combustion of unburned
HC and CO in the exhaust manifold and increase the warm-up rate of the catalyst. By means of an
electrical pump, secondary air is injected into the exhaust system, preferably in close proximity of
the exhaust valve. Together with the oxygen of the secondary air and the hot exhaust components
of HC and CO, an advanced reaction ahead of the catalyst can bring about an efficient increase in
the exhaust temperature which helps the catalyst to heat up quicker. The exothermic reaction that
occurs is dependent on several parameters (secondary air mass, location of secondary air injection,
engine A/F ratio, engine air mass, ignition timing, manifold and headpipe construction, etc.), and
ensuring reproducibility of functions demands detailed individual application for each vehicle or
engine design.

             4. Insulated or Dual Wall Exhaust System

       Insulating the exhaust system is another method of furnishing heat to the catalyst to decrease
light-off time. Similar to close-coupled catalysts, the principle behind insulating the exhaust system
is to conserve heat generated in the engine for aiding the catalyst warm-up.  Through the use of
laminated thin-wall  exhaust pipes, less heat will be lost in the exhaust system, enabling  quicker
catalyst light-off.

       D.  Improvements in Engine Calibration Techniques

       Of all the technologies discussed above, one of the most important emission control strategies
is not  hardware-related.  Rather,  it's the software and,  more specifically, the algorithms and
calibrations contained within the software that are used in the power-train control module (PCM)
which  control how the various engine and  emission control components and systems  operate.
Advancements in software along with refinements to existing algorithms and calibrations can have
a major impact in reducing emissions. Confidential discussions between manufacturers and EPA
suggest that manufacturers believe emissions can be further reduced by improving and updating their
calibration techniques.  As computer technology and software continues to advance, so does the
ability of the automotive engineer to use these advancements in ways to better optimize the emission
control systems.  For example, as  processors become faster, it is possible to perform calculations
quicker, thus allowing for faster response times for things such as fuel and spark control.  As the
PCM becomes more powerful with greater memory  capability, algorithms can  become  more
sophisticated. Manufacturers have found that as computer processors, engine control sensors and

                                                                                     57

-------
Regulatory Impact Analysis
actuators, and computer software become more advanced, and, in conjunction with their growing
experience with developing calibrations, as time passes, their calibration skills will continue to
become more refined and robust, resulting in even lower emissions.

        Manufacturers have suggested to EPA that perhaps the single most effective method for
controlling NOx  emissions  will be tighter A/F  control which could be  accomplished with
advancements in calibration techniques without necessarily having to use advanced technologies,
such as UEGO sensors. Manufacturers have found ways to improve calibration strategies such that
meeting federal cold CO requirements, as well as, complying with LEV standards, have not required
the use of advanced hardware, such as EHCs or adsorbers.

       Since emission control calibrations are typically confidential, it is difficult to predict what
advancements will occur in the future, but it is clear that improved calibration techniques and
strategies are a very important and viable method for further reducing emissions.

       E. Advanced Technology

       Thus far,  the technology assessment  has focused  on conventional  emission  control
technology for vehicles with gasoline-powered spark ignition engines.  There are a number of
advanced technologies in the near horizon that may be capable even further reductions in emissions.
Examples of such technologies are fuel cells, electric vehicles, and hybrid vehicles.

       Fuel cell technology converts such fuels as methanol, natural gas, and gasoline into electrical
energy without generating the pollutants associated with internal-combustion engines. A fuel cell
is made of a thin plastic film sandwiched between two plates. Hydrogen fuel and oxygen from the
air are electrically combined in the fuel cell to produce electricity. Typically, the only by-products
are heat and water vapor.  A fuel cell coupled with an electrically powered drive-train is essentially
a quite, zero-emissions vehicle.

       Electric vehicles use electric motors to power the wheels. The electric motors are powered
by packs of batteries stored underneath the vehicle. These vehicles use many newer technologies,
such as advanced charging and regenerating systems as well as vehicle structural design.  Battery
technology, which has been the major technical limitation to date, has been and will be the focus of
much developmental work. Improved nickel-metal hydride and lithium ion batteries are two of the
battery types being analyzed for use in electric vehicles produced in the near future.

       Hybrid vehicles are typically powered by a combination of two powertrain systems. There
is usually a low or zero emitting main powertrain system (e.g., battery-powered electric motors) that
powers the vehicle during steady-state operation, when power demands are low. When more power
is required to accelerate or drive up a hill, an axillary powertrain, usually a small displacement
internal combustion engine is used. The engine may be diesel-powered, or some derivative thereof,
or an alternative-fuel powered spark ignition engine that is low emitting. Because the engine used
is small and low polluting, and the majority of operation uses the non-engine powertrain, hybrid
vehicles have the potential to be very low emitting vehicles.

                                                                                      58

-------
                                                 Chapter 3:  Technological Feasibility
       F. Technologies In-use On Current Otto-cycle HD Engines

       Otto-cycle engine manufacturers are producing heavy-duty engines equipped with substantial
emission controls. Table 3-8 provides a list of some key technologies currently being used for HD
engine emissions control. Manufacturers have introduced improved systems as they have introduced
new or revised engine models.  These systems can provide very good emissions control and many
engines are being certified to levels of less than half the current standards. Many of the technologies
have been carried over from light-duty applications.

                                       Table 3-8:
	Key Technologies for Current Engines	
 Sequential Fuel Injection/electronic control
 3 way catalyst
 pre and post catalyst heated exhaust gas oxygen sensors
 Electronic EGR
 Secondary air injection
 Improved electronic control modules
       Improving fuel injection  has  been proven to be an effective and durable strategy for
controlling emissions and reducing fuel consumption from gasoline engines. Improved fuel inj ection
will result in better fuel atomization and a more homogeneous charge with less cylinder-to-cylinder
and cycle-to-cycle variation of the air-fuel ratio. These engine performance benefits will increase
as technology advances allow fuel to be injected with better atomization.  Increased atomization of
fuel promotes more rapid evaporation by increasing the surface area to mass ratio of the inj ected fuel.
This results in a more homogeneous charge to the  combustion  chamber and more complete
combustion.  Currently, sequential multi-port fuel injection  (SFI) is used in most,  if not all,
applications because of its proven effectiveness.

       One  of the most effective means of reducing  engine-out NOx emissions is exhaust gas
recirculation (EGR). By recirculating spent exhaust gases into the combustion chamber, the overall
air-fuel mixture is diluted, lowering peak combustion temperatures and reducing NOx. Exhaust gas
recirculation is currently used on heavy-duty gasoline engines as a NOx control strategy. Many
manufacturers now use electronic EGR in place of mechanical back-pressure designs. By using
electronic solenoids to open and close the EGR valve, the flow of EGR can be more precisely
controlled.

       EPA believes that the most promising overall emission control strategy for heavy-duty
gasoline engines is the combination of a three-way catalyst and closed loop electronic control of the
air-fuel ratio.  Control of the air-fuel ratio is important because the three-way catalyst  is only


                                                                                     59

-------
Regulatory Impact Analysis
effective if the air-fuel ratio is at a narrow band near stoichiometry. For example, for an 80 percent
conversion efficiency of HC, CO, and NOx with a typical three-way catalyst, the air-fuel ratio must
be maintained within a fraction of one percent of stoichiometry.  During transient operation, this
minimal variation cannot be maintained with open-loop control. For closed-loop control, the air-fuel
ratio in the exhaust is measured by an oxygen sensor and used in a feedback loop.  The throttle
position, fuel inj ection, and spark timing can then be adjusted for given operating conditions to result
in the proper air-fuel ratio in the exhaust. Most if not all engines have been equipped with close loop
controls.  Some engines have been equipped catalysts that are achieving catalyst efficiencies in
excess of 90 percent.  This is one key reason engine and vehicle certification levels are very low.
In addition, electronic control can be used to adjust the air-fuel ratio and spark  timing to adapt to
lower engine temperatures, therefore controlling HC emissions during cold start operation.

       All FID engines are equipped with three-way catalysts.  Engine may be equipped with a
variety of different catalyst sizes and configurations. Manufacturers choose catalysts to fit their
needs for particular vehicles. Typically, federal vehicle catalyst systems are a single converter or
two converters in series or in parallel.  A converter is constructed of a substrate, washcoat, and
catalytic material.  The substrate may be metallic or ceramic with a flow-through design similar to
a honeycomb.  A high surface area coating, or washcoat, is used to provide a suitable surface for the
catalytic material. Under high temperatures, the catalytic material will increase the rate of chemical
reaction of the exhaust gas constituents. Catalyst systems on HD vehicles tend to be large with fairly
low precious metal loading. Catalyst volumes are typically 80 to 90 percent of engine  volumes.
Precious metal loadings are in the range of 1 to 4 grams per liter (g/1).

       Significant changes in catalyst formulation have been made in recent years and additional
advances in these areas are still possible. Platinum, Palladium and Rhodium (Pt, Pd, and Rh) are the
precious metals typically used in catalysts. Historically, platinum has been  widely used. Today,
palladium is being used much more widely due to its ability to withstand very  high exhaust
temperatures. In fact, some HD vehicles currently are equipped with palladium-only catalysts. Other
catalysts contain all three metals or contain both palladium and rhodium. Some manufacturers have
suggested that they will use Pd/Rh in lieu of tri-metal or conventional Pt/Rh catalysts for underfloor
applications.   Improvements  in  substrate and washcoat materials and technology have also
significantly improved catalyst performance.

       Tables 3-9 and 3-10 provide certification results from either the 1998 or 1999 model year for
various engines and vehicles. The engine data is from EPA certification data and the vehicle data
comes from California Medium-duty Vehicle certification data. California vehicles were certified
to the Tier 1 standards.  The table provide and indication of the  emissions levels that have been
achieved through the application of these technologies.
                                                                                       60

-------
                          Chapter 3:  Technological Feasibility
                 Table 3-9:
1998 or 1999 Model Year Certification Data (g/mile)
Manufacturer
Chrysler
Ford
GM
Model
Ram 3 500 Cab Chassis
Ram 3 500 Cab Chassis
Ram 3 500 Cab Chassis
Ram 2500 Pickup
Ram 3 500 Pickup
F250/F350
F250/F350 Dual rear wheel
E250 Econoline
E350
E250 Strip Chassis
E350
E350
E350
F250/F350
F250/F350 Dual rear wheel
K2500 Suburban
K2500 Pickup
K3500 Pickup
K3500 Pickup
C/K2500 4WD Pickup
C/K2500 2WD Pickup
C/K2500, 3500, Suburban,
Engine
size
8.0
8.0
8.0
8.0
8.0
5.4
6.8
5.4
5.4
4.2
6.8
6.8
6.8
5.4
6.8
5.7
5.7
5.7
7.4
6.0
6.0
6.0
GVWR
11,000
11,000
11,000
8,800
10,500
8,800-
9,700
8,800-
11,000
8,550
9,100
8,550
9,400
9,300
9,300
8,800-
9,700
8,800-
11,000
8,600
8,600
10,000
10,000
8,600
8,600
8,600-
NOx
(120k)
0.6
0.9
0.7
0.9
0.9
0.5
0.5
0.209
0.212
0.273
0.289,
0.446
0.278
0.654
0.161
0.299
0.308
0.364
0.209
0.212
0.273
0.6
0.6
0.6
0.5
0.4
0.5
0.3
0.5
HC
(120k)
0.23
0.24
0.24
0.24
0.19
0.21
0.19
0.21
0.301
0.314
0.263
0.295
0.300
0.263
0.283
0.111
0.270
0.296
0.276
0.301
0.314
0.263
0.22
0.2
0.27
0.16
0.14
0.12
0.13
0.15
                                                            61

-------
Regulatory Impact Analysis
                                      Table 3-10:
                 1998/1999 Model Year Engine Certification Data (g/bhp-hr)
Manufacturer
Chrysler
Ford
GM
Engine size
5.9
8.0
5.4
6.8
6.8
4.3
5.7
5.7
6.0
7.4
7.4
NOx
3.8
1.2
0.4
0.1
0.4
1.1
1.2
1.7
0.4
2.3
0.7
HC
0.4
0.2
0.1
0.1
0.1
0.3
0.1
0.2
0.1
0.3
0.4
       G. Chassis-based standards

       EPA is extending the California LEV standards nationwide. California began requiring some
vehicles to meet LEV standards in 1998 and the phase-in will be complete in 2001.  We have based
our technological feasibility assessment and technology projections primarily  on the mix of
technologies being used to achieve California LEV emissions levels. Cold start emissions contribute
to a larger portion of the emissions measured over the chassis-based test procedure compared to the
engine-based  test procedure.   This will  likely influence  some  of the technology  choices
manufacturers make in response to the chassis-based standards.

       Of the anticipated changes, enhancements to the catalyst systems are expected to be most
critical. Catalyst configurations are likely to continue to vary widely among the manufacturers
because manufacturers must design the catalyst configurations to fit the vehicles. One  potential
change is that  manufacturers may move the catalyst closer to the engine (close-coupled) or may place
a small catalyst close to the engine followed by a larger underfloor catalyst. These  designs provide
lower cold start emissions because the catalyst is closer to the  engine and warms up more quickly.
       Typically, the catalyst systems used in HD applications have a large total volume but with
lower precious metal content per liter compared to light-duty catalyst systems.  For 2004, we are
projecting an increase in overall precious metal loading of about 50 percent for a catalyst loading of
between 4 to 5 g/1.  We are not expecting significant increases in total catalyst volume.  The trend
                                                                                     62

-------
                                                 Chapter 3: Technological Feasibility
toward increased use of Pd and Rh is also expected to continue.  Close-coupled catalysts would
likely be Pd only.

       Calibration changes will also be important.  The engine and catalyst systems must be
calibrated to optimize the performance of the systems as a whole. Post catalyst oxygen sensors will
allow further air fuel control. Manufacturers are moving to more powerful computer systems and
EPA expects this trend to continue. Other technologies such as insulated exhaust systems may also
be used in some cases to reduce cold start emissions.

       As shown in Table 3-9, HD vehicles in California have typically been certified with full life
emissions levels in the 0.3 - 0.5 g/mile range for NOx and the 0.1 - 0.3 g/mile range for NMOG.
These levels are well within the LEV standards and provide manufacturers with head room or
compliance cushion. We expect manufacturers would equip vehicles with very similar technologies
to meet our new standards.

       H. Engine-based Standards

       As shown in Table 3-10, a few engine families are currently certified with NOx emissions
levels close to the current standards. Many others are certified with emissions levels of less than half
the standard.  Manufacturers have begun to apply advanced system designs to their heavy-duty
applications. Some newer engine families have been certified with emissions levels of 0.5 g/bhp-hr
combined NOx plus NMHC. These engines and systems feature precise air/fuel control and catalyst
designs comparable to the catalyst systems being used in LEV applications.  Based on industry input,
we believe that manufacturers will  continue the process of replacing their old engine families with
advanced engines over the next several years. New and more advanced engines are being introduced,
and we anticipate that they will be  capable of achieving our new standards.

       Catalyst systems with increased precious metal loading will be the critical hardware change
for meeting the  standards being finalized in this rule. Catalyst system volumes and precious metal
loading are likely to be similar to the  systems  discussed above for the chassis-based standards.
Engines used in vehicles above  14,000 pounds may have more rigorous duty cycles which may lead
to some catalyst enhancements.  A small increase in precious metal loading over that used in chassis-
based systems may be needed to  ensure the thermal durability of the system.  Palladium and
palladium/rhodium catalyst formulations are expected. There is likely to be less use of close coupled
systems compared with chassis-based certifications because of durability concerns. Also, there is
less emphasis on cold start emissions with the engine test than with the  chassis test.  Advanced
washcoats including layering may also be used to enhance durability.

       Optimizing and calibrating the catalyst and engine systems as a whole will also be important
in achieving the levels required by the standards.  Precise air/fuel control is critical to meeting these
standards. Increased use of air injection to control cold  start emissions may occur,  especially to
reduce NMHC emissions during cold start operation.  Also, improved EGR systems  and retarded
spark timing may be needed to reduce engine out NOx emissions levels.
                                                                                      63

-------
Regulatory Impact Analysis
       Manufacturers have noted on several occasions that they target emissions certification levels
of about half the standard. Manufacturers noted that they maintain this cushion between the standard
and their certification level in part due to the potential for in-use deterioration of catalysts and
oxygen sensors beyond that captured during the certification process. Catalysts experience wide
variations in exhaust temperature due to the wide and varied usage of vehicles in the field.  Some
vehicles may experience more severe in-use operation than is represented by the durability testing
conducted for engine certification. Manufacturers have argued that EPA should not set new standards
based on certification data because certification levels do not account for severe in-use deterioration.
Taking manufacturer practices into account, EPA would expect that engines certified in the 0.5
g/bhp-hr NOx plus NMHC range would meet a 1.0 g/bhp-hr standard.

       Catalyst system durability is a key issue in the feasibility of the  standards.  Historically,
catalysts have deteriorated when exposed to very high temperatures and this has long been a concern
for heavy-duty work vehicles.  Manufacturers have often taken steps to protect catalysts by ensuring
exhaust temperatures remain in an acceptable range. Catalyst technologies in use currently are much
improved over the catalysts used only a few years ago.  The improvements have come with the use
of palladium, which has superior thermal stability, and  through much improved  washcoat
technology. The use of rhodium with palladium will also enhance performance of the catalyst.  The
catalysts have been shown to withstand temperatures  typically experienced in  HD applications.
Manufacturers also continue  to limit exhaust temperature extremes not  only to protect catalyst
systems but also to protect the engine. EPA requirements allow manufacturers to take necessary
steps to protect engine and emission control systems  from high temperatures.

       In addition to general  comments noted above regarding the need for compliance cushion,
manufacturers presented EPA with an analysis of the otto-cycle engine emissions standards for 2004.
The analysis assumed:

       NOx catalyst efficiency of 90.9 percent at the end of the engine's useful life;

•      An engine-out NOx level of 12 g/bhp-hr;

•      A cushion of .3 g/bhp-hr for engine variability and a safety margin of 20  percent of the
       standard;

       Tailpipe NMHC levels of 15 percent of the NOx level (.26 g/bhp-hr).

       Based on these assumptions, manufacturers recommended a 2.0 g/bhp-hr NMHC plus NOx
standard.j  Manufacturers noted that a catalyst efficiency of about 97 percent would be needed to
meet a 1.0 g/bhp-hr standard and that their assessments of post-2000 catalysts indicate worst case
performance well  below this level. The manufacturers'  recommended 2.0 g/bhp-hr standard seems
to indicate that compliance cushions greater than half the standard are needed.
(j)  12.0 g engine out x.091 for catalyst efficiency + 0.65 for compliance cushion = 1.74 g NOx. The difference
between 2.0g and 1.74 g is reserved for NMHC emissions.)
                                                                                       64

-------
                                                 Chapter 3:  Technological Feasibility
       Manufacturers state that their catalyst assumptions represented catalyst deterioration based
on worst case vehicle operation (highly loaded operation, high exhaust temperatures). Details of the
catalyst were not available except that manufacturers stated that the catalyst represented post-2000
catalyst technology. Due to the lack of detail, it is difficult to evaluate the assumption. However,
EPA believes that this assumption is somewhat conservative given the recent developments in
catalyst technology, the lead time available, and methods available to protect catalysts under worst
case vehicle operation.

       Engine-out NOx levels are also critical to the  analysis.  In their analysis, manufacturers
assumed engine-out NOx levels of 12 g/bhp-hr, based on manufacturer development data for one
engine.  EPA does not believe that the engine-out NOx level of 12 g/bhp-hr is a reasonable or
representative assumption. Other available data indicates that several engines have engine-out NOx
emissions well below this level in the 6 to 10 g/bhp-hr range. Also, a previous assessment of engine
standards presented to EPA by one manufacturer assumed much lower engine-out NOx levels.k EPA
does not believe that the current standards have encouraged manufacturers to place a high priority
on engine-out emissions levels. For recent engines, catalysts have provided the majority of needed
emissions control.

       EPA also further considered the engine variability factor  of 0.3  g/bhp-hr built into the
manufacturers analysis.  The analysis as presented assumes a 12 g/bhp-hr engine-out NOx level.
Manufacturer data for the developmental engine suggests that 12 g/bhp-hr is the worst case engine-
out level anticipated (the actual highest test point recorded was 12.65).  It appears to EPA that
manufacturers double counted engine variability by using the worst case engine data and an engine
variability factor.  Using engine-out NOx levels of 12 g in the analysis but without the engine
variability factor yields a NOx + NMHC level of 1.6 g/bhp-hr. Without including a safety margin,
which may be appropriate considering the analysis is already based on worst case engine and catalyst
assumptions, the level would be 1.3 g/bhp-hr. To reach the 1.0 g/bhp-hr level with this engine and
a 20 percent safety margin,  a catalyst efficiency  of 94  percent would be needed.  The catalyst
efficiency would need to be  93 percent if the 20 percent safety margin were not included in the
analysis.

       EPA believes that the standards will require manufacturers to focus some effort on engine-out
emissions control and that engine-out NOx levels in the 6 to 8 g/bhp-hr are reasonably achievable.
Some engines  are already in this range.  For other engines, some recalibration of engine systems
including the EGR system and perhaps some modest hardware changes to those systems would be
necessary. EGR plays a key role in reducing engine-out NOx and system redesign may allow more
effective use of this technology.

       When coupled with a catalyst with worst case efficiencies in the 91 to 93 percent range, these
engines could  achieve the standards.  Of course with higher catalyst efficiencies, manufacturers
(k)  The details of this analysis are considered Confidential Business Information.

                                                                                       65

-------
Regulatory Impact Analysis
would not have to achieve lower NOx engine-out levels. Catalyst efficiencies of about 93 percent
would allow manufacturers to maintain compliance margins in the range of 25 and 45 percent of the
standard. EPA believes these margins are sufficient considering the analysis is also based on worst
case catalyst efficiencies.

       To help address phase in concerns that could arise for manufacturers, EPA is finalizing a
modified ABT program for engines.  The averaging, banking, and trading (ABT) program can be an
important tool  for  manufacturers  in  implementing  a  new  standard.   The program allows
manufacturers to comply with the more stringent standards by introducing emissions controls over
a longer period of time, as opposed to during a single model year. Manufacturers plan their product
introductions well in advance. With ABT, manufacturers can better manage their product lines so
that the new standards don't interrupt their product introduction  plans. Also, the program also
allows manufacturers to focus on higher sales volume vehicles first and use credits for low sales
volume vehicles. EPA believes manufacturers have significant opportunity to earn credits in the pre-
2004 time frame.

       Considering all of these factors, EPA believes that  the 1.0 g/bhp-hr NOx plus NMHC
standard is the appropriate standard for HD otto-cycle engines in the 2004 time frame. Certification
levels of 0.5 g/bhp-hr NOx plus NMHC have been achieved on  recently introduced engines of
various sizes. EPA believes that the standard provides sufficient opportunity for manufacturers to
maintain a compliance margin. As manufacturers continue with normal product plans between now
and 2004, improved engines will continue to replace older models.  The ABT program is available
for manufacturers who have not completely changed over to new engine models by 2004. ABT
provides manufacturers with the opportunity to earn credits  prior to 2004 and use the credits to
continue to offer older engine models that have not yet been redesigned or retired by 2004.

