Exhaust Emission Rates for Heavy-Duty
Onroad Vehicles in MOVES3
&EPA
United States
Environmental Protection
Agency
-------
Exhaust Emission Rates for Heavy-Duty
This technical report does not necessarily represent final EPA decisions
or positions. It is intended to present technical analysis of issues using
data that are currently available. The purpose in the release of such
reports is to facilitate the exchange of technical information and to
inform the public of technical developments.
Onroad Vehicles in MOVES3
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
NOTICE
4>EPA
United States
Environmental Protection
Agency
EPA-420-R-22-031
November 2022
-------
Table of Contents
1 Introduction 1
1.1 Pollutant 2
1.2 Emission Process 2
1.2.1 Running Exhaust 2
1.2.2 Start Exhaust 3
1.2.3 Extended Idle and Auxiliary Power Exhaust 3
1.2.4 Crankcase Exhaust 4
1.2.5 Evaporative and Brake and Tire Wear Emissions 4
1.3 Fuel Type 5
1.4 Regulatory Class 5
1.5 Model Year Groups 7
1.6 Operating Modes 7
1.7 Vehicle Age 13
1.8 Updates for MOVES3 14
2 Heavy-Duty Diesel Exhaust Emissions 16
2.1 Running Exhaust Emissions 16
2.1.1 Nitrogen Oxides (NOx) 16
2.1.2 Particulate Matter (PM2 5) 55
2.1.3 Total Hydrocarbons (THC) and Carbon Monoxide (CO) 78
2.1.4 Energy 92
2.1.5 Evaluation of Fleet-average Running Rates with Real-World Measurements 104
2.2 Start Exhaust Emissions 105
2.2.1 THC, CO, and NOx 106
2.2.2 Particulate Matter (PM2.5) 114
2.2.3 Adjusting Start Rates for Soak Time 116
2.2.4 Start Energy Rates 127
2.3 Extended Idling Exhaust Emissions 130
2.3.1 1960-2006 Model Years 131
2.3.2 2007-2060 Model Years 134
2.3.3 Model Year Trends 146
iii
-------
2.3.4 Extended Idle Energy Rates 148
2.4 Auxiliary Power Unit Exhaust 149
2.5 Glider Vehicle Emissions 156
3 Heavy-Duty Gasoline Exhaust Emissions 158
3.1 Running Exhaust Emissions 158
3.1.1 THC, COandNOx 158
3.1.2 Particulate Matter (PM2.5) 178
3.1.3 Energy 188
3.2 Start Emissions 193
3.2.1 THC, CO, and NOx 194
3.2.2 Particulate Matter (PM2.5) 212
3.2.3 Soak Time Adjustments 213
3.2.4 Start Energy Rates 214
4 Heavy-Duty Compressed Natural Gas Exhaust Emissions 217
4.1 Running Exhaust Emission Rates 219
4.1.1 1960-2009 Model Years 219
4.1.2 2010-2060 Model Years 232
4.1.3 Model Year Trends 234
4.2 Start Exhaust Emission Rates 239
4.3 Ammonia Emissions 239
5 Heavy-Duty Crankcase Exhaust Emissions 240
5.1 Background on Heavy-Duty Diesel Crankcase Emissions 240
5.2 Modeling Crankcase Emissions in MOVES 241
5.3 Heavy-Duty Diesel Crankcase Emissions 242
5.3.1 1960-2007 Model Years 243
5.3.2 2007-2009 Model Years 245
5.3.3 2010-2060 Model Years 253
5.3.4 Glider Crankcase Emissions 264
5.4 Heavy-Duty Gasoline and CNG Crankcase Emissions 265
6 Nitrogen Oxide Composition 267
6.1 Heavy-Duty Diesel 268
6.2 Heavy-Duty Gasoline 269
iv
-------
6.3 Compressed Natural Gas 269
7 Appendices 271
Appendix A Calculation of Accessory Power Requirements 272
Appendix B Tampering and Mal-maintenance for Diesel Running Exhaust 273
Appendix C Tampering and Mal-maintenance for MY 2007 and Later Diesel Extended Idle 292
Appendix D Pre-2007 Model Year Extended Idle Data Summary 295
Appendix E Developing Pre-2007 Model Year HD Diesel PM2.5 Emission Rates for Missing Operating
Modes 300
Appendix F Heavy-Duty Gasoline Start Emissions Analysis Figures 302
Appendix G Selection of Fixed Mass Factor (fSCaie) values for MY 2010+ Heavy-Duty Vehicles 307
Appendix H THC and CO Emisssion rates from 2010 and Later Model Year Heavy-duty Vehicles from
the HDIUT 319
Appendix I Analysis of 2010 and Later Model Year Heavy-duty Gasoline Emission Rates 325
Appendix J Comparing Glider Vehicle and MOVES Model Year 2000 Heavy Heavy-Duty Emission Rates
336
Appendix K PM Composition Measurements from Auxiliary Power Units 341
Appendix L Crankcase Emissions from Glider Trucks 342
8 References 345
v
-------
List of Acronyms
ABT
emissions averaging, banking and trading program
A/C
Air Conditioning
ACES
Advanced Collaborative Emission Study (CRC)
APU
auxiliary power units
ARCO
Atlantic Richfield Company
BC
black carbon
bhp
brake horsepower
BTU
British Thermal Unit
CARB
California Air Resources Board
CBD
Central Business District
CFR
Code of Federal Regulations
ch4
methane
CNG
Compressed Natural Gas
CO
carbon monoxide
C02
carbon dioxide
CRC
Coordinating Research Council
CTI
Cleaner Trucks Initiative
DB
database
DOC
diesel oxidation catalysts
DOE
U.S. Department of Energy
DPF
diesel particulate filter/periodic trap oxidizer
EC
elemental carbon
ECOSTAR
gaseous and exhaust flow measurement system
ECU
Engine Control Unit
EFEE
Engine, Fuel and Emissions Engineering Inc.
EGR
exhaust-gas recirculation
-------
EMFAC CARB emissions factors model
EPA U.S. Environmental Protection Agency
ESC European Stationary Cycle
FEL family emission limit
FHWA Federal Highway Administration
FID Flame Ionization Detection
FTP Federal Test Procedure
g Grams
GDI Gasoline Direct injection engines
GHG Greenhouse Gases
g/mi Grams per mile
GPS Global Positioning System
GVWR Gross Vehicle Weight Rating
THC Hydrocarbons
HD Heavy-Duty
HDIU Heavy-Duty Diesel In-Use
HDT Heavy-Duty Truck
HFC Hydrofluorocarbon
H-GAC Houston-Galveston Area Council
HHD Heavy-Heavy-Duty Class 8 Trucks (GVWR > 33,000 lbs)
HHDD Heavy Heavy-Duty Diesel
HN02 nitrous acid (HONO)
HP horsepower
hr hour
HV heating value
H20 water
l/M Inspection and Maintenance program
IUVP In-Use Verification Program
kJ Kilojoules
vii
-------
kW Kilowatt
LHD Light-Heavy-Duty
LHD2b3 Light-Heavy-Duty Class 2b and 3 Truck (8,500 < GVWR < 14,000 lbs)
LHD45 Light Heavy-Duty Class 4 or 5 Truck (14,000 < GVWR < 19,500 lbs)
LHDDT Light Heavy-Duty Diesel Truck
LNT Lean NOx Trap aftertreatment
MDPV Medium-Duty Passenger Vehicle
MECA The Manufacturers of Emission Controls Association
MEMS Mobile Emissions Measurement System
mg milligram
MHD Medium-Heavy-Duty Class 6 and 7 Trucks (19,500 < GVWR < 33,000 lbs)
MOBILE6 EPA Highway Vehicle Emission Factor Model, Version 6
MOVES Motor Vehicle Emission Simulator Model
MOVES201X Motor Vehicle Emission Simulator Model development version
MSOD Mobile Source Observation Database
MY model year
MYG model year group
NCHRP National Cooperative Highway Research Program
NCP nonconformance penalty
NDIR non-dispersive infrared
NFRAQS Northern Front Range Air Quality Study
NH3 ammonia
NMHC Non-Methane Hydrocarbon
NMOG non-methane organic gases
NonEC non-elemental carbon
NonECnonS04PM non-elemental carbon non-sulfate particulate matter
NonECPM non-elemental particulate matter carbon
NO nitric oxide
NOx nitrogen oxide
viii
-------
NOy combined NOx and NOz compound
NOz nitrous oxide
N02 nitrogen dioxide
NREL National Renewal Energy Laboratory
NTE Not-to-Exceed
NYSDEC New York Department of Environmental Conservation
N20 nitrous oxide
OBD On-Board Diagnostics
OC oxidation catalyst
OEM Original Equipment Manufacturer
OM operating mode
OMNMHCE organic material non-methane hydrocarbon equivalent
PCV positive crankcase ventilation
PEMS portable emissions measurement system
PERE Physical Emission Rate Estimator
PHA Port of Houston Authority
PM Particulate Matter
PM2.5 fine particles of particulate matter with aerodynamic diameters < 2.5 pim
PM10 particles of particulate matter with aerodynamic diameters < 10 pim
ROVER EPA dataset measurement collection system
RPM revolutions per minute
SCAQMD South Coast Air Quality Management District
SCR selective catalytic reduction
S04 sulfate
STP scaled tractive power
ST01 258-second driving cycle
T&M Tampering and Maintenance
TC total carbon
TEOM Tapered Element Oscillating Microbalance
ix
-------
THC
Total Hydrocarbon (FID detection)
TOG
Total Organic Gases
TTI
Texas Transportation Institute
TWC
three-way catalysts
UDDS
Urban Dynamometer Driving Schedule
UL
useful life
ULSD
Ultra Low Sulfur Diesel
VIN
Vehicle Identification Number
VIUS
Vehicle Inventory and Use Survey
VMT
Vehicle Miles Traveled
VOC
Volatile Organic Compounds
VSP
vehicle specific power
WMATA
Washington Metropolitan Area Transit Authority
WVU
West Virginia University
ZML
zero-mile emissions level
X
-------
1 Introduction
The United States Environmental Protection Agency's Motor Vehicle Emission Simulator (MOVES) is a set
of modeling tools for estimating air pollution emissions produced by onroad (highway) and nonroad mobile
sources. MOVES estimates the emissions of greenhouse gases (GHGs), criteria pollutants and selected air
toxics. The MOVES model is currently the official model for use for state implementation plan (SIP)
submissions to EPA and for transportation conformity analyses outside of California. The model is also the
primary modeling tool for estimating the impact of mobile source regulations on emission inventories.
This report describes the analyses conducted to generate exhaust emission rates and energy rates
representing exhaust emissions and energy consumption for heavy-duty vehicles in MOVES3. Heavy-duty
vehicles in MOVES are defined as any vehicle with a Gross Vehicle Weight Rating (GVWR) above 8,500 lbs.
Emission rates for criteria pollutants (THC, CO, NOx, and PM2.5) are stored in the "EmissionRateByAge"
table in the MOVES database according to the following:
• Pollutant
• Emission process
• Fuel type
• Regulatory class
• Model year group
• Operating mode
• Vehicle age
Energy emission rates are stored in the "EmissionRate" table, which is similar to the "EmissionRateByAge"
table, except emission rates are not differentiated by vehicle age. The MOVES framework and additional
details regarding the "EmissionRateByAge" and "EmissionRate" table are discussed in the report
documenting the rates for light-duty vehicles.8
In Section 1, we provide more background on the factors used to estimate heavy-duty exhaust emissions in
the "EmissionRateByAge" and "EmissionRate" tables. We then discuss the major updates made to the
heavy-duty emissions in MOVES3.
Sections 2 through 4 document the tailpipe exhaust emission rates for heavy-duty diesel, heavy-duty
gasoline, and heavy-duty CNG vehicles. Section 5 documents the crankcase emission rates used for each
fuel type of heavy-duty vehicles. Section 6 documents the methods used to estimate nitric oxide (NO),
nitrogen dioxide (N02), and nitrous acid (HN02 or HONO) emissions from NOx emissions using ratios.
1
-------
1.1 Pollutant
This report discusses the development of tailpipe exhaust emission rates for total hydrocarbons (THC),
carbon monoxide (CO), nitrogen oxides (NOx), fine particulate matter, defined as particulate matter with
mean aerodynamic diameter less than 2.5 microns (PM2.5), and energy consumption (in units of kJ).
Total hydrocarbons (THC) is the measurement of hydrocarbons from a flame ionization detector.1 From
THC emissions, MOVES generates other estimates of hydrocarbon and organic gas missions, including
volatile organic compounds (VOCs), methane (CH4) and total organic gases (TOG). MOVES then uses VOC
emission rates to estimate individual toxic compounds such as formaldehyde and benzene. The derivation
of the factors used to compute aggregate measures of organic gases and individual toxic emissions are
available in the Speciation1 and Toxics2 MOVES Reports.
MOVES reports PM2.5 emissions in terms of elemental carbon (EC) and the remaining non-elemental carbon
PM (nonECPM). This heavy-duty report covers the derivation of EC/PM fractions used to estimate
elemental carbon (EC), and the remaining non-elemental carbon PM (nonECPM). MOVES also estimates 18
PM subspecies beyond elemental carbon, including organic carbon, sulfate, nitrate, and other trace
elements and ions through the use of speciation profiles as documented in the Speciation Report.1
From NOx, MOVES estimates NO, N02, and HONO emissions as documented in Section 6.
Nitrous Oxide (N20) emissions from heavy-duty vehicles remained unchanged from MOVES2009 and later
versions and are documented in the MOVES3 Greenhouse Gas and Energy Report.3 MOVES estimates C02
emissions from the energy rates documented in this report, using conversion factors documented in the
Greenhouse Gas and Energy Report. Ammonia (NH3) emission rates for heavy-duty vehicles were
developed for MOVES2010 and are documented in a separate report,4 except CNG rates which are
documented here in Section 4.3.
1.2 Emission Process
MOVES models vehicle emissions from fourteen different emission processes as listed in Table 1-1. This
report covers the emission rates for the exhaust emission processes (running exhaust, start exhaust,
extended idle exhaust, auxiliary power exhaust, crankcase running exhaust, crankcase start exhaust, and
crankcase extended idle exhaust). We discuss the different processes below:
1.2.1 Running Exhaust
The running exhaust process occurs as the vehicle is operating on the road either under load or in idle
mode. MOVES3 includes off-network idling as a new source of vehicle activity, but it is not a new emission
process. Off-network idling is classified as idle mode activity within the running emission process. The
running process is further delineated by 23 operating modes as discussed in Section 1.6.
2
-------
1.2.2 Start Exhaust
The start exhaust process is the incremental emissions that occur when starting a vehicle, including the
incremental emissions that occur after the engine start before the aftertreatment system is fully
functional. Ideally, start emission rates are calculated as the difference in emissions measured between
two otherwise identical drive cycles, where the first cycle includes a start and in the second drive cycle the
vehicle is already running and in a warmed-up condition as shown in Equation 1-1.
g
Start emission rate ( )
start
Equation 1-1
= (emissions from drive cycle; with start
— emissions from same drive cycle; without start and with vehicle in warm condition)
Starts also have operating modes to characterize different amount of soak time (time since the vehicle has
last been running before being started again). Cold starts, (or starts after a long soak period) generally have
higher emission rates than warm starts (starts after a short soak period), due to additional fueling needed
due to increased condensation of fuel at colder engine temperatures, and because the catalytic
aftertreatment needs to reach a warm temperature to be fully operational. MOVES defines eight operating
start operating modes based on soak time as discussed in Section 1.6.
Operationally, we typically don't have two identical drive cycles that fit the conditions of Equation 1-1, and
cold starts in this report are calculated using Equation 1-2. Equation 1-2 is a reasonable approximation
since cold start emissions are typically much higher than hot-start emission.
g
Operationally — defined cold start emission rate ( ) =
stcirt
Equation 1-2
= (emissions from drive cycle; with cold start
— emissions from drive cycle; with hot start and vehicle in warm condition)
1.2.3 Extended Idle and Auxiliary Power Exhaust
The extended idle exhaust process in MOVES occurs during periods of hotelling, when long-haul trucks are
used during rest periods, such as when a vehicle is parked for the night and left idling. Extended idle in
generally defined to cover idling periods for longer than one hour. Extended idle can result in different
emissions than incidental idle that occurs during running operation because the engine may be operated at
a higher engine speed and the exhaust aftertreatment system may be too cool to operate at its full
efficiency.
3
-------
Auxiliary power exhaust are emissions that come from diesel-powered generators that power the truck's
accessory loads, sometimes are used in place of the main engine during periods of hotelling.
Documentation of the extended idle and auxiliary power exhaust emissions for heavy-duty diesel trucks
are in Section 2.3.
1.2.4 Crankcase Exhaust
Crankcase exhaust emissions (for running, start, and extended idle) include combustion products and oil
that are vented from the engine crankcase to the atmosphere. Crankcase emissions are estimated for THC,
CO, NOx, PM2.5 emissions and their chained pollutants as discussed in Section 5. We do not estimate energy
or C02 emissions because crankcase emissions are a small contribution to the total C02 emissions.
Crankcase emissions are significant sources of THC and PM2.5 emissions from heavy-duty diesel engines.
The emission rates for all four pollutants for all heavy-duty source types and fuels are discussed in Section
5.
1.2.5 Evaporative and Brake and Tire Wear Emissions
Estimation of evaporative hydrocarbon emissions from heavy-duty gasoline vehicles is described in the
evaporative report.5 MOVES does not estimate evaporative emissions for diesel-powered vehicles, but
does estimate fuel spillage emissions which are part of the refueling emissions documented in the
evaporative report.5 Brake and tire wear emission rates from heavy-duty vehicles are discussed in the
Brake and Tire Wear Report.10
Table 1-1 Emission Processes for Onroad Heavy-Duty Vehicles
processID
processName
Covered
in this
report?
1
Running Exhaust
Y
2
Start Exhaust
Y
9
Brakewear
N
10
Tirewear
N
11
Evap Permeation
N
12
Evap Fuel Vapor Venting
N
13
Evap Fuel Leaks
N
15
Crankcase Running Exhaust
Y
16
Crankcase Start Exhaust
Y
17
Crankcase Extended Idle Exhaust
Y
18
Refueling Displacement Vapor
Loss
N
19
Refueling Spillage Loss
N
90
Extended Idle Exhaust
Y
91
Auxiliary Power Exhaust
Y
4
-------
1.3 Fuel Type
This report is primarily organized around the exhaust emission rates by fuel type: heavy-duty diesel
(Section 2), heavy-duty gasoline (Section 3), and heavy-duty CNG (Section 4). These three fuel types can be
modeled in all the heavy-duty sourcetypes, with the exception of long-haul combination trucks which can
only model diesel vehicles. Note that the emissions from the heavy-duty sector predominantly come from
diesel vehicles and the majority of the data analyzed were from diesel vehicles. MOVES also models E85
and electricity fuel types for light-duty vehicles, but these fuel types are not available to be modeled for
heavy-duty vehicles.9
1.4 Regulatory Class
The MOVES regulatory classes group vehicles that have similar emission rates. The MOVES heavy-duty
regulatory classes are largely determined based on gross vehicle weight rating (GVWR) classifications,
because the heavy-duty emission standards are based on GVWR as shown in Table 1-2.
There are additional criteria that define two of the heavy-duty regulatory classes in MOVES. Urban Bus
vehicles are distinguished from other heavy heavy-duty vehicles (GVWR >33,000 lbs.) because they have
tighter PM emission standards for the 1994 through 2006 model years.6 Urban bus is a regulatory class
that is defined by its intended use as well as the GVWR. EPA regulations define urban buses as "heavy
heavy-duty diesel-powered passenger-carrying vehicles with a load capacity of fifteen or more passengers
and intended primarily for intra-city operation."7.
In MOVES, gliders (regClassID 49) are defined as heavy heavy-duty diesel trucks with an old powertrain,
combined with a new chassis and cab assembly. Currently in MOVES, gliders are limited to the diesel long-
haul and combination short-haul source types. As discussed in Section 2.5, the emissions are equivalent to
MY 2000 HHD diesel vehicles.
MOVES classifies vehicles of similar activity, usage patterns, and body type into source use types, often
simply referred to as "source types." The MOVES source types are defined in the Population and Activity
Report.9 As shown in Table 1-2, vehicles of a regulatory class may be mapped to multiple source types.
Likewise, each source type used to model heavy-duty vehicles includes population from several different
regulatory classes. For example, single unit short-haul trucks (sourceTypelD 52), include vehicles from the
following regulatory classes: LHD2b3, LHD45, MHD, and HHD.
5
-------
Table 1-2 Regulatory Classes for Heavy-Duty Vehicles
Regulatory Class
Description
regClassName
regClassID
Gross Vehicle
Weight Rating
(GVWR) [lb.]
Possible Source Types
(sourceTypelD)
Light Heavy-Duty
Class 2b and 3
trucks
LHD2b3a
41
8,501-
14,000
Passenger Trucks (31), and
Light-Commercial Trucks (32),
Buses (41, 43), and Single Unit
Trucks (51, 52, 53, 54)
Light Heavy-Duty
Class 4 and 5
Trucks
LHD45
42
14,001 -
19,500
Buses (41, 42, 43) and Single
Unit Trucks (51, 52, 53, 54)
Medium Heavy-
Duty (Class 6 and
7 Trucks)
MHD
46
19,501-
33,000
Buses (41,42,43), Single Unit
Trucks (51, 52, 53, 54), and
Combination Trucks (61, 62)
Heavy Heavy-Duty
(Class 8 Trucks)
HHD
47
>33,000
Buses (41, 42, 43), Single Unit
Trucks (51, 52, 53, 54), and
Combination Trucks (61, 62)
Urban Bus
Urban Bus7
48
> 33,000
Transit Bus (42)
Gliders (Class 8
Trucks)
Glider Vehicles
49
> 33,000
Combination Trucks (61, 62)
a In M0VES3, we consolidated the MOVES2014 regulatory classes LHD<10K (regClassID 40) and LHD<14K (regClassID
41) into LHD2b3 (regClassID 41). In MOVES2014, LHD<10K was used to model Class 2b trucks among the light-duty
source types (31,32), and LHD<14K was used to model Class 2b and 3 trucks among the heavy-duty source types
(41,42 ,43, 51,52,53, 54). In MOVES2014, the emission rates in LHD<10K and LHD<14K were intended to be the
same. However, separate regulatory classes were needed because MOVES2014 assigned the/Sca/e- power scaling
factor by sourcetype only, so the fSCaie values (stored in the sourcellseTypePhysics table) had to differ between the 2b
trucks mapped to the light-duty sourcetypes (passenger and light-commercial trucks) and the 2b3 trucks among the
heavy-duty sourcetypes (single unit trucks, refuse, motorhomes and buses). Because they had different/Sca/es values,
they needed different emission rates. However, in MOVES3, we can now specify different road-load coefficients,
vehicle weights and/sCa/e values by both source type and regulatory class. Thus, we can model LHD2b3 in both the
light-duty (e.g., Source type 31, passenger trucks) and heavy-duty sourcetypes (e.g., Sourcetype 52, single-unit short-
haul trucks) using the same/sca/e value and the same emission rates, such that the LHD<10K regulatory class is no
longer needed. The vehicles that were classified as LHD<10K in MOVES2014 are now classified as LHD2b3 in MOVES3.
6
-------
1.5 Model Year Groups
MOVES model year groupings are designed to represent major changes in emission rates due to changing
vehicle and aftertreatment technologies introduced in response to EPA emission standards. Model year
groups in MOVES can represent a single model year (e.g., 2007), or a range of model years (e.g., 2030-
2060). The emission rates discussed in the following sections are discussed in terms of model year ranges.
The model year groups cover all the model years between 1960 through 2060. When data are limited or
unavailable for model year groups, we make assumptions about the impact of emission standards and
vehicle aging to estimate those emission rates.
Emission rates in MOVES are stored by regulatory class, fuel type, model year, and operating mode. To
calculate emissions from each process, MOVES sums the product of the emission rate for each operating
mode by the time spent in each operating mode.
For example, the activity basis for running process is source hours operating (SHO). The running process is
divided into 23 operating modes (as shown in Table 1-4). Using Equation 1-3, the total running emissions is
calculated by summing the product of the emission rates with the fraction of time spent in each operating
mode (the operating mode distribution). This is multiplied by the total hours (SHO) spent in this emission
process.
Running Emissions
MOVES performs the calculations shown in Equation 1-3, at a detailed level that accounts for each factor
that impacts the emission rates (e.g., model year, vehicle age, fuel type, regulatory class) and the operating
mode distribution (source type, roadtype, average speed (which varies across hour of the day)). Then, the
emissions can be aggregated to different levels (e.g., by sourcetype). Similar equations can be constructed
for other process, the equation for starts is shown in Equation 1-4, where the starts are classified into eight
operating modes (Table 1-5).
Start Emissions
1.6 Operating Modes
= SHO
Equation 1-3
i=1
= Starts
8
Equation 1-4
x (Operating Mode Distributioni x Emission Ratei)
i=1
7
-------
The operating modes for running exhaust are defined in terms of power output (with the exception of the
idle and braking modes). For light-duty vehicles, the parameter used is known as vehicle-specific power
(VSP), which is calculated by normalizing the continuous power output for each vehicle to its own weight
as shown in Equation 1-5 (hence the term "vehicle-specific"). As discussed in the light-duty emission rate
report,8 VSP is a robust predictor of vehicle emissions. In the laboratory, light-duty vehicles are tested on
full chassis dynamometers, and emission standards are in units of grams per mile. The emission standards
are largely independent of the weight (and other physical characteristics) of the vehicle.
In developing emission rates for MOVES, light-duty emissions data from individual vehicles are assigned to
VSP operating mode bins using Equation 1-5, using the individual vehicle's measured weight as the source
mass, and ideally using vehicle-specific road load coefficients. In contrast, when MOVES calculates VSP
from driving cycles and assigns operating modes for an entire source type, the average source type mass
and average road load coefficients are used instead.
Avt + Bv} + Cvf + m ¦ vt(at + a ¦ sindt) ^ ^ . r
VSPt = — ———- — Equation 1-5
m
Where:
VSPt = vehicle specific power at time t [kW/ton]
A = the rolling resistance coefficient [kW-sec/m],
B = the rotational resistance coefficient [kW-sec2/m2],
C = the aerodynamic drag coefficient [kW-sec3/m3],
m = mass of individual test vehicle [metric ton],
vt = instantaneous vehicle velocity at time t [m/s],
ot= instantaneous vehicle acceleration [m/s2]
g = the acceleration due to gravity [9.8 m/s2]
sin 0t = the (fractional) road grade at time t
For heavy-duty vehicles, we classify running exhaust using Scaled Tractive Power (STP) as shown in
Equation 1-6 using road-load coefficients. STP is equivalent to VSP, except the power for all vehicles within
the same regulatory class and model year are scaled using a fixed mass factor, rather than the individual
weight of the vehicle. The fscaie is used to bring the numerical range of tractive power from heavy-duty
vehicles into the same numerical range as the VSP values when assigning operating modes. When
developing emission rates for MOVES, operating modes are assigned to individual vehicles using both the
individual truck mass, m, and the common fscaie value used for all heavy-duty vehicles from the same
regulatory class, source type and model year group. Because a common fscaievalue is used, individual
8
-------
vehicles assigned to the same STP-defined operating mode bin are producing the same absolute tractive
power, regardless of differences in their individual source masses.
Avt + Bvf + Cv} + m ¦ vt(at + q ¦ sindt)
STPt = —- - - ——- - Equation 1-6
fscale
Where:
5TPt = the scaled tractive power at time t [scaled kW or skW]
fscaie = fixed mass factor (see Table 1-3)
Other variables as previously defined in Equation 1-5
When MOVES estimates STP and assigns operating mode distributions for the heavy-duty source
types, Equation 1-6 uses the average source type mass (m) for each regulatory class, source type,
and model year group in the numerator and uses the common fscaie value for the regulatory class
and model year group which was also used in the emission rate analysis. At County and Default
National Scale, MOVES uses the instantaneous speed (vt) from the second-by-second driving
cycles associated with average speed, source types, and road types. For Project Scale, MOVES can
use user-supplied driving cycles and grade. The default average speed driving cycles and the load
road coefficients are discussed in the Population and Activity Report.9
The equation for STP is generalized below in Equation 1-7, with units in scaled kW or skW:
STP = —Equation 1-7
fscaie
Where: Paxie is the power demand at the axle for the heavy-duty truck.
As presented in Equation 1-6, Paxie can be estimated using the road-load coefficients from chassis-
dynamometer tests (for example, pre-2010 heavy-duty diesel THC, CO, and PM2.5 emission rates
in Sections 2.1.2 and 2.1.3). Road-load coefficients can also be used to estimate power demand
from onroad tests when more accurate power demand measurements are not available (for example,
the 2010+ heavy-duty gasoline emission rates in Section 3.1).
The Paxie can be estimated from an engine dynamometer or from an engine control unit (ECU) for chassis or
onroad testing by measuring the engine power and estimating the accessory loads and powertrain
efficiencies for the vehicle.
For onroad tests, measuring power from the ECU is generally more accurate than estimating power from
road-load coefficients. Unlike a generic road-load equation where vehicle characteristics, such as
9
-------
aerodynamic drag and rolling resistance are assumed, the ECU measures engine speed and calculates
torque directly during the test, avoiding the need to capture the impact of wind speed and wind direction,
as well as weight and the road grade throughout the entire test cycle and route. Wind can have a
significant impact on power needs, and the payload of heavy-duty vehicles can be greater than the vehicle
weights itself, while also varying significantly over the test. Thus, for onroad tests, we generally use power
calculated from the ECU measurements, because the vehicle and environmental characteristics determine
the axle power (Section 2.1.1.3).
The use of STP instead of VSP is preferable for modeling heavy-duty vehicles emissions because heavy-duty
vehicle emissions are strongly correlated with power output. Heavy-duty vehicles are regulated using
engine dynamometer tests, and emissions standards are in units of grams per brake-horsepower-hour
(g/bhp-hr). Additionally, each heavy-duty regulatory class contains a wide variety of truck sizes, truck
weight, power ratings, and in-use payloads. Using STP, we can scale the heavy-duty emission rates to
different power outputs that were not measured in our emission rate database. The sample of trucks we
used to develop emission rates for each regulatory class has a limited number of trucks and loads
compared to the in-use fleet, which may not be representative of the average vehicle weight and power
output of the in-use fleet. The use of VSP would require the sample of vehicles to match the average
vehicle weights and load to accurately estimate average in-use emission rates. By using STP, MOVES scales
the measured emissions to match the estimated weight and power output of the modeled in-use fleet.
The fscaie values can be considered as a surrogate for the average mass of heavy-duty vehicles
within each regulatory class, as reported in Table 1-3. For the pre-2010 emission rates, we used an
fscaie equivalent to the average mass of the light-commercial trucks (2.06 metric tons) for the
LHD2b3 and LHD45 emission rates. For the other heavy-duty source types, a single fscaie (17.1
metric tons) was used, which provided emissions within each operating mode bin for the largest
heavy-duty truck vehicles.
However, in our MOVES3 analysis of in-use data from recent trucks, we found that an fscaie of 17.1
metric tons limited most of the real-world activity to low and medium power bins within a speed-
bin, especially for the light-heavy and medium-heavy-duty regulatory classes. In MOVES3, we
revised the fscaie values for MY 2010 and later to provide more resolution in the fscaie by regulatory
class. Derivation of the new fscaie values is described in Appendix G.
10
-------
Table 1-3 Fixed Mass Factor, fscaie (metric tons)
Regulatory Class (regClassID)
MY 1960-2009
MY 2010+
LHD2b3 (41)
2.06
5.0
LHD45 (42)
2.06
5.0
MHD (46)
17.1
7.0
HHD (47)
17.1
10.0
Urban Bus (48)
17.1
10.0
Glider (49)
17.1
17.1
NOTE: OpMode-based emission rates CANNOT be compared directly between regClasses or model years (MYs) with
different fscaie values. For example, the OpMode 14 emission rates for MY 2012 MHD (fscaie = 7) cannot be directly
compared to the same OpMode rates from MY 2009 MHD (,fscaie = 17.1) or MY 2012 HHD (fscaie =10). That is because
data assigned to an OpMode based on different fscaie values will have different absolute power (numerator of Equation
1-6 and Equation 1-7). When using vehicle mass in the denominator (Equation 1-5), this is not an issue because the unit
is kW/ton and the power is normalized to the mass of the vehicle. However, when using fscaie in the denominator, as is
the case for all heavy-duty vehicles, the unit is scaled kW and there is no normalization to a physical quantity.
The operating modes bins for running exhaust are shown in Table 1-4.
11
-------
Table 1-4 Operating Mode Definition for Running Exhaust for Heavy-Duty Vehicles
OpModelD
Operating Mode
Description
Scaled Tractive
Power (STP,, skW)
Vehicle Speed
(vt, mph)
Vehicle Acceleration
including grade
(mph/sec)1
0
Deceleration/Braking2
at +g-sin(6t) < -2.0 OR
[at +g-sin(6t) < -1.0
AND
at-1 +g-sin(6t-i) < -1.0
AND
at-2 +g-sin(dt-2) < -1.0)
1
Idle
vt< 1.0
11
Coast
STP,< 0
1 < vt< 25
12
Cruise/Acceleration
0 < STP,< 3
1 < vt< 25
13
Cruise/Acceleration
3 < STP,< 6
1 < vt< 25
14
Cruise/Acceleration
6 < STP,< 9
1 < vt< 25
15
Cruise/Acceleration
9 < STP(< 12
1 < vt< 25
16
Cruise/Acceleration
12 < STP,
1 < vt< 25
21
Coast
STP(< 0
25 < vt< 50
22
Cruise/Acceleration
0 < STP,< 3
25 < vt< 50
23
Cruise/Acceleration
3 < STP,< 6
25 < vt< 50
24
Cruise/Acceleration
6 < STP,< 9
25 < vt< 50
25
Cruise/Acceleration
9 < STP(< 12
25 < vt< 50
27
Cruise/Acceleration
12
-------
Table 1-5 Operating Modes for Start Emissions (as a function of soak time)
Operating Mode
Description
101
Soak Time < 6 minutes
102
6 minutes < Soak Time < 30 minutes
103
30 minutes < Soak Time < 60 minutes
104
60 minutes < Soak Time < 90 minutes
105
90 minutes < Soak Time <120 minutes
106
120 minutes < Soak Time < 360 minutes
107
360 minutes < Soak Time < 720 minutes
108
720 minutes < Soak Time
Extended idle exhaust and diesel auxiliary power unit (APU) exhaust are each modeled in MOVES
with a single operating mode (opModelDs 200 and 201, respectively).
1.7 Vehicle Age
In MOVES, the start and running emission rates for THC, CO, NOx, and PM2.5 are stored in the
"emissionRateByAge" table by age group, meaning that different emission rates can be assigned to
different aged vehicles of the same model year, regulatory class, fuel type and operating mode. MOVES
uses six different age classes to model the age effects, as shown in Table 1-6. The effects of age on the
emission rates were developed separately for gasoline and diesel vehicles.
Table 1-6 MOVES Age Group Definitions
Lower
Upper
ageGroupID
bound
bound
(years)
(years)
3
0
3
405
4
5
607
6
7
809
8
9
1014
10
14
1519
15
19
2099
20
None
For diesel running exhaust, we estimated the effects of tampering and mal-maintenance (T&M) on
emission rates as a function of age. Tampering refers to intentional disabling or modifying the vehicle
engine, control systems, and/or exhaust aftertreatment systems that results in increased emissions. Mal-
maintenance refers to lack or improper maintenance of the engine and aftertreatment, including
neglecting to repair broken or mal-functioning engine and aftertreatment parts, which increase emissions.
Based on surveys and studies, we developed estimates of frequencies and emission impacts of specific
emission control component malfunctions, and then aggregated them to estimate the overall emissions
13
-------
effects for each pollutant (see Appendix B). We adopted this approach due to the lack of adequate data to
directly estimate the deterioration for heavy-duty vehicles, and because the effects of T&M are believed to
be the dominant source of emission deterioration on fleet-wide heavy-duty diesel emissions. MOVES
currently does not allow explicit modeling of tampered vehicles - we hope to address this in future
versions of MOVES as more data on tampering become available.
Start emissions are generally a small contributor to total exhaust emissions from heavy-duty diesel
vehicles. No T&M effects are currently applied to the diesel start emissions.
For gasoline heavy-duty vehicles, the age effects are estimated directly from the a sample of emissions
data, or are adopted from light-duty deterioration as discussed in Section 3.1.1.1. These effects are applied
to both running and start exhaust emissions.
Energy rates are stored in the "EmissionRate" table, where are not distinguished by age. This table includes
THC, CO, NOx, PM2.5, and ammonia (NH3) emission rates for extended idle and auxiliary power units (APU),
nitrous oxide (N20) rates for start and running emissions, and tire and brake wear emission rates. We
calculate T&M effects for the diesel extended idle emission rates without regard to vehicle age as
discussed in Section 2.3.2.3. This report documents the THC, CO, NOx, and PM2.5 emissions from extended
idle and APU usage, however, the documentation of heavy-duty nitrous oxide3 and ammonia4 and tire and
brake wear10 emission rates are documented in separate MOVES reports.
1.8 Updates for MOVES3
This report documents the heavy-duty emission rates in MOVES3, which was updated from both
MOVES201411 and the MOVES version used to support the Heavy-duty Greenhouse Gas Phase 2
rulemaking12.
In summary, the following emission rates were updated in MOVES3 from MOVES2014:
• Heavy-Duty Diesel Vehicles
o NOx, PM2.5, THC, CO, and running exhaust emission rates for model year (MY) 2010 and later
(Sections 2.1.1.5, 2.1.2.2, and 2.1.3.2)
o Energy rates for MY 2010 and later vehicles, including the impact of the Heavy-Duty
Greenhouse Gas Phase 2 rulemaking (Sections 2.1.4.3 and 2.1.4.3.2)
o Elemental carbon fraction of PM2.5 exhaust emissions from pre-2007 MY (Section 2.1.2.1.8)
o THC, CO, NOx, and PM2.5 start exhaust emission rates for MY 2010 and later (Section 2.2)
o Extended idle and auxiliary power unit (APU) exhaust THC, CO, NOx, and PM2.5 emission rates
for all model years (Section 2.3 and 2.4)
o Running, start, extended idle, and APU exhaust emissions from glider trucks (Section 2.5)
14
-------
o Updated crankcase emissions for 2007 and later vehicles (Sections 5.3.2, 5.3.3, and 5.3.4)
• Heavy-Duty Gasoline Vehicles
o THC, CO, NOx, and PM2.5 running exhaust emission rates for MY 2010 and later (Sections
3.1.1.2.1 and 3.1.2.2)
o Energy rates for MY 2010 and later, including the impact of the Heavy-Duty Greenhouse Gas
Phase 2 rulemaking (sections 3.1.3.2 through 3.1.3.2.2)
• Heavy-Duty Compressed Natural Gas (CNG) Vehicles (Section 4)
o THC, CO, NOx, and PM2.5 running exhaust emission rates for MY 2007 and later
o Emissions for vehicles using CNG fuel are now calculated for most heavy-duty source types, not
just transit buses
The significant updates made to MOVES3 since the MOVES2014 release were peer-reviewed under EPA's
peer review guidance.13 In 2017, we peer-reivewed an internal version referred to as "MOVES201X" that
included the updates to diesel running and extended idle emission rates and updated energy consumption
rates due to the incorporation of the Heavy-duty Greenhouse Gas Phase 2 rulemaking.14 In 2019, we
conducted a peer-review of the updates made for the preliminary Cleaner Trucks Initiative (CTI) analysis,
which included further updates to the heavy-duty diesel emission rates, incorporation of glider vehicles,
and updated heavy-duty gasoline and CNG emission rates.15 In 2020, we conducted a peer-review of the
updated diesel crankcase emission rates incorporated into MOVES3.16 Materials from each peer review,
including peer-review comments and EPA responses are located on the EPA's science inventory webpage.
15
-------
2 Heavy-Duty Diesel Exhaust Emissions
This section details our analysis to develop emission rates for heavy-duty diesel vehicles. Four emission
processes (running, extended idling, starts, and auxiliary power unit exhaust) are discussed.
2.1 Running Exhaust Emissions
The analysis for running exhaust emissions requires accurate second-by-second measurements of emission
rates and parameters that can be used to estimate the tractive power exerted by a vehicle. This section
describes how we analyzed continuous second-by-second heavy-duty diesel emissions data to develop
emission rates applied within the predefined set of operating modes (Table 1-4). Stratification of the data
sample and generation of the final MOVES emission factors was done according to the combination of
regulatory class (shown in Table 1-2) and model year group. As mentioned in Section 1.6, the emission
rates were developed using scaled-tractive power (STP), using the power scaling factors shown in Table
1-3.
2.1.1 Nitrogen Oxides (NOx)
For NOx rates, we stratified heavy-duty vehicles into the model year groups listed in Table 2-1. These
groups were defined based on changes in NOx emissions standards and the outcome of the Heavy-Duty
Diesel Consent Decree17, which required additional control of NOx emissions during highway driving for
model years 1999-and-later. This measure is referred to as the "Not-to-Exceed" (NTE) limit.
16
-------
Table 2-1 Model Year Groups for NOx Analysis Based on Emissions Standards
Model year group
Standard (g/bhp-
hr)
NTE limit (g/bhp-hr)
Pre-1988
None
None
1988-1989
10.7
None
1990
6.0
None
1991-1997
5.0
None
1998
4.0
None
1999-2002
4.0
7.0 HHD; 5.0 other reg. classes
2003-2006
2.41
1.5 times the standard or family emission limit
(FEL)
(or 1.25 standard or FEL, when FEL > 1.50 g/bhp-
hr)
2007-2009
1.21,2
2010+
0.2
Notes:
1 NMHC+NOx Standard
2 Assumes Phase-in of NOx standard
2.1.1.1 Data Sources
In modeling NOx emissions from HHD, MHD, LHD, and urban buses, we relied on the following data
sources:
ROVER. This dataset includes measurements collected during onroad operation using the ROVER system, a
portable emissions measurement system (PEMS) developed by the EPA. The measurements were
conducted by the U.S. Army Aberdeen Test Center on behalf of U.S. EPA.18 This program started in October
2000. Due to time constraints and data quality issues, we used only data collected from October 2003
through September 2007. The data was compiled and reformatted for MOVES analysis by Sierra
Research.19 EPA analyzed the data and developed the emission rates. The data we used represents
approximately 1,400 hours of operation by 124 trucks and buses of model years 1999 through 2007. The
vehicles were driven mainly over two routes:
• "Marathon" from Aberdeen, Maryland, to Colorado and back along Interstate 70
• Loop around Aberdeen Proving Grounds in Maryland
Consent Decree Testing. These data were conducted by West Virginia University using the Mobile
Emissions Measurement System (MEMS).20,21,22 This program was initiated as a result of the consent
17
-------
decree between the several heavy-duty engine manufacturers and the US government, requiring the
manufacturers to test in-use trucks over the road. Data was collected from 2001 through 2006. The data
we used represented approximately 1,100 hours of operation by 188 trucks of model years 1994 through
2003. Trucks were heavily loaded and tested over numerous routes involving urban, suburban, and rural
driving. Several trucks were re-acquired and tested a second time after 2-3 years. Data were collected at 5-
Hz frequency, which we averaged around each second to convert the data to a 1.0-Hz basis.
Heavy-Duty Diesel In-Use Testing (HDIUT). The manufacturer-run in-use testing program for heavy-duty
diesel vehicles was promulgated in June 2005 to monitor the emissions performance of heavy-duty engines
operated under a wide range of real world driving conditions within the engine's useful life.23 It requires
each manufacturer of heavy-duty highway diesel engines to assess the in-use exhaust emissions from their
engines using onboard, portable emissions measurement systems (PEMS) during typical operation while on
the road. The PEMS unit must meet the requirements of 40 CFR 1065 subpart J. The in-use testing program
began with a mandatory two-year pilot program for gaseous emissions in calendar years 2005 and 2006.
The fully enforceable program began in calendar year 2007 and is ongoing. The vehicles selected for
participation in the program are within the engine's useful life, and generally, five unique vehicles are
selected for a given engine family. This dataset includes results for HHD, MHD, and LHD vehicles. The
HDIUT data are publically available on EPA's website.24
The data available for use in MOVES2014 were collected during calendar years 2005 through 2010 and
represent engines manufactured in model years 2003 to 2009. For MOVES3, we evaluated data from
engines selected for testing in calendar years 2010 through 2018. These engines cover model years 2010 to
2016. The MY 2010+ data set included 40 unique engine families and 372 vehicles. There are about 10
million seconds of quality-assured second-by-second data covering about 85,000 miles of instrumented
travel. The operational conditions include a wide range of driving speeds, transient and steady-state
conditions, engine loads, and exhaust temperature conditions that have implications for emissions control
efficacy, particularly for NOx.25 For the HHD class, out of a total 159 vehicles selected for testing during
2010-2016, 109 were line-haul, 46 were delivery, and the remaining were marked as "Other" in the
metadata. Since the HDIUT data is measured and submitted by the manufacturer and the test vehicles are
required to be free of any tampering or mal-maintenance, we can safely assume that they represent zero-
mile vehicles for the purpose of assigning base rates and applying the tampering and mal-maintenance
effects. We plan to expand the characterization of the MY 2010+ HDIUT data set, in a future update to this
report, by adding summary information on vehicle age distribution, odometer reading, idling time
duration, and operating mode based time and miles travelled.
Houston Drayage Data. In coordination with the Texas Commission on Environmental Quality, the
Houston-Galveston Area Council, and the Port of Houston Authority, EPA conducted a study collecting
emissions data from trucks in drayage service using portable emission measurement systems (PEMS) from
December 2009 to March 2010.26 The trucks studied were diesel-fueled, heavy heavy-duty trucks used to
transport containers, bulk and break-bulk goods to and from ports and intermodal rail yards to other
locations. These trucks conducted the majority of their travel on short-haul runs, repeatedly moving
18
-------
containers across fixed urban routes. Note that only small fractions of trucks involved in drayage service
are dedicated solely to this function, with most trucks spending large fractions of their time performing
other types of short-haul service. No specific drive cycles were used and all PEMS testing was based on
actual in-use loads and speeds.
A summary of vehicles by model years for the above-mentioned datasets is provided in Table 2-2.
Table 2-2 Numbers of Diesel Vehicles from the ROVER, WVU MEMS, HDIUT, and Houston Drayage Programs by
Model Year Group
Regulatory Class
Data Source
MYG
HHD
MHD
LHD1
BUS
ROVER and
Consent Decree
Testing
1991-1997
19
-
-
2
1998
12
-
-
-
1999-2002
78
30
-
25
2003-2006
91
32
-
19
HDIUT
2003-2006
40
25
15
-
2007-2009
68
71
24
-
2010+2
194
74
94
10
Houston Drayage
1991-1997
8
-
-
-
1998
1
-
-
-
1999-2002
10
-
-
-
2003-2006
8
-
-
-
Note:
1 LHD45 and LHD2b3 vehicles were grouped together and classified as LHD vehicles for analysis of rates. The only
difference in the LHD45 and LHD2b3 emission rates is the application of production volume weighting and T&M
effects summarized in Table 2-10.
2 New data used to update heavy-duty diesel MY 2010 and beyond rates in MOVES3.
For the pre-2010 emission rates, we used a combination of the ROVER, Consent Decree Testing and the
HDIUT testing to estimate the emission rates. The Houston Drayage Study was used as comparison study
only. Additional details are provided in Section 2.1.1.4. The HDIUT data are used exclusively for the 2010
and later model year emission rates as discussed in Section 2.1.1.4.
19
-------
2.1.1.2
Measurement Accuracy and Quality Assurance
PEMS devices continue to make improvements that affect measurement accuracy, such as sensor
response, sample conditioning, and noise reduction. The data sets represent the accuracy of the
instruments at the time of measurement. In compliance determinations, when determining whether the
tested vehicle meets the in-use emissions standard or not, an "accuracy margin for portable in-use
equipment" (commonly referred to as measurement allowance) is added onto the standard; increasing the
vehicle compliance margin. The accuracy margins vary by model year and type of measurement method
and are described in 40 CFR 86.1912. This is done to prevent measurements that are biased-high from
affecting the compliance decision. However, since the true value for each second of data is unknown and
errors could be biased either high or low, the in-use emission rates used in MOVES from each of these data
sets are not adjusted to reflect the measurement allowance.
From each data set, we used only tests we determined to be valid. For the ROVER dataset, we eliminated
all tests with any reported problems, including GPS malfunctions, PEMS malfunctions, etc., whether or not
they affected the actual emissions results. For HDIUT data for MY2009 and earlier and Houston Drayage,
the time-alignment was visually confirmed by comparing relevant time-series plots, such as exhaust mass-
flow rate vs. C02 concentration, and exhaust-mass flow rate vs. engine speed, as recorded by the engine
control unit (ECU). Data was generally aligned within one second. When an issue with the time-alignment
was found, efforts were made to realign the data as much as possible. As our own high-level check on the
quality of PEMS and ECU output, we then eliminated any trip from ROVER, HDIUT (MY 2009 and earlier),
and Houston Drayage where the Pearson correlation coefficient between C02(from PEMS) and engine
power (from ECU) was less than 0.6. The correlation check removed approximately 7 percent of the ROVER
and HDIUT (MY 2009 and earlier) data. All the data from Houston Drayage met the criteria for correlation
between C02 and engine power. In addition, data were excluded from the analysis when the vehicle speed
was not available due to GPS and/or ECU malfunctions, when no exhaust flow was reported, and when a
periodic zero correction was being performed on gas analyzers. For the WVU (West Virginia University)
MEMS data, WVU itself reported on test validity under the consent decree procedure and no additional
detailed quality checks were performed by EPA. Table 2-2 shows the total distribution of vehicles by model
year group from the emissions test programs above following evaluation of the validity of the data.
In analyzing the HDIUT data for MY 2010+, we checked the time-alignment and deleted any second of data
that met any of the following conditions: (1) instrument was undergoing zeroing, as marked by the zero
flag field; (2) engine RPM was below 500; (3) vehicle speed was missing or below zero; (4) acceleration was
missing; (5) engine torque was missing; (6) measured exhaust flow rate was missing, or less than or equal
to zero; and (7) as catch-all, if the calculated STP and OpMode were invalid numbers. We did not verify the
accuracy of exhaust flow rate measurement and C02 measurement using techniques such as carbon-
balance versus ECU reported fuel rate data. Such verifications are assumed to have been done (by the
manufacturer) before data is submitted to EPA since they are required by 40 CFR 1065 subpart J.
20
-------
2.1.1.3 Calculation of Operating Modes
As discussed in Section 1.6, we prefer to estimate tractive power from engine data collected during real-
world operation rather than using the road-load equation. To do so, we first identified the seconds in the
data that the truck was either idling or braking based on acceleration and speed criteria shown in Table
1-4. For all other operation, engine speed coeng and torque Teng from the ECU were used to determine
engine power Peng, as shown in Equation 2-1.
\ng ®eng^eng Equation 2-1
We then determined the relationship between the power required at the wheels of the vehicle and the
power required by the engine. We first had to account for the losses due to accessory loads during
operation. Some accessories are engine-based and are required for operation. These include the engine
coolant pump, alternator, fuel pump, engine oil pump, and power steering. Other accessories are required
for vehicle operation, such as cooling fans to keep the powertrain cool and air compressors to improve
braking. The third type of accessories is discretionary, such as air conditioning, lights, and other electrical
items used in the cab. None of these power loads are subtracted in the engine torque values that are
output from the engine control unit.
We calculated accessory load requirements and mapped them to STP bins in a number of steps. First, we
grouped the accessories into five categories: cooling fan, air conditioning, engine accessories, alternator
(to run electrical accessories), and air compressor. We identified where the accessories were
predominately used on a vehicle speed versus vehicle load map to properly allocate the loads. For
example, the cooling fan will be on at low vehicle speed where the forced vehicle cooling is low and at high
vehicle loads where the engine requires additional cooling. The air compressor is used mostly during
braking operations; therefore, it will have minimal load requirements at high vehicle speeds. Table 2-3
identifies the predominant accessory use within each of the vehicle speed and engine load map areas.
At this point, we translated the vehicle speed and engine load map into engine power levels. The engine
power levels were aggregated into low (green), medium (yellow) and high (red) as identified in Table 2-3.
Low power means the lowest third, medium is the middle third, and high is the highest third, of the
engine's rated power. For example, for an engine rated at 450 hp, the low power category would include
operation between 0 and 150 hp, medium between 150 and 300 hp, and high between 300 and 450 hp. So,
for example, when vehicle operation is in middle of the engine load map and vehicle speed is low or mid
speed, the engine power level is in medium (yellow) band and the active accessory loads are as listed in the
respective cells. However, for the same engine load map operation (mid) with vehicle speed at high, the
engine power level is high (red) and active accessory loads are as listed in the cell. Some accessory loads,
21
-------
such as cooling fan, are absent from cells with the same engine power level (identified by color) based on
the reasons given in the previous paragraph.
Table 2-3 Accessory Use as a Function of Speed and Load Ranges, Coded by Power Level
Vehicle Speed (mph)
Low (0-25)
Mid (25-50)
High (50+)
Engine Load Map
Low
Cooling Fan
Air Cond.
Engine Access.
Alternator
Air Compress
Air Cond.
Engine Access.
Alternator
Air Compress
Air Cond.
Engine Access.
Alternator
Mid
Cooling Fan
Air Cond.
Engine Access.
Alternator
Air Compress
Cooling Fan
Air Cond.
Engine Access.
Alternator
Air Compress
Air Cond.
Engine Access.
Alternator
High
Cooling Fan
Air Cond.
Engine Access.
Alternator
Air Compress
Cooling Fan
Air Cond.
Engine Access.
Alternator
Air Compress
Cooling Fan
Air Cond.
Engine Access.
Alternator
Next, we estimated the power required when the accessory was "on" and percentage of time this
occurred. The majority of the load information and usage rates are based on information from "The
Technology Roadmap for the 21st Century Truck."27
The total accessory load is equal to the power required to operate the accessory multiplied by the percent
of time the accessory is in operation. The total accessory load for each cell in Table 2-3 is equal to the sum
of each accessory load. The calculations are included in Appendix A. The total accessory loads P/oss,OCc listed
below in Table 2-4 are subtracted from the engine power determined from Equation 2-1 to get net engine
power available at the engine flywheel.
For pre-2010 model years, LHD losses were set to zero. The losses for MY 2010+ LHD vehicles were
estimated by adjusting the MHD vehicle losses as such: (1) removed the loss for air compressor; (2) no
change to air condition loss; (3) scaled the losses for cooling fan, alternator, and engine accessories by 5/7
(where 5 and 7 are rough estimates of the average engine displacement in liters for LHD and MHD engines,
respectively). Based on these adjustments, the LHD losses are estimated to be approximately 60 percent of
22
-------
MHD losses for the low power band and 70 percent for the mid and high power bands. We acknowledge
this calculation relies on a number of assumptions, but we believe it is a step forward from having LHD
losses equal to zero for all model years (as was the case in MOVES2014).
Table 2-4 Estimates of Accessory Load in kW by Engine Power Level
Engine Power Level
(of rated power)
HHD
MHD
LHD1
(pre-2010)
LHD1
(2010+)
Urban
Bus
Low (0 - 1/3rd)
8.1
6.6
0.0
4.1
21.9
Medium (l/3rd to
2/3rd)
8.8
7.0
0.0
4.8
22.4
High (2/3rd to 1)
10.5
7.8
0.0
5.5
24.0
Note:
1 In MOVES2014, the accessory load losses for LHD were assumed to be zero for all model years. In MOVES3, MY
2010+ LHD data (Table 2-2) is analyzed with non-zero accessory load losses. However, for pre-2010 MY LHD, we
continue to assume zero accessory load loss.
We then adjusted for the driveline efficiency, accounting for efficiency losses in the wheel bearings,
differential, driveshaft, and transmission. The efficiency values were determined through literature
searches. Driveline efficiency rjdnveUne varies with engine speed, vehicle speed, and vehicle power
requirements. Using sources available in the literature, 28.29.30.31.32.33.34.35.36 we estimated an average value
for driveline efficiency. Table 2-5 summarizes our findings.
23
-------
Table 2-5 Driveline Efficiencies Found in the Literature
Vehicle Type
Data Source
Driveline Efficiency
General Truck
Barth (2005)
80-85%
General Truck
Ludic (2001)
75-95%
HDT
Rakha
75-95%
NREL (1998)
91%
Goodyear Tire Company
86%
Ramsay (2003)
91%
21st Century Truck (2000)
94%
HDT Single Drive/direct
SAE J2188 Revised Oct.
2003
94%
HDT Single Drive/indirect
92%
HDT Single Drive/double
indirect
91%
HDT Tandem Drive/direct
93%
HDT Tandem
Drive/indirect
91%
HDT Tandem
Drive/double indirect
89%
Bus
Prtichard (2004):
Transmission Eff.
96%
Hedrick (2004)
96%
MIRA
80%
Based on this research, we used a driveline efficiency of 90 percent for all HD regulatory classes. Equation
2-2 shows the translation from engine power Peng to axle power Pox/e. MOVES uses the Pi0SSiacc from Table
2-4 for each regulatory class and engine power-level (high, medium, or low).
24
-------
Paxle 11cLrivelineiPeng Ploss,acc) Equation 2-2
Finally, we scaled the axle power using Equation 2-3, and the STP-scaling factors/sco;e presented in Table
1-3 for every second of data.
STP=—Equation 2-3
^scale
We then constructed operating mode bins defined by STP and vehicle speed according to the methodology
outlined earlier in Section 1.5. It is possible that future test programs might acquire accessory load
information from the ECU and axle efficiency data available through certification information during the
HD GHG Phase 2 compliance program.
2.1.1.4 1960-2009 Model Years
The 1960-2009 model year emission rates were originally developed for MOVES2010 using the ROVER and
Consent Decree Testing data and then evaluated using MY 2003-2009 HDIUT testing data and the Houstan
Drayage Program data.37,38 In MOVES2014, the emission rates were considered for an update using the
HDIUT or Houston Drayage Program data if:
1. MOVES2010 rates were not based on actual data, and
2. The comparison to the new independent data shows that more than a half of MOVES2010 emission
rates are outside the boundary of the 95 percent confidence intervals of the independent data.
Using this methodology, the emission rates for HHD MY 2003-2006 vehicles and LHD2b3 and LHD45 MY
2003-2006 vehicles were updated using the MY 2003-2006 HDIUT data.b The Houston Drayage Program
was not used to estimate any of the emission rates in MOVES.
Table 2-10 outlines the data sets used to estimate emission rates for each model year group and regulatory
class. ROVER and Consent Decree testing data was used to estimate the following regulatory class and
model year groups combinations: HHD 1991-1997, 1998, 1999-2002, 2003-2006, MHD 1999-2002, 2003-
b This analysis is described in more detail in the MOVES2014 heavy-duty exhaust report However, since we have
updated the/sCa/e values used to assign STP bins, the MOVES3 rates are no longer directly comparable (see
Presentation by Choi et al. (201237).
25
-------
2006c, and Urban Bus 1991-1997, 1999-2002, 2003-2006. The HDIUT was used to estimate emissions for
HHD 2007-2009 and LHD 2003-2006.
Emissions in each data set were reported in grams per second. To calculate MOVES heavy-duty exhust
emission rates, we first averaged all the 1-Hz NOx emissions by vehicle and operating mode because we did
not believe the amount of driving done by each truck was necessarily representative. Then, the emission
rates were again averaged by regulatory class and model year group. For trucks MY 2009 and older, these
data sets were assumed to be representative and each vehicle received the same weighting. Equation 2-4
summarizes how we calculated the mean emission rate for each stratification group (i.e., model year
group, regulatory class, and operating mode bin).
ynveh 1^1 = 1 j> i |
]=1 I Uj J Equation 2-4
Tp
"veh
Where:
rij = the number of 1-Hz data points (for a given operating mode bin) for each vehicle j,
fiveh = the total number of vehicles,
rpjj = the emission rate of pollutant p for vehicle j at second /',
Y = the mean emission rate (meanBaseRate) for pollutant p (for a given model year group,
P
regulatory class and operating mode bin).
We calculated a mean emission rate, denoted as the "meanBaseRate" in the MOVES emissionRateByAge
table, for each combination of regulatory class, model year group, and operating mode bin combination.
95% confidence intervals of the mean emission rate were calculated by accounting for the variability of the
averages across different trucks in a regulatory class and model year group. Examples of mean emission
rates are displayed in Figure 2-1. As expected, the emissions increase with power, with the lowest
emissions occurring in the idling/coasting/braking bins.
c For 2003-2006 MHD, the emission rates are different than the 2003-2006 HHD emissions for operating modes 0,1,
11, 21, and 33. For the other operating modes, the emission rates are equivalent to the HHD emission rates.
26
-------
The data included in the emissions analysis does not cover all operating modes or regulatory classes and
model year combinations needed for MOVES. In the following sections, we discuss the methods used to fill
missing operating mode bins, and missing regulatory class and model year combinations. In addition, we
also estimate the impact of low- NOx rebuilds which were not included in the sampled vehicles. To do so,
we rely on the heavy-duty diesel emission standards, as well as engineering knowledge and test data of
emission control technologies that were implemented or forecasted to be implemented to meet the
standards.
2.1.1.4.1 Light Heavy-Duty Class 2b3 and Classs 4&5 Trucks
As described in Section 1.5, the LHD regulatory classes were redefined for MOVES3, and the fscaie value for
2009-and-earlier LHD2b3 and LHD45 regulatory classes is now 2.06 metric tons.d This differs from
MOVES2014 where the value for these vehicles was 17.1 metric tons. Thus, it was not possible to carry
over the emission rates from MOVES2014. Instead, in MOVES3, the 2009-and-earlier LHD2b3 and LHD45
regulatory classes were assigned the same emission rates as the LHD<10K in MOVES2014.
2.1.1.4.2 High-Power Operating Modes
As described in Section 1.5, fscaie values for MHD and HHD trucks were not changed for model years 2009-
and-earlier. Thus, for MHD and HHD trucks, the maximum operating mode (opModelD = 40) represents a
tractive power greater than 513 kW (STP= 30 skW x 17.1). This value exceeds the capacity of most HHD
vehicles; MHD vehicles and buses exert even less power. Nearly all of the HHD activity occurs in modes 0,
1, 11-16, 21-28, and 33-38, with activity for buses and MHD vehicles usually occurring over an even smaller
range.
To estimate emission rates in the modes beyond the ranges of available data, for each model year group
we linearly extrapolated the rates from the highest operating mode in each speed range where significant
data were collected. In most cases, this mode was mode 16 for the lowest speed range, mode 27 or 28 for
the middle speed range, and mode 37 or 38 for the highest speed range. For each of these operating
modes, work-specific emissions factors (g/kW-hr) were calculated using the midpoint STP (Table 1-4).
Then, these emissions factors were multiplied by the midpoint STP of the higher operating modes (e.g.,
modes 39 and 40 for speed>50 mph) to input emission rates for the modes lacking data. For the highest
bins in each speed range, a "midpoint" STP of 33 skW (564.3 kW) was used. Equation 2-5 displays an
example calculation of the emission rate for opModelD 40, using a mean emission rate from opModelD 37,
for a given regulatory class and model year group.
d This is consistent with the 2.06 fscaie used to develop LHD2b3 and LHD45 emission rates in MOVES2010.
27
-------
^oijMocLcID 40 l
Equation 2-5
ropModelD 37 /
Figure 2-1 shows NOx emission rates by operating mode for HHD trucks in MY 2002. The mean emission
rates for the highest STP operating mode bins (30 and 40) are extrapolated using the method explained
above. In addition, the confidence intervals for the extraplolated bins are copied from the closest bin with
collected data. However, because the data are extrapolated, the uncertainty of these rates is larger than
what is shown by the copied confidence intervals.
Figure 2-2 displays the MY 2002 MHD and Urban Bus regulatory classes emission rates, with the error bars
removed for clarity. For these vehicle, less data was captured at the highest operating mode bins, and
more of the high speed, high power emission rates were extrapolated, which explains the strictly linear
trend in the emission rates between operating mode 27 -30 and 37-40.
4000
3500
3000
2500
x
22000
(0
01
E 1500
1000
500
» ¦ »
—i 1—i—i 1—i—i 1—i 1—i—i 1—i—i 1—i 1—i—i 1—i—i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating Mode
Figure 2-1 NOx Emissions by Operating Mode from HHD Trucks for Model Year 2002. Error Bars represent the 95
percent confidence interval of the Mean
28
-------
6000
5000
"Si 4000
00
01
re
* 3000
re
41
2000
1000
¦ MHD
Bus
~ HMD
~
~ ¦
I 4
~ ¦
. * '
¦ I*
1 1 1 1 ! S 1 i m I 1 1 1 1 1 1 1 J i 1 1 1 ! 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating Mode
Figure 2-2 NO* Emissions by Operating Mode from MHD, HHD, and Urban Bus Regulatory Classes for Model Year
2002.
2.1.1.4.3 Missing Regulatory Class and Model Year Combinations
For regulatory class and model year combinations with missing data, we set the emissions equal to
regulatory class/model year combinations which had equivalent emission standards or proportionally
adjusted the existing emissions data using ratios of certification data or vehicle emission standards as
specified in Table 2-1. For HHD model year groups 1988-1989 and 1990, we increased the 1991-1997
model year group emission rates by a factor proportional to the increase of the certification levels as
analyzed for MOBILES.39 On average, the MY 1990 and MY 1988-1989 rates are 1.055 times and 1.367
times the baseline rates of MY 1991-1997, respectively. We applied the 1988-1989 emission rates to model
years 1987 and earlier. For MHD vehicles from MY 1960-1997, we used the HHD emission rates.
For model year 1998, data existed for HHD trucks but not for buses. In these cases, we calculated the rates
for Urban Buses by multiplying the Urban Bus emission rates for 1999-2002 by the ratio of HHD emission
rates between the 1998 and the 1999-2002 model year groups, as shown in Equation 2-6.
HHD rates 1998
Urban Bus ratesAqqR = - x Urban Bus rates-,ooQ_7nn? Equation 2-6
1998 HHD rates1999-2002 1999 2002
29
-------
For LHD2b3 and LHD45 vehicles, no data were available for the pre-2003 model year vehicles. For MY
1999-2002, we analyzed MHD engine data using an fSCaie of 2.06 as appropriate for LHD vehicles (See Table
1-3). For MY 1998, we copied the LHD emission rates from MY group 1999-2002. We also used thse rates
as base rates to back-cast emission rates for the 1991-1997 model years, using the ratio of emission
standards between these two model-year groups (5.0 g/bhp-hr over 4.0 g/bhp-hr) such that the 1991-1997
rates are 1.25 times the rates for MY 1998.
Table 2-10 provides a summary of the assumptions used to estimate emission rates for regulatory class-
model year groups with missing data.
The default emission rates for HHD and MHD diesel vehicles in MOVES for model years 1991 through 1998
are intended to include the effects of defeat devices as well as the benefits of heavy-duty low-NOx rebuilds
(commonly called reflash) that occurred as the result of the heavy-duty diesel consent decree.17 Reflashes
reduce NOx emissions from these engines by reconfiguring certain engine calibrations, such as fuel
injection timing.