IV. On-board Diagnostics for HD  Diesel and  Otto-cycle Engines

       To meet customer demands, manufacturers of heavy-duty  engines currently use on-board
diagnostics (OBD) to electronically monitor engine parameters to ensure proper engine performance
and to assist in malfunction diagnostics and repair90.  Because EPA expects manufacturers to
implement electronically controlled emission control  strategies such as EGR and fuel injection rate
shaping, EPA is promulgating OBD requirements for heavy-duty engines used in vehicles up to
14,000 pounds, gross vehicle weight (GVW) to ensure that emission-control components meet
certain performance standards.  These requirements  are intended to ensure that emission-control
components remain effective in-use. The California Air Resources Board (CARB)  has already
implemented similar requirements.

       EPA believes that the new requirements are already technologically feasible. All classes of
HD vehicles currently employee some form of on-board diagnostics for performance purposes, and
many of these systems are highly sophisticated.  In  addition, HD vehicles up to 14,000 pounds
already have to meet regulatory OBD requirements in California.  Finally, federal and California
emission driven OBD regulatory requirements have been in place for Otto-cycle and diesel light-duty
vehicles for a number of years.  The technology  necessary  to perform OBD of HD vehicles is

                                                                                     66

-------
                                                Chapter 3:  Technological Feasibility
available today. The new emission control technologies employed in 2004 will also lend themselves
easily to OBD.  For example, LD vehicle manufacturers have been monitoring EGR systems for
OBD for a number of years.

       As discussed previously, EPA does not expect diesel engine manufacturers to utilize
aftertreatment devices in order to  achieve the 2004  HD  standards,  or the 2007  supplemental
requirements.  However, in the past engine manufacturers  have used diesel oxidation to provide
typically a 20 to 30 % reduction in PM on some light- and medium-heavy duty engine families. For
these diesel oxidation catalysts, a complete failure of the catalysts would not result in an exceedance
of the 1.5 times the standard threshold, therefore monitoring of catalyst performance would not be
a requirement. For PM traps and lean NOx catalysts, neither technology is anticipated for the 2004
model  year.   However, in the event a manufacturer did employee either of these types  of
aftertreatment devices, EPA believes a back-pressure sensor would be feasible to monitor a PM trap
for a complete failure of the trap, and either a chemical sensor (such as the oxygen sensors used for
gasoline 3-way catalysts)  or potentially a temperature sensor could be  used to monitor  the
performance of a lean NOx catalyst.

       The final rule requires that PM traps whose failure would result in an exceedance of 1.5 times
the PM standard must be monitored. However, the rule does not require that the monitor detect an
exceedance of 1.5 times the standard threshold. Rather, the requirement is to detect a complete
failure of the device. We define complete failure as a sudden drop in exhaust back-pressure below
that of a clean or unloaded trap under monitoring conditions specified by the manufacturer. Current
pressure monitoring sensors are clearly capable of performing this detection.

       Direct emission measurement has been identified as an important technology to achieve
diesel engine closed-loop feedback control and to achieve after-treatment OBD. Researchers already
have achieved promising results on  a compact NOx sensor  that  is capable of measuring real-time
NOx within 10% accuracy of laboratory-grade instruments under a wide range of operating
conditions, including the temperature, pressure, and oxygen concentration typical of diesel engine
exhaust. This breakthrough technology could be used for closed-loop control, and, because it can
accurately measure NOx in the 100 ppm range, it would enable monitoring of NOx aftertreatment
technologies.91'92   The most  recent of these papers (Kato et. al., 1999) provides an in depth
discussion the accuracy, repeatability, and durability of an on-board NOx sensor,  as well strategies
for using the sensor for closed loop control and OBD monitoring of an active lean NOx absorber.

       The federal requirements for OBD, as they  exist today,  require manufacturers to monitor
emission related  powertrain  components,  OBD  does not monitor  actual regulated  pollutant
emissions. It is possible that in the future the on-board measurement of actual emission performance
may become  feasible. EPA is following the development of  a number of emerging on-board
emission measurement technologies which may lend themselves to regulatory requirements in the
future. These technologies include in-cylinder measurement devices, on-board PM measurement
devices, and predictive emission measurement systems such as neural networks.   Crank-angle
resolved pressure and/or temperature measurements would allow for NOx emission prediction, based
on the current understanding  of NOx formation.93  Piezo-electric  and infrared  pressure sensing

                                                                                     67

-------
Regulatory Impact Analysis
technologies are currently used to measure crank-angle resolved in-cylinder pressure.  Based on
recent advances in sensor durability,94'95 EPA expects that future advances could allow their use on-
board.  Lastly, neural networks have recently demonstrated a technique for accurately predicting
emissions based solely on currently measured engine parameters.  One study has shown excellent
correlation  between  predicted NOx  and PM measurement  with  respect to actual emissions
measurements.96
                                                                                     68

-------
                                              Chapter 3:  Technological Feasibility
Chapter 3 References

l.Heywood J.B.: Internal Combustion Engine Fundamentals, McGraw-Hill, Inc, New York, p.
590, 1988.

2.Dodge L.G., D.M. Leone, D.W. Naegeli, D.W. Dickey, K.R. Swenson: "A PC-Based Model
for Predicting NOx Reductions in Diesel Engines," SAE paper 962060, p. 149, 1996.

3.Ibid.

4.1bid

5. Dickey D.W., T.W. Ryan HI, A.C. Matheaus: "NOx Control in Heavy-Duty Engines-What is
the Limit?", SAE paper 980174, 1998. Dickey; and, ZelenkaP., H. Aufmger, W. Reczek, W.
Cartellieri: "Cooled EGR-A Key Technology for Future Efficient HD Diesels," SAE paper
980190, 1998.

6. Kohketsu S., K. Mori, K. Sakai, T. Hakozaki: EGR Technologies for a Turbocharged and
Intercooled Heavy-Duty Diesel Engine," SAE paper 970340,  1997; Baert R., D.E. Beckman,
A.W.M.J. Veen: "EGR Technology for Lowest Emissions," SAE paper 964112, 1996; and,
Heavy-duty Engine Working Group, Mobile Source Technical Advisory Subcommittee of the
Clean Air Act Advisory Committee, "Phase 2 of the EPA HDEWG Program - Summary
Document", available in EPA Air Docket A-98-32.

7.Dickey D.W., T.W. Ryan IE, A.C. Matheaus: "NOx Control in Heavy-Duty Engines-What is
the Limit?", SAE paper 980174, p. 9, 1998.

S.Zelenka P., H. Aufmger, W. Reczek, W. Cartellieri: "Cooled EGR-A Key Technology for
Future Efficient HD Diesels," SAE paper 980190, p. 47, 1998.

9.Kohketsu S., K. Mori, K. Sakai,  T. Hakozaki: EGR Technologies for a Turbocharged and
Intercooled Heavy-Duty Diesel Engine," SAE paper 970340, p. 97, 1997.

lO.ZelenkaP., p. 46.

11.Kohketsu S., p. 97.

12.Baert R., D.E. Beckman, A.W.M.J. Veen: "EGR Technology for Lowest Emissions," SAE
paper 964112, p. 183, 1996.

13.Ibid

14.Kohketsu S., p. 100.

IS.ZelenkaP., p. 49.
                                                                                 69

-------
Regulatory Impact Analysis
16.BaertR., p. 184.

IT.Kohketsu S., p. 100.

IS.ZelenkaP., p. 47.

19.Kohketsu S., p. 102.

20.BaertR., pp. 181,182.

21.Ibid.

22.ZelenkaP. pp. 48,50.

23. Ibid.

24.Kreso A.M., J.H. Johnson, L.D. Gratz, S.T. Bagley, D.G. Leddy: "A Study of the Vapor- and
Particle-Phase Sulfur Species in the Heavy-Duty Diesel Engine EGR Cooler," SAE paper
981423, pp. 1-2,7,  1998.

25. Ibid.

26.McKinley T.L: "Modeling Sulfuric Acid Condensation in Diesel Engine EGR Coolers," SAE
paper 970636, p. 211, 1997.

21.Ibid.

28.Kreso A.M., p. 7.

29.Ibid.

SO.Banzhaf M., R. Lutz: "Heat Exchanger for Cooled Exhaust Gas Recirculation," SAE paper
971822, pp. 4,6, 1997.

Sl.ZelenkaP., p. 50.

32.Banahaf M., p. 3,4.

33.1bid.

34.Dekker H.J., W.L. Sturm: "Simulation and Control of a HD Diesel Engine Equipped with
New EGR Technology," SAE paper 960871, p. 214, 1996.

35.ZelenkaP., p. 48.

36.Kohketsu S. p.98.
                                                                                  70

-------
                                              Chapter 3:  Technological Feasibility
37.BaertR., p. 179.

38. Anon., "Electronic Throttle Valve From Pierburg,", Diesel Progress, August, 1998. Copy
Available in EPA Air Docket A-98-32, Docket Item #U-D-10.

39.Fielden, G., Lucas Inc., Vandalia, OH: Phone conversation with Matt Spears,
EPCD/OMS/EPA, January 7,1999. Summary of phone call available in EPA Air Docket A-98-
32, Docket Item # II-E-01.

40.Boehner W., K. Hummel: "Common Rail Injection System for Commercial Diesel Vehicles",
SAE paper 970345, p. 140, 1997.

41.  Dickey, SAE 981074

42.  Boehner, 970345; and UchidaN, K. Shimokawa, Y. Kudo, M. Shimoda: "Combustion
Optimization by Means of Common Rail Injection System for Heavy-Duty Diesel Engines", SAE
paper 982679, 1998.

43.  Heywood, J.B., Internal Combustion Engine Fundamentals. McGraw-Hill, Inc., New York,
p. 643-644, 1988.

44.Guerrassi N., P. Dupraz: "A Common Rail Injection System for High Speed Direct Injection
Diesel Engines", SAE paper 980803, p. 19, 1998.

45.BoehnerW., p. 134.

46.Youngblood, J.R.: "Cummins New Midrange Fuel System", SAE TOPTEC presentation,
"Diesel Technology for the New Millennium", April 21-22, 1998. Available in EPA Air Docket
A-98-32, Docket Item H-D-Ol.

47. Stover, T.R., D.H. Reichenbach, E.K. Lifferth: "The Cummins Signature 600 Heavy-Duty
Diesel Engine", SAE paper 981035, p. 8, 1998.

48.Osenga M.: "Cat Gears Up Next Generation Fuel Systems", Diesel Progress, North American
Ed., P. 82, August 1998. Available in EPA Air Docket A-98-32, Docket Item #U-D-03.

49.1bid., pp. 88-89.

SO.Ibid., pp. 82-87.

Sl.Youngblood, J.R.

52.1kegami, M., K. Nakatani,  S. Tanaka, K. Yamane: "Fuel Injection Rate Shaping and Its  Effect
on Exhaust Emissions in a Direct-Injection Diesel Engine Using a Spool Acceleration Type
Injection System", SAE paper 970347, p. 163,  1997.
                                                                                 71

-------
Regulatory Impact Analysis
53.1bid, p. 174.

54.Uchida N, K. Shimokawa, Y. Kudo, M. Shimoda: "Combustion Optimization by Means of
Common Rail Injection System for Heavy-Duty Diesel Engines", SAE paper 982679, p. 3, 1998.

SS.DickeyD.W., p. 13.

56.UchidaN., p. 4.

57.Coldren D.R.: "Advanced Technology Fuel System for Heavy Duty Diesel Engines", SAE
paper 973182, p. 10, 1997.

SS.DickeyD.W., p. 9.

59.    Johnson, J., Bagley, S., Gratz, L., Leddy, D., "A Review of Diesel Particulate Control
Technology and Emissions Effects - 1992 Horning Memorial Award Lecture", SAE Paper
940233, 1994.

60.    Meeting between EPA and the Manufacturers of Emission Controls Association, April
1995.

61.    Voss, K., Bulent, Y., Hirt, C, and Farrauto, R., "Performance Characteristics of a Novel
Diesel Oxidation Catalyst", SAE Paper 940239, 1994.

62.    Johnson, J., Bagley, S., Gratz, L., Leddy, D., "A Review of Diesel Particulate Control
Technology and Emissions Effects - 1992 Horning Memorial Award Lecture", SAE Paper
940233, 1994.

63.    Kawanami, M., Horichi, M., Klein, H., Jenkins, M., "Development of Oxidation and de-
NOx Catalysts for High Temperature Exhaust Diesel Trucks, SAE Paper 981196.

64.    "Demonstration of Advanced Emission Control  Technologies Enabling Diesel-Powered
Heavy-Duty  Engines to Achieve Low Emission Levels: Interim Report Number 1: Oxidation
Catalyst Technology" prepared for the Manufacturers of Emission Control Associations,
December 1998, copies available in EPA Air Docket A-98-32, Docket Item U-D-07.

65.    Hawker, P., et. al., "Effect of a Continuously Regenerating Diesel Particulate Filter on
Non-Regulated Emissions and Particulate Size Distribution", SAE paper 980189, February 1998.

66.    "Demonstration of Advanced Emission Control  Technologies Enabling Diesel-Powered
Heavy-Duty  Engines to Achieve Low Emission Levels: Interim Report Number 2: Diesel
Particulate Filter Technology" prepared for the Manufacturers of Emission Control Associations,
December 1998, copies available in EPA Air Docket A-98-32, Docket Item U-D-08.

67. Wall, J.C., Cummins Engine Co., Diesel Fuel Composition for Future emissions
Regulations.  Panel discussion, SAE International Fall Fuels and Lubricants Meeting and


                                                                                  72

-------
                                               Chapter 3: Technological Feasibility
Exposition, October 21, 1998.

68. Wall, J.C., Cummins Engine Co., Diesel Fuel Composition for Future emissions
Regulations. Panel discussion, SAE International Fall Fuels and Lubricants Meeting and
Exposition, October 21, 1998.

69. Peters, A., et al., Catalytic NOx Reduction on a Passenger Car Diesel Common Rail Engine.
SAE Technical Paper Series, No. 980191, 1998.

70. Peters, A., et al., Catalytic NOx Reduction on a Passenger Car Diesel Common Rail Engine.
SAE Technical Paper Series, No. 980191, 1998.

71. Wall, J.C., Cummins Engine Co., Diesel Fuel Composition for Future emissions
Regulations. Panel discussion, SAE International Fall Fuels and Lubricants Meeting and
Exposition, October 21, 1998.

72. Engler, B.H., et al., Catalytic Reduction of NOx with Hydrocarbons Under Lean Diesel
Exhaust as Conditions.  SAE Technical Paper Series, No. 930735, 1993.

73. Mitsubishi Motors, 4G93 GDI Engine Technical Information: Reduced Exhaust Emissions
1997.

74. Engler, B.H., et al., Catalytic Reduction of NOx with Hydrocarbons Under Lean Diesel
Exhaust as Conditions.  SAE Technical Paper Series, No. 930735, 1993.

75. Brown, K.F., Diesel After treatment Technologies. Presentation at the SAE Topical
Technical Seminar Diesel Technology for the New Millennium, Schaumburg, IL, April 22, 1998.

76.    Heavy-duty Engine Working Group, Mobile Source Technical Advisory Subcommittee of
the Clean Air Act Advisory Committee, "Phase I Final Report", April, 1997, available in EPA
Air Docket A-98-32, Docket Item # H-A-02.

77.    Lee, D., Ryan, T., "Fuel Property Effects on NOx Emissions in Year 2004 Emissions
Class Diesel Engines - Final Report", December 5, 1996, Southwest Research Institute, available
in EPA Air Docket A-98-32, Docket Item # H-A-03

78.    Heavy-duty Engine Working Group, Mobile Source Technical Advisory Subcommittee of
the Clean Air Act Advisory Committee, "Phase 2 of the EPA HDEWG Program  - Summary
Document",  available in EPA Air Docket A-98-32.

79.Banzhaf M., p. 4.

80. See "Summary of CBI information regarding proposed HD Supplemental Test
Requirements", available in EPA Air Docket A-98-32.
                                                                                   73

-------
Regulatory Impact Analysis
81.  See ""Summary of Conference Call between US EPA and Honeywell Turbocharging
Systems on December 22, 1999 regarding 2004 On-highway Heavy-duty Diesel Proposal",
available in EPA Air Docket A-98-32.

82.BaertR., p. 175.

83.DekkerHJ., pp. 209,212.

84.Kohketsu S., pp. 97, 104.

SS.ZelenkaP., pp. 45,  56.

86.  Energy and Environmental Analysis, "Benefits and Cost of Potential Tier 2 Emission
Reduction Technologies", Final Report, November 1997.

87. California Air Resources Board, "Low Emission Vehicle and Zero Emission Vehicle
Program Review, Staff Report", November 1996.

88. California Air resources Board, "Proposed Amendments to California Exhaust and
Evaporative Emissions Standards and Test Procedures for Passenger Cars, Light-duty Trucks,
and Medium-duty Vehicles, LEVH", Staff Report, September 18, 1998.

89. Arcadis Geraghty and Miller, "Cost Estimates for Heavy-duty Gasoline Vehicles" Final
Report, September 30, 1998.

90.  EPA Technical Memorandum "Documentation of Sophisticated On-board Diagnostic
Systems on Current Heavy-duty Diesel Engines", Todd Sherwood, Available in EPA Air Docket
A-98-32.

91.Kato N., H. Kurachi, Y. Hamada: "Thick Film ZrO2 NOx Sensor for the Measurement of
LowNOx  Concentration",  SAE paper 980170, pp. 76-77, 1998.

92.  Kato N., N. Kokune, B. Lemire, T. Walde: "Long term stable NOx sensor with integrated in-
connector control electronics", SAE paper 1999-01-0202, also see memorandum from Mr. Todd
Sherwood to EPA Air Docket A-98-32 summarizing this paper

93.Dodge L.G, pp.  148-149.

94. Atkins R. A, C.E. Lee, H.F.  Taylor: "Fiber-Optic In-cylinder Pressure Sensor Developed",
Diesel & Gas Trubine Worldwide, p. 16b, April 1995.

95.Poorman T.J., L. Xia, M.T. Wlodarczyk:  "Ignition system-embedded fiber-optic combustion
pressure sensor for engine control and monitoring", SAE paper 970845, p.l, 1997.

96. Atkinson C. : "Emissions Prediction for On-Board Diagnostics and Engine Control", SAE
TOPTEC presentation, "Diesel Technology for the New Millennnium", April 21-22, 1998, Air
                                                                                  74

-------
                                            Chapter 3:  Technological Feasibility
Docket A-98-32, item no. H-D-02.
                                                                             75

-------
Regulatory Impact Analysis
    CHAPTER 4:   ECONOMIC IMPACT OF HD DIESEL

                                STANDARDS

I.  Methodology

       EPA previously analyzed the costs of the 2004 FTP heavy-duty diesel standards for the 1997
FRM.  That economic analysis was based on a study conducted by ICF Incorporated and Acurex
Environmental Corporation, which analyzed the potential costs of a wide variety of technologies.1
This current analysis is generally a re-analysis of those previous analyses (unless noted otherwise),
but also addresses new requirements such as the NTE requirements. The reader should refer to the
previous analyses for additional information and background. In the previous analyses, all costs
were described in terms of 1995 dollars.  Where these costs were used in this analysis, they were
adjusted upward by 9.3 percent to be equivalent to 1999 dollars. This adjustment was based on the
Consumer Price Index. (Note: This adjustment was not made in the Draft RIA.) As was done in the
proposal, EPA is proj ecting costs assuming that testing will be completed in time for the 2004 model
year, even though the  supplemental requirements are being finalized for the 2007 model year. We
believe that many manufacturers will choose (as a convenience) to incorporate the minor calibration
changes necessary to comply with these requirements during the  2004  model year, rather than to
modify their 2004 designs for the 2007 model year. Since this assumption means that manufacturers
would incur the testing costs three years earlier than required, it results in a slight increase in the net
present value of the costs.

       While the following analysis is based on a relatively uniform emission control  strategy for
designing the different categories of engines, this is not intended to suggest that a single combination
of technologies will actually be used by all manufacturers.  In fact, depending on basic engine
emission characteristics, EPA expects that control technology packages will gradually be fine-tuned
to each application. Furthermore, EPA expects manufacturers to use averaging, banking, and trading
programs as a means to deploy varying degrees of emission control  technologies on different
engines.  EPA nevertheless believes that the projections presented here provide a cost estimate
representative of the different approaches manufacturers may ultimately take.

       Costs of control include variable costs (for incremental hardware costs, assembly costs, and
associated markups) and fixed costs (for tooling,  R&D, and certification).  Variable costs  are
marked up  at  a  rate  of 29 percent to account for manufacturers' overhead and profit.2  For
technologies sold by a supplier to the engine manufacturers, an additional 29 percent markup is
included for the supplier's overhead and profit. Estimated variable costs for new technologies (i.e.,
EGR and VGT) include a ten percent markup to account for increased warranty costs.  Fixed costs
for R&D are assumed to be incurred over the seven-year period from 1996 through 2002, tooling and
certification costs are assumed to be incurred one year ahead of initial production. Fixed costs are
increased by seven percent for every year before the start of production  to reflect the time value of
money. This total preproduction cost  is then amortized at the same rate over a five-year period
during  which the manufacturer would be able to recoup the fixed costs.  The analysis also includes

                                                                                   76

-------
                                   Chapter 4: Economic Impact of HDDE Standards
consideration of lifetime operating costs where applicable. Projected costs were derived for three
service classes of heavy-duty diesel vehicles, as depicted in Table 4-1.

       In some cases, EPA expects that there may be significant overlap between technologies
needed to reduce NOx  emissions for compliance with 2004 model year  standards and those
technologies that offer other benefits for improved fuel economy and engine performance or for
better control of HC or paniculate emissions. In the absence of future standards, manufacturers
would have continued research on and eventually deploy many technological upgrades to improve
engine performance or more cost-effectively control emissions. Specifically, this is most appropriate
for the application of VGT and improved fuel injection technologies, as is discussed in Sections
A(3) and A(4) of this chapter. For those cases, EPA is assuming that only a fraction of the fixed and
variable costs are attributable to emission control.

                                       Table 4-1:
                          Service Classes of Heavy-Duty Vehicles
Service Class
Light
Medium
Heavy
Vehicle Class
2B-5
6-7
8
GVWR (Ibs.)
8,500- 19,500
19,501 -33,000
33,001 +
II. Technologies for Meeting the New Standards

       The following discussion provides a description and estimated costs for those technologies
EPA projects will be needed to comply with the new emission standards.  EPA believes that a small
set of technologies represent the primary changes manufacturers must make to meet the standards
in this rule. Other technologies applied to heavy-duty engines, before or after implementation of new
emission standards, will make smaller secondary contributions to controlling NOx orHC emissions
and are therefore considered secondary improvements for this analysis.  In this category are design
changes such as improved oil control, optimized catalyst designs, and variable-valve timing. Lean
NOx catalysts are also considered secondary technologies in this analysis, not because NOx control
is an incidental benefit, but because it appears unlikely that they will be part of 2004 model year
technology packages. Modifications to fuel injection systems will also continue independently of
new standards, though some further development with a focus on reducing NOx or HC emissions
would be evaluated.  While a few engines must reduce HC emission levels, EPA expects the
combination of technologies selected for meeting NOx and  particulate emission standards to be
sufficient for adequate control of HC emissions.
                                                                                     77

-------
Regulatory Impact Analysis
       The technology analysis includes an analysis of the baseline technology being used by
manufacturers to meet the 1998 emission standards and future technologies that will be used to
improve engine designs through model year 2003. Specification of the future technologies is based
on an  observation of current trends in heavy-duty engine technology.  The baseline control
technologies  being assumed  for  engines meeting 1998  emission standards  in 2003 include
technologies that contribute directly to lower NOx emissions and a variety of engine improvements
with only secondary benefits for  NOx control.  The  assumed  baseline  scenario includes  full
utilization of electronic controls and unit injectors. Except for urban bus engines, one-third to one-
half of diesel engines are expected to include unit injectors designed to  operate independently of
engine speed; one example of such an injector is the Hydraulically-activated, Electronically-
controlled Unit Injector (HEUI), whichis currently manufactured for several Caterpillar and Navistar
engine models. Another example  is  the newer, more advanced, Next Generation Electronic Unit
Injector (NGEUI) developed by Caterpillar. Also, these engine models are assumed to have some
basic manipulation of the fuel inj ection profile (for "rate shaping"). Variable-geometry turbochargers
are expected for several engine lines as manufacturers aim for better performance and fuel economy,
and potentially for additional  braking capacity.  Light  and medium heavy-duty engines may be
modified to further reduce the contribution of lubricating oil to parti culate emissions. Manufacturers
may also pursue variable-valve timing or upgrade to four valves per cylinder for improved engine
performance. While EPA is not assuming EGR to be included among the baseline technologies,
EPA recognizes that some manufacturers may actually incorporate EGR into future engines to offset
fuel consumption increases associated with the 1998 NOx standard (due to injection timing retard).
For example, DDC recently announced the introduction of their new Series 50 heavy-duty bus and
truck engine, which is being equipped with EGR and VGT  in the 2000 model year.3  DDC
specifically noted, in its announcement, that this was being done as an alternative to retarded
injection timing in order to minimize fuel consumption.  Thus, this assumption, that 100 percent of
the cost of adding EGR is attributable to compliance with the standards in this rule, is conservative
and actual compliance costs for the these standards maybe significantly lower than is estimated here.

       Compliance costs for 2004  and later  model year engines are based  on an assumed
combination of primary technology upgrades.  Modifications to basic engine design features can
improve intake air characteristics  and  distribution during combustion.  Manufacturers are also
expected to use upgraded electronics and advanced fuel-inj ection techniques and hardware to modify
various fuel injection parameters for higher pressure, further rate shaping, and some split injection.
EPA also expects that all engines will incorporate cooled exhaust gas recirculation and many will
incorporate variable geometry turbochargers.   The  costs of these individual  technologies are
considered in the following paragraphs and summarized in Table 4-2.  The costs of secondary
improvements are not included in this analysis since  they are  not expected to be needed for
compliance with these emission standards. The reader is referred to the  RIA for the June 2, 2000
NPRM that proposed new standards for the 2007 model year (Docket A-99-06) for more information
regarding the potential costs of these secondary technologies. In that NPRM, EPA projected that
many of these technologies would  be available for the 2007 model year.
                                                                                      78

-------
                                    Chapter 4: Economic Impact of HDDE Standards
                                        Table 4-2
                             2004 Model Year Cost Estimates

                        Light Heavy-Duty Diesel Vehicles (Dollars per Engine)
Item
Cooled EGR (high-flow)
Combustion optimization
Improved fuel injection
Variable geometry turbochargers
Onboard diagnostics
Emission map testing
Certification
Fixed
Cost
42
22
9
15
1
2
2
Variable Cost
215
0
135
188
0
0
0
Operating Cost
8
0
0
0
0
0
0
Fraction of Cost
For Emissions*
100%
100%
50%
50%
100%
100%
100%
* Costs listed in the table are the full costs for adding each of the technologies. However, because
both fuel inj ection improvements and variable geometry turbochargers provide performance benefits
not related to emissions control, and because these technologies may be in use prior to 2004, only
fractions of the full costs are used for the cost-effectiveness analysis.

                       Medium Heavy-Duty Diesel Vehicles (Dollars per Engine)
Item
Cooled EGR (high-flow)
Combustion optimization
Improved fuel injection
Variable geometry turbochargers
Onboard diagnostics
Emission map testing
Certification
Fixed
Cost
103
55
20
35
0
5
2
Variable Cost
242
0
127
248
0
0
0
Operating Cost
49
0
0
0
0
0
0
Fraction of Cost
For Emissions*
100%
100%
50%
50%
100%
100%
100%
* Costs listed in the table are the full costs for adding each of the technologies. However, because
both fuel inj ection improvements and variable geometry turbochargers provide performance benefits
not related to emissions control, and because these technologies may be in use prior to 2004, only
fractions of the full costs are used for the cost-effectiveness analysis.
                                                                                      79

-------
Regulatory Impact Analysis
                        Heavy Heavy-Duty Diesel Vehicles (Dollars per Engine)
Item
Cooled EGR (high-flow)
Combustion optimization
Improved fuel injection
Variable geometry turbochargers
Onboard diagnostics
Emission map testing
Certification
Fixed
Cost
103
55
20
35
0
5
9
Variable Cost
336
0
140
338
0
0
0
Operating Cost
104
0
0
0
0
0
0
Fraction of Cost
For Emissions*
100%
100%
50%
50%
100%
100%
100%
* Costs listed in the table are the full costs for adding each of the technologies. However, because
both fuel inj ection improvements and variable geometry turbochargers provide performance benefits
not related to emissions control, and because these technologies may be in use prior to 2004, only
fractions of the full costs are used for the cost-effectiveness analysis.

       A.  Primary Technologies

       The  following  discussion presents the projected  costs  of the primary technological
improvements expected for complying with the new emission standards, first for fixed costs, then
for hardware and operating costs  of the individual technologies.

       The  cost analysis anticipates an extensive ongoing research  program to develop these
technologies. While most of this R&D will be needed to develop new technologies for  reducing
emissions, some  will  be needed  to  verify  emission performance  for compliance  with the
supplemental standards and OBD requirements. R&D costs account for over 90 percent of the total
fixed  costs per engine  detailed in Table 4-2.   Retooling is another fixed cost factored into the
analysis. Retooling costs will be incurred about one year before initial production and are discounted
accordingly.

       Manufacturers will also incur costs for certifying the range of engine families to the emission
standards.   EPA previously developed a detailed methodology for calculating certification costs.4
Adjusting those figures to convert them to 1999 dollars (using the Consumer Price Index) results in
an estimated certification cost of $250,000 per engine family. This estimate, which is used here, is
the same estimate that was used in the 1997 analysis (after being adjusted to 1999 dollars).  This is
because EPA believes that the new supplemental steady-state certification requirement will not
significantly affect certification costs. Certification costs will be incurred on average one year before
the beginning of production, so the calculated cost is increased by seven percent. Distributing those
costs across the different engine categories, amortizing the costs over five years, and dividing by the
number of projected 2004 model year sales for each category results in per-engine costs between $2
and $9 for each category of heavy-duty diesel vehicles.
                                                                                       80

-------
                                   Chapter 4: Economic Impact of HDDE Standards
             1.  Exhaust Gas Recirculation

       Exhaust gas recirculation (EGR) is projected to be the most important area of technology
development that will enable manufacturers to achieve the targeted NOx emission levels.  Unlike
most other technological developments, which are largely further optimization of the baseline
technologies discussed in Chapter 3.n. A, introduction of EGR would be a step change in the design
of heavy-duty diesel engines. The technological challenges facing the manufacturers in developing
EGR for diesel engines are described in Chapter 3.II.B. While some research remains to optimize
EGR  systems for maximum  NOx-control effectiveness with minimum  negative impacts on
performance and durability, current developments show great promise  for substantial emission-
control improvements with EGR systems.

       According to the Acurex cost report, the typical cost to manufacturers of adding the hardware
for a high-flow cooled EGR system is estimated to range from $140 to $220 (in 1995 dollars) per
engine depending on the  service class. Factoring in the fixed costs, the appropriate markups, and
inflation results in an increased purchase price of $257, $345, and $439 for light, medium, and heavy
heavy-duty diesel vehicles, respectively.

             2.  Combustion Optimization

       Manufacturers can make a variety of changes to the basic engine  design that do not require
additional components.  Programming the engine's  electronic controls, optimizing intake air
characteristics and distribution, and making changes to piston bowl shape, the compression ratio, and
the injection timing strategy add little or no variable cost, but require significant expenses for R&D
and retooling. According to the Acurex cost report, total costs for these improvements would be
$5 million per engine line. For the different classes of vehicles, this translates to an incremental cost
between $22 and $55 per engine.

             3.  Improved Fuel Injection

       Manufacturers  are expected to  improve their fuel injection systems by increasing fuel
injection pressure, improving spray patterns, and adding rate shaping and split injection capability;
however, much of this improvement is expected to  occur independently  of the new emission
standards.5'6'7'8'9'10  For cam-driven electronic unit injection systems, the expected fuel  system
improvements will require stronger and better performing fuel injectors and solenoids.  Advanced
systems such as FtEUI or NGEUI technology require various reinforcements and better high-pressure
oil pumps and solenoid valves.  Common rail injection systems are similar enough to HEUI designs
that the cost estimate would mirror that for FtEUI systems.

       Incremental costs for this set of fuel injector improvements are roughly proportional to the
number of cylinders in an engine. EPA calculated  typical costs for these improvements using the

                                                                                    81

-------
Regulatory Impact Analysis
information contained in the Acurex report.  Light heavy-duty vehicles, typically equipped with
eight-cylinder engines, have an estimated total cost of $144 per engine, which is an average for the
different hardware configurations.  Medium and heavy heavy-duty vehicles, with six-cylinder
engines, would have a cost between $147 and $160 per engine. These cost estimates are based on
the cost estimates in the Acurex cost report, assuming that half of light and medium heavy-duty
engines, and that two-thirds  of heavy heavy-duty engines will have cam-driven unit injectors (and
that the remainder will have common rail, HEUI, or similar systems), and by adjusting to  1999
dollars. For this analysis EPA is assuming that 50 percent of the costs for these improvements are
attributable to emission control.  This is because EPA believes that manufacturers would make these
improvements for many of their engines, even in the absence of these emission standards, to reduce
fuel consumption and improve engine performance.11

             4. Variable  Geometry Turbochargers

       For several years research has focused on improving turbocharger designs to reduce response
time and increase compressor efficiency. One such design, the variable-geometry turbocharger, is
more complex than existing  turbochargers, but offers two primary operating enhancements: boost
pressure is maintained over a wider range of engine operation and response time is reduced. These
improvements contribute to lower exhaust emissions and provide control of airflow needed for
engines with EGR. Variable-geometry turbochargers require more parts and more assembly time,
resulting in a variable cost to manufacturers as high as $200 to $300 per engine according to the
Acurex cost report. However, EPA has become aware of new simpler designs for VGT systems that
are expected to be less expensive than the systems considered by Acurex.  Thus EPA has revised the
Acurex estimate by reducing assembly costs by 70 percent, and eliminating the actuator costs.12 The
revised estimates of the variable cost increase to manufacturers for VGT (relative to current
technology turbochargers)  range from $90 to $150.  Fixed costs for R&D and retooling  were
estimated at about $3.5 million per engine line. Combining the costs with the appropriate markups,
and adjusting for inflation results in costs of $203, $283, and $373 for light, medium and heavy
heavy-duty engines, respectively. For this analysis, however, 50 percent of these costs are assumed
to be attributable to emission control. As with the expected fuel injection improvements,  EPA
believes that manufacturers would make these improvements for many of their engines, even in the
absence of these emission standards, to reduce fuel consumption and improve engine performance.3
An EPA technical memo to the docket for this rulemaking contains additional discussion of the
Agency's 50 percent cost estimate for both improved electronic fuel injection, including a cost
sensitivity analysis detailing what impact this estimate has on the  standards cost-effectiveness.11
This cost sensitivity is also summarized in Table 8-13 of Chapter 8, Section IV of this RIA.

             5. Onboard Diagnostics

       Manufacturers are  not expected to make significant hardware modifications in response to
the OBD requirements for vehicles at or under 14,000 pounds. This is because, even without the
OBD regulations, manufacturers would monitor emission control components to ensure proper
engine performance.  In fact, manufacturers  already use onboard monitors for fuel injectors for
current engines. However, manufacturers are expected to incur additional costs for emission testing

                                                                                    82

-------
                                   Chapter 4:  Economic Impact of HDDE Standards
of representative engine configurations in various malfunction modes. Based on EPA's engineering
judgement, we estimate this testing will add $1 to the fixed costs for light heavy-duty engines, but
will not affect variable or operating costs. Even though only those light heavy-duty engines for
vehicles at or under 14,000 pounds would be subject to the OBD requirements, EPA is applying this
cost to all light heavy-duty engines for this analysis.  (Note: the Draft RIA analysis estimated this
cost to be $3 using an incorrect amortization calculation.)

              6. Engine Map Testing

       While manufacturers are not expected to make  significant hardware modifications in
response to the new supplemental standards, they are expected to conduct extensive steady-state and
transient cycle emission testing (i.e., testing at speeds and loads represented by the new supplemental
test cycles) as part of their R&D efforts. This will add $2, $5, and $5 to the fixed costs for  light,
medium, and heavy heavy-duty engines, respectively, but will not affect variable or operating costs.
(Note: the  Draft RIA analysis estimated this cost  to be $9, $23, and $23 using an incorrect
amortization calculation.)

              7.  Total Technology Package Costs

       The estimated incremental costs of these primary technologies depend on several judgements
about which technologies  will be used.  For example,  predicting  precisely  how  much  these
technologies will impact engine-out PM emissions is  difficult. If engine-out PM emissions are
increased, then manufacturers may need to increase the  use of aftertreatment.  This increases
hardware costs and there would be a greater potential for increased operating expenses.

       EPA believes it is not appropriate to assign  the full cost of fuel system upgrades or the
addition of VGT to the new emission standards.  As is discussed in Sections A.3 and A.4 of this
chapter, much of the anticipated improvements will come independently of the new standards and
any remaining system improvements for 2004 and later model year vehicles will provide benefits
beyond lower NOx emissions.  The resulting calculation  of total incremental cost for the set of
primary technologies, summarized in Table 4-2, shows the expected increase in purchase price due
to the new emission standards. Projected cost increases are $485, $657 and $803 for light, medium,
and heavy and heavy-duty vehicles, respectively for the 2004 model year.

       B.  Operating Costs

       EPA has assessed the potential for increased  operating costs, as described below, first for
EGR-related maintenance,  then for fuel economy.  EGR has the potential,  if not developed and
implemented properly, to increase operating costs, either by increasing fuel consumption or requiring
additional maintenance to avoid accelerated engine  or component  wear.  While it is possible to
develop scenarios and estimate the impact on operating costs of current diesel EGR concepts, this

                                                                                     83

-------
Regulatory Impact Analysis
is of minimal value due to the expected continuing development of these technologies. One major
focus of the R&D conducted over the next several years will be to resolve potential operating cost
impacts related to the use of EGR; thus the current state of the technology is not representative of
what is expected for 2004.

       While engine-out particulate emissions are dramatically lower than only a few years ago,
recirculating even a small amount of particulate matter through an engine introduces a concern for
engine durability. To prevent wear, manufacturers might specify more frequent oil change intervals
or a greater oil sump volume to accommodate any effects of acidity or particulate agglomeration in
the oil. However, EPA expects manufacturers to make a great effort to minimize any potential new
maintenance burden for the end user. Alternatively, changing fuel or oil formulations may be the
most  cost-effective way to reduce the  potential for particulate-related wear.  EPA  originally
proposed in the  1996  NPRM that manufacturers would be able to keep engine costs lowest by
investing $ 10 million to $ 15 million industry-wide in research to address these concerns (about $25
per engine when amortized over the fleet).  This estimate was based on EPA's engineering
judgement and prior experience.  Although manufacturers have had two opportunities to provide
information in their public comments, they have not provided any information that allow the Agency
to improve this estimate. Therefore, we continue to believe that it is the best available estimate of
the cost of addressing this issue.  To include the affect of improved materials resulting from the
R&D effort, the analysis incorporates a 2 percent increase in the cost of engine oil. The increased
expense of oil changes over the lifetime of vehicles ranges from $8 to $33 per engine (net present
value at the point of sale).

       In addition, EPA has included a cost for preventive maintenance, at the time of rebuild, to
ensure that EGR systems will not malfunction. EPA data show that nearly all engines from heavy
heavy-duty vehicles and 65 percent of those from  medium heavy-duty vehicles are rebuilt.13
Rebuilding engines from light heavy-duty vehicles is rare. EPA estimates that engine rebuild occurs
at 240,000 miles for medium heavy-duty vehicles, at 500,000 miles for heavy heavy-duty vehicles,
and at 300,000 miles for urban buses.  These mileage figures represent an approximate average
across the various applications within each service class, which experience widely differing mileage
accumulation rates. For example, garbage trucks have much different operating characteristics than
line-haul trucks.  According to the MOBILE model, these mileage figures translate into a rebuild in
the eleventh year for both truck categories.  As mentioned in the Acurex report,  rebuild procedures
for EGR systems will likely include solvent cleaning of the  EGR tubing and replacement of the
electronic control valve. Removal, cleaning, and replacement of the tubing are estimated to take 30
minutes at a $71 per hour labor rate.  Replacing the control valve on an aftermarket basis is expected
to cost three times the manufacturers' long-term direct cost, or $73 and $105 for medium and heavy
heavy-duty vehicles, respectively. Calculated in terms of net present value at the point of sale, the
net effect of EGR servicing comes to about $55 for medium heavy-duty vehicles and $71 for heavy
heavy-duty vehicles and urban buses.

       While EPA believes that sufficient R&D and the use of cooled EGR will address other
operating cost concerns, the Acurex report includes a cost estimate for increasing oil sump volumes
by 10 percent to address maintenance concerns with EGR. Oil sump volumes currently range from

                                                                                     84

-------
                                   Chapter 4:  Economic Impact of HDDE Standards
4 gallons for light heavy-duty diesel vehicles to 11 gallons for heavy heavy-duty vehicles, so the cost
impact varies greatly by vehicle category.  By calculating a cost at each oil change  as vehicles
accumulate mileage and discounting the life-cycle costs to the point of sale, Acurex estimated that
the cost of increasing oil sump volumes by 10 percent would cost $25, $55, and $145 for light,
medium, and heavy heavy-duty vehicles, respectively (in 1995 dollars). These cost are presented
here only for informational purposes, and are not included in the cost-effectiveness analysis.

       As is discussed in Chapter 3, while EGR has the potential to incur a fuel economy penalty,
it will probably be more than offset by improvements in fuel injection and the use of VGT. In fact,
EPA believes that the combined effect of these three technologies may decrease fuel consumption
by as much as  1.5 percent. EPA estimated the cost savings for each one percent decrease in fuel
consumption, based on a diesel fuel cost of $ 1.00 per gallon. Calculated as a net present value at the
point of sale, these estimates are $102, $178, and $891 for light, medium, and heavy heavy-duty
vehicles, respectively.  This sensitivity with respect to changes in fuel consumption varies so much
by vehicle category because of the widely differing mileage accumulation rates for different vehicle
categories. As discussed in section IV of Chapter 8, we have also performed a sensitivity analysis
to estimate the impact on costs if this rulemaking resulted in an increase in fuel consumption for HD
diesel engines. Table 8-11 estimates the impact on the per-vehicle cost increase for a one percent
increase in fuel consumption, and Table 8-12 estimates the increase in per-vehicle cost-effectiveness
for a one percent increase in fuel consumption.

       C. Secondary Technologies

       In the 1997 FRM, EPA analyzed the potential costs of secondary technologies (i.e.,  those
technologies that may potentially be available, but that EPA was projecting would not be used by
manufacturers to comply with the 2004 standards). EPA is not revising this analysis of secondary
technologies for this technology review rulemaking. The reader is referred to the RIA for the 1997
FRM for more information regarding this analysis.    However, new cost information has been
recently presented to the Agency which will be presented here.

       As discussed in Chapter 3, the Manufacturers of Emission Control Associations (MEC A) has
recently undertaken a test program at Southwest Research Institute to evaluate the emission benefit
potential of several aftertreatment devices.  Specifically, MECA members provided to  SwRI a
number of diesel oxidation catalysts  (DOC), particulate traps, and urea-based selective catalytic
reduction systems (SCR). As discussed in Chapter 4, DOC's have been used in the past for some
light- and medium-heavy duty engine families in order to comply with the 0.1 g/bhp-hr PM standard
which began in 1994, and for urban buses to comply with the 0.05 g/bhp-hr PM standard. It is likely
some number of engine families would continue to rely on DOC's for modest PM reductions.  As
discussed in Chapter 3, technical issues remain to be solved before PM traps or SCR systems  could
be considered feasible for wide spread us in the U. S. HD diesel market, and we believe it is unlikely
                                                                                     85

-------
Regulatory Impact Analysis
manufacturers will use either  of these technologies in 2004.  However, it appears that these
technologies could be available for the 2007 model year.

III.  Summary of Costs

       The per-vehicle cost figures presented above are used in Chapter 9 to calculate the cost-
effectiveness of the program by comparing to emission reductions over the lifetime of each vehicle
category for those engines covered by the new standards.  Included in that calculation are the
following modifications for later model year production.

       First, manufacturers recover their initial fixed costs for tooling, R&D, and certification over
a five-year period. Fixed costs are therefore applied only to the first five model years of production.

       The second modification is related to the effects of the manufacturing learning curve. This
is a well documented and accepted phenomenon dating back to the 1930s.  The general concept is
that unit costs decrease as cumulative production increases. Learning curves are often characterized
in terms of a progress ratio, where each doubling in cumulative production leads to a reduction in
unit cost to a percentage "p" of its former value (referred to as a "p cycle").  The organizational
learning which brings about a reduction in total cost is caused by improvements in several areas.
Areas involving direct labor and material are usually the source of the greatest savings.  These
include, but are not  limited to, a reduction  in the number or complexity of component parts,
improved component production, improved assembly speed and processes, reduced error rates, and
improved manufacturing process. These all result in higher overall production, less scrappage of
materials and products, and better overall quality.