Since defeat devices were in effect mostly during highway or steady cruising operation, we assumed that
NOx emissions were elevated for only the top two speed ranges in the running exhaust operating modes
(>25mph). To modify the relevant emission rates to represent reflash programs, we first used emission
rates from model year 1999 (the first model year with not-to-exceed emission limits, see Table 1-2) to
calculate baseline ratios of the emission rates for operating modes 27 and 37 to the rate for opMode 16 (as
discussed in the beginning of Section 2.1.1.4). We then multiplied the MY 1999 ratios by the emission rates
in operating mode 16 for model years 1991 through 1998, to get estimated "reflashed" emission rates for
operating modes 27 and 37. This step is described in Equation 2-7 and Equation 2-8. To estimate
"reflashed" rates in the remaining operating modes, we multiplied the reflashed rates by ratios of the
remaining operating modes to mode 27 for MY 1991-1998, as shown in Equation 2-9 and Equation 2-10.
The second step preserves the relationship among operating mode 21 and operating modes 21-30, and
among operating mode 37 and operating modes 31-40, that existed in the original 1991-1998 data.
Where: operating
2.1.1.4.4 Defeat Device and Low-NOx Rebuilds for 1991-1998 Model Year HHD and
MHD
modes (OM) 21-
30
Equation 2-7
30
-------
1~re flash, 91-98,0Mx 're flash, 91-98,27
r91-98,0Mx
^91 —98,27
Equation 2-8
Where: operating
modes (OM) 31-
40
rreflash,91-98,37 ~ r91-98,16
r1999,37
71999,16
Equation 2-9
Equation 2-10
Because the reflash occurred over time after the engines were sold, we phase-in the reflash rates with age.
An EPA assessment shows that about 20 percent of all vehicles eligible for reflash had been reflashed by
the end of 2008.40 We assumed that vehicles were reflashed at a steady rate from the time of the consent
decree (1999/2000 calendar year), such that in 2008, about 20 percent had been reflashed. Figure 2-3
displays the results of the reflash calculations on the HHD fleet-average emission rate for operating mode
37. For model years 1994-1997 and 1998, we approximated a linear increase in reflash rate from age zero
to the age 20+ age group. For model years 1991-1993, the adjustments do not start until calendar year
2001, when the 1991 model years exist in the age 10-15 group. When all of the 1991-1998 vehicles have
reached the age 20+ group (CY 2018), MOVES assumes that close to 30% of all the 1991-1998 engines have
been reflashed. The reflashed HHD 1991-1998 MY emission rates were also applied to the 1991-1998 MHD
diesel emission rates. Note that there are no tampering and mal-maintenance aging effects in MOVES for
pre-2010 NOx emission rates. Thus, the only change in the NOx emission rates by vehicle age for these
model years is due to reflashing.
31
-------
2540
2520
2500
2480
£ 2460
cD
w244o
g 2420
H 2400
2380
2360
2340
'¦Q..£
. s
•O-'G..
"G"G.,
| *' V
|
—•—
Aodel Year 1991-1993
vlodel Year 1994-1997
—•—Model Year 1998
\
i • • • •
\
\
\ m m » m t
0
10
15
Age (years)
20
25
30
Figure 2-3, HHD NOx emission rates for OpModelD 37 for Model Years 1991-1998 Adjusted for Low-NOx Rebuilds by
Vehicle Age
2.1.1.4.5 2007-2009 Model Year HHD, MHD, and Urban Bus
The 2007 Heavy-Duty Rule107 required the use of ultra-low sulfur diesel fuel; this fuel enabled diesel
engines with diesel particulate filters to reach the 0.01 g/bhp-hr PM standard beginning in 2007. In
addition, the 2007 Heavy-Duty Rule107 established much tighter NOx emission standards (0.2 g/bhp-hr).
While the NOx standard went into effect for MY 2007 at 0.2 g/bhp-hr, it was phased in over a three-year
period ending in MY 2010. Rather than phasing in the aftertreatment technology needed to meet the new
standard, most manufacturers chose to meet a 1.2 g/bhp-hr standard for MY 2007-2009 (down from 2.4
g/bhp-hr in 2006), which did not require NOx aftertreatment. For the 2007-2009 HHD, we used the data
from the HDIUT program. For the NQX emission rates within the 2007-2009 model year group for MHD and
Urban Bus, we assumed that the NOx emission rates were 50 percent lower than the corresponding Rover
and Consent Decree derived 2003-2006 emissions (proportional to the reduction in the NOx emission
standards mentioned above). The MHD MY 2007-2009 rates are consistent with NOx emission rates
measured from 2007-2009 MHD trucks measured in HDIUT.38
2.1.1.4.6 2007-2009 Model Year LHD45 and LHD2b3
For LHD2b3 trucks in 2007-2009, we accounted for the penetration of Lean NOx Trap technology. Cummins
began using Lean NOx Trap (LNT) aftertreatment starting in 2007 in engines designed to meet the 2010
standardin vehicles such as the Dodge Ram. This technology allows for the storage of NOx during fuel-lean
32
-------
operation and conversion of stored NOx into N2 and H20 during brief periods of fuel-rich operation. In
addition, to meet particulate standards in MY 2007 and later, heavy-duty vehicles are equipped with diesel
particulate filters (DPF). The DPF must be regenerated at regular intervals to remove and combust
accumulated PM to relieve backpressure and ensure proper engine operation. This step requires high
exhaust temperatures. However, these conditions adversely affect the LNT's NOx storage ability, resulting
in elevated NOx emissions.
In order to determine the fraction of time that DPFs spend in PM regeneration mode, in 2007, EPA
acquired a truck equipped with a LNT and a DPF and performed local onroad measurements using portable
instrumentation and chassis dynamometer tests. We distinguished regimes of PM regeneration from
normal operation based on operating characteristics, such as exhaust temperature, air-fuel ratio, and ECU
signals. During the testing conducted onroad with onboard emission measurement and on the chassis
dynamometer, we observed a PM regeneration frequency of approximately 10 percent of the operating
time.
Emissions from this vehicle were not directly used to calculate emission rates, because only one vehicle
was tested. Rather, adjustments were made to the MOVES2010® MY 2003-2006 model year group to
develop emission rates for this model year group and regulatory class. During PM regeneration, we
assumed that the LNT did not reduce emissions from 2003-2006 levels. During all other times, we assumed
that emissions were reduced by the percent reduction in the certification standards from 2003-2006 levels
(i.e., 90 percent). These assumptions result in an estimated NOx reduction of 81 percent for LNT equipped
trucks from 2003-2006 to 2007-2009, as shown in Equation 2-11.
LNT NOx emissions
Baseline LHD2b3 (2003 — 2006) NOx emissions
(LNT normal emissions \
= (normal op. frequency) x ( baseline ) Equation
/baseline emissions\ 2-11
+ (DPF reg. frequency) x -
V baseline emission J
= (0.90) X (0.10) + (0.10) X (1) = 0.19
We weighted the average rates for the LHD2b3 regulatory class (regClassID 41) for model years 2007-2009
assuming that LNT-equipped trucks account for 25 percent of the LHD2b3 diesel vehicles. We assume that
the remaining 75 percent of MY 2007-2009 LHD2b3 diesel trucks will not have LNT aftertreatment and will
e In MOVES2014, we updated the diesel NOx emission rates for 2003-2006 based on the HDIUT program.
33
-------
exhibit a 50% NOx reduction from the 2003-2006 model year emission rates as was assumed for the HHD
and MHD described in the previous section. Overall, these assumptions result in a 58 percent reduction in
NOx emission rates in the model year 2007-2009 emission rates from the MOVES2010 MY 2003-2006 NOx
emission rates as shown in Equation 2-12.
2007 — 2009 LHD2b3 NOx emissions
2003 — 2006 LHD2b3 NOx emissions
LNT NOx emissions \
= (LNT market share) —— r TTr^„, „ ,T^ Equation
V2003 — 2006 LHD2b3 NOx emissionsJ
+ (non
— LNT market share)
2-12
2007 — 2009 emission standerds \
2003 — 2006 NOx emissions standerdsJ
= (0.25) x (0.19) + (0.75) x (0.5) = 0.4225
In the absence of other data, we applied the LHD2b3 emission rates to LHD45 vehicles. Newer data shows
that LNT is not being used in LHD45 vehicles, however, we have not updated this assumption because the
LHD results compare well to the HDIUT data.37
2.1.1.5 2010-2060 Model Years
In MOVES3, the MY 2010 and later emission rates for HHD, MHD, Urban Bus, and LHD45, and LHD2b3 are
based on analysis of the HDIUT data and model-year specific production volume weighting for each model
year from 2010 through 2018. The rates for HHD, MHD, and the two LHD classes use data from vehicles
with HHD, MHD, and LHD engines, respectively. The NOx emission rates are projected to remain constant
for MY 2018 and later vehicles for regulatory classes HHD, MHD, Urban Buses, and LHD45. The LHD2b3
trucks are projected to have a further decrease in NOx emissions through the implementation of the Tier 3
program as discussed in Section 2.1.1.5.5.
In calculating the mean emission rates for MY 2010+ vehicles, we made several additions to the method
presented for the pre-2010 model years described in Section 2.1.1.4:
1. For a given regulatory class (HHD, MHD, LHD), we grouped the vehicles in the HDIUT data set into
three NOx family emission limit (FEL)f groups as detailed in Section 2.1.1.5.1 below.
f A Family Emission Limit is the maximum emission level established by a manufacturer for the certification of an
engine family.
34
-------
2. Within the N0X FEL group and regulatory classes, we grouped vehicles into model year groups as
detailed in Section 2.1.1.5.2.
3. We calculated the operating mode-based average emission rate for each vehicle by regulatory
class, NOx FEL group, and model year group (Equation 2-13). Then, we calculated the operating
mode-based average emission rate for all vehicles in the NOx FEL group and model year group
(Equation 2-14).
4. We weighted the operating mode-based average emission rates for each of the NOx FEL groups,
and model year group within each regulatory class by the model year specific production volumes
of the engines in the NOx FEL group to the total production volume of the regulatory class
(Equation 2-15). Thus, we created operating mode-based average emission rates for each model
year and regulatory class.
f-.p, ^isec ERpol,OM,C,FEL,MYG,veh,sec n i4_. 1->
zRpol,OM,C,FEL,MYG,veh ~ bquatlOn Z-13
secC0unt
T.vehERpol,OM,C,FEL,MYG,veh r .. „ .
ERPoI,om,c,fel,myg = 7 Equation 2-14
V^flcount
ERpoi,OM,c,MYG,MY — / I ERpoi,OM,c,FEL,MYG * v ox/ ) Equation 2-15
\ 2-iFel c,my,fel /
FEL
Where:
C = Regulatory class (LHD, MHD, HHD, and Urban Bus)
ERx,y,z = Emission rate in mass/time. The subscripts show the categorization.
FEL = NOx FEL group of engine (0.20 g/bhp-hr, 0.35 g/bhp-hr, and 0.50 g/bhp-hr)
MYG = Model year group (2010-2013, 2014-2016)
MY = Model year
OM = Running exhaust emissions operating mode
pol = Pollutant (NOx, THC, CO)
PVc,my,fel = Production volume by class, model year, and FEL group
sec; secCOUnt = a second of data (for a given i/eh and OM); number of seconds in that category
veh; vehcount = a vehicle (in the class and FEL grouping); number of vehicles in that category
35
-------
Figure 2-4 displays the average N0X emission rates using the described method for HHD model year 2013
vehicles. To calculate the 95% confidence intervals for the model years, the confidence intervals were first
calculated for each FEL group, treating the mean emission rate by operating mode from each individual
truck as an independent random variable. Then, the confidence interals from the different FEL groups were
weighted together using the production volumes, similar to how the means were calculated.
0.10 -
0.08 -
< 0.06 -
35
x
O
Z 0.04 -
0.02 -
0.00 -
Figure 2-4
More details about the selection of the NOx FEL groupings, updated fscaie values, and the methods for the
production volume weighting are provided in Section 2.1.1.5.1 through 2.1.1.5.4 below.
2.1.1.5.1 NOx Family Emission Level Groups
We grouped engines, within a regulatory class, by their NOx FEL. These groups are shown in Table 2-6.
Table 2-6 NOx Family Emission Limit (FEL) based Groups for LHD, MHD, HHD, and Urban Bus Classes in the HDIUT
Data
Group Name
Range of NOx FEL (g/bhp-hr)
Lower Limit
(Excluded)
Upper Limit
(Included)
0.20
0.00
0.20
¦ HDIUT, MY 2013
i
i
il
I
T
i i
i
,
.
.ill
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
NOx Emissions by Operating Mode from HHD Trucks for Model Year 2013. Error Bars represent the 95
percent confidence interval of the Mean
36
-------
0.35
0.20
0.35
0.50
0.35
0.50
Each test vehicle, within a regulatory class, was assigned to one of these three groups. These groupings
were applied not only to NOx, but to all pollutants for emission rate calculations. We chose to use NOx as
the basis for creation of these groups because data for NOx FEL is most abundant in the heavy-duty engine
certification database and, for MY 2010-2015 engines, the biggest technology changes and tailpipe exhaust
emissions impacts are due to emissions control measures for NOx. We arrived at the specific group bins
based on the spread of NOx FELs for MY 2010-2015 engine families reported in the certification database
available at the time of the analysis (not just the engine families tested under the HDIUT program). The
NOx FELs for MY 2010-2015 HD diesel engine families in the certification database are shown in Figure 2-5.
As highlighted by the shaded rectangles, most of the engine families are concentrated in the 0.05-0.20,
0.30-0.35, and 0.45-0.50 bands and this trend guided our bin choices, represented by the curly braces, for
the three NOx FEL groups.
100
V)
a>
E
05
LL.
V
c
lo
*—i
o
rsi
rv O
tM)
C
o
rsi
>
so
80
60
40
* ^ 20
^
Vp-^' --v
-------
Urban Bus application; and there is no separate NOx standard (for MY 2010+) for the Urban Bus regulatory
class.
Table 2-7 Number of 2010 and Later Model Year HDIUT Vehicles by NOx FEL Group and Model Year Groups1 Used
for Emission Rate Analysis
NO. FEL Group
0.2
0.35
0.5
MY
MY
MY
MY
2010-
2014-
2010-
2010-
2013
2016
2016
2016
Total
LHD2
52
27
0
15
94
MHD
19
23
23
9
74
HHD3
78
50
31
35
194
Total
149
100
54
59
362
NOTE:
1 THC, CO, and CO2 emission rates were analyzed using the same model year groups. Sample size is generally the same, with a few
exceptions for vehicles with invalid pollutant measurements for one or more pollutant. For example, one of the MHD 0.35 NOx FEL
vehicles did not have NOx measurements.
2 LF1D data for the 0.2 FEL group includes MY up to MY 2016. MHD and HHD data for the 0.2 FEL group only include up to MY
2015
3 The HHD 0.35 group contains five MY 2009 vehicles, which have similar NOx emission rates as the other vehicles in this group.
The average NOx emission rates for the 0.2 NOx FEL Group for model year 2010-2013, compared to the
0.35 and 0.5 NOx FEL Groups are shown below for LHD (Figure 2-6), MHD (Figure 2-7) and HHD (Figure 2-8).
Because the 0.35 and 0.5 NOx FEL groups do not have a model year distinction like the 0.2 NOx FEL group,
the figures display all the available MY 2010 and later vehicles for the 0.35 and 0.5 NOx FEL groups.
As shown, the NOx emission rates in the 0.2 NOx FEL Group and model year 2010-2013 are consistently
lower than those in the 0.35 and 0.50 NOx FEL groups, with a few minor exceptions. In addition, for many
operating mode and regulatory class combinations, the NOx emission rates in the 0.2 NOx FEL Group are
statistically significantly lower than the NOx emission rates in the 0.35 and 0.5 NOx FEL Groups. However,
no such trend exists in the NOx emission rates for the 0.35 and the 0.5 NOx FEL groups. In general, the NOx
emission rates in the 0.35 and the 0.5 NOx FEL groups are not statistically different than one another by
operating mode for the MHD and HHD regulatory classes.
38
-------
0.10 -
0.08 :
< 0.06 :
¦SB
x
O
Z 0.04 -
0.02 -
0.00
Figure 2-6 Average LHD NO* Emission Rates by Operating Mode for the 0.2 NOx FEL for MY 2010-2013 and the 0.5
NO* FEL for MY 2010-2015. Error Bars are 95% Confidence Intervals of the Mean
¦ LHD FEL 0.20, MY 2010-2013
¦ LHD FEL 0.35, N=0
N=52
¦ LHD FEL 0.50, MY 2010-2015
N=15
J
..ililili
i
..ihilii
i
1
i
.1,1,
1
i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
0.10 -
0.08 :
< 0.06 -
3B
x
O
z 0.04 -
0.02 -
0.00 -
Figure 2-7 Average MHD NOx Emission Rates by Operating Mode for the 0.2, 0.35 and 0.50 NOx FEL Groups for MY
2010-2013 Vehicles. Error Bars are 95% Confidence Intervals of the Mean
i 1 r
¦ MHD FEL 0.20, MY 2010-2013' N=23 I
I I I
¦ MHD FEL 0.35, MY 2010-2013^ N=23 j
p MHD FEL 0.50, MY 2010-2013- N=9 ;
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
39
-------
0.20
i 1
I HHD FEL 0.20, MY 2010-2013- N=78
I HHD FEL 0.35, MY 2010-2015' N=31
0.15 -
0.10 -
0.05 -
0.00
I HHD FEL 0.50, MY 2010-2015
Ms.
N=35
Jj
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 2-8 Average HHD NOx Emission Rates by Operating Mode for the 0.2 NOx FEL for MY 2010-2013 and the 0.35
and 0.5 NOx FEL for MY 2010-2015. Error Bars are 95% Confidence Intervals of the Mean
2.1.1.5.2 Model Year Groups within the 0.2 NOx FEL Group
We grouped the vehicles within the 0.2 NOx FEL Group further into 2010-2013 and 2014 and later model
year groups to account for differences in emissions performance of more recent engines and
aftertreatment systems. Both of the 2017 peer-reviewers recommended that MOVES3 consider varying
emission rates by model year for MY 2010 and later trucks, citing studies50,51,52 that supported the claim
that "since the inception of SCR conrol in 2010, significant advancements in thermal management, urea
dosage strategy, and SCR formulations have taken place."14 For MOVES3, we were able to incorporate data
from MY 2014-2016 engines to evaluate this recommendation.
Table 2-7 displays the model year groupings used in developing MOVES3 emission rates by NOx FEL groups
and pollutant. For NOx, THC, CO, and C02, we grouped 0.2 NOx FEL group into two model year groups given
the sufficient sample size. Within the 0.35 and 0.5 NOx FEL groups, we only have a single model year group
(2010 and later) due to the smaller sample size of HDIUT test vehicles in these groups. For example, for the
MHD regulatory class, the HDIUT data set only includes MY 2010-2013 vehicles in the 0.35 and 0.5 NOx FEL
Group (see Table 2-23). Aside from the small sample size, we believe it is defensible to have single model
year groups for the 0.35 and 0.5 NOx FEL groups for two additional reasons. First, with the exception of the
0.35 NOx FEL Group for MHD, heavy-duty vehicle manufacturers are no longer producing engines certified
in the 0.35 or 0.5 NOx FEL groups beyond MY 2017. Second, we anticipate the MY 2014-2017 MHD engines
certified to the 0.35 NOx FEL group would have similar emission control technology as earlier MY 2010-
40
-------
2013 vehicles certified to the same emission levels. For these reasons, we believe it is reasonable to use
only one model year group for the 0.35 and 0.5 NOx FEL groups, within each regulatory class.
A comparison of the emission rates for the 0.2 NOx FEL Group by model year group and regulatory class are
shown in Figure 2-9 through Figure 2-11. For the LHD vehicles, the NOx emission rates for MY 2014 and
later are lower than the MY 2010-2013 vehicles, with some of the differences being statistically significant.
For the MHD and HHD regulatory classes, no significant differences in the NOx emission rates between the
two model year groups are observed across operating modes. In contrast, significant differences are
observed for THC and CO emissions for the MHD and HHD regulatory classes between the MY 2010-2013
and MY 2014-2015 model year groups as discussed in Section 2.1.3.2. Even though the differeces in NOx
emission rates across regulatory classes and the model year groups were not statistically significant, we
separated the 0.2 l\IOx FEL group for all three regulatory classes into two model year groups (2010-2013
and 2014 and later) to be consistent in our analysis for all pollutants, and to account for the potential
differences in real-world l\IQx emissions performance due to the updated engine and aftertreatment
systems
0.030
0.025
_0.020
33
x 0.015
O
0,010
0.005
0.000
Figure 2-9 NOx emission rates for the MY 2010-2013 and MY 2014-2016 vehicles in the LHD 0.20 NOx FEL Group
MY 2010-2013
MY 2014-2016
JiJi
Jl
i<
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
41
-------
0.030
0.025
--C.
M 0.020
CD
03
^ 0.015
c
o
—
:««o.oio
E
UJ
0.005
0.000
MY 2010-2013
MY 2014-2015
ii
k
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 2-10 NOx emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the MHD 0.20 NO* FEL Group
0.08
0.07
0.06
-"C
ao
0.05
X
O
2
0.04
0.03
0.02
0.01
0.00
I MY 2010-2013
MY 2014-2015
i
JL
ii
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 2-11 NO NO* emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the HHD 0.20 NO* FEL
Group
2.1.1.5.3
Weighting by Production Volume
We collected production volume data by the same regulatory classes (LHD, MHD, HHD, Urban Bus) and
NOx FEL groups (0.20, 0.35, 0.50) that we used for the emission rate analysis. We then combined the NOx
FEL group-based rates (averaged within the model year groups described above) with the production
42
-------
volume data (distinct for each model year) to create emission rates unique to each model year. We did this
for each model year from 2010 through 2018 (the last model year for which we have production volume
data). For MY 2019 and later, we used the same production volume weighting as MY 2018. The per-model-
year production volume weighting, by regulatory class and NOx FEL groups, is shown in Figure 2-12. This
method allows us to better represent the technology adoption landscape in the heavy-duty domain. For
example, for HHD, model years 2010 through 2013 had engines with NOx FEL in the 0.50 group (0.35
g/bhp-hr < NOx FEL < 0.50 g/bhp-hr), but starting with model year 2014, there are no engines in the 0.50
group. Compared to engines in the 0.20 group and 0.35 group, engines in the 0.50 group predominantly
use a different emissions control strategy to reduce tailpipe NOx emissions. Thus, our approach using the
NO* FEL groups and per-model-year production volume correctly captures the prevalence and influence
(on emissions) of different technologies in the fleet. Production volume percent contributions of the three
NOx FEL groups sum to 100 percent of the production volume for each regulatory class and model year.
120
100
>
SO
a.
"ro
60
£
40
o
se
20
0
-20
LHDD
—« ~
-»-D.20
-©-0.35
-CI-0.50
2010 2011
2012 2013 2014 2015
2016
2017 2018
Model Year
120
100
>
B0
a.
IB
60
h-
40
o
SS
20
0
-20
MHDD
-•-0.20
-*—0.35
-O-Q.50
2010 2011 2012 2013
2014 2015 2016 2017 2018
Model Year
120
100
>
SO
a.
IB
60
K
h-
H—
40
O
SS
20
0
-20
\/
URBU /
->-0.20
-~-0.35
-O—0.50
2010 2011 2012 2013
2014
2015
2016
2017 2018
Model Year
Model Year
Figure 2-12 Production Volume Contribution of Heavy-Duty Diesel Engine Families by NO* Family Emission Limit
Group for Each Regulatory Class
2.1.1.5.4
Operating Modes arid fscaie Values
For the updates to THC, CO, NOx, PM2.5, and energy rates for MY 2010+ HD vehicles, we used new/sco/e
values (see Table 1-3 and Appendix G) that allowed all OpModes to be populated with rates based on real-
43
-------
world data. Thus, there was no need to perform the hole-filling approach used to populate the high-power
operating modes for the pre-2010 MY vehicles (see 2.1.1.4.2).
Figure 2-13 to Figure 2-15 show the effect of the new/sco;e values on OpMode coverage using the example
of NOx emission rates for the vehicles in the NOx FEL=0.20 group for LHD, MHD, and HHD regulatory
classes, respectively.8 Note that the absolute mass/time OpMode-based emissions rates between the two
series based on different fscaie cannot be compared. The main benefit of the new/sco/e values is that all the
operating modes are populated even if the trends may not be perfectly monotonically increasing for each
pollutant in a regulatory class. The comparison is similar for the 0.35 and 0.50 NOx FEL groups. Note the
final rates input into MOVES are estimated as production volume weighted rates from each of the NOx FEL
groups.
0.025 -
0.020 -
0.015 -
an
X
o
Z 0.010 -
0.005 -
0.000 -
Figure 2-13 Effect of MOVES2014 (17.1) and MOVES3 (5.00)/sra/e Values on OpMode Coverage for NOx Emission
Rates for Light Heavy-Duty Vehicles in the NOx FEL = 0.20 Group
¦ L H D DO. 20_f s=5.00
HLHDD 0.20 fs=17.1
3
E
E
¦
I
~
! 1
II
run
11111M
:
111111
:
i
1
i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
g These plots were generated using model year 2010-2015 HDIUT data that was available at the time of the analysis.
The final emission rates were developed using the complete HDIUT data that includes both MY 2010-2015 and MY
2016-2018.
44
-------
M
0.060
0.050 -
0.040 -
0.030 -
0.020 -
0.010
0.000
¦ MHDD_0.20_fs=7.00
0MHDD 0.20 fs=17.1
I i I
ill
I
11
1
I
. 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 2-14 Effect of MOVES2014 (17.1) and MOVES3 (7.00) fscaie Value on OpMode Coverage for NOx Emission
Rates for Medium Heavy-Duty Vehicles in the NOx FEL = 0.20 Group
0.080
0.070 H
0.060
5 0.050 H
M
0.040
o
2 0.030 -
0.020 -
0.010 -
0.000
¦ H H D DO. 20_f s=10.00
BHHDD 0.20 fs=17.1
MB.
ll
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 2-15 Effect of MOVES2014 (17.1) and MOVES3 (10.00) fscaie Value on OpMode Coverage for NOx Emission
Rates for Heavy Heavy-Duty Vehicles in the NOx FEL = 0.20 Group
45
-------
2.1.1.5.5 Incorporation of Tier 3 Standards for LHD2b3
In addition to regulating light-duty vehicles, the Tier 3 vehicle emission standard41 also applies to chassis-
certified light heavy-duty diesel vehicles, which constitute the majority of diesel vehicles within class 2b
and 3 vehicles (LHD2b3).h For these LHD2b3 diesel vehicles, reductions in emission rates attributable to the
Tier 3 standards are applied only to rates for NOx. For THC and CO emissions, the emission rates measured
in the HDIUT program, as described in Section 2.1.3.2, imply that current levels on the FTP cycle are
substantially below the Tier 3 THC and CO standards. For example, when MOVES rates are used to simulate
FTP cycle for NMHC, the result is a rate of approximately 0.05 grams per mile, while the simulated FTP
estimate for CO is less than 1.0 gram/mile. Consequently, we assumed that no additional reductions in THC
and CO emissions would be realized through implementation of the Tier 3 standards on LHD diesel
vehicles.
By contrast, the Tier 3 NOx standard results in a reduction of emissions from diesel vehicles in regulatory
classes LHD2b3. Because emission standards tend to impact start and running emissions differently, we
applied a greater portion of the reduction to running emissions and a smaller reduction to start emissions
These reductions were phased-in over the same schedule as for gasoline vehicles, as detailed in Table 2-8.
The derivation of the phase-in assumptions is discussed in the MOVES2014 heavy-duty exhaust report.38
h To estimate the impact of Tier 3 on the LHD2b3 emission rates, we assumed that all LHD2b3 diesel vehicles and
engines are chassis-certified although there are several 2b3 diesel engines produced by Cummins that are engine-
certified subject to the heavy-duty engine emission standards. All Class 4 and heavier engines are engine-certified and
therefore, not subject to the Tier 3 standards.
46
-------
Table 2-8 Phase-in Assumptions for Tier 3 NOx Standards for Light Heavy-Duty Diesel Vehicles
Model Year
Phase-in
fraction {%)
Reduction in Running
Emission Rate (%)1
Reduction in Start
Emission Rate (%)1
2017
0
0.0
0.0
2018
49
30.1
11.2
2019
62
38.1
14.2
2020
75
46.1
17.2
2021
87
53.5
19.9
2022
100
61.5
22.9
Note:
1 These reductions are based on comparison of Tier 3 standards against Tier 2 standards.
In generating the reduced rates for running operation, the starting point (or pre-Tier 3 baseline) are the
LHD2b3 rates for MY2017. The ending point, representing full Tier 3 control, was model year 2022. The MY
2022 rates were calculated by multiplying the rates for MY2017 by 0.3855. This fraction reflects the
percent reduction in running emission rates for MY2022 as shown in Table 2-8.
In addition to tightening emission standards, the Tier 3 regulations require an increase in the regulatory
useful life from 120,000 miles to 150,000 miles. An increase in the useful life is interpreted as an
improvement in durability, which is expressed in MOVES through a delay in deterioration effects. The
deterioration effects for the diesel LHD2b3 MY 2010 and later vehicles are derived from the tampering and
mal-maintenance effects discussed in the next section. To express this effect, the diesel LHD2b3 NOx
emission rates estimated for the 0-3 year age group are replicated to the 4-5 year age group, i.e., the onset
of deterioration is delayed two years until the 6-7 year age group (as shown in Figure 2-16). This effect is
phased-in for model years 2018-2021 and fully implemented for MY2022+. The reduction in grams per mile
from the Tier 3 rulemaking across model years is displayed in Figure 2-19.
47
-------
I
[
1
*
-V
. /
¦ *
*
¦ ¦
¦
~ ¦
,
¥
f
0
2
4
6
8
10
12
14
1
6
18
20
22
24
model_years
— 2015-2017
— ¦ 2018
— 2019
— 2020
¦ ¦ 2021
¦ ¦ 2022+
Vehicle age, years
Figure 2-16 NOx Emission Rates for Running Exhaust Operation in a Single Operating Mode (27) vs. Age for LHD2b3
Diesel Trucks for 2017 through 2022 and Later Model Years
2.1.1.6 Tampering and Mal-maintenance
MOVES accounts for the fleet-average increase in emissions with vehicle age. For heavy-duty diesel
vehicles, the increase in aging is modeled through tampering & mal-maintenance (T&M) adjustment
factors. We assume that the T&M adjustment factors are the dominant cause of emission increases with
age for heavy-duty diesel vehicles and that our estimated T&M adjustment factors also account for the
emission increases from normal aging and deterioration of properly maintained engines and exhaust
aftertreatment systems.
Table 2-9 shows the estimated aggregate NOx emissions percent increases due to T&M adjustment factors
by regulatory class and model year group. As described in Appendix B, the T&M adjustment factors in
Table 2-9 are calculated by combining information regarding the assumed frequency rate of an equipment
failure at the useful life of the engine with the estimated emission impact of the equipment failure. We
assume that emissions begin to deteriorate at the age vehicles pass the warranty requirements and
increase to the full T&M adjustment factors at the age the vehicles reach the useful life mileage
requirement (Figure B-l). Because the warranty period for the HHD, MHD, and Urban Buses are
anticipated to be reached within the first three years (Table B-l), the T&M adjustments factors are applied
starting with the age 0-3 group for these vehicles, and the LHD45 and LHD2b3 vehicles in the age 4-5
group.
48
-------
We account for the emission increases with age by multiplying the MY 2007-2009 LHD emission rates, and
MY 2010 and later emission rates by the corresponding T&M adjustment factors and scaled aged effects
(Equation 7-3). Both the 2007-2009 LHD emission rates and the MY 2010 rates are assumed to represent
zero-mile emission rates. The 2007-2009 LHD emission rates are derived from scaling emission factors by
certification level. The MY 2010 and later heavy-duty emission rates are based on the HDIUT data which
tested well-maintained vehicles. The MOVES emission rates for regulatory classes with the same zero-mile
emission rates (Table 2-10) are different due to the T&M NOx effects (Table 2-9) and phase-in of T&M
effects by age (Table B-3).
Table 2-9 NOxT&M adjustment factor (percent) by regulatory class
Model years
LHD2b3 trucks (%)
LHD45 trucks (%)
MHD, HHD, Urban Bus (%)
1994-1997
0
0
0
1998-2002
0
0
0
2003-2006
0
0
0
2007-2009
4
4
0
2010-2012
56
77
77
2013+
56
58
58
The T&M adjustments for NOx are zero for some of the model year groups because these vehicles lack the
heavy-duty advanced aftertreatment that we assume are most affected by tampering and
malmaintenance. The LHD vehicles have different T&M NOx increases than HHD, MHD, and Urban Bus
vehicles due to the 25 percent assumed penetration of lean NOx trap (LNT) aftertreatment within the
LHD2b3 vehicles beginning in MY 2007-2009, consistent with the assumptions previously made in Section
2.1.1.4.6. Certification data shows that LNT is not actually being used in LHD45 vehicles, however, we have
not updated this assumption used to derive the T&M adjustment factors on the rates because the resulting
LHD45 2007-2009 emission rates compare well to the HDIUT data for 2007-2009 vehicles.37
The T&M values for model year 2010 and later vehicles account for implementation of OBD. For LHD2b3
trucks, OBD systems were assumed to be fully implemented in MY 2010. For Class 4 through 8 trucks,
(LHD45, MHD, HHD), we assumed there would be a phase-in period from MY 2010 to 2012 where one-
third of those trucks were equipped with OBD systems. In MY 2013 and later, all trucks have OBD systems.
These OBD adoption rates have been incorporated into the tampering and mal-maintenance emission
increases in Table 2-9 based on the assumptions and calculations detailed in Appendix B.
Figure 2-17 displays the NOx rates in gram per mile by age and regulatory class for model year 2015
vehicles. Due to faster mileage accumulation, the HHD trucks tend to reach their useful life age quicker,
49
-------
with the maximum emission rates at the 4-5 age group, while the Urban Bus do not reach the useful life
until age 10-14 group. The gram per mile (g/mile) emission rates were calculated outside the model by
using operating mode distributions and average speeds for each regulatory class estimated from a national
MOVES run. The emission rates do not include any adjustments to the rates due to fuel effects or humidity
that are applied during MOVES run, however they do incorporate differences due to activity.1
Despite the significant T&M adjustment factors for MY 2010+ heavy-duty diesel vehicles, the fully-aged
NOx emissions of MY 2010+ vehicles are significantly lower than the emission rates of MY 2009 and earlier
vehicles. As shown in Figure 2-19, the age 0-3 MY 2009 NOx g/mile emissions are over 100% higher than
the MY 2010 vehicles, while the T&M adjustment factors are less than 100% (Table 2-9). Thus, MOVES
estimates that, on average, a fully-aged MY 2010 heavy-duty truck has lower NOx emissions than a new
MY 2009 heavy-duty truck.
' For example, the MY 2015 NOx emission rates by operating mode for LHD2b3 and LHD45 diesel vehicles are the
same, but Figure 2-17 shows different gram per mile emission rates due to vehicle and activity differences, including
heavier weights of LHD45 vehicles (source mass), and because the two regulatory classes are distributed differently
among the source types in MOVES, which have different operating mode distributions. See the MOVES3 Population
and Activity Report9 for more information. The zero-mile HHD and Urban Bus NOx emission rates by operating mode
are equivalent. However, the emission rates in the age groups: 0-3, 4-5, 6-7, and 8-9 are different because they have
a different phase-in of the T&M effects. The difference between HHD and Urban Bus in grams per mile for ages 10+
are due to differences in the operating mode distributions and average speeds.
50
-------
3-
Reg Class
— HHD8
Urban Bus
— MHD67
LHD45
LHD2b3
0-3 4-5 6-7 8-9 10-14 15-19 20+
Age (years)
Figure 2-17 Heavy-duty Diesel NOx Emission Rates (g/mile) by Age for Model Year 2015 by Regulatory Class
Estimated using Nationally Representative Operating Mode Distribution
2.1.1.7 Summary
Table 2-10 summarizes the methods used to estimate emission rates for each regulatory class and model
year group combination. The emission rates in MOVES are based on the analysis of ROVER, Consent Decree
testing data, and HDIUT data. Using the HDIUT data, we updated the HHD, MHD, LHD, and Urban Bus rates
for MY 2010+ vehicles in MOVES3.
51
-------
Table 2-10 Summary of Methods for Heavy-Duty Diesel NOx Emission Rate Development for Each Regulatory Class
and Model Year Group
Model
year group
HHD
(regClass 47)
MHD
(regClass 46)
Urban Bus
(regClass 48)
LHD45
(regClass 42)
LHD2b3
(regClass 41)
1960-
1989,
1990
HHD 1991-1997
rates
proportioned to
ratio of
certification
levels
Same rates as
HHD
Urban Bus 1991-
1997 rates
proportioned
using ratio of
HHD certification
levels
LHD 1991-1993 rates proportioned to
LHD certification levels
1991-1997
Data analysis3 0,
with
adjustments for
Low-NOx
rebuilds
Same rates as
HHD
Data analysis3
LHD 1999-2002 rates proportioned to
1991-1997 FTP standards per Table 2-1
1998
Data analysis3 0,
with
adjustments for
Low-NOx
rebuilds
Same rates as
HHD
Urban Bus 1999-
2002 rates
proportioned
using ratio of
HHD 1998 rates
to HHD 1999-
2002 rates
Same rates as 1999-2002
1999-2002
Data analysis3 0
Data analysis3
Data analysis3
MHD data analyzed with 2.06 fSCaie
2003-2006
Data analysis3 0
Data analysis3 0'
Data analysis3
Data analysis b
2007-2009
Data analysis'5
MHD 2003-
2006 rates
proportioned to
FTP standards
per Table 2-1°
Urban Bus 2003-
2006 rates
proportioned to
FTP standards per
Table 2-1
Percent reductions from the
MOVES2010 LHD 2003-2006 rates
(Section 2.1.1.4.6)°
2010-
2018
HHD data
analysis'5 with
MY specific
production
volume
weighting
MHD data
analysis'5 with
MY specific
production
volume
weighting
Same as HHD
except T&M
adjustment
factors specific to
Urban Bus
LHD data analysis'5
with MY specific
production
volume weighting;
T&M specific to
LHD45
LHD data analysis'5
with MY specific
production
volume weighting;
T&M specific to
LHD2b3 & Tier 3
reductions starting
in MY 2018
52
-------
Model
year group
HHD
(regClass 47)
MHD
(regClass 46)
Urban Bus
(regClass 48)
LHD45
(regClass 42)
LHD2b3
(regClass 41)
2019-2060
Same as HHD
MY 2018
Same as MHD
MY 2018
Same as Urban
Bus MY 2018
Same as LHD45
MY 2018
Same baseline as
LHD2b3 MY 2018
with Tier 3
reductions phase-
in from MY 2019-
2022
Notes:
a Analysis based on ROVER and Consent Decree testing data
b Analysis based on HDIUT data
c Confirmed by HDIUT and/or Houston Drayage Program data
The role of the model year groups, representing a rough surrogate for technology or standards, can be
seen in Figure 2-19, which shows NOx emission rates for 0-3 age group by model yearand regulatory class
estimated in grams per mile (g/mile) using nationally representative operating mode distributions and
average speeds. MOVES model estimates drastic reductions in NOx emissions from pre-1990 technologies
to modern 2018 and later model year engines. Some of the differences in the emission rates for the 0-3
age group are caused by different phase-in of T&M adjustment factors, which began to be applied in the
first age group for HHD, MHD, and Urban Bus regulatory classes (Section 2.1.1.5.5). Because the nationally
representative operating modes are different across model years and regulatory classes, some of the
differences between model years and the regulatory classes are due to activity differences and not the
emission rates by operating mode.1
The figure also shows NOx emissions for "gliders", which are trucks with a new chassis and cab assembly
and equipped with a rebuilt engine typically without an exhaust aftertreatment system. These emissions
are of similar magnitude to emission rates from the model year 2000 HHD vehicles as discussed in Section
2.5.
Figure 2-19 shows the model year trend for NOx in gram per mile for the 2007 and later model years. The
reduction in LHD2b3 emission rates due to the phase-in of the Tier 3 standards between 2018 and 2022 is
evident. All other heavy-duty emission rates by operating mode are unchanged beginning in model year
2018 as documented in Table 2-10. The slight reductions observed in the gram per mile emission rates for
all heavy-duty vehicles (including gliders) in model years 2014 through 2027 are due to changes in the
operating mode distribution from lower weights, lower rolling resistance, and improved aerodynamics of
trucks implemented in the Phase 1 and Phase 2 Heavy-duty Greenhouse Gas Regulations as documented in
the MOVES3 Population and Activity Report.9
53
-------
5
0
>
03
a: 20-
c
o
C/3
tt>
LU
o 10"
o-
1970
1980
1990
2000 2010
Model Year
2020
2030
Reg Class
Gliders
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
Figure 2-18 Heavy-duty Diesel NOx Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions
Figure 2-19 Heavy-duty Diesel NOx Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions for Model Year 2007 through 2030
Reg Class
— HHD8
Urban Bus
MHD67
LHD45
— LHD2b3
2010
2015 2020
Model Year
2025
2030
54
-------
2.1.2 Particulate Matter (PM2.s )
PM2.5 refers to particulate matter with a mean aerodynamic diameter less than 2.5 microns. Particulate
matter is a complex mixture of particles that are composed of one or many of the following substances:
organic material, metals, elements, and ions, including sulfate. Measurements from which the MOVES
PM2.5 emission rates are based on are typically filter-based, which include the mass of all the chemical
components in the particle-phase. MOVES reports PM emissions in terms of elemental carbon (EC) and the
remaining non-elemental carbon PM (nonECPM).
As described above for NOx, the heavy-duty diesel PM2.5 emission rates in MOVES are a function of: (1)
fueltype, (2) regulatory class, (3) model year group, (4) operating mode, and (5) age group. We classified
heavy-duty diesel exhaust PM emission data into the following model year groups for purposes of emission
rate development. These groups are generally based on the introduction of emissions standards for heavy-
duty diesel engines. They also serve as a surrogate for continually advancing emission control technology
on heavy-duty engines. For example, MY 2010 and beyond is defined as a separate group even though the
PM standard is unchanged from the previous group MY 2007-2009. This is because, starting with MY 2010,
the wide adoption of SCR systems and improvements in DPFs likely resulted in a shift in tailpipe
PM2.5emissions. Other, secondary reasons include wider availability of PM2.5data in the HDIUT data set
(section 2.1.1.1) for MY 2010-2011 and the/scofe updates to all HD regClasses for MY 2010 and later (see
section 2.1.1.4.2 and Appendix G). Table 2-11 shows the model year group ranges and the applicable
brake-specific emissions standards.
Table 2-11 Model Year Groups Used for Analysis Based on the PM Emissions Standard
Model Year Group
Range
PM Standard (g/bhp-hr)
1960-1987
No transient cycle
standard
1988-1990
0.60
1991-1993
0.25
1994-1997
0.10
1998-2006
0.10
2007-2009
0.01
2010+
0.01
Section 1.2.1 and Section 2.1.2.2 discuss the derivation of the MY 1960-2009 and 2010-2060 heavy-duty
diesel PM2.5 emission rates, respectively. The discussion of the tampering and mal-maintenance factors
applied to emissions in both model year groups are discussed in Section 2.1.2.1.
55
-------
2.1.2.1 1960-2009 Model Years
The PM2.sdata from was analyzed in several steps to obtain MY 1960-2009 PM2.5 emission rates. First, STP
operating mode bins were calculated from the chassis dynamometer data from the CRC E55/59. Second,
continuous PM2.5 data measured by the TEOM was normalized to gravimetric PM filters. Third, MOVES
PM2.5 emission rates were calculated for the STP operating mode bins for the available regulatory class and
model year combinations. Then, steps were taken to estimate missing operating modes, regulatory classes,
and model years from the E55/59 program. In addition, we estimate the EC/PM fraction and adjust the
emission rates to account for tampering and mal-maintenance. The E55/59 data and analysis steps are
explained in detail in the following subsections.
2.1.2.1.1 Data Sources
All of the data used to develop the MOVES PM2.5 emission rates for MY 1960-2009 was generated based on
the CRC E-55/59 research program.42 The following description in the "Compilation of Diesel Emissions
Speciation Data - Final Report"43 provides a good summary of the program:
The objective of the CRC E55/59 test program was to improve the understanding of the California
heavy-duty vehicle emissions inventory by obtaining emissions from a representative vehicle fleet,
and to include unregulated emissions measured for a subset of the tested fleet. The sponsors of
this project include CARB, EPA, Engine Manufacturers Association, DOE/NREL, and SCAQMD. The
project consisted of four segments, designated as Phases 1, 1.5, 2, and 3. Seventy-five vehicles
were recruited in total for the program, and recruitment covered the model year range of 1974
through 2004. The number and types of vehicles tested in each phase are as follows:
• Phase 1: 25 heavy heavy-duty (HHD) diesel trucks
• Phase 1.5: 13 HHD diesel trucks
• Phase 2: 10 HHD diesel trucks, 7 medium heavy-duty (MHD) diesel trucks,
2 MHD gasoline trucks
• Phase 3: 9 MHD diesel, 8 HHD diesel, and 2 MHD gasoline
The vehicles tested in this study were procured in the Los Angeles area, based on model years
specified by the sponsors and by engine types determined from a survey. WVU measured
regulated emissions data from these vehicles and gathered emissions samples. Emission samples
from a subset of the vehicles were analyzed by Desert Research Institute for chemical species
detail. The California Trucking Association assisted in the selection of vehicles to be included in this
study. Speciation data were obtained from a total of nine different vehicles. Emissions were
measured using WVU's Transportable Heavy-Duty Vehicle Emissions Testing Laboratory. The
laboratory employed a chassis dynamometer, with flywheels and eddy-current power absorbers, a
full-scale dilution tunnel, heated probes and sample lines and research grade gas analyzers. PM
was measured gravimetrically. Additional sampling ports on the dilution tunnel supplied dilute
56
-------
exhaust for capturing unregulated species and PM size fractions. Background data for gaseous
emissions were gathered for each vehicle test and separate tests were performed to capture
background samples of PM and unregulated species. In addition, a sample of the vehicles received
Tapered Element Oscillating Microbalance (TEOM) measurement of real time particulate
emissions.
The HHDDTs were tested under unladen, 56,000 lb., and 30,000 lb. truck load weights. The driving
cycles used for the HHDDT testing included:
• AC50/80;
• UDDS;
• Five modes of an HHDDT test schedule proposed by CARB: Idle, Creep, Transient, Cruise,
and HHDDT_S (a high-speed cruise mode of shortened duration);
• The U.S. EPA Transient test.
The CARB HHDDT test cycle is based on California truck activity data and was developed to
improve the accuracy of emissions inventories. It should be noted that the transient portion of this
CARB test schedule is similar but not the same as the EPA certification transient test.
The tables below provide a greater detail on the CRC E-55/59 data used in the analysis. Both the number of
tests and number of vehicles by model year group and regulatory class (MHD, HHD) are provided in Table
2-12.
Table 2-12 Vehicle and Test Counts by Regulatory Class and Model Year Group
Regulatory
Model Year
Number of
Number of
Class
Group
tests
vehicles
1960 - 1987
82
7
1988 - 1990
39
5
MHD
1991 - 1993
22
2
1994 - 1997
39
4
1998 - 2006
43
5
2007 - 2009
0
0
1960 - 1987
31
6
1988 - 1990
7
2
HHD
1991 - 1993
14
2
1994 - 1997
22
5
1998 - 2006
171
18
2007 - 2009
0
0
Counts of tests are provided by test cycle in Table 2-13.
57
-------
Table 2-13 Vehicle Test Counts by Test Cycle
Test Cycle
Number of tests
CARB-T
71
CARB-R
66
CARB-I
42
UDDS_W
65
AC5080
42
CARB-C
24
CARBCL
34
MHDTCS
63
MHDTLO
23
MHDTHI
24
MHDTCR
29
2.1.2.1.2 Calculate STPfrom Second-By-Second Data
For each second of operation on the chassis-dynamometer the instantaneous scaled tractive power (STPt)
was calculated using Equation 1-6, and then subsequently classified to one of the 23 operating modes
defined above in Table 1-4.
The values of coefficients A, B, and C are the road-load coefficients pertaining to the heavy-duty vehicles as
determined through previous analyses for EPA's Physical Emission Rate Estimator (PERE).44 The chassis
dynamometer cycles used in E55/59 include the impact of speed, acceleration, and loaded weight on the
vehicle load, but grade effects are not included and the grade value is set equal to zero in Equation 1-6.
Note that this approach differs from the NOx emission rates analysis described in Section 2.1.1.3, since the
particulate data was collected on a chassis dynamometer from vehicles lacking electronic control units
(ECU). We have not formally compared the results of the two methods of calculating STP. However, on
average, we did find the operating-mode distributions to be similar between the two calculation methods
for a given vehicle type. For example, we found that the maximum STP in each speed range was
approximately the same.
2.1.2.1.3 Compute Normalized TEOM Readings
The TEOM readings were obtained for a subset of tests in the E-55/59 test program. Only 29 vehicles had a
full complement of 1-Hz TEOM measurements. However, the continuous particulate values were modeled
for the remaining vehicles by West Virginia University, and results were provided to EPA. In the end, a total
of 56 vehicles and 470 tests were used in the analysis out of a possible 75 vehicles. Vehicles and tests were
58
-------
excluded if the total TEOM PM2.5 reading was negative or zero, or if corresponding full-cycle filter masses
were not available. Table 2-14 provides vehicle and test counts by vehicle class and model year. The MHD
(Class 6 and Class 7) trucks tested in the study included seven Class 6 and eleven Class 7 vehicles,
representing only a limited model years.
Table 2-14 Vehicle and Test Counts by Heavy-Duty Class and Model Year
Model Year
HDD Class 6/7(MHD)
HDD Class 8 (HHD)
No. Vehicles
No. Tests
No. Vehicles
No. Tests
1969
-
-
1
6
1974
1
10
-
-
1975
-
-
2
10
1978
-
-
1
5
1982
1
5
-
-
1983
1
10
1
6
1985
1
28
1
10
1986
1
3
1
4
1989
11
1
4
1990
1
12
1
3
1992
1
11
1
11
1993
1
11
1
3
1994
1
9
3
15
1995
2
24
3
13
1998
2
20
3
28
1999
-
-
3
43
2000
2
18
5
44
2001
1
5
2
21
2004
-
-
4
29
2005
-
-
1
6
Since the development of MOVES emission rates is cycle independent, all available cycles/tests which met
the above requirements were utilized. As a result, 488,881 seconds of TEOM data were used. The process
required that each individual second-by-second TEOM rate be normalized to its corresponding full-cycle
filter mass, available for each combination of vehicle and test. This step was necessary because individual
TEOM measurements are highly uncertain and vary widely in terms of magnitude (extreme positive and
negative absolute readings can occur). Kinsey et al. (2006)45 demonstrated that time-integrated TEOM
measurements compare well with gravimetric filter measurements of diesel-generated particulate matter.
Equation 2-16 shows the normalization process for a particular one second TEOM measurement.
P ^filter j
^normalized,!,; =V/D,, —PMjE0M,j,i Equation 2-16
2/ r™TEOM,i
Where:
59
-------
i = an individual 1-Hz measurement (g/sec),
j = an individual test on an individual vehicle,
PMteomjj = an individual TEOM measurement on vehicle j at second /',
PMfnterj = the total PM2.5 filter mass on vehicle j,
P M normalized, ij = an estimated continuous emission result (PM25) emission result on vehicle j at
second /'.
2.1.2.1.4 Compute Average Rates by MOVES Operating Mode
After normalization, the data were classified into the 23 operating modes defined in Section 1.6 by
regulatory class and model-year group. Mean average results, and standard deviation for PM2.5 emission
values were computed in terms of g/hour for each operating mode. In cases where the vehicle and TEOM
samples were sufficient for a given mode (based on the number of points within each operating mode bin),
these mean values were adopted as the MOVES emission rates for total PM2.5.
2.1.2.1.5 Populating Missing and High-Power Operating Modes
As detailed in Appendix E, a log-linear regression was performed on the existing PM2.5 data against STP to
fill in emission rates for missing operating mode bins. Similar to the NOx rates for MY 2009 and older
vehicles, emission rates were extrapolated for the highest STP operating modes.
2.1.2.1.6 LHD and Urban Bus Emission Rates
The PM2.5 emission rates for LHD and Urban Buses are based on the available TEOM data collected on MHD
and HHD vehicles. We believe this is reasonable because the certification standards in terms of brake
horsepower-hour (bhp-hr) are the same for LHD, MHD, and HHD regulatory classes.
The following steps were conducted to adjust the emissions estimated from the MHD and HHD regulatory
classes because the data were not analyzed for the/scofe.used for LHD. The emission rates of pre-2010 LHD
(LHD2b3 and LHD45 (regClassID 41 and 42) are based on an fSCaie of 2.06 as discussed in Section 1.5,
whereas MHD and HHD are based on an fSCaie of 17.1. The PM2.5 emission rates for the pre-2010 LHD
regulatory classes are derived from the VSP-based MHD PM2.5 emission factors derived from the E55/59
TEOM data as analysed for MOVES2009.46
With VSP-based emission rates, the power of the vehicle is scaled to the mass of the individual tested
vehicle. Because LHD have lower vehicle weights and power outputs than the MHD and HHD vehicles, we
60
-------
scaled the VSP-emission rates down to the power requirements of the LHD vehicles. To estimate the
LHD2b3 and LHD45 PM2.5 emission rates, we multiplied the VSP-based MHD PM2.5 emission rates by a
factor of 0.46 obtained from the MOBILE6.2 heavy-duty conversion factors47, which accounts for the lower
power requirements per mile (bhp-hr/mile) of light heavy-duty trucks versus MHD trucks. This scaling
factor estimates VSP-based emissions rates for LHD vehicles. We are approximating the STP-based MOVES
emission rates for LHD vehicles using these VSP-based rates.J Equation 2-17 used to derive the PM2.5
emission rates for LHD regulatory class is shown below:
LHDPM2 5 emission rate = 0.46 X MHD (VSPbased)PM2,5 emission rate Equation 2-17
Despite the uncertainty used in this approach, the representative PM2.5gram per mile emission rates show
reasonable trends (see Figure 2-27). LHD2b3 and LHD45 emission rates for pre-2010 model years are lower
than the MHD and HHD emission rates, but higher than the Urban Bus emission rates.
Urban Bus emission rates are generally assumed to be the same as the HHD emission rates. However, for
some model year groups, the Urban Buses are subject to a different emission standard, so we adjust the
emission rates by applying a ratio of the EPA certification standards. Table 2-15 displays the model years
for which the Urban Bus regulatory class has different PM emission standards from other heavy-duty
compression-ignition engines. For these model years (1991-2006), the Urban Bus PM emission rates are
equal to the HHD emission rates multiplied by the ratio in emission standards. In addition, the Urban Buses
have different emission deterioration effects as discussed in Appendix B.l.The gram per mile emission
rates for Urban Bus presented in Figure 2-27 show expected trends, with lower emission rates than the
other regulatory classes starting in MY 1991 through 2006.
Table 2-15 Urban Bus PM Standards in Comparison to Heavy-Duty Highway Compression Engine Standards
Engine
Model Year
Standard for Heavy-
Duty Highway
Compression-Ignition
Engines (g/bhp-hr)
Standard for
Urban Buses
(g/bhp-hr)
Ratio in
standards
1991-19933
0.25
0.1
0.4
1994-1995
0.1
0.07
0.7
J When this approximation was conducted in MOVES2010, the/Sca/e of 2.06 matched the average mass of LHD vehicles
for the source types. This approximation need to be revisited now that we have updated the mass of LHD vehicles in
MOVES3 to range from 3.5 to 7.8 metric tons.
61
-------
1996-2006
0.1
0.05
0.5
Note:
a The 0.1 g/bhp-hr US EPA Urban Bus standard began with model year 1993. In California, the 0.1 g/bhp-hr Urban Bus
standard began in 1991. MOVES assumes all Urban Buses met the stricter CA standard beginning in 1991.
2.1.2.1.7
Model Year 2007-2009 Vehicles (with Diesel Particulate Filters)
The heavy-duty diesel emission regulations were made considerably more stringent for PM2.5 emissions
starting in model year 2007 - even considering the phase-ins and average banking and trading program,
the emission standard fell by a factor of ten from 0.1 g/bhp-hr to 0.01 g/bhp-hr. This increase in regulatory
stringency required the use of diesel particulate filter (DPF) systems on heavy-duty diesels. As a result, the
PM2.5 emission performance of diesel vehicles has improved dramatically.
At the time of analysis (originally done for MOVES2014, but carried over to MOVES3 for the 2007-2009
vehicles), no continuous PM2.5 emissions data were availableon the 2007-2009 model-year vehicles.
However, heavy and medium heavy-duty diesel PM2.5 data were available from the EPA engine certification
program on model years 2003 through 2007. These data provided a snapshot of new engine emission
performance before and after the introduction of particulate trap technology in 2007 and made it possible
to determine the relative improvement in PM2.5 emissions from model years 2003 through 2006 to model
year 2007. This same relative improvement was applied to the existing, operating mode-based, 1998-2006
model year PM2.5 emission rates to estimate in-use rates for MY 2007-2009 vehicles.
An analysis of the certification data is shown in Table 2-16 below. It suggests that the particulate trap
reduced PM2.5 emissions by a factor of 27.7. This factor is considerably higher than the relative change in
the certification standards (i.e., a factor of 10). The reason for the difference is that the new trap-equipped
vehicles certify at emission levels which are much lower than the standard, and thus, create a much larger
compliance marginthan previous technologies could achieve.
As an additional check on the effectiveness of the trap technology, EPA conducted some limited in-house
testing of a Dodge Ram truck, and carefully reviewed the test results from the CRC Advanced Collaborative
Emission Study (ACES) phase-one program, designed to characterize emissions from diesel engines
meeting 2007 standards. The results from these studies demonstrated that the effectiveness of working
particulate traps is very high.48
Table 2-16 Average Certification Results for Model Years 2003-2007
Certification Model Year
Mean
St. Dev.
n
62
-------
(g/bhp-hr)a
2003
0.084
0.014
91
2004
0.088
0.013
59
2005
0.085
0.014
60
2006
0.085
0.014
60
2007
0.003
0.002
21
Note:a Average ratio from MYs 2003-2006 to MY 2007 is 27.7.
2.1.2.1.8 Elemental Carbon and Non-Elemental Carbon Emission Factors
Particulate matter from conventional (pre-2007) diesel engines is largelycomposed of elemental carbon.
Elemental carbon is often used synonymously with soot and black carbon. Black carbon is important
because of its negative health effects and its environmental impacts as a climate forcer.49 Elemental
carbon from vehicle exhaust is measured with filter-based measurements using thermal optical methods.
Continuous surrogate measures of elemental carbon can also be made with photoacoustic instruments.
MOVES models Total PM2.s emissions by vehicle operating mode using elemental carbon (EC) and non-
elemental particulate matter carbon (NonECPM), as shown in Equation 2-18.
PM2.5 = EC + NonECPM Equation 2-18
By having emission rates for EC and nonECPM for each operating mode, the MOVES design permits the EC
fraction of PM (EC/PM) to vary for each operating mode. In practice, the data used to develop EC and
nonECPM emission rates does not support such fine resolution, and the EC/PM is the same across all the
running exhaust operating modesexcept for the idle operating mode.
For pre-2007 diesel trucks, we developed EC and nonECPM emission rates by applying EC/PM fractions to
the modal-based emission rates. For the idle operating mode (opModelD 1), we applied an EC/PM fraction
of 46.4 percent from the PM25 speciation profile developed from the idle mode from the UDDS tests from
the E55/59 program. For all the other operating modes within the running emission process, we used an
EC/PM fraction of 79.0 percent from the PM25 speciation profile developed from the transient mode of the
UDDS tests from the E55/59 program. The development of the pre-2007 PM25 speciation profiles from the
E55/59 program are documented in the Onroad Speciation Report.1
63
-------
For 2007-2009 DPF-equipped diesel engines, we used the EC/PM fraction of 9.98 percent measured in
Phase 1 of the Advanced Collaborative Emissions Study (ACES) Report.48 Diesel particulate filters
preferentially reduce elemental carbon emissions, resulting in the low percentage of elemental carbon
emissions. The average EC/PM fraction is based on four engines run on the 16-hour cycle which composes
several different operating cycles. Because the fraction is based upon a range of driving conditions, we
applied the constant 9.98 percent EC/PM fraction across all operating modes for the 2007+ diesel
emissions rates, including the idle operating mode (opModelD 1).
Figure 2-20 and Figure 2-21 show the operating mode trend for PM2.5 with the EC and nonECPM fractions
As with NOx, the highest operating modes in each speed range will rarely be attained due to the power
limitations of heavy-duty vehicles and the high fscaie used for these model years, but are included in the
figures for completeness. At high speeds (greater than 50 mph; operating modes > 30), the overall PM2 5
rates are lower than the other speed ranges. For pre-2007 model years (Figure 2-20), the PM2.5 rates are
dominated by EC (except for the idle operating mode, opModelD 1). With the introduction of DPFs in
model year 2007 (Figure 2-21), we model the large reductions in overall PM2.5 rates and the smaller
relative EC contribution to PM emissions. Figure 2-29 shows the PM2.5 gram per mile emission rates
separated into the elemental carbon and non-elemental carbon fractions for HHD vehicles across model
years.
1 EC
InonEC
Mil
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 2-20 MHD Diesel PM2.5 Emission Rates for MY 2006 (age 0-3) by Operating Mode
64
-------
1.4
1.2
M
5! 0.8
0.6
c
TO
CU
0.4
0.2
I EC
InonEC
Mil
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 2-21 MHD Diesel PM2.5 Emission Rates for MY 2007 (age 0-3) by Operating Mode
2.1.2.2 2010 and Later Model Years
The MY 2010+ HDIUT data set described in Section 2.1.1.1 and Table 2-2 was used to update PM2.5
emissions rates for MY 2010+ vehicles. Operating modes (Table 1-4) were assigned to the 1 hz data using
the method to calculate STP described in Section 2.1.1.3, and the updated fscaie values developed for the
2010+ MY NOx analysis described in Section 2.1.1.4.2 and Appendix G.
2.1.2.2.1 Estimating Base Rates
As compared to the NOx, THC, and CO data, the PM2.5 emissions data reported from the HDIUT program
had many missing, negative and zero values. When the PM2.5 emissions data are distributed over the NOx
FEL groups, the number of vehicles with valid measurements (non-zero or non-negative) by regulatory
class and model year group was very limited as shown in Table 2-17. The number of vehicles with valid
measurements for each model year group and regulatory class are significantly smaller than the total
number of vehicles tested, with the exception of the HHD 2014-2015 model year group.
65
-------
Table 2-17. HDIU Vehicles with Valid PM2.5 Measurements By Regulatory Class, Model Year Group and NOx FEL
Group
Valid Measurements by NOx FEL
Group
Total
Valid
Total
Vehicles
Tested
Reg
Class
Model Year
Group
0.2
0.35
0.5
LHD
2010-2013
52
0
4
56
64
2014-2016
19
0
2
21
32
MHD
2010-2013
20
6
6
32
55
2014-2015
17
61
61
29
51
HHD
2010-2013
57
9
26
92
139
2014-2015
48
4
0
52
55
Note:
1 Due to an absence of MHD 2014-2015 vehicles certified to 0.35 and 0.5 NOx NOx in the HDIUT, we replicated the
HDIUT MHD 2010-2013 vehicles certified to 0.35 and 0.5 NOx FEL to the 2014-2015 model year group.
When evaluating the data by operating mode, the data are even more sparse. Table 2-17 shows the total
number of vehicles tested by model year group and regulatory class that had valid (positive) PM2.5
emissions data in at least one operating mode. However, the number of vehicles with valid measurements
is less when the data is evaluated by operating mode. Table 2-18 shows the minimum and maximum range
of valid vehicle measurements by individual operating modes. The operating modes with the
smallestnumber of valid measurements tended to be for the high-power operating modes (opModelD 29,
30, 39 and 40). The highest number of valid measurements tend to be for the high speed (>50 mph) at
moderate power (opModelD 33, 35, 37, and 38).
We addressed the issue of data sparsity in the PM25 emissions data by not using NOx FEL based grouping
and production volume weighting. By aggregating the data across NOx FEL groups, there was sufficient data
to divide the data into the two model year groups used for the NOx analysis: MY 2010-2013 and 2014-
2015/2016. Note that for the current HDIUT dataset, LHD includes MY 2016 vehicles, but MHD and HHD
only include up to MY 2015.
66
-------
Table 2-18. Range of Valid Vehicle Measurements by Operating Mode by Regulatory Class and Model Year Group
Model Year
Group
Total Valid Vehicle
Measurements
(in at least one
operating mode)
Minimum Valid
Vehicle
Measurements for
any operating
mode
Maximum Valid Vehicle
Measurements for any
operating mode
LHD
2010-2013
56
11
52
2014-2016
21
3
18
MHD
2010-2013
32
4
27
2014-2015
29
3
26
HHD
2010-2013
92
17
83
2014-2015
52
4
45
The operating mode-based PM2.5 emission rates were estimated using Equation 2-19 and Equation 2-20,
which are similar to Equation 2-13 and Equation 2-14 used for the development of NOx emission rates
except the FEL grouping factor is removed. Zero or negative emission rates are as treated as missing
values. Despite the sparseness of the data by operating mode, due to the revised fscaie factors for 2010 and
later vehicles (2.1.1.5.4), in general, there was no need for estimating rates for missing high power
operating modes.k
^isec ERpol,OM,C,MYG,veh,sec .- .. „ _
ERpoi,OM,c,MYG,veh = Equation 2-19
S€Ccourlt
^'iveh ERpol,OM,C,
ERPoI,om,c,myg = '—""-veh Equation 2-20
vehCOunt
Where:
C = Regulatory class (LHD, MHD, HHD, and Urban Bus)
k The MHD 2010-2013 vehicles were an exception, missing data in operating modes 29, 30, and 40. Using the methods
outlined in Appendix 1.3 the emission rates in operating mode 29 and 30 were set equal to 28, and 40 was set equal to
39.
67
-------
ERx,y,z = Emission rate in mass/time. The subscripts show the categorization
MYG
= Model year group (2010-2013 or 2014-2016)
MY
= Model year
OM
= running exhaust emissions operating mode
pol
= Pollutant (PM2.5)
sec; seCcount = a second of data (for a given 1/eh and OM); number of seconds in that category
veh; vehcount = a vehicle (in the class); number of vehicles in that category
Similar to NOx, the PM2.5 rates for LHD2b3 and LHD45 are identical and based on the combined LHD class
vehicles in HDIUT, while the MHD rates are based on MHD class vehicles, and HHD and Urban Bus rates are
based on HHD class vehicles.
Because we did not use production volume weighting by NOx FEL, the PM2 5 emission rates are the same
for each model year within the two model year groups. The 2014-2015 model year group sample contains
a higher penetration of HHD vehicles certified to the 0.2 NOx FEL group, than the 2010-2013 model year
HHD group (Table 2-17) which is consistent with the trend in production volumes (Figure 2-12). The LHD
vehicles had a similar proportion of vehicles certified to the 0.2 and 0.35 NOx FELs within the 2010-2013
and 2014-2016 model year groups. The PM2.5 emission rates decrease for each of the regulatory classes
between the MY 2010-2013 and 2014 and later model years as observed in Figure 2-28.
There was no valid HDIUT PM data for 2014-2015 MY MHD vehicles within the 0.35 and 0.50 FEL groups.