       Companies and industry sectors learn differently. In a 1984 publication, Button and Thomas
reviewed the progress ratios for 108 manufactured items from 22 separate field studies representing
a variety of products and services.14'15 The average progress ratio for the whole data was slightly
higher than 80 percent, which supports the commonly used p value of 80 percent, i.e., each doubling
of cumulative production reduces the former cost level by 20 percent. As each successive p cycle
takes longer to complete, production proficiency generally reaches a relatively stable plateau, beyond
which increased production does not necessarily lead to markedly decreased costs.  In their article,
Button and Thomas emphasize the importance of understanding the dynamics underlying learning
processes.

       EPA applied a p value of 20 percent beginning in 2004 in this analysis.  That is, the variable
costs were reduced by 20  percent for each doubling of cumulative production. However, to avoid
overly optimistic projections, EPA included several additional constraints. Using one year as the
base unit of production, the first doubling would occur at the start of the 2006 model year and the
second doubling at the start of the 2008  model year. To be conservative, EPA incorporated the
second doubling at the start of the 2009 model  year.  Recognizing that the learning curve effect may
not continue indefinitely with ongoing production, EPA used only two p cycles.
                                                                                      86

-------
                                    Chapter 4:  Economic Impact of HDDE Standards
       EPA believes the use of the learning curve is appropriate to consider in assessing the cost
impact of heavy-duty engine emission controls. The learning curve applies to new technology, new
manufacturing operations, new parts, and new assembly operations.  Heavy-duty diesel engines
currently do not use EGR of any type today (hot, cooled, or cooled and filtered). This is therefore
a new technology for heavy-duty diesel engines and will involve new manufacturing operations, new
parts, and new assembly operations. Since this will be a new and unique product, EPA believes this
is an optimal situation for the learning curve concept to apply. Opportunities to reduce unit labor
and material costs and increase productivity (as  discussed above) will be great.  EPA believes a
similar opportunity exists for fuel systems on heavy-duty diesel engines.  While all diesel engines
have high-pressure fuel inj ection systems, the changes envisioned for common rail and unit inj ection
systems require fundamental redesign of system hardware. These new parts and new assemblies will
involve new manufacturing operations. As manufacturers gain experience with these  new systems,
comparable learning is expected to occur with respect to unit labor and material costs.  These
changes require  manufacturers to  start new production procedures, which, over time, will be
improved with experience.

       Table 4-4 lists the projected schedule of costs over time for each category of heavy-duty
diesel vehicles. The estimated long-term cost savings would reduce the impact on the total cost of
heavy-duty vehicles by about half.

       Characterizing these estimated costs in the context of their fraction of the total purchase price
and life-cycle operating costs is helpful in gauging the economic impact of the standards. Table 4-5
presents the baseline costs for each vehicle category, as developed by ICF.

IV. Aggregate Costs to Society

       The above analysis develops per-vehicle cost estimates for each vehicle class.  With current
data for the size and characteristics of the heavy-duty vehicle fleet and projections for the future,
these costs can be translated into a total cost to the nation for the emission standards in any year.
The sales for the different categories of heavy-duty diesel engines that would be covered by the rule
based on the 1995  model year were  determined using production information  provided by
manufacturers to EPA and were assumed to grow at a linear rate of two percent from the 1995 levels.
The result of this analysis is a projected total cost starting at $479 million in 2004. Per-vehicle cost
savings over time reduce projected costs to a minimum value of $248 million in 2009, after which
the growth in truck population leads to an increase to $325 million in 2020. Total costs  for these
years are presented by vehicle class in Table 4-6.

       The incremental cost associated with oil changes is incorporated on an annual basis for each
vehicle category. Incremental costs related to rebuild are not include in 2004 or 2009,  since the first
rebuilds would be expected after 2009. In 2020, incremental rebuild costs are applied to the vehicles
                                                                                      87

-------
Regulatory Impact Analysis
that would be rebuilt in that year. Maintenance costs are projected to be over $44 million per year
by 2020.
                                       Table 4-4
                    Projected Long-Term Diesel Engine/Vehicle Costs
                     (net present value at point of sale in 1999 dollars)
Vehicle Class
Light heavy-duty
Medium heavy-duty
Heavy heavy-duty
Model
Year
2004
2006
2009
2004
2006
2009
2004
2006
2009
Change
—
20 percent learning curve applied to
variable costs
Fixed costs expire; 20 percent learning
curve applied to variable costs
—
20 percent learning curve applied to
variable costs
Fixed costs expire; 20 percent learning
curve applied to variable costs
—
20 percent learning curve applied to
variable costs
Fixed costs expire; 20 percent learning
curve applied to variable costs
Purchase
Price
485
410
241
657
571
275
803
688
368
Life-cycle
Operating
Cost (NPV)
8
8
8
49
49
49
104
104
104
                                       Table 4-5
             Baseline Costs for Heavy-Duty Engines and Vehicles (1995 dollars)
Vehicle Class
Light heavy-duty
Medium heavy-duty
Heavy heavy-duty
Urban Bus
Engine Cost
$7,800
$12,400
$21,700
$22,000
Vehicle Cost
$22,504
$46,132
$96,490
$224,000
Operating Costs
$12,450
$31,242
$108,027
$437,153
                                                                                    88

-------
                Chapter 4: Economic Impact of HDDE Standards
                   Table 4-6
Estimated Annual Costs for Improved Heavy-Duty Vehicles
Year
2004
2009
2020
Category
Light heavy-duty
Medium heavy-duty
Heavy heavy-duty
Total Annual Cost
Light heavy-duty
Medium heavy-duty
Heavy heavy-duty
Total Annual Cost
Light heavy-duty
Medium heavy-duty
Heavy heavy-duty
Total Annual Cost
Cost Elements (millions of 1999 dollars)
Fixed
36
38
59
133
0
0
0
0
0
0
0
0
Variable
125
71
149
344
86
49
104
239
101
58
121
280
Operation
0.3
0.3
1.2
1.7
1.6
1.3
6.0
8.9
3.4
9.3
32
45
Total
161
109
210
479
88
50
110
248
105
67
153
325
                                                          89

-------
Regulatory Impact Analysis
Chapter 4 References

1.  Estimated Economic Impact of New Emission Standards for Heavy-duty On-Higway
Engines", Accurex Environmental Corporation, March 31, 1997.

2. "Update of EPA's Motor Vehicle Emission Control Equipment Retail Price Equivalent (RPE)
Calculation Formula," Jack Faucett Associates, Report No. JACKFAU-85-322-3, September
1985.

3.  "Detroit Diesel Unveils Year 2000 Series 50 Bus and Coach Engine", DDC Press Release,
March 17, 2000. Available in EPA Air Docket A-98-32, Docket Item #IV-G-07.

4.  Draft Regulatory Impact Analysis and Oxides of Nitrogen Pollutant Specific Study, p. 3-29
ff, October 1984.

5.  SAE paper 973182, "Advanced Technology Fuel System for Heavy-duty Diesel Engines"

6.  Diesel Progress, August 1998, "CAT Gears Up Next Generation Fuel Systems", available in
EPA Air Docket A-98-32, Docket Item #II-D-03.

7.  Diesel Progress, August 1998, "Next Generation MEUI-B to Debut in 2001", available in
EPA Air Docket A-98-32, Docket Item #II-D-03.

8.  Diesel Progress, October 1998, "No Mistaking New Cummins ISL Engine", available in EPA
Air Docket A-98-32, Docket Item #II-D-04

9.  "Cummins New Midrange Fuel System", presented by John Youngblood,  Cummins Engine
Company, at the SAE Diesel Technology TOPTEC, April 22, 1998, available in EPA Air Docket
A-98-32, Docket Item #U-D-01.

10. "GM Full-Size Trucks Make 'Power Play' with Gasoline and Duramax Diesel Teamed with
New Automatics and Six-Speed Manual", R. Wilson, Diesel Progress.  January 2000. Available
in EPA Air Docket A-98-32, Docket Item #IV-G-06.

11. "Costs and Benefits of VGT and Improved Fuel Injection with Sensitivity Analysis for On-
highway Heavy-duty Diesel 2004 Emission Standards", C. Moulis, Docket Item #A-98-32-IV-B-
O

12. "Summary of Phone Conversation Between Matthew Spears, US EPA and Dr. S.M. Shahed,
Allied Signal Corporation, on January 5, 1999 Regarding Advances in Variable Nozzle
Turbocharger (VNT) Design and Manufacturing", Docket #A-98-32-IV-B-4.

13."Heavy Duty Engine Rebuilding Practices," Draft EPA Report by Karl Simon and Tom
Strieker, March 21, 1995.
                                                                                 90

-------
                          Chapter 4: Economic Impact of HDDE Standards
;f3^&Q8fo(a!^%i]itMg^
                          Air Docket A-98-32.
                                                                  91

-------
Regulatory Impact Analysis
    CHAPTER  5:  ECONOMIC IMPACTS OF HD OTTO-

                          CYCLE STANDARDS

       This chapter contains an analysis of the economic impacts of the new emission standards for
heavy-duty Otto-cycle vehicles and engines.  First, a brief outline of the methodology used to
estimate the economic impacts is presented, followed by a summary of the technology packages that
are expected to be used to meet the standards.   Next, the projected costs of the individual
technologies is presented, along with a discussion of fixed costs such as research and development
(R&D), tooling and certification costs. Following the discussion of the individual costs components
is a summary of the projected per-vehicle cost of the new regulations. Finally, an analysis of the
aggregate cost to society of the regulations is presented. The costs  presented here are in 1999
dollars.

I.  Methodology for Estimating Costs

       Using  the  information  on emission  reduction technology presented in Chapter  4,  we
identified packages of technologies that would be likely to be used by the manufacturers to comply
with the emission standards. These technology packages are those which are being implemented to
meet California's low emission vehicle (LEV) standards. To identify the required technologies and
to quantify the costs of most of these technologies, we relied on the contracted study of heavy-duty
gasoline vehicle technology conducted by Arcadis Geraghty & Miller (hereafter referred to as the
Arcadis report).1   Information in the Arcadis  report regarding technology costs and current
nationwide vehicle and California LEV technologies (including such things as catalyst sizes and
loadings, as well as estimates of the percentages of vehicles that will require  certain technologies)
were obtained though a series of confidential discussions with vehicle and engine manufacturers,
catalyst and equipment suppliers, and other relevant entities.  Costs for  onboard refueling vapor
recovery (ORVR) equipment were taken from the final Regulatory Impact Analysis for ORVR
controls and updated for purposes of this analysis, and are not from the Arcadis report.2

       The costs of meeting the emission standards include both variable costs (incremental
hardware costs, assembly costs, and associated markups) and fixed costs  (tooling, R&D, and
certification costs).  Variable costs  are  marked up at a rate of  29  percent to account  for
manufacturers' overhead and profit.3 For a technology which is sold by a  supplier to the vehicle or
engine manufacturer an additional 29 percent markup is included to cover the suppliers' overhead
and profit. The exception to this is for precious metals. Vehicle manufacturers typically provide
catalyst suppliers with precious metals for use in the catalysts the suppliers manufacture.  Thus, the
additional 29 percent supplier markup is not applied to the cost of precious  metals. Fixed costs were
increased by seven percent for every year before the  start of production to reflect the time value of
money, and are then recovered with a five year amortization at the same rate.

II.  Technology Packages for Compliance with the Regulations
                                                                                  92

-------
	Chapter 5:  Economic Impacts of HP Otto-cycle Standards

       The various technologies that could be used to comply with the regulations were discussed
in the previous chapter on technological feasibility. We expect that the technology mixes being used
to meet the California LEV standards fairly accurately represent those that will be used to comply
with the federal standards beginning with the 2005 model year.  Thus, in developing costs for the
associated technologies we looked at the current technology used on HD Vs and compared that to the
technologies being used to meet the LEV standards in California. Table 5-1 shows both the current
baseline and expected technologies for complete vehicles. Table 5-2 shows the current baseline and
expected technologies for the engine-based standards.  The information contained in these tables
comes form the Arcadis report and is derived from conversations with the vehicle manufacturers
concerning their estimated configurations to be used to meet the new standards.  These tables only
show the technologies which are expected to change in some way from their current design or be
applied to different percentages of the fleet than they are currently. Technologies which are currently
being widely used on these vehicles such as  sequential multi-port fuel injection and EGR,  while
important to meeting the standards, are not expected to be fundamentally changed in their design,
or be utilized in  different percentages of the fleet than they currently are.   Thus, while  such
technologies will continue to be used on these vehicles, they are not included in these tables.
However, in some cases the cost of optimizing such technologies is included in the cost estimates
and are discussed in the following section.
                                                                                     93

-------
Regulatory Impact Analysis
                                      Table 5-1
         Current and Expected Technology Packages for Complete Vehicle Standards
Technology
Catalysts1
Oxygen sensors
ECM
Adaptive learning
Individual cylinder A/F
control
Leak free exhaust
Insulated exhaust
Secondary air injection
ORVR
Baseline Federal
60% single underfloor
40% dual underfloor
70% dual heated
10% triple heated
20% four heated
50% 32 bit computers
50% 16 bit computers
0%
0%
90%
0%
20%
0%
Estimated 2005
13% single underfloor
50% dual underfloor
37% dual close-coupled and
dual underfloor
13% dual heated
87% four heated
100% 32 bit computers
80%
10%
100%
40%
30%
100%2
1.  In addition to the change in catalyst configurations shown, we
precious metal compositions and loadings will change.
2.  ORVR only applies to complete vehicles 10,000 Ibs GVWR
100% application to those vehicles in 2006.
expect that catalyst washcoat and

and under, and is phased in, with
                                                                                   94

-------
           	Chapter 5:  Economic Impacts of HP Otto-cycle Standards

                                       Table 5-2
           Current and Expected Technology Packages for Engine-based Standards
Technology
Catalysts1
Oxygen sensors2
ECM
Improved fuel control
Secondary air injection
Baseline Federal
60% single underfloor
40% dual underfloor
70% dual heated
10% triple heated
20% four heated
50% 32 bit computers
50% 16 bit computers
50%
20%
Estimated 2005
13% single underfloor
87% dual underfloor
13% triple heated
87% four heated
100% 32 bit computers
100%
50%
1.  In addition to the change in catalyst configurations shown, we expect that catalyst washcoat and
precious metal compositions and loadings will change.
2.  OBD only applies to HDGEs under 14,000 Ibs GVWR (approximately 60 percent of HDGEs).

III.  Technology Costs

       The following  sections outline in detail the costs of new technologies and the costs of
improvements to existing technologies we expect will be used to comply with the standards.

       A. Improved Catalysts

       Improvements in catalyst systems fall into two broad categories: changes in catalyst system
configuration and changes in the catalyst precious metal and washcoat compositions and loadings.
In addition estimating costs for these improvements, we estimated the costs of substrates and
packaging (cans) for the improved catalysts.

              1. Changes in Catalyst Configurations

       Currently, all non-California Otto-cycle HDVs either have single or dual underfloor catalyst
configurations. Under the single underfloor catalyst system the exhaust from both banks of engine
cylinders "Y" into a single catalyst. With the dual underfloor catalyst system each bank of engine
cylinders exhaust into their own catalyst. Currently 60 percent of vehicles utilize the single catalyst
approach with the remaining 40 percent utilizing the dual catalyst approach.  We expect that the
usage of the single floor catalyst system will drop to 13 percent as a result of the new standards, and
usage of the dual catalyst system will drop to 50 percent.  We expect that the remaining 37 percent
of vehicles will utilize dual underfloor catalysts in conjunction with dual close-coupled catalysts.
The costs of the single underfloor catalyst and the dual underfloor catalysts were analyzed for both
the baseline (i.e., current) scenario and for enhanced versions used in compliance with the standards.
                                                                                     95

-------
Regulatory Impact Analysis
The cost dual underfloor/dual close-coupled catalyst system was only analyzed in an enhanced
configuration for use in compliance with the standards since there are currently no close-coupled
systems in wide use outside of California.  Since the required catalyst size tend to be a function of
engine size, we analyzed catalysts for two engine sizes, standard and large. For purposes of
developing an average per-vehicle cost we weighted the costs of the two catalyst systems assuming
that 75 percent of HDVs would be representative of the standard engine size and that the remaining
25 percent would be representative  of the large engine size.

              2.  Changes in Precious Metals

       The catalyst enhancements referred to in the previous paragraph consist of changes in the
catalyst precious metal and washcoat compositions and loadings. Vehicle catalysts have typically
used some combination of platinum (Pt), palladium (Pd) and rhodium (Rh).  These precious metals
account for a significant portion of the catalyst cost. Historically, a Pt/Rh  combination has been
used, although Pd has been used in much greater quantities (up to 100 percent). Pd is more thermally
stable than Pt and Rh and is therefore  a good choice for applications which see a high degree of
thermal loading, such as close-coupled catalysts. Currently, federally-certified HDVs typically have
a precious metal mix of 6.7 grams (g) Pd for each g of Rh,  with no Pt. This is generally applied at
loading of 2 grams per liter (g/L) of total catalyst substrate volume.  However, Pd usage is going up.

       We used a 10/1 ratio of Pd to Rh as it baseline assumption. Currently, enhanced underfloor
catalysts being used in California are loaded at 3 to 6 g/L of substrate volume at a Pd/Rh ratio of 10
to 1.  Close-coupled catalysts are typically 100 percent Pd loaded at 5 to 8 g/L of substrate volume.
Current federally-certified HDVs tend to have rather large  catalysts with fairly low precious metal
loadings. Thus, we expect that no increase in catalyst volume will be required to comply with the
standards. Rather,  the improvements will  center on the precious metals and washcoats, as well as
the movement toward increased use  of close-coupled catalysts.  In cases where close-coupled
catalysts will be utilized, we are assuming that total catalyst volume will remain unchanged, and that
the size of the underfloor catalysts will be reduced from the baseline size by the volume of the close-
coupled catalysts. We are assuming that, on average, the new standards will be met using a Pd/Rh
combination in a 10 to 1 ratio and at a loading of 4 g/L for underfloor catalysts. For close-coupled
catalysts we assumed that 100 percent Pd will be used at a loading of 6.25 g/L.

       Precious metal prices have shown some volatility in recent years. In order to smooth out this
volatility, as  well as insure an  uninterrupted supply of precious metals, vehicle  manufacturers
typically buy  precious metals in bulk and supply them to the catalyst manufacturers.  It is for this
reason that the 29  percent supplier markup that we are applying to products supplied to the
manufacturers by component suppliers is not being applied to the cost of precious metals.  We chose
to use the same precious metal spot  prices (i.e., short-term,  or daily, prices) for the purposes of this
analysis that we recently used in support of the Tier 2 emission standards for light-duty vehicles,
light-duty trucks, and medium-duty passenger vehicles. These are $868 per troy ounce for Rh and
$390 per troy ounce for Pd. We believe that the  spot prices used in the Tier 2 rulemaking are
appropriate here for two reasons.  First, the manufacturers affected by today's regulations also
                                                                                      96

-------
	Chapter 5:  Economic Impacts of HP Otto-cycle Standards

produce vehicles which must comply with the Tier 2 regulations.  Second, today's regulations and
the Tier 2 regulations take effect in the same time frame.

             3.  Changes in Washcoat

       In addition to the changes to precious metals just discussed, we expect that the new standards
will also result in changes to the catalyst washcoat compositions and loadings.  Current washcoats
are typically a combination of a cerium oxide blend (ceria) and aluminum oxide (alumina). Current
ratios of these tow components  range from 75 percent ceria/25 percent alumina to 100 percent
alumina. We assumed a 70/30 ceria to alumina ratio to represent the current baseline configuration.
Of the two common washcoat components, ceria is more thermally stable and, thus, is expected in
higher concentrations in close-coupled catalysts.  The We assumed that a slightly higher ceria
concentration (75/25  ratio of ceria to alumina) will be used in compliance with the vehicle-based
standards and that an even higher ceria concentration (80/20 ratio of ceria to alumina) will be used
in compliance with the engine-based standards.

       Current washcoat loadings range from 160 to 220 g/L of catalyst substrate volume.  For
simplicity we assumed an average loading of 190 g/L for the baseline configuration, and that this
loading would remain unchanged for compliance with the vehicle-based standards. For compliance
with the engine-based standards, we assumed that the washcoat loading will increase to 220 g/L. In
addition, we expect that a new technique of layering the washcoat and precious metals will be
employed for compliance with the engine-based standards.   Currently, the precious metals and
washcoat are applied to the catalyst substrate in a single slurry. Under the layering approach there
is a separate slurry for each precious metal, with the second slurry being applied after the first dries.
This process allows for more reaction surface area, resulting in a more efficient catalyst.  Table 5-3
provides a summary of the precious metal and washcoat compositions and loadings for the current
baseline vehicle, as well as those expected to be used in compliance with the vehicle-based and
engine-based  standards.

                                       Table 5-3
   Current and Projected Catalyst Precious Metal and Washcoat Compositions and Loadings

Precious Metals
Washcoat
Baseline
Pd/Rh=10/l
Loading = 2.1 g/L
30% Alumina/
70% Ceria
Loading = 190 g/L
Vehicle-based
Pd/Rh=10/l
Loading = 4.0 g/L1
25% Alumina/
75% Ceria
Loading = 190 g/L
Engine-based
Pd/Rh= 10/1
Loading = 4.5 g/L
20% Alumina/
80% Ceria
Loading = 220 g/L
1.  For close-coupled catalysts we assumed 100 Percent Pd at a loading of 6.25 g/L.

             4.  Substrates
                                                                                     97

-------
Regulatory Impact Analysis
       The substrate that the precious metals and washcoat are affixed to are typically ceramic
substrates of 400 cells per inch.  Increasing efforts are going into developing metallic substrates,
which offer better temperature and vibration stability, as well as requiring less precious metal loading
to achieve the same emission benefits. However, information from catalyst suppliers suggested that
the increased costs of the metal substrates will tend to cancel out any savings in precious metal costs,
as discussed in the Arcadis report..  Thus, we assumed that the current ceramic substrate  would
continue to be used in compliance with the standards.  Based on cost data obtained from catalyst
substrate suppliers, the following linear relationship

                            C = $4.67V+$1.50
where:
       C = cost to the vehicle manufacturer from the substrate supplier
       V = substrate volume in liters

was developed in the Arcadis  report and is accurate for ceramic substrates sized from 0.5 L  to 4 L.
Generally, catalyst substrates for HDVs are manufactured in bricks no larger than 2.5 L,  with a
catalyst of greater than 2.5 L being comprised of more than one brick.

              5. Packaging

       The final cost component of the catalyst system is the can. The catalyst substrate is typically
packaged in a can made of 409 stainless steel and around 0.12 centimeters thick (18 gauge).  The
cost of the can is a very small portion of the overall catalyst cost.

       The following tables (Tables 5-4, 5-5 and 5-6) show our estimates of the total catalyst system
cost  for each of the three configurations  previously discussed, and for the current, baseline
formulation as well as the formulations projected to be used to comply with the vehicle-based and
engine-based requirements. No baseline costs are shown in Table 5-6 (dual underfloor plus dual
close-couple catalyst system) because these systems are not currently in wide use on federally-
certified vehicles. These tables show the estimated costs rounded to the nearest dollar.
                                                                                       98

-------
                        Chapter 5:  Economic Impacts of HP Otto-cycle Standards

                                     Table 5-4
               Estimated Catalyst Costs of Single Underfloor Catalyst System

Engine Size
Catalyst
Volume (L)
Substrate
Washcoat
Precious
Metals
Can
Total Material
Cost
Labor
Labor
Overhead
@40%
Supplier
Markup
@29%'
Manufacturer
Cost
Baseline
Standard
4.8
$26
$19
$141
$5
$191
$4
$1
$16
$212
Large
5.8
$31
$22
$170
$5
$228
$4
$1
$18
$251
Projected Vehicle-based
Standards
Standard
4.8
$26
$19
$268
$5
$318
$4
$1
$16
$339
Large
5.8
$31
$22
$324
$5
$382
$4
$2
$19
$407
Projected Engine-based
Standards
Standard
4.8
$26
$23
$302
$5
$356
$6
$2
$18
$382
Large
5.8
$31
$27
$365
$5
$428
$6
$2
$21
$457
1 The supplier markup is not applied to the cost of the precious metals because the precious metals
are supplied by the vehicle manufacturer.
                                                                                99

-------
Regulatory Impact Analysis
                                      Table 5-5
                Estimated Catalyst Costs of Dual Underfloor Catalyst System

Engine Size
Catalyst
Volume (L)
Substrate
Washcoat
Precious
Metals
Can
Total Material
Cost
Labor
Labor
Overhead
@40%
Supplier
Markup
@29%'
Manufacturer
Cost
Baseline
Standard
4.8
$26
$19
$141
$5
$191
$5
$2
$17
$215
Large
5.8
$31
$22
$170
$6
$229
$6
$2
$19
$256
Projected Vehicle-based
Standards
Standard
4.8
$26
$19
$268
$5
$318
$7
$3
$17
$345
Large
5.8
$31
$22
$324
$6
$383
$8
$3
$20
$414
Projected Engine-based
Standards
Standard
4.8
$26
$23
$302
$5
$356
$11
$4
$20
$391
Large
5.8
$31
$27
$365
$6
$429
$12
$5
$23
$469
1 The supplier markup is not applied to the cost of the precious metals because the precious metals
are supplied by the vehicle manufacturer.
                                                                                 100

-------
                         Chapter 5:  Economic Impacts of HP Otto-cycle Standards

                                      Table 5-6
     Estimated Catalyst Costs of Dual Underfloor Plus Dual Close-coupled Catalyst System

Engine Size
Catalyst Volume (L)
Substrate
Washcoat
Precious Metals
Can
Total Material Cost
Labor
Labor Overhead @40%
Supplier Markup
©29%'
Manufacturer Cost
Projected Vehicle-based Standards
Standard
4.8
$29
$20
$314
$6
$369
$15
$6
$22
$412
Large
5.8
$34
$24
$370
$7
$435
$16
$6
$25
$482
1 The supplier markup is not applied to the cost of the precious metals because the precious metals
are supplied by the vehicle manufacturer.

       B. Exhaust Gas Recirculation (EGR)

       Electronically controlled EGR is currently used on about 85 percent of non-California Otto-
cycle HDVs.  Those manufacturers that do not currently employ EGR on their federally certified
vehicles are not expected to utilize it to comply with the standards. Thus, the percentage of the fleet
with EGR is not expected to change as a result of the standards. However, some improvements to
flow control are expected to be made to those EGR systems that are currently being used, primarily
to comply with the new OBD requirements. In addition to minor changes in control algorithms, we
expect minor changes to the EGR valve and gasket, as well as the EGR flow sensor. These changes
are expected  to cost  from $5 to $12 per vehicle.  For simplicity we  assumed that the EGR
improvements will cost $7 per EGR-equipped vehicle for the purposes of this analysis.

       C. Secondary Air  Injection

       The hardware cost for vehicles which utilize secondary air injection to reduce HC and CO
is expected to be about $67 per vehicle. We expect that the usage rate of secondary air inj ection will
rise from its current use on about 20 percent of Otto-cycle HDVs to about 30 percent as a result of
the standards.
                                                                                  101

-------
Regulatory Impact Analysis
       D. On-board Diagnostics

       On-board diagnostics systems are currently required in California (OBD II).  Although not
required federally, many non-California HDVs do have some form of OBD system. This is primarily
for ease of manufacturing. On lines where vehicles are being assembled for both the California and
non-California markets it is often easier to simply install the California hardware on all vehicles
rather than install different parts on vehicles depending on their destination. This is especially true
of parts such as electronic control modules, some form of which needs to be installed regardless of
the vehicle's destination.  It is less true of hardware needed specifically for the California market that
is  not required outside of California, such as additional oxygen sensors for catalyst monitoring.
Thus, the changes required to implement OBD nationwide are not extensive.  The  main cost
components associated with adopting OBD nationwide are as follows:

       •       Oxygen sensors/catalyst efficiency monitoring
             Evaporative emissions purge and leak
       •      Electronic control module improvements
             Manifold  vapor pressure sensor improvements

Each of these OBD cost components is discussed in the following sections.  A National OBD
program is only being applied to HDGVs weighing 14,000 Ibs GVWR or less.  Thus, only 60 percent
of HDGVs certified according to the engine-based program would be required to comply with the
OBD requirements.  This is reflected in the cost summary table later in this chapter.

              1.  Oxygen Sensors/Catalyst Efficiency Monitoring

       The OBD requirements, as well as the expected changes in catalyst configuration, will result
in  changes in the number and placement of oxygen sensors in the exhaust system.  Oxygen sensors
in  non-California are typically only placed before the catalyst and used for closed loop air/fuel ratio
control. Compliance with the OBD requirements will require the use of oxygen sensors both before
and after the catalysts, to be used to monitor catalyst efficiency in addition to controlling air/fuel
ratio.

       Heated oxygen sensors are used for both California and non-California vehicles. We also
expect them to be used in compliance with the standards. Heated oxygen sensors have an average
manufacturer's cost of $10 to $15.  Thus, for simplicity we used a manufacturer's cost of $13 for
each sensor for this analysis.

       Oxygen sensors are currently required downstream of the catalyst only on California vehicles.
However, many non-California vehicles are equipped with downstream sensors as a way of reducing
part complexity when they are manufactured on the same production line as vehicles destined for
California.  Of non-California vehicles, one-sixth of single underfloor catalyst vehicles and half of
dual underfloor catalyst vehicles have downstream oxygen sensors. However, the OBD requirements
(as well as the expected changes in  catalyst configurations) will result in 80 percent of HDGEs
subject to the OBD requirements needing an average of two additional oxygen sensors.
                                                                                    102

-------
	Chapter 5: Economic Impacts of HP Otto-cycle Standards

             2. Evaporative Emissions Purge and Leak

       The OBD provisions include a requirement for evaporative emissions control system purge
and leak detection. The most common method of performing these functions is to close off the vent
solenoid, use manifold vacuum to purge vapors from the evaporative control system, close the vapor
management valve and monitor the system vacuum using a fuel tank pressure transducer.  Any
change in the vacuum over time indicates a leak with the rate of vacuum loss related to the size of
the leak.

       The additional costs associated with this system include those for the canister vent solenoid,
the fuel tank pressure transducer, tubing and wiring, and programming of the electronic control
module. The manufacturer costs are $11 for the canister vent solenoid and $16 for the fuel tank
pressure transducer.  Wiring and labor bring the average unit cost of this system to around $31.

             3. Electronic Control Module Improvements

       Although almost all vehicles use 16 bit electronic control modules (ECMs), there is a gradual
change toward 32 bit processors on  California vehicles.  We expect that many non-California
vehicles will have 32 bit processors as well in order to reduce parts complexity.  Thus, we assumed
that, as a baseline, 50 percent of non-California vehicles will be equipped with 32 bit processors
prior to 2005. We expect that all vehicles will be equipped with 32 bit processors in order to comply
with the standards. We expect that this move to 32 bit processors will result in a $21 increase over
the baseline vehicle. However, the need for 32 bit processors is only partly driven by the OBD
requirements. The lower emission limits will also result in a move to more powerful ECMs.  Thus,
we are assigning half of the incremental cost of the improved ECM to the OBD requirements and
the other half to the exhaust emission  standard requirements for all covered vehicles. For engines
we are assigning half of the ECM cost to OBD for engines under 14,000 Ib GVWR (about 60 percent
of Otto-cycle HDEs) and none of the ECM cost to OBD for those over 14,000 Ib GVWR.

             4. Manifold Vapor Pressure Sensor Improvements

       We expect that the  OBD requirements will result in improved exhaust gas recirculation
(EGR) flow control.  This will require improvements to the manifold vapor pressure sensor at a cost
of $6 per EGR-equipped vehicle.

       E.  Exhaust Systems

       We expect that insulated exhaust systems will be used in  close-coupled catalyst-equipped
vehicles in order to improve catalyst light-off time. We estimate that such systems will cost $42 per
vehicle. Since we project that 40 percent of chassis-based vehicles will use close-coupled catalysts,
the cost per vehicle on average will be $17.

       F.  Electronic Control Module
                                                                                  103

-------
Regulatory Impact Analysis
       The projected improvements to electronic control modules (ECMs) were discussed in the
earlier section on OBD systems. As was discussed there, half of the cost of the ECM improvements
will be a result of the OBD requirements and half will be a result of the lower exhaust emission
standards for vehicles and those engines subject to the OBD requirements. 1  Thus, we project that
ECM improvements due to the increased stringency of the exhaust emission standards will result in
a $11 per vehicle increase for vehicles, and engine subject to the OBD requirements.  For those
engines over 14,000 Ib GVWR (which are not subject to the OBD requirements) we are assigning
the entire $21 cost of the ECM improvements to the emission standards.

       G. Onboard Refueling Vapor Recovery

       We estimated the costs for onboard refueling vapor recovery (ORVR) equipment by updating
the estimates of ORVR technology cost that were developed in the Regulatory Impact Analysis for
the original ORVR regulations, as referenced at the beginning of this chapter.  While we did not
ultimately adopt ORVR requirements for any heavy-duty vehicles in the original rulemaking, we did
estimate the cost of such controls separately for both light heavy-duty vehicles (8,501 through 14,000
Ibs. GVWR) and heavy heavy-duty vehicles  (14,001 Ibs. GVWR and greater). For this analysis we
assumed that the technology required to meet the standards has not changed since the original ORVR
analysis was done, and that the nature of the light heavy-duty fleet (in terms of percentage of vehicles
with one versus two fuel tanks, etc.) also has not changed. Despite specific requests for comment
on these assumptions in the proposal, we received no comments disputing them.  Thus, we simply
took the cost estimates for light heavy-duty vehicles from the original analysis and adjusted them to
account for inflation. EPA believes this is reasonable because the vehicles we are applying ORVR
requirements to in this action (complete  vehicles from 8,501 through 10,000 Ibs. GVWR) are a
subset of the light heavy-duty vehicle class analyzed for the original rulemaking. The original per-
vehicle cost estimates (in 1992 dollars) were $6.29 forvariable cost and $2.60 for fixed cost. Using
the Consumer Price Index to account for  inflation, these costs were adjusted (to 1999 dollars) to
$7.50 for variable cost and $3.00 for fixed cost.

       In addition to variable and fixed costs, ORVR also has an associated operating cost benefit,
which takes into account both a the fuel economy penalty of the added weight of the hardware and
the much larger fuel economy benefit that comes from recovering refueling vapors and using them
in the engine. In the analysis for the original ORVR regulations this operating cost was estimated
to be a $5.50 per-vehicle lifetime credit for light heavy-duty vehicles. The credit was conservatively
calculated assuming that Stage n refueling vapor recovery controls would not be discontinued. Since
the value of this credit is entirely dependent on the price of gasoline, it was not updated to account
for inflation because the price of gasoline has not generally risen with inflation. Although gasoline
prices have risen significantly in recent months, it is too early to tell whether this is a long term trend.
Thus, a lifetime operating credit of $5.50 per vehicle is used in this analysis. Using the recent higher
gasoline prices would have resulted in a higher per vehicle credit.

IV. Fixed Costs
                                                                                    104

-------
	Chapter 5:  Economic Impacts of HP Otto-cycle Standards

       The fixed costs are broken into four main components: research and development, tooling,
certification, and in-use testing.  Of these four, only certification and in-use testing costs apply to
vehicle-based certifications. In-use testing costs do not apply to engine-based certifications. These
costs are discussed individually in the following sections.

       A. R&D and Tooling Costs

       The vehicle-based standards  will  essentially require the  application of California LEV
technology to HDVs nationally. Since this technology has already been developed and is being
implemented and the tooling is in place to make California vehicles on the same assembly lines as
non-California vehicles, there are no R&D or tooling costs associated with the vehicle-based
requirements. However, in the case of the engine-based standards, we  expect that some R&D and
new tooling will be required. We believe that, on average, R&D costs for a single engine family will
be about $3 million, and that tooling costs will be about $75 thousand per engine family. Assuming
that annual sales per engine family average 25 thousand units and that these costs are recovered over
a five year period, we estimate that the R&D and tooling costs will be  $34 per engine for the first
five years of the program.

       B. Certification Costs

       We relied on a previous analysis for estimating certification costs.4 Updating those costs for
inflation using the Consumer Price Index results in an estimated certification cost of $252,750 per
engine family. Certification costs will be incurred on average one year before the start of production.
Thus, this cost is increased by seven percent.  Summing the costs separately for engine families
certified to the chassis-based  and engine-based and amortizing them over five years results in
projected per-vehicle certification costs of $1 for chassis-based configurations and $6 for engine-
based configurations.

       C. In-use Testing Costs

       Using cost information developed in support of our CAP 2000  regulations, we project that
the in-use testing requirement will cost $1 per vehicle.  We used the cost from the CAP 2000
rulemaking of $4,600 per test.  We used the Consumer Price Index to inflate this cost from 1998 to
1999 dollars Assuming seven engine families and ten tests per engine family, total in-use testing
costs would be $334,000. Using assumed complete heavy-duty vehicle sales of just under 300,000
in 2005 yields a per vehicle in-use testing cost of just over $1.  This cost will be incurred indefinitely
for each year of production, rather than being recovered over five years as with the other fixed costs.
In other words, while other fixed costs such as tooling and R&D are incurred only once and then
recovered over a five year period, the in-use testing costs are incurred every year.

V. Summary of Costs

       Table 5-7 contains a summary of per-vehicle costs associated  with the new standards for
Otto-cycle heavy-duty vehicles and engines.  The various hardware cost components include the
                                                                                    105

-------
Regulatory Impact Analysis
manufacturers' 29 percent markup. These costs are presented as incremental cost increases from the
cost of current vehicle emission control systems.

                                       Table 5-7
       Summary of Incremental Costs to Meet the Otto-cycle Vehicle Emission Standards

Catalyst
On-board Diagnostics
ORVR
Other Emissions Hardware
Total Hardware
Fixed Costs
Operating Costs (ORVR)
Total Incremental Cost
Chassis-based Standards
$163
$68
$8
$41
$280
$5
-$6
$279
Engine-based Standards
$187
$41
—
$28
$256
$40
—
$296
VI. Aggregate Cost to Society

       In addition to the per vehicle costs just described, we also calculated the aggregate cost to
society. This was done by combining the per vehicle costs with assumed future sales of HDVs. The
sales for the different categories of heavy-duty Otto-cycle vehicles and engines that would be
covered by the rule based on the 1995 model year were determined using production information
provided by manufacturers to EPA and were assumed to grow at a linear rate of two percent from
the 1995 levels.  The results of this analysis are summarized in Table 5-8. The recovery of fixed
costs results in slightly reduced costs beginning in 2010, after which costs begin to rise in accordance
with projected increased sales. The aggregate costs represent a combined estimate of the fixed costs
as they are allocated over the first five years of sales (with the exception of fixed costs for in-use
testing, which continue indefinitely), variable costs assessed at the point of sale, and operating costs
(primarily in the form of fuel cost savings) for ORVR-equipped vehicles (calculated to net present
value and applied at the point of sale).
                                                                                     106

-------
            Chapter 5: Economic Impacts of HP Otto-cycle Standards
                        Table 5-8
Aggregate Cost to Society of the Heavy-duty Otto-cycle Requirements
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
Cost (Smillion)
$110
$117
$124
$126
$129
$124
$126
$128
$131
$133
$135
$137
$139
$141
$144
$146
$148
$150
$152
$154
                                                                107

-------
Regulatory Impact Analysis
Chapter 5 References

l."Cost Estimates for Heavy-duty Gasoline Vehicles," Arcadis Geraghty & Miller, Final Report,
September 30, 1998.

2. "Final Regulatory Impact Analysis: Refueling Emission Regulations for Light Duty Vehicles
and Trucks and Heavy Duty Vehicles," U.S. EPA, January, 1994.

3. "Update of EPA's Motor Vehicle Emission Control Retail Price Equivalent (RPE) Calculation
Formula," Jack Faucett Associates, Report No. JACKFAU-85-322-3, September 1985.

4. "Draft Regulatory Impact Analysis and Oxides of Nitrogen Pollutant Specific Study," p. 3-29
ff, October 1984.
                                                                                 108

-------
	Chapter 6:  Environmental Impact of HP Diesel Standards


        CHAPTER 6:  ENVIRONMENTAL  IMPACT OF

                        HD DIESEL STANDARDS

I.  Introduction

       This chapter describes the expected environmental impacts of the new heavy-duty diesel
engine NMHC plus NOx emissions standards described in the preamble. Specifically, this chapter
includes an estimated nationwide NOx, VOC, and PM10 inventory for 2000, heavy-duty diesel engine
NOx, NMHC, and PM inventory projections for future years (with and without additional control),
estimates of the impacts of the standards on typical vehicles over their lifetime, and a discussion of
the environmental effects of the emissions reductions.1

       While the new standards are combined NMHC plus NOx levels, we consider the NMHC and
NOx emissions impacts separately.  Given the technologies we expect manufacturers to use on
heavy-duty diesel engines to comply with the new standards, we model the fleet-average impact of
the combined standard  as being equivalent to a 2.3 g/bhp-hr NOx standard and a 0.2 g/bhp-hr
NMHC standard. We base these emission factors on the judgement that manufacturers will find it
easier to design for low NMHC to give them more flexibility for their NOx calibrations.  This is
consistent with statements made in informal discussions with engine manufacturers.

       We emphasize,  however, that this  is only  an analytical  approach;  we  expect that
manufacturers will optimize each family uniquely with respect to the combined standards, balancing
the sometimes competing effects on NMHC and NOx control technologies. Thus individual engine
families may have emission levels different from the fleet-average emissions we use in this analysis.
It is also important to note that we are modeling the environmental impacts of the supplemental
testing requirements beginning in calendar year 2004, because we believe that manufacturers will
design most, if not all, of their engine models to comply with these requirements in model year 2004.
This assumption, which is consistent with the assumptions made for the economic analysis in
Chapter 4, should not significantly affect the results of the cost-effectiveness analysis in Chapter 8.

       The inventory analysis described below builds on the inventory analysis in the Regulatory
Impact Analysis associated with the 1997 Final Rulemaking for heavy-duty diesel engines (HDDE).l
However,  we use recent studies to improve our understanding of the emissions impact of mobile
sources. We discuss these studies and their effects on the calculated HDDE emissions inventories
in this chapter.
(1) Three terms are used in this chapter to describe organic emissions: "total hydrocarbons" (THC or HC), volatile
organic compounds"(VOC), and "nonmethane hydrocarbons" (NMHC). THC refers to the organic emissions from
an engine as measured by the test procedures of 40 CFR 86. VOC refers to organic emissions excluding
compounds that have negligible photochemical reactivity, primarily methane and ethane (see 40 CFR 51.100).
NMHC refers to the difference obtained by subtracting methane from total hydrocarbons.  Since the ethane content
of emissions is very small from diesel engines, organic emissions measured as NMHC are approximately the same
as when measured as VOC.

                                                                                   109

-------
Regulatory Impact Analysis
II. Description of Calculation Method

       In modeling emissions from heavy-duty diesel engines, our intent is to be consistent with the
upcoming MOBILE6 model. MOBILE6 is the upcoming version of the MOBILE model that we
historically use  to  develop  calendar  year  specific  emission  factors for on-highway vehicles.
However, it does not have the capability to analyze all of the scenarios  needed to support the
rulemaking. Consequently, we developed a spreadsheet model which provides consistent results
with the MOBILE model, and has the needed capability.

       A. General Equation

       We divide HDDEs into four distinctive classes for the  purpose of inventory calculations.
Table 6-1 presents these classes which have different characteristics due to the difference in their
size and use. Later in this chapter, we discuss some of these differences as they apply to emission
modeling. Our standards apply throughout an engine's regulatory useful life. Therefore, emissions
may be cleaner earlier in an engine's life  and dirtier later in its life due to deterioration.  We use
regulatory useful life in our modeling as the point in the engine's life at which the engine just meets
the emissions standards with an assumed compliance margin.

                                        Table 6-1
	HDDE Classes and Regulatory Useful Life	
         Class                  Description*                Regulatory Useful Life**
      Light HDDE          8,501-19,500 Ibs. GVWR             110,000 mi 710 yrs

    Medium HDDE       19,501-33,000 Ibs. GVWR            185,000 mi 710 yrs

     Heavy HDDE           > 33,000 Ibs. GVWR         435,000 mi 710 yrs 7 22,000 hrs

       Urban Bus          characterized by application      435,000 mi 710 yrs 7 22,000 hrs
  * GVWR refers to gross vehicle weight rating; "urban bus" does not generally include school buses or inter-city buses.
  ** Whichever occurs first; for the purposes of these calculations, we use 290,000 miles for urban buses.
       For our calculations, emissions from HDDEs are primarily a function of per-engine emission
factors, in-use deterioration, and vehicle miles traveled.  Equation 1 presents the basic calculation
we use to determine emissions from HDDEs in short tons per year.  Following this section, we
supply more detail on the components of this equation.


TonsCY = (454 x 2000)-' x Eclass{ VMTx CFxEM¥/age [(ZML^ + DET^^x 7Fage]} (1)

where:
                                                                                      110

-------
                          Chapter 6: Environmental Impact of HP Diesel Standards

       TonsCY - emissions for a given calendar year expressed in short tons
       class - LHDDE, MHDDE, HHDDE, and urban bus
       VMT- total vehicle miles traveled in a given calendar year by class
       CF - conversion factor from g/bhp-hr to g/mi by class
       MY/age - distribution of vehicles in a calendar year by vehicle age
              - zero-mile emission level in g/bhp-hr for a given model year engine
              agg - emissions deterioration as a function of model year and vehicle age
       TFage - travel fraction of vehicles from each model year in a given calendar year
       (454 x 2000) •' - conversion from grams to short tons

       B.  Conversion Factors

       Our engine standards are in terms of grams of pollutant per work performed. We use these
units because we believe they best characterize emissions for an engine-based emission standard.
However, we use vehicle miles traveled (VMT) to characterize heavy-duty engine operation in our
emission inventory calculations.  We believe that we can more accurately determine VMT for
FIDDVs than we can determine the work performed by FIDDEs.