Because there are 2014-2015 MY MHD vehicles certified to these higher levels in the production volume
data (Figure 2-12), we supplemented the 2014-2016 MY MHD data with data from the MY 2010-2013 MHD
vehicles in the 0.35 and 0.5 FEL groups. As such, the MHD vehicles have a similar proportion of vehicles
certified to the 0.35 and 0.5 NOx FEL Groups within the 2010-2013 and 2014-2015 model year groups.
Figure 2-22 through Figure 2-24 display the average PM2.5 emission rates by regulatory class, model year
group, and operating mode. The error bars are the 95% confidence intervals of the mean, calculated by
treating the mean emission rates from each vehicle within each operating mode as an independent
random variable. While the average PM2.5 emission rates by individual operating mode bin and regulatory
class generally are not significantly different between the 2010-2013 and 2014-2016 model year groups,
when you look across the operating modes, there is a fairly consistent decrease in PM2.5 emission rates in
the 2014-2016 compared to the 2010-2013 model year groups. The observed decrease in PM2.5 emission
68
-------
rates in the 2014+ model year emission rates is consistent with the decrease in extended idle PM2.5
emissions observed with the full-phase in of SCR engines' (Section 2.3.2.2).
0.60 q
0.50 ^
0.40 j
:
W)
E
0.30 ^
-
0.
0.20 -
0.10 ^
0.00 :
I MY 2010-2013
I MY 2014-2016
i
i ll I
I I 1.1.1,
Lii
i ii ii
0 1 1 11 12 13 14 15 16 ! 21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 2-22 PM2.s Emissions by Operating Mode for LHD Model Year Groups 2010-2013 and 2014-2016
1 As discussed in Section 2.3.2.2, we believe the reduction in THC and PM2.5 with the SCR systems is because the SCR
aftertreatment classification is a surrogate for the combined engine control and aftertreatment system used with SCR
equipped trucks that have a large impact on THC emissions. With the use of SCR, engines can be calibrated to run
leaner, producing lower engine-out PM2.5- Additionally, SCR systems rely on oxidation catalysts and/or catalyzed DPFs
to convert NO to N02, which also reduces PM25 tailpipe emissions.
69
-------
0.80 :
0.70 :
0.60 :
^ 0.50 :
00
E, 0.40
| 0.30 :
0.20 :
0.10 :
0.00
Li
I MY 2010-2013
l MY 2014-2015
1,1. LI
0 1
11 12 13 14 15 16
21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 2-23 PM2.5 Emissions by Operating Mode for MHD Model Year Groups 2010-2013 and 2014-2015
-il.
Ji
I MY 2010-2013
1 MY 2014-2015
n ¦ I IS
0 1 11
12 13 14 15 16
21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 2-24 PM2.5 Emissions by Operating Mode for HHD Model Year Groups 2010-2013 and 2014-2015
As observed in Figure 2-22 through Figure 2-24, the PM2.5 emission rates are highest for the MHD vehicles,
followed by the HHD, and then the LHD vehicles. The corresponding uncertainty of the MHD rates is also
largest compared to the other rates. We do not have a reason to suspect the accuracy of the HDIUT PM2.5
measurements from the MHD vehicles, and we can only speculate on the reasons the MHD emission rates
would be higher than the other regulatory classes. For example, the fraction of vehicles produced in the
70
-------
less stringent 0.35 and 0.5 NOx FEL groups is higher for the MHD than most of the other regulatory class
and model year combinations, but not all (Table 2-17).
Because much of the HDIUT PM data are missing or reported as zero, and given the additional uncertainty
regarding the MHD rates, we compared our HDIUT-based PM2.5 rates against values reported in the
literature. As shown in Figure 2-28, the MOVES age 0-3 PM2.5 rate ranges from 3 to 26 mg/mile for a MY
2010-2014 HD vehicle and from 2 to 7 mg/mile for MY 2014+. Other studies have reported PM2.5 rates in
the range of 1-7 mg/mi for MY 2010+ vehicles equipped with DPF and SCR and certified to NOx standard of
0.20 g/bhp-hr.50'51,52 The rates from the MOVES run and other studies are dependent on driving cycle,
however, since the MOVES rates are generally within the range of reported values, we believe it is
reasonable to use the HDIUT-based PM2.5 data for the update for all regulatory classes.
The MY 2014-2016 emission rates are applied to all future model yearsfor each regulatory class. The Tier 3
rulemaking sets PM FTP emission standards for Class 2b and Class 3 of 8 mg/mile and 10 mg/mile
respectively.53 Because the age 0-3 LHD2b3 MOVES emissions rates are well below the Tier 3 standard, we
do not estimate a reduction in PM2.5 rates with the phase-in of Tier 3 in MY 2018 and later vehicles as we
do for NOx emissions (Section 2.1.1.5.5).
2.1.2.2.2 Elemental Carbon and Non-Elemental Carbon Emission Rates
The EC (9.98 percent) and non-EC (90.02 percent) fractions for MY 2010 and later are unchanged from MY
2007-2009 analysis described in section 2.1.2.1.8. As discussed in the Speciation Report1, the ACES Phase 1
is deemed to be an appriorate source of the EC/PM fractions for 2010+ engines which continue to use DPF
technology along with other selective catalyctic converters for controlling NOx. One of the reasons why we
deemed applying the ACES Phase 1 PM speciation preferable to using the ACES Phase 2 profile (which is
tested on MY 2011 diesel engines and aftertreatment systems) is because the ACES Phase 2 program did
not include DPF regeneration events that can have a large impact on PM composition, as discussed in the
MOVES speciation report.1
2.1.2.2.3 DPF Regeneration Events
The MOVES 3 emission rates include active DPF regeneration effects because the HDIUT data set includes
active regeneration activity, but MOVES does not model active regeneration explicitly. To do this, we
would like to have detailed information on the frequency and emission effect of real world regeneration
events by operating mode and regulatory class. Until we have that kind of data and see a need for that
detail in MOVES, we assume that the emission rates in MOVES for MY 2010+ HD vehicles reasonably
capture the average effect of active DPF regeneration events.
71
-------
To assess the amount of active regeneration activity in the HDIUT data, we examined the ECU codes.
Modern DPFs have catalyzed substrate that allow them to undergo passive regeneration when the vehicle
is operating at high-speeds and/or high-loads such that the exhaust temperature is sufficient to induce the
regeneration. The passive regeneration events are "silent" and happen in the background without any
regeneration code in the ECU data. On the other hand, active regeneration events happen when the ECU
actively raises the temperature in the exhaust so that the soot captured in the DPF can be combusted. One
way to increase the temperature is to inject additional fuel which gets burned off and raises the
temperature. These events can raise the PM25 concentrations considerably, but may only occur
infrequently. We analyzed the "Regen_Signal" column in the quality-assured 1 hz emissions data files for
77 vehicles in the HHD 0.20 NOx FEL group to estimate the frequency and count of regeneration events. It
is our understanding that the "Regen_Signal" flag only accounts for active regen events. There were 11
vehicles with the Regen_Signal set to "Y" and the regen events totaled 60,576 seconds, which is about 18%
of the data from just those 11 vehicles and 3% of the data from all 77 vehicles. Future work could evaluate
whether the active regeneration observed in the HDIUT data is consistent with other studies and whether
the PM2.5 second-by-second measurements accurately capture the elevated PM2.5 concentrations that
occur during active regeneration events."1
2.1.2.1 Tampering and Mal-maintenance
Tampering refers to intentional disabling or modifying the vehicle engine, control systems, and/or exhaust
aftertreatment systems that results in increased emissions. Mal-maintenance refers to lack or improper
maintenance of the engine and aftertreatment, including neglecting to repair broken or mal-functioning
engine and aftertreatment parts, which also increases emissions. MOVES uses the same methodology to
apply tampering and mal-maintenance (T&M) adjustment factors to both the pre-2010 and 2010 and later
PM2.5 emission rates
We followed the same tampering and mal-maintenance methodology and analysis for PM2.5 as we did for
NOx, as described in Appendix B. We account for the emission increases with age by multiplying the zero-
mile emission rates by T&M adjustment factors and scaled aged effects (Equation 7-3). The MOVES T&M
adjustment factors on PM2.5 emissions over the fleet's useful life are shown in Table 2-19 and derived in
Appendix B.8. The value of 89 percent for 2010-2012 model years reflects the projected effect of heavy-
duty on-board diagnostic deterrence/early repair of T&M effects. It is an eleven percent improvement
from model years which do not have OBD (i.e., 2007-2009). The 67 percent value for 2013+ is driven by the
assumed full-implementation of the OBD in 2013 and later trucks, which assumes a 33 percent decrease in
T&M emission effects.
m As discussed in the MOVES speciation report1, the PM2.5 composition can change significantly during regeneration
events. Second-by-second PM2.5 measurements made with a photoacoustic or optical method are dependent on the
properties of the PM2.5 composition
72
-------
Table 2-19 Tampering and Mal-maintenance Adjustment Factors for PM2.5 over the Useful Life2
Model Year Group
Increase in PM2.5 Emissions (%)
Pre-1998
85
1998 - 2002
74
2003 - 2006
48
2007 - 2009
100
2010-20121
89 (HHD, MHD, LHD45, and Bus)
67 (LHD2b3)
2013+
67
Note:
1 LHD2b3 achieve full OBD adoption in MY 2010. HHD, MHD, LHD45, and Bus are at partial (33%) and full OBD
adoption in MY 2010-2012 and MY 2013, respectively.
2 Useful life varies by regulatory class (Table B-3)
2.1.2.1.1 1960-2009 Model Years
The CRC E-55/59 emissions data set used for the pre-2010 emission rates was collected during a limited
calendar year period, yet MOVES requires data from a complete range of model year/age combinations.
For example, for the 1981 through 1983 model year group, the primary dataset contained data which was
in either the 15-to-19 or the 20+ age groups. However, for completeness, MOVES must have emission rates
for these model years for age groups 0-3, 4-5, 6-7, etc. In populating the emission rates in MOVES, we used
the age group that had the most data in each regulatory class and model year group combination. Then,
we used the T&M methodology discussed in the previous section to model age and model year group
combinations.
One criticism of the T&M approach is that it may double count the effect of T&M on the fleet because the
primary emission measurements, and base emission rates, were made on in-use vehicles that may have
had maintenance issues during the testing period. This issue would be most acute for the 2007 and later
model year vehicles where all of the deterioration is subject to projection. However, for 2007 and later
model year vehicles, the base emission rates start at low levels, and represent vehicles that are considered
to be free from the effects of T&M.
Figure 2-25 shows the estimated PM2.5 emission rates from MY 2006 heavy-duty diesel vehicles, accounting
for the effect of tampering and mal-maintenance. The different ages at which each regulatory class
reaches full useful life can be observed as the age when the emissions reach their maximum value. The
age 0-3 emission rates for the 2006 and earlier model year were extrapolated using the T&M adjustment
73
-------
factors since the majority of these engines were older than three years when tested in the E-55/59
program.
Reg Class
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
o
LO
CNJ
CL
0.0-
0-3
4-5
6-7 8-9 10-14 15-19 20+
Age (years)
Figure 2-25 Heavy-duty Diesel PM2.5 Emission Rates (g/mile) by Age Group and Regulatory Class for Model Year
2006 using Nationally Representative Operating Mode Distribution
As mentioned in Section 2.1.1.1, the vehicles in HDIUT program are generally well-maintained and
therefore, are used to represent zero-mile emission rates in MOVES. As such, we apply the T&M
adjustment factors shown in Table 2-19 and scaled age effects (Table B-3) to estimate emission rates for
the different ages. Note that for regulatory class LHD45, MHD, HHD and Urban Bus, the PM2.5 emission
rates in the MOVES database for all age groups for MY 2013+ have lower T&M age adjustments than their
counterpart rates for MY 2010-2012 due to HD OBD phase-in (see Section 2.1.2.1 and Appendix B).
Figure 2-26 shows the impact of tampering and mal-maintenance on PM2.5 emission rates by vehicle age
for MY 2015 heavy-duty vehicles. Again, the rate at which emissions increase toward their maximum varies
by regulatory class.
2.1.2.1.2
2010-2060 Model Years
74
-------
0.020-
0)
E
3
cd 0.015
-4—'
CD
(Z
c
o
(/)
cn
E 0.010
LU
-£=
X
LU
aj
-*—>
° 0.005
ln
CN
2
CL
o.ooo-
0-3 4-5 6-7 8-9 10-14 15-19 20+
Age (years)
Figure 2-26 Heavy-duty Diesel PM2.s Emission Rates (g/mile) by Age Group and Regulatory Class for Model Year
2015 using Nationally Representative Operating Mode Distribution
2.1.2.2 Model Year Trends
Figure 2-27 displays the PM2.5 rates by model year and regulatory class for 0-3 age group estimated in
grams per mile (g/mile) using nationally representative operating mode distributions and average speeds.
MOVES models a very large decrease in PM2.5 emission rates starting in model year 2007 (decrease on
order of ~10 to 40 times), when all regulatory classes are assumed to have implemented diesel particulate
filters, with the exception of gliders (Section 2.5). As discussed in Section 2.1.1.7, some of the variation
between regulatory classes is also due to differences in the application of T&M adjustment factors and
differences in the operating mode distributions and average speeds.
Figure 2-28 provides resolution to the model year changes in PM2.5 emission rates for the 2007 and later
model years. Further reductions in PM2.5 emissions are observed for each regulatory class between the
2007-2009 and the 2014 and later emission rates. The higher rates for MY 2010 and later MHD vehicles
stem directly from the HDIUT data as shown in Figure 2-23 and discussed in Section 2.1.2.2.1. The minor
variation in the gram per mile emission rates within the 2010-2013 and 2014+ model year groups by model
year and regulatory class are due to differences in operating mode distributions.
Reg Class
— HHD8
Urban Bus
MHD67
LHD45
— LHD2b3
75
-------
Q) 1.5-
3
CD
-i—"
ro
m
c
o
CO
CO
E
LU
-C
X
LU
1.0-
3 0.5-
o
I—
in
og
o.o-
1970
1980
1990
2000
Model Year
2010
2020
2030
Reg Class
Gliders
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
Figure 2-27 Heavy-duty Diesel PM2.5 Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions
76
-------
2010 2015 2020 2025 2030
Model Year
Figure 2-28 Heavy-duty Diesel PM2.s Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions for MY 2007 through 2030
3
cd 0.03"
-I—'
03
(Z
c=
o
CO
CO
0.02-
X
LU
"en
o
I—
Ln
og
0.01-
Reg Class
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
Figure 2-29 shows the PM2.5 emission rates separated into elemental carbon (EC) and non-elemental
carbon (nonEC) fractions for age 0-3 HHD diesel vehicles using nationally representative operating mode
distributions and average speeds. The EC fraction stays constant until model year 2007, when it is reduced
to less than ~10 percent due the implementation of diesel particle filters.
77
-------
1970 1980
1990 2000 2010
Model Year
2020 2030
Figure 2-29 Heavy Heavy-duty (HHD) Diesel PM2.5 Emission Rates by Elemental Carbon (EC) and Non-Elemental
Carbon (nonEC) Fraction for the 0-3 Age Group by Model Year using Nationally Representative Operating Mode
Distributions
2.1.3 Total Hydrocarbons (THC) and Carbon Monoxide (CO)
While diesel engine emissions of THC and CO are important, they are not the largest contributors to mobile
source THC and CO emission inventories. Regulations of non-methane hydrocarbons (NMHC), combined
with the common use of diesel oxidation catalysts have yielded reductions in both THC and CO emissions
from later model year heavy-duty diesel engines. As a result, data collection efforts typically do not focus
on THC or CO from heavy-duty engines, and less data is available. As discussed in Section 1.1, this report
discusses the derivation of total hydrocarbons (THC), from which MOVES estimates other hydrocarbons
and organic gaseous pollutants.
2.1.3.1 1960-2009 Model Years
We used emissions data combined with emissions standards to develop appropriate model year groups.
Since standards did not change frequently in the past for either NMHC or CO, we created fewer model year
groups than we did for NOx and PM. The MOVES THC and CO model year groups are:
• 1960-1989
• 1990-2006
78
-------
2007-2009
2.1.3.1.1 Data Sources
The heavy-duty diesel THC and CO emission rate development followed a methodology that resembles the
light-duty methodology8, where emission rates were calculated from 1-hz data produced from chassis
dynamometer testing. Data sources were all heavy-duty chassis test programs:
1. CRC E-55/5942: As mentioned earlier, this program represents the largest volume of heavy-duty
emissions data collected from chassis dynamometer tests. All tests were used, not just those using
the TEOM. Overall, 75 trucks were tested on a variety of drive cycles. Model years ranged from
1969 to 2005, with testing conducted by West Virginia University from 2001 to 2005.
2. Northern Front Range Air Quality Study (NFRAQS)54: This study was performed by the Colorado
Institute for Fuels and High-Altitude Engine Research in 1997. Twenty-one HD diesel vehicles from
model years 1981 to 1995 selected to be representative of the in-use fleet in the Northern Front
Range of Colorado were tested over three different transient drive cycles.
3. New York Department of Environmental Conservation (NYSDEC)55: NYSDEC sponsored this study
to investigate the nature and extent of heavy-duty diesel vehicle emissions in the New York
Metropolitan Area. West Virginia University tested 25 heavy heavy-duty and 12 medium heavy-
duty diesel trucks under transient and steady-state drive cycles.
4. West Virginia University: Additional historical data collected on chassis dynamometers by WVU is
available in the EPA Mobile Source Observation Database.
The pre-2010 onroad data used for the NOx analysis was not used since THC and CO were not collected in
the MEMS program, and the ROVER program used the less accurate non-dispersive infrared (NDIR)
technology instead of flame-ionization detection (FID) to measure HC. To keep THC and CO definitions and
data sources consistent, we only used chassis test programs which measured THC using a FID exclusively
for the analysis. Time-series alignment was performed using a method similar to that used for light-duty
chassis test data.
79
-------
Table 2-20 Numbers of Vehicles by Model Year Group, Regulatory Class, and Age Group
Model year
group
Regulatory class
Age group
0-3
4-5
6-7
8-9
10-14
15-19
20+
HHD
58
19
16
9
16
6
7
MHD
9
6
5
4
12
15
6
1960-2002
LHD45
2
1
LHD2b3
6
Bus
26
1
3
2003-2006
HHD
6
2007-2009
HHD, MHD, LHD45,
LH2b3, Bus
No vehicles for this model year group. Rates for
this model year group are based on MY 2003-
2006 with 80 percent reduction.
2.1.3.1.2 1960-2006 Model Years
Similar to the analysis done for PM2.5, for each second of operation on the chassis dynamometer, the
instantaneous scaled tractive power (STPt) was calculated using Equation 1-6 and the second was
subsequently classified to one of the 23 operating modes defined in Table 1-4. We used the same track-
load coefficients, A, B, and C pertaining to heavy-duty vehicles that were used in the PM2.5 analysis.
Using the methods introduced in the NOx analysis, we averaged emissions by vehicle and operating mode.
We then averaged across all vehicles by model year group, age group, and operating mode. In populating
the emission rates in MOVES, we used the age group that had the most data in each regulatory class and
model year group combination. These age groups are shown in Table 2-21. We then used the T&M effects
discussed in Section 2.1.3.3 to extrapolate the emission rates for each age group. For missing operating
modes, we extrapolated using STP as was discussed for NOx in Section 2.1.1.4.2. For the 1960-2002 group,
data for the HHD and Urban Bus regulatory classes were combined because they have the same CO and
NMHC emission standards, although they have separate age effects as discussed in Section 2.1.3.3.
Table 2-21 Age Groups for which THC and CO Emission Rates are Populated Directly Based on the Data
Regulatory class
Model year group
Age group
HHD/Urban Bus
1960-2002
0-3
MHD
1960-2002
15-19
LHD2b3
1960-2002
0-3
HHD
2003-2006
0-3
80
-------
With limited data on LHD45 vehicles, we applied the LHD2b3 emissions data to all LHD vehicles. We also
applied the LHD emission rates from 1960-2002 to the LHD 2003-2006 model year group. For 2003-2006
MHD and Urban Bus regulatory classes emission rates, we applied the HHD 2003-2006 emission rates.
Figure 2-30 and Figure 2-31 show the rates for MHD and HHD for MY 2002 for THC and CO, respectively,
based on the methods described above. The THC and CO mean emission rates increase with STP, though
there is much higher uncertainty than for the NOx rates (Figure 2-1). This pattern could be due to the
smaller data set or may reflect a less direct correlation of THC and CO to STP as is observed for the 2010
and later model year rates.
jig
41
re
100
90
80
70
60
50
™ 40
01
30
20
10
0
~ MHD
¦ HHD/Bus
i i
-| 1 1 1 1 1 r
~i 1 1 1 1 1 1 r
~i 1 1 1 1 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 2-30 THC Emission Rates [g/hr] by Operating Mode for Model Year 2002 and Age Group 0-3. Error Bars
Represent the 95 Percent Confidence Interval of the Mean
81
-------
600 n
500 -
-T 400 H
l59
a»
¦*-»
(0
O
u
c
O)
300 -
2 200 -
100 -
~ MHD
¦ HHD/Bus
i i
«¦51
Hi
n 1 1 1 1 1 r
iii
il
1 1 1 1 1 1 1 r
i—i—i—i—i—i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 2-31 CO Emission Rates [g/hr] by Operating Mode for Model Year 2002 and Age Group 0-3. Error Bars
Represent the 95 Percent Confidence Interval of the Mean
2.1.3.1.3 2007-2009 Model Years
With the increased use of diesel oxidation catalysts (DOCs) in conjunction with DPFs, we assumed an 80
percent reduction in zero-mile emission rates for both THC and CO for 2007-2009 model years. The
derivation of the T&M effects for 2007-2009 model years are presented in Table 2-22 and discussed in
Appendix Section B.9. As shown in Figure 2-42, the CO emission rates developed using this assumption are
significantly lower than the model year 2010 and later emission rates that were developed based on the
HDIUT data, and should be re-evaluated in future versions of MOVES.
2.1.3.2 2010 and Later Model Years
We used the MY 2010+ HDIUT data set, using the same vehicles as used for NOx and described in Section
2.1.1.1 and Table 2-2. The HDIUT dataset includes vehicles in the HHD, MHD, LHD45, LHD2b3 and Urban
Bus regulatory classes. The HDIUT emission measurements are made using instruments that conform to
the requirements described in 40 CFR Part 1065, which require the use of a flame ionization detector (FID)
for measuring total hydrocarbons (THC)56 and a non-dispersive infrared (NDIR) analyzer for carbon
monoxide (CO)57
82
-------
The THC and CO emission rates have more uncertainty than the NOx emission rates, which suggests a less
direct correlation of THC and CO to STP. Nevertheless, we followed the analysis methodology used for MY
2010+ NOx rates as described in Sections 1.6 (calculation of STP and assignment of operating modes),
2.1.1.5 (calculation of mean emission rates), 2.1.1.5.1 (NOx FEL groups) and Appendix G (selection of fscaie).
Figure 2-32 and Figure 2-33 display the HHD THC and CO emission rates estimated from the HDIUT data by
the NOx FEL Groups used to develop the MY 2010-2013 emission rates. Comparisons of the THC and CO
emission rates by NOx FEL Groups for the LHD and MHD regulatory classes are provided in Appendix H.
These comparisons show that there are significant differences among the emission rates in different NOx
FEL groups for THC emissions. The THC emission rates in the 0.2 and 0.35 NOx FEL group are lower than the
THC emission rates from the vehicles in the 0.5 NOx FEL groups for each regulatory class, with the
differences being the most significant for MHD (Figure 7-9) and especially HHD (Figure 2-32). For CO, there
is not a consistent trend among the different NOx FEL groups and regulatory classes. Regardless, we have
analyzed the CO emission rates using the NOx FEL groups for consistency.
0.020
0.015 -
bjo
™ o.oio H
u
0.005
0.000
ii HHD FEL 0.20, MY 2010-2013,
I
k HHD FEL 0.35, MY2010-2015,
I
fa HHD FEL 0.50, MY2010-2015,
N=78
N=31
N=35
- i —t —j . ¦ I
Ji
il
ji
JI
i
il
i
Li
il
0 1, 11 12 13 14 15 161 21 22 23 24 25 27 28 29 30; 33 35 37 38 39 40
MOVES OpMode
Figure 2-32 Average HHD THC Emission Rates by Operating Mode for the 0.2 NO* FEL for MY 2010-2013 and the
0.35 and 0.5 IMOx FEL for MY 2010-2015. Error Bars are 95% Confidence intervals of the Mean
83
-------
MOVES OpMode
Figure 2-33 Average HHD CO Emission Rates by Operating Mode for the 0,2 NOx FEL for MY 2010-2013 and the 0.35
and 0.5 NOx FEL for MY 2010-2015. Error Bars are 95% Confidence Intervals of the Mean
Figure 2-34 and Figure 2-35 display the production-weighted average emission rates for THC and CO
emissions for model year 2013 HHD trucks. Production-weighted averages are calculated for each model
year between model year 2010 and 2018 using the production volumes displayed in Figure 2-12.
0.0050
0.0040
0.0030
5 0.0020
0,0010 ¦
0.0000
LI
I HDIUT, MY 2013
I I
ii
11
I
. I
0 1,11 12 13 14 IS 16 21 22 23 24 25 27 28 29 30 i 33 35 37 38 39 40
MOVES OpMode
Figure 2-34 THC Emissions by Operating Mode from HHD Trucks for Model Year 2013. Error Bars represent the 95
percent confidence interval of the Mean
84
-------
0.09 -
0.03 -
i i
I HDIUT, MY 2013
i
ii
i
0 1 ; 11 12 13 14 15 16 i 21 22 23 24 25 27 28 29 30 i 33 35 37 38 39 40
MOVES OpMode
Figure 2-35 CO Emissions by Operating Mode from HHD Trucks for Model Year 2013. Error Bars represent the 95
percent confidence interval of the Mean
Figure 2-36 and Figure 2-37 display the comparison of the MY 2010-2013 and MY 2014-2015 groups within
the HHD 0.2 NOx FEL Groups. In general, the newer vehicles (MY 2014-2015) have lower THC and CO
emission rates than the corresponding MY 2010-2013 emission rates. Similar model year trends are
observed for LHD THC emissions, and MHD THC and CO emissions in Appendix H. As discussed in Section
2.1.1.5.2 regarding NOx emissions, we attribute the model year differences within the 0.2 NOx FEL Group
to improved emission control hardware and engine and aftertreatment operation.
85
-------
0.0040
0.0035
— 0.0030
i/i
no
— 0.0025
U
I- 0.0020
0.0015
0.0010
0.0005
0.0000
Figure 2-36 THC emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the HHD 0.20 NOx FEL Group
¦ MY 2010-2013
¦ MY 2014-2015
|
ii ii ii li ii ii
1 ii ii ii ii ii Ii
I
I 1
T I
1
i. .. ii Ii 1
||
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
0.14
0.12
0.10
ln
0.08
O
<-> 0.06
0.04
0.02
0.00
Figure 2-37 CO emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the HHD 0.20 NOx FEL Group
LHD2b3 and LHD45 emission rates are based on the vehicles with "LHD" service class in the HDIUT data set.
Urban Bus emission rates are based on HHD vehicles in the HDIUT data set. MHD and HHD emission rates
are based on the emission rates from those vehicle classes. The
HDIUT data set predominantly contains vehicles in the 0-3 age group with only a handful vehicles in the 4-5
age group. Since the HDIUT data is measured and submitted by the manufacturer and the test vehicles are
required to be free of any tampering or mal-maintenance, we can safely assume that they represent zero-
mile vehicles for the purpose of assigning base rates and applying the tampering and mal-maintenance
effects.
MY 2010-2013
MY 2014-2015
i1 i1
jl
11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
86
-------
A comparison of HDIUT-based THC and CO emission rates for MY 2010+ heavy-duty vehicles by regulatory
class are shown in Figure 2-41 and Figure 2-42, respectively. The THC rates, generally low for diesel
vehicles, are comparable to MY 2007-2009 rates for both MHD and HHD. However, for CO, the HHD rates
for MY 2010+ are significantly higher compared to MY 2007-2009, but are comparable to the pre-2007
data which are based on emission measurements. The variation in the 2010-2018 rates reflects the model
year variation in the production volume by NOx FEL group, and use of the different 0.2 FEL NOx model year
group between 2010-2013 and 2014 and later.
In the 2017 review of a draft version of this report,14 we received a comment that single-cell-NDIR-based
CO measurements suffer from severe drift that is not corrected by zero and span checks because the
calibration gases are dry, while vehicle tailpipe exhaust gases are not dry. Based on the HDIUT data, it is
not possible for us to determine if MY 2010+ CO emission rates are affected by the alleged drift in the CO
measurements. We looked at the CO emissions for each of 93 vehicles in the HHD 0.20 FEL group (from the
2010-2016 selection years) and confirmed the high average CO rate is not due to a few outliers. Further,
the CO emission rate for the MHD and LHD vehicles is significantly lower (see Figure 2-42). Based on the
available data and trends, we are unable to confirm whether or not the high CO emissions for the HHD
vehicles is real or an artifact of CO sensor drift. In Section 2.1.5, we demonstrated that the fleet-average
heavy-duty CO emission rate estimates from MOVES compare well with measurements from heavy-duty
exhaust plume capture and tunnel measurement campaigns conducted in 2015 and 2017, which increased
our confidence that the CO emission rates measured from HDIUT are reasonable. Thus, we decided to
accept the reported HDIU CO emission rates as valid.
2.1.3.3 Tampering and Mal-maintenance
For all model years, we applied tampering and mal-maintenance effects to adjust emissions from the
measured age to all age groups, lowering emissions for younger ages and raising them for older ages, using
the methodology described in Appendix B. We applied the tampering and mal-maintenance effects shown
below in Table 2-22 to CO and THC.
Table 2-22 Tampering and Mal-maintenance Effects for THC and CO over the Useful Life
Model years
Increase in THC and CO Emissions (%)
Pre-2003
300
2003 - 2006
150
2007 - 2009
150
2010-20121
29 (HHD, MHD, LHD45, and Bus)
22 (LHD2b3)
87
-------
2013+
22
Note:
1 LHD2b3 achieve full OBD adoption in MY 2010. HHD, MHD, LHD45, and Bus are at partial (33%) and full OBD
adoption in MY 2010-2012 and MY 2013, respectively.
While LHD2b3, LHD45 and MHD vehicles share the same pre-2010 MY fully deteriorated emission rates for
THC and CO, they deteriorate differently as they age. Table B-3 estimates the degree of T&M that occurs
by age by using the warranty and full useful life requirements for each heavy-duty regulatory class with the
average mileage accumulation rates. We multiplied these increases by the T&M age-based adjustment
factors shown in Table B-3 and applied the result to the zero-mile (or age 0) emissions rate to get the
emissions rate by age group using Equation 7-3.
Figure 2-38 and Figure 2-39 show THC and CO emission rates by age group. Due to our projections of T&M
effects, there are large increases as a function of age. Additional data collection would be valuable to
determine if real-world deterioration effects are consistent with those in MOVES.
88
-------
100
90
80
IT 70
.c
^ 60
01
2 50
u
X
c 40 -
(U
0)
^ 30
20
10
0
S
¦ HHD
~ MHD 1
A Bus
0-3 4-5 6-7 8-9 10-14
Age group [years]
15-19
20+
Figure 2-38 THC Emission Rates [g/hr] by Age Group for Model Year 2002 and Operating Mode 24. Error Bars
Represent the 95 Percent Confidence Interval of the Mean
600
500
400
ai
(V
i-
O
u
300
m
0)
s 200
100
!¦
¦ HHD
~ MHD
* Bus
0-3
4-5 6-7 8-9
10-14 15-19
Age group [years]
20+
Figure 2-39 CO Emission Rates [g/hr] by Age Group for Model Year 2002 and Operating Mode 24. Error Bars
Represent the 95 Percent Confidence Interval of the Mean
89
-------
2.1.3.4
Model Year Trends
Figure 2-40 through Figure 2-42 display the THC and CO emission rates by model year and regulatory class
for age group 0-3 estimated in grams per mile (g/mile) using nationally representative operating mode
distributions and average speeds. As discussed in Section 2.1.1.7, some of the minor variation in the gram
per mile emission rates within the model year groups groups and between regulatory classes are due to
differences in operating mode distributions. Differences in the emission rates for age group 0-3 between
regulatory classes are also due to different application of the T&M adjustment factors (Section 2.1.3.3). For
example, as discussed in Section 2.1.3.1.3, the zero-mile MY 2003-2006 emission rates by operating mode
are equivalent for HHD, MHD, and Urban Bus, but the T&M adjustment factors are applied differently for
each regulatory class, and the operating mode distributions are difference, resulting in the differeng gram
per mile emission rates observed for 2003-2006 in Figure 2-40 and Figure 2-42.
The MY 2007-2009 emission rates reflect the use of diesel oxidation catalysts and are derived by reducing
the CO and THC emissions in MY 2003-2006 by 80 percent and applying the model-year and regulatory
class specific T&M adjustment factors. For MY 2010-2018, the significant variation in the emission rates by
model year are due to the model year specific production volumes of the NOx FEL Group and the model
year split of the 2010-2013 and 2014 and later 0.2 NOx FEL group (Section 2.1.1.5.2 and Section 2.1.3.2).
For example, the spike in THC emissions observed in the 2010 model year HHD vehicles (Figure 2-41) is
explained by the high THC emissions of the 0.5 NOx FEL group (Figure 2-32), and the high production
volumes of the 0.5 NOx FEL engines in MY 2010 (~50% of the total HHD, see Figure 2-12).
90
-------
0.5-
| 0.4-
5
(D
-t—«
CO
01 0.3-
c:
o
CO
w
£
LU 0 2 —
O
X
c/)
03
o
«o.i-
1
¦ -W»
^^5
o.o--
1970
1980
1990
2000
Model Year
2010
2020
2030
Reg Class
Gliders
— HHD8
Urban Bus
MHD67
LHD45
— LHD2b3
Figure 2-40 Heavy-duty Diesel THC Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions
0.15-
3
0)
ro
02 0.10-
C
o
CO
co
LU
O
X
a) 0.05-
co
0
~CD
-t—<
o
Reg Class
— HHD8
Urban Bus
— MHD67
LHD45
— LHD2b3
o.oo-
2010
2015 2020
Model Year
2025
2030
91
-------
Figure 2-41 Heavy-duty Diesel THC Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions (Model Year 2007-2030 only)
2-
3
CD
-I—"
CO
a:
c
o
CO
CO
E
LU
O
O
Reg Class
Gliders
HHD8
Urban Bus
— MHD67
LHD45
— LHD2b3
1970 1980 1990 2000 2010
Model Year
2020
2030
Figure 2-42 Heavy-duty Diesel CO Emission Rates (g/mile) for the 0-3 Age Group by Regulatory Class using
Nationally Representative Operating Mode Distributions
2.1.4 Energy
2.1.4.1 1960-2009 Model Years
2.1.4.1.1
LHD
In MOVES3, the energy rates for LHD (LHD2b3 and LHD45) for pre-2010 MY diesel vehicles are unchanged
from MOVES2010a. In MOVES2010, the energy rates for LHD2b3 regulatory class, along with the light-duty
regulatory classes (regClasslDs 20 and 30), varied by fueltype, model year group, engine technology, and
"size weight fraction" as discussed in the MOVES2010a energy updates report.58 The energy rates in
MOVES2010a were simplified to be a single set of energy rates for each regulatory class, fuel type and
model year combination by weighting across engine size, engine technology, and vehicle weight according
to the default population in the MOVES2010 sample vehicle population table. The resulting C02 (g/mile)
emission rates and fuel economy values (miles per gallon) calculated from the energy rates using nationally
representative operating mode distributions and average speeds are shown in Figure 2-46 and Figure 2-47.
92
-------
Because this approach uses highly detailed data, coupled with information on the vehicle fleet that varies
for each model year, model year variability was introduced into the energy rates used in MOVES.
2.1.4.1.2 MHD, Urban Bus, and HHD
The data used to develop NOx rates was used to develop running-exhaust energy rates for the MHD, Urban
Bus, and HHD vehicles. The energy rates were based on the same data (Section 2.1.1.1), STP structure and
calculation steps as in the NOx analysis (Sections 2.1.1.3 and 2.1.1.4); however, unlike NOx, we did not
classify the energy rates by model year, regulatory class, or by age, because neither variable had a
significant impact on energy rates or C02.
In MOVES, C02 emissions were used as the basis for calculating energy rates. To calculate energy rates
(kJ/hour) from C02 emissions (Equation 2-21), we used a heating value (HV) of 138,451 kJ/gallon and C02
fuel-specific emission factor (fC02) of 10,180 g/gallon59 for conventional diesel fuel.
HV
^energy ^C02 ~c Equstion 2-21
JC02
The energy rates for the MHD, Urban Bus, and HHD vehicle classes are shown in Figure 2-43. Compared to
other emissions, the uncertainties in the energy rates are smaller, in part because there is no classification
by age, model year, or regulatory class. Thus, the number of vehicles used to determine each rate is larger,
providing for a greater certainty of the average mean energy rate.
Operating mode-based energy consumption rates are the same across MHD, Urban Bus, and HHD
regulatory classes. However, the distribution of time spent in the operating mode varies between these
regulatory classes based on differences in their activity and tractive power demand. Thus, the C02 (g/mile)
emission rates and fuel economy values (miles per gallon) calculated from the energy rates using nationally
representative operating mode distributions differ by regulatory class as shown in Figure 2-46 and Figure
2-47.
93
-------
6
5
~ • *
-i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 2-43 Diesel running exhaust energy rates for MHD, HHD, and Urban Buses for 1960-2009 model years. Error
bars represent the 95 percent confidence interval of the mean
2.1.4.2 2010-2013 Model Years
The MY 2010+ HDIUT dataset described in Section 2.1.1.1 and Table 2-2 included C02 emissions data,
which was used to update the energy rates. The energy rates are derived using the C02 rates and the
conventional diesel specific values for carbon content (0.0202 g/KJ) and a HV of of 138,451 kJ/gallon,
yielding a C02 fuel-specific emission factor (fC02) of 10,255 g/gallon. MOVES uses these same values to
calculate C02 emissions from the energy rates of vehicles using conventional diesel fuel - this methodology
is described in the MOVES GHG and Energy Consumption report3.
The 2010-2013 model year energy rates were calculated using the NOx FEL production volume and model
year group splits used for estimating the MY 2010+ NOx rates as described in Section 2.1.1.5. The energy
rates for the 0.2 NOx FEL group are based only on MY 2010-2013 vehicles. The energy rates for the 0.35
and 0.5 NOx FEL group were developed using vehicles sampled between 2010-2016 model years. Figure
2-44 shows the mean HHD C02 emission rates for the NOx FEL Groups used to estimate the MY 2010-2013
emission rates. As shown, C02 emission rates are a strong function of STP operating mode, and there is
significantly less variability in the C02 emission rates between the sampled vehicles compared to other
pollutants, as evidenced by the small confidence intervals. Even though there is little difference between
the C02 emission rates among the different FEL groups, we used the FEL production volume and model
94
-------
years splits to estimate the C02 emissions, to be consistent with our analysis of the THC, CO, and NOx
emission rates.
^ HHD FEL 0.20, MY 2010-2013,
!¦ HHD FEL 0.35, MY 2010-2015,
p HHD FEL 0.50, MY 2010-2015,
J"- zls.
Li
J1L
0 1 , 11 12 13 14 15 16
N=78
N=31
N=35
JiL
21 22 23 24 25 27 28 29 30 ; 33 35 37 38 39 40
MOVES OpMode
Figure 2-44 Average HHD C02 Emission Rates by Operating Mode for the 0.2 NOx FEL for MY 2010-2013 and the 0.35
and 0.5 NOx FEL for MY 2010-2015. Error Bars are 95% Confidence intervals of the Mean.
As shown in Table 2-23, the majority of vehicles within the NOx FEL groups of 0.35 and 0.5 include MY
2010-2013 vehicles, for both HHD and MHD vehicles. As discussed in the next subsection, we expect
improved energy effiencies in MY 2014 and later vehicles due to the phase-in of the Phase 1 Heavy-duty
Greenhouse Gas Emission Standards. Due to the small sample of vehicles in the 0.35 and 0.5 NOx FEL
groups, we assume that the MY 2010-2016 energy rates are representative of both the MY 2010-2013 and
the 2014-2016 model year groups.
95
-------
Table 2-23 HDIU Vehicles with Valid C02 Measurements By Regulatory Class, Model Year Group and NOx FEL Group
NOx FEL Group
Total
Valid
Tested
Vehicles
Reg
Class
Model Year
Group
0.2
0.35
0.5
LHD
2010-2013
52
0
10
62
64
2014-2016
27
0
5
32
32
MHD
2010-2013
21
23
9
53
55
2014-2015
19
0
0
19
19
HHD
2010-2013
78
26
35
139
139
2014-2015
44
5
0
49
55
Using this method, the energy rates for each model year are unique based on NOx FEL based production
volume weighting, as can be observed in the model year variability among the MY 2010-2013 C02 (g/mile)
emission rates and fuel economy values (miles per gallon) shown in Figure 2-46 and Figure 2-47.
2.1.4.3 2014-2060 Model Years
2.1.4.3.1 LHD45, MHD, Urban Bus, and HHD Energy Rates
In developing the MY 2014-2060 running energy rates for LHD45, MHD, Urban Bus, and HHD, we also used
the NOx FEL groups, model year groups, and production volume weights as discussed in the previous
section (Section 2.1.4.2). The MY 2014-2018 running energy rates were calculated as a weighted average
using the MY 2014-2015/2016 rates from the 0.2 NOx FEL group, and the MY 2010-2016 rates from the
0.35 and 0.5 NOx FEL groups. Although the 0.35 and 0.5 NOx FEL groups contain measurements from MY
2010-2013 vehicles (Table 2-23), applying the MY-specific production volume weighting of the FEL groups
means that the MY 2014-2018 emission rates are primarily or entirely based on the data from the MY
2014-2015/2016 vehicles in the 0.2 NOx FEL group. The MY 2014-2018 running energy rates developed in
this step are considered the "baseline" - the reductions in energy rates expected from the Medium- and
Heavy-Duty Greenhouse Gas (GHG) Phase 1 Rule60 were applied to the "baseline" energy rates as
described below.
MOVES accounts for the improved fuel efficiency achieved by the HD GHG Rulemakings in two ways. First,
the running, start, and extended idle rates for total energy consumption are reduced to be consistent with
the HD GHG rules. Second, the truck weights and road-load coefficients are updated to reflect the lower
vehicle curb weights through lightweighting of materials, lower resistance tires, and improved
aerodynamics of the vehicle chassis. Vehicle weights and road-load coefficients are discussed in the
Population and Activity Report.9
96
-------
The HD GHG Phase 1 rule60 was implemented starting with 2014 model year and increased in stringency
through model year 2018. The reductions in start and running energy rates reflect the improvements
expected from improved energy efficiency in the powertrain. The estimated reductions for heavy-duty
diesel energy rates from the HD GHG Phase 1 rule are shown in Table 2-24.
In MOVES3, the MY 2014-2016 running energy rates were not adjusted for the HD GHG Phase 1
rulemaking, because the impact of Phase 1 is assumed to be included in the measurements from the MY
2014-2015/2016 vehicles sampled from the HDIU program. Instead, we renormalized the Phase 1 GHG
reductions using the MY 2014-2016 as the baseline using Equation 2-22.
Renormalized Phase 1 reductions in yeart
1 — (reductions in yeari) Equation
1 — (Average reductions in 2014 thru 2016) 2-22
For example, the renormalized reductions for LHD and MHD in 2017-2020 are calculated as:
Renormalized Phase 1 reductions for LHD and MHD in 2017 thru 2020
1 — (9%) 91% c ..
= 1 =1 = 1 - 96% = 4% Equation
1 - (5%) 95% 2-23
We applied the renormalized reductions to estimate the MY 2017 and later running energy rates, as shown
in Table 2-24. As discussed in Section 2.2.4, because the start energy rates were not updated with more
data from model year 2014-2016 vehicles, the reduction in energy consumption from starts due to HD
GHG Phase 1 rule was modeled by directly using the reductions estimated from the rule.
97
-------
Table 2-24 Estimated Reductions in Diesel Engine Energy Consumption Rates from the HD GHG Phase 1 Program61
Regulatory
Class
Fuel
Model
Years
Estimated Reduction from the
MY 2013 Baseline
(applied to starts)
MOVES3 Renormalized
Reductions to MY 2014-
2016 Energy Rates
(applied to running)
HHD and
Urban Bus
Diesel
2014-2016
3%
-
2017
6%
3%
LHD and
MHD
Diesel
2014-2016
5%
-
2017-2020
9%
4%
MOVES3 also incorporates the Medium- and Heavy-Duty GHG Phase 2 rule.62 The Phase 2 program begins
in 2018 model year for trailers and in 2021 for the other categories, while phasing in through model year
2027. These Phase 2 standards continue indefinitely after model year 2027. The programs break the
diverse truck sectors into three distinct categories, including:
• Line haul tractors and trailers (combination trucks source types in MOVES)
• Heavy-duty pickups and vans (passenger truck and light-commercial trucks)
• Vocational trucks (buses, refuse trucks, motorhomes, single-unit trucks)
The Phase 2 Rule set separate standards for engines and vehicles and ensured improvements in both. It
also set separate standards for fuel consumption, C02, N20, CH4 and HFCs."
Because the Phase 2 rulemaking set different standards for vocational vehicles and tractor-trailers and
because single-unit vocational vehicles and tractor-trailers are mapped to the same regulatory classes
(MHD and HHD) under the default MOVES framework, it was necessary to create a new
EmissionRateAdjustment table with sourceTypelD as a primary key. The EmissionRateAdjustment table
alllows MOVES3 to model the final standards for vocational vehicles and tractor-trailers simultaneously.
The EmissionRateAdjustment table includes the following data fields, many shared with the EmissionRate
table:
1) polProcessID (primary key)
2) sourceTypelD (primary key)
3) regClassID (primary key)
4) fuelTypelD (primary key)
" HFCs are not modeled in MOVES, and the N20 and CH4 standards are not considered technology forcing on
emissions.
98
-------
5) beginModelYearlD (primary key)
6) endModelYearlD (primary key)
7) emissionRateAdjustment
8) dataSourcelD
Table 2-25 summarizes the energy rate reductions stored in the EmissionRateAdjustment table which are
applied to the running rates in MOVES3 for MY 2018 and later heavy-duty diesel vehicles.
Table 2-25 Estimated Reductions in Diesel and CNG Engine Energy Consumption Rates due to the HD GHG Phase 2
Program62
Vehicle Source Type
(Source Type ID)
Fuel
Model years
Reduction from MY 2017
Energy Rates
Long-haul Combination
Truck
(62)
Diesel
2018-2020
1.0%
2021-2023
7.9%
2024-2026
12.4%
2027+
16.3%
Short-haul Combiation
Truck
(61)
Diesel & CNG
2018-2020
0.6%
2021-2023
7.4%
2024-2026
11.9%
2027+
15.0%
Other Bus, School Bus,
Refuse Truck, Single-Unit
Short-Haul, Single-Unit
Long-Haul, Motorhomes
(41, 43, 51, 52, 53, 54)
Diesel & CNG
2021-2023
7.8%
2024-2026
12.3%
2027+
16.0%
Transit Bus
(42)
Diesel & CNG
2021-2023
7.0%
2024-2026
11.8%
2027+
14.4%
Thus, for LHD45, MHD, HHD and Urban Bus, the running energy rates for MY 2018 and later model years
are estimated with a chain of calculations starting with the HDIUT-based estimates by operating mode and
regulatory class for MY 2018, then reduced by applying the HD GHG Phase 1 reduction in Table 2-24 and
further reduced by applying the HDGHG Phase 2 reductions listed in Table 2-25 for model year 2018 and
later. The reductions shown in Table 2-25 are a combination of improvements to the engine and other
systems excluding aerodynamics and tire rolling resistances. The projected improvements due to
aerodynamics and tire rolling resistance are reflected in new road load coefficients, as described in the
Population and Activity Report.9 The reductions in energy use applied from both the Phase 1 and Phase 2
HD rulemakings to the MY 2014-2018 energy rates developed using HDIUT data are displayed in Figure
2-45.
99
-------
CD
-t-J
03
0c
-------
Renormalized Phase 1 reductions for LHD2b3 in 2018 thru 2020 =
1 -15% 1 -15% 85% c
=1 =1 = 1 — 88% = 12% Equatio
1 ^2.3% + 3% + 6%j 1-3.8% 96.2% 2.24
Table 2-27 shows the projected improvements in C02 emissions due to the HD GHG Phase 2 program for
diesel and gasoline LHD2b3 vehicles. Since nearly all HD pickup trucks and vans will be certified on a chassis
dynamometer, the C02 reductions for these vehicles are not treated as separate engine and road-load
reduction components, but represented as total vehicle C02 reductions. MOVES models the HD pickup
truck/van standards by lowering the energy rates stored in the emissionRate table. No change is made to
the road-load coefficients or weights of passenger or light-duty truck source types. Instead of using the
EmissionRateAdjustment table, the energy consumption rates for LHD2b3 were lowered by the
percentages shown in Table 2-26 and Table 2-27 for the corresponding model years.
The heavy-duty regulations apply to all the vehicles in the LHD2b3 regulatory class, with the exception of
medium-duty passenger vehicles (MDPVs). The fuel economy of MDPVs are covered by the Light-duty GHG
rule.63 However, MDPVs make a minor contribution to the population of LHD2b3, and we apply the HD
GHG energy reductions to all vehicles within LHD2b3.
Table 2-26 Estimated Total Vehicle Reductions in Energy Consumption Rates for LHD2b3 Diesel Vehicles due to the
HD GHG Phase 1 Program
Regulatory
Class
Fuel
Model years
Reduction from MY
2013 Energy Rates
MOVES3
Renormalized
Reductions to MY
2014-2016 Energy
Rates (applied to
running)
LHD2b3
Diesel
2014
2.3%
-
2015
3%
-
2016
6%
-
2017
9%
5%
2018-2020
15%
12%
101
-------
Table 2-27 Estimated Total Vehicle Reductions in Energy Consumption Rates for LHD2b3 Diesel and Gasoline due to
the HD GHG Phase 2 Program
Regulatory Class
Fuel
Model years
Reduction from MY
2020 Emission Rates
LHD2b3
Gasoline
2021
2.50%
and Diesel
2022
4.94%
2023
7.31%
2024
9.63%
2025
11.89%
2026
14.09%
2027+
16.24%
2.1.4.4 Model Year Trends
Figure 2-46 and Figure 2-47 display the C02 (g/mile) emission rates and fuel economy values calculated in
MOVES from the energy rates using the carbon content and energy density conversion factors for
conventional diesel fuel as documented in the MOVES3 Greenhouse Gas and Energy Report.3 The C02
(g/mile) emission rates and fuel economy values are estimated using nationally representative operating
mode distribution and average speed values. The figures show that, since model year 2010, there are
decreasing trends in C02 (g/mile) with corresponding increases in fuel economy, due to the lower MOVES
energy consumption rates as well as the lower source mass values and improved road load coefficients
estimated vehicles meeting both Phase 1 and Phase 2 Heavy-Duty Greenhouse Gas Standards. The energy
rates by operating mode are constant for model year 2027-2060. However, some of the small differences
in C02 (g/mile) and fuel economy values observed within model year groups and regulatory classes are due
to differences in the nationally representative operating modes across model years due changing fractions
of regulatory classes among different source types.
102
-------
Reg Class
Gliders
— HHD8
Urban Bus
MHD67
LHD45
LHD2b3
1500-
•| 1000--
LU
OJ
O
O
o
S 500"
CL
c/>
o
1970
1980
1990
2000 2010
Model Year
2020
2030
Figure 2-46. Heavy-duty Diesel C02 Emission Rates (g/mile) by Regulatory Class using Nationally Representative
Operating Mode Distributions
103
-------
25
1 15-
CO
0
1
CD
CL
w
-------
Table 2-28 shows that CO emission factors compare quite well between the different studies. The NOx
values are comparable to the Peralta CA location, but significantly lower than the Ft. McHenry location.
THC are also below the Peralta, CA measurements. PM2.5 is lower than the Ft. McHenry estimates, but
within the standard error of the winter measurements, and close to the 95% confidence range of the
summer measurements (approximately two times the standard error). Given the expected differences in
vehicle operation and fleet composition, the comparisons increased our confidence that MOVES is
estimating representative in-use running emission factors for heavy-duty diesel vehicles.
Table 2-28. Comparison of MOVES Emissions with Remote Sensing and Tunnel Measurements.
Peralta CA 2017
HDV (Haugen et
al. 2018)
Winter Ft.
McHenry MD
2015 (Wang
et al. 2019)
Summer Ft.
McHenry MD
2015 (Wang
et al. 2019)
MOVES3
National
Heavy-duty
Fleet 2016
THC
2.2 ± 0.4
NA
NA
0.68
CO
5.9 ±0.9
4.6 ± 2.0
7.5 ± 2.6
5.2
NOx
12.4 ±0.6
29.6 ±4.7
17.9 ± 1.4
12.2
PM2.5
NA
0.81 ±0.89
0.61 ±0.11
0.36
Note: the error terms are the standard error of the mean based on individual vehicle measurements for the Peralta
location and sampling periods for the Ft. McHenry Tunnel.
2.2 Start Exhaust Emissions
The start process occurs when the vehicle is started and the engine is not fully warmed up. For modeling
purposes, we define start emissions as the increase in emissions due to an engine start. Operationally, we
estimate difference in emissions between a test cycle with a cold start and the same cycle with a hot start.0
As explained in Section 1.2.2, we define eight stages which are differentiated by soak time length (time
duration between engine key off and engine key on) between a cold start (> 720 minutes of soak time) and
a hot start FTP (< 6 minutes of soak time). More details on how start emission rates are calculated as a
function of soak time, can be found later in this section and in the MOVES light-duty exhaust emission rate
report.8 The impact of ambient temperature on cold starts is discussed in the Emission Adjustments
MOVES report.64
The next subsections discuss the derivation of heavy-duty diesel start emissions by pollutant and model
year group. Start emissions are currently a small contributor to total exhaust emissions from heavy-duty
diesel vehicles. No T&M or other age effects are currently applied to the diesel start emissions.
105
-------
2.2.1 THC, CO, and N0X
The pre-2010 model year emissions are discussed in Section 2.2.1.1 and 2010+ model year emission rates
are discussed in Sections 2.2.1.2.
2.2.1.1 1960-2009 Model Years
For light-duty diesel vehicles, start emissions are estimated by subtracting FTP bag 3 emissions from FTP
bag 1 emissions. Bag 3 and Bag 1 are collected on the same dynamometer cycle, except that Bag 1 starts
with a cold start, and Bag 3 begins with a hot start.0 A similar approach was applied for LHD vehicles tested
on the FTP and ST01 cycles, which also have separate bags measuring cold and hot start emissions over
identical drive cycles. Data from 21 LHD diesel vehicles, ranging from model years 1988 to 2000, were
analyzed. No classifications were made for model year or age due to the limited number of vehicles. The
results of this analysis for THC, CO, and NOx are shown in Table 2-29.
Table 2-29 Average Start Emissions Increases (g/start) for pre-2010 Model Year Light Heavy-Duty Diesel Vehicles for
Regulatory Class LHD2b3 and LHD45 (regClassID 41 and 42)
THC
CO
NOx
0.13
1.38
1.68
For pre-2010 model year HHD and MHD trucks, analogous data were unavailable. To provide at least a
minimal amount of information, we measured emissions from a 2007 Cummins ISB which is used in both
LHD and MHD vehicles on an engine dynamometer at the EPA National Vehicle and Fuel Emissions
Laboratory in Ann Arbor, Michigan. Among other idle tests, we performed a cold start idle test at 1,100
RPM lasting four hours, long enough for the engine to warm up. Essentially, the "drive cycle" we used to
compare cold start and warm emissions was the idle cycle, analogous to the FTP and ST01 cycles used for
LHD vehicles. Emissions and temperature stabilized about 25 minutes into the test. The emission rates
through time are shown in Figure 2-48. The biggest drop in emission rate over the test was with CO,
whereas there was a slight increase in NOx (implying that cold start NOx is lower than running NOx), and an
insignificant change in THC.
° As discussed in Section 1.2.2, ideally, bag 3 would not include a start, but only include running emissions.
Operationally, we use bag 3 with a hot-start because that is the available data, and we assume that the hot-start
emissions are small in comparison to the cold-start emissions, and thus have minimal impact on the cold-start
estimate. Our estimates of emissions by soak time in Sections 2.2.3.1 and 2.2.3.2 support this assumption; for pre-
2010, hot-start THC and CO emissions are less than 10% of the cold-start emissions, and NOx hot-start emissions are
less than 20% of cold-start emissions. For 2010+ emissions the hot-start emissions for THC, CO, and NOx are less than
1% of the cold-start emissions.
106
-------
time [hrs]
Figure 2-48 Trends in the Stabilization of Idle Emissions from a Diesel Engine Following a Cold Start (from a 2007
Cummins ISB Measured on an Engine Dynamometer)
We calculated the area under each curve for the first 25 minutes and divided by 25 minutes to get the
average emission rate during the cold start idle portion. Then, we averaged the data for the warm idle
portion using the remaining portion of the test (215 minutes). We then calculated the the difference
between cold start and warm idle over a 25-minute period of the elevated cold starts as shown in Equation
2-25.
Grams per Start =
= stabilization time x (cold start average rate — hot running average rate)
Equation 2-25
/ 25 240 \
I v emissions v emissions \
= 25 minutes x
r1 emissions r1
2j 25 2j 215 I
\t=0 t=25 /
The results are shown in Table 2-30. The measured THC increment is zero. The NOx increment is negative
since cold start emissions were lower than warm idle emissions.
107
-------
Table 2-30. Cold-start Emissions Increases (g/start) in Grams on the 2007 Cummins ISB
THC
CO
NOx
0.0
16.0
-2.3
We also considered NOx data from University of Tennessee,65 which tested 24 trucks with PEMS at
different load levels during idling. Each truck was tested with a cold start going into low-RPM idle with air-
conditioning on. We again used Equation 2-25 to integrated the emissions over the warm-up period to get
the total cold start idling emissions. We calculated the warm idling emissions by multiplying the reported
warm idling rate by the stabilization time. We used the stabilization period from our engine dynamometer
tests (25 minutes). Then, we subtracted the cold start-idle emissions from the warm idle emissions to
estimate the cold start increment. We found that several trucks produced lower NOx emissions during cold
start (similar to our own work described above), and several trucks produced higher NOx emissions during
cold start. Due to these conflicting results, and the recognition that many factors affect NOx emission
during start (e.g., air-fuel ratio, injection timing, etc.), we set the default NOx cold-start increment to zero.
Table 2-31 shows our final MOVES inputs for HHD and MHD diesel start emissions increases from our 2007
MY in-house testing. Due to the limited data, the emission rate is constant for all pre-2010 model years
and ages.
Table 2-31. MOVES Inputs for Pre-2010 HHD and MHD Diesel Start Emissions (grams/start) for Regulatory Class 46,
47, and 48. No Differentiation by Model Year or Age.
THC
CO
NOx
0.0
16.0
0.0
As discussed in the Emission Adjustments Report64, MOVES applies an additive adjustment to diesel THC
cold-start emissions for ambient temperatures below 72 F. Thus, despite a pre-2010 baseline THC start
emission rate of zero, MOVES estimates positive THC start emissions from heavy-duty diesel vehicles at
ambient temperatures below 72 F. No temperature adjustments are applied to CO, PM2.5, or NOx diesel
start emissions because no clear trend was found with the data.
2.2.1.2 2010-2060 Model Years
The cold start emissions for 2010 model year and later LHD, MHD, and HHD diesel engines have been
updated for MOVES3 based on new data. However, because of the small sample size and lack of real-world
data, notable uncertainty about real world heavy-duty diesel start emissions remains.
Similar to the approach taken for light-duty vehicles, the cold start emissions are defined as the difference
in emissions between a test cycle with a cold start and the same test cycle with a hot start. Heavy-duty
108
-------
diesel engines are certified using the Heavy-Duty Diesel Engine Federal Test Procedure (FTP) cycle66. The
test procedure for certification requires that manufacturers run the engine over the FTP cycle with a cold
start and then repeat the cycle with a warm start. Starting in model year 2016, EPA began collecting
certification data that contained separate cold and hot results for each engine certified. The data that was
analyzed for this MOVES3 update includes the following engine families from 2016 and 2017 model years
shown in Table 2-32.
Table 2-32 Engine Data Analyzed to Estimate the Cold Start Emission Rates for HD Diesel Engines
Category
Number of Engines
Manufacturers
LHD
5
Ford, Isuzu, Hino, FPT
MHD
6
Ford, Hino, Cummins, Detroit Diesel
HHD
11
Cummins, PACCAR, Detroit Diesel,
Volvo, Hino
The certification data was used to determine the grams emitted per cold start using Equation 2-26.
Grams per Start
= [Cold FTP Emission Results (g/(hp — hr))
— Hot FTP Emission Results (g/(hp — hr))] Equation 2-26
* FTP Cycle Work (hp — hr)
The amount of work (hp-hr) performed over the FTP cycle is required to convert the FTP emission results in
grams per horsepower-hour into grams, but it is not provided as part of the certification data submitted by
the manufacturers to EPA. Furthermore, the FTP cycle work is unique to each engine and is generally
calculated based on the engine's maximum speed, curb idle speed, and the maximum torque curve.
Therefore, we needed to develop a surrogate from the information that is provided by manufacturers for
certification for each engine. We determined that the rated power of an engine correlates well to the FTP
cycle work. This analysis was based on FTP cycle work and rated power data from ten HD engines. As
shown in Figure 2-49, the FTP cycle work is approximately a linear function of the engine's rated power.
For the calculation of cold start emissions for each engine analyzed, the FTP cycle work (hp-hr) was
estimated for the engine based on its rated power using the equation in Figure 2-49 - 0.0599 (hr) times the
rated power (hp) plus 4.4297 (hp-hr).
109
-------
FTP workvs Rated Power - Diesel
40
35
ISO
f-
£
20
15
250 300 350 400 450 500 550
Rated Power (hp)
Figure 2-49: Relationship between HD Diesel Engine Rated Power and FTP Cycle Work
2.2.1.2.1 Heavy-heavy duty
Analysis of cold and hot start FTP certification data from eleven HHD diesel engines determined the grams
per start for THC, CO, NOx, and PM2.5. The average and standard deviation of the THC, CO, and NOx
emission levels of the eleven engines are shown in Table 2-33. The PM2.5 emissions are summarized in
Table 2-37. The sample included both MY2016 and MY2017 engines, ranging in displacement between 7.7
and 14.9 liters, and in rated power between 260 and 605 HP. The default cold start emissions values for
MOVES3 are the mean values shown in the table. The THC and NOx cold start emissions for HHD diesel
engines are higher than those in MOVES2014, while the CO emissions are lower.
Table 2-33: Cold Start Emissions for MY2010 and Later Heavy Heavy-Duty Diesel Engines
Grams per Start
THC
CO
NOx
Mean
0.08
6.6
8.4
Standard Deviation of Data
0.1
5.6
1.7
2.2.1.2.2 Medium-heavy duty
The certification data from six MHD diesel engines were used to develop the THC, CO, and NOx grams
emitted per start. The average and standard deviation of the emissions from the six engines are shown in
Table 2-34. The sample included MY2016 and MY2017 engines, ranging in displacement between 5.1 and
8.9 liters, and in rated power between 230 and 380 HP. The default values for MOVES3 are the mean
values shown in the table. Similar to the HHD engines, the THC and NOx cold start emissions for the MHD
diesel engines are higher than those in MOVES2014, while the CO emissions are lower.
V = 0.055!** + 4.4297
RJ
: = 0.8503
•
•
9
110
-------
Table 2-34 Cold Start Emissions for MY2010 and Later Medium Heavy-Duty Diesel Engines
Grams per Start
THC
CO
NOx
Mean
0.20
2.5
6.4
Standard Deviation of Data
0.2
2.7
1.8
2.2.1.2.3 Light-heavy duty
Analysis of five LHD diesel engines from the certification data determined the grams per start for THC, CO,
and NOx shown in Table 2-35. The sample included MY2016 and MY2017 engines, ranging in displacement
between 3.0 and 6.7 liters, and in rated power between 161 and 330 HP. The default values for MOVES3
are the mean values. The CO and NOx cold start emissions are higher than those in MOVES2014, while the
THC emissions are lower.
Table 2-35 Cold Start Emissions for MY2010 and Later Light Heavy-Duty Diesel Engines (LHD45 and LHD2b3)
Grams per Start
THC
CO
NOx
Mean
0.005
2.47
6.77
Standard Deviation of Data
0.11
2.61
2.24
We are applying the new cold start THC, CO, and NOx emission rates from the 2016 MY and 2017 MY
engines to all 2010 MY and newer engines. The latest tier of HD diesel emission standards completed
phase-in in 2010 MY and the aftertreatment systems on these engines are similar and generally include
both a diesel particulate filter and selective catalytic reduction system.
2.2.1.2.4 Incorporation of Tier 3 Standards for Light Heavy-Duty Diesel
The Tier 3 exhaust emission standards affect light heavy-duty diesel vehicles in the LHD2b3 regulatory class
(regClassID 41). Reductions are applied to start rates for NOx only, phasing in from MY2018 to MY2021 as
previously described for running emissions in Section 2.1.1.5.5. No reductions are applied to THC and CO
rates.
2.2.1.2.5 Model Year Summary
Figure 2-50 through Figure 2-52 display the displays the cold start (operating mode 108) emission rates
across model years for heavy-duty diesel vehicles. The figures show the large difference in start emission
111
-------
rates before and after model year 2010. Model year 2010 corresponds to the implementation of (SCR)
aftertreatment, as well as the different datasets and methodologies.
112
-------
0.20
¦E
CO
+J
O)
0.15-
ro
a:
c
o
CO
0
E
LD
O
X
CO
O 0
"(0
-I—"
o
.10-
.05-
Reg Class
Gliders
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
o.oo-
1980
2000
Model Year
2020
Figure 2-50 Heavy-duty Diesel THC Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year
15-
E
ro
5?
S
0)
ro
a:
c
o
CO
CO
10-
LU 5-
O
O
Reg Class
Gliders
HHD8
Urban Bus
MHD67
LHD45
LHD2b3
1980
2000
Model Year
2020
Figure 2-51 Heavy-duty Diesel CO Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year
113
-------
.CO
3
0)
§5.0-
c
o
V>
to
LU
X
O
2.5-
o.o-
1980
2000
Model Year
2020
Reg Class
Gliders
HHD8
-+¦ Urban Bus
MHD67
LHD45
LHD2b3
Figure 2-52 Heavy-duty Diesel NOx Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year
2.2.2 Particulate Matter (PM2.s)
2.2.2.1 1960-2010 Model Years
Data for particulate matter start emissions from heavy-duty vehicles are limited. Typically, heavy-duty
vehicle emission measurements are performed on fully warmed up vehicles. These procedures bypass the
engine crank and early operating periods when the vehicle is not fully warmed up.
Data for model year 2009-and-earlier vehicles was only available from engine dynamometer testing
performed on one heavy heavy-duty diesel engine, using the FTP cycle with particulate mass collected on
filters. The engine was manufactured in MY 2004. The cycle was repeated six times, under both hot and
cold start conditions (two tests for cold start and four replicate tests for hot start). The average difference
in PM2.5 emissions (filter measurement - FTP cycle) was 0.11 grams. The data are shown in Table 2-36.