       To apply VMT to our emissions calculations, we need emission factors in terms of grams per
mile. Therefore, in our calculations, we convert the g/bhp-hr figures to g/mi. Because large engines
typically perform more work in a mile  of travel than  small engines, we use separate conversion
factors (CF) for each class of FIDDEs.  These numbers are reported in units of bhp-hr/mi and are
based on work performed for MOBILE62. The conversion factor is a function of fuel density, brake
specific fuel consumption and fuel economy.  Table 6-2 presents the CFs we use for 1996 and later
model year engines.  For older engines, the CFs may vary, some higher and others lower than the
values in Table 6-2.

                                       Table 6-2
                        Conversion Factors for HDDEs (bhp-hr/mi)
LFIDDE
1.23
MFIDDE
2.25
FfflDDE
2.97
Urban Bus
4.68
       C. Vehicle Miles Traveled

       To determine the tons of emissions in a given calendar year we need to know the total VMT
for that calendar year and the travel fraction of each model year of engines.  The travel fraction is
important because engines produced before and after a new standard goes into effect will have
different emission levels.

       Total Miles
                                                                                   111

-------
Regulatory Impact Analysis
       To determine the tons of emissions in a given calendar year we need to know the total VMT
for that calendar year and the travel fraction of each model year of engines. The travel fraction is
important because engines produced before and after a new standard goes into effect will have
different emission levels.

       To calculate nationwide emissions from HDDEs, we multiply the total VMT by the emission
factors for HDDEs. For this analysis, we base the nationwide annual VMT on Federal Highway
Administration estimates of annual VMT by highway category and vehicle type. We extrapolate the
VMT estimates out to 2030 using linear growth at approximately 3 percent for all classes of heavy
duty diesel vehicles.

       We also split this VMT by MOBILE class and fuel type. To split the VMT by class and fuel
type, we use information on engine registrations by class and per-vehicle operation in miles per year
collected for use in MOBILE6.3 We use the products of the vehicle registrations and per-vehicle
operation to determine the VMT fractions. Table 6-3 presents the resulting breakdown of VMT by
class.
                                        Table 6-3
             Total VMT by Class for Heavy Duty Diesel Vehicles [million miles]
Calendar Year
2000
2005
2010
2015
2020
2030
LHDDV
41,158
48,475
55,713
63,550
71,386
87,060
MHDDV
37,013
43,593
50,102
57,149
64,197
78,292
HHDDV
143,974
169,355
194,643
222,022
249,401
304,160
Urban Bus
2,753
3,242
3,726
4,250
4,774
5,823
       These HDDE VMT estimates are higher than those used in our proposal for 2004 heavy-duty
engine standards due to the use of new and updated MOBILE6 estimates of the fraction of total VMT
that is heavy-duty and the fraction of heavy-duty VMT that is diesel (see Chapter 6 of the draft RIA
for a discussion of the VMT estimates used in the proposal, available in EPA Air Docket A-98-32,
docket item in-B-01).  The predominant changes were to increase VMT estimates of light heavy-
duty vehicles and the diesel fraction of heavy-duty vehicles, both of which are consistent with recent
trends. The net result is that if the MOBILE values4 are used to calculate diesel fuel consumption,
they agree very well with Federal Highway Administration estimates.5'6  This gives  us added
confidence that these new estimates are accurate.
       Travel Fraction
                                                                                     112

-------
	Chapter 6: Environmental Impact of HP Diesel Standards

       Travel fraction refers to the percentage of total miles driven in a given calendar year coming
from each surviving model year of vehicles. In determining the travel fraction of vehicles by age,
we considered both the survival rates of HDDVs and the average annual mileage accumulation rates
by age. The survival rates give us the distribution of the number of vehicles of each model year in
a given calendar year.  HDDEs are operated less as they age; therefore, we consider the miles
traveled by age when determining our travel fraction. We use the age distributions and VMT by age
rates developed for MOBILE6.7 Table 6-4 presents survival distribution and mileage accumulation
rates by age.
                                                                                   113

-------
Regulatory Impact Analysis
                                       Table 6-4
                         Survival Distribution of HDDEs by Age
Vehicle
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Light
0.740
1.000
0.932
0.870
0.811
0.756
0.705
0.657
0.613
0.572
0.533
0.497
0.464
0.432
0.403
0.376
0.351
0.327
0.305
0.284
0.265
0.247
0.231
0.215
0.207
0.194
0.177
0.167
0.153
0.119
Survival Distributions
Medium Heavy
0.535
1.000
0.935
0.875
0.818
0.765
0.715
0.669
0.626
0.585
0.547
0.512
0.478
0.447
0.418
0.391
0.366
0.342
0.320
0.299
0.280
0.262
0.245
0.229
0.218
0.204
0.191
0.179
0.165
0.151
0.535
1.000
0.935
0.875
0.818
0.765
0.715
0.669
0.626
0.585
0.547
0.512
0.478
0.447
0.418
0.391
0.366
0.342
0.320
0.299
0.280
0.262
0.245
0.229
0.218
0.204
0.191
0.179
0.165
0.151
Bus
0.500
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.999
0.996
0.989
0.970
0.925
0.832
0.662
0.413
0.197
0.161
0.132
0.108
0.089
0.072
0.059
0.065
0.033
0.033
0.033
0.016
0.016
Mileage Accumulation Rates
Light Medium Heavy Bus
28,951
26,479
24,226
22,173
20,301
18,593
17,035
15,613
14,314
13,128
12,043
11,052
10,146
9,317
8,558
7,864
7,227
6,645
6,111
5,622
5,173
4,762
4,384
4,038
3,720
3,427
3,159
2,913
2,686
2,477
36,493
33,203
30,221
27,519
25,069
22,849
20,836
19,012
17,359
15,861
14,502
13,271
12,155
11,145
10,228
9,397
8,644
7,962
7,342
6,782
6,274
5,814
5,396
5,017
4,674
4,363
4,082
3,826
3,595
3,385
113,208
102,211
92,288
83,332
75,250
67,954
61,369
55,424
50,059
45,214
40,840
36,892
33,327
30,107
27,200
24,575
22,204
20,063
18,129
16,382
14,804
13,379
12,091
10,928
9,877
8,928
8,069
7,294
6,595
5,962
45,171
43,731
42,337
40,987
39,681
38,416
37,191
36,005
34,857
33,746
32,670
31,629
30,620
29,644
28,699
27,784
26,898
26,041
25,211
24,407
23,629
22,875
22,146
21,440
20,757
20,095
19,454
18,834
18,234
17,652
       To calculate the annual VMT by age for an average HDDV, we multiply the vehicle survival
distribution by the vehicle mileage accumulation by age. To get the travel fraction, we divide the
annual VMT by the total average lifetime miles. Table 6-5 presents the annual VMT by age for an
average HDDV and the total average lifetime miles.
                                                                                   114

-------
                         Chapter 6: Environmental Impact of HP Diesel Standards
                                     Table 6-5
                       Average Annual VMT by Age for HDD Vs
Vehicle Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Total
LHDDE
21,426
26,479
22,591
19,281
16,461
14,059
12,011
10,265
8,776
7,505
6,421
5,495
4,704
4,028
3,450
2,957
2,534
2,173
1,863
1,598
1,372
1,177
1,011
868
771
664
559
485
411
294
201,689
MHDDE
19,511
33,203
28,263
24,067
20,503
17,476
14,904
12,718
10,859
9,279
7,934
6,790
5,816
4,987
4,280
3,678
3,164
2,725
2,350
2,030
1,756
1,522
1,321
1,149
1,020
892
730
684
593
512
244,716
HHDDE
60,517
102,211
86,321
72,905
61,577
52,011
43,934
37,113
31,353
26,487
22,378
18,908
15,976
13,499
11,407
9,640
8,147
6,885
5,819
4,918
4,157
3,514
2,971
2,511
2,177
1,845
1,445
1,306
1,090
903
713,926
Urban Bus
22,579
43,731
42,337
40,987
39,681
38,416
37,191
35,995
34,847
33,707
32,547
31,273
29,692
27,427
23,876
18,381
11,115
5,136
4,070
3,228
2,559
2,028
1,605
1,275
1,353
655
634
614
297
288
567,521
III.  Total Nationwide Inventories

      This section looks at 2000 emission inventories of NOx, NMHC and PM from HDDEs as
well as projected inventories for NOx and NMHC. We present our projected baseline and controlled
emissions inventories in addition to our anticipated benefits.
                                                                               115

-------
Regulatory Impact Analysis
       A. Current Inventories

       The 1997 Trends Report estimates total nationwide emissions of NOx, VOC, and PM.  The
purpose of including these inventories here is to show the relative importance of heavy-duty sources.
The highway emissions were estimated using EPA's emissions factor models MOBILE5a (NOx and
NMHC) and PARTS (PM) and information from the Federal Highway Administration's Highway
Administration's Highway Performance Monitoring System and the 1980 U.S. census.  More
information about the methodologies used to estimate the mobile source emissions, as well as the
other emissions, can be found in the Trends Report.

       Due to recent information, we adjust the HDV NOx and VOC estimates to reflect the changes
described above and those described in the next chapter. In addition, we modify the nonroad
inventories to be consistent with recently finalized rules for land-based nonroad diesel engines8 and
for marine  diesel engines.9  Light duty vehicle estimates reflect emission  inventory estimates
projected in the recent Tier 2 rule10 with a small adjustment in light-duty VMT due to the updated
total VMT splits by class (see section HC).

       The national NOx, VOC, and PM10 inventories for HDVs are summarized in Table  6-6.
These data indicate that emissions from "current" heavy-duty diesel vehicles account for about 17
percent of total NOx emissions and 1.7 percent of total VOC emissions.  The PM numbers presented
in this table represent total vehicle emissions, which includes brake wear, and exhaust. Excluding
fugitive dust and wind erosion, HDDVs account for about 1.0 percent of total PM emissions.  The
PM data presented in Table 6-6 is for informational purposes, this rule does not establish new PM
standards for HD vehicles.

                                       Table 6-6
                       2000 National NOx, VOC, and PM Emissions
                              (thousand short tons per year)
Emission Source
Light-Duty Vehicles
Heavy-Duty Diesel Vehicles
Heavy-Duty Gasoline Vehicles
Nonroad
Other
Total Nationwide Emissions
NOx
4,005
4,181
298
5,343
10,656
24,485
VOC
3,965
282
241
2,485
9,567
16,540
PM10
96
131
8
642
8,206
9,044
       B. NOx Emission Projections and Impacts
                                                                                   116

-------
	Chapter 6: Environmental Impact of HP Diesel Standards

       We calculate NOx emissions using the same methodology as was used for the 1997 RIA.
However, this analysis uses new conversion factors, scrappage rates, and vehicle miles traveled as
described above. Baseline EFs and DFs for 1988 to 2003 model year engines are based on a report
which  considered certification  data from 1988 to the present.11  These  emission factors were
developed for use with MOBILE6.  For earlier years, we continue to use the EFs and DFs in
MOBILESb. As discussed in the introduction to this chapter, we modified the controlled emission
factor for NOx.

       For both the baseline and controlled EFs, we assume a compliance margin of 8%. We base
this compliance margin on historical certification data showing past practices. In other words, we
assume that the manufacturers will conservatively design their engines to be 8% below the standards.
Therefore, for a NOx standard of 2.3 g/bhp-hr, we use a level of 2.12 g/bhp-hr for the deteriorated
emission level at the regulatory useful life of the engine. Table 6-7 presents baseline and controlled
EFs and DFs for HDDEs. For the purposes of the HDDE inventory calculations, EF refers to the
emission factor at the end of the regulatory useful life.

                                       Table 6-7
                     NOx Emission Factors and Deterioration Rates for
                        2004 and Later Heavy-Duty Diesel Engines



Baseline
Controlled
Zero-Mile Level [g/bhp-hr]
LHDDE MHDDE HHDDE Urban
Bus
3.26 3.69 3.68 3.90
2.10 2.10 1.99 2.21
DR [g/bhp-hr per 10,000 miles]
LHDDE MHDDE HHDDE Urban
Bus
0.001 0.001 0.003 0.000
0.001 0.001 0.003 0.000
       In our current analysis of HDDE emissions, we may underestimate emissions due to engine
deterioration in-use. We believe, that current modeling only represents properly maintained engines
but may not be representative of in-use malmaintenance or tampering. One study12 shows much
larger deterioration rates for HDDEs. However, in the time frame of this rule, we have not been able
to fully consider this study or assess other relevant information. In our future modeling efforts, we
intend to strengthen this part of our analysis.

       The EFs and DFs in Table 6-7 above do not account for excess emissions from engines,
produced in the past, that operate with higher NOx in-use than on the certification test cycle which
were at issue in the  1998 consent decree between EPA and a number of HDDE manufacturers.
Although it does not affect the emissions benefits from this rule, we add excess emissions to our
annual NOx projections for some HDDE engines built in the 1988 through 1998 model years to
accurately represent the entire HDDE inventory over time. We use the excess emissions inventory
developed by the EPA's Office of Enforcement and Compliance Assurance.13
                                                                                   117

-------
Regulatory Impact Analysis
      Figure 6-1 shows our national projections of total NOx emissions with and without the new
engine controls.  The new standards should result in about a 43% reduction in NOx from new
engines. Table 6-8 presents these projections with the estimated NOx benefits for selected years.

          Figure 6-1: Projected National Exhaust NOx Emissions from HDDEs

,
,
CO
,>»4,
(/>
o
+J O
t °'
O
.G
w 2
X
O
z 1
i,

nnn nnn -,
uuu,uuu
nnn nnn
uuu,uuu
nnn nnn
UUU,UUU
nnn nnn
uuu,uuu

nnn nnn
uuu,uuu
n
20


k _^-*-*^^


^^*^*~^-« •-!



I I I
00 2005 2010 2015 20


r

—•—Baseline
, -"—Controlled



20
                                      Table 6-8
               Estimated National NOx Emissions and Benefits from HDDEs
                              (thousand short tons per year)
Calendar Year
2005
2010
2015
2020
2030
Baseline
3,410
3,470
3,830
4,250
5,130
Controlled
3,150
2,570
2,490
2,600
3,000
Benefit
266
900
1,340
1,650
2,130
                                                                                  118

-------
	Chapter 6: Environmental Impact of HP Diesel Standards

       C. NMHC Emission Projections and Impacts

       Estimates of the impact of the new standards on NMHC emissions are described below.
Heavy-duty engines do not currently have applicable NMHC standards, so the discussion in this
section focuses on total hydrocarbons. For this analysis, we assume that the effect of the combined
standards is equivalent to 0.2 g/bhp-hr NMHC-only standards. Emissions are modeled using the
same methodology as in the 1997 RIA with the updates described earlier.

       It should be noted that the analysis of the NMHC emission impacts is limited to a large extent
by the difficulty in projecting what the NMHC emissions from heavy-duty engines will be in the
future in the absence of new standards. This difficulty arises because NMHC emission levels from
heavy-duty diesel engines are largely the incidental result of a variety of other engine design
constraints, and thus are highly variable.  As is described below, the fact that total HC emissions
from current engines are  so far below the applicable HC standards, and that they vary among
different  engine families by more than an order of magnitude, is evidence of the incidental nature
of HC emission reductions. Although the current HC standard is much higher than the baseline HC
emission factors in our calculations, the actual certification levels for HDDEs  have shown
historically low HC emissions compared to the standards.

       Baseline EFs and DFs for 1988 to 2003 model year engines are based on a report which
considered certification data from 1988 to the present.14  These emission factors were developed for
use in MOBILE6. For earlier model years, we rely on the EFs and DFs in MOBILESb.  Table 6-9
presents the baseline and controlled emission factors and deterioration rates. As with NOx, we
assume a compliance margin of 8 percent.

                                      Table 6-9
                    NMHC Emission Factors and Deterioration Rates for
                        2004 and Later Heavy-Duty Diesel Engines

Baseline
Controlled
Zero-Mile Level [g/bhp-hr]
LHDDE MHDDE HHDDE Urban
Bus
0.26 0.31 0.22 0.08
0.17 0.17 0.14 0.08
DR [g/bhp-hr per 10,000 miles]
HDDE Urban Bus
0.001 0.000
0.001 0.000
       Figure 6-2 shows our national projections of total NMHC emissions with and without the
new engine controls. The new standards should result in about a 32% reduction in NMHC from new
engines. Table 6-10 presents these proj ections with the estimated NMHC benefits for selected years.
                                                                                   119

-------
Regulatory Impact Analysis
         Figure 6-2: Projected National Exhaust NMHC Emissions from HDDEs
  re
  0)
  c
  o
  o
  I
             2000
                                                                       Baseline
                                                                       Controlled
2005
2010
2015
2020
                                     Table 6-10
           Estimated National NMHC Emissions and New Benefits from HDDEs
                             (thousand short tons per year)
Calendar Year
2005
2010
2015
2020
2030
Baseline
253
251
296
339
409
Controlled
238
202
223
249
292
Benefit
14
49
73
90
117
IV.  Per Vehicle Emission Impacts

      Using the emissions factors and lifetime vehicle miles traveled described above, lifetime
emissions can be calculated for individual heavy-duty diesel vehicles. Table 6-11 presents the
lifetime benefits associated with this new control program on aper-vehiclebasis. Because emissions
                                                                                120

-------
                          Chapter 6: Environmental Impact of HP Diesel Standards
reductions are considered to be more valuable in the present than in the future, we present these
benefits both with and without a seven percent discount on the value of emissions reductions.

                                      Table 6-11
                    Per-Vehicle Average Lifetime Emission Reductions
                 Due to the New Standards for Heavy-Duty Diesel Engines
Vehicle
Category
LHDDV
MHDDV
HHDDV
Urban Bus
Undiscounted Reductions (Ibs.)
NOx NMHC
634 49
1,930 170
7,890 374
9,887 —
Discounted Reductions (Ibs.)
NOx NMHC
423 33
1,270 112
5,308 251
5,848 —
V.  Environmental Impacts of Emission Reductions

       A. Ozone Impacts

       We expect the effect of the reduced NOx on ozone concentrations to vary geographically.
In general, when fully phased-in, the effect of this action in most nonattainment areas should be a
reduction in ozone concentrations on the order of a few percent. It should be noted, however, that
the  potential exists for a few localized areas to  actually experience slight  increases in ozone
concentrations as a result of NOx emission reductions. The effect of the NMHC reductions on ozone
concentrations will be positive, though relatively small.

       B. Air Toxics

       The term "hydrocarbons" includes many different molecules. Speciation of the hydrocarbons
would show that many of the molecules are those which are considered to be air toxics including
benzene, formaldehyde, acetaldehyde, and 1,3-butadiene. Speciated hydrocarbon data was collected
for heavy-duty diesel engines.15'16'17'18  According to this data, hydrocarbons from a HDDV include
approximately 1.1 percent benzene, 7.8 percent formaldehyde, 2.9 percent acetaldehyde, and 0.6
percent 1,3-butadiene. Table 6-12 shows the estimated air toxics reductions  associated with the
hydrocarbon reductions in this rule.
                                                                                   121

-------
Regulatory Impact Analysis
                                       Table 6-12
                    Estimated Annual Air Toxics Reductions [short tons]
Year
2005
2010
2015
2020
Benzene
159
541
799
990
Formaldehyde
1,130
3,840
5,670
7,020
Acetaldehyde
419
1,430
2,110
2,610
1,3 -Butadiene
87
295
436
540
       C. Other Impacts of Emission Reductions

       The discussion in this section was contained in the RIA from the 1997 rulemaking for HD
diesel standards which established the 2004 HDDE NMHC+NOx standard, and the estimates
contained here were not re-analyzed  for the affirmation of these standards contained in this
rulemaking activity.  The expected reductions in NOx emissions should also positively affect
visibility, acid deposition, and estuary eutrophication. Both NO2 and nitrate particulates are optically
active, and in some urban areas, NO2 and nitrate particulates can be responsible for 20 to 40 percent
of the visible light extinction. The effect of this action on visibility should be small, given that it is
expected to reduce overall NOx emissions by several  percent.  For example, we expect the new
engine controls to result in about five percent less total  NOx in the year 2020, and therefore would
be expected to  decrease haze by  about one percent where NO2 and nitrate particulates cause 20
percent of the haze.

       We also expect the new standards to provide benefits with respect to acid deposition.  The
1.7 million ton per year reduction expected in 2020 as a result of this action is greater than the
400,000 ton per year reduction expected from Phase I of the Agency's acid rain NOx control rule (59
FR 13538, March 22, 1995), which was considered to be a significant step toward controlling the
ecological damage caused by acid deposition.  It is not clear, however, that reducing emissions of
NOx from ground-level sources such as heavy-duty vehicles is truly equivalent to reducing NOx
emissions from elevated smokestacks,  since NOx emitted higher into the atmosphere is likely to
travel further downwind, undergoing additional reactions before deposition. In any case, it is clear
that there will be some significant reduction in the adverse effects of acid deposition as a result of
this rule.

       This action should also lead to a reduction in  the nitrogen loading of estuaries.  This is
significant since high nitrogen loadings can lead to eutrophi cation of the estuary, which causes
disruption in the ecological balance. The effect should be most significant in areas heavily affected
by atmospheric NOx emissions. One such estuary is Chesapeake Bay, where as much as 40 percent
of the nitrogen loading may be caused by atmospheric deposition.

VI. Summary


                                                                                     122

-------
                          Chapter 6: Environmental Impact of HP Diesel Standards
       The proj ected total NOx and NMHC emission reductions as a result of this action are shown
in Figure 6-4. NOx reductions are projected to be about 1 .7 million tons per year in 2020. NMHC
reductions are projected to be much smaller, about 90,000 tons per year in 2020, which would be
much less than one percent of the national NMHC (or VOC) inventory.  These emission reductions
are expected to contribute very significantly towards reducing and controlling ambient ozone levels
in the future, counteracting the expected effects of new sources and growth in the vehicle miles
traveled. The new controls would also result in benefits with respect to visibility, acid deposition,
and estuary eutrophi cation.

              Figure 6-4:  Projected Benefits of Control for NOx and NMHC
      200,000
      175,000
      150,000
   "w"
   I 125,000
      100,000
   o
                                                  2,250,000

                                                -- 2,000,000

                                                - 1,750,000
                                                -- 1,500,000  of
                                                            1
                                                  1,250,000  t
                                                            o
                                                  1,000,000  OL
                                                            x
                                                -- 750,000   i
                                                                        500,000

                                                                      -- 250,000

                                                                        0
             2000
2005
2010
2015
2020
2025
2030
                                                                                    123

-------
Regulatory Impact Analysis
Chapter 6 References

1.  "Control of Emissions of Air Pollution from Highway Heavy-Duty Engines; Final Rule,"
Federal Register, pp.54694-54730, October 21, 1997.

2.  "Update Heavy-Duty Engine Emission Conversion Factors for MOBILE6: Analysis of Fuel
Economy, Non-Engine Fuel Economy Improvements, and Fuel Densities," U.S. Environmental
Protection Agency, EPA420-P-98-014, May 1998.