114
-------
Table 2-36 Average PM2.5 emissions (grams) from MY 2004 HHD diesel engine tested on the FTP Cycle
PM2.5 emissions (grams)
Cold start FTP average
1.93
Warm start FTP average
1.82
Cold start - warm start
0.11
We use the difference between the cold start and warm start bags to represent the cold start (g/start_ in
MOVES.0 We applied this value to 1960 through 2006 model year vehicles. For 2007 through 2009 model
years, we applied a 90 percent reduction to account for the expected use of DPFs, leading to a
corresponding value of 0.011 g/start. The value is the same for all heavy-duty diesel regulatory classes.
As introduced in Section 2.1.2.1.8, in MOVES, the PM2 5 emission rates are estimated as the elemental
carbon (EC) and non-elemental carbon PM (nonEC). We estimated the EC and nonEC from the total PM2.5
starts rates by applying the EC/PM fraction of 46.4 percent from the PM2.5 speciation profile developed
from the idle mode of the UDDS tests from the E55/59 program for pre-2007 trucks.1 For all 2007+ trucks,
we apply the EC/PM fraction of 9.98 percent from the PM2.5 speciation profile developed from trucks
equipped with diesel particulate filters.1
2.2.2.2 2010-2060 Model Years
The cold start emissions for 2010 model year and later LHD, MHD, and HHD diesel engines have been
updated in MOVES3 based on new data. We updated the cold start particulate matter emission rates
based on the certification data and data analysis methods discussed in Section 2.2.1.2. The resulting cold
start emission rates for each HD diesel engine regulatory group are shown in Table 2-37. For LHD diesel
vehicles, the certification data yielded zero PM2.5 start emissions. We attribute the zero start to the
uncertainty of the data (note the standard deviation shown in Table 2-37 is of similar magnitude or greater
than the HHD and MHD). Instead of using the certification test data, we used the data from MHD diesel to
represent the LHD diesel PM2.5 emission rate in MOVES3, because of the overlap in engines and
aftertreatment systems between the two categories.
Table 2-37: Cold Start PM2.5 Emission Rates for Heavy-Duty Diesel Emissions for 2010+ MY
Grams per Start
HHD
MHD
LHD Test Data
LHD for MOVES3
Mean of Data
0.013
0.008
0.000
0.008
Standard Deviation of
Data
0.029
0.017
0.010
Note:
a Instead of using the test data, we used the data from MHD diesel to represent the LHD diesel PM2.5 emission rate in
MOVES3 as noted above.
115
-------
We are applying the new cold start PM2.5 emission rates from the model year 2016 and 2017 engines to MY
2010 and newer engines because the PM standards are the same and all the MY 2010 and later engines
generally include both a diesel particulate filter (DPF) and selective catalytic reduction (SCR) system.
2.2.2.3 Model Year Summary
Figure 2-52 and Figure 2-53 display the cold start (operating mode 108) emission rates across model years
for heavy-duty diesel vehicles. As expected, large reductions are shown in model year 2007 with the
implementation of diesel particulate filters. Further changes are due to the incorporation of the 2010 and
later certification data.
0.125-
co
0
O)
0)
-t—'
(U
cr
c
o
'(/)
CO
.100-
0.075-
LU
X
LU
03
o
LO 0
CN
0.050-
025-
0.000-
1980
2000
Model Year
2020
Reg Class
Gliders
HHD8
Urban Bus
MHD67
LHD45
— LHD2b3
Figure 2-53 Heavy-duty Diesel PM2.5 Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year. Urban Bus and HHD are equivalent. MHD, LHD45, LHD2b3 are equivalent.
2.2.3 Adjusting Start Rates for Soak Time
The discussion to this point has concerned the development of rates for cold start emissions from heavy-
duty diesel vehicles. In addition, it was necessary to derive rates for additional operating modes that
account for shorter soak times. As with light-duty vehicles, we accomplished this step by applying soak
fractions.
116
-------
In the MOVES input database, operating modes for start emissions are defined in terms of soak time
preceding an engine start. The "cold-start" is defined as a start following a soak period of at least 720
minutes (12 hours) and is represented as opModelD=108. An additional seven modes are defined in terms
of soak times ranging from 3 min up to 540 min (opModelD = 101-107). Table 1-5 describes the different
start-related operating modes in MOVES as a function of soak time. The distribution of vehicle start activity
among the start operating modes is described in the MOVES Vehicle Population and Activity report.9
2.2.3.1 Adjusting Start Rates for Soak Time - MY 2009 and Earlier
The soak adjustment ratios we used for THC, CO, and NOx for MY 2009 and older HD diesel vehicles are
illustrated in Figure 2-54 below. Due to limited data, we applied the same soak ratios that we applied to
1996+ MY light-duty gasoline vehicle as documented in the light-duty emission rate report.8 The soak
adjustments are taken from the non-catalyst soak adjustments derived in a CARB report67 and reproduced
in a MOBILE6 report.68
Soak Time (minutes)
Figure 2-54. Soak Adjustment Ratios Applied to Cold-Start Emissions (opModelD = 108) to Estimate Emissions for
shorter Soak Periods (operating modes 101-107). This figure is reproduced from the Light-Duty Emissions Report8
For light heavy-duty vehicles (regulatory classes LHD2b3 and LHD45), the soak ratios apply to the cold
starts for THC, CO and NOx. For medium and heavy heavy-duty vehicles (regulatory classes MHD, HHD, and
Urban Bus), only the CO soak ratios are applied to the cold-start emissions, because the base cold start THC
and NOx emission rates for medium and heavy heavy-duty emission rates are zero (see Section 2.2.1.1).
The start emission rates entered into MOVES for 2009 and older model year heavy-duty vehicles, derived
from applying the soak ratios are displayed in Table 2-38 for THC, CO, and NOx.
117
-------
Table 2-38. Heavy-Duty diesel THC, CO, and NOx Start Emissions (g/start) by Operating Mode for 2009 and Earlier
Model Year Vehicles
THC
CO
NOx
opModelD
LHD1
Other HD2
LHD
Other HD
LHD
Other HD
101
0.0052
0
0.055
0.64
0.275
0
102
0.0273
0
0.276
3.2
0.760
0
103
0.0572
0
0.607
7.04
1.350
0
104
0.0780
0
0.869
10.08
1.481
0
105
0.0832
0
1.007
11.68
1.481
0
106
0.0949
0
1.090
12.64
1.468
0
107
0.1183
0
1.256
14.56
1.376
0
108
0.1300
0
1.380
16
1.298
0
Notes:
1 LHD refers to LHD2b3 and LHD45
2 Other HD refers to the medium heavy-duty, heavy heavy-duty, and urban bus regulatory classes
The PM2.5 start rates by operating mode are given in Table 2-39 below. They are estimated by assuming a
linear decrease in emissions with time between a full cold start (>720 minutes) and zero emissions at a
short soak time (< 6 minutes).
118
-------
Table 2-39. Particulate Matter Start Emission Rates (g/start) by Operating Mode (soak fraction) for all HD Diesel
vehicles through MY 2009
Operating
Mode
1960-2006
MY
2007-2009
MY
101
0.0000
0.00000
102
0.0009
0.00009
103
0.0046
0.00046
104
0.0092
0.00092
105
0.0138
0.00138
106
0.0183
0.00183
107
0.0549
0.00549
108
0.1099
0.01099
2.2.3.2 Adjusting Start Rates for Soak Time - MY 2010 and Later
As described in the preceding section, the start rates are based on data collected from light-duty vehicles
in the 1990's. The question arose as to whether they could be considered applicable to heavy-duty diesel
vehicles with aftertreatment systems designed to meet the 2007/2010 exhaust emissions standards. To
address this question, we initiated a research program in 2016, with the goal of examining the
relationships between soak time and start emissions for a set of heavy-duty vehicles. Two new test
programs were conducted to revise the 2010 MY and later soak curves for heavy-duty diesel vehicles in
MOVES3. The testing consisted of both chassis and onroad testing of MY 2015 and MY 2016 vehicles.
The first test program included a MY 2015 day-cab tractor with a MY 2015 HHD diesel engine tested on a
heavy-duty chassis.69 The vehicle was relatively new and had 10,000 miles on the odometer. The testing
consisted of running two repeats of a transient drive cycle developed by the National Renewable Energy
Laboratory (NREL). The vehicle speed trace is shown below in
Figure 2-55. Prior to each soak test, the vehicle was first run through two of the NREL cycles. Then the
engine was shut off for a specified amount of time to reflect the soak periods shown in Figure 2-55. At least
two repeats were conducted for each soak period. The emission measurements included dilute gaseous
measurements and triplicate particulate matter filters.
119
-------
Double NREL Cycle
60
1/JrH ¦J^iNCOmOsl^Os£i'Hr*v 00 ^ Q
rHrHiHrHiH rl rl rl
Time (seconds)
Figure 2-55 National Renewable Energy Laboratory's Heavy-Duty Vocational Transient Cycle
The MOx, CO, THC, and PM2.s emission results in terms of grams or mg per mile from the tests over a range
of soak periods are shown in Figure 2-56 through Figure 2-59.
8.0
7.0
^ 6.0
j= 5.0
_bD
£ 4.0
o
jj 3.0 1
i§ 2.0 |
1.0
0.0 1
c
NOx
%•
•
•
•
•
•
•
•1
f
I
I
m
> 10 20 30 40 50 50 70 B0 90 It
Soak Time (hours)
Figure 2-56 MY 2015 Heavy-Duty Vehicle NOx Emissions by Soak Time
120
-------
0.06
0.05
OJ
~ 0.04
£ 0.03
o
i/i
~ 0.02
lu
0.01
o.oo t
t
HC
f
*
»
9
•
•
•
•
•
•
•
• •
•1
t
) 10 20 30 40 50 50 70 SO 90 100
Soak Time (hours)
Figure 2-57 MY 2015 Heavy-Duty Vehicle THC Emissions by Soak Time
CO
0.50
0.45
„ 0.40
aj
~ 0.35
S °'30
£ 0.25
2 0.20
~ 0.15
m 0.10
0.05 ,
0.00
c
•
•
•
A A
V
•
—•—
•
n
!•
t
10
20 30
40 50 60
Soak Time (hours)
70
so
90
100
Figure 2-58 MY 2015 Heavy-Duty Vehicle CO Emissions by Soak Time
121
-------
a r.
PM
8.0
7.0
aj
~ 6.0
If 5.0
I <° '
.2 3.0 ,
' <
2.0 '
1.0
0.0
(
•
L.
y
%n
•
1
/
•
• *
X)
10 20 30 40 50 60 70 B0 90 It
Soak Time (hours)
Figure 2-59 MY 2015 Heavy-Duty Vehicle PM2.5 Emissions by Soak Time
In addition to the chassis testing, onroad testing was conducted using a portable emissions measurement
system (PEMS).70 The emissions data gathered by the PEMS in this test program only included the gaseous
emissions, not PM data. A MY 2016 work van with a diesel engine was tested on the road. The vehicle was
soaked and started within a laboratory under controlled temperatures. All onroad testing occurred with
ambient temperatures over 50 degrees F. Each test began with 10 seconds of idle followed by driving a
defined "soak route." A typical vehicle speed profile from the route is shown in Figure 2-60. The route
consisted of approximately 700 seconds of driving in a neighborhood/urban environment over
approximately 2.7 miles.
122
-------
Soak Route Speed Profile
TdfrCouster iseeomJii
Figure 2-60 Onroad Soak Drive Route
The emission results, in terms of total emissions over the route, from the onroad tests are shown in
Figure 2-61 through Figure 2-63.
3.0 ,
2.5
3
c 20
o
ui
J2 1.5
E
Lli
~2 1-0
£
0.5
0.0
(
Total NOx Over Cycle
•
•
•
a
•
•
•
w
>
> 5 10 15 20 25 30
Soak Time (hours)
Figure 2-61 MY 2016 Heavy-Duty Vehicle NOx Emissions by Soak Time
123
-------
0.S
0.7
"to 0.6
i/i
J 05
i/i
l/i 0.4
E
- 0.3
ra
£ 02
0.1
0.0 '
(
Total HC Over Cycle
•
•
••
• •
• * •
A
f
> 5 10 15 20 25 30
SoakTime (hours)
Figure 2-62 MY 2016 Heavy-Duty Vehicle THC Emissions by Soak Time
o n
Total CO Over Cycle
S.O
— 7.0
£ 6.0
D
S 5.0
i/i
E 4.0
LJ
ra 3.0
+j
,o
2.0
1.0
0.0
(
•
A
*
•
•
A
A A
W
• • •
m
f
w
D
5 10 15 20 25 3
SoakTime (hours)
Figure 2-63 MY 2016 Heavy-Duty Vehicle CO Emissions by Soak Time
The soak emission adjustment ratios were calculated using a multi-step process based on the chassis test
and onroad test results. First, the total emissions over the route or drive cycle were averaged for each soak
period for each pollutant (NOx, THC, CO) for each vehicle. Then the start emissions for each soak period
were determined by subtracting the average total emissions from the tests with the 3 minute soak time
124
-------
from the emissions from the specific soak period. The ratios for soak period operating modes 102 through
108 were calculated based on the average start emissions of the soak period divided by the average start
emissions of the cold start (>12 hours) soak period. The soak fractions for the operating mode 101 were
determined by extrapolating the value from the operating mode 102 result using the proportional
difference in time between the midpoints of each operating mode 101 and 102 soak times. In other
words, soak fraction for operating mode 102 was multiplied by the ratio of 3 minutes divided by 18
minutes (the midpoint times of operating mode 101 and 102). The NOx, CO, and THC soak period ratio
results for each vehicle are shown below in Figure 2-64.
HD Diesel Truck Start Emission Ratio
1.20
1.00
.2 0.80
to
cm
O 0.50
"175
j—
£ 0.40
0.20
0.00
/
m "1'
-V
/
$
/ f
4
,*r.
0 100 200 300 400 500
Soak Time (min)
600
700
800
HD Van NOx
-•— Day Cab NOx
HD Van CO
¦ Day Cab CD
HD Van HC
--•--Day Cob HC
Figure 2-64 Soak Emission Ratios from a MY 2015 HD Day-Cab and a MY 2016 HD Van
The 2010 MY and later heavy-duty diesel soak ratios for MOVES3 were determined by averaging the results
from the two trucks. The resulting soak adjustment ratios are shown in Table 2-40. The soak adjustment
ratios are applied to all heavy-duty diesel regulatory classes because the two trucks tested cover the range
of HD diesel regulatory classes.
125
-------
Table 2-40 HD Diesel Engine Soak Ratios for MY 2010 and Newer
Operating
Mode
Description
NOx
CO
THC
101
Soak Time < 6 minutes
0.01
0.00
0.00
102
6 minutes < Soak Time < 30 minutes
0.04
0.03
0.02
103
30 minutes < Soak Time < 60 minutes
0.13
0.06
0.05
104
60 minutes < Soak Time < 90 minutes
0.33
0.02
0.24
105
90 minutes < Soak Time < 120 minutes
0.40
0.12
0.36
106
120 minutes < Soak Time < 360 minutes
0.37
0.32
0.33
107
360 minutes < Soak Time < 720 minutes
0.62
0.38
0.55
108
720 minutes < Soak Time
1.00
1.00
1.00
The PM2.5 start rates by operating mode for MY 2010 and newer vehicles are given in Table 2-41 below.
They were updated in MOVES3 using a linear interpolation based on the new cold start data (certification
data discussed in Section 2.2.2.2) for Operating Mode 108. They are estimated by assuming a linear
decrease in emissions with time between a full cold start (>720 minutes) and zero emissions at a short soak
time (< 6 minutes). This approach is consistent with the approach taken for MY 2009 and older vehicles, as
described in Section 2.2.3.1. We did not revise the approach because we obtained PM2.5 data for only one
of the trucks and it showed mixed soak effect results.
Table 2-41 PM2.5 Start Emission Rates (g/start) by Regulatory Class and Operating Mode (soak fraction) for all MY
2010 and newer HD Diesel Vehicles
Operating
Mode
HHD and
Urban Bus
MHD
LHD2b3 and
LHD45
101
0.00000
0.00000
0.00000
102
0.00163
0.00100
0.00100
103
0.00325
0.00200
0.00200
104
0.00488
0.00300
0.00300
105
0.00650
0.00400
0.00400
106
0.00813
0.00500
0.00500
107
0.00975
0.00600
0.00600
108
0.01300
0.00800
0.00800
2.2.3.3 Adjusting Start Rates for Ambient Temperature
The ambient temperature effects in MOVES are used to estimate the impact ambient temperature has on
cooling the engine and aftertreatment system on vehicle emissions. The temperature effect is the greatest
for a vehicle that has been soaking for a long period of time, such that the vehicle is at ambient
temperature. Accordingly, the impact of ambient temperature should be less for vehicles that are still
warm from driving. The emission adjustments report discusses the impact of ambient temperature on cold
126
-------
start emission rates (operating mode 108).64 The ambient temperature effects for starts with warm and
hot soaks (operating mode 101-107) are documented below and recorded in the MOVES
startTempAdjustment table.
Because the THC temperature effects in MOVES are modeled as additive adjustments, the adjustment
calculated for cold starts needs to be reduced for warm and hot starts. Due to lack of data, we multiply the
soak fractions described earlier in Figure 2-54 for pre-2007 trucks by the additive cold temperature effect
for the 12-hour cold start (operating mode 108) to obtain cold start temperature adjustments for the
warm and hot soaks starts (operating mode 101 through 107) for all model years.p The additive cold start
adjustment for THC emission factors are displayed in Table 2-42, along with the soak fractions applied.
These additive THC starts are applied to all diesel sources in MOVES, including light-duty diesel (regulatory
class LDV and LDT). There are currently no diesel temperature effects in MOVES for PM2 5, CO, and NOx.
Table 2-42 THC Diesel Start Ambient Temperature Adjustment by Operating Mode
Operating
mode ID
Start Temp Adjustment
Soak fraction
101
-0.0153x(Temp-75)
0.38
102
-0.0152x(Temp-75)
0.37
103
-0.0180x(Temp - 75)
0.44
104
-0.0201x(Temp - 75)
0.5
105
-0.0211x(Temp - 75)
0.52
106
-0.0254x(Temp - 75)
0.62
107
-0.0349x(Temp - 75)
0.86
108
-0.0406x(Temp - 75)
1
2.2.4 Start Energy Rates
p The temperature effects from pre-2010 technology engines are applied to all model years. We plan to update the
temperature effects by operating mode for 2010 and later model year vehicles in future version of MOVES using the
data from 2010 and later engines.
127
-------
The start energy rates (in units of kJ) were developed for MOVES200471, and updated in MOVES2010 as
documented in the MOVES2010a energy updates report.58 Figure 2-65 displays the cold starts in grams of
C02 emissions calculated from the energy rates using the carbon content for conventional diesel fuel as
documented in the MOVES3 Greenhouse Gas and Energy Report.3
As shown, there is more detail in the pre-2000 energy rates. The spike in C02 g/start for model years 1984-
1985 reflects variability in the data used to derive starts, which was consistent with the more detailed
approach used to derive the pre-2000 energy rates in MOVES2004. The only updates to the start energy
rates post-2000 is the impact of the Phase 1 Heavy-Duty GHG standards, which began phase-in in 2014 and
have the same reductions as the running energy rates as presented in Table 2-24 and Table 2-26. It is
worth noting that unlike the Phase 1 HD GHG standards, the technologies projected for meeting the Phase
2 HD GHG standards are not expected to have an impact on start energy rates. Therefore, the start energy
rates are constant after MY 2018 (the first year of full phase-in of the HD Phase 1 rule).
500-
¦e
CO
-52
O)
400
03
cr
•£ 300-
(/)
CO
200-
LU
CN
O
O
o
' i
a)
£-100-
o
Reg Class
Gliders
— HHD8
Urban Bus
— MHD67
LHD45
LHD2b3
1980
2000
Model Year
2020
Figure 2-65 Heavy-Duty Energy Cold Start C02 Rates (Operating Mode 108) by Model Year and Regulatory Class
The start energy rates are adjusted in MOVES to account for increased fuel consumption required to start a
vehicle at cold ambient temperatures. The temperature effects are documented in the MOVES2004 Energy
Report.71 Additionally, the energy consumption is reduced for starts that occur when the vehicles have
soaked for a short period of time. The soak fractions used to reduce the cold start energy consumption
emission rates are provided in Table 2-43. These fractions are used for all model years and regulatory
classes of diesel vehicles.
128
-------
Table 2-43 Fraction of Energy Cconsumed at Start of intermediate Soak Lengths compared to the Energy Consumed
at a Full Cold Start (operating mode 108)
Operating
Mode
Description
Fraction of energy consumption
compared to full cold start
101
Soak Time < 6 minutes
0.013
102
6 minutes < Soak Time < 30 minutes
0.0773
103
30 minutes < Soak Time < 60 minutes
0.1903
104
60 minutes < Soak Time < 90 minutes
0.3118
105
90 minutes < Soak Time < 120 minutes
0.4078
106
120 minutes < Soak Time < 360 minutes
0.5786
107
360 minutes < Soak Time < 720 minutes
0.8751
108
720 minutes < Soak Time
1
The energy rates for heavy-duty starts have not been updated due to relatively small contribution the
starts have to the energy inventory. Table 2-44 displays the relative contribution of total energy
consumption estimated from a national run of MOVES for calendar year 2016, using a draftversion of
MOVES3 used for the preliminary CTI analysis.15 As shown, the estimated energy consumed due to starts is
very small in comparison to the energy use of running activity.
Table 2-44. Relative contribution of total energy consumption from each pollutant process by regulatory class for
heavy-duty diesel vehicles in calendar year 2016
processID
processName
LHD<14K
LHD45
MHD
HHD
Urban
Bus
Gliders
1
Running
Exhaust
98.5%
99.3%
99.42%
98.85%
99.7%
98.63%
2
Start Exhaust
1.5%
0.7%
0.55%
0.10%
0.3%
0.05%
90
Extended Idle
Exhaust
0.03%
1.03%
1.27%
91
Auxiliary Power
Exhaust
0.00%
0.03%
0.05%
129
-------
2.3
Extended Idling Exhaust Emissions
In the MOVES model, extended idling is idle operation characterized by long duration idle periods (e.g., > 1
hourq), typically overnight, including higher engine speed settings and extensive use of accessories by the
vehicle operator. Extended idling most often occurs during rest periods by long-haul trucking operators
where the truck is used as a residence (sometimes referred to as "hotelling"). Operators idle to power
accessories such as air conditioning systems or heating systems. Heavy-duty engine and truck
manufacturers recommend trucks not idle at low engine speeds for extended periods, because it can
"create engine wear and carbon soot buildup in the engine and components."72 Additionally, idling for
extended periods allows the vehicle's exhaust to cool below the effective temperature required for
emission aftertreatment systems in modern trucks such as selective reduction catalysts, and diesel
oxidation catalysts. As a result, extended idle is treated as a separate emission process in MOVES which
uses a different emission rate than the idling that occurs during the running emission process.
Extended idling does not include vehicle idle operation that occurs during normal road operation, such as
idling at a traffic signal or the "off-network" idle that migh occur during a delivery. Although frequent stops
and idling can contribute to overall emissions, these modes are included in the normal vehicle hours of
operation. Extended idling is characterized by idling periods that last hours, rather than minutes.
In the MOVES model, diesel long-haul combination trucks (sourceTypelD 62) are the only fuel and source
type combination assumed to have extended idling activity. These trucks are only associated with diesel
MHD, HHD and Gliderr regulatory classes. As an alternative to extended idling, long-haul truck operators
can also use auxiliary power units (APUs) to power their cabin and accessories during hotelling. The
emission rates for auxiliary power units (APUs) are discussed in Section 2.4.
Extended idle emission rates for THC, CO, NOx and PM2.5 were updated in MOVES3 for all model years.
Energy rates were updated for 2007 and later model years. Separate analyses were conducted using
different data sets to derive extended idle emission rates for pre-2007 (Section 2.3.1) and 2007 and later
long-haul combination trucks (Section 2.3.2). For each range of model years, MOVES applies different data
and assumptions regarding the impact of accessory use, frequency of high idle engine speed, and impacts
of tampering and mal-maintenance to calculate extended idle emission rates.
q The default hotelling activity in MOVES3 is estimated from telematics data assigning all idle events > 1 hr from long-
haul combination trucks to extended idling.9
r Glider extended idle emission rates are documented in Section 2.5
130
-------
2.3.1 1960-2006 ModeI Years
The MOVES extended idling emission rates for pre-2007 model years were derived from data collected in
several distinct test programs under different types of idle conditions. For MOVES3, weightings were
adjusted from those in previous versions of MOVES to better account for new information on typical
extended idling engine idling speeds and loads. These adjustments are described below in Section 2.3.1.2.
Appendix D summarizes the data and calculations for the pre-2007 model years.
2.3.1.1 Data Sources
The references included in this section provide more detailed descriptions of the data and how the data
were obtained:
• Testing was conducted on 12 heavy-duty diesel trucks and 12 transit buses in Colorado by
McCormick et al.73 Ten of the trucks were Class 8 heavy-duty semi-tractors, one was a Class 7
truck, and one of the vehicles was a school bus. The model years ranged from 1990 through 1998.
Typical Denver area wintertime diesel fuel was used in all tests. Idle measurements were collected
during a 20-minute time period. All testing was done at 1,609 meters above sea level (high
altitude).
• Testing was conducted by EPA on five trucks in May 2002 (Lim et al.).74 The model years ranged
from 1985 through 2001. The vehicles were put through a battery of tests including a variety of
idling conditions.
• A total of 63 trucks (nine in Tennessee, 12 in New York and 42 in California) were tested over a
battery of idle test conditions including with and without air conditioning (Irick et al.).75 Not all
trucks were tested under all conditions. Only results from the testing in Tennessee and New York
are described in the IdleAire report (Irick et al.).75
• The California test data was collected on 42 diesel trucks in parallel with roadside smoke opacity
testing (Lambert)76. All tests conducted by the California Air Resources Board (CARB) at a rest area
near Tulare, California in April 2002 are described in the Lambert76 Clean Air Study. All analytical
equipment for all testing at all locations was operated by Clean Air Technologies.
• Fourteen trucks were tested as part of the E-55/59 Coordinating Research Council (CRC) study of
heavy-duty diesel trucks with idling times either 900 or 1,800 seconds long.77
• The National Cooperative Highway Research Program (NCHRP)78 obtained the idling portion of
continuous sampling during transient testing to determine idling emission rates on two trucks.
• A total of 33 heavy-duty diesel trucks were tested in an internal study by the City of New York
(Tang et al.)79. The model years ranged from 1984 through 1999. One hundred seconds of idling
were added at the end of the WVU five-mile transient test driving cycle.
• A Class 8 Freightliner Century with a 1999 engine was tested using EPA's onroad emissions testing
trailer based in Research Triangle Park, North Carolina (Brodrick).80 Both short (10 minute) and
longer (five hour) measurements were made during idling. Some testing was also done on three
older trucks.
131
-------
• Five heavy-duty trucks were tested for particulate and N0X emissions under a variety of conditions
at Oak Ridge Laboratories (Storey et al.).81 These are the same trucks used in the EPA study (Lim et
al.).
• The University of Tennessee (Calcagno et al.) tested 24 1992 through 2006 model year heavy-duty
diesel trucks using a variety of idling conditions including variations of engine idle speed and load
(air conditioning).65
2.3.1.2 Analysis
We used the data sources referenced above to estimate the emission rates for particulate matter (PM2.5),
oxides of nitrogen (NOx), hydrocarbons (THC), carbon monoxide (CO) and carbon dioxide (C02). The data
were grouped by truck and bus and by idle speed and accessory usage to develop emission rates
representative of extended idle emissions.
The important conclusion from the analysis was that truck operator behavior plays an important role when
assigning emission rates to periods of extended idling. Factors such as accessory use and engine idle speed,
which are controlled by operators, affect engine load and emission rates during extended idling. The
impacts of other factors, such as engine size, altitude, model year within MOVES groups, and test cycle are
negligible.
We first evaluated the studies on engine idle speed. NREL's review of owner's manuals found that several
heavy-duty engine manufacturers recommend use of fast idle (> 1000 rpm) if the engine needs to idle for
extended periods.82 In a 2004 UC-Davis survey (Lutsey et al. 2004), respondents' average engine idle speed
was 866 rpm, with small peaks around 650 and 1000 rpm.83 About one-third of the respondents indicated
they changed their idle speed from its usual setting, which is consistent with the distribution of the
responses where about one-third of the idle engine speeds reported were 1000 rpm or faster. A 2015
study by Hoekzema (2015)84 suggested that even fewer trucks operated in a high idle condition. Drivers
surveyed for this study reported high idle operation (> 1000 rpm) just 18 percent of the time during idling
periods of an hour or more. Additionally, Hoekzema (2015) cited similar studies representing 764 trucks
that averaged engine speeds of 886 rpm during extended idle. Therefore, in MOVES3, we reduced the
amount of high idle from 100 percent assumed in MOVES2014 to 33 percent, to better match the
references noted above.
The use of accessories (e.g., air conditioners, heaters, televisions, etc.) provides recreation and comfort to
the operator and increases load on the engine. There is also a tendency to increase idle speed during long
idle periods for engine durability. The emission rates estimated for the extended idle in MOVES assume
both accessory use and engine idle speeds set higher than used for "curb" (non-discretionary) idling. We
classify the extended idling that does not employ high speed idle without additional auxiliary loads as "curb
idle."
132
-------
Emissions data from the references in the data sources section (2.3.1.1) was classified into one of three
idle conditions. The first condition, which has a low engine speed (<1,000 rpm) and no air conditioning is
representative of curb idle. The second condition is representative of extended idle with higher engine
speed (>1,000 rpm) and no air conditioning. The third represents an extended idle condition with higher
engine speed (>1,000 rpm) and air conditioning. For the purpose of this analysis, the load placed on the
engine due to air conditioning is assumed to represent all forms of accessory load that may be used during
hotelling.
Note that some of the idle tests are of short duration. We believe it is reasonable to classify the short-
duration tests as curb idle in our calculations of extended idle emissions. We are using the short-duracton
idle tests from the pre-2007 MY vehicles because idle emissions stabilize more quickly than later model
years because the pre-2007 vehicles lack the emission aftertreatment technologies that can lose
effectiveness as exhaust cools during longer idle periods.
For 1990 and earlier, we developed curb idle emission rates based on the analysis of the 18 heavy-duty
diesel trucks from 1975-1990 model years used in the CRC E-55/59 study and one MY 1985 truck from the
Lim study. The curb idle rates were then adjusted using ratios from 1991-2006 trucks to estimate the
elevated NOx emission rates characteristic of higher engine speed and accessory loading of extended idle.
In particular, as summarized in the tables in Appendix D, data from 188 vehicles were used to estimate
curb idle NOx emission rates for 1991-2006 model year heavy-duty diesel trucks. The curb idle NOx
emission rate of 91 g/hr was calculated by weighting the average NOx emission rate from each test by the
number of vehicles tested. Four studies and results from 31 vehicles included higher idle engine speed and
air conditioner use, which resulted in a weighted idle NOx emission rate of 227 g/hr. The ratio of the 1991-
2006 MY NOx emission rate from curb idle to idle with high engine speed and A/C was applied to the 1990
and earlier model year curb idle rate to get the calculated 1990 and earlier NOx emission rate with high
engine speed and A/C. A similar strategy was applied to the THC, CO, and C02 emission rates for 1990 and
earlier model years.
For both the MY 1960-1990 and 1991-2006 vehicles, using the data summarized in Appendix E, adjusted
emission rates were calculated for each pollutant by weighting the overall "high speed idle, A/C on" results
by 0.33 and the "low speed idle, A/C off" (i.e., curb idle) results by 0.67 to account for the fraction of idling
at high and low engine speeds.
The NOx, THC, CO, and PM2.5 emission rates from this data analysis are primarily from diesel HHD trucks. In
MOVES2014, we calculated the MHD extended idle emission rates as half of the corresponding HHD
emission rates. However, a study by Khan et al. (2009)85 found that MHD and HHD trucks had similar
133
-------
emission rates during extended idle. Consequently, MOVES3 applies the same extended idle emissions
rates to MHD and HHD, as shown in Table 2-45.
MOVES stores PM25 emission rates according to elemental carbon (EC) and non-elemental carbon
(NonECPM), but the data sources used to calculate the extended idle emission rates reported only total
PM2.5. As mentioned in Section 2.1.2.1.8, an EC/PM fraction of 46.4 percent is applied for the running
exhaust idle operating mode (opModelD 1), and we also apply it to extended idle. The resulting EC and
NonECPM rates are also shown in Table 2-45.
No adjustment to the rates are made to account for tampering and mal-maintenance (T&M) because the
pre-2007 trucks do not have the exhaust aftertreatment technologies that are anticipated to see large
emission increases when they are tampered or mal-maintained. While the 188 trucks used for these
estimates may not fully represent real-world emssion deterioration, they do include real-world vehicles at
a variety of ages and conditions and thus it would be "double-couting" to apply the exhust running T&M
effects to these rates.
Table 2-45. Pre-2007 Extended idle emission rates (g/hour) in MOVES by pollutant for MHD and HHD
Model Year
Groups
NOx
THC
CO
PM2.5
EC
Non-ECPM
Pre-1991
69.3
49.8
50.8
5.39
2.50
2.89
1991-2006
136
25.6
55.0
2.48
1.15
1.33
2.3.2 2007-2060 Model Years
The 2007 and later model year extended idle emission rates were updated in MOVES3 using the following
data sources and analysis.
134
-------
2.3.2.1
Data Sources
The extended idle emission rates for model year 2007 and later heavy-duty diesel combination long-haul
trucks (sourceTypelD 62) diesel emission rates in MOVES3 were updated based on two test programs
measuring extended idle emissions from HHD diesel trucks. The Texas Transportation Institute (TTI) tested
extended idle emission from 15 heavy-duty diesel tractors ranging from model year 2005s to 2012.86
Another study conducted by California Air Resources Board (ARB)87 tested five tractors (engine model
years 2007 and 2010). As discussed in the analysis section (Section 2.3.2.2), the four MY 2005 and 2006
engines included in the TTI study are included in the development of the 2007 and later model year
emission rates for THC, CO, NOx, and energy becaue there is no noticeable differences in the emission rate
for these model years from comparable MY 2007 and later engines. For PM2.5, these engines are only used
for comparison and to develop T&M adjustment factors.
The study (TTI or ARB), engine model year, engine manufacturer, odometer, the NOx certification level,
California Clean Idle certification, and engine aftertreatment are listed for each of the trucks in Table 2-46.
The last three columns in Table 2-46 are taken from the California Executive Order certification database.88
NOx certification level (g/bhp-hr) is the standard to which the engine was certified. Some 2010 and later
engines were certified above the 0.2 g/bhp-hr NOx 2010 federal standard due to the emissions averaging,
banking and trading (ABT) program, and EPA allowance of nonconformance penalty (NCP) engines in
2012.89 In these cases, the family emission limit for which the vehicle was certified is reported in Table
2-46. California Clean Idle Certification was implemented in 2008 and allows engines that are certified to a
30 g/hr idle NOx standard to idle beyond the 5-minute idle limit initiated in 2008 in California. The
aftertreatment column in Table 2-46 indicates whether the engine was certified with an oxidation catalyst
(OC), diesel particulate filter or periodic trap oxidizer (DPF), and/or selective catalytic reduction (SCR)
system.
s Although 2005-2006 model year engine data was available at the time of the MOVES3 MY 2007+ analysis, we lacked
the time and resources to incorporate them into the pre-2007 emission rates.
135
-------
Table 2-46. HHD Diesel Tractors Used to Update the MY 2007 and Later Extended Idle Emission Rates
Study
Engine
MY
Engine
Odometer
NOx cert
(g/bhp-hr)
Clean
Idle
Certified?
Aftertreatment
TTI
2005
Caterpillar
484,550
2.4
No
OC
TTI
2006
Cummins
505,964
2.4
No
TTI
2006
Volvo
640,341
2.4
No
TTI
2007
Cummins
406,740
1.2
No
OC, DPF
ARB
2007
Cummins
390,000
2.2
No
OC, DPF
ARB
2007
DDC
10,700
1.2
No
OC, DPF
TTI
2008
Cummins
353,945
2.4
Yes
OC, DPF
TTI
2008
Mack
82,976
1.2
Yes
DPF
TTI
2009
Mack
96,409
1.2
Yes
OC, DPF
TTI
2010
Mack
89,469
0.2
Yes
OC, DPF, SCR
TTI
2010
Navistar
73,030
0.5
Yes
OC, DPF
TTI
2010
Navistar
57,814
0.5
Yes
OC, DPF
TTI
2010
Navistar
10,724
0.5
Yes
OC, DPF
ARB
2010
Cummins
13,500
0.35
Yes
OC, DPF, SCR
ARB
2010
Navistar
70,000
0.5
Yes
OC, DPF
ARB
2010
Volvo
68,000
0.2
Yes
OC, DPF, SCR
TTI
2011
Mack
95,169
0.2
Yes
OC, DPF, SCR
TTI
2012
Mack
6,056
0.2
Yes
OC, DPF, SCR
TTI
2012
Mack
11,989
0.2
Yes
OC, DPF, SCR
TTI
2012
Mack
25,148
0.2
Yes
OC, DPF, SCR
The 15 trucks from the TTI program were tested in an environmental chamber under hot and cold
conditions to represent summer conditions in Houston, TX and winter conditions in the Dallas-Fort Worth
area. The test data we used in this analysis were the measurements taken after a twelve-hour soak, where
the vehicle had idled for at least one hour, and the vehicle had reached a 'stabilized' idling condition. The
vehicles were tested at the engine load required to run the heater or air conditioning under the cold winter
or hot summer conditions (see Table 2-47) but were not commanded to be in the high idle state.
136
-------
While the TTI tests included idling after different soak lengths and 'commanded high idle' for engines
capable of idling with an engine speed approximately 400 rpm higher than their standard idle speed, we
decided not to use the 'commanded high idle' emission rates for several reasons:
1) Six of the fifteen TTI trucks were not able to be commanded into high idle.
2) The 'stabilized' idling emission tests did contain some high idle that appears representative of
automatic engine control strategies for 2007 and later trucks. Two of the trucks included high idle
during the winter stabilized tests due to automatic engine control strategies. We assume that for
2007 and later technology trucks, operators and manufacturers rely on automatic engine control
strategies rather than the vehicle operators to employ high idle conditions'. Because most of the
engines did not use high engine speeds to power the heater/air conditioner during the
winter/summer conditions, we assume this engine operation of MY 2007 and later trucks is also
representative of in-use operation.
3) The emissions impact of "commanded" high idle versus stabilized idle was not as pronounced as
observed in the pre-2007 trucks. For the TTI study, the high idle NOx rates were only ~36 percent
higher than the stabilized emission rates. By using the stabilized emission rates, we are using
emission rates that are not much different than the "commanded" high idle emission rates.
For these reasons, the summer and winter stabilized conditions were deemed to be the best estimate of
real-world extended idle emissions. The 'stabilized' idle emission rates (g/hr) for the winter and summer
conditions, are reported in Figure 2-66 through Figure 2-70.
'As discussed earlier, our assumptions for pre-2007 trucks are different.
137
-------
Table 2-47. Ambient Test Conditions for the TTI Extended Idle Tests
Test ID
Temperature
Relative Humidity
Auxiliary Load
Hot (Summer)
100'F (37.8TC)
70%
Air conditioning
Cold (Winter)
3Cf F (-l.rc)
N/A
Heating System
ARB tested five trucks on a chassis dynamometer on the ARB HHDDT 4-mode cycle, reporting the g/hr
results from the 10-minute 'Idle' mode. Before testing the 'Idle' mode, the vehicle was first warmed on a
pre-conditioning cycle, and then soaked for 10-20 minutes.90 Additional test conditions were not reported
by ARB, but we assumed that the ARB vehicles were tested at moderate temperatures, with no auxiliary
loading. Thus, we treated the ARB data as more representative of an extended idling truck that did not
require significant A/C or heating system auxiliary loading on the engine, where the extended idling
occurred shortly after active driving by the main engine.
2.3.2.2 Analysis
To develop the revised extended idle emission rates for MOVES3, we averaged the emission rate from each
of the tests, within model year ranges that represent engine and aftertreatment technology groups that
have similar impacts on extended idle emissions. Where possible, we used all 35 tests (15 trucks x 2
conditions = 30 TTI tests, and 5 ARB tests). Because there were more TTI tests, the average within each
model year group is weighted significantly towards the TTI tests. We chose to weight each test equally,
because we believe the TTI data are more representative of real-world extended idle conditions, because
they were tested with auxiliary loads at non-standard 'lab' temperatures.
The individual test results and the average emission rates by model year group are presented in the
following figures (Figure 2-66 to Figure 2-70). Within each figure, the tests are distinguished according to
the test condition - 'hot' and 'cold' conditions represent the tests from the TTI test program; 'lab' test
condition are the tests from the ARB test program. Additionally, we indicate if the test was from a truck
equipped with SCR or not, which we found was the most useful aftertreatment classifier to determine
engine model year groups.
For C02, CO, and NOx, we do not model any increase in emissions to account for deterioration, including
tampering, of the engines or emission control systems, because we did not observe strong effects of the
emission control on the extended idle emission rates for these pollutants - the aftertreatment technology
(oxidation catalyst, selective catalytic reduction systems) may not be fully functional during the extended
idle conditions, due to lower exhaust temperature occurring at extended idle. On the other hand, for THC
and PM2.5 emissions, we adjust the model year group emission rates to account for deterioration of the
aftertreatment systems, as discussed in more detail below.
138
-------
Figure 2-66 displays the C02 individual test results. No trend with respect to aftertreatment or model year
is observed (nor was one expected). The emissions from cold tests tend to be higher than the hot tests,
which are both higher than the ARB laboratory tests. Two of the cold tests have extended idle emission
rates > 10,000 g/hr which is likely due to higher engine rpm for these engines during the cold tests. TTI
observed that some engines have an engine control strategy, termed "cold ambient protection," which
increases the idle engine speed at cold temperature to warm the coolant temperature and protect against
engine wear. We calculated an average C02 extended idle emission rate for all 2007 and later trucks by
using all the data and treating each test equally across all model years.
The C02 extended idle emission rate is used to derive the energy and fuel consumption extended idle rate
of 97,084 kJ/hr and 0.71 gallons-diesel/hr, respectively. We used the conversion factor of 0.0736 g C02/kJ
and 10,045 g C02/gallon from B3.4 biodiesel (3.4% percent biodiesel blend) highway diesel reported from
the MOVES GHG and Energy Report.3
C02
15000-
3
w10000-
c
o
w
OT
'E
a;
2 5000-
condition
s
i
A
A
•
hot
•
cold
•
lab
SCR
•
0
A
1
A
~
2006
2008
2010
2012
Figure 2-66. C02 Emission Rates from the TTI and ARB Programs by Engine Model Year, and Average Emission Rate
(line) based on all the data.
Note:
Within "condition," "hot" refers to the summer conditions from the TTI tests, "cold" refers to the winter conditions
from TTI, and "lab" refers to the laboratory tests conducted by ARB. For SCR, 0 means the truck does not have a
selective catalytic reduction system (SCR), and 1 means the truck has SCR.
Figure 2-67 displays the CO individual test results. No trend is observed with respect to model year or use
of aftertreatment. The laboratory ARB tests are lower than the TTI tests, which could be due to the lower
139
-------
fuel consumption of the tests. The CO emission rate is slightly lower than the MOVES3 emission rate for
1990-2006 MY of 55 g/hr. Similar to C02, a single average emission rate is calculated for all the tests results
and is applied to all 2007 and later model years.
CO
•
•
•
: =
• •
~
1 1
•
~
A
A
t
~
condition
•
hot
•
cold
•
lab
SCR
•
0
A
1
2006
2008
2010
2012
2014
Figure 2-67. CO Emission Rates from the TTI and ARB Programs by Engine Model Year, and Average Emission Rate
(line) Based on All the Data
Figure 2-68 displays the NOx individual test results. We initially expected the data to show a decrease in the
extended idle emission rates beginning in MY 2008 to account for the California Clean Idle Certification (all
MY 2008 and later trucks were clean-idle certified). However, no reduction was observed. We also
expected to observe a decrease in 2012, with the full implementation of SCR, but this was also not the
case. Therefore, we calculated average NOx emission rates for two model year groups (2005-2009) and
(2010 and later model years) as represented by a solid line in Figure 2-68. The MY 2005-2009 rates
calculated in this analysis are applied to the 2007-2009 model years. Given the variability of the data, the
2007-2009 average rate of 100 g/hr compares well to the MY 1991-2006 rate of 136 g/hr shown in Figure
2-71.
140
-------
NOx
A
A
s
A
A
±
condition
•
hot
•
cold
•
lab
SCR
•
0
~
1
2006
2008
2010
2012
Figure 2-68. NOx Emission Rates from the TTI and ARB Programs by Engine Model Year, and Average Emission Rates
for 2005-2009 and the 2010-2012 Engine Model Years (lines)
Figure 2-69 displays the THC individual test results. The results are displayed with the SCR aftertreatment,
rather than according to the use of an oxidation catalyst aftertreatment. The use of SCR corresponded
better to THC emissions than the reported use of an oxidation catalyst. We believe the SCR aftertreatment
classification is a surrogate for the combined engine control and aftertreatment system used with SCR
equipped trucks that have a large impact on THC emissions. For example, with the use of SCR, engines can
be calibrated to run leaner, which reduces engine-out THC emissions. Additionally, SCR systems rely on
oxidation catalysts, or catalyzed DPFs to convert NO to N02, which also reduces the THC tailpipe emissions.
We calculated average emission rates for three model year groups 2005-2009, 2010-2012 and 2013 and
later model years. The 2005-2009 model year vehicles include a combination of DPF and non-DPF
equipped trucks" and are used to represent the 2007-2009 emission rates in MOVES. The 2010-2012
represents DPF equipped trucks, with some penetration of SCR equipped trucks. The model year group
representing 2013 and later model years was developed because starting in 2013, Navistar began certifying
u The 2005-2009 THC rates here are ~3 times smaller than the MOVES3 THC rates for MY 1990-2006 derived in
Section 2.3.1.2, which may be due to the small sample size of overlapping model year vehicles (3 MY 2005-2006
trucks) in the TTI study.
141
-------
a heavy heavy-duty diesel (HHDD) engine equipped with SCR aftertreatment. In 2014 and 2015, Navistar
and all other engine manufacturers certified all their HHDD engines equipped with SCR aftertreatment.91
Therefore, emission rate for the 2013+ model year group was estimated by averaging the rates of all the
SCR equipped trucks in the data set, even though the dataset did not include any data on 2013 and later
model year engines.
THC
Figure 2-69. THC Emission Rates from the TTI and ARB Programs by Engine Model Year, and Average Emission Rates
for 2005-2009, 2010-2012, and 2013 + (SCR only) Engine Model Years (lines)
Figure 2-70 displays the PM2.s individual test results. The ARB tests reported zero emission or "Not
Reported due to PM collection failure" for the five ARB tests, and thus, only the TTI data was used to
develop the PM2.5 extended idle emission rates. For the same reasons provided for the THC results, the use
of an SCR-equipped engine and aftertreatment systems should also have a significant impact on the PM2.5
emissions. Additionally, and as expected, the implementation of diesel particulate filters starting in 2007
model year had a significant impact on the PM2 5 emissions.
We grouped the individual emission tests into four model year groups: 2005-2006 (pre-DPF), 2007-2009
(DPF, pre-SCR), 2010-2012 (DPF and phase-in of SCR) and 2013 and later model years (SCR only). Because
the MY 2005-2006 PM2.5 emission rates are significantly different than the MY 2007-2009 emission rates,
they are grouped separately. The 2005-2006 rates from this study are not used to update the pre-2007
142
-------
PM2.5 emission rates/ The other model years and aftertreatment groups are used to estimate the MOVES
emission rates for MY 2007 and later. As for THC, we used the results from the 2010 and later SCR
equipped trucks to calculate PM2 5 emission rate for the 2013 and later model year group.
PM
: i
condition
•
hot
•
cold
•
lab
SCR
•
0
~
1
2006
2008
2010
2012
2014
Figure 2-70. PM2.5 Emission Rates from the TTI Program by Engine Model Year, and average Emission Rates Using
for 2005-2006, 2007-2009, 2010-2012, and 2013 + (SCR only) Engine Model Years (lines)
PM2.5 emission rates in MOVES are composed of elemental carbon (EC) and non-elemental carbon PM
(nonEC). The TTI study measured total PM2.5 emissions, but not EC. We used the EC/PM fractions from the
sources listed in Table 2-48 to estimate the EC and PM2.5 emission rates.
v The MY 2005-2006 PM2.5 emission rates measured from the TTI data are only ~3 times higher than the MY 2007-
2009 PM2.5 rates, and roughly ~10 times smaller than the PM2 5 rates for MY 1990-2006 (2.5 g/hr). We would expect a
larger decrease in PM25 emission rates with the use of DPF as discussed in Section 2.3.3. Differences could be due to
PM sampling methods, or variation in the truck emissions given the small sample size of 2005-2006 model year trucks
in the TTI study. As mentioned above, we did not update the pre-2007 PM25 emission rates in MOVES with
informatin from the TTI dataset due to limitations on time and resources.
143
-------
Table 2-48. Baseline elemental carbon to PM2.5 fraction assumed for extended idling
Model Year
Group
EC/PM
Source
Pre-2007
0.26
MOVES2014 Extended
Idling11™
2007-2009
0.10
ACES Phase I90
2010+
0.16
ACES Phase II92
2.3.2.3 Tampering and Mal-maintenance
As discussed in Section 2.3.1.2, we did not incorporate tamperaing and mal-maintenance effects on the
pre-2007 extended idle rates. For the 2007 and later extended idle rates, we incorporatd the effects of the
effect of tampering and mal-maintenance (T&M) for two reasons:
1. The twenty vehicles used to estimate the extended idle emission rates did not appear to include
any tampered or mal-maintened vehicles with elevated emission rates. In addition, 14 of the 20
vehicles had odometer readings with less than 100,000 miles (Table 2-46).
2. The 2007 and later technology includes aftertreatment technology, including diesel oxidation
catalysts (DOC) and diesel particulate filters (DPF). We anticipate that the failure of these after-
treatment systems would significantly increase extended idle emissions if they were tampered or
mal-maintaned.
We incorporated the T&M effects for extended idle exhaust using different data and methodology than
was used to derive the tailpipe exhaust emission rates for two reasons:
1. Extended idle emissions in MOVES are stored in the EmissionRate table, and are not distinguished
by vehicle age, as the running and start exhaust emission factors. To fit the current MOVES
structure, we incorporated the effects of T&M into a single emission rate by model year that
applies to all vehicle ages.
w The pre-2007 EC/PM ratio for extended idling has subsequently been updated in MOVES3 to be 46.4% as discussed
in Section 2.3.1, but it was not updated for this analysis.
144
-------
2. We are less confident in the application of the the emission effects of T&M failures estimated for
running emissions in Appendix B to extended idling emission. For example, we do not think failure
of selective catalytic reduction (SCR) aftertreatments systems should impact extended idling NOx
emission rates as much as running exhaust emissions, because the SCR systems is not fully
operational during long idling periods. Instead, we estimated the effects of T&M on 2007 and later
extended idle emissions using pre-2007 extended idle emissions as surrogate values for 2007 and
later extended idle emission with failed aftertreatment systems.
As shown in the figures above, the THC and PM2.s emissions showed the largest reductions in extended idle
emissions with newer model year vehicles. We believe that the reductions are due primarily to the
continued effectiveness of the catalyzed diesel particulate filter even during extended idling conditions.
For the MOVES extended idle THC and PM2.s emission rates, we included an estimate of the impact of
deterioration and failure of the diesel particulate filters in calculating the 2007-2009, 2010-2012, and
2013+ model year group emission rates as discussed in Appendix C, and displayed in Table 2-49. As shown,
the MOVES EC/PM emission rates for MY 2007+ trucks are slightly higher than the 'Baseline' EC/PM
fractions in Table 7-2, because the fleet emissions are assumed to include some emission contribution
from trucks with failed DPFs, which have a higher EC/PM fraction.
Table 2-49. Extended Idle Emission Rates for 2007 and Later Model Year Heavy-Duty Vehicles
Model Year
Group
C02
(g/hr)
CO
(g/hr)
NOx
(g/hr)
THC
(g/hr)
PM2.5
(g/hr)
EC
(g/hr)
nonEC
(g/hr)
EC/PM
2007-2009
7151
39.3
100.5
8.5
0.087
0.012
0.076
0.13
2010-2012
7151
39.3
42.6
2.7
0.034
0.006
0.028
0.18
2013+
7151
39.3
42.6
1.6
0.021
0.004
0.017
0.20
2.3.2.4 MHD Regulatory Class
The extended idle emission rates for MHD are assumed to be the same as HHD for the following two
reasons. First, MHD trucks are estimated to account for only 5 percent of long-haul combination trucks in
the US and therefore, they are a minor contributor to the emissions from extended idling trucks. Second,
Khan et al. 200993 evaluated extended idle emission rates of pre-2007 MHD engines and did not observe a
pronounced difference in extended idle emission rates between MHD and HHD trucks. Without any
extended idling data on 2007 and later model year MHD trucks, we felt it was most defensible to keep the
MHD emission rates the same as the HHD emission rates.
145
-------
2.3.3 Model Year Trends
Extended idle emission rates were updated for all model years, as described in Sections 2.3.1 and 2.3.2.
Figure 2-71 through Figure 2-74 illustrate the extended idle emission rates in MOVES3 for regClasslDs 46
and 47.
As shown, the NOx and the CO extended idle emission rates have a relatively small decrease between the
pre-2007 and the 2007+ model years. For THC and PM2.5, we observe large decreases starting in MY 2007,
which is consistent with our understanding of the effect of diesel particulate filters. We observed a
decrease by ~29 times in extended idle PM2.5 rates between the pre-2007 and post-2007 extended idle
rates corresponding to the implementation of the DPFs, which is consistent with the ~27 decrease in PM2.5
running exhaust emission rates from PM2.5 certification data as discussed in Section 2.1.2.1.7.
Extended Idle NOx Emission Rate
150
P 125
O)
- 100
ro
* 75
c
o
v) 50
w
L§ 25
0
1960 1980 2000 2020 2040
Model Year
Figure 2-71. Extended Idle NOx Emission Rates for HHD and MHD Diesel Vehicles by Model Year
L
146
-------
Extended IdleTHC Emission Rate
150
£ 125
O)
V100
ro
& 75
c
I 50
w 25
1
L,
1960
1980
2000
Model Year
2020
2040
Figure 2-72. Extended Idle THC Emission Rates for HHD and MHD Diesel Vehicles by Model Year
150
£ 125
O)
2 100
"5
* 75
c
o
c/) 50
en
i§ 25
Extended Idle CO Emission Rate
1960
1980
2000
Model Year
2020
2040
Figure 2-73 Extended Idle CO Emission Rates for HHD and MHD Diesel Vehicles by Model Year
147
-------
Extended Idle PM Emission Rates
5
4
L_
.c
3 3
B
ro
Qt
d
o
'(J)
(/>
"E
HI
2
1
Elemental Carbon
Non-Elemental Carbon
PM
r
1960
1980
2000
Model Year
2020
2040
Figure 2-74. Extended Idle PM2.s Emission Rates HHD and MHD Diesel Vehicles by Model Year
2.3.4 Extended Idle Energy Rates
The pre-2007 extended idle energy emission rates are unchanged from those originally developed for
MOVES2004 and are documented in the Energy and Emissions Report71, and are displayed in Figure 2-75.
The extended idle energy consumption rates are the same for regulatory class MHD and HHD diesel
vehicles. The extended idle energy rates for 2007+ trucks were updated in MOVES3 and estimated using
the C02 emission rates presented in Table 2-49 and are also plotted in Figure 2-75. The extended idle
energy consumption rates are the same for regulatory class MHD and HHD diesel vehicles.
148
-------
1.56+05
o
1.00+05
cl>
4—'
ra
i_
c
o
'(/)
.
O)
CD
5.0e+04
0.0©+00
regClassNarne
HHD8
MHD67
1970 1980 1990 2000 2010
modelYearlD
2020
2030
Figure 2-75. Extended Idle Energy Emission Rates for HHD and MHD Diesel Trucks
2.4 Auxiliary Power Unit Exhaust
Auxiliary power unit (APU) exhaust is a separate emission process in MOVES. APU usage only applies to the
vehicles with hotelling activity, which are the heavy-duty regulatory classes (MHD, HHD, and Gliders)
within the combination long-haul truck source type (sourceTypelD 62). The APU emission rate for MHD,
HHD and glider regulatory classes are the same for each model year. The projected use of APUs during
hotelling due to the HD GHG Phase 2 program, shown below in Table 2-50, were used to revise the
"hotellingactivitydistribution" table in MOVES362, as is also discussed in the Population and Activity
Report.9
149
-------
Table 2-50: Projected APU Use during Extended Idling for Combination Long-Haul Tractor-Trailers
Vehicle Type
Model years
Diesel APU
Penetration
Battery APU
Penetration
Combination
Long-Haul
Trucks
2010-2020
9%
0%
2021-2023
30%
10%
2024-2026
40%
10%
2027+
40%
15%
For MOVES3, the APU emission rates have been updated to reflect new standards, data, and analysis. The
APU emission rates were updated in MOVES3 based on two studies that measured in-use APU emission
rates. The Texas Transportation Institute (TTI, 2014)94 tested two diesel APU systems with and without
diesel particulate filters at ambient temperatures of lOCf F and Cf F. The exhaust emission rates (THC, CO,
C02, and NOx) and the exhaust flow rates were measured using an ECOSTAR gaseous portable emission
measurement system. The PM mass was measured using a BG-3 partial flow dilution and filter sampling
system. Limitations of the TTI study are discussed in the HD GHG Phase 2 MOVES documentation.100 "
The second study used to update APU emission rates was by Frey and Kuo (2009),95 who tested two APU
systems (APU ID 2 and 3), equipped with 2006 Kubota Z482 engines. The APU systems were tested at a
range of electric output loads to obtain the fuel consumption relationship with the electric power
demands, and the fuel-based emission rates. The study measured the in-use APU electric loads from a fleet
of 20 vehicles (10 trucks equipped with each APU system) for over a year. They then used the relationship
between electric power demand and the fuel-based emission factors with the average energy use of the
APU system to estimate average APU (g/hr) emission rates of C02, CO, NOx, THC, and PM for both a mild
temperature (50-68' F) scenario and a high temperature (lOOf F+) scenario. Frey and Kuo 2009 reported a
PM emission rate, but the emission rate is 'inferred from the literature' because their PM measurements
were semi-qualitative.
An additional two studies were used as a source of data to compare and evaluate the APU emission rates
obtained from the studies mentioned above. TTI 20 1 296 conducted testing of two APU systems using their
environmental chamber at both lOOf F and Of F. The APU systems (APU 4 and 5) manufacturer, engine
make and model year were maintained confidential in the report. Storey et al. 200397 tested a Pony Pack
APU System (APU ID 6), equipped with a Kubota Z482 engine, in an environmental chamber at both 90f F
x Problems in testing meant only one of the APU systems could be used. Additionally, PM composition (EC/PM
fraction) was measured on tests with errors in the exhaust flow measurement. The PM emission rates determined
invalid for these tests were excluded and repeated, but the PM composition measurements from these tests were
considered valid and were not repeated.
150
-------
and Cf F. This is one of the studies used by Frey and Kuo 200995 to determine the PM emission factor for
the APU's tested in their study. The engine year, engine displacement, and engine power were not
reported in the TTI 2012 and Storey et al. 2003 studies. For this reason, these studies were used only as
comparative data sets.
Table 2-51. APU Engines and Studies Used in This Analysis
APU ID
Engine Model
Engine
Year
Displacement
(L)
Power
(HP/kW)
Tier
Study
1
Kubota Z482
2011
0.48
14.2/11
Tier 4
TTI 201494
2
Kubota Z482
2006
0.48
10.9/8.1
Tier 2
Frey and Kuo
200995
3
Kubota Z482
2006
0.48
10.9/8.1
Tier 2
Frey and Kuo 2009
4
Confidential Information
TTI 201296
5
Confidential Information
TTI2012
6
Kubota Z482
Storey et al. 200397
Table 2-52 contains the in-use emission rates measured from reviewed APU systems. As shown, the
emission and fuel rates for the APUs measured in the TTI 2014, and Frey and Kuo 2009 (APU ID 1, 2 and 3)
compare well with the APU emission rates reported from TTI 2009 Storey et al. 2003 (APU ID 4, 5, and 6).
The impact of the DPF is clearly shown on the PM emission rates from APU ID 1, as expected. However,
there does not appear to be a substantial impact of the DPF on the gaseous emissions (C02, CO, NOx, and
THC). Additionally, no notable emission effects are observed with respect to the nonroad emission
standard tier or engine model year.
The impact of ambient temperature can be observed within individual studies. For APU ID 2 and 3, the C02,
and fuel consumption are higher at the hot ambient temperatures compared to the mild conditions, which
is expected. However, there is no consistent trend between hot and cold conditions, when the APU is
required to either cool or heat the tractor cabin. For APU ID 1 and 4, the cold temperatures had higher C02
emissions and fuel use. For APU ID 5 and 6, the hotter temperatures had higher C02 emissions and fuel
use.
For criteria emissions (CO, NOx, THC, and PM), there are conflicting trends with respect to ambient
temperature. For APU 2 and 3, NOx and PM emissions are higher at the hot conditions compared to mild
conditions, consistent with the higher fuel use. However, CO shows lower emissions at hot conditions, and
151
-------
THC shows a mixed trend. For the other studies, there is no consistent trend in the criteria emissions
between the hot and cold conditions.
Table 2-52. In-Use APU Emission Rates
APU
o
u
CO
NOx
THC
PM
Fuel
Ambient
Temperature
DPF
ID
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(gal/hr)
condition
CF)
present
1
4340
7.3
18.6
1.35
0.96
0.43
Cold
0
No
1
4270
5.1
20.0
0.73
0.02
0.43
Cold
0
Yes
1
2820
6.2
23.5
1.35
0.56
0.29
Hot
100
No
1
2800
5.2
23.7
1.52
0.03
0.28
Hot
100
Yes
2
3000
20.4
6.3
1.4
1
0.3
Mild
60a
No
3
2500
7.2
13.4
1.3
0.8
0.25
Mild
60
No
2
3900
13.9
11.5
1.5
1.3
0.38
Hot
100
No
3
3600
6.3
20.2
1
1.2
0.36
Hot
100
No
4
3100
5.8
19
1.3
1.23
0.3
Hot
100
No
5
3600
7.3
24
0.8
0.58
0.35
Hot
100
No
4
4000
3.9
22
1.2
0.75
0.39
Cold
0
No
5
2800
24
14
2.4
0.98
0.28
Cold
0
No
6
2146
25
8.7
7.8
0.48
0.22
Cold
0
No
6
2351
10.8
11.4
4.2
1.00
0.24
Hot
90
No
Note:
a Frey and Kuo 2009 report the mild condition for auxiliary loads on the trucks is for ambient temperatures ranging
from 10-20"C (50-68 °F)
Because the only notable trend in the APU emissions data was the large decrease in PM emission rates
with the use of a DPF, we developed "no DPF" baseline MOVES emission rates using the "no DPF" results
from TTI, 2014 and Frey and Kuo, 2009 (APU ID 1, 2, and 3). We first averaged the emission rates within the
cold, hot, and mild conditions as shown in Table 2-53.
152
-------
Table 2-53. Average APU Emission Rates from non-DPF APU IDs 1, 2, and 3 according to Cold, Hot, and Mild
Ambient Conditions
co2
(g/hr)
CO
(g/hr)
NOx
(g/hr)
THC
(g/hr)
PM
(g/hr)
Fuel
(gal/hr)
Ambient
condition
Temperature
CF)
DPF
present
4340
7.27
18.59
1.35
0.96
0.43
Cold
0
No
3440
8.80
18.41
1.28
1.02
0.34
Hot
100
No
2750
13.80
9.85
1.35
0.90
0.28
Mild
60
No
Next, we calculated a fleet-average APU emission rate. Similar to our treatment of the extended idle
emission rates, we equally weighted the different ambient conditions. For APUs, we weighted each
ambient condition (Cold, Hot, and Mild) equally in developing the fleet-average emission rate shown in
Table 2-55.
We estimated elemental carbon (EC) fraction of PM from composition measurements made on APU ID 1 as
reported in Appendix K. For each test, we calculated the elemental carbon/total carbon ratio, and then
averaged the ratio across all cold and hot tests, separately for the DPF and the non-DPF tests as shown in
Table 2-54. We assumed that total carbon (TC) is a reasonable approximation of the total PM2.5 emissions
from the APU, and we used the EC/TC ratio from the non-DPF tests as the source of the EC/PM fraction to
derive the EC and nonEC emission rates in Table 2-55.
Table 2-54. Average Elemental Carbon/Total Carbon Ratio for APU ID 1 without and with a Diesel Particulate Filter
(DPF)
EC/TC ratio
APU 1 non-DPF
0.138
APU 1 DPF
0.073
Table 2-55. Fleet-Average Non-DPF Equipped APU Emission Rates in MOVES3
C02
(g/hr)
CO
(g/hr)
NOx
(g/hr)
THC
(g/hr)
PM2.5
(g/hr)
EC
(g/hr)
NonEC
(g/hr)
EC/PM
Fuel
(gal/hr)
3510
10.0
15.6
1.3
0.96
0.13
0.83
0.14
0.35
The HD GHG Phase 2 rule implements a phase-in standard that requires APUs installed in new tractors to
meet lower PM standards from MY 2021 through MY 2024 (beyond the Tier 4 nonroad standards).98 The
153
-------
APU PM standards along with the current Tier 2 and Tier 4 nonroad standards for nonroad diesel engines
8
-------
Table 2-57 APU Emission Rates in MOVES3 with APU PM Controls in the HD GHG Phase 2 Program
Model Year
C02
CO
NOx
THC
PM2.s
EC
NonEC
EC/PM
Fuel
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(g/hr)
(gal/hr)
2010-20201
3510
10.0
15.6
1.3
0.96
0.13
0.83
0.14
0.35
2021-2023
3510
10.0
19.5
1.3
0.32
0.044
0.28
0.14
0.35
2024-2050
3510
10.0
15.6
1.3
0.021
0.0015
0.019
0.073
0.35
Note:
1 The default APU allocation in MOVES assigns APU usage beginning in model year 2010. If MOVES users specify
APU usage in years previous to 2010, it will use the 2010-2020 APU emission rate.
155
-------
2.5
Glider Vehicle Emissions
"Glider vehicles" or "Gliders" refer to vehicles with old powertrain (engine, transmission and/or rear axle)
combined with a new chassis and cab assembly. Most gliders are Class 8 heavy heavy-duty vehicles. They
typically use model year 2001 or older remanufactured engines that do not have to use emissions controls
such as DPF or SCR needed to meet the stringent PM and NOx standards starting MY 2007+. 101
Starting with MOVES3, we model the emission impacts of the glider vehicles as a separate regulatory class
(regClassID 49) because their population became significant starting with model year 2008 as described in
the Population and Activity Report.9
For modeling purposes, all glider vehicles are presumed to be combination trucks (sourceTypelD 61 and
62) running on diesel fuel. As detailed in 0, EPA's in-house glider vehicle emission testing data102 suggest
that glider emissions have similar THC, NOx, PM2.5, and C02 running exhaust emission rates to the MOVES
model year 2000 heavy heavy-duty vehicles (regClassID 47). CO from glider vehicles is higher. Based on
this analysis, the MOVES running, start, and extended idling exhaust rates for gliders of all model yearsv are
set equal to those of the model year 2000 heavy heavy-duty vehicles.