3.  "Update of Fleet Characterization Data for Use in MOBILE6 - Final Report," U.S.
Environmental Protection Agency, EPA420-P-98-016, June 1998.

4."Update Heavy-Duty Engine Emission Conversion Factors for MOBILE6: Analysis of Fuel
Economy, Non-Engine Fuel Economy Improvements, and Fuel Densities," U.S. Environmental
Protection Agency, EPA-420-P-98-014, May 1998.

5.      Petroleum Supply Annual, 1998, Volume #1, Energy Information Administration (EIA),
       Department of Energy (DOE), DOE/EIA-0340(98)/1.

6.      Draft MOBILE6 fuel economy estimates.

7.  "Fleet Characterization Data for MOBILE6: Development and Use of Age Distributions,
Average Annual Mileage Accumulation Rates and projected Vehicle Counts for Use in
MOBILE6," U.S. Environmental Protection Agency, EPA420-P-99-011, April 1999.

8.  "Control of Emissions of Air Pollution from Nonroad Diesel Engines; Final Rule," Federal
Register, pp. 56967-57023, October 23, 1998.

9.  "Control  of Emissions of Air Pollution from New CI Marine Engines at or Above 37 kW;
Final ¥M\Q" Federal Register, p. 73300, December 29, 1999.

10. "Control of Air Pollution from New Motor Vehicles: Tier 2 Motor Vehicle Emissions
Standards and Gasoline Sulfur Control Requirements," Regulatory Impact Analysis, EPA420-R-
99-023, U.S. Environmental Protection Agency, December 1999.

11. "Update of Heavy-Duty Emission Levels (Model Years 1988-2004+) for Use in MOBILE6,"
U.S. Environmental Agency, EPA420-R-99-010, April 1999.

12. "Modeling Deterioration in Heavy-Duty Diesel  Particulate Emissions," Engine, Fuel, and
Emissions Engineering, Incorporated, prepared for the U.S. Environmental Protection Agency,
September 30, 1998.

13. Memo from Mike Samulski to Docket A-99-06, "OECA Estimates of Defeat Device NOx
Emissions," November 10, 1999.
                                                                                124

-------
	Chapter 6:  Environmental Impact of HP Diesel Standards

14.  "Update of Heavy-Duty Emission Levels (Model Years 1988-2004+) for Use in MOBILE6,"
U.S. Environmental Agency, EPA420-R-99-010, April 1999.

15.  Springer, K., "Investigation of Diesel-Powered Vehicle Emissions VII," U.S. EPA, EPA-
460/3-76-034, 1977.

16.  Springer, K., "Characterization of Sulfates, Odor, Smoke, POM and Particulates from Light
and Heavy-Duty Engines - Part IX," U.S. EPA, EPA-460/3-79-007, 1979.

17.  Bass, E., and Newkirk, M., "Reactivity Comparison of Exhaust Emissions from Heavy-Duty
Engines Operating on Gasoline, Diesel, and Alternative Fuels," Southwest Research Institute,
SwRI9856, 1995.

18.  College of Engineering - Center for Environmental Research and Technology, "Evaluation
of Factors that Affect Diesel Exhaust Toxicity," submitted to California Air Resources Board,
Contract No. 94-312, 1998.
                                                                                125

-------
Regulatory Impact Analysis
  CHAPTER 7:  ENVIRONMENTAL IMPACT OF THE HD

                   OTTO-CYCLE  STANDARDS

I. Introduction

      This chapter describes the expected environmental impacts of the new exhaust and ORVR
standards for  heavy-duty gasoline engines and vehicles described  in the previous chapters.
Specifically, this chapter includes a description of how heavy-duty gasoline vehicle emission factors
were developed, the per-vehicle exhaust emission reductions due to the new standards over the life
of heavy-duty gasoline vehicles, the estimated exhaust NOx and NMHC emission inventories from
heavy-duty gasoline vehicles, and the exhaust emission benefits from the new exhaust standards.
Last of all, the chapter concludes with a description of the emission benefits from the new ORVR
requirements for Class 2b heavy-duty gasoline vehicles.

      In evaluating the environmental impact of today's heavy-duty gasoline engine and vehicle
standards for 2005 and later, we developed estimates of exhaust NOx and NMHC emissions from
HDGVs (excluding California) both with and without the effect of the standards.  The analysis
performed to estimate the emission reductions from HD gasoline vehicles and engines in this final
rule is identical to the analysis performed for the Agency's recently announced proposal to reduce
emissions from HD gasoline engines in the 2007 time frame (published on June 2, 2000 (65 FR
35430)). This analysis is  different than the analysis we performed for the proposed rulemaking. In
the proposal we used the EPA MOBILES emission model, with in-use adjustment factors developed
specifically for the proposal. As discussed in the RIA, the draft MOBILE6 emission rates for HD
gasoline engines and vehicles have been completed, so we use those emission rates in this final rule.
Because MOBILE6 is not complete, we  used the updated emission rates from MOBILE6 in
MOBILES for our analysis.  The EPA report in which these emission rates are reported has gone
through an external stakeholder review."1 For this final rule we use zero-mile and deterioration rates
for 1988 and later model  year HD gasoline exhaust emissions developed for the draft MOBILE6
emission model. The impact of this change on this final rule, as compared to the proposal, was to
decrease the estimated in-use emission rates, for both the baseline and controlled scenarios, for 1998
and later model year HD gasoline engines. Full details of the environmental impact analysis can be
found in Chapter 7  of the RIA. The following paragraphs summarize the key results.

II. Exhaust NOx and NMHC Emission Factors

      A. Baseline Emission Rates (Zero-Mile Levels and Deterioration Rates)

      To determine the  impact of the standards, we first estimate the  emission levels of vehicles
currently in the fleet and then estimate the emission levels of vehicles that will meet the new
(m)  "Update of Heavy-Duty Emission Levels (Model Years 1988-2004+) for Use in MOBILE6", EPA document
EPA-420-R-99-010.

                                                                               126

-------
                         Chapter 7: Environmental Impact HP Otto-cycle Standards
standards. For the emission rates of engines currently in the fleet, we use the recently updated zero-
mile level and deterioration rates for 1988 and later model years that were developed for the
upcoming MOBILE6 model.1 (For pre-1988 model year heavy-duty gasoline vehicles, we use the
standard  MOBILES emission rates.) Because MOBILE6 is not yet complete, we perform our
emissions calculations by running MOBILES with the updated emission rates.  Table 7-1 presents
the zero-mile level in grams per brake horsepower-hour (g/bhp-hr) and the deterioration rate in
g/bhp-hr per 10,000 miles for 1988 and later model year heavy-duty gasoline engines.

                                       Table 7-1
               Baseline Exhaust Emission Rates for 1988 and later Model Year
                              Heavy-Duty Gasoline Engines
Model Year
1988-1989
1991
1991-1997
1998-2004
Zero-Mile Level
g/bhp-hr
NOx NMHC
4.96 0.62
3.61 0.35
3.24 0.33
2.59 0.33
Deterioration Rate
g/bhp-hr per 10,000 miles
NOx NMHC
0.044 0.023
0.026 0.023
0.038 0.021
0.038 0.021
       B. Conversion Factors

       Up until this rulemaking,  we expressed the emission standards for heavy-duty gasoline
engines in units of g/bhp-hr.  To convert the emissions of engines certified to g/bhp-hr standards to
g/mi levels, we multiply the g/bhp-hr levels by a conversion factor that is expressed in units of bhp-
hr/mi.  The conversion factor is a function of fuel density, brake specific fuel consumption and fuel
economy. We recently updated the conversion factors for heavy-duty engines.2 Table 7-2 contains
the conversion factors assumed in this analysis for heavy-duty gasoline engines.

                                       Table 7-2
              Conversion Factors for Heavy-Duty Gasoline Engines (bhp-hr/mi)
Class 2b
1.096
Class 3
1.150
Class 4
1.134
Class 5
1.324
Class 6
1.311
Class 7
1.446
Class 8a
1.540
       C. Control Emission Rates (Zero-Mile Levels and Deterioration Rates)

       Using the 120,000 mile useful life for heavy-duty gasoline engines, we estimate that a typical
1998 and later model year heavy-duty gasoline engine would emit NOx at roughly 3.0 g/bhp-hr, or
                                                                                   127

-------
Regulatory Impact Analysis
75 percent of the level of the standard of 4.0 g/bhp-hr.   This is based on current and historical
certification data for heavy-duty Otto-cycle engines, which shows manufacturers have relied on a
large compliance margin when certifying,  as well as statements from the engine manufacturers
indicating they attempt to target certification levels as low as one-half the emission standard (see the
discussion in Chapter 3(in)(H)  for additional discussion of the  manufacturers statements on
certification levels).  Assuming manufacturers maintain the same  amount of cushion below the
standard, we estimate the end of useful life level emissions levels associated with the new standards.
From these reduced levels, we determine the corresponding zero-mile levels and deterioration rates
assuming the ratio of the zero-mile level emissions to the deterioration rate (for 1998 engines) stays
the same as shown in Table 7-1.  Table 7-3 presents the resulting controlled zero-mile levels and
deterioration  rates for the three classes of heavy-duty gasoline engines and vehicles used in this
analysis (i.e., Class 2b complete vehicles, Class 3 complete vehicles, and incomplete HDGVs).

                                        Table 7-3
         Estimated Controlled Exhaust Emission Rates for 2005 and later Model Year
                        Heavy-Duty Gasoline Engines and Vehicles
Vehicle Category
Class 2b Completes
Class 3 Completes
Incomplete HDGVs
Zero-Mile Level
grams per mile (g/mi)
NOx NMHC
0.574 0.119
0.638 0.140
0.674 0.117
Deterioration Rate
g/mi per 10,000 miles
NOx NMHC
0.008 0.008
0.009 0.009
0.009 0.007
       D. Emission Rate Adjustments

       In the Draft RIA for the NPRM, we adjusted the emission rates in an attempt to quantify
increases in in-use emissions due to fuel that differs from the test fuel, to driving that differs from
the test cycle, fuel that differs from the test fuel, and to account for "high emitters" in the fleet
(engines that have much higher emissions than certification due to  emission control failure).
However, we remove these adjustments from the final analysis.

       MOBILES includes adjustment factors for HDGVs to account for in-use fuel with different
properties than certification test fuel. However, this adjustment is primarily driven by higher sulfur
levels seen in-use.  Due to the recent Tier 2 standards for light-duty vehicles which included low
sulfur fuel requirements, this adjustment is no longer needed once the low sulfur fuel goes into effect
in 2004.  Therefore we t urn off this fuel adjustment for 2004 and later calendar years. As a result,
the projected emission reductions from HDGVs, which begin in 2005, are not affected by this
adjustment.
                                                                                     128

-------
	Chapter 7: Environmental Impact HP Otto-cycle Standards

       MOBILES includes speed correction factors which adjust the emission rates relative to the
average speeds above and below the average speed of the federal test procedure. These adjustments
consider the type of operation that is typically seen at these various average speeds such as stop-and-
go driving at low average speeds. However, these correction factors are not specifically designed
to account for off-cycle effects such as control strategies that are designed to the test procedure that
may not work as  well off the test procedure.   We plan to work with stakeholders to explore
augmenting our existing test procedures to address off-cycle concerns.  If appropriate we will
develop a regulatory proposal to address this issue.  For this reason we do not include a further
adjustment for in-use operation in our final analysis.

       MOBILES does not include high-emitter adjustments for HDGVs. In addition, no high-
emitter adjustments have been developed for the upcoming MOBILE6 model. Our NPRM analysis
used high-emitter adjustments that were based on information collected on light-duty vehicles and
may not represent heavy-duty vehicle emissions. For the final analysis we believe it is appropriate
to be consistent with the upcoming MOBILE6 model. In the future, we intend to look into this issue
further.

III.  Per-Vehicle  Exhaust NOx and NMHC Emission  Reductions

       To determine the cost-effectiveness of the new standards, we estimate the per-vehicle
emissions and emission reduction over the lifetime of typical heavy-duty gasoline vehicles. The
following sections presents the per-vehicle emission reduction analysis for three sub-categories of
heavy-duty gasoline vehicles (Class 2b completes, Class 3 completes, and incomplete HDGVs).

       A. Per Vehicle Emission  Rates

       To estimate the per-vehicle lifetime emission reduction from the new standards, we first
estimated the emission rates of pre-control engines (i.e., model year 1998-2003  model years) and
controlled engines (i.e., model year 2004 and later).  Table 7-4 presents the zero-mile levels and
deterioration rates that we use in our analysis.  This is the same as Tables 7-1 and 7-3, except that
the baseline rates are broken out by class and expressed in units of g/mi using the conversion factors
in Table 7-2.
                                                                                    129

-------
Regulatory Impact Analysis
                                      Table 7-4
                Final Exhaust Emission Rates for Pre-control and Controlled
                       Heavy-Duty Gasoline Engines and Vehicles
Vehicle
Category
Class 2b
Completes
Class 3
Completes
Incomplete
HDGVs
Model Year
Grouping
1998-2004
2005+
1998-2004
2005+
1998-2004
2005+
Zero-Mile Level,
grams per mile (g/mi)
NOx NMHC
2.839 0.362
0.574 0.119
2.979 0.380
0.638 0.140
3.435 0.438
0.674 0.117
Deterioration Rate
g/mi per 10,000 miles
NOx NMHC
0.042 0.023
0.008 0.008
0.044 0.024
0.009 0.009
0.050 0.028
0.009 0.007
       B. Mileage Accumulation and Scrappage Rates

       Table 7-5 presents the HDGV mileage accumulation rates and scrappage rates used in this
analysis.  The mileage accumulation rates come from our recently updated rates for heavy-duty
gasoline vehicles developed for the MOBILE6 emissions model.3 We took the scrappage rates from
aNational Highway Traffic Safety Administration (NHTSA) study. These scrappage rates are based
on light-duty truck (LDT) scrappage rates.4 (The scrappage rate represents the fraction of engines
still in the fleet at a given age.) The NHTSA study did not include information on HDGVs.  We
believe the LDT scrappage rates would be similar to those for most HDGVs since three-quarters of
all HDGV sales are in the Class 2b truck category, which is the weight category just above the LDT
cutoff of Class 2a trucks.
                                                                                  130

-------
                        Chapter 7: Environmental Impact HP Otto-cycle Standards
                                      Table 7-5
    Annual Mileage Accumulation, Scrappage, and Composite Mileage Accumulation Rates
                            for Heavy-duty Gasoline Vehicles
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25+
Class 2b/3
Annual Mileage
19,977
18,779
17,654
16,596
15,601
14,666
13,787
12,961
12,184
11,454
10,768
10,122
9,516
8,946
8,409
7,905
7,432
6,986
6,568
6,174
5,804
5,456
5,129
4,822
4,533
Class 4+
Annual Mileage
21,394
19,692
18,125
16,683
15,356
14,134
13,010
11,975
11,022
10,145
9,338
8,595
7,911
7,282
6,703
6,169
5,679
5,227
4,811
4,428
4,076
3,752
3,453
3,178
2,926
Scrappage Rate
0.998
0.995
0.989
0.980
0.967
0.949
0.924
0894
0.857
0.816
0.795
0.734
0.669
0.604
0.539
0.476
0.418
0.364
0.315
0.271
0.232
0.198
0.169
0.143
0.648
       Table 7-6 contains the annual mileage accumulation rates for typical Class 2b/3 vehicles and
typical incomplete vehicles factoring the effect of scrappage. (For the incomplete vehicles, We
sales-weight the mileage accumulation rates for Class 2b/3 and Class 4+ vehicles in Table 7-8 based
on sales data on incomplete vehicles submitted by manufacturers to EPA.)
                                                                                  131

-------
Regulatory Impact Analysis
                                     Table 7-6
               Annual Mileage Accumulation Rates (Factoring in Scrappage)
                        for Typical Heavy-duty Gasoline Vehicles

Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25+
Lifetime Mileage
Class 2b/3 Complete
Vehicle Annual Mileage
19,937
18,685
17,460
16,264
15,086
13,918
12,739
11,587
10,442
9,346
8,561
7,430
6,366
5,403
4,532
3,763
3,107
2,543
2,069
1,673
1,347
1,080
867
690
2,937
197,832
Incomplete
Vehicle Annual Mileage
20,524
19,062
17,653
16,299
14,988
13,709
12,441
11,221
10,028
8,903
8,089
6,964
5,921
4,986
4,151
3,420
2,802
2,277
1,839
1,477
1,180
940
749
592
2,505
192,722
      C.  Per-vehicle Lifetime Emissions and Emission Reductions

      Table 7-7 presents the NOx and NMHC emissions from typical heavy-duty gasoline vehicles
over the life of the vehicle.  We determine these levels by combining the emission rate information
contained in Table 7-7 with the mileage accumulation rate information contained in Table 7-9.
                                                                                132

-------
                         Chapter 7: Environmental Impact HP Otto-cycle Standards
                                      Table 7-7
          Lifetime NOx and NMHC Emissions from Heavy-duty Gasoline Vehicles
Vehicle
Category
Class 2b
Completes
Class 3
Completes
Incomplete
HDGVs
Model Year
Grouping
1998-2004
2005+
1998-2004
2005+
1998-2004
2005+
Undiscounted,
Emissions,
NOx
0.71
0.14
0.74
0.15
0.83
0.16
Lifetime
tons
NMHC
0.13
0.04
0.13
0.04
0.83
0.16
       Table 7-8 presents the expected per vehicle NOx and NMHC emission benefits for heavy-
duty gasoline vehicles from the new exhaust emission standards, both undiscounted and discounted
(at a rate of seven percent). In addition to the three subclasses of heavy-duty gasoline vehicles, Table
7-8 also contains the reductions for all HDGVs calculated on a sales-weighted basis from the three
individual categories.

                                      Table 7-8
                        Per Vehicle Exhaust Emission Reductions
                from the Heavy-duty Gasoline Engine and Vehicle Standards
Vehicle Category
Class 2b Completes
Class 3 Completes
Incomplete HDGVs
All HDGVs
Undiscounted Lifetime
Emission Reductions, tons
NOx NMHC
0.57 0.09
0.60 0.10
0.67 0.11
0.60 0.09
Discounted Lifetime
Emission Reductions, tons
NOx NMHC
0.38 0.05
0.40 0.06
0.45 0.07
0.40 0.06
IV.  HDGV Exhaust Inventory and Reductions

       To estimate the exhaust NOx and NMHC inventories from heavy-duty gasoline vehicles, we
calculate the average emissions of all heavy-duty gasoline vehicles in the fleet for a variety of years.
                                                                                  133

-------
Regulatory Impact Analysis
To estimate the fleet average emissions for heavy-duty gasoline vehicles, we ran the MOBILESb
emissions model with the updated information on emission levels and vehicle usage characteristics
as described in Sections II and HI of this chapter. We multiply these resulting fleet average emission
levels by the estimated fleetwide vehicle miles traveled (VMT) for heavy-duty gasoline vehicles for
the corresponding year to yield the exhaust emission inventories.

       We use the same methodology as described in Chapter 6 for HDDEs to determine the annual
VMT.  We use Federal  Highway Administration estimates of total VMT  then use information
collected for MOBILE6 to split by class and fuel type. We exclude miles traveled by medium-duty
passenger vehicles because they are covered by the Tier 2 FRM for light-duty vehicles.

       Table  7-9 presents the exhaust  NOx and NMHC fleet average emissions, VMT, and
inventories from heavy-duty gasoline vehicles both without the new standards and with the new
standards taking effect in the 2004 model year.  The inventories presented in Table 7-9  represent
nationwide inventories excluding California.  California is excluded to simplify this analysis since
they are claiming reductions from their own emission program for these engines. A more detailed
description of the inventory development has been placed in the docket for this rulemaking.5

                                       Table 7-9
 Fleetwide Exhaust NOx and NMHC Emission Factors, Vehicle Miles Traveled, and Inventories
                   from Heavy-duty Gasoline Vehicles (49-state analysis)

Pollutant Calendar
Year

NOx 2000
2005
2010
2015
2020
2030
NMHC 2000
2005
2010
2015
2020
2030
49-state
VMT
(1010 miles)

6.68
7.83
9.03
10.27
11.52
14.01
6.68
7.83
9.03
10.27
11.52
14.01
Heavy-duty Gasoline Vehicle Fleet Emission Levels
without new
g/mi 10
4.05
3.65
3.47
3.34
3.31
3.29
1.70
1.06
0.74
0.66
0.66
0.66
standards
3 short tons
298
315
345
378
420
507
125
91
73
75
84
102
with the
g/mi
4.05
3.50
2.18
1.46
1.12
0.83
1.70
1.04
0.59
0.43
0.38
0.34
new standards
103 short tons
298
302
217
165
142
127
125
90
59
49
48
52
       Table 7-10 contains the estimated exhaust NOx and NMHC emission reductions due to the
new standards for heavy-duty gasoline vehicles. As noted above, the reductions are for the entire
                                                                                    134

-------
                        Chapter 7: Environmental Impact HP Otto-cycle Standards
United States excluding California. Figures 7-1 and 7-2 present the heavy-duty gasoline vehicle
exhaust NOx and NMHC inventories, respectively.

                                    Table 7-10
                49-State Exhaust Emission Reductions due to the Standards
                     for Heavy-duty Gasoline Engines and Vehicles
Calendar Year
2005
2010
2015
2020
2030
Emission
NOx
13
128
213
278
380
Benefits (thousand short tons)
NMHC
1
14
26
35
50
          Figure 7-1: Projected 49-State Exhaust NOx Emissions from HDGVs
   600,000

   500,000
C*
re
> 400,000

I
e  300,000
   "  200,000
   O
       100,000
                          •Baseline
                          • Controlled
             o
             2000     2005     2010     2015     2020     2025     2030
                                                                               135

-------
Regulatory Impact Analysis
         Figure 7-2: Projected 49-State Exhaust NMHC Emissions from HDGVs

        140,000

        120,000
         20,000
               0
                   Baseline
                   Controlled
               2000
2005
2010
2015
2020
2025
2030
V. ORVR Benefits

       Along with the new exhaust standards, we are requiring ORVR regulations for Class 2b
heavy-duty gasoline vehicles.  Back in the early 1990s, we proposed, but never finalized, ORVR
requirements for heavy-duty gasoline vehicles.6 For this analysis, we rely on the earlier analysis to
estimate the HC benefits of ORVR requirements.

       Because many areas of the country have Stage II vapor recovery" on fuel pumps at the gas
station, we developed an estimate of the HC benefits that were attributable to the ORVR equipment.
For this analysis, we assume that Stage n will remain in place in the areas that currently have Stage
II controls even after the ORVR requirements for light-duty vehicles and trucks have finished taking
effect.  This assumption lowers the benefits attributable to the ORVR requirements and likely results
in a conservative estimate of benefits and cost-effectiveness as well.