For example, Figure 2-76 shows a comparison of the running exhaust emission rates (for age 0-3 group) of
regClass 47 (heavy heavy-duty) vs. regClass 49 (glider vehicles) for selected pollutants and model year
groups. The rates for the two regulatory classes are identical for model year 2000. For later model years,
however, the emissions rates for regular heavy heavy-duty vehicles are significantly lower due to more
stringent emission standards, whereas the rates for glider vehicles stay the same at the model year 2000
levels.
The auxiliary power unit (APU) exhaust emission rates of the glider vehicles, on the other hand, are set
equal to those of regular (non-glider) heavy heavy-duty vehicle fleet. This is consistent with our
assumption that glider vehicles have the same vehicle characteristics as regular heavy heavy-duty vehicles
for non-powertrain components, and thus, have the same APU, aerodynamics, rolling resistance, brake and
tire wear.
y Glider emission rates are the same for all model years from 1960 through 2060. As discussed in the activity report,
glider activity in MOVES begins in model year 2008.
156
-------
1
3000-
2000-
1000-
't—'
€
CT
£ 100-t
CD
q:
S> 50-)
CO
CD
t=
CD
0)
E 30-
04
20-
10-
0-
MY2000
i II II II II .. II II II
¦¦ .. .. II II II II .. II II II
Mil
regClassID ¦ 47 ¦ 49
MY2008
. .i.. >1 if ll ll ll .. J il ll I
. ... I I I I . I I I
-i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r
D « 2 ; « is S S S 15 S KS Sfi S S §
.... ..i .1
j.1.1.
.nil
. I I
MY2010+
. ¦.* .1 .1 il il ll __ .1 .1 .1 .1 il I
. . I I I I . I I I
.i .1 .1 .1 il I
l—i—I—r—i—i—i—i—r—i—i—i—i—i—I—i—i—t
.1.1 J.I . I I I
I I I
I I I
i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r
- n n w n n S «s or-^rr«-»-NNNNP(WWNflii>nii)S(ii«
opModelD
4 Figure 2-76, Comparison of the running exhaust emission rates (0-3 age group) of HHD (regClassID 47) vs. Gliders (regClassID 49) for selected pollutants
5 (NOx, ECPM, NonECPM) and model year groups
157
-------
3 Heavy-Duty Gasoline Exhaust Emissions
The discussion of heavy-duty gasoline vehicles first covers running exhaust emissions (Section 3.1),
followed by start emissions (Section 3.2). Within each emission process, we discuss the derivation of the
emission rates by pollutant and model year group. As gasoline-fueled vehicles are a small percentage of
the heavy-duty vehicle fleet, the amount of data available on their emissions is more limited.
3.1 Running Exhaust Emissions
3.1.1 THC, CO and NOx
The heavy-duty gasoline running rates were analyzed in three stages. The MY 1960-2007 emission rates
were originally developed in MOVES2010. In MOVES2014, we updated the MY 2008-2009 heavy-duty
gasoline rates to account for the Tier 2 and 2007 heavy-duty rulemakings. In MOVES3, we updated the MY
2010 and later emission rates based on the latest testing data. The analysis of PM2.5 running exhaust
emission rates are discussed separately in Section 3.1.2 because it used separate data and analyses than
for the gaseous pollutants.
3.1.1.1 1960-2007 Model Years
The heavy-duty gasoline emission rates for model year 2007 and earlier were carried over from previous
versions of MOVES. They are based on analysis of four medium heavy-duty gasoline trucks from the CRC E-
55 program and historical data from EPA's Mobile Source Observation Database (MSOD)103, which has
results from chassis tests performed by EPA, contractors and outside parties. The heavy-duty gasoline data
in the MSOD is mostly from pickup trucks which fall mainly in the LHD2b3 regulatory class. Table 3-1 shows
the total number of vehicles in these data sets. In the real world, most heavy-duty gasoline vehicles fall in
either the LHD2b3 or LHD45 class, with a smaller percentage in the MHD class. There are very few HHD
gasoline trucks now in use.
Table 3-1 Distribution of Vehicles in the Data Sets by Model Year Group, Regulatory Class and Age Group
Model year
Regulatory
Age group
group
class
0-5
6-9
1960-1989
MHD
2
LHD2b3
10
1990-1997
MHD
1
LHD2b3
33
19
1998-2002
MHD
1
LHD2b3
1
158
-------
Similar to the HD diesel PM, THC, and CO analysis described above, the chassis vehicle speed and
acceleration, coupled with the average weight for each regulatory class, were used to calculate STP
(Equation 1-6). To supplement the available data, we examined engine certification data as a guide to
developing model year groups for analysis. Figure 3-1 shows averages of certification results by model
year.
30 T
25
20 -
¦Q
"5i
O
o
tr 15 —
10 -
5 4""1 1 £ L
z
I
i CO
l Nox
~ HC
5
I
"i{if
^ " -- -- -- ~
1980
1985
1990
1995
Model year
2000
—it—- -—4r—
2005
1.4
b 1.2
1
0.8
h 0.6
h 0.4 x
- 0.2
- 0
-0.2
2010
Figure 3-1 Brake-Specific Certification Emission Rates by Model Year for Heavy-Duty Gasoline Engines
Based on these certification results, we decided to classify the data into the coarse model year groups
listed below.
• 1960-1989
• 1990-1997
• 1998-2007
Unlike the analysis for HD diesel vehicles, we used the age effects present in the data itself. We did not
incorporate external tampering and mal-maintenance assumptions into the HD gasoline rates. Due to the
sparseness of data, we used only the two age groups listed in Table 3-1, and applied the same age effects
to all the heavy duty regulatory classes.
159
-------
3.1.1.1.1 LHD
The emission rates for LHD (LHD2b3 and LHD45, regClassID 41 and 42, respectively) were analyzed by
binning the emission measurements using the STP with a fixed mass factor of 2.06 (Table 1-3). Figure 3-2
shows all three pollutants vs. operating mode. In general, emissions follow the expected trend with
increasing STP, though the trend is most pronounced for NOx. As expected, NOx emissions for light heavy-
duty gasoline vehicles are much lower than for light heavy-duty diesel vehicles.
4000-
3000
2000
1000
0
150
~
1 t
itiit
3
O
O)
« 100
2
c
.2 50
V)
E
m 0
500
400
300
200
100
0
~
• A
«
~ A * * A * A
! • • • s J ' A i i i * 1 ) 1 1 I ~ ~ *
HC
* ¦#
~
~
•
A
A
A
~
~
A
•
~ 4 .
X
I # '
; i - ' i |1 * ¦ "
iii*1 * A
«¦1
~ *
I I I I
0 1 11 1213141516212223242527282930333537383940
opModeD
Model.Years
~ 1960-1989
* 1990-1997
¦ 1998-2007
Figure 3-2. Emission Rates by Operating Mode for MY Groups 1960-1989,1990-1997, and 1998-2007 at Age 0-3
Years for LHD2b3 and LHD45 Vehicles
Table 3-2 displays the multiplicative age effects by operating mode for LHD gasoline vehicles. The relative
age effects are derived from the sample of vehicle tests summarized in Table 3-1. The multiplicative age
160
-------
effects are used to estimate the aged emission rates (ages 6+) years from the base emission rates (ages 0-
5) for THC, CO, and NOx. These multiplicative age effects apply to all model year groups between 1960 and
2007. As discussed earlier, we derived multiplicative age effects from the pooled data across the three
model year groups and regulatory classes due to the limited data set.
161
-------
Table 3-2 Relative Age Effect on Emission Rates between Age 6+ and Age 0-5 for LHD Gasoline Vehicles in Model
Years 1960-2007
OpModelD
THC
CO
NOx
0
2.85
1.45
1.67
1
2.43
1.79
1.85
11
3.12
1.66
1.88
12
2.85
2.05
1.69
13
3.55
2.68
1.48
14
3.43
2.84
1.46
15
3.37
3.03
1.26
16
3.76
3.88
1.06
21
2.78
1.67
1.42
22
2.64
1.64
1.36
23
2.96
1.67
1.32
24
2.83
1.62
1.21
25
3.23
2.79
1.43
27
3.21
3.20
1.21
28
3.20
4.04
1.11
29
3.00
3.90
1.05
30
2.55
2.56
1.05
33
1.95
2.00
1.77
35
2.67
2.20
1.59
37
2.80
2.24
1.42
38
2.46
2.06
1.34
39
2.46
2.30
1.27
40
2.47
2.59
1.17
Figure 3-2 illustrates the emissions trends by age group for the 1998-2007 model year group. Since we did
not use the tampering and mal-maintenance methodology as we did for diesels, the age trends reflect our
162
-------
coarse binning with age. For each pollutant, only two distinct rates exist - one for ages 0-5 and another for
age 6 and older.
2500-
2000-1
1500 "
1000-
500-
o-
A A
~. *
a ~ *¦ A ~ A a
A A * *
~ ~ ~ • ~
13
O
JZ
75)
-------
emission rates is that the regulatory class MHD and HHD emission rates were analyzed using STP operating
modes with a fixed mass factor of 17.1. The resulting MHD and HHD emission rates for THC, CO, and NOx
for each model year group are presented in Figure 3-4.
3
o
.c
-------
Table 3-3 Relative Age Effect on Emission Rates between Age 6+ and Age 0-5 for MHD and HHD Gasoline Vehicles in
All Model Years 1960-2060
OpModelD
THC
CO
NOx
0
2.85
1.45
1.67
1
2.43
1.79
1.85
11
3.12
1.66
1.88
12
3.36
3.12
1.13
13
3.53
3.16
1.11
14
3.53
3.16
1.11
15
3.53
3.16
1.11
16
3.53
3.16
1.11
21
2.78
1.67
1.42
22
3.08
2.59
1.23
23
2.97
3.31
1.05
24
1.80
1.54
1.03
25
1.80
1.54
1.03
27
1.80
1.54
1.03
28
1.80
1.54
1.03
29
1.80
1.54
1.03
30
1.80
1.54
1.03
33
2.45
2.41
1.33
35
2.16
2.41
1.19
37
2.16
2.41
1.19
38
2.16
2.41
1.19
39
2.16
2.41
1.19
40
2.16
2.41
1.19
Figure 3-5 displays the resulting emission rates by operating mode bin and age group for the LHD45, MHD,
and HHD gasoline vehicles, which were calculated by applying the multiplicative age effects in Table 3-3.
165
-------
7500
5000
2500
0
o
-C
15)
a>
re
c
o
if)
LU
A
A
+
o
o
A
A
* «
A
A
A
A
A
* . * ± ~ ~ * * *
1
~
t t
A
A
A
A
I
o
A
•
*
A *
A *
A
i * *
A a. * •
A * ~ *
A i * *
*
r
A
A ~
A
*
4
A
*
L
NOx
A
~
*
4-
A
*
k A * *
*
A *
•
c
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
age
* 6+
• 0-5
opModelD
Figure 3-5. Emission Rates by Operating Mode and Age Group for MY 1998-2007 Vehicles in Regulatory Class MHD
and HHD Gasoline Vehicles
3.1.1.2 2008-2009 Model Years
3.1.1.2.1 LHD
The MY 2008 and 2009 LHD emission rates are updated from the MY 2007 LHD emission rates to account
for the phase-in of the Tier 2 and HD 2007 rulemaking which set emissions standards for medium-duty
passenger vehicles (MDPV), Class 2b, and Class 3 chassis-certified vehicles. Medium duty passenger
vehicles fall within the LHD2b3 regulatory class in MOVES. The useful life emission standards for these
vehicles are shown in Table 3-4.
166
-------
Table 3-4 Useful Life FTP Standards from the Tier 2 Rulemaking104 and the HD 2007 Rule106
MDPV
8.5k- 10K
10K-14K
(Tier 2 Bin 5)
(Class 2B)
(Class 3)
Units
g/mile
g/mile
g/mile
Fully Phased in MY
2009
2009
2009
THC
0.09 NMOG
0.195 NMHC
0.230 NMHC
CO
4.2
7.3
8.1
NOx
0.07
0.2
0.4
This section documents the THC, CO and NOx emission rates for regulatory class LHD2b3 vehicles in model
years 2008 and 2009. In conducting this analysis, we lacked any modal data on LHD vehicles and therefore,
we ratioed the modal emission rates measured from light-duty vehicles by the difference in standards.8 By
MY 2008, the certification results demonstrated that LHD2b3 were nearing the emission levels of light-duty
vehicles certified to the Tier 2 Bin 8 standard11. Consequently, we relied on the MOVES2014 analysis of in-
use Tier 2 Bin 8 vehicles conducted for the light-duty emission rates.8 We applied this analysis to derive
MY 2009 emission rates, then calculated MY 2008 rates by interpolating between MY 2007 and MY 2009.
Although the light-duty rates are based on VSP, rather than STP, adapting them for the LHD2b3 rates was
deemed an acceptable approximation because the gasoline LHD2b3 gasoline vehicles are chassis-certified
to distance-based standards (g/mi). Accordingly, the vehicle emissions rates are less dependant on the
individual power and weight of the vehicle, and should scale approximately to the g/mile emission
standards.2
Based on these assumptions, we scaled modal rates for Tier 2 Bin 8 vehicles by the ratio of FTP standards
to the calculated aggregate LHD2b3 standards documented in the MOVES2014 heavy-duty exhaust
report.11,33 Table 3-5 displays the aggregated LHD2b3 standards, Bin 8 FTP standard and the ratio between
the standards by pollutant.
z This approximation needs to be revisited in the future now that we have updated the mass of LHD vehicles in
MOVES3 to range from 3.5 to 7.8 metric tons9, which differs from the/sca/e value of 2.06 metric tons.
33 This analysis assumed that 5% of the gasoline LHD2b3 engines were engine-certified, but, actually, all engine-
certified LHD2b3 vehicles are diesel-fueled. However, the engine-certification standard has a small impact on the
calculated aggregated standard conducted for MOVES2014.
167
-------
Table 3-5 Aggregate LHD2b3 Standard Ratios against Bin 8 Modal Rates
Aggregate LHD2b3
FTP standard
Bin 8 FTP
standard
Aggregate
LHD2b3/Bin 8
NMOG
0.18
0.1
1.8
CO
7.49
3.4
2.2
NOx
0.22
0.14
1.6
We took an additional step to "split" these ratios into "running" and "start" components, such that the
running rates increased twice as much as the start rates, while maintaining the same simulated value for
the FTP composite. This split ratio is consistent with typical emission reduction trends, where running
emissions are reduced about twice as much as start emissions.8 The "split" ratios for running and start,
which were applied to the light-duty Tier 2 Bin 8 vehicle emission rates are shown in Table 3-6.
Table 3-6 Ratio Applied to Light-Duty Tier 2 Bin 8 Emission Rates to Estimate Regulatory Class LHD2b3 Emission
Rates for 2008-2009 MY
THC
CO
NOx
Running
2.73
2.73
1.95
Start
1.37
1.37
1.00
We also adopted the light-duty deterioration effects and applied them to the MY 2009 regulatory class
LHD2b3 (regClassID 41) emission rates. The light-duty emission rates have age effects that change with
each of the 6 age groups in MOVES, as shown in Table 3-7.
168
-------
Table 3-7 Multiplicative Age Effect used for Running Emissions for Regulatory Class LHD2b32009 Model Year
ageGroupID
THC
CO
NOx
3
1
1
1
405
1.95
2.31
1.73
607
2.80
3.08
2.21
809
3.71
3.62
2.76
1014
4.94
4.63
3.20
1519
5.97
5.62
3.63
2099
7.20
6.81
4.11
After applying the steps described above (scaling the emission factors by ratio of FTP standards, and
applying light-duty deterioration trends), we restricted the scaled data so that the individual emission rates
by operating mode were never higher than MY 1998-2007 regulatory class LHD2b3 rates. This step
essentially "capped" the emission rates, such that none of the modal rates for MY 2009 are higher than
their counterparts for MY 2007 and earlier. MY 2008 rates are interpolated between MY 2007 and MY
2009 emission rates as discussed later.
This final step "capped" the model year 2009 emission rates in the highest operating modes, as shown in
Figure 3-6.. For THC, emission rates in operating modes 28-30 and 38-40 were capped for some or all age
groups by the pre-2007 emission rates. For CO, emission rates in 12 of the 23 running operating modes (1,
16, 23-24, 27-30, 35-40) were capped by the pre-2007 rates. None of the NOx emission rates were
impacted by this step. Figure 3-6. shows the regulatory class LHD2b3 model year 2008-2009 emission rates
for CO, THC, and NOx. In the figure, rates "capped" by the pre-2007 rates exhibit the uncharacteristic
"stairstep" deterioration trends. Even with the "capping" effects, the rates for regulatory class LHD2b3
(regClassID 41) are higher than those for light-duty trucks (regClassID 30), with a few exceptions. The few
exceptions are some of the age-dependent THC and or CO emission rates in operating modes 1, 30, 38, 39,
and 40. However, the majority of emission rates are considerably higher for the heavy-duty (LHD2b3) than
for the light-duty trucks. Similarly, when the FTP is simulated from the resulting rates, estimated
composites are substantially higher for LHD2b3 than for light-duty trucks.
The Light-duty Tier 2 standards shown in Table 3-4 phase-in at a rate of 50 percent in MY 2008 and are
considered fully phased in by MY 2009.105 For estimating emission rates in MOVES, we used the same
assumptions to estimate the MY 2008 emission rates. The MY 2008 running emission rates are
interpolated between the MOVES 2007 and 2009 emission rates by operating mode and age group.
169
-------
Figure 3-6. Age Effects for CO, THC, and NOx Emission Rates for Regulatory Class LHD2b3 (regClassID 41) Vehicles in
Running Operating Modes for MY 2007, 2008 and 2009
2500-
2000-
1500-
1000-
500-
V—
0
3
100
O
JZ
O)
75
-------
there is relatively little data on these vehicles, and therefore, the current 2008 and 2009 model year
emission rates are from MOVES2010.106 The rates are modeled by applying a a 70 percent reduction to the
MY 2007 running rates starting in MY 2008, which is consistent with the emission standard reduction with
the "Heavy-Duty 2007 Rule."107,cc The 2008 and 2009 model year emission rates have two age groups (0-5,
and 6+) and the same relative multiplicative age effects as the pre-2007 emission rates (Table 3-3).
3.1.1.3 2010-2060 Model Years
In MOVES3, we updated the THC, CO, and NOx emissions rates for MY 2010 and later vehicles for all
gasoline heavy-duty regulatory classes. The updated rates are based on analysis of real-world PEMS-based
emissions measurement data from two engine-certified and one chassis-certified heavy-duty gasoline
vehicles (Table 3-8) with model years between 2015-2017. As explained in the PM2.s section (Section
3.1.2.2), we also conducted chassis-dyanamometer laboratory testing on these vehicles, but used the
PEMS gaseous emissions data because it better represents emissions in the real-world.
The Ford and Isuzu vehicles used the most popular engine configurations for recent model year heavy-duty
gasoline Class 4 vehicles. Each of the HD gasoline vehicles had three-way catalyst (TWC) technology to
control THC, CO, and NOx emissions. However, one key difference compared to light-duty gasoline vehicle
TWC configuration is that the engine-certified HD gasoline vehicles do not use a close-coupled TWC. There
might also be differences in catalyst precious metal loading and in-cylinder combustion control for
maximum TWC efficacy. The reason for these differences is that engine-certified and chassis-certified
gasoline spark-ignited vehicles have to meet different standards.
Table 3-8 Summary of MY 2015-2017 Heavy-Duty Gasoline Vehicles with Real-World PEMS-based Emissions
Measurement Data
Vehicle
Engine
Test Weight (lbs)
Cert-
ification
Make
Model
MY
Odometer
(miles)
GVWR
(lbs)
GCWR
(lbs)
Family
Displ
(L)
Low
High
Isuzu
NPR
2015
48,000
14,500
20,500
FGMXE06.0584
6.0
8,620
12,940
Engine
Ford
E450
2016
31,000
14,500
-
GFMXE06.8BWZ
6.8
9,320
13,080
Engine
RAM
3500
2017
32,000
13,300
19,900
HCRXD06.45W0
6.4
14,557
18,020
Chassis
cc The engine-certified standards for heavy-duty gasoline were reduced by 93% (THC), 80% (NOx), and 61% (CO) with
the MY 2008 standard.
171
-------
The testing was conducted by US EPA over various test cycles in the Ann Arbor, Michigan area. The test
matrix covered a range of vehicle operation that included:
1. Two idling tests of 15- or 30-minutes duration
2. Seven on-road driving routes that cover the full range of power demand by including transient
low- and medium-speed urban driving to steady-state high-speed highway driving
3. Soak times ranging from zero minutes (hot start) to 720 minutes (cold start)
4. Vehicle weight at low or high (Table 3-8)
5. Air conditioning on or off
6. Cabin windows down or up
A total of 202 tests across vehicles and operation modes was available for data analysis. These tests
covered about 412,000 seconds of post-QA operation. We removed the effect of warm and cold starts
from the operation since the running emissions are intended to be just the hot running operation; details
are discussed in Appendix 1.1. After removal of vehicle operation related to start emissions, the final data
set used for just the hot running emissions rates update was about 390,000 seconds.
The following steps were used to calculate the operating mode-based emission rates for each age and
regulatory class of LHD, MHD, and HHD:
1. Assign operating modes as per the method described above for diesel vehicles and calculate the
average rate per operating mode per test per vehicle
2. Calculate the average operating mode-based rate per vehicle (using only vehicle specific tests)
3. Estimate emission rates for operating modes with limited or missing data.
4. Calculate the operating mode-based emission rate as the production weighted average of the
three test vehicles.
5. Adjust emission rates by vehicle age.
In Step 1, the operating modes (Table 1-4) were assigned to the 1-hz data using the STP equation (Equation
1-6) with road-load coefficients for single-unit short-haul truck (sourceType 52) for the 2014-2020 model
year range as defined in the sourceusetypephysics table in MOVES3 database. The coefficients for single-
unit short-haul trucks are the same for all the regulatory classes within this sourcetype (LHD2b3, LHD45,
MHD, and HHD). The road-load coefficient values used are:
rollingTermA = 0.596526 [kW.sec/m]
rotatingTermB = 0 [kW.sec2/m2]
dragTermC = 0.00160302 [kW.sec3/m3]
172
-------
For vehicle mass, we used the actual test weight (Table 3-8). Road-grade was not available, so it was set to
zero. The entire data set was analyzed with the new/scofe values (Table 1-3) of 5 (LHD2b3 and LHD45), 7
(MHD), and 10 (HHD). The selection of these new/sco;e values was based on the diesel HDIUT dataset and is
described in Appendix G.
In Step 2, we averaged according to operating mode for each vehicle. In Appendix 1.2, we compared the
emission rates among the three vehicles by operating mode. Significant differences are observed between
the vehicles, however no consistent differences were noted across operating modes and pollutants
between the two engine-certified vehicles and the chassis-certified vehicle.
In Step 3, we estimated emission rates for high power operating modes with limited or missing data from
regulatory class MHD and HHD, due to the larger fSCaie values used for these operating modes. In these
cases, we aggregated the data across the nearest high-power operating modes with sufficient data, and set
the emission rates to be equivalent across the aggregated bins. Additional details and examples are
discussed in Appendix 1.3.
In Step 4, we calculated a weighted average of the emission rates from the three vehicles using the
production volumes of each of the tested engines. Ideally, the emission rates for each regulatory class
(LHD2b3, LHD45, MHD, and HHD) would be estimated from test data collected from vehicles of that
regulatory class, or estimated separately for the engine-certified (LHD45 and heavier) and chassis-certified
vehicles (LHD2b3). However, due to the small sample size (including only one LHD2b3 vehicle), we used the
same weighting of the three vehicles for all the regulatory classes. The production volumes of the RAM
3500 vehicle are only a minor fraction of the combined production of the Ford and Isuzu engine volumes.
As such, the production weighting is most representative of LHD45 emission rates.dd
Because we use the same production volume weighting for all the regulatory classes, the base_emission
rates for MY 2010+ LHD2b3 and LHD45 are identical. However, the LHD2b3 rates are further modified by
applying the Tier 3 reductions phased-in from MY 2018 to 2022 (Section 3.1.1.3.1). The only difference
between the LHD45, MHD, and HHD emission rates is the/SCate used to estimate the emission rate by
operating mode, and methods used to estimate high-power operating modes conducted in Step 4 (Details
in Appendix 1.3).
In Step 5, we applied the MHD/HHD age effects shown in Table 3-3 to all gasoline heavy-duty regulatory
classes, including LHD2b3 and LHD45. We did not use the LHD2b3/LHD45 specific age effects shown in
dd Sales of Class 2b gasoline trucks are much larger than for Class 3, 4, 5, and 6.
173
-------
Table 3-2. Both of these age effects tables are based on the same data set (Table 3-1) with the difference
being only the/scofe used while assigning the data to operating modes. Applying the LHD or LD (Table 3-7)
age effects to rates developed using HD data and different/sco;e ranges could over- or under-estimate the
increases in emissions from aging. Ideally, LHD2b3 emission rates and age effects would be derived from
chassis-certified heavy-duty gasoline vehicles.
3.1.1.3.1 LHD2b3 2018-2060 Model Years
The LHD2b3 vehicles are subject to the Tier 3 light-duty standards starting in MY 2018. To calculate
emission rates for MY 2018 and later, we applied reductions representing the Tier-3 phase-in for MY 2018-
2022 for LHD2b3 vehicles (as shown in Table 3-9) to the emission rates representing MY 2010-2017
estimated from the above. The reductions for each model year during the phase-in were estimated by
extracting the corresponding rates for MY 2007-2022 from the MOVES2014 database, and calculating the
fractions relative to MY 2017. The basis and rationale for the Tier 3 reductions for gasoline LHD2b3 vehicles
developed for the Tier-3 rulemaking are documented in the MOVES2014 heavy-duty exhaust report.108
The LHD2b3 MY 2018+ rates contain the same heavy-duty gasoline age effects as were applied to the MY
2010-2017 rates (Table 3-3). The resulting emission rates for THC, CO and NOx are shown in Figure 3-7
through Figure 3-9.
Table 3-9 Tier 3 Reductions by Model Year for Gasoline LHD2b3
Model Year
THC
CO
NOx
2018
35%
38%
41%
2019
44%
48%
52%
2020
53%
59%
63%
2021
62%
68%
74%
2022-2060
71%
78%
85%
3.1.1.4 Inspection and Maintenance Program Effects for LHD2b3 Gasoline Vehicles
In MOVES3.1, we updated the meanBaseRatelM values for THC, CO, and NOx running exhaust emission
rates to better reflect our understanding how state and local inspection and maintenance (l/M) programs
reduce gaseous emissions from LHD2b3 gasoline vehicles.
As background, in addition to the meanBaseRates described in Section 1, the MOVES emissionRateByAge
table includes a field, meanBaseRatelM, that is used to estimate emissions under a relevant l/M program.
These calculations are explained in more detail in the MOVES Adjustments report.64
174
-------
In MOVES3.0.4 and earlier versions of MOVES, we set the values of meanBaseRatelM for all HD exhaust
rates to equal the associated meanBaseRate—essentially assuming no benefit from HD l/M programs. For
MOVES3.1, we reconsidered this choice for gasoline LHD2b3 trucks (regClass 41). Given the similarity of
the engine technology and aftertreatment systems between gasoline LHD2b3 and light-duty trucks
(regClass 30), and the similarity in the way these trucks are tested in contemporary l/M programs, we
updated the gasoline LHD2b3 HC, CO and NOx running exhaust values for meanBaseRatelM to reflect the
same proportional reduction (that is meanBaseRatelM/meanBaseRate) that we model for each operating
mode bin and age for the light-duty trucks. For more detail on the meanBaseRate and meanBaseRatelM
values for light-duty trucks, see the MOVES3 LD report.8
3.1.1.5 Model Year Trends
Figure 3-7 through Figure 3-9 display the THC, CO, and NOx non-IM running exhaust emission rates by
model year and regulatory class (HHD and Urban Bus). The emission rates are estimated in grams per mile
(g/mile) using nationally representative operating mode distributions and average speeds. The model year
groups used to estimate the emission rates are evident: 1960-1989, 1990-1997, 1998-2007, 2008-2009,
and 2010-2060. Note that not all the changes in the gram per mile emission rates are due to changes in the
operating mode specific emission rates. For example, the MY 1995-1997 operating mode specific emission
rates are the same as the 1990-1995 emission rates for all regulatory classes. However, there is an
observed spike in the HHD gram per mile THC emission rate, which is attributed to a shift in the
distribution of HHD gasoline activity among different source types in MOVES.
Figure 3-7 shows that the THC emission rates follow decreasing trends with model year that correspond
with tighter emission factors. The drop in emission rates in model year 2008-2009 is attributed to the
different methodology used to develop those rates discussed earlier. Even though the increasing trend
between 2008-2009 and 2010 and later model year groups may not be intuitive from a technical
perspective, we have increased confidence that the MY 2010 and later THC emission rates represent the
real-world emissions since they were developed based on in-use testing of MY 2010 and later vehicles.
175
-------
O)
CO
cr
(/)
CO
LU
O
x 0.5
(/>
03
CD
~co
-i—"
o
Hi
Lj
I
o.o-
1970 1980 1990 2000 2010
Model Year
2020
2030
Reg Class
HHD8
MHD67
LHD45
— LHD2b3
Figure 3-7. Heavy Duty Gasoline Running Exhaust THC Emission Rates for the 0-3 Age Group by Model Year and
Regulatory Class using Nationally Representative Operating Mode Distributions
The CO emission rates are shown in Figure 3-8. The CO emission rates for LHD2b3 vehicles (the largest
regulatory class of heavy-duty gasoline) follow a generally decreasing trend with model year. The trends
for LHD45 and MHD show unexpected variation across model years, including an increase in CO emission
rates for LHD45 and MHD vehicles. We have the most confidence in the most recent model year data, and
the variability in the model year trends reflects uncertainty in the earlier heavy-duty gasoline emission
rates.
176
-------
Reg Class
HHD8
— MHD67
LHD45
LHD2b3
Figure 3-8. Heavy Duty Gasoline Running Exhaust CO Emission Rates for the 0-3 Age Group by Model Year and
Regulatory Class using Nationally Representative Operating Mode Distributions
0-
1970
1980
1990
2000 2010
Model Year
2020
2030
Figure 3-9 shows that the NOx emission rates follow decreasing trends with model years that correspond
with tighter emission standards.
177
-------
1970
1980
1990
2000
Model Year
2010
2020
2030
Reg Class
HHD8
— MHD67
— LHD45
— LHD2b3
Figure 3-9. Heavy Duty Gasoline Running Exhaust NOx Emission Rates for the 0-3 Age Group by Model Year and
Regulatory Class using Nationally Representative Operating Mode Distributions
3.1.2 Particulate Matter (PM2.s)
The available studies from which to develop PM25 emission for heavy-duty gasoline are partiuclarly limited.
This includes limitations on second-by-second data from which to develop operating mode specific rates,
as well as studies representative of in-use and fleet average emissions. At the same time, heavy-duty
gasoline is a relatively small contributor to the total PM2.5 emissions inventory when compared to heavy-
duty diesel and light-duty gasoline. As a result, the limited analysis conducted for MOVES2010 has been
carried over into MOVES3 for the 2009 and earlier vehicles as discussed in Section 3.1.2.1. For MOVES3, we
have updated the 2010 and later model year heavy-duty gasoline emission rates to be based on heavy-
duty diesel rates as discussed in Section 3.1.2.2.
3.1.2.1 1960-2009 Model Years
For MOVES, the MY 1960-2009 heavy-duty gasoline PM2.5 emission rates were calculated by multiplying
the MOVES2010b light-duty gasoline truck PM2.5 emission rates by a factor of 1.40, as explained below.
Since the MOVES light-duty gasoline PM2.5 emission rates comprise a complete set of factors classified by
particulate sub-type (EC and nonECPM), operating mode, model year and regulatory class, the heavy-duty
PM2.5 emission factors are also a complete set. No change to the PM emission rates is made between MY
2003 and 2009, because the HD 2007 Rule PM standards are not expected to change in-use emissions for
178
-------
medium- and heavy-duty gasoline vehicles. As presented in the next subsection, the simulated age 0-3 HD
gasoline MY 1960-2009 emission rates on the UDDS is ~6.6 mg/mile, while the standard for 2008+ spark-
ignition vehicles is 20 mg/mile107
3.1.2.1.1 Data Sources
The factor of 1.4 used to convert light-duty gasoline PM rates to heavy-duty rates was developed based on
PM2.5 emission test results from the four heavy-duty gasoline trucks tested in the CRC E55-E59 test
program. The specific data used were collected on the UDDS test cycle. Each of the four vehicles in the
sample received two UDDS tests, conducted at different test weights. Other emission tests using different
cycles were also available on the same vehicles but were not used in the calculation. The use of the UDDS
data enabled the analysis to have a consistent driving cycle. The trucks and tests are described in Table
3-10.
Table 3-10 Summary of Data Used in HD Gasoline PM Emission Rate Analysis
Vehicle
MY
Age
Test cycle
GVWR
[lb]
PM2.5 mg/mi
1
2001
3
UDDS
12,975
1.81
2001
3
UDDS
19,463
3.61
2
1983
21
UDDS
9,850
43.3
1983
21
UDDS
14,775
54.3
3
1993
12
UDDS
13,000
67.1
1993
12
UDDS
19,500
108.3
4
1987
18
UDDS
10,600
96.7
1987
18
UDDS
15,900
21.5
The table shows the four vehicles, two of which are quite old and certified to fairly lenient standards. A
third truck is also fairly old at twelve years and certified to an intermediate standard. The fourth is a
relatively new truck at age three and certified to a more stringent standard. No trucks in the sample are
certified to the Tier 2 or equivalent standards.
Examination of the heavy-duty data shows two distinct levels: vehicle #1 (MY 2001) and the other three
vehicles. Because of its lower age (3 years old) and newer model year status, this vehicle has substantially
lower PM emission levels than the others, and initially was separated in the analysis. The emissions of the
other three vehicles were averaged together to produce these mean results:
179
-------
Mean for Vehicles 2 through 4:
Mean for Vehicle 1:
3.1.2.1.2
65.22 mg/mi Older Group
2.71 mg/mi Newer Group
LHD
To compare these rates with rates from light-duty gasoline vehicles, we simulated UDDS cycle emission
rates based on MOVES2010b light-duty gas PM25 emission rates (with normal deterioration assumptions)
for light-duty gasoline trucks (regulatory class LDT). The UDDS cycle represents standardized operation for
the heavy-duty vehicles.
The simulated light-duty UDDS results were then compared to the results from the four heavy-duty gas
trucks in the sample. Emission rates from the following MOVES model year groups and age groups for light-
duty trucks were used:
• MY group 1983-1984, age 20+
• MY group 1986-1987, age 15-19
• MY group 1991-1993, age 10-14
• MY group 2001, age 0-3
The simulated PM25 UDDS emission factors for the older light-duty gas truck group using MOVES2010b are
38.84 mg/mi (ignoring sulfate emissions which are on the order of lxlO"4 mg/mile for low sulfur fuels). This
65.22^
value leads to the computation of the ratio:—-—— = 1.679 •
38-84^
The simulated PM2 5 UDDS emission rates for the newer light-duty gas truck group are 4.687 mg/mi using
MOVES2010b. Ignoring sulfate emissions, which are in the order of lxlO"5 mg/mile for low sulfur fuels, this
2 m8
value leads to the computation of the ratio:—-—2^ = 0 578 •
4.687^
The newer model year group produces a ratio which is less than one and implied that large trucks produce
less PM2.5 emissions than smaller trucks. This result is counter-intuitive and is the likely result of a very
small sample and a large natural variability in emission results.
Thus, all four data points were retained and averaged together by giving the older model year group a 75
percent weighting and the newer model year group (MY 2001) a 25 percent weighting. This is consistent
with the underlying data sample. It produces a final ratio of:
180
-------
Ratiofinal = RatioolderWtFrac + Rationewer(l — WtFrac)
= 1.679x0.75 + 0.578x0.25 = 1.40
Equation 3-1
We then multiplied this final ratio of 1.40 by the light-duty gasoline truck PM2.5 rates to calculate the input
emission rates for heavy-duty gasoline PM2.5 rates.
This approach is similar to how the LHD THC, CO, and NOx emissions for MY 2008 and 2009 were estimated
by using the light-duty gasoline truck emissions as the basis, with VSP-based light-duty rates applied as
STP-based LHD2b3 emission rates.This assumption was deemed an acceptable approximation because the
LHD2b3 gasoline vehicles are chassis certified to distance-based standards (g/mi). Accordingly, the vehicle
emissions rates are less dependent on the individual power and weight of the vehicle, and should scale
approximately to the the g/mile emission standards.ee
3.1.2.1.3 MHD and HHD
For MHD and HHD regulatory classes, the emission rates are based on a fscaie of 17.1. The LHD emission
rates are based on the light-duty truck rates, with an fSCaie of 2.06.
We used an indirect approach to derive MHD and HHD PM2.5 emission rates from the LHD emission rates.
We assume that the relationship of total hydrocarbon (THC) between emission rates based on an fSCaie of
2.06 and 17.1 is a reasonable surrogate to map PM2.5 emission rates from an fSCaie of 2.06 and 17.1 because
both pollutants are products of incomplete fuel combustion and unburned lubricating oil. For the
mapping, we first calculated the emission rate ratio for THC emissions for each operating mode between
regulatory class MHD (regClassID 46) and LHD2b3 (regClassID 41). We then multiplied this ratio by the EC
and nonEC PM2.5 emission rates in regulatory class LHD2b3 (regClassID 41) to obtain EC and nonEC
emission rates based on the 17.1 fscaie used in the heavier regulatory classes (RegClassID 46 and 47). An
example of the regulatory class LHD2b3 EC emission rates, 17.1/2.06 fSCaie THC ratios, and the calculated
17.1 fscaie based EC emission rates are displayed in Table 3-11. No reductions are made between 2003 and
ee This approximation needs to be revisited in the future now that we have updated the mass of LHD vehicles in
MOVES3 to range from 3.5 to 7.8 metric tons9, which differs from the/sCa/e value of 2.06 metric tons.ff For example,
the LHD gasoline PM2.5 age 0-3 emission rates for model year 2016 are on average 5.5 mg/mile and 7 mg/mile for
LHD2b3 and LHD45, respectively, using nationally representative operating mode distributions (See Figure 3-11). In
contrast, the MHD gasoline PM2.5 rates are lower than the comparable MHD diesel PM2.5 emission rates.
181
-------
2009, because the 2007 HD rule is not anticipated to cause reductions in heavy-duty gasoline PM2.5
emissions.
182
-------
Table 3-11. Derivation of MHD and HHD Elemental Carbon Emission Rates from LHD2b3 Rates using/sra/e 17.1/2.06
THC emission ratios. Using Model Year 2001 as an Example
opModelD
LHD2b3EC
emission
rates (mg/hr)
fscale
17.1/2.06
THC
emission
ratios
MHD and HHD EC
emission rates
(mg/hr)
0
0.59
1.000
0.59
1
0.54
1.000
0.54
11
0.60
1.000
0.60
12
0.79
2.263
1.78
13
1.38
3.677
5.08
14
2.62
5.095
13.37
15
5.55
5.443
30.22
16
64.52
5.427
350.13
21
8.38
1.000
8.38
22
2.92
1.154
3.37
23
2.08
2.173
4.52
24
2.92
2.825
8.24
25
10.94
4.842
52.95
27
20.50
7.906
162.10
28
126.42
8.796
1,112.05
29
523.16
6.471
3,385.32
30
2,366.75
7.102
16,809.50
33
26.59
2.121
56.40
35
10.76
4.780
51.42
37
13.29
4.010
53.28
38
43.61
8.979
391.56
39
75.73
9.522
721.06
40
74.96
5.300
397.26
183
-------
The resulting PM2 5 emissions by regulatory class for LHD, MHD and HHD are shown in Figure 3-10. In
general, PM2.5 emission rates are of similar magnitude for each regulatory class between model year 1980
and 2009. There is significant variation in the model years, with some unexpected trends (e.g., LHD45 has
higher emission rates than HHD and MHD for most of these model years). These unepectected trends and
variation in the emission rates across model years and regulatory class reflect uncertainties in deriving the
pre-2010 emission rates heavy-duty gasoline from light-duty gasoline data and THC surrogate values.
3.1.2.2 2010 and Later Model Years
The real-world PEMS-based emissions measurement data from two engine-certified and one chassis-
certified heavy-duty gasoline vehicles used to update the THC, CO, and NOx emission rates (Section
3.1.1.2.1) did not include PM2.5- Lacking appropriate PM data by operating mode, we populated the MY
2010+ HD gasoline PM2.5 rates by copying MY 2010+ HD diesel PM2.5 rates from a draft MOVES3 version
used for the preliminary Cleaner Trucks Initiative (CTI) analysis.15
This decision was supported by analysis of laboratory chassis tests. Gravimetric filter-based PM2.5
emissions measured from the three HD gasoline vehicles (described in Section 3.1.1.2.1) over various
chassis-dynamometer tests are shown in Table 3-12 The average PM2.5 rate over all vehicles and test cycles
is 1.35 mg/mi. The average PM2.5 emission rate for MY 2016 age 0-3, LHD diesel (comparable to the tested
gasoline vehicles) using nationally representative operating mode distributions and average speeds is 1.4
mg/mile (See Figure 2-28). Since those numbers are comparable given the uncertainty of the PM2.5
emission rates, and the fact that no modal HD gasoline PM2.5 data available, we decided to use the HD
diesel PM2.5 rates for HD gasoline. These rates also include the tampering and mal-maintenance age effects
for model year 2010 and later (see Appendix B.8).
Table 3-12 PM2.5 Emissions for Lab-Based Cycles for HD Gasoline Vehicles1
Vehicle
FTP
HWFET
LA92
Supercycle
Average
2015 ISUZU NPR
1.74
0.75
1.69
2.73
1.64
2016 Ford E450
0.53
0.55
1.55
2.51
1.17
2017 RAM 3500
1.68
0.40
1.43
1.35
1.34
Average
1.36
0.57
1.53
2.24
1.35
Note:
1 The vehicles are described in section 3.1.1.2.1.
The draft diesel LHD2b3 and LHD45 PM2.5 rates were copied to the gasoline LHD2b3 and LHD45 rates,
respectively, from a MOVES version used for the preliminary CTI analysis15. Since the diesel MHD rates
184
-------
were notably higher than the diesel LHD and HHD rates, the diesel HHD rates were used for gasoline MHD
and HHD. Note that after this analysis, the heavy-duty diesel PM2.5 emission rates were updated to
account for the updated HDIUT sample and model year split described in Section 2.1.2.2. For this reason,
the zero-mile PM2.5 emission rates from heavy-duty gasoline are constant for 2010 and later model years,
whereas the heavy-duty diesel PM2.5 emission rates are reduced starting in model year 2013, and the
heavy-duty gasoline rates for 2013 and later are generally higher than the comparable heavy-duty diesel
rates.ff LHD45 and LHD2b3 are higher than than the HHD rates for all 2010 and later years, whereas this
trend is only seen in the initial model years (2010-2013) of the heavy-duty diesel PM25 rates from the
HDIUT program (Figure 2-28). We intend to update the HD gasoline rates to be consistent with the
updated HD diesel rates in an upcoming MOVES update.
The gasoline rates were copied from the diesel rates as PM25, and then allocated to EC and nonECPM using
gasoline-specific fractions based on the Kansas City study of light-duty cars and trucks as described in the
MOVES3 Speciation Report.1 Because the diesel EC (9.98 percent) and nonEC (90.02 percent) split of PM2.5
differs from gasoline EC (14 percent) and nonEC (86 percent), the EC and nonECPM emissions rates stored
in the MOVES database are also quite different than the diesel emission rates. Figure 3-12 shows the EC
and nonECPM emission rates for gasoline LHD45 vehicles by model year for age 0-3 vehicles.
The MOVES heavy-duty gasoline PM2.5 emission rates are constant for MY 2013 and later model years.
There are differences between the 2010-2012 and 2013-2060 model year groups due to different
tampering and mal-maintenance assumptions applied to the diesel emission rates (see Section 2.1.2.1),
which primarily impact the ages 4-5 and older ages. The Tier 3 rulemaking sets PM FTP emission standards
for Class 2b and Class 3 of 8 mg/mile and 10 mg/mile, respectively, which began phase-in starting with
model year 2018 vehicles.53 We did not model reductions in the gasoline PM2.5 emission rates with the
phase-in of Tier 3, because the data on the tested heavy-duty gasoline vehicles (Table 3-12) suggests that
the heavy-duty gasoline vehicles are well in compliance with the Tier 3 standard. In addition, the diesel
rates on which the gasoline rates are based also are well in compliance with the Tier 3 standards as
discussed in Section 2.1.1.5.5.
ff For example, the LHD gasoline PM2.5 age 0-3 emission rates for model year 2016 are on average 5.5 mg/mile and 7
mg/mile for LHD2b3 and LHD45, respectively, using nationally representative operating mode distributions (See
Figure 3-11). In contrast, the MHD gasoline PM2.5 rates are lower than the comparable MHD diesel PM2.5 emission
rates.
185
-------
3.1.2.3
Model Year Trends
Figure 3-10 and Figure 3-11 display the PM2.s rates by model year and regulatory class for age 0-3 age
group estimated in grams per mile (g/mile) using nationally representative operating mode distributions
and average speeds.
_0)
|
30.15
0)
ra
ce
c
o
« 0.10
e
LU
sz
X
LU
2 0.05
o
H
U*>
CN
1
A
¦iiiiiiii
}
/ \ A
L U\
o.oo-
1970
1980
1990
2000
Model Year
2010
2020
2030
Reg Class
— HHDS
MHD67
— LH045
LHD2b3
Figure 3-10. Heavy Duty Gasoline Running Exhaust PM2.5 Emission Rates for the 0-3 Age Group by Model Year and
Regulatory Class using Nationally Representative Operating Mode Distributions
186
-------
U)
-------
0.06-
5
0)
ro 0.04-
Ol
w
w
LU
0.02-
o.oo-
EC
NonEC
IlllllllllllllWH
1970 1980 1990 2000 2010
Model Year
2020
2030
Figure 3-12 Heavy Duty Gasoline Running Exhaust PM2.5 Emission Rates by Elemental Carbon and Non-Elemental
Carbon (nonEC) Fractions for the 0-3 Age Group by Model Year and Regulatory Class using Nationally
Representative Operating Mode Distributions
3.1.3 Energy
3.1.3.1 1960-2009 Model Years
3.1.3.1.1 LHD
The energy rates for gasoline LHD (LHD2b3 and LHD45 regulatory classes) pre-2009 energy rates are
unchanged from MOVES2010a. In MOVES2010a, the energy rates for LHD2b3 and LHD45, along with the
light-duty regulatory classes, were consolidated across weight classes, engine size and engine technologies,
as discussed in the MOVES2010a energy updates report58.
3.1.3.1.2 MHD and HHD
The energy rates for gasoline MHD and HHD pre-2009 energy rates are unchanged from MOVES2014. The
rates were developed using the same data set we used to develop the THC, CO, and NOx exhaust emission
rates. Similar to the analysis for the diesel running exhaust energy rates, we made no distinction in rates by
model year, age, or regulatory class. To calculate energy rates (kJ/hour) from C02 emissions, we used a
heating value (HV) of 122,893 kJ/gallon and C02 fuel-specific emission factor (fC02) of 8,788 g/gallon for
gasoline (see Equation 3-20). STP was calculated using Equation 1-6. Figure 3-13 presents the gasoline
running exhaust energy rates in MOVES for these regulatory classes.
188
-------
6
5
—i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating mode
Figure 3-13. Gasoline Running Exhaust Energy Rates for MHD (1960-2009) and HHD (1960-2009)
A linear extrapolation to determine rates at the highest operating modes in each speed range was
performed analogously to diesel energy and NOx rates (see Section 2.1.1.4.2).
3.1.3.2 2010-2060 Model Years
The real-world PEMS-based emissions measurement data from two engine-certified and one chassis-
certified heavy-duty gasoline vehicles used to update the THC, CO, and NOx emission rates (Section
3.1.1.2.1) included C02 emissions data which was used to update the energy rates. The energy rates are
derived using the measured C02 values and the conventional gasoline specific values for carbon content
(0.0196 g/KJ) and oxidation fraction (1.0) and the molecular mass of C02 (44), and atomic mass of Carbon
(12). These values are described in the MOVES GHG and Energy Rates report.3
When calculating the operating mode-based energy rates for high-power operating modes with limited or
missing data, we extrapolated using STP values using the method described in Section 2.1.1.4.2.
For LHD2b3, the energy rates are identical for MY 2010-2013. For LHD45, MHD, and HHD, the energy rates
are identical for MY 2010-2015.
189
-------
3.1.3.2.1 LHD2b3 2014-2060 Model Years
The LHD2b3 gasoline energy rates are reduced to incorporate the impacts of the Phase 1 and Phase 2
Heavy-duty Greenhouse Gas rules. The LHD2b3 gasoline rates are adjusted from the 2010-2013 model year
rates using the gasoline reductions documented in Table 2-26 (Phase 1) and Table 2-27 (Phase 2) in Section
2.1.4.3.
Table 3-13 Estimated Total Vehicle Reductions in Energy Consumption Rates for LHD2b3 Gasoline Vehicles due to
the HD GHG Phase 1 Program
Regulatory
Class
Model years
Reduction from MY
2013 Energy Rates
LHD2b3
2014
1.5%
2015
2%
2016
4%
2017
6%
2018-2020
10%
The HD GHG Phase 1 reductions for the affected model years are incorporated into the energy rates in the
emissionRate table in the MOVES database. The adjustments for HD GHG Phase 2 are applied at run-time
using the values in the emissionRateAdjustment table in the MOVES database.
3.1.3.2.2 LHD45, MHD, and HHD 2016-2060 Model Years
Updates to the energy rates were made to the heavy-duty gasoline energy rates for model years 2016-
2020 based on the Phase 1 Medium and Heavy-Duty Greenhouse Gas Rule109 discussed in Section 2.1.4.3
and shown in Table 3-14.
Table 3-14 Heavy-Duty Gasoline Reductions due to the Heavy-Duty GHG Phase 1 Rule109
Regulatory Class
Model
Years
CO2 Reduction From 2013
Baseline
LHD45, MHD, HHD
2016-2020
5%
The energy rates for 2021 model year and beyond were updated in MOVES3 to reflect the C02 emission
reductions expected from the Heavy-Duty GHG Phase 2 rule, as shown in Table 3-15, which have separate
reductions for vocational and combination trucks.
190
-------
As noted above, the HD GHG Phase 2 reductions to energy rates are not incorporated into the energy rates
in the emissionRate table in the MOVES database, but are applied at run-time using the values in the
emissionRateAdjustment table in the MOVES database.
Table 3-15 Heavy-Duty Gasoline Reductions due to the Heavy-Duty GHG Phase 2 Rule110
Source Type
(SourceTypelD)
Regulatory Class
Model
Years
CO2 Reduction from
2017 Baseline
Other Bus, School Bus,
Refuse Truck, Single-Unit
Short-Haul, Single-Unit
Long-Haul, Motorhomes
(41, 43, 51, 52, 53, 54)
LHD45
2021-2023
6.9%
2024-2026
9.8%
2027+
13.3%
Other Bus, School Bus,
Refuse Truck, Single-Unit
Short-Haul, Single-Unit
Long-Haul, Motorhomes
(41, 43, 51, 52, 53, 54)
MHD and HHD
2021-2023
6.9%
2024-2026
9.8%
2027+
13.3%
Short-haul Combination
Trucks (61)
MHD and HHD
2018-2020
0.6%
2021-2023
7.4%
2024-2026
11.9%
2027+
15.0%
3.1.3.3 Model Year Trends
Figure 3-14 and Figure 3-15 display the C02 (g/mile) emission rates and fuel economy values calculated
from the energy rates using the carbon content and energy density conversion factors for conventional
gasolinegg as documented in the MOVES3 Greenhouse Gas and Energy Report.3 The C02 (g/mile) emission
rates and fuel economy values are estimated using nationally representative operating mode distribution
and average speed values. Figure 3-15 displays the significant decrease in fuel economy in model year 2010
due to the updated data and analysis incorporated for model year 2010 trucks. The large change in fuel
economy is not anticipated to be real, but an artifact of the using the updated data and analysis.
gg Using the energy content of conventional gasoline (E0), the fuel economy is ~4% higher than is estimated using the
energy content of E10 gasoline. Note that E10 is estimated to be the dominant gasoline fuel sold in 2008 and later.
MOVES has the same carbon content for both fuels, so there is no estimated impact on the C02 g/mile.
191
-------
The LHD emission rates show substantial variability in the early model years (pre-1985) - we do not expect
LHD vehicles to have lower fuel economy than MHD and HHD for these years, but have not revisited these
emission rates due to the small number of pre-1985 gasoline vehicles remaining in the onroad fleet. As
discussed in Section 2.1.4.1, the detailed methodology used in MOVES2004 (which modeled different
emission rates according to vehicle weights, engine technologies, and engine sizes) introduced variability
into the energy rates within the current MOVES regulatory class emission rates for pre-2010 LHD.
The figures display that, since model year 2010, there are decreasing trends in C02 (g/mile) with
corresponding increases in fuel economy, due to the lower energy rates as well as lower source mass
values and improved road load coefficients estimated with the Phase 1 and Phase 2 heavy-duty
greenhouse gas rulemaking. The energy rates by operating mode are constant for model year 2027-2060.
However, some small differences in C02 (g/mile) or fuel economy values observed within model year
groups and regulatory classes with the same energy rates are due to differences in the nationally
representative operating modes, which are different across model years due to changing fractions of
regulatory classes among different source types.
2000-
Reg Class
HHD8
— MHD67
LHD45
LHD2b3
0-
1970
1980
1990
2000
Model Year
2010
2020
2030
Figure 3-14. Heavy-Duty Gasoline Running Exhaust C02 Emission Rates by Model Year and Regulatory Class
192
-------
o-
1970 1980 1990 2000 2010 2020 2030
Model Year
Figure 3-15. Heavy-Duty Gasoline Miles per Gallon by Model Year and Regulatory Class
3.2 Start Emissions
Representative in-use data on vehicle start emissions for heavy-duty gasoline vehicles is even less common
than running data. While some data was available (Table 3-17, Table 3-22), the MOVES analysis also relies
on deterioration patterns from light-duty vehicles, as well as ratios to the relevant engine emission
standards. For LHD2b3 gasoline vehicles, manufacturers comply with chassis (g/mile) emission standards.
For the larger regulatory classes, engine emission standards apply. We used the engine emission standards
to estimate differences in emissions between the LHD2b3 regulatory class and the heavier regulatory
classes. Most of this analysis has been carried over from MOVES2010b and MOVES2014, but the cold start
emissions for LHD45, MHD, and HHD gasoline engines of 2008 model year and later have been updated for
MOVES3 based on recent certification data.
The heavy-duty spark ignition engine emissions standards111 for the Federal Test Procedure (FTP) are
shown in Table 3-16. Note that the standards for model years 1990 through 2004 for CO and THC vary by
weight class, but not by model year, whereas those for NOx vary by model year, but not by weight class.
Also, for model years 2005-2007, a single standard is applied for NMHC+NOx, but by 2008, separate but
lower standards are again in effect. Note also that by model year 2008, the standards for the three gaseous
pollutants are the same across regulatory class.
193
-------
Table 3-16 FTP Standards (g/hp-hr) for Heavy-Duty Gasoline Engines for Model Years 1990-2008+111
Model-Year
Group
GVWR < 14,000 lb
(LHD2b3)
GVWR > 14,000 lb
CO
NMHC1
NOx
CO
NMHC1
NOx
1990
14.4
1.1
6.0
37.1
1.9
6.0
1991-1997
14.4
1.1
5.0
37.1
1.9
5.0
1998-2004
14.4
1.1
4.0
37.1
1.9
4.0
2005-2007
14.4
1.01
37.1
1.02
2008+
14.4
0.14
0.20
14.4
0.14
0.20
Note:
1 Non-methane hydrocarbons standard expressed as NMHC + NOx
3.2.1 THC, CO, and NOx
The heavy-duty gasoline vehicle start emissions for MOVES regulatory class LHD2b3 and LHD45 vehicles
are discussed in Section 3.2.1.1. Section 3.2.1.2 discusses the development of the rates for MOVES
regulatory class MHD and HHD gasoline vehicles. In Section 3.2.1.3, we summarize and compare the two
sets of start emission rates for THC, CO and NOx. Soak time adjustments are detailed in Section 3.2.3.
3.2.1.1 LHD2b3
For LHD2b3, the gaseous emission rates for MY 1960-2004 are based on data analysis of test data, and the
MY 2005+ emission rates are based on ratioing the pre-2005 rates based on the emission standards.
3.2.1.1.1 1960-2004 Model Years
To develop start emission rates for MY 1960-2004 heavy-duty gasoline-fueled vehicles, we extracted data
available in EPA's Mobile-Source Observation Database (MSOD).103 These data represent aggregate test
results for heavy-duty spark-ignition (gasoline powered) engines measured on the Federal Test Procedure
(FTP) cycle. The GVWR for all trucks was between 8,500 and 14,000 lbs, placing all trucks in the LHD2b3
regulatory class. The 1960-2004 LHD2b3 start rates are unchanged from LHD2b3 start emission rates in
MOVES2010b.
Table 3-17 shows the model-year by age classification for the data. The model year groups in the table
were designed based on the progression in NOx standards between MY 1990 and 2004. Standards for CO
and THC are stable over this period, until MY 2004, when a combined NMHC+ NOx standard was
introduced. However, no measurements for gasoline HD trucks were available for MY2004 and later.
194
-------
Start emissions are not dependent on power, and therefore, the emission rates do not need to be
calculated differently to distinguish different fscaie values as was done for running exhaust rates. As
discussed later, start emission rates are separated by regulatory classes to account for differences in the
emission standards and/or available test data.
Table 3-17 Availability of Emissions Start Data by Model-Year Group and Age Group for LHD2b3 Vehicles
Model-year Group
Age Group (Years)
Total
0-3
4-5
6-7
8-9
10-14
1960-1989
19
22
41
1990
1
29
30
1991-1997
73
59
32
4
168
1998-2004
8
8
Total
81
59
33
52
22
247
3.2.1.1.2 Estimation of Mean Rates
As with light-duty vehicles, we estimated the "cold-start" as the mass from the cold-start phase of the FTP
(bag 1) less the "hot-start" phase (Bag 3). As a preliminary exploration of the data, we averaged by model
year group and age group and produced the graphs shown in Appendix F. Sample sizes were small overall
and very small in some cases (e.g., 1990, age 6-7) and the behavior of the averages was somewhat erratic.
In contrast to light-duty vehicle emissions, strong model-year effects were not apparent. This may not be
surprising for CO or THC, given the uniformity of standards throughout. This result was more surprising for
NOx, but model year trends are no more evident for NOx than for the other two. Broadly speaking, it
appeared that an age trend may be evident.
If we assume that the underlying population distributions are approximately log-normal, we can visualize
the data in ways that illustrate underlying relationships. As a first step, we calculated geometric mean
emissions, for purposes of comparison to the arithmetic means calculated by simply averaging the data.
Based on the assumption of log-normality, the geometric mean (xg) was calculated in terms of the
logarithmic mean (x/) as shown in Equation 3-2.
X = elnX/ Equation 3-2
This measure was not appropriate for use as an emission rate, but was useful in that it represents the
"center" of the skewed parent distribution. As such, it was less strongly influenced by unusually high or
195
-------
outlying measurements than the arithmetic means. In general, the small differences between geometric
means and arithmetic means suggest that the distributions represented by the data do not show strong
skew in most cases. Because evidence from light-duty vehicles suggested that emissions distributions
should be strongly skewed, this result implied that these data are not representative of "real-world"
emissions for these vehicles. This conclusion appeared to be reinforced by the values in Figure 7-3 which
represent the "logarithmic standard deviation" calculated by model-year and age groups. This measure (s/),
is the standard deviation of natural logarithm of emissions (x/). The values of s/ were highly variable, and
generally less than 0.8, showing that the degree of skew in the data was also highly variable as well as
generally low for emissions data; e.g., corresponding values for light-duty running emissions are generally
1.0 or greater. Overall, review of the geometric means confirmed the impression of age trends in the CO
and THC results, and the general lack of an age trend in the NOx results.
Given the conclusion that the data as such are probably unrepresentative, assuming the log-normal parent
distributions allowed us to re-estimate the arithmetic mean after assuming reasonable values for s/. For
this calculation, we assumed values of 0.9 for CO and THC and 1.2 for NOx. These values approximate the
maxima seen in these data.
The re-estimated arithmetic means were calculated from the geometric means, by adding a term that
represents the influence of the "dirtier" or "higher-emitting" vehicles, or the "upper tail of the
distribution," as shown in Equation 3-3.
xa = xge 2
sf
— Equation 3-3
For purposes of rate development using these data, we concluded that a model-year group effect was not
evident and re-averaged all data by age group alone. Results of the coarser averaging are presented in
Figure 3-16 with the arithmetic mean (directly calculated and re-estimated) and geometric means shown
separately.
196
-------
10 12
14
Age (years)
Figure 3-16. Cold-start FTP Emissions for Heavy-Duty Gasoline Trucks, Averaged by Age Group Only (g = Geometric
Mean, a= Arithmetic Mean Recalculated from xt and 5/)
197
-------
We then addressed the question of the projection of age trends. As a general principle, we did not allow
emissions to decline with age. For THC and NOx, we assumed the emission rates stabilized at the maximum
level reached at the 6-7 and 8-9 age groups, respectively as shown in Table 3-18. For CO emissions, we
kept the age trends as they were, since there was only a slight decrease in CO emissions after the
maximum was reached in the 8-9 age group.
We calculated standard errors for each mean in a manner consistent with the re-calculation of the
arithmetic means. Because the (arithmetic) means were recalculated with assumed values of s/, it was
necessary to re-estimate corresponding standard deviations for the parent distribution s, as shown in
Equation 3-4.
After recalculating the standard deviations, the calculation of corresponding standard errors was simple.
Because each vehicle is represented by only one data point, there was no within-vehicle variability to
consider, and the standard error could be calculated as s/^fn. We divided the standard errors by their
respective means to obtain CV-of-the-mean or "relative standard error." Means, standard deviations and
uncertainties are presented in Table 3-18 and in Figure 3-17. Note that these results represent only "cold-
start" rates (opModelD 108). Soak time adjustments other start opModes are detailed in Section 3.2.3.
3.2.1.1.3 Estimation of Uncertainty
Equation 3-4
198
-------
Table 3-18. Cold-Start Emission Rates (g) for Heavy-Duty Gasoline Trucks, by Age Group (Italicized Values
Replicated from Previous Age Groups)
Age Group
n
Pollutant
CO
THC
NOx
Means
0-3
81
101.2
6.39
4.23
4-5
59
133.0
7.40
5.18
6-7
33
155.9
11.21
6.12
8-9
52
190.3
11.21
7.08
10-14
22
189.1
11.21
7.08
Standard Deviations
0-3
108.1
6.82
8.55
4-5
142.0
7.90
6-7
166.5
11.98
12.39
8-9
203.2
11.98
14.32
10-14
202.0
11.98
14.32
Standard Errors
0-3
12.01
0.758
0.951
4-5
18.49
1.03
1.18
6-7
28.98
2.08
2.16
8-9
28.18
2.08
1.99
10-14
43.06
2.08
1.99
199
-------
300
250
@200
$ 150
2
8 ioo
P=
50
(a) CO
k
10 15
Age (years)
20
25
18
16
14
@12
t
10
TD
8
O
o
6
CL
t
4
2
0
(b
; ino
^
10 15
Age (years)
20
25
12 -
10 -
s 8
* «
ex e
2
1<
(c) NQx
10 15
Age (years)
20
25
Figure 3-17. Cold-Start Emission Rates for Heavy-Duty Gasoline Trucks, with 95 Percent Confidence Intervals
The steps described so far involved reduction and analysis of the available emissions data. In the next step,
we describe approaches used to impute rates for model years not represented in these data. For purposes
of analysis, we delineated four model year groups: 1960-2004, 2005-2007, 2008-2017 and 2018 and later.
The rates above were used for the 1960-2004 model year group. We describe the derivation of rates for
the remaining groups below.
200
-------
3.2.1.1.4 2005-2007 Model Years
For the 2005-2017 model year emission rates, we applied reductions to the 1960-2004 emission rates, by
comparing the standards between the two model year ranges. For CO, the approach was simple. We
applied the age zero values in Table 3-18 to the 2005-2007 model year group. The rationale for this
approach is that the CO standards do not change over the full range of model years considered.
For THC and NOx, we imputed values for the 2005-2007 and 2008-2017 model-year groups by multiplying
the age zero values for the 1960-2004 emission rates in Table 3-18 by ratios expressed in terms of the
applicable standards. Starting in 2005, a combined THC+NOx standard was introduced. It was necessary for
modeling purposes to partition the standard into THC and NOx components. We assumed that the
proportions of NMHC and NOx would be similar to those in the 2008 standards, which separate NMHC and
NOx while reducing both.
We calculated the THC value by multiplying the 1960-2004 value by the fraction /Hc as shown in Equation
3-5.
(
f =.
•'hc
0-14g/hp-hr V )
(0.14 + 0.20) g/hp-hrjv s v '
1.1 g/hp - hr
= 0.37
Equation 3-5
This ratio represents the component of the 2005 combined standard attributed to NMHC. We calculated
the corresponding value for NOx as shown in Equation 3-6.
-^NOx —
^ 0.20 g/hp-hr
(0.14 + 0.20) g/hp - hr
A
1.0 g/hp - hr
4.0 g/hp - hr
= 0.147
Equation 3-6
For these heavy-duty rates, we neglected the THC/NMHC conversions, to which we gave attention for
light-duty.
3.2.1.1.5 2008-2017 Model Years
For the 2008-2017 model years, the approach to projecting rates was modified to adopt two refinements
developed for light-duty rates. First, start emission rates for the LHD2b3 gasoline vehicles were estimated
from composite rates by applying the "start split-ratio" shown in Table 3-6 to a set of rates representing
light-duty trucks in Tier-2/Bin 8. Second, we applied the age effects used for light-duty truck (LDT) start
emission rates in a draft MOVES3 version used for the preliminary Cleaner Trucks Initiative (CTI) analysis.
201
-------
In that version, we updated the deterioration effects for LDT start NOx exhaust from MOVES2014112,hh by
applying the ratios shown in Table 3-19.
Table 3-19. Deterioration Reduction Ratio for NOx Starts from MY 2009-2017 LDT Vehicles
ageGroupID
Draft MOVES3 Verson
/MOVES2014
3
1.00
405
0.85
607
0.79
809
0.73
1014
0.62
1519
0.62
2099
0.62
For THC and CO, the age effects are unchanged from the LDT multiplicative age effects from
MOVES2014.113 The resulting multiplicative age effects for start emission rates for LHD2b3 vehicles used in
MOVES3 for model years 2009-2017 are shown in Table 3-20. The start emission rates for model year 2008
are estimated by averaging the MY 2007 and 2009 emission rates across all age groups and operating
modes assuming a phase-in of 50% of the Tier 2 standards and the HD 2007 Rule in MY 2008 as we
assumed for LHD2b3 gasoline running emissions as discussed in Section 3.1.1.2.1. The relative age effects
for LHD2b3 MY 2008 and MY 2009-2017 are shown Figure 3-21.
hh In MOVES3, we incorporated additional updates to the start deterioration rates (including for NOx) for LDT as
documented in the MOVES3 light-duty exhaust emission rate report.8
202
-------
Table 3-20 Multiplicative Age Effect Used for Start Emissions for Gasoline LHD2b3 Vehicles for 2009-2017 Model
Years Adopted from the Deterioration Effects for Light-Duty Trucks
ageGroupID
THC
CO
NOx
3
1
1
1
405
1.65
1.93
1.47
607
2.20
2.36
1.74
809
2.68
2.54
2.01
1014
3.30
3.00
2.00
1519
3.66
3.35
2.26
2099
4.42
4.06
2.56
Using the LDT deteriorartion rates for starts results in the MOVES3 start emission rates for MY 2010+
gasoline LHD2b3 vehicles having a higher relative deterioration than running emission rates (compared to
Table 3-3)." We recognize this is inconsistent with our knowledge of light-duty start deterioration.8 We
plan to address this data gap with data collected on LHD2b3 in future versions of MOVES.
3.2.1.1.6 Incorporating Tier 3 Standards: 2018 and Later Model Years
Emission rates representing the phase-in of Tier-3 standards for the start-exhaust process were developed
for MOVES2014 as described in gasoline running emissions section of the MOVES2014 heavy-duty exhaust
report.108 Like the MY 2008-2017 rates, the LHD2b3 Tier 3 start rates are based on light-duty truck emission
rates scaled to higher emission standards for the LHD2b3 regulatory class. The reduction in start emissions
due to Tier 3 is relatively lower than the reductions in running emissions presented in Section 3.1.1.3.1.
rhe LHD2b3 start rates during and following the Tier 3 phase-in have relatively lower deterioration than
the start rates for the model years preceding the onset of the phase-in (MY 2008-2017) as documented in
the MOVES2014 light-duty exhaust report.113,jJ For MOVES3, we adjusted the NOx start emission rates from
those developed in MOVES3 by applying the deterioration ratios in Table 3-19 to the MOVES2014 NOx
start rates. The LHD2b3 cold start rates for THC, CO and NOx after the complete phase-in of Tier 3 phase-in
" The updated MY 2010 and later heavy-duty running gasoline rates (including LHD2b3) use the heavy-duty age
effects as discussed in Section 3.1.1.2.1 and 3.1.1.3.1.
JJ In MOVES3, the deterioration effects for all model year light-duty vehicles were updated based on updated data and
analyis. The light-duty Tier 3 emission rates no longer have different deterioration values. We plan to update the
LHD2b3 start deterioration effects to be consistent with the light-duty vehicles in an upcoming version of MOVES.
203
-------
model year 2022 are shown below in Table 3-21 The age effects of the phase-in years of Tier 3 (MY 2018-
2021) are a weighted average of the MY 2010-2017 and the MY 2022 start emission rates using the phase-
in assumptions documented in the MOVES2014 heavy-duty exhaust report.108
Table 3-21 Multiplicative Age Effect Used for Start Emissions for Gasoline LHD2b3 Vehicles for 2022-2060 Model
Years
ageGroupID
THC
CO
NOx
3
1
1
1
405
1.54
1.73
1.38
607
1.94
1.97
1.57
809
2.26
1.96
1.74
1014
2.78
2.33
1.72
1519
3.09
2.59
1.95
2099
3.73
3.15
2.21
3.2.1.1.7 Inspection and Maintenance Program Effects for LHD Gasoline Vehicles
In MOVES3.1, we updated the meanBaseRatelM values for THC, CO, and NOx start exhaust emission rates
to better reflect the l/M programs for LHD2b3 gasoline vehicles.
Due to a lack of data and analysis, in MOVES3.0.4 and earlier versions of MOVES, we assumed no benefit
from l/M programs for starts. However, for the same reason provided in Section 3.1.1.4, we updated the
gasoline LHD2b3 HC, CO and NOxstart values for MeanBaseRatelM in MOVES3.1 to reflect the same
proportional reduction (that is meanBaseRatelM/meanBaseRate) that we model for each operating
modebin and age for the light-duty trucks. For more detail on the meanBaseRate and meanBaseRatelM
values for light-duty trucks, see the MOVES3 LD report.8
3.2.1.2 LHD45, MHD, and HHD
The start emission rates from LHD45, MHD, and HHD gasoline vehicles differ from the rates for LHD2b3.
The following two subsections document the emission rates for 1960-2007 model years (Section 3.2.1.2.1)
and 2008+ model years (Section 3.2.1.2.2).
3.2.1.2.1 1960-2007 Model Years
Since bag data were lacking for MY 1960-2007 vehicles in classes LHD45 and MHD, we estimated cold start
values relative to the LHD2b3 start emission rates.
204
-------
For CO and THC, we estimated rates for the heavier vehicles by multiplying them by ratios of standards for
the heavier class to those for the lighter class. The value of the ratio for CO based on 1990-2004 model
year standards is shown in Equation 3-7.
37.1 g/hp - hr
/rn = =2.58 Equation 3-7
co 14.4g/hp -hr q
The corresponding ratio for THC for 1990-2004 model year vehicles is 1.73, as shown in Equation 3-8.
1.9 g/hp -hr
HC = T~i—7u vT = " Equation 3-8
1.1 g/hp - hr
The ratios derived in the previous two equations (2.58 and 1.73) were applied to estimate the start
emission rates for 1960-2004 and 2005-2007 model year groups for the LHD45, MHD, and HHD gasoline
vehicles (Table 3-24 ). Note that the ratios for CO and THC do not vary by model year group because the
standards do not; See Table 3-15.
For MY 1960-2007, NOx start emission rates for medium and heavy-duty vehicles are equal to the LHD2b3
start emission rates, because the same standards apply to all the HD regulatory classes. The approaches for
all three regulatory classes in all model years are summarized in Table 3-24 .
3.2.1.2.2 2008-2060 Model Years
The cold start emissions for 2008 model year and later LHD45, MHD, and HHD gasoline engines have been
updated for MOVES3 based on new data. Similar to the approach taken for light-duty vehicles and for
diesel vehicles (see Section 2.2.1.2), the cold start emissions are calculated as the difference in emissions
between a test cycle with a cold start and the same test cycle with a hot start. Heavy-duty gasoline engines
are certified using the Heavy-Duty Gasoline Engine Federal Test Procedure (FTP) cycle.114 The test
procedure for certification requires that manufacturers run the engine over the FTP cycle with a cold start
and then repeat the cycle with a warm start. Starting in model year 2016, EPA began collecting certification
data that contained separate cold and hot results for each engine certified. The data that was analyzed for
this MOVES3 update includes the following engine families from the 2016 and 2017 model years shown in
Table 3-22.
205
-------
Table 3-22 Engine Data Analyzed to Revise the Cold Start Emission Rates for HD Gasoline Engines
Category
Number of Engines
Manufacturers
LHD45,
MHD, HHD
Gasoline
3
Ford, GM, Powertrain
Integration
The certification data was used to determine the grams emitted per cold start using Equation 3-9.
Grams per Start
= [Cold FTP Emission Results (g/(hp — hr))
Equation 3-9
— Hot FTP Emission Results(g/(hp — hr))]
* FTP Cycle Work (hp — hr)
The amount of work (hp-hr) performed over the FTP cycle is not provided as part of the certification data
submitted by the manufacturers to EPA. We only had cycle work data from one 19.3 hp-hr HD gasoline
engine. While we acknowledge that FTP cycle work is unique to each engine because it is created based on
the engine's maximum speed, curb idle speed, and the maximum torque curve, we estimated cycle work
for all HD gasoline engines using our one engine data source.
The analysis of cold and hot start FTP emissions data from three HD gasoline engines determined the
grams per start for THC, CO, NOx, and PM2.5- The mean and standard deviation of the THC, CO, NOx, and
PM2.5 emission levels for the three engines are shown in Table 3-23. The MY 2016 and 2017 engines ranged
in displacement between 5.4 and 7.2 liters, and ranged in rated power between 297 and 332 HP. The new
default cold start emissions values for MOVES3 are the mean values shown in Table 3-23. The THC, NOx
and PM2.5 cold start emissions for HD gasoline engines are higher compared to MOVES2014, while the CO
emissions are lower.
Table 3-23 Cold Start Emissions for MY 2008 and Later Heavy-Duty Gasoline Engines
Grams per Start
THC
CO
NOx
PM2.s
Mean
5.57
31.5
1.88
0.084
Standard Deviation
0.6
6.36
1.04
0.049
We applied the same relative age deterioration for the 2008+ model years starts for THC, CO and PM2.5 as
was used for the previous model year groups (which is based on the gasoline LHD2b3 1960-2004 model
years). For NOx, we applied the relative age deterioration as was used for LHD2b3 vehicles for MY 2008 and
206
-------
later vehicles shown in Table 3-20 and Table 3-21.kk The start rates for THC, CO, and NOx for this model
year group for each age are graphed in Figure 3-21.
3.2.1.3 Summary
Table 3-24 summarizes the data and methods used to estimate THC, CO, and NOx start emission rates from
heavy-duty gasoline vehicles as discussed in Sections 3.2.1.1 and 3.2.1.2. Figure 3-18 through Figure 3-20
displays the cold start (operating mode 108) emission rates across model years for heavy-duty gasoline
vehicles.
kk In a future update to MOVES, we intend to update the HD gasoline deterioration to be consistent with the updates
made to the LD rates in MOVES3, as well as apply a consistent approach for HD gasoline emissions deterioration for
both start and running deterioration and for all pollutants.
207
-------
Table 3-24 Summary of Cold Start Emission Rates for Heavy-Duty Gasoline Vehicles
Regulatory
Class
Model Year
Group
CO
THC
NOx
LHD2b3
1960-2004
Data analysis, values from
Table 3-18
2005-2007
Data analysis, values
from
Table 3-18
Reduce in proportion
to standards from 1960-2004
2008 - 2017
Section 3.2.1.1.5 Based on Tier 2 Bin 8 LDT rates and deterioration
2018 +
Section 3.2.1.1.6. Based on LDT rates, adjusted to account for Tier 3
standards and assumed lower deterioration
LHD45,
MHD, HHD
1960-2004
Increased
in proportion
to standards from LHD2b3
Same values as
LHD2b3
2005-2007
Increased in proportion
to standards from LHD2b3
Same values as
LHD2b3
2008 +
Updated based on FTP certification data,
deterioration based on the 1960-2004
LHD2b3 data
Updated based on FTP
certification data,
deterioration based
on the 2008+ LHD2b3
vehicles
Note:
Soak time adjustments are detailed in Section 3.2.3.
The outcomes of the methods described in the table above are summarized graphically in Figure 3-18
through Figure 3-20 for cold-start emissions. The decline in start emissions with the adoption of more
stringent standards begins with the reduction in model year 2005 and ends at the completion of the phase-
in of Tier 3 standards for LHD2b3 vehicles in model year 2022. Note that there is a slight increase in THC
start emissions for LHD45 vehicles in model year 2008, which is the first model year using the new start
certification data discussed above in Section 3.2.1.1.5.
208
-------
tl
03
J5 9-
3
0
•*—>
CO
o R_
to o
tn
LU
O
X
CO
ro 3-
O
"ro
-l—'
o
Reg Class
HHD8
— MHD67
LHD45
LHD2b3
1980
2000
Model Year
2020
Figure 3-18 Heavy-duty Gasoline THC Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year. LHD45, MHD and HHD are equivalent.
209
-------
t 200 —
CO
-t—<
U)
3
0)
¦*—>
co
QL
n.
o
en
¦| 100-
UJ
o
o
Reg Class
HHD8
— MHD67
LHD45
LHD2b3
1980
2000
Model Year
2020
Figure 3-19 Heavy-duty Gasoline CO Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year. LHD45, MHD and HHD are equivalent.
210
-------
JO
O)
res
o;
c
o
\r>
w 2 -
E
LU
X
O
Reg Class
— HHD8
MHD67
LHD45
— LHD2b3
1980
2000
Model Year
2020
Figure 3-20 Heavy-duty Gasoline NOx Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class and
Model Year. LHD45, MHD and HHD are equivalent.
2004
20-
15-
10-
5-
bob-
-r
400-
ro
300-
cn
200-
U)
100-
o-
6
4-
2-
o-l
2007
2008
2017
2022
0 4 8 12 16 20 240 4 8 12 16 20 240 4 8 12 16 20 240 4 8 12 16 20 240 4 8 12 16 20 24
Vehicle age, years
Reg Class - LHD2b3 - LHD45
Figure 3-21 Heavy-duty Gasoline Cold-Start Rates (opModelD 108) vs. Vehicle Age for Select Model Years LHD45,
MHD and HHD are equivalent.
211
-------
3.2.2 Particulate Matter (PM2.s)
Data on PM2.5 start emissions from heavy-duty gasoline vehicles were unavailable, so these emissions were
extrapolated as described below.
3.2.2.1 LHD2b3
For LHD2b3 vehicles, we used the multiplicative factor from the running exhaust emissions analysis of 1.40
(derived in Equation 3-1 in Section 3.1.2.1.2) to scale up start emission rates from light-duty trucks (LDT)
for model years 1960-2003.
For 2004+ model years, the LHD2b3 start emission rates are 1.4 times the model year 2003 LDT emission
rates. We project constant start emissions using the 2003 model year emission rates rather than scaling to
the LDT PM2.5 rates with the 2004 and later model years because the LD rates increase due to the updated
data on emission rates and sales penetration of gasoline direct injection technology, and subsequently,
decrease beginning in model year 2018 with the implementation of the Tier 3 Vehicle Emissions and Fuel
Standards Program. We are not confident that such patterns will apply to HD gasoline due to limited data
regarding heavy-duty PM2.5 rates and uncertainty regarding (a) the expected penetration of gasoline direct
injection technology in heavy-duty gasoline vehicles and (b) the impact of Tier 3 on HD gasoline PM2.5
emissions (see Section 3.1.2.2).
The start PM2.5 emission rates for heavy-duty gasoline vehicles exhibit the same relative effects of soak
time, and deterioration as the LDT PM2.5 start emission rates.
3.2.2.2 LHD45, MHD, and HHD
Due to a lack of PM2.5 start data, we use the same PM2.5 emission rates for LHD2b3 for all heavy-duty
gasoline for MY 1960-2007. For MY 2008 and later, for LHD45, MHD, and HHD, we updated the PM2.5 start
emissions data using certification data presented in Table 3-23. This causes the start emissions to increase
significantly for LHD45, MHD, and HHD between MY 2007 and MY 2008 as shown in Figure 3-22. We
continue to apply the same age adjustments to the start PM2.5 as the LDT emissions due to lack of data. We
caution there is considerable uncertainty in the start heavy-duty gasoline PM2.5 emission rates, especially
for pre-2007 model years.
212
-------
3.2.2.1
Model Year Summary
Figure 3-22 displays the cold start emission rates across model years for heavy-duty gasoline vehicles. For
the LHD45, MHD and HHD vehicles, we have more confidence in the emission rates from the 2010 and
later model year groups since they are based on certification results from these engines.
0.125
o.ooo-
1980 2000 2020
Model Year
Figure 3-22. Heavy-duty Gasoline PM2.s Cold-Start Emission Rates (g/start) for Age Group 0-3 By Regulatory Class
and Model Year. LHD45, MHD and HHD are equivalent
-z
CO
3? 0.100-
o>
0)
ro
a:
c:
o
CO
10
0.075-
LU
-g 0.050"
LU
"ro
o
£ 0.025"
c\i
Reg Class
HHD8
MHD67
LHD45
LHD2b3
3.2.3 Soak Time Adjustments
To estimate the start emissions at various soak lengths, we apply the same soak fractions to the cold start
emissions that we apply to 1996-2003 MY light-duty gasoline vehicle as documented in the light-duty
emission rate report8 and shown in Figure 2-54. These are the same adjustments used for heavy-duty
gasoline vehicles in MOVES2014.
To evaluate these adjustment ratios for MOVES3, we considered recent start emission rate data from one
heavy-duty gasoline truck. The data was gathered using PEMS using the procedure and methods discussed
in Section 2.2.3.2. The vehicle tested was a 2012 MY box truck with a gasoline engine. Figure 3-23 shows
the results from the testing as compared to the MOVES adjustments. Because the trend in the soak time
213
-------
effects is similar to the values used in MOVES2014, and because we only had new data from one truck,
MOVES3 retains the start emission adjustement ratios used in MOVES2014.
HD Gasoline Truck Start Emission Ratio
100 200 300 400 500 600 700 800
Soak Time (min)
New Data NOx —•— New Data CO —New Data HC
¦+— MOVES NOx —•— MOVES CO —•— MOVES HC
Figure 3-23 HD Gasoline Start Emission Ratio Compared to Recent Data
3.2.4 Start Energy Rates
The heavy-duty gasoline start energy rates were originally derived in MOVES2004, and updated in
MOVES2010a as described in the corresponding reports.58 Figure 2-26 displays the C02 (g/mile) emission
rates for cold start (operating mode 108) calculated from the energy rates using the carbon content of
conventional gasoline as documented in the MOVES3 Greenhouse Gas and Energy Report.3 As shown,
there is substantial variability in the start rates between 1974 and 2000. As discussed in Section 2.1.4.1,
the detailed methodology used in MOVES2004 (which modeled different emission rates according to
vehicle weights, engine technologies, and engine sizes) introduced variability into the energy rate within
the current MOVES regulatory class emission rates.
214
-------
Table 3-25 displays the relative contribution of running and start operation to total energy consumption
from the heavy-duty gasoline regulatory classes from a national run for calendar year 2016 using a draft
version of MOVES3 used for the preliminary CTI analysis15. Like diesel vehicles, starts from gasoline vehicles
are estimated to be a relatively small contributor to the total energy demand of vehicle operation. Due to
the small contribution to the total energy inventory, we have not prioritized updating the heavy-duty
gasoline start emissions rates.
Table 3-25 Relative Contribution of Total Energy Consumption from Each Pollutant Process by Regulatory Class for
Heavy-Duty Gasoline Vehicles in Calendar Year 2016
processID
processName
LHD2b3
LHD45
MHD
HHD
1
Running Exhaust
97.8%
99.2%
99.0%
99.2%
2
Start Exhaust
2.2%
0.8%
1.0%
0.8%
The HD gasoline start energy rates are reduced for shorter soak times using the same factors for diesel
vehicles, as presented in Table 2-43. The energy rates also increase with cold temperatures using the
temperature effects documented in the 2004 Energy Report.71
The start energy rates include the projected impact of the Phase 1 Heavy-Duty GHG standards, which
began phasing-in in 2014 and have the same reductions as the running energy rates, as presented in Table
2-24 and Table 2-26. As discussed in Section 2.2.4, the start energy rates are not projected to change due
to the HD GHG Phase 2 standards.
215
-------
500-
-e
to
.CO
s
0
-*—>
03
CT
c
o
400
§ 300
C/3
C/3
LU
CN
O
O
o
200-
•v
Reg Class
— HHD8
MHD67
LHD45
LHD2b3
&100-
o
1980
2000
Model Year
2020
Figure 3-24 Heavy-Duty Gasoline Cold Start C02 Emission Rates (g/start) by Model Year and Regulatory Class
(OpmodelD 108)
216
-------
4 Heavy-Duty Compressed Natural Gas Exhaust Emissions
While natural gas lacks the ubiquitous fueling infrastructure of gasoline, compressed natural gas (CNG),
propane, and liquefied natural gas have grown as transportation fuels for public transit, government, and
corporate fleets. Such fleets typically utilize centralized, privately-owned refueling stations. Fleet vehicles
are operated as back-to-base, which means the vehicles return to the same base location each day for
refueling. Within this segment, some of the most prevelant use of in CNG vehicles has occurred among city
transit bus fleets and in solid waste collection or refuse truck fleets.115 Figure 4-1 displays the fraction of
heavy-duty CNG fueled-vehicles by source type and model year estimated in the default national activity
database in MOVES3.9
30.00%
25.00%
o
¦
tj
CZ
£ 20.00%
o
cc
§• 15.00%
a,
jy
•§ 10.00%
0
£
U 5.00%
0.00%
1980
Figure 4-1 Fraction of Heavy-Duty CNG fueled-vehicles in MOVES3 by Source Type by Model Year
MOVES2014 modelled only CNG transit buses. In MOVES3, we allow the modeling of CNG fuel for most
heavy-duty source types. Long-haul combination trucks (source type 62) are still diesel-only in MOVES
because of the difficulties in accurately modeling hotelling for non-diesel vehicles. We hope to improve this
in a future version of MOVES.
The CNG transit buses are mapped to the urban bus regulatory class. The CNG vehicles in other heavy-duty
source types are mapped to the HHD regulatory class ). However, the base emission rates for the two
—•—Intercity Bus
Transit Bus
—School Bus
Refuse Truck
-•—Single Unit Short-liaul Truck
Single Unit Long-haul Truck
—•—Combination Short-liaul Truck
1990 2000 2010 2020 2030
Model Year
217
-------
regulatory classes are identical. Thus, any differences in CNG emissions between source types is due to
differences in population and activity.
Much of the analysis for CNG emissions, especially for older model years, is unchanged from MOVES2014.
Important updates for MOVES3 include:
• Two new model year groups, 2007-2009 and 2010+, to replace the 2007+ emission rates in
MOVES2014
• Emissions rates for MY2010+ based on real-world CNG vehicle emissions data.
• For pre-2010 model years, we still estimate emissions using vehicle certification data, but we now
use all HD CNG engine emissions data within a model year group. In MOVES2014, the certification
emission rate was limited to engine families classified as urban bus.
As noted above for diesel and gasoline vehicles, MOVES methane emissions are not estimated using
emission rates. Rather, methane is estimated in relation to THC, using ratios stored in the
MethaneTHCratio table in MOVES. The ratios are categorized by fuel type, pollutant process, source type,
model-year group, and age group. MOVES multiplies the THC rate by the corresponding ratio from the
"methanethcratio" table to calculate the CH4 rate. The methane fraction from CNG vehicles is 89% and 96%
for model year groups 1960-2004 and 2002-2060 respectively, as documented in the Speciation report.1
All criteria emission rates are dependent on vehicle age, and thus are stored in the emissionRateByAge
table.
Total energy consumption is age independent, and therefore, stored in the EmissionRate table. Some of
the published studies did not report total energy consumption directly, so it was necessary to compute
energy from a stoichiometric equation based on the carbon content in the emitted pollutants or from
reported values of miles per gallon equivalent of diesel fuel. In the former case, we used 0.8037 as the
carbon fraction coefficient for non-methane hydrocarbons (NMHC) when the bus was equipped with an
oxidation catalyst and 0.835 without due to high ethene levels, using speciation profiles from Ayala et al.
(2003)116 discussed later in this section. All other conversion factors to energy were taken from Melendez
et al. (2005).120
On a similar note, MOVES does not report particulate matter (PM2.5) as a single rate; it reports one rate for
PM from elemental carbon (EC) of 2.5 microns or less, and another rate for non-elemental carbon of 2.5
microns or less. These separate rates for PM (EC) and PM (NonEC) from the emissionRateByAge table are
added together for a total PM2 5 rate used for comparison to the measurements.
218
-------
4.1
Running Exhaust Emission Rates
The pre-2010 running emission rates are relatively unchanged from MOVES2014", and are based on cycle
average rates as discussed in Section 4.1.1. In MOVES3, we updated the running exhaust emission rates for
model year 2010 and later CNG vehicles using second-by-second in-use emission measurements from
heavy-duty vehicles as discussed in Section 4.1.2.
4.1.1 1960-2009 Model Years
Ideally, MOVES modal emission rates would be developed through analysis of second-by-second data of
vehicles of the appropriate regulatory class, model year, and age. Unfortunately, such data are not readily
available for all model years.
In particular, data at multiple ages that can be used to determine emission deterioration, and second-by-
second data that can be used to establish STP trends was very limited for MY 2009 and earlier CNG
vehicles. Thus, for MOVES, we applied STP and age trends from MHD gasoline vehicles to cycle-based
certification results. The following sections describe the available data and the methods to calculate the
adjustment ratios.
4.1.1.1 CNG Chassis Dynamometer Measurements
Chassis data was collected from programs that were conducted at several research locations around the
country on heavy-duty chassis dynamometer equipment. In our analysis for MOVES2014, we compiled 34
unique dynamometer measurements. Data from newer studies such as Clark et al. (2007)127 would provide
further validation and refinement to the rates discussed in this report, however they have not been
incorporated here.
The data considered consisted of distance-specific running emissions rates for each of the following
pollutants and total energy:
1. oxides of nitrogen (NOx)
2. carbon monoxide (CO)
3. particulate matter (EC + non-EC)
" The only change was to limit the certification data used to derive the 2007-2009 model year emission rates to the
2007-2009, rather than 2007-2017 as discussed in Section 4.1.1.2.3.
219
-------
4. total hydrocarbons (THC)
5. methane (CH4)
6. total energy consumption
This data was collected on two driving cycles, the Central Business District (CBD) and Washington
Metropolitan Area Transit Authority (WMATA).
The CBD cycle is defined as a driving pattern with constant acceleration from rest to 20 mph, a short cruise
period at 20 mph, and constant deceleration back to rest, repeated for 600 seconds (see Figure 4-2).117 The
WMATA cycle was developed using GPS data from city buses in Washington, DC, and has higher speeds
and greater periods of acceleration than the CBD cycle (see Figure 4-3).
CBD
0 100 200 300 400 500 600
Time, s
Figure 4-2 Driving Schedule Trace of the Central Business District (CBD) Cycle118
60
£ 50
1
40
a 30
:«
4>
3
>
20
10
I J nl .l v__J I
fU
\ 1 Al lift ll I 1 . I Jir
500
1000
Time (sec)
1500
2000
220
-------
Figure 4-3 Driving Schedule Trace of the Washington Metropolitan Area Transit Authority (WMATA) Cycle120
Table 4-1 shows a summary of the number of unique CNG bus measurements by driving cycle for each
study. Navistar published a similar study of CNG and diesel buses in 2008, and this analysis shares many of
the same sources.119 All of the vehicles were in service with a transit agency at the time of testing. The
number of unique measurements are typically equal to the number of vehicles tested and the
measurements were typically reported as averages based on multiple runs with the same vehicle and
configuration over a specific driving cycle with the exception of measurements reported by Ayala et al.
(2002)122 and Ayala et al. (2003).116 In the Ayala et al. (2002) study the 2000 model year CNG bus was
tested and then retested after approximately two months of service, which we treated as independent
measurements. Ayala et al. (2003) retested the same 2000 CNG bus as in their previous study; however,
the bus had accumulated an additional 35,000 miles and was serviced by the OEM to be equipped with an
oxidation catalyst that was later removed for baseline testing. Ayala et al. (2003) conducted duplicate tests
under each vehicle/aftertreatment configuration, which we considered four independent measurements.
Table 4-1. Summary of External Emissions Testing Programs by Driving Cycle and Number of Unique Measurements
and their Corresponding Model Years
Paper/Article
Lead Research Unit
Driving
Cycle(s)
Model Year
(Number of
Measurements)
Melendez
2005120
National Renewable Energy Laboratory
(NREL)
WMATA
2001 (4),
2004 (3)
Ayala 2003116
California Air Resources Board (CARB)
CBD
2000 (4),
2001 (2)
LeTavec 2002121
Atlantic Richfield Company (ARCO)
CBD
2001 (1)
Ayala 2002122
CARB
CBD
2000 (2)
Lanni 2003123
New York Department of Environmental
Conservation
CBD
1999 (3)
McKain 2000124
West Virginia University (WVU)
CBD
1999 (3)
Clark 1997125
WVU
CBD
1996 (10)
McCormick
1999126
Colorado School of Mines
CBD
1994 (2)
TOTAL
(34)
221
-------
4.1.1.2 Determining Model Year Groups
Model year groups are intended capture differences in vehicles over time while still being manageable
from a computational viewpoint. Model year groups are defined based on availability of measurement
data (see Table 4-1), emissions standards, and/or new vehicle technologies that affect real-world
emissions.
4.1.1.2.1 1994-2001 Model Years
We evaluated the measured criteria pollutant rates (NOx, CO, PM2.5, and THC) to establish model year
groups and chose to group all the CBD measurements from the literature into one model year group,
spanning from MY 1994 to MY 2001. Note that we decided to exclude one of the studies that had four MY
2001 buses tested on the WMATA cycle from this part of the analysis. This was done because inclusion
increased the complexity of analysis by having to deal with two driving cycles within a model year group
while providing only an incremental increase in sample size.
222
-------
4.1.1.2.2
2002-2006 Model Years
Of the surveyed data, only one study had vehicles newer than MY 2001.mm-127 This paper, a joint study
between NREL and WMATA, had three MY 2004 vehicles. The MY 2004 vehicles have a visibly different
emissions profile than the other vehicles. While these buses were only tested on the WMATA cycle, they
were all equipped with oxidation catalysts and had substantially lower emissions, particularly for PM2.5,
compared to the 1994-2001 buses tested on the CBD cycle. As a result, we created a model year group
from MY 2002 to MY 2006 based on the MY 2004 buses tested on the WMATA cycle. This MY group ends
before MY 2007 when a new series of stringent emission standards went into effect, as described below.128
4.1.1.2.3 2007-2009 Model Years
MOVES2014 had a single set of emissions for 2007-and later buses. In MOVES3, we created two groups,
MY 2007-2009 and MY 2010+ (noted as MY 2010-2017 when comparing certification data). We decided to
split the groups in this way because: (a) changes to fSCaie values starting MY 2010 (see Appendix G) requires
rates to be re-analyzed using 1 hz data; (b) the HDIUT data set includes real world data on MY 2010+ CNG
vehicles; (c) certification data showed a significant difference between the average emissions rates for NOx
and CO between these two model year groups (but note that certification data is not used in developing
the rates for MY2010+), and (d) this allows for better representation of differences in combustion and
aftertreatment technology, such as stoichiometric-combustion with three-way catalysts (TWC) that
became more prevalent starting year 2010.
Certification emission data for natural gas heavy-duty vehicles are publicly available by model year on the
EPA's Office of Transportation and Air Quality website.129 Analysis of these data showed that from MY
2002 to MY 2017, there have been changes in average certification levels for all the pollutants considered
in this report. In particular, NOx and PM2.5 levels have dropped dramatically. This effect is largely
attributable to increasingly stringent emission standards, which have affected both diesel and CNG
engines.
Emission rates from analysis of certification data and number of CNG engine families in the certification
data are shown in Table 4-2 below. The current, and historically most stringent, heavy-duty compression-
ignition NOx standard of 0.20 g/bhp-hr was fully phased in by 2010 and MY 2010+ heavy-duty CNG engines
are required to meet this standard (even if they are not compression-ignition). Thus, the average NOx
certification value for the MY 2010-2017 group is considerably lower compared to the MY 2007-2009
mm Several papers have discussed more recent vehicles. Examples include Clark et al. (2007).126 Data from these
newer studies would provide further validation and refinement to the rates discussed in this report, however, time
and resources were not available to complete a re-analysis for MOVES3.
223
-------
group. At the same time, and mostly to meet the new N0X standard, heavy-duty CNG engines transitioned
from lean-burn to stoichiometric-combustion with TWC. This technology transition is the likely reason for
the increase in THC and CO certification emissions rates from MY 2007-2009 to MY 2010-2017.
The differences in MOVES emission rates across all model years are discussed below in Section 4.1.3.
Table 4-2 Model Year Group Based Certification Emission Rate for Heavy-Duty CNG Engine Families
Model Year
Group
Number of Engine
Familiesab
Certification Emission Rate (g/bhp-hr)c
NOx
CO
PM2.s
NMHCd
2002-2006
22
1.208
1.355
0.0078
0.147
2007-2009
30 (24 for PM2.5)
0.6123
1.940
0.0042
0.063
2010-2017®
159 for NOx and CO, 153
for THC, and 120 for
PM2.5
0.1051
4.413
0.0028
0.044
Notes:
a For MY 2002-2006, the number of engine families is based on HD CNG urban bus regulatory class. For MY
2007-2009 and MY 2010-2017, the number of engine families is based on all FID CNG engine families.
b Some engine families did not report emission data for TF1C and/or PM2.5.
c MY 2002-2006 group emission rates are projected sales weighted average of FID CNG urban bus certification
emission rates. MY 2007-2009 and 2010-2017 group emission rates are simple average of all FID CNG
certification emission rates (no weighting for projected sales).
d Certification data has measurements of organic material non-methane hydrocarbon equivalent (OMNMF1CE).
For this analysis they were treated as NMF1C values.130
e Only shown for comparison. Certification data for MY 2010-2017 is not used in developing MY 2010+ rates,
which are based on MY 2010+ CNG vehicles in the F1DIUT data set.
4.1.1.3 Creating Comparable MOVES Gasoline Emissions
Section 1.6 explains how MOVES operating modes relate to scaled tractive power (STP). Because we
lacked data on age and STP trends for pre-2010 CNG vehicles, we applied adjustments based on the rates
for gasoline MHD vehicles. To do this, we compared CNG emission data collected on the CBD and WMATA
cycles to what MOVES estimated for MHD gasoline vehicles on those same cycles.
Because the pre-2009 CNG vehcles form a small and diminishing portion of the MOVES3 fleet, we have not
updated this analysis which relied on emissions, vehicle and activity information from MOVES2010b.
224
-------
This approach requires converting activity on the CBD and WMATA bus driving cycles to MOVES operating
mode distributions, and then simulating MHD gasoline emissions on those same operating mode
distributions.
4.1.1.3.1 Operating Mode Distributions for Transit Bus Drive Cycles
The MOVES2010b project level importer was used to input the second-by-second drive cycle for the CBD
and WMATA drive cycles. For each, a single link was created, with the test cycle entered as a drive trace.
Running MOVES2010b generated the operating mode distribution, which is created by allocating the time
spent in each operating mode according to the cycle speed and acceleration, as shown in Figure 4-4 and
Figure 4-5. The derivation of scaled tractive power (STP) and operating mode attribution for heavy-duty
vehicles are discussed earlier in this report, in Section 1.5. Road grade is set to zero because these are
chassis dynamometer runs.
Since STP is dependent on mass (among other factors), the average vehicle inertial test mass for each cycle
was inserted into the MOVES2010b sourceUseType table in place of the default transit bus mass to ensure
a more accurate simulation- 14.957 metric tons for the CBD and 16.308 metric tons for the WMATA,
compared to the MOVES2010b default of 16.556 metric tons. The STP calculations used the road-load
coefficients from MOVES2010b for transit buses, assuming the coefficients (A, B, and C) were similar to
those of the tested buses.
225
-------
350
300
— 250
51 200
(U
£ 150
P 100
50
0
a
11
T I I I I I I I I I I I I I I I I I
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating Mode
Figure 4-4 Operating Mode Distribution for the CBD Cycle
900
800
700
~ 600
8. 500
£ 400
1= 300
200
100
I
m
i i i i i i i i i i i i i i i i i i i i i i i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
Operating Mode
Figure 4-5 Operating Mode Distribution for the WMATA Cycle
4.1.1.3.2 Simulating Cycle Average Emission Rates
With the operating mode distributions determined above, and the emission rates in the MOVES2010b
database, we simulated the gasoline MHD emissions for each pollutant for each cycle. Dividing by the
cycle total distance we calculated the simulated cycle-average distance-specific rate for that cycle (ESimcycie,
g/mile), as shown in Equation 4-1. Using this method, the simulated cycle emission aggregates were
calculated as a function of the following parameters:
226
-------
• fuel type,
• driving cycle,
• age group,
• regulatory class,
• model year, and
• pollutant and process.
„ ZioM Rp,OM * T0M,cycle _ . . .
Ep.simcycie = Equation 4-1
cycle
Where:
Dcyde = distance of the cycle, in miles
Rp,om = emission rate of pollutant p in operating mode OM, in g/hr
Tom,cycle = time spent in operating mode OM for given cycle, in hr
We compared the MOVES2010b simulated MHD gasoline rates with the published chassis dynamometer
measurements. We also specified the age group and model year to match individual vehicles in the testing
programs from the literature on CNG transit buses.
4.1.1.4 Emissions Rates by Model Year Group
To estimate emission rates for pre-2010 CNG vehicles, we applied STP and age trends from MHD gasoline
vehicles to cycle-based CNG certification results. Mathematically, this is the same as applying a CNG
adjustment to MHD gasoline emission rates, which is how the analysis is described in the sections below.
Due to limited data on older vehicles in the literature, the ratios (shown in Table 4-3) developed
using vehicles in the 0-3 age group have been applied to all other age groups. In addition, we
assumed that CNG vehicles exhibit the same deterioration trend as medium heavy-duty gasoline
trucks (Table 3-3 in Section 3.1.1.1.2 forTHC, CO and NOx, and Section 3.1.2.1.3 for PM2.5).
4.1.1.4.1 1960-2006 Model Years
227
-------
The operating mode based emissions rates for MOVES2010bnn MHD gasoline vehicles were adjusted by the
ratio of cycle-average emissions rates from chassis dynamometer measurements to simulated cycle
modeling (see Section 4.1.3). For MY 1994-2001 and MY 2002-2006, the adjustment ratios were based on
the CBD cycle and WMATA cycle, respectively.
For each model year group, a central model year was selected as the source for the MHD gasoline
operating mode based rates. For MY group 1994-2001, we used MHD gasoline rates from MY 1997
because it is one of the median years in the group. Alternatively, we could have used the other median
year, MY 1998. Even though the average rate for MY 1998 was significantly lower (44 percent of that of
MY 1997), based on Equation 4-2, we expect minimal differences in the final estimated CNG rates (Rcng
term) whether we use MY 1997 or MY 1998 as the median year since the lower operating mode rates (RMdg
term) will lead to lower simulated cycle-average rate (Emdg term), which in turn will lead to larger
adjustment ratio (ECng/Emdg). For MY group 2002-2006, we used MHD gasoline rates from MY 2004
because that was the model year of the engine in each of the CNG vehicles measured on the chassis
dynamometer (the MY 2001 vehicles were not included in this group). See Equation 4-2 and Equation 4-3
for MY groups 1994-2001 and 2002-2006, respectively.
We assumed that the MY 1993 and earlier CNG vehicles have the same emission rates as MY group 1994-
2001.
n _ n ECNG CBD 1994_2ooi Equation 4-2
KCNG,OM,1994-2001 — KMDG,OM,1997 * P
^MDG.simCBD.iggy
Rcng,OM,2002-2006 — RmDG,OM,2004 * '
Ecng,wmata,2004 Equation 4-3
E]VIDG,simWMATA,2004
Where:
Rcng,om,myg = operating mode based emissions rate for CNG vehicles for model year group (MYG)
1994-2001 or MY 2002-2006, in g/hr
Rmdg,om,my = operating mode based emissions rate for MHD gasoline vehicles for model year 1997
or 2004 (corresponding to MYG), in g/hr
Ecng,cycle,myg = Chassis dynamometer cycle-average emissions rate for MY 1994-2001 or 2004 CNG
buses tested on a CBD or WMATA cycle, respectively, in g/mile. See Table 4-3.
nn The PM exhaust emission rates were subsequently updated from MOVES2010b for MHD gasoline vehicles for pre-
2010 model years.
228
-------
EMDG,simcycie,MY = Simulated cycle-average emissions rate for MY 1997 or 2004 MHD gasoline vehicles
for CBD or WMATA cycle, respectively, in g/mile. This cycle-average rate is
calculated using the Rmdg,om,my operating mode rates. See Table 4-3.
4.1.1.4.2 2007-2009 Model Years
Due to lack of published data on MY 2007-2009 in-use vehicles, we used certification emissions rates,
shown in Table 4-2, to scale the operating mode based emissions rates. Certification emissions rates are
reported in grams per brake horsepower-hour (g/bhp-hr) and are not directly used in formulating MOVES
emission rates because they do not include real-world effects such as deterioration131 which were present
in the chassis dynamometer measurements used to estimate emissions rates for MY 1994-2001 and MY
2002-2006. So, we created scaling factors that we could apply to the MY 2002-2006 emissions rates to
estimate rates for MY 2007-2009. This scaling factor is the right-most term in Equation 4-4 shown below.
n _ n Ecng WMATA 2oo4 Ccng,2007-2009 Equation 4-4
KCNG,OM,2007-2009 — KMDG,OM,2004 * P * r
tMDG,simWMATA,2004 LCNG,2002-2006
Where:
Ccng,2007-2009 = Average certification emission rate of all heavy-duty CNG engine families of model
year MY 2007-2009 in g/bhp-hr
Ccng,2002-2006 = Projected sales weighted average certification emission rate for CNG urban bus
engine families in MY 2002-2006, in g/bhp-hr
The adjustment ratio for energy consumption for MY 2002-2006 (Equation 4-3) is applied to all model
years in 2007-2009. For MY 2007+, we did not scale the energy consumption rates like we did for criteria
pollutants (Equation 4-4) because even though we have certification data on C02 emission rates for 2007-
2009 model years, we do not have certification data on C02 emission rates for MY 2002-2006. As a result,
MY 2007-2009 energy consumption rates are identical to the MY 2002-2006 rates.
229
-------
4.1.1.4.3 Ratio Summary
Table 4-3 Ratios Applied to MHD Gasoline Rates to Compute CNG Rates
Ecng, Cycle-Average Chassis Dynamometer Measurement Rates (g/mile)
MY
Age
Group
Cycle
NOx
CO
PM_NonEC
u
LU
1
CL
THC
TOTAL
ENERGY
(BTU/mi)
1994-2001
0-3
CBD
20.8
9.97
0.037
0.0038
13.2
42782
2002-2006
0-3
WMATA
9.08
2.17a
0.0039
0.0005
11.2
40900
Emdg , Simulated Cycle-Average Medium Heavy-Duty Gasoline Rates (g/mile)
MY
Age
Group
Simulated
Cycle
NOx
CO
PM_NonEC
u
LU
1
CL
THC
TOTAL
ENERGY
(BTU/mi)
1997
0-3
CBD
9.63
62.4
0.0024
0.0002
1.84
31137
2004
0-3
WMATA
5.45
18.9
0.0035
0.0003
1.43
35489
Ratios Applied to the Medium Heavy-Duty Gasoline Rates to Create CNG Rates
MY
Age
Group
MHD
Gasoline
MYb
NOx
CO
PM_NonECf
u
LU
1
CL
THC
TOTAL
ENERGY
1994-2001c
all
1997
2.16
0.160
15.5
21.6
7.17
1.37
2002-2006
C
all
2004
1.67
0.115
1.09
1.87
7.79
1.15
2007-2009
d
all
2004
0.842
0.157
0.587
1.01
3.34
1.15
2010+®
Age 0- 3 (and 4-5) rates are based on analysis of 1 hz data from MY 2010+ CNG
vehicles in the HDIUT data set. Ages 6+ apply deterioration factors to age 0-3
rates as described in main text.
Notes:
a The measured CO rate (0.14 g/mi) was uncharacteristically low and thus determined to be an outlier and not used.
Each of the three post-2001 vehicles in this study had the same MY 2004 engine (John Deere 6081H). This
engine's CO certification rate was a full order of magnitude lower than certification rate of other MY 2004 engine
models, and was not supported by additional test results. We adjusted the WMATA chassis dynamometer CO rate
by the ratio between the sales-weighted average CO certification level of all MY 2004 CNG engine models and the
CO certification level for the MY 2004 John Deere 6081H engine.
b Model year of the medium heavy-duty gasoline operating mode rates to which the pollutant-specific ratios are
applied
230
-------
cThe ratios are calculated using Equation 4.2 or Equation 4.3 and the ECNG and EMDG values in this table
dThe ratios are calculated using Equation 4.4, the Ecng and Emdg values in this table, and the Ccng values in Table
4.2
e Energy consumption rates for MY 2014-2017 and MY 2018+ are reduced as per heavy-duty GHG Phase 1 and
Phase 2 rules, respectively. See main text for details.
f The PM2.5 exhaust emission rates have been subsequently updated in MOVES3 from MOVES2010b for MHD
gasoline vehicles for MY 1997. The ratios presented here are applicable to the MOVES2010b MHD gasoline rates
and the MOVES3 CNG emission rates.
231
-------
4.1.2 2010-2060 Model Years
Running emission rates for MY 2010 and later CNG vehicles were based on information from in-use trucks,
and thus, unlike the calculations for earlier model years, the rates by operating mode could be calculated
directly. We then applied factors to account for deterioration with vehicle age, and adjustments to energy
rates to account for the phase-in of heavy-duty greenhouse gas standards.
4.1.2.1 Base Emission Rates
Running emission rates for MY 2010 and later CNG vehicles were based on information from in-use trucks.
To develop MY 2010+ emissions rates (for THC, CO, NOx, and PM2.5) and energy consumption rates, we
used the MY 2010+ CNG vehicles in the HDIUT data set. At the time of analysis, there were five MY 2011
CNG vehicles and six MY 2014 vehicles. These 11 vehicles are all stoichiometric-combustion with TWC and
are certified at or below the 0.20 g/bhp-hr standard.
After quality assurance, the 1 hz data set included about 310,000 seconds of operation. Operating modes
(Table 1-4) were assigned to the 1 hz data using the method to calculate STP described in section 2.1.1.3.
The analysis used updated/sco;e values described in section 2.1.1.4.2 and Appendix G and thus, there was
no need for hole-filling of missing operating modes. The operating mode-based rates were calculated using
fscaie = 10. The rates for regClass 47 and 48 are identical.
Unlike the analysis method for HD diesel (described in section 2.1.1.5), the method for HD CNG did not use
the NOx FEL based grouping since all 11 vehicles are in the same NOx FEL group. As a result, the zero-mile
(age 0) THC, CO, NOx, and PM2.5 rates for CNG are identical for all model years starting 2010 (unlike HD
diesel where they change for each model year in 2010-2015 based on production volume differences
between the NOx FEL groups).
4.1.2.2 Age-based deterioration factors
THC, CO, and NOx age-based deterioration factors for MY 2010+ CNG vehicles in MOVES3 are unchanged
from MOVES2014. In MOVES2014, these factors were set as equal to the factors for MY 2010+ HD gasoline
vehicles, which in turn are identical to and based on MY 1960-2007 HD gasoline vehicles. There is no
deterioration for age groups 0-3 and 4-5 and the deterioration factor (per operating mode) is same across
age groups for ages 6+ but varies between operating modes within an age group. These deterioration
factors are described in Table 3-3 in Section 3.1.1.1.2.
For PM2.5, in MOVES3, ages 0-3 and 4-5 have no deterioration and the MOVES2014 light-duty PM2.5
deterioration factor for age 6-7 is applied to all CNG PM2.5 emission rates for ages 6+, thus making the PM2.5
232
-------
and gaseous pollutant methods more (but not fully) aligned. Note that, unlike gaseous pollutants, the
PM2.5 deterioration factor does not vary between operating modes for a given age group. See Section
3.1.2.1.3 for more details and Table 4-4 for a comparison between MOVES3 and MOVES2014.
Table 4-4 Age-based Deterioration Factor for PM2.5 Emission Rates for HHD and Urban Bus CNG Vehicles in Model
Year 2010+
Age
MOVES2014
MOVES31
0-3 (Baseline)
1.00
1.00
4-5
1.57
1.00
6-7
1.75
1.75
8-9
1.96
1.75
10-14
2.38
1.75
15-19
3.14
1.75
20+
4.15
1.75
Note:
1 When recreating the deterioration factor, for age 6+, from the age-group based default emissions rates in the MOVES
database, the ratios will not be exactly 1.75 because the final rates (with deterioration factors already applied) are
rounded to a set precision before submission to the database.
4.1.2.3 Application of Heavy-Duty Greenhouse Gas Phase 1 and Phase 2 Rules
To model energy consumption in MOVES3, we split the CNG MY 2010+ group into MY 2010-2013, MY
2014-2017, and MY 2018+ groups. The MY 2010-2013 energy consumption rates are identical across these
model years and based solely on the HDIUT data set analysis. For MY 2014-2017, the CNG energy
consumption rates of MY 2013 are reduced by the percentage reduction assigned to HHD vehicles in the
Greenhouse Gas Emissions Standards and Fuel Efficiency Standards for Medium and Heavy-Duty Engines
and Vehicles Phase 1 rule60 (see Table 2-24). Similarly, for MY 2018 and later, using MY 2017 rates as base
year, the energy consumption rates of CNG vehicles are further reduced as per the Greenhouse Gas
Emissions and Fuel Efficiency Standards for Medium and Heavy-Duty Engines and Vehicles — Phase 2 rule62
(see Table 2-25). Note that the Phase 1 reduction for CNG vehicles is identical across all allowed source
type and regulatory class combinations. However, for the Phase 2, different reductions for CNG vehicles
are applied by source type and regulatory class (see Table 2-25). The anticipated improvements in fuel
efficiency from the Phase 2 rules are stored in the EmissionRateAdjustment table.
The GHG Phase 1 and 2 reductions reflect the percent improvements projected from the rules
based on engine technology improvements to diesel engines. In making these projections, we
233
-------
assumed the HD GHG rules lead to the same reductions in the energy rates for CNG vehicles as for
heavy-duty diesel. In reality, manufacturers of CNG vehicles can meet the standards by lowering
both CH4 and CO2 emissions, and the reductions in fuel consumption (and CO2 emission rates)
between CNG and diesel vehicles will likely differ. Future MOVES versions may update the energy
consumption rates and CH4 emission rates with data from MY 2015 and later CNG vehicles that
comply with the GHG standards.
4.1.3 Model Year Trends
Figure 4-6 through Figure 4-10 display the TFIC, CO, NOx, PM2.5, and C02 running exhaust emission rates by
model year and regulatory class (HHD and Urban Bus). The emission rates are estimated in grams per mile
(g/mile) using nationally representative operating mode distributions and average speeds. The change in
emissions at MY 2010 coincides with both a change in our analysis methodology and a shift in the CNG
vehicle fleet from lean-burn combustion to stoichiometric combustion with a three-way catalyst.
Figure 4-6 shows a significant increase in the TFIC emissions between the MY 2007-2009 and MY 2010 and
later vehicles. Because MOVES uses the same methane fraction for 2002 and later CNG vehicles, the CFI4
and non-methane hydrocarbons (NMFIC) emissions follow the same trend. The significant increase in TFIC,
CFI4, and NMFIC starting in MY 2010 with the increased penetration of CNG vehicles with stoichiometric-
combustion engines and three-way catatalyst (TWC) is not supported by the certification data presented in
Table 4-2 nor in recent studies comparing stochiometric and lean-burn combustion CNG engines. 132,133 The
differences in the methodologies and limitations in the pre-2010 data likely contributed to apparent
increase in TFIC emissions starting in 2010 model year in MOVES.
234
-------
10.0-
3
0
-•—»
03
cr
c
o
V)
cn
E
LU
o
X
CO
CO
CD
la
+j
o
7.5-
5.0-
2.5-
Reg Class
HHD8
Urban Bus
\J
0.0-
2000
2010
Model Year
2020
2030
Figure 4-6. Heavy Duty CNG THC Exhaust Emission Rates for the 0-3 Age Group by Model Year and Regulatory Class
using Nationally Representative Operating Mode Distributions
Like THC, CO also shows a significant increase in emission rates with the updated analysis of the MY 2010+
vehicles (See Figure 4-7). However, the increase in CO is supported by certification data (Table 4-2) and in
more recent testing comparing stoichiometric-combustion with TWC based CNG buses.132,133
235
-------
10-
w
w
LU
o
o
Reg Class
— HHD8
Urban Bus
2000
2010
Model Year
2020
2030
Figure 4-7. Heavy Duty CNG CO Exhaust Emission Rates for the 0-3 Age Group by Model Year and Regulatory Class
using Nationally Representative Operating Mode Distributions
Figure 4-8 shows consistent decreases in NOx emission rates from older to new model years. The trends in
NOx emissions are consistent with the certification data (Table 4-2) and recent studies.132,133
236
-------
Reg Class
— HHD8
Urban Bus
2000
2010
Model Year
2020
2030
Figure 4-8. Heavy Duty CNG NOx Exhaust Emission Rates for the 0-3 Age Group by Model Year and Regulatory Class
using Nationally Representative Operating Mode Distributions
Figure 4-9 shows significant decreases in the PM2.5 emission rates between the 2001 and 2002 model year
emission rates based on the chassis-cycle average PM2.5 emission rates. An increase in PM2 5 emission rates
is shown from model year 2009 and 2010, which is inconsistent with the certification data (Table 4-2). The
two studies evaluating stoichiometric and lean-burn showed mixed results, with one study showing
stoichiometric engines with TWC emitting lower PM2.5 emissions133 and the other study showing
stoichiometric engines emitting higher PM2.5 rates.132
237
-------
5
-------
1000-
1500
0
(T3
a:
c
o
CO
w
E
LU
CN
O
o
•C 500-
0)
CL
O
Reg Class
— HHD8
Urban Bus
2000
2010
Model Year
2020
2030
Figure 4-10. Heavy Duty CNG C02 Exhaust Emission Rates by Model Year and Regulatory Class using Nationally
Representative Operating Mode Distributions
We have more confidence in the model year 2010 and later emission rates because the emissions rates are
derived directly from in-use second-by-second data. Unfortunately, the differences in the methodology
likely contributed to the differences in emission rates for THC and possibly PM2.s that are not explained by
shift to stoichiometric TWC vehicles for 2010 and later vehicles. Future updates to MOVES could revisit the
pre-2010 CNG exhaust running emission rates to address these inconsitencies.
4.2 Start Exhaust Emission Rates
In the absence of any measured start exhaust emissions from CNG vehicles, their start rates are copied
from the pre-2010 model year heavy-duty diesel start rates for all pollutants including energy rates.
MOVES still estimates that the majority of emissions from CNG vehicles are from running emissions, which
are based on CNG test programs. We acknowledge that the diesel start rates may not accurately represent
CNG start emissions.
4.3 Ammonia Emissions
Similar to carbon monoxide emissions, CNG vehicles with stoichiometric three-way catalysts have been
shown to emit higher ammonia emissions than CNG vehicles with lean-burn combustion.132,133 Since
ammonia measurements were not part of the HDIUT program used to update criteria pollutant and
particulate matter emission rates, the ammonia emission rates for CNG vehicles are simply set equivalent
to the ammonia emission rates from heavy-duty gasoline vehicles.4
239
-------
5 Heavy-Duty Crankcase Exhaust Emissions
Crankcase exhaust emissions, also referred to as crankcase blowby, are combustion gases that pass the
piston rings into the crankcase and are subsequently vented to the atmosphere. Crankcase blowby
includes oil-enriched air from the turbocharger shaft, air compressors, and valve stems that enters the
crankcase. The crankcase blowby contains combustion generated pollutants, as well as oil droplets from
the engine components and engine crankcase.134
5.1 Background on Heavy-Duty Diesel Crankcase Emissions
Federal regulations permit 2006-and-earlier heavy-duty diesel-fueled engines equipped with
"turbochargers, pumps, blowers, or superchargers" to vent crankcase emissions to the atmosphere.135
Crankcase emissions from pre-2007 diesel engines were typically vented to the atmosphere using an open
unfiltered crankcase system, referred to as a 'road draft tube'.134 Researchers have found that crankcase
emissions vented to the atmosphere can be the dominant source of diesel particulate matter
concentrations measured within the vehicle cabin 136 137 13S.
Beginning with 2007-model-year heavy-duty diesel vehicles, federal regulations no longer permit crankcase
emissions to be vented directly to the atmosphere, unless they are included in the certification exhaust
measurements.139 Many manufacturers have adopted open crankcase filtration systems.134These systems
vent the exhaust gases to the atmosphere after the gases have passed a coalescing filter which removes oil
and a substantial fraction of the particles in the crankcase blowby.134 In the ACES Phase 1 program, four MY
2007 diesel engines from major diesel engine manufactures (Caterpillar, Cummins, Detroit Diesel, and
Volvo) all employed filtered crankcase ventilation systems.140
A summary of published estimates of diesel crankcase emissions as percentages of the total emissions
(tailpipe + crankcase) are provided in Table 5-1. For the pre-2007 diesel technologies, hydrocarbon and
particulate matter emissions have the largest contributions from crankcase emissions. There is a
substantial decrease in PM2.5 emissions beginning with the 2007 model year diesel engines. The
aftertreatment technologies required for 2007-and-later engines reduces the tailpipe emissions more than
the crankcase emissions, resulting in an increase in the relative crankcase contribution for THC, CO, and
PM2.5 emissions.
240
-------
Table 5-1 Literature Review on the Contribution of Crankcase Emissions to Diesel Exhaust (Tailpipe + Crankcase)
Study
Model
Year
Tailpipe
Exhaust
Aftertreat
ment
#
Engines
or
Vehicles
THC
CO
NOx
PM
Hare and Baines, 1977143
1966,
1973
None
2
0.2%-
3.9%
0.01%-
0.4%
0.01%-
0.1%
0.9%-
2.8%
Zielinska et al. 2008136
Ireson et al. 2011137
2000,
2003
None
2
13.5% -
41.4%
Clark et al. 2006142
Clark etal. 2006141
2006
None
1
3.6%
1.3%
0.1%
5.9%
Khalek et al. 200948
("ACES Phase 1")
2007
DPF-
equipped
4
84.6%
33.5%
0.007%
44.4%
NVFEL Testing
2015,
2018
SCR-DPF
2
19.0%-
57.8%
14.2%-
76.7%
2.3%-
7.5%
Not
measure
d
Note:
The crankcase ratios shown here are a fraction of tailpipe and crankcase exhaust. The crankcase ratios used in MOVES
are a ratio of crankcase to tailpipe exhaust.
5.2 Modeling Crankcase Emissions in MOVES
MOVES calculates crankcase emissions using a gaseous and a particulate matter crankcase emission
calculators. Within these calculators, crankcase emissions are calculated in relationship to tailpipe exhaust
emissions. In MOVES, the tailpipe exhaust processes are running exhaust, start exhaust, and extended idle
exhaust (processID 1, 2, 90). The corresponding crankcase emission processes are crankcase running
exhaust, crankcase start exhaust, and crankcase extended idle exhaust (processID 15, 16, 17).
The gaseous crankcase calculator chains the crankcase emission rates to the tailpipe exhaust emission
rates, but it does not change the tailpipe exhaust emission rates. Unlike the gaseous calculator, the
particulate matter calculator accounts for tailpipe MOVES exhaust emission rates in the emissionratebyage
table (processID 1, 2, 90) that may include crankcase emissions and has the ability to divide the exhaust
emission rates into components representing the contributions from tailpipe exhaust and crankcase
emissions. Exhaust ratios are used to adjust the particulate matter tailpipe exhaust emission rates to
account for the crankcase contribution. For most model years, the particulate matter tailpipe exhaust
241
-------
emission rates only include tailpipe emissions and the exhaust ratios are set equal to one as shown in the
subsequent sections. However, the 2007-2009 model year tailpipe particulate matter emission rates are
assumed to include crankcase emissions, and the exhaust ratios are less than one to account for the
contribution as discussed in Section 5.3.2. More details on the crankcase calculator provided in the MOVES
Speciation Report.1
The crankcase ratios for non-methane hydrocarbons (NMHC), NOx, and PM2.5 are used to estimate the
crankcase emissions for each subspecies that is chained to their respective primary pollutant as shown in
Table 5-2. The PM10 crankcase emission rates are subsequently estimated from the PM2.5 exhaust and
crankcase emission rates using PM10/PM2.5 emission ratios as documented in the MOVES Speciation
Report.1
Table 5-2 Speciation of Chained Crankcase Pollutants from the Primary Pollutants
Primary Pollutant
Chained Crankcase Pollutant
Non-methane
hydrocarbons
(NMHC)
Benzene, Dibenzo(a,h)anthracene gas, 1,3-Butadiene, Fluoranthene gas,
Formaldehyde, Acenaphthene gas, Acetaldehyde, Acenaphthylene gas, Acrolein,
Anthracene gas, 2,2,4-Trimethylpentane, Benz(a)anthracene gas, Ethyl Benzene,
Benzo(a)pyrene gas, Hexane, Benzo(b)fluoranthene gas, Propionaldehyde,
Benzo(g,h,i)perylene gas, Styrene, Benzo(k)fluoranthene gas, Toluene, Chrysene
gas, Xylene, Fluorene gas, non-methane organic gas (NMOG),
lndeno(l,2,3,c,d)pyrene gas, volatile organic carbon (VOC), Phenanthrene gas,
Naphthalene gas, Pyrene gas
NOx
NO, N02, HONO
PM2.5
EC, S04, H20(aerosol), NonECNonS04PM
By using crankcase to tailpipe emission ratios to estimate crankcase emission rates, MOVES implicitly
assumes that any increase in emissions due to aging and deterioration also occurs for crankcase emissons.
The data sets used to derive the crankcase emission rates for 2007 and later heavy-duty diesel engines are
based on engines and vehicles with relatively low miles and no expected deterioration in the emisson
control system. If crankcase emissions do not exhibit the same increase in deterioration as tailipipe
emissions, this method would lead to an overestimation of crankcase emissions in older vehicles..
5.3 Heavy-Duty Diesel Crankcase Emissions
Different crankcase emission ratios are derived for pre-2007, 2007-2009, and 2010 and later heavy-duty
vehicles as documented in the following sections. The model year distinctions capture large changes in the
242
-------
requirements for crankcase emissions, as well as the control of tailpipe exhaust emissions which impact
the crankcase to tailpipe emission ratios.
As discussed in the following sections, the data on crankcase emissions are limited. For example, the 2010
and later heavy-duty diesel crankcase emission rates are based on tests from two vehicles. As such, the
data have considerable uncertainties. In general, the crankcase emission rates are only a small fraction of
the tailpipe emissions as presented in the following sections, so the uncertainties in crankcase emissions
do not extend to total exhaust emissions. For glider trucks, we had no crankcase emissions data, we
conducted sensitivity testing to evaluate the impact of the assumptions used in MOVES on the heavy-duty
emissions inventory as discussed in Section 5.3.4.
5.3.1 1960-2007 Model Years
Table 5-3 displays the crankcase/tail-pipe emission ratios used for pre-2007 diesel exhaust. For THC, CO,
and NOx, we selected the values measured on the MY2006 diesel engine reported by Clark et al. 2006.142
These values compare well with the previous HC, CO, NOx values reported much earlier by Hare and Baines
(1977),143 which represent much older diesel technology. The similarity of the crankcase emission ratios
across several decades of diesel engines suggests that for pre-2007 diesel engines, crankcase emissions can
be well represented as a fraction of the exhaust emissions. The THC crankcase ratios presented in Table 5-3
are also used for methane, total organic gases (TOG), non-methane hydrocarbons (NMHC) and all the
chained pollutants to NMHC listed in Table 5-2.
For PM2.5 emissions, we use a crankcase/tailpipe ratio of 20 percent. The 20 percent ratio falls within the
range of observations from the literature on diesel PM emissions. Zielinska et al. 2008136 and Ireson et al.
2011137 reported crankcase contributions to total PM2.5 emissions as high as 40 percent. Jaaskelainen
(2012)134 reported that crankcase can contribute as much as 20 percent of the total emissions from a
review of six diesel crankcase studies. Similarly, an industry report estimated that crankcase emissions
contributed 20 percent of total particulate emissions from 1994-2006 diesel engines.144 The crankcase
emission ratios shown Table 5-3 are applied to running, start and extended idle exhaust to estimate the
corresponding crankcase exhaust emissions.
243
-------
Table 5-3 MOVES pre-2007 Diesel Crankcase Ratios for HC, CO, NO*, and PM2.5
Pollutant
Crankcase/Tailpipe
Ratio (MOVES inputs)
Crankcase/(Crankcase + Tailpipe)
Ratio
THC
0.037
0.036
CO
0.013
0.013
NOx
0.001
0.001
PM2.5
0.200
0.167
Note:
MOVES uses a crankcase/tailpipe ratios. We also calculated the crankcase to total exhaust ratio (crankcase + tailpipe)
to compare the MOVES inputs to the values reported in the literature
As outlined in the MOVES Speciation Report, MOVES does not apply the crankcase/tailpipe emission ratio
in Table 6-3 to the total exhaust PM2.5 emissions. MOVES applies crankcase/tailpipe emission ratios to
PM2.5 subspecies: elemental carbon PM2.5, sulfate PM2.5, aerosol water PM2.5, and the remaining PM
(nonECnonS04PM). This allows MOVES to account for important differences in the PM speciation between
tailpipe and crankcase emissions.
The pre-2007 diesel ratios are derived such that the total crankcase PM2.5/tailpipe exhaust PM2.5 ratio is 20
percent, and the crankcase emissions EC/PM fraction reflects measurements from in-use crankcase
emissions. Zielinska et al. 2008136 reported that the EC/PM fraction of crankcase emissions from two pre-
2007 diesel buses is 1.57 percent. Tailpipe exhaust from pre-2007 diesel engines is dominated by
elemental carbon emissions from combustion of the diesel fuel, while crankcase emissions are dominated
by organic carbon emissions largely contributed by the lubricating oil.136 137 The crankcase emission factors
shown in Table 5-4 are derived such that the crankcase PM2.5 emissions are 20 percent of the PM2.5 exhaust
measurements and have an EC/PM split of 1.57 percent. The start and extended idle crankcase ratios are
the same, because the pre-2007 start and extended idle exhaust EC/PM are both 46.4% (as documented in
Section 2.2.2 and Section 2.3.1). The running exhaust EC/PM ratio is 79% (Section 2.1.2.1.8).
244
-------
Table 5-4. MOVES Exhaust and Crankcase Ratios for Pre-2007 Diesel by Pollutant, Process, and Model Year Group
for PM2.5 Species
Pollutant
Process
Start
Running
Extended Idle
EC
1
1
1
nonECnonS04PM
Tailpipe Exhaust
1
1
1
S04
1
1
1
H20
1
1
1
EC
0.007
0.004
0.007
nonECnon SO4PM
Crankcase
0.367
0.937
0.367
S04
0.367
0.937
0.367
H20
0.367
0.937
0.367
5.3.2 2007-2009 Model Years
As discussed in the background section above, the 2007 heavy-duty diesel emission regulations impacted
the technologies used to control exhaust and crankcase emissions. The regulations also expanded the
types of emissions data included in certification tests by including crankcase emissions in the regulatory
standards which previously included only tailpipe emissions. Because heavy-duty diesel engine
manufacturers are using open-filtration crankcase systems, the crankcase emissions are included in the
emission certification results. In MOVES, the base exhaust rates for 2007 to 2009 diesel engines are based
on certification test results for PM2.5 as discussed in Section 5.3.2.
The crankcase ratios for 2007-2009 HDD emissions are based on the ACES Phase 1 study140, which tested
four MY 2007 engines from different manufactures (Caterpillar, Cummins, Detroit Diesel and Volvo). The
ACES Phase 1 engines and exhaust control systems were new, and underwent a 125 hours of degreening
process before the test program. Thus, they represent low-mileage, properly functioning heavy-duty diesel
engine emissions. In reporting the emission rates, the engines were anonymized as A, B, C, D and a backup
engine B' which was tested at a secondary site. The B' backup engine is the same make, model as engine B.
The ACES study conducted samples of the hot FTP cycles that included tailpipe exhaust emissions and
repeat tests that included both the tailpipe exhaust and the crankcase exhaust routed to the sampling
system. The crankcase emission rates for each engine were calculated by taking the difference in the total
mass emitted between hot-FTP cycles with and without crankcase routed into the sampling system and
then dividing by the testing time as shown in Equation 5-1.