       Table 7-11 presents the assumptions we use in estimating the  per-vehicle HC emission
benefits attributable to the new ORVR requirements for Class 2b heavy-duty gasoline vehicles and
the estimated benefits. The gram per gallon (g/gal) refueling HC emission benefit is taken from
(n) Stage II vapor recovery systems on fuel pumps use hoods over the refueling nozzle to collect vapor from
vehicles during refueling.
                                                                                    136

-------
	Chapter 7: Environmental Impact HP Otto-cycle Standards

Table 4.10 of the above mentioned rulemaking. The gallon/mile (gal/mi) Class 2b fuel consumption
value is taken from the MOBILE6 Conversion Factor report referenced earlier. We determined the
benefits over the lifetime mileage accumulation of a typical Class 2b heavy-duty gasoline vehicle
as specified in Table 7-6 of this chapter on both an undiscounted basis and a discounted basis (at a
rate of seven percent).

                                     Table 7-11
      Determination of Per-Vehicle Hydrocarbon Benefits from the ORVR Requirements
     	for Class 2b Heavy-duty Gasoline Vehicles	
      Refueling Hydrocarbon Emission Benefit Rate                  2.42 g/gal

      Class 2b Heavy-duty Gasoline Vehicle Fuel Consumption      0.0987 gal/mi

      Lifetime Undiscounted Emission Benefit                       0.052 tons

      Lifetime Discounted Emission Benefit                         0.033 tons
                                                                                  137

-------
Regulatory Impact Analysis
Chapter 7 References

1.  "Update of Heavy-Duty Emission Levels (Model Years 1988- 2004+) for Use in MOBILE6,"
EPA Report No. EPA420-R-99-010, Christian Lindhjem and Trade Jackson, U.S EPA, QMS,
AMD, April 1999.

2.  "Update Heavy-Duty Engine Emission Conversion Factors for MOBILE6," prepared by
Arcadis for EPA, May 1998.

3. "Update of Fleet Characterization Data for Use in MOBILE6," prepared by Arcadis for EPA,
May 1998.

4. "Updated Vehicle Survivability and Travel Mileage Schedules," U.S. Department of
Transportation, National Highway Traffic Safety Administration, November 1995.

5.  "Development of Heavy-Duty Gasoline Emissions Inventories for the Tier 2/Sulfur NPRM,"
EPA memo from John W. Koupal to Docket A-97-10, March 26, 1999.

6.  "Final Regulatory Impact Analysis: Refueling Emission Regulations for Light Duty Vehicles
and Trucks and Heavy Duty Vehicles," U.S EPA, OAR, QMS, RDSD, SRPB, January 1994.
                                                                               138

-------
                                                      Chapter 8:  Cost-Effectiveness
       CHAPTER 8:   COST-EFFECTIVENESS FOR HD

       DIESEL AND  OTTO-CYCLE REQUIREMENTS

       This chapter assesses the cost-effectiveness of the requirements for new heavy-duty engines,
including the new standards, OBD, useful life, allowable maintenance, in-use testing, and rebuild
provisions. This analysis relies in part on cost information from Chapters 4 and 5 and emissions
information from Chapters 6 and 7 to estimate the cost-effectiveness of the provisions in terms of
dollars per ton of total emission reductions.

       Separate analyses were performed for otto-cycle engines and diesel engines. The analysis
presented  in this chapter for heavy-duty diesel vehicles is an updated version  of the analysis
performed for the 1997 FRM. Both the otto-cycle and diesel analyses were performed on a per-
vehicle basis using total  costs and total NOx  plus NMHC emission reductions over the typical
lifetime of heavy-duty vehicle, discounted at a rate of seven percent to the beginning of the vehicle's
life. Analyses of the fleet cost-effectiveness for 30 model years after the new engine standards take
effect are also presented.

       The following section describes the cost-effectiveness of the new engine NOx and NMHC
standards for the various categories of heavy-duty diesel vehicles noted above.  As discussed in
Chapters 5 and 6, the estimated cost of complying with the provisions varies depending on the model
year under consideration.    Therefore, the  following  section  presents the  per-vehicle  cost-
effectiveness results for the different model years during which the costs are expected to change.
Just as the emission standard combines NOx  and NMHC emissions, the cost-effectiveness of
adopting the new standard is calculated by dividing the combined NOx and  NMHC emission
reductions into the cost of compliance.

       Also presented  is the  fleet cost-effectiveness over the first 30 model  years after the new
engine standards take  effect  (i.e., model years 2004  through 2033).  These cost-effectiveness
numbers are calculated by weighting the various model year per-vehicle cost-effectiveness results
by the fraction of the total 30 model year sales they represent. The sales for the different categories
of heavy-duty diesel engines that would be covered by the rule based on the 1995 model year were
determined using production information provided by manufacturers to EPA and were assumed to
grow at a linear rate of two percent from the  1995 levels. It is important to  note that 30-year
estimates are discounted so that they emphasize the higher costs which occur during the first several
years of these programs.

I.  Cost-Effectiveness of the Diesel Requirements

       Tables 8-1, 8-2, and 8-3 contain the total net present value costs based on the information
presented in Chapter 4, the lifetime emission reductions as presented in Chapter 6, and the resulting
cost-effectiveness values  for light-, medium-, and heavy-heavy duty diesel vehicles, respectively.
Tables 8-1, 8-2, and 8-3 also  contain the fleet cost-effectiveness covering the first 30 model years
                                                                                   139

-------
Regulatory Impact Analysis
after the new engine standards take effect (i.e., model years 2004 through 2033). As was noted in
Chapters 4 and 6,  we are modeling both the costs and benefits of the supplemental  testing
requirements as if they begin in model year 2004.  We are doing this  because we expect that
manufacturers will design most, if not all, of their engine models to comply with these requirements
in 2004.
                                      Table 8-1
                  Cost-Effectiveness for Light Heavy-Duty Diesel Vehicles
Model Year
Grouping
2004-05
2006-08
2009+
30 Year Fleet
Total NPV
Costs per
Vehicle
$493
$417
$249
—
Discounted
Lifetime Reduction (tons)
NOx
0.232
—
NMHC
0.018
—
Discounted
Per- Vehicle
Cost-
Effectiveness
($/ton of
NOx+NMHC)
$1,969
$1,668
$995
$1,230
                                      Table 8-2
                Cost-Effectiveness for Medium Heavy-Duty Diesel Vehicles
Model Year
Grouping
2004-05
2006-08
2009+
30 Year Fleet
Total NPV
Costs per
Vehicle
$706
$620
$323
—
Discounted
Lifetime Reduction (tons)
NOx
0.764
—
NMHC
0.067
—
Discounted
Per- Vehicle
Cost-
Effectiveness
($/ton of
NOx+NMHC)
$849
$746
$389
$506
                                                                                  140

-------
                                                        Chapter 8:  Cost-Effectiveness
                                        Table 8-3
                  Cost-Effectiveness for Heavy Heavy-Duty Diesel Vehicles
Model Year
Grouping
2004-05
2006-08
2009+
30 Year Fleet
Total NPV
Costs per
Vehicle
$907
$792
$472
—
Discounted
Lifetime Reduction (tons)
NOx
3.189
—
NMHC
0.151
—
Discounted
Per- Vehicle
Cost-
Effectiveness
($/ton of
NOx+NMHC)
$272
$237
$141
$174
       Table 8-4 contains the total net present value costs, the lifetime emission reductions, and the
resulting cost-effectiveness values for all heavy-duty diesel vehicles.  In determining the cost-
effectiveness for all heavy-duty diesel vehicles, the cost and emission reductions for all heavy-duty
diesel vehicles were determined by weighting the corresponding light, medium, and heavy heavy-
duty diesel vehicle results by the respective sales estimates for each year.

                                        Table 8-4
                   Cost-Effectiveness for All Heavy-Duty Diesel Vehiclesa
Model Year
Grouping
2004-05
2006-08
2009+
30 Year Fleet
Total NPV
Costs per
Vehicle
$682
$591
$342
—
Discounted
Lifetime Reduction (tons)
NOx
1.365
—
NMHC
0.074
—
Discounted
Per- Vehicle
Cost-
Effectiveness
($/ton of
NOx+NMHC)
$474
$410
$238
$296
         combined cost-effectivness weighted by distribution of vehicles by class
                                                                                      141

-------
Regulatory Impact Analysis
II. Cost-Effectiveness of the Otto-cycle Requirements

       A. Exhaust Emission Standards

       We analyzed the cost-effectiveness of the new exhaust emission standards for three different
categories of heavy-duty Otto-cycle vehicles.  The three categories analyzed were incomplete
vehicles, Class 2b complete vehicles, and Class 3 complete vehicles. Tables 8-5 through 8-7 contain
the discounted lifetime per-vehicle cost  based on the information in Chapter 5, the discounted
lifetime emission reductions as presented in Chapter 7, and the resulting cost-effectiveness values
for the three categories of heavy-duty Otto-cycle vehicles. Each of the tables also contains the fleet
cost-effectiveness covering the first 30 model years after the new standards take effect (i.e., model
years 2005 through 2034). Table 8-8 contains the cost-effectiveness of the new  standards for all
categories of heavy-duty Otto-cycle vehicles combined.  A copy of the spreadsheet prepared for the
heavy-duty Otto-cycle vehicle cost-effectiveness analysis has been placed in the public docket for
the notice of new rulemaking.1 The reader is directed to the spreadsheet for a complete version of
the cost-effectiveness calculations.

                                       Table 8-5
              Cost-Effectiveness for Incomplete Heavy-Duty Otto-cycle Vehicles
Model
Year
Grouping
2005-09
2010+
30 Year
Fleet
Total
NPV
Cost per
Vehicle
$296
$256
—
Discounted Lifetime
Reduction (tons)
NOx
0.45
—
NMHC
0.07
—
Discounted
Per-Vehicle
Cost-Effectiveness
($/ton of
NOx+NMHC)
$565
$489
$511
                                                                                     142

-------
                                           Chapter 8: Cost-Effectiveness
                            Table 8-6
Cost-Effectiveness for Class 2b Complete Heavy-Duty Otto-cycle Vehicles
Model
Year
Grouping
2005-09
2010+
30 Year
Fleet
Total
NPV
Cost per
Vehicle
$274
$273
—
Discounted Lifetime
Reduction (tons)
NOx
0.38
—
NMHC
0.05
—
Discounted
Per-Vehicle
Cost-Effectiveness
($/ton of
NOx+NMHC)
$635
$633
$634
                            Table 8-7
 Cost-Effectiveness for Class 3 Complete Heavy-Duty Otto-cycle Vehicles
Model
Year
Grouping
2005-09
2010+
30 Year
Fleet
Total
NPV
Cost per
Vehicle
$274
$273
—
Discounted Lifetime
Reduction (tons)
NOx
0.40
—
NMHC
0.06
—
Discounted
Per-Vehicle
Cost-Effectiveness
($/ton of
NOx+NMHC)
$596
$594
$595
                                                                       143

-------
Regulatory Impact Analysis
                                       Table 8-8
                 Cost-Effectiveness for All Heavy-Duty Otto-cycle Vehicles
Model
Year
Grouping
2005-09
2010+
30 Year
Fleet
Total
NPV
Cost per
Vehicle
$281
$268
—
Discounted Lifetime
Reduction (tons)
NOx
0.40
—
NMHC
0.06
—
Discounted
Per-Vehicle
Cost-Effectiveness
($/ton of
NOx+NMHC)
$612
$586
$598
       B. Refueling Emission Standards

       We also separately analyzed the cost-effectiveness of the new onboard vapor recovery
requirements  for complete Class 2b heavy-duty Otto-cycle vehicles.  Table 8-9 contains the
discounted lifetime per-vehicle cost based on the information in Chapter 5, the discounted lifetime
emission reductions as presented in Chapter 7, and the resulting cost-effectiveness values for the new
ORVR requirements for complete Class 2b heavy-duty otto-cycle vehicles.

                                      Table 8-9
          Discounted, Lifetime Cost-effectiveness of the New ORVR Requirements
                   for Complete Class 2b Heavy-duty Otto-cycle Vehicles
Model Year
Grouping
2005-2009
2010+
Discounted
Lifetime Cost
$5
$2
Discounted Lifetime NMHC
+ NOx Emission Reductions
0.035 tons
0.035 tons
Discounted Lifetime Cost-
effectiveness
$141/tonofNMHC
$56/ton NMHC
                                                                                   144

-------
                                                       Chapter 8:  Cost-Effectiveness
III.  Other Benefits

       In addition to the primary benefit of reducing ozone within and transported into urban ozone
nonattainment areas, the NOx reductions from the new engine standards are expected to have other
benefits as well. These other benefits, which are discussed in Chapter 2, include impacts with
respect to agricultural yields, visibility, soiling (due to secondary paniculate), and ecosystems (e.g.,
through the reduced effects of acid deposition and eutrophication).  These benefits are real, and they
have monetary value. For the 1997 FRM for on-highway FID diesels, an EPA contractor report from
1996 was used to estimate the monetary value of a number of these other benefits.2 However, EPA
has been reevaluating the techniques used to estimate  the value of these benefits since 1996.
Therefore, it is not presented here.

IV.  Cost-Effectiveness Sensitivity Analyses

       The following section presents an analysis of the sensitivity of the cost-effectiveness results
for heavy-duty diesel vehicles to different assumptions regarding the impact of the new standards
on fuel economy or other costs. As noted in Chapter 4, EPA is projecting a  small increase in
operating costs associated with oil changes and EGR maintenance. However, based on the analysis
discussed in Chapter 3 and 4, as well as the substantial lead time available for R&D, EPA is not
projecting other significant maintenance costs, or losses in fuel economy or engine durability. Even
if such impacts were to occur for a few engines, EPA believes that they could be eliminated with
additional R&D, and would thus be short-term in nature. Nevertheless, as a sensitivity analysis, EPA
estimated the discounted per-vehicle lifetime cost associated with a 1.0 percent fuel economy penalty
calculated over the typical lifetime of each class of heavy-duty diesel vehicles.  These costs are
shown in Table 8-11. To calculate the effect on cost-effectiveness of the new standards with the fuel
economy penalty, the fuel economy penalty costs in Table 8-11  were divided by the emission
reductions (as presented in Table 8-4). Table 8-12 contains the resulting discounted per-vehicle cost-
effectiveness numbers.
                                       Table 8-11
                Increase in Discounted Per-Vehicle Lifetime Operating Costs
                    Associated with a One Percent Fuel Economy Penalty
                       For Diesel Fuel Cost of One Dollar per Gallon
Light HD
$102
Medium HD
$178
Heavy HD
$891
                                                                                     145

-------
Regulatory Impact Analysis
                                      Table 8-12
              Effect on Cost-Effectiveness for All Heavy-Duty Diesel Vehicles
                         For a One Percent Fuel Economy Change


Fuel Price
($/Gal)



$1.00
$1.50


Average
NP V Fuel
Cost


$390
$585
Discounted
Lifetime Reduction
from New
Standards (tons)



NOx
1.365
1.365


NMHC
0.074
0.074

Increase in Per- Vehicle
Cost-Effectiveness
($/ton) for Each
One Percent Increase in
Fuel Consumption

$271
$407
       EPA performed a similar sensitivity analysis to show the effect of assuming that only 50
percent of the costs for VGT and improved fuel injection are attributable to emission control. In this
sensitivity analysis, EPA included the full costs for VGT and improved fuel inj ection in the estimates
of per vehicle costs, and recalculated the total cost effectiveness of the program. The results are
shown in Table 8-13.  The effect of this assumption can be seen by comparing this table with Table
8-4.
                                      Table 8-13
                   Cost-Effectiveness for All Heavy-Duty Diesel Vehicles
                 Assuming Full Costs for VGT and Improved Fuel Injection

Model Year
Grouping

2004-05
2006-08
2009+
30 Year
Fleet

Total
NPV
Costs per
Vehicle

$896
$766
$466
	

Discounted
Lifetime Reduction
(tons)

NOx
1.365
	


NMHC
0.074
	

Discounted
Per-Vehicle
Cost-
Effectiveness
With Full Costs
($/ton)

$623
$532
$324
$396

Discounted
Per-Vehicle
Cost-
Effectiveness
With Partial
Costs
(From Table 8-4)
$474
$410
$238
$296

                                                                                   146

-------
                                                      Chapter 8:  Cost-Effectiveness
V.   Comparison  of Cost-Effectiveness  with Other  Mobile  Source NOx Control
       Strategies

       In an effort to evaluate the cost-effectiveness of the new standards, EPA has summarized the
cost-effectiveness results for other recent EPA mobile source rulemakings that required reductions
in NOx emissions, the primary focus of the new standards. Both EPA and states have previously
adopted numerous NOx emission control measures, and remaining measures may be more expensive
than those previously employed. As additional control measures are applied, more expensive ones
may be necessary over time.  Table 8-14 summarizes the cost-effectiveness results from the Clean
Fuel Fleet Vehicle Program, Phase II of the Reformulated Gasoline Program, Tier 2 and Tier 3
Standards for Nonroad Diesel Engines, and Standards for Locomotives.

       The projected long-term cost-effectiveness of the diesel vehicles  is $296 per ton of NMHC
and NOx. The projected long-term cost-effectiveness of the Otto-cycle vehicles is $598. The cost-
effectiveness of these standards in this rule falls within the range of these other programs. The cost-
effectiveness of these standards compare favorably with the other programs listed in Table 8-14.
                                      Table 8-14
       Summary of Cost-Effectiveness Results for Recent EPA Mobile Source Programs
EPA Final Rule
Clean Fuel Fleet Vehicle Program
(Heavy-duty)
Reformulated Gasoline — Phase II
Nonroad Diesel Engines — Tiers 2
and 3
Locomotives
Pollutants Considered
in Calculations
NOx
NOx
NMHC+NOx
NOx
Cost-Effectiveness
($/ton)
$1,300-1,500
(1994 dollars)
$5,000
(1990 dollars)
$410-600
(1995 dollars)
$160
(1997 dollars)
                                                                                   147

-------
Regulatory Impact Analysis
Chapter 8 References
1.  "Cost Effectiveness Analyses of Proposed Heavy-Duty Gasoline Engine and Vehicle
Standards," EPA memo from Phil Carlson to Docket A-98-32.

2.  See Chapter 7, Section 1 of "Final Regulatory Impact Analysis: Control of Emissions of Air
Pollution from Highway Heavy-Duty Engines", September, 1997. Available in EPA Air Docket
A-95-27, Docket Item # V-B-01.
                                                                                148

-------
                               APPENDIX A

            ANNUALIZED COST EFFECTIVENESS ANALYSIS
This appendix contains tables that show EPA's estimates of 20-year annualized costs and emission
reductions. These data come from the analyses contained in Chapters 4 through 8.  The analyses
assume a 7 percent discount rate.
                                                                                A-l

-------
Regulatory Impact Analysis


Annual Diesel
NOx and NMHC
Benefits (Tons) and Costs
With Annualized 20-Year Cost-Effectiveness
(All numbers were rounded after calculation)
Year
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023

NOx
(tons)
97,000
266,000
417,000
555,000
681,000
795,000
900,000
999,000
1,093,000
1,180,000
1,260,000
1,336,000
1,407,000
1,474,000
1,537,000
1,597,000
1,654,000
1,709,000
1,762,000
1,813,000

NMHC
(tons)
5,000
14,000
23,000
30,000
37,000
43,000
49,000
55,000
60,000
64,000
69,000
73,000
76,000
80,000
84,000
87,000
90,000
93,000
96,000
99,000
20-Year Total
Total
(tons)
103,000
280,000
440,000
586,000
718,000
838,000
949,000
1,054,000
1,152,000
1,244,000
1,329,000
1,409,000
1,483,000
1,554,000
1,621,000
1,684,000
1,744,000
1,802,000
1,858,000
1,912,000
= 23,760,000
20-Year Annualized NPV =

20-Year Cost-Effectiveness
Discounted 20-Year $/ton
Undiscounted 20-Year $/ton
(NOx+NMHC) =
(NOx+NMHC) =
Discounted Benefits
(tons)
103,000
262,000
384,000
478,000
548,000
597,000
632,000
656,000
670,000
677,000
676,000
669,000
658,000
645,000
629,000
610,000
591,000
570,000
550,000
529,000
11,134,000
1,05 1,000 tons
$360
$283
Total Costs
(Undiscounted)
$479,000,000
$489,000,000
$427,000,000
$436,000,000
$444,000,000
$248,000,000
$253,000,000
$258,000,000
$262,000,000
$267,000,000
$296,000,000
$301,000,000
$306,000,000
$311,000,000
$316,000,000
$320,000,000
$325,000,000
$330,000,000
$335,000,000
$339,000,000
$6,741,000,000
$379,000,000
Discounted
Costs
$479,000,000
$457,000,000
$373,000,000
$355,000,000
$338,000,000
$177,000,000
$169,000,000
$161,000,000
$153,000,000
$145,000,000
$151,000,000
$143,000,000
$136,000,000
$129,000,000
$122,000,000
$116,000,000
$110,000,000
$104,000,000
$99,000,000
$94,000,000
$4,011,000,000

                                                                            A-2

-------
Annual Otto-Cycle NOx and NMHC Benefits (Tons) and Costs
        With Annualized 20-Year Cost-Effectiveness
        (All numbers were rounded after calculation)
Year
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024

NOx
(tons)
13,000
39,000
63,000
86,000
107,000
128,000
146,000
165,000
181,000
197,000
213,000
224,000
235,000
248,000
263,000
278,000
288,000
298,000
308,000
318,000

NMHC
(tons)
3,000
7,000
11,000
15,000
19,000
23,000
26,000
30,000
33,000
36,000
39,000
42,000
44,000
47,000
50,000
52,000
54,000
56,000
58,000
60,000
20-Year Total
Total
(tons)
16,000
46,000
74,000
101,000
126,000
151,000
172,000
195,000
214,000
233,000
252,000
266,000
279,000
295,000
313,000
330,000
342,000
354,000
366,000
378,000
= 4,503,000
20-Year Annualized NPV =

20-Year Cost-Effectiveness
Discounted 20-Year $/ton
Undiscounted 20-Year $/ton
(NOx+NMHC) =
(NOx+NMHC) =
Discounted Benefits
(tons)
16,000
43,000
65,000
82,000
96,000
108,000
115,000
121,000
125,000
127,000
128,000
126,000
124,000
122,000
121,000
120,000
116,000
112,000
108,000
105,000
2,080,000
196,000 tons
$710
$598
Total Costs
(Undiscounted)
$110,000,000
$117,000,000
$124,000,000
$126,000,000
$129,000,000
$124,000,000
$126,000,000
$128,000,000
$131,000,000
$133,000,000
$135,000,000
$137,000,000
$139,000,000
$141,000,000
$144,000,000
$146,000,000
$148,000,000
$150,000,000
$152,000,000
$154,000,000
$2,694,000,000
$139,000,000
Discounted
Costs
$110,000,000
$109,000,000
$109,000,000
$103,000,000
$98,000,000
$89,000,000
$84,000,000
$80,000,000
$76,000,000
$72,000,000
$69,000,000
$65,000,000
$62,000,000
$59,000,000
$56,000,000
$53,000,000
$50,000,000
$47,000,000
$45,000,000
$43,000,000
$1,478,000,000

                                                                                  A-3

-------