245
-------
CCemissions,i = Exhaustwithcc J - Exhaustw/oCC i Equation 5-1
Where:
CCemissions,i= crankcase emissions (grams per hour) for engine i
Exhaustwithcc phot-FTP cycle average emission emissions (grams per hour) for each engine i with
the crankcase routed into the sampling system
Exhaustw/oCC ;= hot-FTP cycle average emission emissions (grams per hour) for each engine i
without the crankcase
Then the crankcase emission rates were averaged together in Equation 5-2 with the backup engine being
treated as additional tests of engine B.
, CCengineB "I" CCengineB'.sitel C-'engineB',site2 , nn , nn ^
l^^engineA "¦ o ¦" ^^engineC ^^engineDy _
CCACESavg = T Equation
5-2
Since ACES Phase 1 hot-FTP contained a hot-start we assume starts are accounted for in the running
values, and thus for these model years, the crankcase ratio for starts is zero. Based upon data from 2010+
HDD vehicles using the same crankcase technology, the crankcase emission rates for extended idling are
similar to the running rates, so we also use the hot-FTP rates from ACES Phase 1 for the crankcase
extending idling rates. The crankcase rates are listed in Table 5-5.
Equation 5-3 is used to calculate TOG crankcase emissions from the methane and NMHC crankcase values,
using the MOVES NMOG/NMHC value for 2007-2009 MY diesel exhaust.1
g g g NMOG
T0G(_) = CH4(_P + NMHC(_p,_
Where each pollutant rate is in g/hr and the ratio of NMOG/NMHC is 1.343 from the hcspeciation table in
MOVES.1
246
-------
Table 5-5 The MY 2007 crankcase exhaust rates (CCACESavg) from ACES Phase 1 FTP cycle used for running and
extended idling (g/hr)
CO
NOx
THC
CH4
NMHC
TOG
Total PM2.5
8.15
0.79
1.76
0.38
1.37
2.23
0.04
The crankcase ratios for the ACES program in Table 5-1 are calculated using Equation 5-4 where the
ExhaustACES is calculated using Equation 5-2, but using Exhaustw/oCC; in place of the crankcase values.
CCACESavg Equation 5-4
ACES ratio rr ,
C^ACESavg + ExhaustACES
We then calculated the crankcase ratios for use in MOVES using the MOVES MY 2007-2009 exhaust base
rates for all diesel HHD vehicles (regClass 47) weighted by the activity in short and long-haul single-unit and
combination trucks (sourcetypes 52,53,61,62) in each operating mode estimated from a MOVES national
scale run. The values for ExhaustMOvES are listed in Table 5-6. The extended idle rates are based on a
single operating mode in MOVES. We used the HHD diesel emission rates to match the HHD engines tested
in the ACES Phase 1 program.
Table 5-6: The MY 2007-2009 MOVES exhaust base rates (g/hr) for running (weighted by operating mode activity)
and extended idling
CO
NOx
THC
CH4
NMHC
TOG
Total PM2.5
Running Exhaust
11.46
288.51
2.93
1.73
1.20
3.34
0.88
Extended Idle Exhaust
39.26
100.45
8.49
5.00
3.49
9.69
0.087
We then used Equation 5-5 to estimate the base crankcase to tailpipe exhaust ratio for 2007-2009 HD
vehicles.
CCACESavg Equation 5-5
^'-'ratio,base — pYu.,11<;t-
nxnaustMOvEs
As mentioned in the background section, many manufactures employ an open crankcase ventilation
system with a coalescing filter, but a substantial fraction opt for a closed crankcase system where the
crankcase vapors are either routed into the engine with the fuel injection or into the exhaust stream
upstream of the aftertreatment. For developing fleet-wide crankcase emissions, we model heavy-duty
247
-------
diesel closed crankcase systems as having zero crankcase emissions, and reduce the base open crankcase
emission ratios to account for the fraction of open crankcase systems in the vehicle fleet.
Available certification data on the prevalence of open and closed crankcase systems are incomplete, but
the available data suggests that most manufacturers use either open or closed systems for all their engines
in a given model year. We assumed that any engine family listed in the certification data with missing
crankcase information was of the same type as all other engine families in that model year for that
manufacturer. This was our best assumption based on the available data, despite finding one conflict.00 We
then used manufacturer production volume data from 2016-2018 (complete data for earlier years was not
available) to weight the number of open and closed systems within vehicle weight categories and across
categories for a heavy-duty diesel total as shown in Table 5-7. In these model years, there were between 9
and 10 engine manufacturers which produced heavy-duty engines, with between 34 and 41 certified
engine families. Using this method, we estimated that 67% of all heavy-duty engines produced in these
model years had open crankcase systems. The average value across model years and regulatory classes
was used for all 2007 and later engines.pp
Table 5-7 Fraction of Engines with Open Crankcase Systems by Vehicle Regulatory Class and Model Year
Model Year
LHD
MHD
HHD
MY Total
2016
0
0.912
0.725
0.696
2017
0
0.919
0.635
0.663
2018
0
0.884
0.640
0.660
average
0
0.905
0.666
0.672
Finally, to estimate the MY 2007-2009 crankcase emission ratios for MOVES, we assume that crankcase
emissions are proportional to the exhaust emissions across regulatory classes and source types. In this
calculation, we must account for the fact that many LHD2b3 trucks are chassis-certified rather than engine-
00 One of the engine manufacturers that we assumed produced closed crankcase systems for all of its 2007 and later
model year engines based on certification data produced one of the open crankcase MY 2007 engine tested in the
ACES Phase 1 program.
pp We recognize the uncertainties in applying open crankcase values from model year 2016-2018 engines to model
year 2007-2009 engines. However, the larger uncertainty in the crankcase emission rates is driven by the measured
crankcase emission rates. Assuming an open crankcase percentage of 67% decreased the crankcase tailpipe emission
rates by roughly one third, which is well within the range of variability of emissions observed in the ACES Phase 1
results.
248
-------
certified, and required to have closed crankcases. To estimate the final crankcase emissions for each
source type, we multiplied the open crankcase ratio, CCratio base, by the open crankcase fraction of all
heavy-duty regulatory classes, OpenCCfrac, and by the fraction of engine-certified vehicles,
engine_certratio Sourcetype modelyear as shown in Equation 5-6.
CCratio,Sourcetype,modelyear
CCratio,base * C)p6nCCfrac Equation 5-6
* sourcetype,modelyear)
Where
CCratio,sourcetype,model year= the crankcase ratio used in MOVES by source type and model year
CCratio,base= the heavy-duty open crankcase ratio calculated from Equation 5-5
OpenCCfrac= the fraction of open crankcase systems determined from certification data and
manufacture production volume as shown in Table 5-7 =0.672
engine_certratio Sourcetype modelyear= the fraction of engine-certified vehicles within each source
type and model year, as explained below.
The MOVES 2007-2009 crankcase ratios for THC, CH4, NMHC, TOG, NOx, CO, and PM2.5 for each process for
heavy-duty source types with 100 percent engine-certified vehicles are shown in Table 5-8.
249
-------
Table 5-8 MOVES 2007 to 2009 Diesel Crankcase/Tailpipe Ratios for the Heavy-duty Source Types
(41,42,43,51,54,61,62) except the short and long-haul single-unit trucks (source types 52,53)
Process
Pollutant
Crankcase/Tailpipe ratio
Baseline
values
(Equation
5-5)
Final values to
account for closed
crankcase
(Equation 5-6)
Crankcase
Running
Exhaust
THC
0.60
0.40
NMHC
1.14
0.77
ch4
0.22
0.15
TOG
0.67
0.45
CO
0.71
0.48
NOx
0.003
0.002
PM2.5
0.04
0.03
Crankcase
Start
Exhaust
THC
0.00
0.00
NMHC
0.00
0.00
ch4
0.00
0.00
TOG
0.00
0.00
CO
0.00
0.00
NOx
0.00
0.00
PM2.5
0.00
0.00
Crankcase
Extended
Idle
Exhaust
THC
0.21
0.14
NMHC
0.39
0.26
ch4
0.08
0.05
TOG
0.23
0.15
CO
0.21
0.14
NOx
0.008
0.005
PM2.5
0.44
0.29
250
-------
For PM2.5 emissions, MOVES applies crankcase ratios to each of the intermediate PM2.5 species (EC,
nonECnonS04PM, S04, and H20). The MOVES PM2.5 speciation profile developed from the ACES Phase 1
study combined the crankcase and tailpipe emissions. As such, MOVES uses the same speciation profile for
both crankcase and tailpipe emissions, and all intermediate PM2.5 species. The resulting crankcase emission
ratios for MY 2007 to 2009 heavy-duty diesel by intermediate PM2.5 species are provided in Table 5-9 and
the tailpipe ratios are calculated as one minus the crankcase ratio. As explained in Section 5.2, the exhaust
crankcase emission factor is less than one for 2007-2009 diesel vehicles to account for the contribution of
crankcase emissions in the base exhaust emission rates.
Table 5-9. MOVES Exhaust and Crankcase Ratios for Model Year 2007-2009 Diesel by Pollutant and Process for
PM2.5 Species for Heavy-duty Source Types (41,42,43,51,54,61,62) except the Short and Long-haul Single-unit Trucks
(source type 52,53)
Process
Model Year Group
Pollutant
Start
Running
Extended Idle
Tailpipe
Exhaust
2007-2009
EC
1
0.971
0.7071
nonECnonS04PM
1
0.971
0.7071
S04
1
0.971
0.7071
H20
1
0.971
0.7071
Crankcase
2007-2009
EC
0
0.0290
0.2929
nonECnon SO4PM
0
0.0290
0.2929
S04
0
0.0290
0.2929
H20
0
0.0290
0.2929
The fractions of engine-certified vehicles in single-unit short- and long-haul trucks (sourceType 52 and 53)
are calculated using Equation 5-7. We assumed that 5.1 percent of the diesel vehicles in LHD2b3 are
engine-certified based on the national vehicle registration data and the vehicle activity data. The
remainder of the LHD2b3 are chassis-certified vehicles which must comply with light-duty standards
requiring closed crankcase systems. The other heavy-duty regulatory classes contain only engine-certified
vehicles. The MOVES sampleVehiclePopulation table is used to estimate the fraction of LHD2b3 vehicles in
each source type for each model year. The fraction of LHD2b3 in diesel-fueled short-haul and long-haul
251
-------
combination trucks (sourceType 52 and 53) by model year are shown in Table 5-10. None of the other
heavy-dutyqq source types include LHD2b3 vehicles for 2007 and later model years.
engine_certratio Sourcetype modelyear
r / n-j Equation 5~/
LHD2b3sourcetType,model year ^ (.1 6fi§ifi6_C6rtratj0 LHDD J J
Where
engine_certratioSourcetypemodelyear= the fraction of engine-certified vehicles within sourcetype 52 and
53 by model year (Column C in Table 5-10)
LHD2b3sourcetype modei ygar= the fraction of LHD2b3 diesel vehicles within each source type and
model year group of diesel vehicles (Column A in Table 5-10)
engine_certratio LHDD= the fraction of engine-certified vehicles within light-heavy-duty diesel
(LHDD) vehicles (Column B in Table 5-10)
qq Note that two of the light-duty source types (passenger truck and light-commercial trucks) contain a small fraction
of LHD2b3 vehicle, but these vehicles are all expected to have closed crankcase systems as documented in the light-
duty exhaust report8
252
-------
Table 5-10 Calculation of the Engine Certification Ratio for Diesel Short- and Long-haul Combination trucks
(sourcetype 52 and 53) for model years 2007 through 2015+
(A)
(B)
(C)
Model Year
Fraction of LHD2b3
Engine cert ratio within LHD2b3
Engine cert ratio
2007
33.3%
5.1%
68.4%
2008
47.3%
5.1%
55.2%
2009
32.9%
5.1%
68.8%
2010
39.8%
5.1%
62.3%
2011
55.2%
5.1%
47.6%
2012
50.5%
5.1%
52.1%
2013
42.3%
5.1%
59.9%
2014
35.8%
5.1%
66.1%
2015+
35.8%
5.1%
66.1%
Using Equation 5-6, we multiply the engine cert ratios shown in Table 5-10 by the crankcase ratios shown
in Table 5-8 and Table 5-9. This reduces the crankcase emission ratios for source type 52 and 53 by
approximately one-third to one-half depending on the model year.
5.3.3 2010-2060 Model Years
For 2010-and-later diesel vehicles, the HDIUT program (see Section 2.1.1.1) is used for the baseline exhaust
emission rates. As the HDIUT program measures tailpipe exhaust emissions from trucks in-use, the
crankcase is not routed to the tailpipe and therefore not accounted for in the data. To account for
crankcase emissions for 2010 and later diesel vehicles, we used direct crankcase measurements of NOx,
THC, CH4, and CO emissions from the US EPA's National Vehicle and Fuel Emissions Laboratory (NFVEL)
testing of two heavy duty trucks (MY 2015 and MY 2019). Each of the trucks had less than 10,000 miles and
represent properly functioning low-mileage heavy-duty vehicles. Testing was conducted on a chassis
dynamometer over a drive cycle that consists of an ARB transient (Phase 1), followed by four repetitions of
the same ARB transient (Phase 2), 10 minutes of idling (Phase 3), and steady-state highway activity at 55
mph and 60 mph (Phase 4). The speed trace and a graphical indication of the testing phases are shown in
Figure 5-1.
253
-------
Figure 5-1 Speed trace of the NVFEL testing program along with each phase
There were a total of thirteen repetitions between the two trucks, with soak times between repetitions of
either between one and three hours (hot start) or greater than twelve hours (cold start) as shown in Table
5-11.
Table 5-11: Testing information for the NVFEL test program
Truck ID
Number of Tests
Dates of Testing
Soak Times
1
6 (3 Hot Start, 3 Cold Start)
May 30, 2017-
June 1, 2017
Hot Starts 1-3 hrs
Cold Starts 12+ hours
2
7 (3 Hot Start, 4 Cold Start)
August 20, 2019
- August 13,
2019
Hot Starts 1-3 hrs
Cold Starts 12+ hours
Table 5-12 reports the averages of the repetitions by truck and phase, and by phase only (by weighting the
two trucks equally).
254
-------
Table 5-12 NVFEL Vehicle average emission rates by testing phase and truck
Crankcase CO
(g/hr)
Crankcase NOx
(g/hr)
Crankcase THC
(g/hr)
Crankcase CH4
(g/hr)
Phase 1 Average
1.0900
0.4427
0.3463
0.0324
Truck 1 average
0.5057
0.5527
0.2373
0.0053
Truck 2 average
1.6742
0.3327
0.4552
0.0595
Phase 2 Average
0.6224
0.9086
0.4537
0.0155
Truck 1 average
0.1916
1.0316
0.2835
0.0001
Truck 2 average
1.0532
0.7855
0.6238
0.0309
Phase 3 Average
0.2461
0.7097
0.4529
0.0058
Truck 1 average
0.1670
0.7633
0.2634
0.0000
Truck 2 average
0.3252
0.6560
0.6424
0.0115
Phase 4 Average
0.2679
1.8447
0.5915
0.0053
Truck 1 average
0.1521
1.7766
0.3711
0.0000
Truck 2 average
0.3836
1.9127
0.8118
0.0106
Average of all phase averages
0.5677
0.9764
0.4611
0.0148
The average running rate was calculated by calculating a weighted average between Phase 2 (36.7%) and
Phase 4 (63.3%). The phase-weighting was calculated to match the national running operating mode
distribution for MY2015, HHD diesel vehicles in short- and long-haul single-unit and combination trucks
(source type 52,53,61,62) as estimated in a MOVES national run using a draft version of MOVES3 used for
the preliminary CTI analysis for calendar year 2015.15 Phase 2 had transient operation below 50 mph and
was matched to operating modes 0-30 (36.7% of total activity) and Phase 4 had constant high speed data
and was matched to operating-modes 33-40 (63.3% of total activity) as shown in Table 5-13. HHD diesel
vehicles in short- and long-haul single-unit and combination trucks vehicles were chosen as to match the
NVFEL testing vehicles.
255
-------
Table 5-13. Assignment of MOVES Operating Modes to Test Phase from the NVFEL data, and Phase weighting
determined from the Operating Mode activity from a National draft MOVES run for MY 2015 HHD Vehicles in the
Short- and Long-haul Single-unit and Combination Trucks Source Types
OpModelD
Operating Mode
Scaled Tractive
Power (STPt,
Vehicle
Speed
% of Total
Activity from
NVEL Test
Phase
Description
skW)
(iitl mph)
MOVES run
256
-------
0
Deceleration/Braking
2.69
1
Idle
vt < 1.0
5.40
11
Coast
STPt< 0
1 < vt < 25
4.55
12
Cruise/Acceleration
0 < STPt< 3
1 < vt < 25
6.62
13
Cruise/Acceleration
3 < STPt< 6
1 < vt < 25
1.64
14
Cruise/Acceleration
6 < STPt< 9
1 < vt < 25
1.00
15
Cruise/Acceleration
9 < STPt< 12
1 < vt < 25
0.77
16
Cruise/Acceleration
12 < STPt
1 < vt < 25
1.36
21
Coast
STPt< 0
25 < vt < 50
3.37
Phase 2
22
Cruise/Acceleration
0 < STPt< 3
25 < vt < 50
1.34
23
Cruise/Acceleration
3 < STPt< 6
25 < vt < 50
1.54
24
Cruise/Acceleration
6 < STPt< 9
25 < vt < 50
1.52
25
Cruise/Acceleration
9 < STPt< 12
25 < vt < 50
1.04
27
Cruise/Acceleration
12 < STPt< 18
25 < vt < 50
1.86
28
Cruise/Acceleration
18 < STPt< 24
25 < vt < 50
0.98
29
Cruise/Acceleration
24 < STPt< 30
25 < vt < 50
0.53
30
Cruise/Acceleration
30 < STPt
25 < vt < 50
0.49
33
Cruise/Acceleration
STPt< 6
50 < vt
16.26
35
Cruise/Acceleration
6 < STPt< 12
50 < vt
14.90
37
Cruise/Acceleration
12 < STPt<18
50 < vt
13.61
Phase 4
38
Cruise/Acceleration
18 < STPt< 24
50 < vt
8.60
39
Cruise/Acceleration
24 < STPt< 30
50 < vt
5.05
40
Cruise/Acceleration
30 < STPt
50 < vt
4.89
The average start rate was calculated by taking the difference between Phase 1 and Phase 2 of each test
and averaging the differences together as shown in Equation 5-8rr.
257
-------
Y"-,fCC u 1 ¦ — CC u , ) Equation 5-8
Zjj = i^v-,v-,phase l,j ^^phase 2,)J ^
^'-'start ~
Where
CCstart= average crankcase (g/start) emission from all the test runs
CCphase i,j = crankcase (g/hr) emission rate from test run j and Phase 1
n= the number of tests (13)
In the case of THC and NOx, Equation 5-8 yielded negative start emission rates, because the average Phase
2 (g/hr) emission rates for these pollutants are higher than the Phase 1 (g/hr) emission rates. There are
physical explanations that can lead to an observed increase in crankcase emission rates in Phase 2.
Lubricating oil and diesel fuel in the crankcase can evaporate as the engine block heats during operation,
which is measured as THC and PM2.5 if the vapors condense upon dilution. For NOx emissions, the engine-
out emissions during the start period can be lower than running period, due to lower air-fuel ratios and
lower in-cylinder temperatures. In fact, MOVES models zero NOx tailpipe starts for pre-2010 heavy-duty
diesel trucks based on observations of negative start emission rates as discussed in Section 2.2.1.2.
Similarly, we set the start crankcase emission rate to zero for THC and NOx. The average CH4 start emission
rate was also positive, but because methane is estimated in MOVES as a fraction of THC, we also set it
equal to zero as shown in Table 5-14.
For the crankcase extended idling rate we used the average of emissions in the idling phase (Phase 3).
These rates also informed the base crankcase rates for NMHC and TOG. We used Equation 5-3 to calculate
TOG crankcase emissions, with the MOVES NMOG/NMHC ratio for 2010+ diesel exhaust (1.085) from the
hcspeciation table in MOVES.1
The crankcase rates are ratioed to the tailpipe emissions from the HDIUT vehicles from the model year
2010-2013 and 2014+ NOx FEL 0.2 groups (See Table 2-7) for each pollutant. These vehicles were chosen as
rr By using g/hr rates, we implicitly assume that the Phase 1 emission rates extend for 1 hour when calculating the
starts, rather than the 11 minutes of Phase 1 (Equation 2-22 show the correct way of calculating starts). This
overestimates the cold starts by roughly a factor of 5 (60/11). This error was overlooked in the peer-review and
noticed after the database was finalized for MOVES3. As such, it will be addressed in a future version of MOVES.
Because heavy-duty crankcase emissions make a very small contribution to CO emissions, the impact on the total
emissions inventory is negligible (< 3% for heavy-duty diesel emissions and < 0.2% of all onroad CO emissions).
258
-------
they comprise the majority of the fleet in most model years (See Figure 2-12). The HDIUT data are used
instead of the tailpipe values from the NVFEL test trucks because 1) they are the basis of the MOVES
tailpipe emission rates, 2) we have much more confidence in the mean HDIUT tailpipe emission rates than
the mean tailpipe emission rates from the two NVEL test-trucks to represent fleet-average rates, and 3) the
variability of the tailpipe measurements from the NVFEL vehicles is generally much greater than the
variability of the crankcase measurements, including for THC emissions.145 Thus, by using tailpe emissions
from MOVES, we yield more stable crankcase ratios.
The ratios for NMHC, TOG, and NOx are applied to the pollutants chained to them as well (Table 5-2).
Table 5-14 Average Crankcase Emission Rates for MOVES Processes from NVFEL data
Process
CO
NOx
THC
ch4
NMHC
TOG
Crankcase Running
Exhaust (g/hr)
0.398
1.501
0.541
0.009
0.5318
0.5861
Crankcase Start
Exhaust (g/start)
0.268
0
0
0
0
0
Crankcase Extended
Idling Exhaust (g/hr)
0.246
0.710
0.453
0.006
0.4471
0.4909
No PM2.5 crankcase emission measurements were made in the NVFEL study, so the 2010 and later model
year calculations use the crankcase PM2.5 rates from ACES Phase 1 report as described for model years
2007-2009 in Section 5.3.2. More information on the PM calculations for model year 2010 and later
vehicles is provided later in this section.
We used the base exhaust emission rates from the NOx FEL 0.2 group for HHD engines in short and long-
haul single-unit and combination trucks (source types 52,53,61,62). These vehicles were chosen from the
full HDIUT data set to match the NVFEL testing vehicles.
259
-------
Table 5-15 The N0X FEL 0.2 MOVES Exhaust Base Rates for Running (Weighted by Operating Mode Activity) and
Extended Idling (g/hr)
Model Year
Group
CO
NOx
THC
ch4
NMHC
TOG
Total PM2.5
Running Exhaust
2010-2013
125.42
62.51
1.50
0.57
0.93
1.58
0.15
2014+
60.06
61.87
1.47
0.56
0.91
1.55
0.12
Extended Idle Exhaust
2010-2013
39.26
42.60
2.75
1.04
1.70
2.89
0.03
2014+
39.26
42.60
1.64
0.62
1.01
1.72
0.02
As was done for the development of the 2007-2009 crankcase emission ratios documented in Section
5.3.2, we assume that the crankcase ratio derived for 2010+ heavy heavy-duty diesel (HHD) engines applies
to the other heavy-duty regulatory classes (MHD, LHD) and source types (other buses, transit buses, school
buses, refuse trucks, and motorhomes), and that crankcase emissions are proportional to the exhaust
emissions across regulatory classes and source use types. Additionally, we use Equation 5-6 to account for
the fraction of open crankcase systems within each source type, with the fraction of open crankcase
systems assumed to be 67.2%. The fraction of engine-certified vehicles among short and long-haul single-
unit trucks (source type 52 and 53) varies by model year, see Table 5-10.
The HHD tailpipe rates have two model year groups 2010-2013 and 2014+, so the ratios differ for these
model year groups as shown in Table 5-16. For long-haul and short-haul single unit trucks (sourceType 52
and 53) these values are multiplied by the fraction of engine certified vehicles in each model year,
provided in Table 5-10.
260
-------
Table 5-16 MOVES 2010+ Diesel Crankcase/Tailpipe Ratios for Heavy-Duty Sourcetype (41,42,43,51,54,61,62)
Except the Short- and Long-haul Single-unit Trucks (Sourcetype 52,53)
261
-------
Crankcase/Tailpipe ratio
Process
Pollutant
Model Year
Group
Baseline
values
(Equation 5-5)
Final values to account for closed crankcase
(Equation 5-6)
THC
2010-2013
0.36
0.24
2014+
0.37
0.25
NMHC
2010-2013
0.57
0.39
2014+
0.58
0.39
ch4
2010-2013
0.016
0.011
2014+
0.016
0.011
Running
TOG
2010-2013
0.37
0.25
2014+
0.38
0.25
CO
2010-2013
0.003
0.002
2014+
0.007
0.005
NOx
2010-2013
0.024
0.016
2014+
0.024
0.016
PM2.s
2010-2013
0.26
0.18
2014+
0.32
0.22
THC
2010-2013
0
0
2014+
0
0
NMHC
2010-2013
0
0
2014+
0
0
Starts
ch4
2010-2013
0
0
2014+
0
0
TOG
2010-2013
0
0
2014+
0
0
CO
2010-2013
0.89
0.60
2014+
0.89
0.60
262
-------
NOx
2010-2013
0
0
2014+
0
0
PM2.5
2010-2013
0
0
2014+
0
0
Extended
Idle
THC
2010-2013
0.17
0.11
2014+
0.28
0.19
NMHC
2010-2013
0.26
0.18
2014+
0.44
0.30
ch4
2010-2013
0.006
0.004
2014+
0.009
0.006
TOG
2010-2013
0.17
0.11
2014+
0.28
0.19
CO
2010-2013
0.006
0.004
2014+
0.006
0.004
NOx
2010-2013
0.017
0.011
2014+
0.017
0.011
PM2.5
2010-2013
1.11
0.75
2014+
1.82
1.22
s noted above, the PM2.5 crankcase emission ratios are ca
culated using the ACES Phase 1 crankcase
emission rates (MY 2007) and the PM2.5 exhaust rates from the 2010-2013 and 2014+ NOx FEL 0.2 groups.
As the tailpipe PM2.5 exhaust emission rates are based upon in-use tailpipe testing for 2010+, the tailpipe
ratio for all processes is 1, and the crankcase PM2.5 ratio is a simple fraction of the tailpipe emissions as
shown in Table 5-17. Since PM2.5 uses ACES Phase 1 data, starts are assumed to be included in the running
and the crankcase starts are set to zero.
263
-------
Table 5-17. MOVES Exhaust and Crankcase Ratios for 2010+ Diesel by Pollutant, Process, and Model Year Group for
PM2.5 Species for the Heavy-duty Source types (41,42,43,51,54, 61,62)
Except the Short- and Long-haul Single-unit Trucks (Source type 52,53)
Process
Model Year Group
Pollutant
Start
Running
Extended Idle
EC
1
1
1
2010-2013
nonECnonS04PM
1
1
1
S04
1
1
1
Tailpipe
H20
1
1
1
Exhaust
EC
1
1
1
2014+
nonECnonS04PM
1
1
1
S04
1
1
1
H20
1
1
1
EC
0
0.1761
0.7465
2010-2013
nonECnonS04PM
0
0.1761
0.7465
S04
0
0.1761
0.7465
Crankcase
H20
0
0.1761
0.7465
EC
0
0.2168
1.2203
2014+
nonECnonS04PM
0
0.2168
1.2203
S04
0
0.2168
1.2203
H20
0
0.2168
1.2203
5.3.4 Glider Crankcase Emissions
Glider trucks in MOVES are anticipated to have emissions equivalent to those from MY 2000 HD diesel
engines (Section 2.5). As such, the crankcase fractions derived for pre-2007 vehicles should also be used to
estimate crankcase emissions from vehicles in the glider regulatory class. However, MOVES models
crankcase emissions by sourcetype (e.g., long-haul combination truck) and model year, rather than by
regulatory class (e.g., Glider, HHD). As such, glider vehicles use the same crankcase ratios as the other HHD
and MHD vehicles in the long-haul and short-haul combination truck sourcetypes and model year groups.
In Appendix L, we evaluated the impact of accounting for gliders in the crankcase ratios for several
calendar years (2015, 2020, 2030, and 2045) using current estimates of glider populations9. The impact on
264
-------
crankcase emissions from gliders is significant for THC, CO, and NOx emissions. However, looking at the
broader inventory, the difference for total emissions of heavy-duty diesel vehicles for CO, NOx and Total
PM2.5 is less than a 0.2%, and for THC, the difference is less than 3%.
By not explicitly accounting for glider crankcase ratios, we are unable to accurately speciate the PM from
glider crankcase emissions (i.e., elemental carbon, organic carbon, and other species). This causes the
EC/PM fraction from crankcase emissions for glider trucks in MOVES3 to be dominated by EC emissions,
whereas the EC/PM fraction of crankcase emissions from pre-2007 HD trucks is anticipated to be less than
1%. When aggregated to all HD regulatory classes and exhaust emission processes, this causes the EC/PM
fraction to be slightly higher in MOVES3, by 1% in 2015 to 6% in 2045. See more details in Appendix L. We
anticipate using this information as we consider revisions to the crankcase calculations in future versions of
MOVES.
5.4 Heavy-Duty Gasoline and CNG Crankcase Emissions
The data on heavy-duty gasoline and CNG crankcase emissions are limited. All 1969 and later spark ignition
heavy-duty engines are required to control crankcase emissions. All gasoline engines are assumed to use
positive crankcase ventilation (PCV) systems, which route the crankcase gases into the intake manifold. For
heavy-duty gasoline engines we use the same values of crankcase emission ratios as light-duty gasoline as
shown in Table 5-18; these are documented in the MOVES light-duty emission rates report.8
For the 1969 and later vehicles, we assume 4 percent of PCV systems fail, which would cause increased
tailpipe emissions for reasons such as misfiring of the engine, lubricating oil in the intake manifold, and
increased deterioration of the three-way catalyst. Although these processes will increase tailpipe
emissions, in MOVES we model this increase of emissions due to a failed PCV as crankcase emissions. We
assume that the elevated emissions due to a failed PCV system would be equivalent to the crankcase
emissions of a pre-1969 vehicles using Equation 5-9. The resulting fleet-wide crankcase to exhaust
emission ratios for 1969 and later vehicles are shown in Table 5-18. Due to limited information, we used
the gasoline heavy-duty crankcase emission factors for heavy-duty CNG engines because the majority of
these engines are spark-ignited.
Crankcase Ratio1969+ = PCV failure rate (4%) x Crankcase Ratioprel969 Equation 5-9
265
-------
Table 5-18 Crankcase to Tailpipe Exhaust Emission Ratio for Heavy-Duty Gasoline and CNG Vehicles for HC, CO,
NOx, and PM2.5
Pollutant
pre-1969
1969 and later
HC
0.33
0.013
CO
0.013
0.00052
NOx
0.001
0.00004
PM (all species)
0.20
0.008
The crankcase and exhaust ratios used by the crankcase calculator for PM2.5 emissions from heavy-duty
gasoline and compressed natural gas vehicles are provided in Table 5-19. No information is available to
estimate separate speciation between exhaust and crankcase, so the factors are the same for all PM
subspecies.
Table 5-19 MOVES Exhaust and Crankcase Ratios for Heavy-Duty Gasoline and CNG Vehicles by Pollutant, Process,
Model Year Group, and Fuel Type, and Source Type for PM2.5 Species
Pollutant
Process
1960-1968
1969-2050
EC
Exhaust
1
1
nonECnonS04PM
1
1
S04
1
1
H20
1
1
EC
Crankcase
0.2
0.008
nonECnonS04PM
0.2
0.008
S04
0.2
0.008
H20
0.2
0.008
266
-------
6 Nitrogen Oxide Composition
This section discusses the values used to estimate nitric oxide (NO), nitrogen dioxide (N02) and nitrous acid
(HONO) from nitrogen oxide (NOx) emissions from heavy-duty vehicles. A similar section on NOx
composition from light-duty emissions is included in the light-duty emissions report. NOx emissions are
reported in mass-equivalent space of N02. In other words, the molar mass of N02 (46 g/mole) is used to
calculate grams of NOx from the molar concentration of NOx.
Nitrogen oxides (NOx) are defined as NO + N02.146,147 NOx is considered a subset of reactive nitrogen
species (NOy) with a nitrogen oxidation state of +2 or greater which contain other nitrogen containing
species (NOz), thus NOy= NOx + NOz.146 NOz compounds are formed in the atmosphere as oxidation
products of NOx.147
Chemiluminescent analyzers used for exhaust NOx measurements directly measure NO, as NO is oxidized
by ozone to form N02 and produces florescent light. Chemiluminescent analyzers measure NOx (NO + N02)
by using a catalyst that reduces the N02to NO in the sample air stream before measurement. N02 is
calculated as the difference between NOx and NO measurements. The NOx converter within
chemiluminescent analyzers can also reduce other reactive nitrogen species (NOz), including HONO to NO.
If the concentrations of NOz interfering species in the sample stream are significant relative to N02
concentrations, then they can bias the N02 measurements high.148
MOVES produces estimates of NO and N02 by applying an NO/NOx or N02/N0X fraction to the NOx
emission rates. The N0/N02and N02/N0X fractions are stored in a MOVES table called nono2ratio. The
nono2ratio enables the nitrogen oxide composition to vary according to source type, fuel type, model year,
and pollutant process. However, for the heavy-duty vehicle source types, the NOx fractions vary only
according to fuel type, model year, and emission process. The NOx fractions in MOVES were developed
from a literature review from emission test programs conducted in the laboratory with constant volume
sampling dilution tunnels.4 As for crankcase emissions, we will consider updating the MOVES structure to
use separate N0/N02 and N02/N0X fractions by regulatory class to accommodate different emissions from
gliders from other heavy-duty regulatory classes.
MOVES also produces estimates of one important NOz species, nitrous acid (HONO), from the NOx values.
HONO emissions are estimated as a fraction (0.8 percent) of NOx emissions from all vehicle types in
MOVES, based on HONO and NOx measurements made at a road tunnel in Europe.149 HONO emissions are
also estimated using the nono2ratio MOVES table. For each source type, fuel type, and emission process,
the NO, N02, and HONO values in the nono2ratio sum to unity.
MOVES users should be aware that the definition of NOx in MOVES (N0+N02+H0N0) is different than the
standard NOx definition of NOx (NO + N02). In MOVES, we include HONO in the NOx values, because the
267
-------
chemiluminescent analyzers are biased slightly high by HONO in the exhaust stream, and HONO is formed
almost immediately upon dilution into the roadway environment from N02 emissions. To avoid
overcounting reactive nitrogen formation, we include HONO in the sum of NOx in MOVES. MOVES users
should consider which measure they would like to use depending on their use-case. For example, for
comparing NOx results with a vehicle emission test program, MOVES users may want to simply use NOx
(pollutantID 3), whereas MOVES users developing air quality inputs of NO, N02, and HONO, should
estimate NOx as the sum of NO + N02 (pollutantlDs 32 and 33), rather than using the direct NOx output in
MOVES (polluantID 3).
Future work is needed to (1) update MOVES to model NOx and HONO fractions according to regulatory
class, (2) update the NOx and HONO fractions in MOVES based on more recent measurements, (3)
reconcile the definition of NOx in MOVES, while also correctly accounting for the emissions of NOz species
that may impact NOx measurements and (4) reconcile measurement differences that may occur between
NOy species measured at the tailpipe, with NOy species measured on road side measurements.150
6.1 Heavy-Duty Diesel
Table 6-1 shows the NOx and HONO fractions for heavy-duty diesel vehicles. The 1960-2006 model year
heavy-duty diesel NOx fractions were estimated as the average reported fraction from three studies of
heavy-duty vehicles not equipped with diesel particulate filters.4 The 2010+ N02fractions are based on the
average from three diesel test programs of diesel vehicles measured with diesel particulate filters. The
2007-2009 values are an average of the 1960-2006 and 2010-2050 values, based on the assumption that
the NOx fractions changed incrementally, as trucks equipped with catalyzed diesel particulate filters were
phased-into the fleet. We intend to incorporate data on NOx fractions from SCR-equipped diesel trucks in
the next version of MOVES.
The NOx fractions are the same across all diesel source types (including light-duty) and across all emission
processes (running, start, extended idle), except for auxiliary power units, which use the conventional NOx
fractions (1960-2006) for all model years because it is assumed that the APUs are not fitted with diesel
particulate filters. The N02fractions originally developed from a MOVES2010 Report4 were reduced by
0.008 to account for the HONO emissions. Because the nono2ratio table is classified by source type, and
not regulatory class, gliders use the same NO/N02 fractions as the other regulatory classes by model year,
even though the pre-2007 NOx fractions are more relevant for this regulatory class.
268
-------
Table 6-1 NOx and HONO Fractions for Heavy-Duty Diesel Vehicles
Model Year
NO
no2
HONO
1960-2006a
0.935
0.057
0.008
2007-2009
0.764
0.228
0.008
2010-2060
0.594
0.398
0.008
Note:
a All Model Year of Auxiliary Power Units (APUs) use the 1960-2006 NOx and HONO fractions
6.2 Heavy-Duty Gasoline
The NOx fractions for heavy-duty gasoline are based on the MOVES values used for light-duty gasoline
measurements. Separate values are used for running and start emission processes. As stated in the
MOVES2010 report4 the values are shifted to later model year groups to be consistent with emission
standards and emission control technologies. These values are shown in Table 6-2 for both light-duty and
heavy-duty gasoline vehicles. The N02 fractions originally developed from the MOVES2010 report4 were
reduced by 0.008 to account for the HONO emissions.
Table 6-2 NO* and HONO Fractions for Light-Duty (Source Type 21, 31, 32) and Heavy-Duty Gasoline Vehicles
(Source Type 41, 42, 43, 51, 52, 53, 54, 61)
Light-Duty
gasoline
model year
groups
Heavy-Duty
gasoline
model year
groups
Running
Start
NO
no2
HONO
NO
no2
HONO
1960-1980
1960-1987
0.975
0.017
0.008
0.975
0.017
0.008
1981-1990
1988-2004
0.932
0.06
0.008
0.932
0.031
0.008
1991-1995
2005-2007
0.954
0.038
0.008
0.987
0.005
0.008
1996-2060
2008-2060
0.836
0.156
0.008
0.951
0.041
0.008
6.3 Compressed Natural Gas
We used the average of three N02/ NOx fraction reported on three CNG transit buses with DDC Series 50 G
engines by Lanni et al. (2003)123 with the 0.008 HONO fraction assumed for other fuel types, to estimate
the NOx fractions of NO, N02, and the HONO fraction. These assumptions yield the NOx and HONO
fractions in Table 6-3, which are used for all model year CNG heavy-duty vehicles.
269
-------
Table 6-3 NOxand HONO Fractions CNG Heavy-Duty Vehicles
Model Year
NO
no2
HONO
1960-2060
0.865
0.127
0.008
270
-------
7 Appendices
271
-------
Appendix A Calculation of Accessory Power Requirements
Table A-l. Accessory Load Estimates for HHD Trucks
VSP
Cooling Fan
Air cond
Air comp
Alternator
Engine
Accessories
Total Accessory Load (kW)
Low
Power (kw)
% time on
19.0
10%
2.3
50%
Off = 0.5 kW
3.0
60%
1.5
100%
1.5
100%
Total (kW)
1.9
1.2
2.0
1.5
1.5
8.1
Mid
Power (kw)
% time on
19.0
20%
2.3
50%
Off = 0.5 kW
2.3
20%
1.5
100%
1.5
100%
Total (kW)
3.8
1.2
0.9
1.5
1.5
CO
CO
High
Power (kw)
% time on
19.0
30%
2.3
50%
Off = 0.5 kW
2.3
10%
1.5
100%
1.5
100%
Total (kW)
5.7
1.2
0.7
1.5
1.5
10.5
Table A-2. Accessory Load Estimates for MHD Trucks
VSP
Cooling Fan
Air cond
Air comp
Alternator
Engine
Accessories
Total Accessory Load (kW)
Low
Power (kw)
% time on
10.0
10%
2.3
50%
Off = 0.5 kW
2.0
60%
1.5
100%
1.5
100%
Total (kW)
1.0
1.2
1.4
1.5
1.5
6.6
Mid
Power (kw)
% time on
10.0
20%
2.3
50%
Off = 0.5 kW
2.0
20%
1.5
100%
1.5
100%
Total (kW)
2.0
1.2
0.8
1.5
1.5
7.0
High
Power (kw)
% time on
10.0
30%
2.3
50%
Off = 0.5 kW
2.0
10%
1.5
100%
1.5
100%
Total (kW)
3.0
1.2
0.7
1.5
1.5
7.8
272
-------
Table A-3. Accessory Load Estimates for Buses
VSP
Cooling Fan
Air cond
Air comp
Alternator
Engine
Accessories
Total Accessory Load (kW)
Low
Power (kw)
% time on
19.0
10%
18.0
80%
Off = 0.5 kW
4.0
60%
1.5
100%
1.5
100%
Total (kW)
1.9
14.4
2.6
1.5
1.5
21.9
Mid
Power (kw)
% time on
19.0
20%
18.0
80%
Off = 0.5 kW
4.0
20%
1.5
100%
1.5
100%
Total (kW)
3.8
14.4
1.2
1.5
1.5
22.4
High
Power (kw)
% time on
19.0
30%
18.0
80%
Off = 0.5 kW
4.0
10%
1.5
100%
1.5
100%
Total (kW)
5.7
14.4
0.9
1.5
1.5
24.0
Appendix B Tampering and Mal-maintenance for Diesel Running
Exhaust
Tampering and mal-maintenance (T&M) effects represent the fleet-wide average increase in emissions as
the fleet ages. In laboratory testing, properly maintained engines often yield very small rates of emissions
deterioration through time. We assume that in real-world use, tampering and mal-maintenance dominate
emissions deterioration over time for heavy-duty diesel vehicles. As a result, MOVES specifically models
the deterioration due to tampering and mal-maintenance, which we assume also includes any other
emission increases due to vehicle aging and deterioration.
The tampering and mal-maintenance methodology was first incorporated into MOVES2010151 from studies
conducted between 1988 and 2007 (See Section B.2. The T&M methodology used in MOVES3 is unchanged
from MOVES2010 along with much of the original T&M assumptions on T&M frequency and T&M
adjustment factors. Slight updates to NOx and PM2.5 T&M adjustment factors were made for
MOVES2014.38 No changes were made to the T&M assumptions or data between MOVES2014 and
MOVES3. We acknowledge that T&M adjustment factors in MOVES should be re-evaluated and updated in
the future, particularly to incorporate data on the durability and emissions performance of advanced
aftertreatment systems on modern heavy-duty diesel vehicles.
This section describes the derivation of T&M emission rates applied to diesel running exhaust. The
estimation of heavy-duty gasoline deterioration is discussed in the derivation of the heavy-duty gasoline
rates (Section 3). The derivation of the T&M effects for diesel extended idle emissions are included in
Appendix C.
273
-------
B.l Modeling Tampering and Mal-maintenance
As T&M affects emissions through age, we developed a simple function of emission deterioration with age.
New vehicles and engines have zero-mile emission rates for each operating mode and maintain that rate
until the age of the vehicle/engine matches the warranty period. Once the warranty period ends, the
emission rate increases linearly until the vehicle/engine reaches its useful life age. At the end of the useful
life, the emissions rates remain constant at a level calculated from the tampering & mal-maintenance
(T&M) adjustment factor.. Figure B-l shows this relationship. The actual emission levels were determined
through data analysis detailed below.
period
Figure B-l. Qualitative Depiction of the Implementation of Age Effects
The T&M adjustment factor is calculated as the sum of the product of the T&M frequency for each
failure i, and the corresponding T&M emission effect, as shown in Equation 7-1.
(T&M frequency; X T&M emission effectp i) Equation 7-1
274
-------
Where:
fT&M= the tampering and mal-maintenance adjustment factor for pollutant p
T&M frequency; = estimated fleet average frequency of a tampering & mal-maintenance
failure i.
T&M emission effect; = estimated emission effect for pollutant p associated with tampering
& mal-maintenance failure i.
The emission rate at the end of useful life is then calculated using Equation 7-2.
E^End of useful life,p,r,o ERzero mile,p,r,o X (l + fr&M.p)
Equation 7-2
Where:
EREnd of warranty,] = the heavy-duty diesel emission rate at the end of warranty for each
pollutant p, regulatory class, r, and operating mode, o
ERZero mile= the zero-mile heavy-duty diesel emission rate for each pollutant p, regulatory
class, r, and operating mode, o
fx&M= the tampering and mal-maintenance adjustment factor for each pollutant p (Equation
7-1)
The useful life refers to the length of time that engines are required to meet emissions standards. We
incorporated this age relationship by averaging emissions rates across the ages in each age group. Mileage
was converted to age with VIUS152 (Vehicle Inventory and Use Survey) data, which contains data on how
quickly trucks of different regulatory classes accumulate mileage. Table B-l shows the emissions warranty
period and approximate useful life requirement period for each of the regulatory classes.
275
-------
Table B-l. Warranty and Useful Life Requirements by Regulatory Class
Regulatory class
Warranty agea
(Requirement:
100,000 miles or 5 years)
Useful life mileage/age
requirement
Useful life
agea
HHD
1
435,000/10
4
MHD
2
185,000/10
5
LHD45
4
110,000/10
4
LHD2b3
4b
110,000/10
4
BUS
2
435,000/10
10
Notes:
aThe warranty age and useful life age here are based on typical miles driven by vehicles in the regulatory class. For
example, HHD vehicles typically accumulate a large number of miles per year (100,000+/yeai"). Thus, HHD vehicles
complete their warranty and useful life requirements based on mileage while the vehicle age is still much below the
requirement.
b LHD2b3 diesel vehicles actually have a warranty of 50K miles. We realized later that MOVES emission rates for LHD
diesel were incorrectly developed using a warranty period of 100K instead of 50K. We will update the warranty period
for LHD diesel in a future update to MOVES.
While both age and mileage metrics are given for these periods, whichever comes first determines the
applicability of the warranty. As a result, since MOVES deals with age and not mileage, we needed to
convert all the mileage values to age equivalents, as the mileage limit is usually reached before the age
limit. The data show that on average, heavy heavy-duty trucks accumulate mileage much more quickly
than other regulatory classes. Therefore, deterioration in heavy heavy-duty truck emissions will
presumably happen at younger ages than for other regulatory classes. Buses, on average, do not
accumulate mileage as quickly. Therefore, their useful life period is governed by the age requirement, not
the mileage requirement.
We use the scaled age effect calculate the age-adjusted emission rates for each age. The scaled age effect,
sa, is calculated using the age of the vehicle in comparison to the warranty and useful life requirements, as
shown in Table B-2. When the vehicle age is between the end of the warranty and the useful life, sa is
interpolated between 0 and 1.
Table B-2. Calculation of sa
Where:
0
age < end of warranty age
276
-------
{age — end of warranty age)
end of warranty age < age < useful life
(Useful life age — end of warranty age)
1
age > useful life
Since MOVES deals with age groups and not individual ages (Table 1-6), the increase in emissions by age
must be calculated by age group. We assumed that there is an even age distribution within each age group
(e.g. ages 0, 1, 2, and 3 are equally represented in the 0-3 age group). We calculated average scaled age
effects for each age group. This is important since, for example, HHD trucks reach their useful life at four
years, which means they will increase emissions through the 0-3 age group. As a result, the 0-3 age group
emission rate will be higher than the zero-mile emission rate for HHD trucks. Table B-3 shows the average
scaled age effect by age group. In this table, a value of 0 indicates no deterioration, or zero-mile emissions
level (ZML), and a value of 1 indicates a fully deteriorated engine, or maximum emissions level, at or
beyond useful life (UL). We multiplied the scaled age factor by the T&M adjustment factor of each
pollutant and age in Equation 7-1.
Table B-3. Scaled Age Effect, sa
Age Group
LHD
MHD
HHD
Bus
0-3
0
0.083
0.25
0.03125
4-5
1
0.833
1
0.3125
6-7
1
1
1
0.5625
8-9
1
1
1
0.8125
10-14
1
1
1
1
15-19
1
1
1
1
20+
1
1
1
1
E^p,r,a,o ERzero mile,p,r,o ^ X ^T&m) Equation 7-3
Where:
ERp,r,o,a = the heavy-duty diesel emission rate for each pollutant p, regulatory class r, age a,
operating mode, o,
ERZero mile= the zero-mile heavy-duty diesel emission rate for each pollutant p, regulatory
class r, operating mode, o
sa= scaled age effect at age a
fx&M= the tampering and mal-maintenance adjustment factor (Equation 7-1)
277
-------
Sections B.2 through B.9 discuss the data sources and assumptions used to determine the T&M failure
frequencies and T&M emission effect for each pollutant and model year range of vehicle, used to derive
the T&M adjustment factor in Equation 7-1.
B.2 Data Sources
EPA used the following information to develop the tampering and mal-maintenance occurrence rates used
in MOVES:
• California's ARB EMFAC2007 Modeling Change Technical Memo153 (2006). The basic EMFAC
occurrence rates for tampering and mal-maintenance were developed from Radian and EFEE
reports and internal CARB engineering judgment.
• Radian Study (1988). The report estimated the malfunction rates based on survey and
observation. The data may be questionable for current heavy-duty trucks due to
advancements such as electronic controls, injection systems, and exhaust aftertreatment.
• EFEE report (1998) on PM emission deterioration rates for in-use vehicles. Their work included
heavy-duty diesel vehicle chassis dynamometer testing at Southwest Research Institute.
• EMFAC2000 (2000) Tampering and Mal-maintenance Rates
• EMA's comments on ARB's Tampering, Malfunction, and Mal-maintenance Assumptions for
EMFAC 2007
• University of California -Riverside (UCR) "Incidence of Malfunctions and Tampering in Heavy-
Duty Vehicles"
• Air Improvement Resources, Inc.'s Comments on Heavy-Duty Tampering and Mal-maintenance
Symposium
B.3 T&M Categories
EPA generally adopted the categories developed by CARB, with a few exceptions. The high fuel pressure
category was removed. We added a category for misfueling to represent the use of nonroad diesel in cases
when ULSD onroad diesel is required. We combined the injector categories into a single group. We
reorganized the EGR categories into "Stuck Open" and "Disabled/Low Flow." We included the PM
regeneration system, including the igniter, injector, and combustion air system in the PM filter leak
category.
EPA grouped the LHDD, MHDD, HHDD, and Diesel bus groups together, except for model years 2010 and
beyond. We assumed that the LHDD group will primarily use Lean NOx Traps (LNT) for the NOx control in
2010 and beyond. On the other hand, we also assumed that Selective Catalyst Reduction (SCR) systems will
be the primary NOx aftertreatment system for HHDD. Therefore, the occurrence rates and emission
impacts will vary in 2010 and beyond depending on the regulatory class of the vehicles.
278
-------
B.4
T&M Model Year Groups
EPA developed the model year groups based on regulation and technology changes.
• Pre-1994 represents non-electronic fuel control.
• 1998-2002 represents the time period with consent decree issues.
• 2003 represents early use of EGR.
• 2007 and 2010 contain significant PM and NOx regulation changes.
• 2010-and later represent heavy-duty trucks with required OBD. This rule began in MY 2010
with complete phase-in by MY 2013. The OBD impacts are discussed in Section B.10.
B.5 T&M Failure Frequency Rates and Differences from EMFAC2007
EPA adopted the CARB EMFAC2007 occurrence rates, except as noted below.
Clogged Air Filter: EPA reduced the frequency rate from EMFAC's 15 percent to 8 percent. EPA reduced
this value based on the UCR results, the Radian study, and EMA's comments that air filters are a
maintenance item. Many trucks contain indicators to notify the driver of dirty air filters and the drivers
have incentive to replace the filters for other performance reasons.
Other Air Problems: EPA reduced the frequency rate from 8 percent to 6 percent based on the UCR
results.
Electronics Failed: EPA continued to use the 3 percent frequency rate for all model years beyond 2010. We
projected that the hardware would evolve through 2010, rather than be replaced with completely new
systems that would justify a higher rate of failure. We assumed that many of the 2010 changes would
occur with the aftertreatment systems which are accounted for separately.
EGR Stuck Open: EPA believes the failure frequency of this item is rare and therefore set the level at 0.2
percent. This failure will lead to drivability issues that will be noticeable to the driver and serve as an
incentive to repair.
EGR Disabled/Low Flow: EPA estimates the ERG failure rate at 10 percent. All but one major engine
manufacturer had EGR previous to the 2007 model year and all have it after 2007, so a large increase in
rates seem unwarranted. However, the Illinois EPA stated that "EGR flow insufficient" is the top OBD issue
found in their LDV l/M program154 so it cannot be ignored.
NOx Aftertreatment malfunction: EPA developed a NOx aftertreatment malfunction rate that is dependent
on the type of system used. We assumed that HHDD will use primarily SCR systems and LHDD will primarily
use LNT systems. We estimated the failure rates of the various components within each system to develop
a composite malfunction rate (Table B-4).
The individual failure rates were developed considering the experience in agriculture and stationary
industries of NOx aftertreatment systems and similar component applications. Details are included in the
chart below. We assumed that tank heaters had a five percent failure rate but were only required in one
third of the country during one fifth of the year. The injector failure rate is lower than fuel injectors, even
279
-------
though they have similar technology, because there is only one required in each system and it is operating
in less severe environment of pressure and temperature. We believe the compressed air delivery system is
very mature based on a similar use in air brakes. We also believe that manufacturers will initiate engine
power de-rate as incentive to keep the urea supply sufficient.
Table B-4. NOx Aftertreatment Failure Rates
Occurrence Rate
SCR
Urea tank
0.5%
Tank heaters
1%
In-exhaust injectors
2%
Compressed air delivery to injector
1%
Urea supply pump
1%
Control system
5%
Exhaust temperature sensor
1%
Urea supply
1%
Overall 13%
LNT
Adsorber
7%
In-exhaust injectors
2%
Control system
5%
Exhaust temperature sensor
1%
Overall 16%
NOx aftertreatment sensor: EPA will assume a 10 percent failure mode for the aftertreatment sensor. We
developed the occurrence rate based on the following assumptions:
• Population: HHDD: vast majority of heavy-duty applications will use selective catalytic reduction
(SCR) technology with a maximum of one NOx sensor. NOx sensors are not required for SCR -
manufacturers can use models or run open loop. Several engine manufacturers representing 30
percent of the market plan to delay the use of NOx aftertreatment devices through the use of
improved engine-out emissions and emission credits.
• Durability expectations: SwRI completed 6000 hours of the European Stationary Cycle (ESC) cycling
with NOx sensor. Internal testing supports longer life durability. Discussions with OEMs in 2007
indicate longer life expected by 2010.
• Forward looking assumptions: Manufacturers have a strong incentive to improve the reliability and
durability of the sensors because of the high cost associated with frequent replacements.
PM Filter Leak: EPA will use 5 percent PM filter leak and system failure rate. They discounted high failure
rates currently seen in the field.
PM Filter Disable: EPA agrees with CARB's 2 percent tamper rate of the PM filter. The filter causes a fuel
economy penalty so the drivers have an incentive to remove it.
280
-------
Oxidation Catalyst Malfunction/Remove: EPA believes most manufacturers will install oxidation catalysts
initially in the 2007 model year and agrees with CARB's assessment of 5 percent failure rate. This rate
consists of an approximate 2 percent tampering rate and 3 percent malfunction rate. The catalysts are
more robust than PM filters, but have the potential to experience degradation when exposed to high
temperatures.
Misfuel: EPA estimated that operators will use the wrong type of fuel, such as agricultural diesel fuel with
higher sulfur levels, approximately 0.1 percent of the time.
B.6 Tampering & Mal-maintenance Failure Frequency Rate Summary
Table B-5. T&M Failure Frequency Rate by Model Year Group
Frequency Rates
1994-97
1998-2002
2003-2006
2007-2009
2010+ HHDT
2010+ LHDT
Timinq Advanced
5%
2%
2%
2%
2%
2%
Timinq Retarded
3%
2%
2%
2%
2%
2%
Injector Problem (all)
28%
28%
13%
13%
13%
13%
Puff Limiter Mis-set
4%
0%
0%
0%
0%
0%
Puff Limiter Disabled
4%
0%
0%
0%
0%
0%
Max Fuel Hiqh
3%
0%
0%
0%
0%
0%
Cloqqed Air Filter - EPA
8%
8%
8%
8%
8%
8%
Wronq/Worn Turbo
5%
5%
5%
5%
5%
5%
Intercooler Cloqqed
5%
5%
5%
5%
5%
5%
Other Air Problem - EPA
6%
6%
6%
6%
6%
6%
Enqine Mechanical Failure
2%
2%
2%
2%
2%
2%
Excessive Oil Consumption
5%
3%
3%
3%
3%
3%
Electronics Failed - EPA
3%
3%
3%
3%
3%
3%
Electronics Tampered
10%
15%
5%
5%
5%
5%
EGR Stuck Open
0%
0%
0.2%
0.2%
0.2%
0.2%
EGR Disabled/Low Flow - EPA
0%
0%
10%
10%
10%
10%
Nox Aftertreatment Sensor
0%
0%
0%
0%
10%
10%
Replacement Nox Aftertreatment Sensor
0%
0%
0%
0%
1%
1%
Nox Aftertreatment Malfunction - EPA
0%
0%
0%
0%
13%
16%
PM Filter Leak
0%
0%
0%
5%
5%
5%
PM Filter Disabled
0%
0%
0%
2%
2%
2%
Oxidation Catalyst Malfunction/Remove - EPA
0%
0%
0%
5%
5%
5%
Mis-fuel - EPA
0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
B. 7 NOx T&M Emission Effects
EPA developed the emission effect from each tampering and mal-maintenance incident from CARB's
EMFAC, Radian's dynamometer testing with and without the malfunction present, Engine, Fuel, and
Emissions Engineering Inc. (EFEE) results, and EPA staff testing experience.
EPA estimated that the lean NOx traps (LNT) in LHDD are 80 percent efficient and the selective catalyst
reduction (SCR) systems in HHDD are 90 percent efficient at reducing NOx.
281
-------
EPA developed the NOx emission factors of the NOx sensors based on SCR systems' ability to run in open-
loop mode and still achieve NOx reductions. The Manufacturers of Emission Controls Association (MECA)
has stated that a 75-90 percent NOx reduction should occur with open loop control and >95 percent
reduction should occur with closed loop control.155 Visteon reports a 60-80 percent NOx reduction with
open loop control.156
In testing, the failure of the NOx aftertreatment system had a different impact on the NOx emissions
depending on the type of aftertreatment. The HHDD vehicles with SCR systems would experience a 1000
percent increase in NOx during a complete failure, therefore we estimated a 500 percent increase as a
midpoint between normal operation and a complete failure. The LHDD vehicles with LNT systems would
experience a 500 percent increase in NOx during a complete failure. We estimated a 300 percent increase
as a value between a complete failure and normal system operation. The values with 0 percent effect in
shaded cells represent areas which have no occurrence rate.
282
-------
Table B-6, NOx T&M Emission Effect by Model Year Group
1994 -87
1998-2002
2003-2006
2007-2009
2010+ HHDT
2010 LHDT
Federal Emission Standard
5.0
5.0
4.0
2.0
0.2
0.2
Tirninq Advanced
60%
60%
60%
60%
6%
12%
Timiriq Retarded
-20%
-20%
-20%
-20%
-20%
-20%
Injector Problem fall)
-5%
-1%
-1%
-1%
-1%
-1%
Puff Limiter Mis-set
0%
0%
0%
0%
0%
0%
Puff Limiter Disabled
0%
0%
0%
0%
0%
0%
Max Fuel Hiqh
10%
0%
0%
0%
0%
0%
Cloqqed Air Filter
0%
0%
0%
0%
0%
0%
Wrong/Worn Turbo
0%
0%
0%
0%
0%
0%
Intercooler Cloqqed
25%
25%
25%
25%
3%
5%
Other Air Problem
0%
0%
0%
0%
0%
0%
Engine Mechanical Failure
-10%
-10%
-10%
-10%
-10%
-10%
Excessive Oil Consumption
0%
0%
0%
0%
0%
0%
Electronics Failed
0%
0%
0%
0%
0%
0%
Electronics Tampered
80%
80%
80%
80%
8%
16%
EGR Stuck Open
0%
0%
-20%
-20%
-20%
-20%
EGR Disabled t Low Flow
0%
0%
30%
50%
5%
10%
Nox Aftertreatment Sensor
0%
0%
0%
0%
200%
200%
Replacement Nox Aftertreatment Sensor
0%
0%
0%
0%
200%
200%
Nox Aftertreatment Malfunction
0%
0%
0%
0%
500%
300%
PM Filter Leak
0%
0%
0%
0%
0%
0%
PM Filter Disabled
0%
0%
0%
0%
0%
0%
Oxidation Catalyst Malfunction/Remove
0%
0%
0%
0%
0%
0%
Mis-fuel
283
-------
Combining the NOx emission effects with the frequency results in the initial T&M adjustment factor shown
in the Table B-7 below. This methodology estimated a small (9-14%) T&M NOx adjustment factor for 2009
and earlier models due to NOx effect of the following failure modes: electronics tampered, timing
advances, intercooler clogged, and ERG disabled/Low Flow. MOVES does not use the estimated NOxT&M
emission effects initially estimated for 2009 and earlier model years, and assumes no NOx increase. This is
incorporated into the 3rd column of Table B-7 labeled with (Remove 2009 and earlier).
MOVES assumes NOx increases only in the 2010+ vehicles with NOx exhaust aftertreatment technologies, in
contrast to the initial estimate of T&M due to a few reasons:
• The Consent Decree Testing conducted by West Virginia University did not show an increase
in NOx emissions with odometer (and consequently, age) during or following the regulatory
useful life.157 Since the trucks in this program were collected from in-use fleets, we do not
believe that these trucks were necessarily biased toward cleaner engines.
• Heavy-duty diesel manufacturers often certify zero or low deterioration factors for these
engines.
• Starting with MY 2010, we expect T&M effects to become much more significant, because a
failure in the NOx aftertreatment system will substantially increase emissions. We decided to
initiate modeling of the NOx T&M adjustment factor with the implementation of the 2010
standards.
Table B-7. Tampering & Mai-Maintenance NOx adjustment factor (fr&M,i\iox) (Percent) for MHD, HHD and Urban
Buses
Model
years
fl&M,NOx,nonOBD
(Initial)
fr&M,NOX ,nonOBD
(Removed 2009
and earlier)
foBD
fr&M,NOX,OBD
1994-1997
10
0
0
-
1998-2002
14
0
0
-
2003-2006
9
0
0
-
2007-2009
11
0
0
-
2010-2012
SCR
87
87
0.33
77
2010-2012
LNT
72
72
1
48
2013+ SCR
87
87
1
58
284
-------
The T&M adjustment factors (fi&M,NOx) are calculated using the N0X emission effects and frequencies shown
above. The fr&M,NOx factors incorporate the OBD assumptions discussed in Section B.10, including the
assumed penetration of OBD (foBD)
Lean NOxtrap (LNT) aftertreatment are assumed to penetrate 25 percent of LHD2b3 trucks starting in
2007, consistent with the assumptions previously made in Section 2.1.1.4.6.
The T&M rates for LHD2b3 in 2007-2009 are calculated by adjusting Equation 3-10 to account for T&M of
LNT aftertreatment, as shown in Equation 7-4:
2007 - 2009 LNT NOx emissions (T&M)
2003 — 2006 LHD2b3 NOx emissions
Equation 7-4
/LNT normal emissions \
= (normal op. frequency) x — - x (T&M effect)
V baseline emissions /
/baseline emissionsx
+ (DPF reg. frequency) x -
V baseline emission /
= (0.90) x (0.10) x (1.72) + (0.10) x (1) x (1) = 0.2548
The ratio of 2007-2009 LHD2b3 (with T&M) over the baseline 2003-2006 NOx rates is calculated by
adjusting Equation 3-11 to account for the T&M effects of LNT, as shown in Equation 7-5.
2007 - 2009 LHD2b3 NOx emissions (T&M)
2003 — 2006 LHD2b3 NOx emissions
/2007 — 2009 LNT NOx emissions (T&M)\
= (LNT market share) T umu? vr. Equation 7-5
V 2003 — 2006 LHD2b3 NOx emissions /
+ (non
N / 2007 — 2009 emission standards \
— LNT market share) ——— :—r~
V2003 — 2006 NOx emissions standards/
=0.25x0.2548 +0.75x0.5=0.4387
285
-------
Then, the overall T&M effect for 2007-2009 LHD2b3 is calculated in Equation 7-6 by dividing Equation B-2
by Equation 3-11.
2007 - 2009 LHD2b3 NOx emissions (T&M)
2007 — 2009 LHD2b3 NOx emissions (zero mile)
Equation 7-6
(:
2003 — 2006 LHD2b3 NOx emissions
2007 - 2009 LHD2b3 NOx (T&M)
)/(¦
2007 - 2009 LHD2b3 NOx (zero mile)
2003 — 2006 LHD2b3 NOx emissions
= 0.4387/0.4225 = 1.04 = 4% increase due to T&M
For 2007-2009, LHD45 uses the same emission rates and T&M factors as for the LHD2b3 rates
As noted earlier, we assume no NOx increase for the pre-2007 for the other heavy-duty regulatory classes.
For 2010+, LHD2b3, we assume that both LNT and SCR equipped vehicles will provide the same level of
control with a 90 percent reduction from 2003-2006 levels (ignoring the PM regeneration NOx benefit for
LNT aftertreatment). Thus, for calculating the T&M NOx effects for 2010-2012, we weighted the LNT-
specific and 2013+SCR-specific T&M effects (from Table B-7) according to the market shares, as shown in
Equation 7-7:
2010+ LHD2b3 NOx emissions T&M
=LNT market share x (2010 thru 2012 LNT T&M) + (non-LNT market share) x Equation 7-7
(2013+SCR T&M)
=0.25x0.48+0.75x0.58=56%
For the MHD, HHD, and Urban Buses regulatory classes we use the SCR T&M effects from Table B-7. We
assume only 33 percent OBD penetration in 2010-2012, and full penetration for 2013+ model years. Note
that for 2010+ LHD45, we use the same T&M as the other heavy-duty regulatory classes, that assume no
LNT penetration, and a phased-in OBD penetration. The NOx T&M adjustment factors used for each MOVES
regulatory classes and model year groups are shown in Table B-8.
286
-------
Table B-8. N0X T&M Adjustment Factors (fr&M,i\iox) by MOVES Regulatory Classes and Model Year Groups
Model Year Group
LH2b3
(RegClass 41)
LHD45
(RegClassID
Other HD
(RegClassID
46,47,48)
Gliders
(RegClass
49)
2007-2009
4%
4%
0%
0%
2010-2012
56%
77%
77%
0%
2013+
58%
58%
58%
0%
B.8 PM T&M Emission Effects
EPA developed the PM emission effects for each tampering and mal-maintenance incident from CARB's
EMFAC, Radian's dynamometer testing with and without the malfunction present, EFEE results, and
internal testing experience.
EPA estimates that the PM filter has 95 percent effectiveness. Many of the tampering and mal-
maintenance items that impact PM also have a fuel efficiency and drivability impact. Therefore, operators
will have an incentive to fix these issues.
EPA estimated that excessive oil consumption will have the same level of impact on PM as engine
mechanical failure. The failure of the oxidation catalyst is expected to cause a PM increase of 30 percent;
however, this value is reduced by 95 percent due to the PM filter effectiveness. We also considered a DOC
failure will cause a secondary failure of PM filter regeneration. We accounted for this PM increase within
the PM filter disabled and leak categories.
The values with 0 percent effect in shaded cells represent areas which have no occurrence rate.
In MOVES2014, we increased the PM emission effect for PM Filter Leaks and PM Filter Tampering for the
2007-2009 and 2010+ model year groups. The PM filter leak was increased from 600 percent to 935
percent and the PM Filter Disabled emission effect was increased from 1000 percent to 2670 percent.
These effects along with the OBD effects discussed in Section B.10 results in a fleet average PM2.5
Tampering & Mal-maintenance effect of 100 percent in 2007-2009 and 89 percent in 2010-2012 (Table
2-19).
287
-------
Table B-9. PM2.5 T&M Emission Effect by Model Year Group
1994-
1997
1998-
2002
2003-
2006
2007-
2009
2010
Federal Emission Standard
0.1
0.1
0.1
0.01
0.01
Timing Advanced
-10%
-10%
-10%
0%
0%
Timing Retarded
25%
25%
25%
1%
1%
Injector Problem
100%
100%
100%
5%
5%
Puff Limiter Mis-set
20%
0%
0%
0%
0%
Puff Limiter Dsabled
50%
0%
0%
0%
0%
Max Fuel High
20%
0%
0%
0%
0%
Clogged Air Filter
50%
50%
30%
2%
2%
Wrong/Worn Turbo
50%
50%
50%
3%
3%
Intercooler Clogged
50%
50%
30%
2%
2%
Other Air Problem
40%
40%
30%
2%
2%
Engine Mechanical Failure
500%
500%
500%
25%
25%
Excessive Oil Consumption
500%
500%
500%
25%
25%
Electronics Failed
60%
60%
60%
3%
3%
Electronics Tampered
50%
50%
50%
3%
3%
EGR Stuck Open
0%
0%
100%
5%
5%
EGR Disabled/Low Flow
0%
0%
-30%
-30%
-30%
NOx Aftertreatment Sensor
0%
0%
0%
0%
0%
Replacement NOx
Aftertreatment Sensor
0%
0%
0%
0%
0%
NOx Aftertreatment Malfunction
0%
0%
0%
0%
0%
PM Filter Leak
0%
0%
0%
935%
935%
PM Filter Disabled
0%
0%
0%
2670%
2670%
288
-------
Oxidation Catalyst
Malfunction/Remove
0%
0%
0%
0%
0%
Mis-fuel - EPA
30%
30%
30%
100%
100%
B.9 THC T&M Emission Effects
EPA estimated oxidation catalysts are 80 percent effective at reducing hydrocarbons. All manufacturers
will utilize oxidation catalysts in 2007, but only a negligible number were installed prior to the PM
regulation reduction in 2007. We assumed that with Tampering and Mal-maintenance, the THC zero level
emissions will increase by 50 percent. This still represents a 70 percent reduction in THC emissions
between zero-mile 2006 emissions and fully deteriorated 2007 vehicles.
We reduced CARB's THC emission effect for timing advancement because earlier timing should reduce
THC, not increase them. The effect of injector problems was reduced to 1000 percent based on EPA's
engineering staff experience. We increased the THC emission effect of high fuel pressure (labeled as Max
Fuel High) to 10 percent in 1994-1997 years because the higher pressure will lead to extra fuel in early
model years and therefore increased THC. Lastly, we used the THC emission effect of advanced timing for
the electronics tampering (0 percent) for all model years. The values with 0 percent effect in shaded cells
represent areas which have no occurrence rate.
289
-------
Table B-10. THC T&M Emission Effect by Model Year Group
1994-97
1998-2002
2003-2006
2007-2009
2010+ HHDT
2010 LHDT
Federal Emission Standard
1.3
1.3
1.3
0.2
0.14
0.14
Timing Advanced
0%
0%
0%
0%
0%
0%
Timing Retarded
50%
50%
50%
50%
10%
10%
Injector Problem (all)
1000%
1000"'.
1000"'.
1000"'.
200"-.'.
200"-.'.
Puff Limiter Mis-set
0%
07:'.
0%
0%
0%
OV:'.
Puff Limiter Disabled
0%
0:<¦
0%
0%
0%
0:<¦
Max Fuel High
10%
0%
0%
0%
07:'.
Clogged Air Filter
0%
0%
0%
0%
0%
0%
Wrong/Worn Turbo
0%
0%
0%
0%
0%
0%
Intercooler Clogged
0%
0%
0%
0%
0%
0%
Other Air Problem
0%
0%
0%
0%
0%
0%
Engine Mechanical Failure
500%
500%
500%
500%
100%
100%
Excessive Oil Consumption
300%
300%
300%
300%
60%
60%
Electronics Failed
50%
50%
50%
50%
10%
10%
Electronics Tampered
0%
0%
EGR Stuck Open
100":'.
1 ()()¦¦•.•.
20%
20%
EGR Disabled / Low Flow
0%
0%
Nox Aftertreatment Sensor
0%
0%
0%
Replacement Nox Aftertreatment Sensor
0%
0%
0%
Nox Aftertreatment Malfunction
0%
0%
0%
PM Filter Leak
0%
0%
0%
PM Filter Disabled
o.:'.
0%
0%
0%
Oxidation Catalyst Malfunction/Remove
0%
0%
50%
50%
50%
Mis-fuel
290
-------
A separate tampering analysis was not performed for CO; rather, the THC effects were assumed to apply
for CO.
Combining all of the emissions effects and failure frequencies discussed in this section, and the OBD effects
discussed in the next section, we summarized the aggregate emissions impacts over the useful life of the
fleet in the main body of the document in Table 2-22 (THC and CO).
B.10 HD OBD impacts
With the finalization of the heavy-duty onboard diagnostics (HD OBD) rule, we made adjustments to 2010
and later model years to reflect the rule's implementation.
Specifically, we reduced the emissions increases for all pollutants due to tampering and mal-maintenance
by 33 percent. Data on the impact of OBD were not available for heavy-duty trucks, and this number is
probably a conservative estimate. This is in addition to the substantial PM2.5 and NOx reductions for 2010
and later vehicles due to the implementation of other standards. We assumed, since the rule phased-in
OBD implementation, that 33 percent of all engines would have OBD in the 2010, 2011, and 2012 model
years, and 100 percent would have OBD by 2013 model year and later. Equation 7-8 describes the
calculation of the percent increase in emission rate through useful life (T&M adjustment factors (fr&M));
where f0BD represents the fraction of the fleet equipped with OBD (0 percent for model years 2009 and
earlier, 33 percent for model years 2010-2012, and 100 percent for model years 2013 and later). The result
from this equation can be plugged into Equation 7-2 to determine the emission rate for any age group.
^T&M,p = fT&M,nonOBD,p X (1 — foBD) + 0-67 X fT&M,nonOBD,p X (foBD) Equation 7-8
Where:
fx&M,p= the tampering and mal-maintenance adjustment factor for pollutant, p, that accounts
for the phase-in of OBD
fx&M,p= the tampering and mal-maintenance adjustment factor for pollutant, p, for engines
without OBD; calculated in Sections B.7 through B.9
f0BD= penetration of the fleet equipped with OBD
These OBD impacts apply to any truck in GVWR Class 4 and above (regulatory classes LHD45, MHD, HHD,
and Urban Bus). LHD2b3 trucks are assumed to have the OBD fully-phased in beginning in model year
2010 as shown in Equation 7-7. As data for current and future model years become available, we may
consider refining these estimates and methodology.
291
-------
Appendix C Tampering and Mal-maintenance for MY 2007 and
Later Diesel Extended Idle
As discussed in Section 2.3.2.3 we assume the failure of diesel particulate filters (DPF) is the primary cause
of T&M effects on emission deterioration in 2007+ extended idle emissions. We made assumptions about
the failure rates of DPFs from in-use trucks based on consultation with several references and staff at the
California Air Resources Board (CARB) as summarized in Table 7-1. We adopted the assumption shared by
CARB staff that 10 percent of 2007-2009 DPFs fail in the real-world, and 5 percent of 2010+ DPFs fail in the
real-world.
Table 7-1. References Used to Support In-Use DPF Failure Rate Assumption for Extended Idling Emissions
Study
Relevant Information
US EPA (2015)108
7% of 2007+ trucks in MOVES are assumed to either have a PM filter leak or have
the PM filter disabled. Current assumption for running exhaust emissions in
MOVES3.
Preble et al.
(2015)158
20% of trucks produce 80% of black carbon (BC) emissions from Port of Oakland
2013 truck fleet, where 99% of the trucks are equipped with DPFs
Bishop et al.
(2014)159
3% of 2007+ trucks at Port of LA have PM emissions 3x the standard. 9% of
2008+ trucks at Cottonwood site have PM emissions 3x the standard
CARB (2015)160
35% to 4% of trucks submitted warranty claims related to the PM filter between
2007 and 2011
CARB (2015)1S0
8% of trucks were classified as high emitters (emitting over 5% opacity) from a
sample of >1,800 trucks test in the snap-idle acceleration test by CARB, about
~l/2 equipped with DPFs
CARB
correspondence
(2016)
~10% of 2007-2009 DPFs and ~5% of 2010+ DPFs to fail in real-world, based on
their observations from warranty claims, snap-idle acceleration opacity tests,
and their review of the Bishop et al. (2014)159 and Preble et al. (2015)158 studies.
To account for the failure of DPF in the THC and PM2.5 emission rates, we used the 2005-2006 average
extended idle emission rates to represent the 'failed' DPF emission rates. We then calculated a
'Deteriorated' emission rate that represents a mix of failed and properly operating systems by assigning
the 'failed' DPF emission rates a weight of 10 percent in the 2007-2009 model year group, and 5 percent
weight in the 2010-2012, and 2013+ model year groups, as shown in Table 7-2. The 'Deteriorated' emission
rate represents the presumed emission rate of fully-aged heavy-duty diesel trucks. Unlike the start and
running MOVES emission rates, extended idle emission rates in MOVES are not distinguished by age. Thus,
these rates are constant with respect to age.
292
-------
Table 7-2. Baseline and deteriorated THC and PM2.s emission rates to account for failure of diesel particulate filters
(DPFs) by model year groups
Engine
Model
Year
Baseline
Deteriorated
THC
(g/hr)
PM2.s
(g/hr)
EC
(g/hr)
nonEC
(g/hr)
Failure
rate
THC
(g/hr)
PM2.s
(g/hr)
EC
(g/hr)
nonEC
(g/hr)
2005-
2006
8.49
0.251
0.065
0.187
-
8.49
0.251
0.065
0.187
2007-
2009
8.49
0.075
0.007
0.067
10%
8.49
0.092
0.013
0.079
2010-
2012
2.53
0.026
0.004
0.022
5%
2.83
0.037
0.007
0.030
2013+
1.38
0.012
0.002
0.010
5%
1.74
0.024
0.005
0.019
We assume that trucks that are under warranty would have substantially fewer aftertreatment failures
than older trucks. Because extended idle rates are modelled as constant with age, to estimate the fleet-
average emission rates used in MOVES, we used the 'Baseline' emission rates to represent trucks that are
within the specified 435,000 miles useful-life of the engine in the US EPA regulations.161 We use the
deteriorated emission rate to represent the years between the regulated "useful life" and the 1,530,000
miles that MOVES models as the mean life-time miles for a long-haul combination trucks. Using the
'deterioration fraction' [(l-.435)/1.53 = 0.72] as the fraction of the vehicle miles traveled during the
deterioration phase, we calculated fleet-average emission rates used for MOVES in Table 7-3. As shown,
the MOVES EC/PM emission rates for MY 2007+ trucks are slightly higher than the 'Baseline' EC/PM
fractions in Table 7-2, because the fleet emissions are assumed to include some contribution of emissions
from trucks with failed DPFs, which have a higher EC/PM fraction.
293
-------
Table 7-3. Emission Rates Calculated from Weighting the 'Baseline' and 'Deteriorated' Emission Rates from Table
7-2 Using the Deteriorated Fraction
Engine
Model
Year
MOVES
Deteriorated
Fraction
THC
(g/hr)
PM2.5
(g/hr)
EC
(g/hr)
nonEC
(g/hr)
EC/PM
2005-2006
-
8.49
0.251
0.065
0.187
0.26
2007-2009
0.72
8.49
0.087
0.012
0.076
0.13
2010-2012
0.72
2.75
0.034
0.006
0.028
0.18
2013+
0.72
1.64
0.021
0.004
0.017
0.20
Although, 2005-2006 model year engine data was used in this analysis, the update itself is limited to the
model year 2007 and later emission rates.
294
-------
Appendix D Pre-2007 Model Year Extended Idle Data Summary
These tables provides additional information on the data used to estimate extended idle emissions for pre-
2007 MY vehicles as described in Section 2.3.1.
295
-------
Idle NOx Rates (gram/hour) Summary
Program
Condition
#Samples
Mean NOX Emiss Rate
1991-2006 Low Speed Idle, A/C Off
McCormick, High Altitude, HDT
Low RPM, AC Off
12
85
Lim, EPA
Low RPM, No access
12
109
Irick, Clean AirTech & IdleAire
49
87
WVU - 1991-2004
Low RPM, AC Off
48
83
WVU, NCHRP
2
47
Tang, Metro NY 1984-1999
33
81
Calcagno
Low RPM, AC Off
27
120
Brodrick, UC Davis
Low RPM, AC Off
1
104
Storey
Low RPM, AC Off
4
126
Overall
188
91
1991-2006 High Speed Idle, A/C Off
Lim, EPA CCD
High RPM, No access
5
169
Calcagno
High RPM, AC Off
21
164
Overall
26
165
1991-2006 High Speed Idle, A/C On
Lim, EPA CCD
High RPM, AC On
5
212
Brodrick, UC Davis
High RPM, AC On
1
240
Calcagno
High RPM, AC On
21
223
Storey
High RPM, AC On
4
262
Overall
31
227
1975-1990 Low Speed Idle, A/C Off
WVU - 1975-1990
Low RPM, AC Off
18
48
Lim, EPA, CCD, 1985 MY
Low RPM, AC Off
1
20
Overall
19
47
1975-1990 High Speed Idle, A/C On (calculated)
Ratio of 1991-2006 "High Idle, A/C On" to "Low Idle, A/C Off"
2.5
Overall (calculated)
115.4
Calculated Extended Idle MYs 1975-1990:
69.3
Calculated Extended Idle MYs 1991-2006:
136.1
296
-------
Idle HC Rates (gram/hour) Summary
Program
Condition
#Samples
Mean HC Emiss Rate
1991-2006 Low Speed Idle, A/C Off
McCormick, High Altitude, HDT
Low Idle, AC Off
12
10.2
WVU - 1991-2004
Low Idle, AC Off
48
9.5
Storey
Low Idle, AC Off
4
28
Overall
64
10.8
1991-2006 High Speed Idle, A/C On
Brodrick, UC Davis
High Idle, AC On
1
86
Storey
High Idle, AC On
4
48
Overall
5
55.6
1975-1990 Low Speed Idle, A/C Off
WVU - 1975-1990
Low Idle, AC Off
18
21
Overall
18
21
1975-1990 High Speed Idle, A/C On (calculated)
Ratio of 1991-2006 "High Idle, A/C On" to "Low Idle, A/C Off"
5.2
Overall (calculated)
108.2
Calculated Extended Idle MYs 1975-1990:
49.8
Calculated Extended Idle MYs 1991-2006:
25.6
297
-------
Idle CO Rates (gram/hour) Summary
Program
Condition
#Samples
Mean CO Emiss Rate
1991-2006 Low Speed Idle, A/C Off
McCormick, High Altitude, HDT
Low Idle, AC Off
12
71
Calcagno
Low Idle, AC Off
27
37
WVU - 1991-2004
Low Idle, AC Off
48
23
Storey
Low Idle, AC Off
4
25
Overall
91
33.6
1991-2006 High Speed Idle, A/C On
Calcagno
High Idle, AC On
21
99
Brodrick, UC Davis
High Idle, AC On
1
190
Storey
High Idle, AC On
4
73
Overall
26
98.5
1975-1990 Low Speed Idle, A/C Off
WVU - 1975-1990
Low Idle, AC Off
18
31
Overall
18
31
1975-1990 High Speed Idle, A/C On (calculated)
Ratio of 1991-2006 "High Idle, A/C On" to "Low Idle, A/C Off"
2.9
Overall (calculated)
91.0
Calculated Extended Idle MYs 1975-1990:
50.8
Calculated Extended Idle MYs 1991-2006:
55.0
298
-------
Idle PM Rates (gram/hour) Summary
Program
Condition
#Samples
Mean PM Emiss Rate
1991-2006 Low Speed Idle, A/C Off
McCormick, High Altitude, HDT
Low Idle, AC Off
12
1.8
Calcagno
Low Idle, AC Off
27
2.55
WVU - 1991-2004
Low Idle, AC Off
48
1.4
Storey
Low Idle, AC Off
4
0.3
Overall
91
1.7
1991-2006 High Speed Idle, A/C On
Calcagno
High Idle, AC On
21
4.11
Storey
High Idle, AC On
4
3.2
Overall
25
4.0
1975-1990 Low Speed Idle, A/C Off
WVU - 1975-1990
Low Idle, AC Off
18
3.8
Overall
18
3.8
1975-1990 High Speed Idle, A/C On (calculated)
Ratio of 1991-2006 "High Idle, A/C On" to "Low Idle, A/C Off"
2.3
Overall (calculated)
8.6
Calculated Extended Idle MYs 1975-1990:
5.4
Calculated Extended Idle MYs 1991-2006:
2.5
299
-------
Appendix E Developing Pre-2007 Model Year HD Diesel PM2.5
Emission Rates for Missing Operating Modes
As noted in Section 2.1.2.1, in cases where an estimated operating mode PM2.5 rate for pre-2007 MY HD
diesel trucks could not be directly calculated from data, we imputed the missing value using a log-linear
least-squares regression procedure. Regulatory class, model year group and speed class (0-25 mph, 25-50
mph and 50+ mph) were represented by dummy variables in the regression. The natural logarithm of
emissions was regressed versus scaled tractive power (STP) to represent the operating mode bins. The
regression assumed a constant slope versus STP for each regulatory class. Logarithmic transformation
factors (mean square error of the regression squared / 2) were used to transform the regression results
from a log based form to a linear form. Due to the huge number of individual second-by-second data
points, all of the regression relationships were statistically significant at a high level (99 percent confident
level). The table below shows the regression statistics, and the equation shows the form of the resulting
regression equation.
Table E-l. Regression Coefficients for HD Diesel Pre-2007 PM2.5 Emission Factor Model
Model-year
group
Speed Class (mph)
Type
Medium
Heavy-Duty
Heavy Heavy-
Duty
1960-87
1-25
Intercept (fio)
-5.419
-5.143
25-50
-4.942
-4.564
50+
-4.765
-4.678
1988-90
1-25
-5.366
-5.847
25-50
-4.929
-5.287
50+
-4.785
-5.480
1991-93
1-25
-5.936
-5.494
25-50
-5.504
-5.269
50+
-5.574
-5.133
1994-97
1-25
-5.927
-6.242
25-50
-5.708
-5.923
50+
-5.933
-6.368
1998-2006
1-25
-6.608
-6.067
25-50
-6.369
-5.754
50+
-6.305
-6.154
STP
Slope (/?i)
0.02821
0.0968
Transformation
Coefficient
0.5864
0.84035
(0.5c2)
ln(PM) = /?0 + y^STP + 0.5cr2
Where :
60 = an intercept term for a speed class within a model year group, as shown in the table above,
300
-------
61 = a slope term for STP, and
o2 = the mean-square error or residual error for the model fit,
STP = the midpoint value for each operating mode (kW/metric ton, see Table 1-4).
301
-------
Appendix F Heavy-Duty Gasoline Start Emissions Analysis Figures
The figures below show heavy-duty gasoline start emissions as mentioned in Section 3.2.1,
Figure 7-1 Cold-Start FTP Emissions for Heavy-Duty Gasoline Vehicles, Averaged by Model-year and Age Groups
FTP Cold—Starts (g). HD SI (hO< = 14Kj
CO slorls vs. Age by MYG
FTP Cold-Starts (g). HO SI (HD< = MK5
THC starts vs. Age by MYG
FTP Cold-Starts (g), HD SI (hO< = 14K)
MQx starts vs. Age by MYG
302
-------
FTP Cold-Starts (g), HD SI (HD< = 14K)
CO GEO-meon siorls vs. Age by MYG
agemld
node I year group B-e-e I960I98S &-S-B 19901990 I—*" 19911997 is-Hr-A 19982004
FTP Cold-Starts (g). HD Si (HD< = 14K)
THC CEO-meon starts vs. Age by MYG
(b) THC
^ /p\[
/ \
pH ' ¦
* ¦ - -- - -
2 3 4 S G 7 8 9 10 II 12 13
agenId
node I year group 19601989 B-e-B 19901990 "•—»—*" 19911997 tctr-A 19982004
FTP Cold-Starts (g). HD SI (HD< = 14K)
NOx GEQ-meon storis vs. Age by MYG
agenid
node! year group e-O-© I9601989 PHD 19901990 I—~ 19911997 A A £> 19982004
re 7-2 Cold-Start FTP Emissions for Heavy-Duty Gasoline Vehicles, GEOMETRIC MEANS by Model-year and Age
Groups
303
-------
CO In.SD vs. Age by MYG
node I year group O O O 19601989 e-B-S 19901990 t—I—~" 19911937 A-te-tt 19982004
FTP Cold—Siarts (g), HD SI (hO< = 14K)
THC In.SD vs. Age by MYG
node 1 year group OOP 19601989 B-B-B 19901990 H—I—h 19911997 AAA 19982001
FTP Cold-Starts (g), HD SI (HD< = 14K)
NOx In.SD vs. Age by MfG
¦node I year group &-&-& 19601989 B-B-B 19901 990 "»—<—h 19911997 H-fr-A 19982004
re 7-3 Cold-start FTP Emissions for Heavy-Duty Gasoline Trucks: LOGARITHMIC STANDARD DEVIATION by
Model-year and Age Groups
304
-------
CO ARITH-mean starts vs. Age by MYG
¦node 1 year group
FTP Cold— Starts (g). HD SI (HD< = 14K)
THC ARITH-mean starts vs. Age by MYG
(b) THC
i 1 : :
mode1yeargroup
13901990 I I I 1991 1997
FTP Cold— Starts (g), HD SI (l-D< = 14K)
NOx ARITH-mean starts vs. Age by MYG
(c) NOx
/ -4 a- ¦ • • 1 • • • i i
* ' 1 : |
model year group ©-©-© I9601989 B-e-S 19301990 +—•—It 1991 1997 tSnS-A 19982004
re F-l. Cold-Start Emissions for Heavy-Duty Gasoline Trucks: RECALCULATED ARITHMETIC MEANS by Model-
year and Age Groups
305
-------
Table F-l Emission Standards for Heavy-Duty Spark-Ignition Onroad Engines
Regulatory Class
Model Year
Emissions Standards (g/hp-hr)
CO
THC
NMHC
X
O
NMHC + NOx
LHD2b3
1990
14.4
1.1
6.0
1991-1997
14.4
1.1
5.0
1998-2004
14.4
1.1
4.0
2005-2007
14.4
1.0
2008+
14.4
0.14
0.20
LHD45, MHD
1990
37.1
1.9
6.0
1991-1997
37.1
1.9
5.0
1998-2004
37.1
1.9
4.0
2005-2007
37.1
1.0
2008+
14.4
0.14
0.20
306
-------
Appendix G Selection of Fixed Mass Factor {fSCaie) values for MY
2010+ Heavy-Duty Vehicles
In MOVES3, for model year 2010 and newer heavy-duty diesel, gasoline, and CNG running-exhaust
emissions operating mode based rates, we discarded the fixed mass factor {fSCaie) value of 17.1 metric tons
used in previous MOVES versions, and used the manufacturer-run HDIUT data to estimate new/SCate values
for LHD, MHD, and HHD weight classes. New/sco;e values were needed because the 17.1 value was too
large, limiting emission rate data to low and medium power operating modes, and requiring gap-filling for
high-power operating modes.
The new fscaie values for MY 2010+ vehicles are 5.00, 7.00, and 10.00 metric tons for LHD (regClass 41 and
42), MHD (regClass 46), and HHD (regClass 47 and 48), respectively. These fscaie values are used when
analyzing the real-world emissions data that leads to the base emission rates in the MOVES database. The
fscaie values are also used by the MOVES model, at run time, to convert vehicle activity to operating mode-
based time distributions. For consistency, operating mode-based emissions rates and time distributions,
for a given regulatory class and model year, must be based on the same/sco/e value. Glider vehicles
(regClass 49) continue to use emission rates from pre-2010 vehicles, and thus their/sco/e value is unchanged
at 17.1. Note that it is not meaningful to compare operating mode based rates based on different/sco;e
values. This appendix describes how we arrived at the/sco/e values.
The entire MY 2010+ HDIUT dataset (Section 2.1.1.1) was analyzed using a range of fSCaie values. For this
exercise we analyzed LHD, MHD, and HHD separately, but within those regulatory classes, we did not
divide the data set by NOx FEL or model years. We included the MOVES2014 value {fSCaie = 17.1) to show
how the operating mode distribution would look for a "business as usual" case. It is expected that the/scofe
for LHD should be lower than MHD, which in turn should be lower than HHD. Our goal was to find/scofe
values that allow the HDIUT data to cover all operating modes, thus reducing the need for gap-filling while
also leaving the highest power operating modes (30 and 40) as not saturated because the HDIUT data is
not expected to have very aggressive operation. If the/sco/e is too big, the high power operating modes are
left vacant. On the other hand, if the/SCate is too small, a lot of the data gets pushed to the high power
operating modes, and the high operating modes no longer capture emissions at only the most extreme and
aggressive operating conditions measured in the HDIUT data set. Both cases are sub-optimal because they
reduce the model's capability to distinguish operating modes in a meaningful way.
When analyzing the HDIUT data for various/scafe values, we estimated the number of vehicles, time, and
mass/time emission rates for criteria pollutants and C02 for each operating mode. Vehicle count and time,
per operating mode, were first cut criteria during the fscaie selection process. We used the C02 mass/time
rates as an additional check because these rates are known to have consistent and predictable
monotonically rising trend within each speed-bin (since higher power demand requires burning more fuel
which leads to more C02).Table G-l through Table G-3 show how the choice of fscaie values would affect the
307
-------
vehicle count and seconds in each operating mode, for LHD, MHD, and HHD, respectively. The number of
seconds is based on the HDIUT-based operating mode time fractions applied to a cycle of one million
seconds. Using a unique but representative/sco;e for each regClass, when combined with a cycle of the same
number of total seconds, should result in similar number of seconds in high power operating modes. In
other words, we expect LHD, MHD, and HHD vehicles in the HDIUT data set to have somewhat similar time
distribution across power modes. Finally, we used the C02 mass/time rate trends as an additional metric to
pick a final fscaie between candidate values that look reasonably good for both vehicle count and time
distribution.
Looking at Table G-l for LHD vehicles, fscaie = 2.06 results in every one of the 64 vehicles having operation in
operating modes 30 and 40 and significantly more seconds of data than operating modes 29 and 39,
respectively. On the other hand, a /sco;e value of 9.00 or 17.1 meant the high power operating modes had
only a couple vehicles and seconds, which is a sign of under-representation in those operating modes.
Thus, a suitable/scofe value, for LHD, should be between 2.06 and 9.00. Based on further analysis, the final
fscaie candidates for LHD were 4.00, 5.00, and 6.00. A value of 4.00 seemed too small because we did not
expect over 40 (out of 64) vehicles to have operation in operating modes 30 and 40. A value of 6.00
seemed too high because it led to only 40 seconds and 135 seconds of data (from a cycle with a million
seconds) in operating modes 30 and 40, respectively. The small sample size was deemed insufficient to
determine robuts emission rates for these operating mode bins. We picked 5.00 as the final fscaie value for
LHD because it resulted in a reasonable number of vehicles and seconds in the high power operating
modes 29, 30, 39, and 40. For confirmation purpose, we also compared the C02 mass/time rates for all the
fscaie values considered during the analysis and Figure G-l shows a comparison between the final
candidates of 4.00, 5.00, and 6.00. As seen in the figure, all three values provide good monotonically
increasing trend. fSCaie 5.00 yields much more aggressive driving behavior in operating modes 30 and 40
compared to/sco;e 4.00, with mean C02 emissions rates approximately 30% higher in these operating
modes. As stated earlier, our objective is to select the/SCate that yields the most aggressive operation in the
highest operating modes while still providing sufficiently robust estimates of the emission rates. Because
fscaie 5.00 still provides mean C02 emissions rates that are robust for the high operating mode bins, /sco;e
5.00 is preferable to an fSCaie of 4.00./sco/e 5.00 provides strong and expected increasing trends in C02
emission rates between operating modes 29 and 30, whereas/sco;e 6.00 does not. We believe this is due to
insufficient data in the highest operating mode bins with/sco/e of 6.00 to determine robust estimates at the
highest operating modes. Thus, we decided to use an fSCaie of 5.00 for LHD in MOVES.
For MHD and HHD, we went through similar reasoning and steps as for LHD. Our final fscaie values for LHD,
MHD, and HHD are 5.00, 7.00, and 10.00, respectively. From Table G-l - Table G-3, these fSCaie values lead
to comparable vehicle count (20-40 % of total vehicles in the regulatory class) and seconds of data (1000-
3000 seconds out of one million) in OpModes 30 and 40.
We did not try to find a precise and even more suitable/sco/e value. Thus, for example, whether fSCaie of 4.80
or 5.20 is better than 5.00, for LHD, was not tested. There are diminishing returns for the extra time and
effort required for that analysis because: (1) the HDIUT data set lacks certain things such as very aggressive
308
-------
operation or malfunctioning vehicles, so a very suitable value of fscaie from this data set might not be as
suitable with another data set; (2) comparing closely spaced fscaie values does not necessarily provide a
clear winner across the board because there's more than one criteria (vehicle count, time, mass/time rates
for various pollutants).
In the 2019 peer-review,15 one of peer-reviewers asked if we have evaluated the time distribution from
real-world data using the proposed fscaie value. In Figure 7-4 through Figure 7-6 we conducted a comparison
of the operating mode distributions measured from the HDIUT dataset and real-world operating modes
estimated from MOVES3 national scale runs for the three evaluated regulatory classes LHD, MHD, and HHD
at the proposed proposed fscaie values. As shown, the distributions are similar between the HDIUT dataset
and national MOVES runs, with most of the data occurring at idle and the operating modes above 50 mph
(opModelD 33-40). One notable difference is for a national scale run MOVES estimates a higher percentage
of activity in the highest power, high speed operating mode bins. This is expected, given that the HDIUT
dataset is expected to under-represent high power operation due to steep grades, high speeds, and heavy-
pay loads (e.g. multiple trailers, over-weight trailers) compared to the in-use fleet. This comparison
supports our logic to select the/SCate that maps only the most aggressive operation from the HDIUT dataset
into the highest MOVES operating mode bins.
The peer-reviewer suggested that we compare the operating mode distribution obtained from the
proposed/sco/e values from other in-use datasets. We agree that this would be useful to better understand
the representativenss of the HDIUT dataset, as well as further evaluate the MOVES default activity
assumptions, including the MOVES heavy-duty driving cycles. In the population and activity report, we
listed this as a project for consideration for future MOVES work.
309
-------
Table G-l Effect of fscaie Value on Vehicle Count and Time for Light Heavy-Duty Vehicles
Number of vehicles1
Number of seconds based on a cycle with one million
seconds1,2
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
OpMode
2.06
4.0
5.0
6.0
9.0
17.1
2.06
4.0
5.0
6.0
9.0
17.1
0
64
64
64
64
64
64
41131
41131
41131
41131
41131
41131
1
64
64
64
64
64
64
358957
358957
358957
358957
358957
358957
11
64
64
64
64
64
64
46235
46235
46235
46235
46235
46235
12
64
64
64
64
64
64
15112
24896
29023
32878
42842
61989
13
64
64
64
64
64
64
10311
14877
16658
18290
20756
16153
14
64
64
64
64
64
43
8109
11394
12304
12431
10835
2387
15
64
64
64
64
63
18
6995
8784
8680
8092
4190
200
16
64
64
64
64
41
1
40203
20778
14065
9039
2107
1
21
64
64
64
64
64
64
45157
45157
45157
45157
45157
45157
22
59
64
64
64
64
64
7153
16366
22038
28166
47554
92823
23
64
64
64
64
64
64
9877
24704
31884
38195
49094
41467
24
64
64
64
64
64
48
12315
25290
28989
30288
26962
6425
25
64
64
64
64
64
22
13307
21400
21274
19931
11739
850
27
64
64
64
64
44
1
25844
28819
25061
18770
5671
1
28
64
64
64
44
22
2
21514
15281
8422
5020
544
5
29
64
64
42
23
0
0
16292
5808
2900
1161
0
0
30
64
43
22
13
2
1
35269
3903
1002
40
5
1
33
64
64
64
64
64
64
26999
42467
55797
73408
142294
267976
35
64
64
64
64
64
46
16820
75707
110862
131366
129751
18220
37
64
64
64
63
44
5
33332
86600
79892
67271
13177
7
38
64
63
62
44
22
1
46641
52178
31818
12100
978
2
39
64
62
41
23
1
1
49374
21416
6167
1940
1
2
40
64
41
23
14
1
1
113054
7852
1685
135
18
13
Notes:
310
-------
1 Values in bold are for final selected fscaie- Shaded cells show instances where using an excessively high/sca/e value
causes data deficit in the higher power operating modes within a speed bin.
2 Number of seconds = Average operating mode time fraction * cycle with one million seconds. The average operating
mode time fraction is the average of the time fraction (for that operating mode) across all vehicles.
MOVES OpMode
Figure G-l Effect of fscaie Value on Coverage and Trends of operating mode Based C02 for Light Heavy-Duty Vehicles
Absolute values of operating mode based emissions rates cannot be compared between series with
different fscaie values.
311
-------
Table G-2 Effect of fscaie Value on Vehicle Count and Time for Medium Heavy-Duty Vehicles
Number of vehicles1
Number of seconds based on a cycle with one million
seconds1,2
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
OpMode
4.0
5.0
6.0
7.0
9.0
17.1
4.0
5.0
6.0
7.0
9.0
17.1
0
58
58
58
58
58
58
36170
36170
36170
36170
36170
36170
1
58
58
58
58
58
58
349622
349622
349622
349622
349622
349622
11
58
58
58
58
58
58
32693
32693
32693
32693
32693
32693
12
58
58
58
58
58
58
20755
23825
26630
29346
34264
48508
13
58
58
58
58
58
58
11234
12492
13612
14397
15495
16683
14
58
58
58
58
58
58
8252
9101
9517
9898
10003
7687
15
58
58
58
58
58
30
6688
7014
7142
7000
6576
1914
16
58
58
58
58
58
16
28357
22855
18385
14645
8948
496
21
58
58
58
58
58
58
44291
44291
44291
44291
44291
44291
22
58
58
58
58
58
58
10132
14200
18827
23955
34436
78404
23
58
58
58
58
58
58
19195
25479
31609
37379
48401
49534
24
58
58
58
58
58
58
21109
27513
32401
34297
30771
19215
25
58
58
58
58
58
37
22136
24498
22647
20428
17498
5111
27
58
58
58
58
58
16
32912
28713
25621
22502
17220
1306
28
58
58
58
58
25
0
18677
15863
13842
10424
4619
0
29
58
58
46
25
10
0
12105
10075
5243
3550
625
0
30
58
39
25
14
0
0
17304
7229
3379
1034
0
0
33
58
58
58
58
58
58
37996
45307
54057
64727
92859
212774
35
58
58
58
58
58
58
40077
63649
87949
107234
130599
92395
37
58
58
58
58
57
15
63932
76693
81453
80818
72135
3197
38
58
58
57
49
30
1
56697
58308
51520
44843
11144
1
39
58
53
43
25
10
1
45255
39067
25466
8307
1628
1
40
53
39
26
14
1
0
64411
25344
7923
2440
2
0
Notes:
312
-------
1 Values in bold are for final selected fscaie- Shaded cells show instances where using an excessively high/sca/e value
causes data deficit in the higher power operating modes within a speed bin.
2 Number of seconds = Average operating mode time fraction * cycle with one million seconds. The average operating
mode time fraction is the average of the time fraction (for that operating mode) across all vehicles.
MOVES OpMode
Figure G-2 Effect of fscaie Value on Coverage and Trends of operating mode Based C02 for Medium Heavy-Duty
Vehicles
313
-------
Table G-3 Effect of fscaie Value on Vehicle Count and Time for Heavy Heavy-Duty Vehicles
Number of vehicles1
Number of seconds based on a cycle with one million
seconds1,2
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
fs
OpMod
e
9.0
10.
0
11.
0
12.
0
14.
0
17.
1
9.0
10.0
11.0
12.0
14.0
17.1
0
159
159
159
159
158
159
18010
18010
18010
18010
18010
18010
1
159
159
159
159
159
159
29766
2
29766
2
29766
2
29766
2
29766
2
29766
2
11
159
159
159
159
158
159
37453
37453
37453
37453
37453
37453
12
159
159
159
159
158
159
24580
25976
27238
28427
30556
33336
13
159
159
159
159
158
159
9472
9547
9576
9576
9551
9524
14
159
159
159
159
158
159
5545
5557
5579
5533
5446
5159
15
159
159
159
159
158
159
3938
3840
3762
3704
3456
2892
16
159
159
159
159
153
152
10041
8657
7422
6337
4568
2666
21
159
159
159
159
158
159
32325
32325
32325
32325
32325
32325
22
159
159
159
159
158
159
12785
14388
15951
17580
20814
25721
23
159
159
159
159
158
159
14276
15457
16513
17395
18748
20117
24
159
159
159
159
158
159
11401
11865
12272
12453
12761
12587
25
159
159
159
159
158
159
8967
9058
9085
9044
8501
8402
27
159
159
159
159
153
154
12410
11927
11767
12026
13569
15804
28
154
154
154
153
134
27
8660
9619
10875
12433
8684
632
29
153
142
122
75
8
0
8905
9822
6562
2329
185
0
30
114
59
10
1
0
0
5861
1127
239
4
0
0
33
159
159
159
159
158
159
11421
4
12621
6
13973
1
15410
1
18609
4
23796
0
35
159
159
159
159
158
159
13910
9
16066
7
17614
4
18613
1
18981
3
17611
1
37
159
159
159
159
153
153
11505
0
10244
0
91446
83420
74122
61219
314
-------
38
154
154
153
152
131
26
55279
52010
50633
47483
27291
2421
39
152
138
122
83
9
0
37885
33033
19207
6576
391
0
40
114
65
11
1
0
0
16174
3344
550
0
0
0
Notes:
1 Values in bold are for final selected fscaie- Shaded cells show instances where using an excessively high/sca/e value
causes data deficit in the higher power operating modes within a speed bin.
2 Number of seconds = Average operating mode time fraction * cycle with one million seconds. The average operating
mode time fraction is the average of the time fraction (for that operating mode) across all vehicles.
MOVES OpMode
Figure G-3 Effect offscaie Value on Coverage and Trends of operating mode Based C02 for Heavy Heavy-Duty
Vehicles
315
-------
1
HHDD Opmode Distributions
Comparison of HDIUT vs. MOVES-CTI
¦ HDIUT ¦ MOVES CTI - CY2027/MY2014
r\ o n i i i
W.
0.30
c 0.25
O
2 0.20
Li-
CU
o 0.15
E
Q.
o 0.10
0.05
n on ll
ll
1
-1
1
m
.1
II J - . -
ll ¦¦ ¦¦ ¦¦ ¦¦ ¦¦ ¦. ¦_
1 il I
0 1
11 12 13 14 15 16
21 22 23 24 25 27 28 29 30
Opmode
33 35
37 38 39 40
3 Figure 7-4. Heavy Heavy-Duty Diesel Operating Mode Distribution compared between the Heavy-Duty In-Use Testing (HDIUT) Program and from MOVES3
4 for a MY 2014 vehicle with an fscaie of 10 metric tons
5
6
316
-------
0.45
0.40
0.35
c
.2 0.30
fo,s
CL>
"E °-20
E
Q- 0.15
O
0.10
0.05
0.00
MHDD Opmode Distributions
Comparison of HDIUT vs. MOVES-CTI
¦ HDIUT ¦ MOVES CTI - CY2027/MY2014
i
1
L
i
i
i
i
1
j_
i
I.
II
il 11 ¦¦ ¦¦ __ ¦¦
li ii ¦¦ .. _
Il .. .
11 12 13 14 15 16
21 22 23 24 25 27 28 29 30 | 33 35 37 38 39 40
Opmode
2 Figure 7-5. Medium Heavy-Duty Diesel Operating Mode Distribution compared between the Heavy-Duty In-Use Testing (HDIUT) Program and from MOVES3
3 for a MY 2014 vehicle with an fSCaie of 7 metric tons
317
-------
0.45
0.40
0.35
c
° 0.30
I 0 25
CD
"8 020
Q- 0.15
O
0.10
0.05
0.00
LHDD Opmode Distributions
Comparison of HDIUT vs. MOVES-CTI
¦ HDIUT ¦ MOVES CTI - CY2027/MY2014
!
-1
1 |
ll ll
II
II ll
la ll ll II ¦¦ I- _
ll II II II ..
1 i 11 12 13 14 15 16 i 21 22 23 24 25 27 28 29 30 ; 33 35 37 38 39 40
Opmode
Figure 7-6. Light Heavy-Duty Diesel Operating Mode Distribution compared between the Heavy-Duty In-Use Testing (HDIUT) Program and from M0VES3SI
for a MY 2014 vehicle with an fscaie of 5 metric tons.
ss The MOVES operating mode distribution excludes class 2b light-heavy-duty vehicles in the passenger truck and light-commercial truck source types The
vehicles included in the HDIU are all engine-certified vehicles and are class 3 or heavier.
318
-------
Appendix H THC and CO Emisssion rates from 2010 and Later Model
Year Heavy-duty Vehicles from the HDIUT
H.l Comparison of THC and CO Emission Rates by NOx FEL Groups for
MY 2010-2013 Vehicles in LHD and MHD
As noted in 2.1.3.2, this appendix section contains figures of the mean THC and CO emission rates by NOx
FEL Group for LHD and MHD. The figures for HHD are included in the main report.
H.l.l LHD
0.007
0.006
0.005
< 0.004
3B
I 0.003
I-
0.002
0.001
0.000
I LHD FEL 0.20, MY 2010-2013
I LHD FEL 0.35, N=0
I LHD FEL 0.50, MY 2010-2015
N=52
N=15
j_I_i
jjjjj
ill
i« i I I
iiiili
1
0 1: 11 12 13 14 15 16; 21 22 23 24 25 27 28 29 30, 33 35 37 38 39 40
MOVES OpMode
Figure 7-7 Average LHD THC Emission Rates by Operating Mode for the 0.2 NOx FEL for MY 2010-2013 and the 0.5
NOx FEL for MY 2010-2016. Error Bars are 95% Confidence Intervals of the Mean
319
-------
0.04
0.03
0.03
< 0.02
3B
3 0.02
0.01
0.01
0.00
Figure 7-8 Average LHD THC Emission Rates by Operating Mode for the 0.2 NOx FELfor MY 2010-2013 and the 0.5
NOx FEL for MY 2010-2016. Error Bars are 95% Confidence Intervals of the Mean
111
il
I LHD FEL 0.20, MY 2010-2013, N=52
I LHD FEL 0.35, N=0 I
I LHD FEL 0.50, MY 2010-2015, N=15
0 1 , 11 12 13 14 15 16 , 21 22 23 24 25 27 28 29 30 , 33 35 37 38 39 40
MOVES OpMode
H.1.2 MHD
1 !
I MHD FEL 0.20, MY 2010-2013, N=23
i MHD FEL 0.35, MY 2010-2013, N=23
I MHD FEL 0.50, MY 2010-2013, N=9
u
MJj
i
Jl
11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 7-9 Average MHD THC Emission Rates by Operating Mode for the 0.2, 0.35 and 0.50 NOx FEL Groups for MY
2010-2013 Vehicles. Error Bars are 95% Confidence Intervals of the Mean
320
-------
MOVES OpMode
Figure 7-10 Average MHD CO Emission Rates by Operating Mode for the 0.2, 0.35 and 0.50 NO* FEL Groups for MY
2010-2013 Vehicles. Error Bars are 95% Confidence Intervals of the Mean
321
-------
H.2 Comparison of THC and CO Emission Rates between MY 2010-
2013 and MY 2014 in the 0.2 N0X FEL Group for LHD and MHD
Figure 7-11 through Figure 2-37 show the comparison of the mean THC and CO emission rates between
model year 2010-2013 and 2014 and later vehicles in the 0.2 NOx FEL Group for the LHD and MHD
regulatory class by operating mode. The Figures for HHD are located in the main report.
H.2.1 LHD
0.0040
0.0035
--- 0.0030
i/>
60
— 0.0025
U
I- 0.0020
0.0015
0.0010
0.0005
0.0000
Figure 7-11 THC emission rates for the MY 2010-2013 and MY 2014-2016 vehicles in the LHD 0.20 NOx FEL Group
¦ MY 2010-2013
1
¦ MY 2014-2016
T
ii
i i
t I
i i 1
T
11
I 1
1
.. ii
.. ii Ii li 1
1
ii ii Ii ii Ii Ii
L i, i,
. i 1
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
322
-------
0.08
0.07
0.06
_ 0.05
CO
-22 0.04
O
U 0.03
0.02
0.01
0.00
Figure 7-12 CO emission rates for the MY 2010-2013 and MY 2014-2016 vehicles in the LHD 0.20 NOx FEL Group
H.2.2 MHD
¦ MY 2010-2013
¦ MY 2014-2016
] T,
,
* T
r
ii is ii i1 i1 i 1
ii ii ii ii i il i
it i i 1
lil
i
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
0.0014
0.0012
^ 0.0010
-2S
^ 0.0008
I-
0.0006
0.0004
0.0002
0.0000
Figure 7-13 THC emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the MHD 0.20 NO* FEL Group
¦ MY 2010-2013
¦ MY 2014-2015
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
323
-------
0.014
0.012
0.010
< 0.008
-2S
3 0.006
0.004
0.002
0.000
Figure 7-14 CO emission rates for the MY 2010-2013 and MY 2014-2015 vehicles in the MHD 0.20 NOx FEL Group
MY 2010-2013
MY 2014-2015
ii
it
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
324
-------
Appendix I Analysis of 2010 and Later Model Year Heavy-duty Gasoline
Emission Rates
This appendix contains additional details of the analysis used to update the 2010 and later model year
heavy-duty gasoline running emission rates documented in Section 3.1.1.2.1.
1.1 Removal of Start Emissions from Real-World PEMS Data in Developing
Heavy-duty Gasoline Running Exhaust Emissions
The running exhaust emissions rates update for 2010 model year and later heavy-duty gasoline vehicles is
meant to include emissions from only the hot-running condition. Thus, ideally, emissions assigned to start
effects should be removed before estimating operating mode-based average rates per test and per vehicle.
This is less of a concern if each test is a full-day of operation since the incremental start emissions might
then be a small fraction of total emissions. However, on-road tests of the three HD gasoline involved drive
cycles that range from 10 to 90 minutes in duration. Also, the idle tests, of 15 or 30 minute duration, need
to have start effects removed to ensure their contribution to OpMode 1 (idle mode) rate is unaffected by
start emissions. Note that the effect of start emissions is modeled as a separate process in MOVES and by
removing them from the running emissions, we are minimizing double-counting.
Start emissions in the Federal Test Procedure are calculated as Bag 1 minus Bag 3 of the FTP cycle, where
Bag 1 is driving after a cold start and Bag 3 is the same cycle as Bag 1 but under hot-stabilized conditions.
This method is not possible in real-world testing because it is not possible to replicate the exact drive cycle
due to varying traffic conditions. Thus, we decided to define start emissions as the incremental emissions
that occur before the TWC reaches the light-off condition where it achieves optimal emissions reduction
efficacy. We define light-off condition as the point when the TWC first reaches 421 °C (790 °F). TWC light-
off temperatures are based on design specifics but are generally in the range of 400 °C. The selection of
421 °C as the criteria is somewhat arbitrary at the very precise level - there is not a good reason why 421
°C is more appropriate than say 410 °C or 430 °C. We picked 421 °C based on visual comparison of a
handful of the on-road tests for each of the three gasoline vehicles to find out at what point the TWC
temperature starts to stabilize. The effect of soak time on time to reach 421 °C catalyst temperature and
grams of emissions assigned to the start effect, thus removed from running exhaust emissions, are shown
in Table 7-4. Figure 7-15 shows the data for NOx. Interestingly, the trend for NOx from on-road testing is
comparable to the trends from previous lab-based testing, shown in Figure 3-23. For the on-road data,
grams of NOx from starts emissions for 105-minute soak is 1.15 times the 720-minute soak. For the same
conditions, the ratios in Figure 3-23 are approximately 1.17 and 1.37 for the data series labeled as
"MOVES" and "New Data", respectively. The trends for THC and CO are also similar between the two
figures.
325
-------
Table 7-4 Time and Pollutant Mass for Driving Assigned to Start Emissions
Soak
Time
(min)
Number
of Tests
Avg. time1 for TWC
to reach 421 °C (sec)
Avg. grams of pollutant removed
NOx
o
u
CO
THC
0
109
78
0.2
356
3
0.4
3
6
42
0.02
213
1
0.03
18
6
63
0.1
265
3
0.3
30
6
91
0.8
427
9
0.8
45
8
114
1.9
493
14
1.6
75
5
122
1.8
470
16
1.8
105
7
102
2.3
463
19
1.9
180
4
107
3.0
531
22
2.7
240
2
94
1.1
424
18
1.9
360
1
1
0.00
0
0
0.00
720
48
125
2.0
662
25
3.3
Note:
1 Of the total 202 tests listed here, in three tests the catalyst never reached 421 °C, so they are not
included in the average time calculation, however, the grams of pollutant removed columns include these
three tests.
326
-------
0.5
o.o
100 200 300 400 500
SoakTime (mins)
600
700
800
Figure 7-15. Grams of NOx from Start Emissions versus Soak Time
327
-------
1.2 Comparison of Heavy-duty Gasoline Emission Rates by Vehicle
Figure 7-16 through Figure 7-18 show the emission rates calculated from the PEMS testing data collected
on three heavy-duty gasoline vehicle. The emission rates in these figures were analyzed using an/sca;eof 5
metric tons used for the LHD2b3 and LHD45 regulatory class emission rates The error bars are the 95%
confidence intervals of the mean calculated by treating the number of routes (R) as independent random
variables.
The Isuzu NPR has the highest NOx emission rates across all operating modes. The Ford E459 and the Ram
3500 have more similar NOx emission rates, except for the high speed and power operating modes
(opModelD 39 and 40).
0.05
0.04 ¦
0.03 -
x
O
0.02 -
0.01 -
0.00
¦ 2016_Ford_E459, R=76
¦ 2015_|SUZU_NPR, R=83
¦ 2017 RAM 3500, R=38
-i- il
ii
i ,i- ii. ii.
I
i
0 1 : 11 12 13 14 15 16: 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure 7-16. Mean Heavy-duty Gasoline NOx Emission Rates by Operating Mode and Vehicle Calculated using/JCO(eof
5 metric tons.
For THC and CO, the Isuzu NPR tends to have lower emission rates. The differences between the vehicles is
more dependent on operating mode. The Ram 3500 has the highest emission rates for the high STP and
high speed operating modes, but is more comparable to the other vehicles at the low STP and low speed
operating mode bins.
328
-------
0.045
0.040
0.035
0.030
IS 0.025
^ 0.020
I-
0.015
0.010
0.005
0.000
' ¦ 2016_Ford_E459, R=76
¦ 2015_lsuzu _NPR, R=83
¦ 2017_RAM_3500, R=38
U- a.
1 ni J
21 22 23 24 25 2 7 28 29 30 33 35 37 38 39 40
i-- I
h. ii. ii- i*i i»j
0 1 : 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 i 33 35 37 38 39 40
MOVES OpMode
Figure 7-17. Mean Heavy-duty Gasoline THC Emission Rates by Operating Mode and Vehicle Calculated using/sco/eof
5 metric tons.
9.00
8.00
7.00
6.00
IS 5.00
8 4.00
3.00
2.00
1.00
0.00
0.60
0.50
0.40 :
0.30 j
0.20 ¦
0.10 :
0.00
¦ 2016_Ford_E459, R=76
¦ 2015_lsuzu_NPR, R=83
¦ 2017_RAM_3500, R=38
¦ i HI
Miii
11 12 13 14 15 16
ill
21 22 23 24 25 27 28 29 30
jJJi
33 35 37 38 39 40
I ¦
0 1 11 12 13 14 15 16 i 21 22 23 24 25 27 28 29 30 ! 33 35 37 38 39 40
MOVES OpMode
Figure 7-18. Mean Heavy-duty Gasoline CO Emission Rates by Operating Mode and Vehicle Calculated using/sco/eOf
5 metric tons.
329
-------
1.3 Extrapolating High-Power Operating Modes
When analyzing the heavy-duty gasoline data for the MHD and HHD regulatory classes (using an/scafeof 7
and 10 metric tons respectively), there was limited or no data for high power operating mode bins. Figure
7-19 shows the mean NOx emission rates by vehicle and operating mode when using an using an/SC£J/eof 10
used for HHD vehicles. Operating modes 29, 30, 39, 40 are missing data from at least one of the tested
vehicles. Operating mode 16 and 28 have limited data from the Ford and Isuzu vehicles (less than 10 routes
had data measured in those operating mode bins). The small amount of data in these operating mode bins
decreases our confidence in the mean operating modes, and can lead to inconsistent trends in emission
rates with power. For example, operating mode 16 has lower mean NOx emission rates than operating
mode 15.
0.05
0.04
0.03
.25
x
O 0.02
0.01
0.00
12016_Ford_E459, R=80
I 2015_lsuzu_NPR, R=84
12017 RAM 3500, R=38
_iL
fj
Ji
K- ifti
o 1
11 12 13 14 15 16
21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 7-19. Mean Heavy-duty Gasoline NQX Emission Rates by Operating Mode and Vehicle Calculated for HHD
using an/sCO/eOf 10 metric tons
In the case of missing data, we used the emission rates of the nearest operating mode bin with data. In
case where there were limited data, we aggregated the averages with data from the next closest bin into
into single averages. Figure 7-20 displays the resulting emission rates by vehicles for NOx for HHD vehicles.
For the Ford and Isuzu vehicles, the following operating modes were aggregated into single averages: 15-
16, 27-30, 38-40. For the RAM vehicle, operating modes 29-30 were aggregated. The same aggregation was
used to calculated the updated THC and CO emission rates.
330
-------
12016_Ford_E459, R=80
l 2015_lsuzu_NPR, R=84
I 2017_RAM_3500, R=38
-.1-i.iiii iiL J.
JLi
Job
J11 III
0 1 I 11 12 13 14 15 16 ; 21 22 23 24 25 27 28 29 30 j 33 35 37 38 39 40
MOVES OpMode
Figure 7-20. Mean Heavy-duty Gasoline NO* Emission Rates by Operating Mode and Vehicle Calculated for HHD
usinga fscaie of 10 metric tons with Aggregated Means for High Power Bins with Limited Data
Figure 7-21 shows the weighted average NOx emission rate calculated by averaging the three vehicles
together according to their production volume sales.
0.040
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0.000
I HHDG, MY 2016
.III
¦ III
• I
0 1
11 12 13 14 15 16
21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 7-21. Weighted Average Heavy-duty Gasoline NOx Emission Rates by Operating Mode for HHD Using
Production Volumes
331
-------
For C02 emission rates, we used a different method than forTHC, CO, and NOx. Rather than aggregate the
emission rates with limited or no data, we extrapolated the higher operating modes (30, 39, and 40) using
the STP values using Equation 2-5. The assumed mid-point STP for each operating mode bin is displayed in
Table 7-5.
332
-------
Table 7-5. Assumed STP Midpoint for Each Operating Mode
OpModelD
STP_midpoint
0
-
1
-
11
-
12
1.5
13
4.5
14
7.5
15
10.5
16
13.5
21
-
22
1.5
23
4.5
24
7.5
25
10.5
27
15
28
21
29
27
30
33
33
3
35
9
37
15
38
21
39
27
40
33
Figure 7-22 displays the initial mean C02 emission rates using an/sctJ;eof 10 metric tons. Note that the
emission rates for C02 are more well behaved than the criteria pollutants, because of this, less of the high
333
-------
power operating modes were replaced with extrapolated rates, than were binned for the THC, CO, and NOx
emission rates. In this case, we decided to only replace one operating mode with limited data (operating
mode 39 for the Isuzu vehicle), because it was only based on one route (which is why there are no error
bars).
I'l I'l
0 1
I 2016_Ford_E459, R=80
I 2015_lsuzu_NPR, R=84
l 2017_RAM_3500, R=38
III IB
11 12 13 14 15 16
a
21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 7-22. Mean Heavy-duty Gasoline C02 Emission Rates by Operating Mode and Vehicle Calculated for HHD
using an/sra/eof 10 metric tons
Figure 7-23 shows the mean C02 emission rates using an fscaie of 10 metric tons with extrapolated emission
rates for the high power bins. For the Ford and Isuzu vehicles operating modes 29 and 30 were
extrapolated from operating mode 28, and operating modes 39 and 40 were extrapolated from 38. For the
RAM, operating mode 30 was extrapolated from 29.
334
-------
,52
(N
o
u
90
80
70
60
50
40
30
20
10
0
-¦ p-i
I 2016_Ford_E459, R=80
I 2015Jsuzu_NPR, R=84
I 2017_RAM_3500, R=38
II1
il
I
ML
n li
0 1 , 11 12 13 14 15 16 | 21 22 23 24 25 27 28 29 30
MOVES OpMode
33 35 37 38 39 40
Figure 7-23. Mean Heavy-duty Gasoline C02 Emission Rates by Operating Mode and Vehicle Calculated for HHD
using an/sco/eOf 10 metric tons with Extrapolated Means for High Power Bins.
Figure 7-24 shows the weighted average C02 emission rate calculated by averaging the three vehicles
together according to their production volume sales. Note that the C02 have stronger increasing trends
with power compared to NO*.
80
60 -
" 40 -|
CM
O
u
20 -
I HHDG, MY 2016
X
X
l
ll
0 1 I 11 12 13 14 15 16 i 21 22 23 24 25 27 28 29 30 I 33 35 37 38 39 40
MOVES OpMode
Figure 7-24. Weighted Average Heavy-duty Gasoline C02 Emission Rates by Operating Mode for HHD Using
Production Volumes
Similar calculations were repeated for the MHD vehicles calculated using an/sca/eof 7 metric tons.
335
-------
Appendix J Comparing Glider Vehicle and MOVES Model Year 2000
Heavy Heavy-Duty Emission Rates
Glider vehicles are new vehicles with older engines. Tailpipe exhaust emissions were measured from two
heavy heavy-duty glider vehicles, one each of MY 2016 and MY 2017, at the US EPA's National Vehicle Fuel
and Emissions Laboratory (NVFEL) heavy-duty chassis dynamometer. Both glider vehicles were sleeper cab
tractors and lacked EGR, DOC, DPF, or SCR emissions control devices. Each glider vehicle was tested on
three drive cycles - (1) cold start Heavy-Duty Vehicle Urban Dynamometer Driving Schedule (UDDS)
sequence; (2) World Harmonized Vehicle Cycle (WHVC) sequence; (3) Super Cycle (a combination of urban
and freeway driving). The vehicles were tested with two sets of road-load coefficients representing 60,000
lbs and 80,000 lbs total vehicle weight. The MOVES operating mode based results shown here are time-
weighted average of the various drive cycle and weight combinations. More details about the vehicles,
drive cycles, test procedures, and results are available in the EPA report for the glider test program.162 The
data that was analyzed for MOVES included only the tests where the vehicles were operating free of any
malfunction indicator lights because MOVES treats tampering and malmaintenance separately.
Since the measured glider emission rates are from only two glider vehicles tested in lab conditions, we
decided not to use those rates as-is to estimate fleet-wide real-world glider vehicle emissions from MOVES.
Instead, we used the data to identify which model year emission rates in MOVES can best represent
contemporary glider vehicles. We know that contemporary glider vehicles typically have re-built
powertrains that do not require DOC, DPF, and SCR controls, and likely lack even EGR systems. This meant
glider vehicles should compare well with MY 2006 or older emission rates. Thus, we compared the
emission rates from the two glider vehicles with regClass 47 (HHD) rates for applicable model year groups
in MOVES. MOVES defines emission model year group by pollutant and has emissions rates for one or
more model year groups within the MY 1998-2006 period. For example, in MOVES, there are different NOx
rates for MY 1998, MY 1999-2002, and MY 2003-2006. On the other hand, PM25 and C02 (from energy)
rates are the same across MY 1998-2006.
A comparison of glider test results versus MOVES base emissions rates for NOx, THC, CO, C02, and PM25
are presented in Figure J-l though Figure J-6. Given that glider vehicles lack emissions control devices that
provide significant reductions in NOx and PM2 5, our primary focus was to find the best model year match
for these pollutants, with THC and CO as secondary comparison to pick between model years that match
equally well for NOx and PM2 5. The decision to pick a representative model year mostly came down to NOx
because emissions rates for PM2 5 are the same for MY 1998-2006. From Figure J-l, it is seen that the NOx
emissions from the measured glider vehicles are best represented by the MY 1999-2002 rates in MOVES.
Similarly, for THC (Figure J-2), MOVES MY 1998-2002 rates are a better match compared to the MY 2003-
2006 rates. The CO rates from the measured glider vehicles are significantly higher than MOVES (Figure
J-3), so it is likely that modeled CO emissions for glider vehicles in MOVES will be underestimated. The C02
rates from glider vehicles are comparable to the MOVES MY 1998-2006 rates (Figure J-4). When comparing
the rates between measured glider vehicles and MOVES, we paid attention to differences in the
336
-------
comparison across operating modes. Typically, within each speed-bin, vehicles spend more time in and
thus generate a larger fraction of total emissions from the low and medium power operating modes versus
high-power operating modes. Using NOx and C02 as examples, the comparison between measured glider
vehicle versus MOVES MY 2000 emission rate for operating modes 27 and 28 are not as good as lower
operating modes. However, the difference in these high power operating modes has a lower impact on
total inventory because vehicles spend less time in the operating mode. For PM2.5, we have gravimetric
filter-based measurement for both gliders and instantaneous measurement using an AVL Micro-Soot
Sensor (MSS) for only Glider-2, as seen in Figure J-5 and Figure J-6. We applied MOVES operating mode
based PM2.5 rates to the operating mode based time distribution for Glider-1 and Glider-2 test cycles to
estimate cycle total emissions for MOVES and compared those to the total PM2.5 from the gravimetric
filter measurements. While Glider-2 filter and instantaneous rates are lower than MOVES MY 1998-2006
rates, the comparison for Glider-1 is much closer. Overall, we believe MOVES MY 1998-2006 PM2.5 rates
are reasonable for use as glider vehicle PM2.5 rates because they are derived from vehicles without DPF, a
key PM2.5 emissions control device that is absent in most glider vehicles.
Based on these comparisons, we decided to apply the MY 2000 HHD (regClass 47) emission rates for THC,
CO, NOx, PM2.5, and Energy (for C02) in MOVES to glider vehicles (regClass 49).
M
x
O
5000
4500
4000
3500
3000
2500
2000
1500
1000
500
0
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
-A- Glider-1
-B Glider-2
MOVES MY1998
-0- MOVESMY1999-2002
-O MOVES MY2003-2006
Figure J-l Comparison of NOx Emission Rates from Glider Vehicle Lab Testing to MOVES regClass 47 (HHD) 1998-
2006 Model Years
337
-------
MOVES OpMode
Figure J-2 Comparison of THC Emission Rates from Glider Vehicle Lab Testing to MOVES regClass 47 (HHD) 1998-
2006 Model Years
900
800
700
— 600
-C
22 500
o
u
400
300
200
100
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
—A- Glider-1
-B- Glider-2
—0—MOVES_MY1998-2002
-O- MOVES_MY2003-2006
¦©—0--€r"°
Figure J-3 Comparison of CO Emission Rates from Glider Vehicle Lab Testing to MOVES regClass 47 (HHD) 1998-2006
Model Years
338
-------
400000
350000
300000
£ 250000
M
g 200000
u
150000
100000
50000
0
-A- Glider-1 q.
-B Glider-2 /I
-e-MOVES_MY1998-2006 J \
*51 \
0 1 11 12 13 14 15 16 21 22 23 24 25 27 28 29 30 33 35 37 38 39 40
MOVES OpMode
Figure J-4 Comparison of C02 Emission Rates from Glider Vehicle Lab Testing to MOVES regClass 47 (HHD) 1998-
2006 Model Years
k Glider 1 O MOVES_1998-2006
¦ Glider 2 OMOVESJL998-2006
30
=• 25
re
2 20
V
s- 15
¦S 10
2
o- 5
0
o
~
O
o
o
~
O
~
o
60,000 lbs
V
V ^ N-
80,000 lbs
30.0
25.0
re
20.0
O
_aj
15.0
>•
10.0
s22
5
5.0
Q.
0.0
o
O
¦
o
o
o
o
¦ ¦
¦ ¦
v
60,000 lbs
80,000 lbs
Figure J-5 Comparison of Gravimetric Filter-based PM2.5 Emissions from Glider Vehicle Lab Testing to MOVES
regClass 47 (HHD) 1998-2006 Model Years
339
-------
MOVES OpMode
re J-6 Comparison of Instantaneous PM2.5 Emissions from Glider-2 Lab Testing to MOVES regClass 47 (HHD)
1998-2006 Model Years
340
-------
Appendix K PM Composition Measurements from Auxiliary Power
Units
Table 7-6 reports the organic carbon (OC), elemental carbon (EC) and total carbon (TC) measurements
conducted in the study conducted by Texas Transportation Institute (TTI, 201494). All the measurements
were collected on APU 1. TTI collected the particulate sample on quartz fiber filters, and Sunset Laboratory
Inc. analyzed the filters using thermal optical reflectance (TOR) using the IMPROVE (Interagency
Monitoring of Protected Visual Environments) procedures. Total Carbon (TC) is the sum of Elemental
Carbon (EC) and Organic Carbon (OC).
Table 7-6. Organic Carbon, Elemental Carbon, and Total Carbon Measurements from the IMPROVE_TOR measured
on APU 1
Sample ID
Min.
DR
Test
OC
OC
uncertainty
EC
EC
uncertainty
TC
TC
uncertainty
EC/TC ratio
(Hg/cm2)
(Hg/cm2)
(Hg/cm2)
(Hg/cm2)
(Hg/cm2)
(Hg/cm2)
APU_005
10
30/1
Hot Test 1
66.35
3.42
12.98
0.75
79.33
4.17
0.16
APU_006
10
30/1
Hot Test 2
65.26
3.36
13.45
0.77
78.70
4.14
0.17
APU_007
10
30/1
Hot Test 3
59.24
3.06
10.51
0.63
69.75
3.69
0.15
APU_009
20
6/1
DPF Hot APU 1
13.85
0.79
0.86
0.14
14.71
0.94
0.06
APU_010
20
6/1
DPF Hot APU 1
14.67
0.83
1.12
0.16
15.79
0.99
0.07
APU_011
20
6/1
DPF Hot APU 1
13.18
0.76
0.93
0.15
14.11
0.91
0.07
APU_012
20
6/1
DPF Cold APU 1
16.62
0.93
1.45
0.17
18.07
1.10
0.08
APU_013
20
6/1
DPF Cold APU 1
15.86
0.89
1.40
0.17
17.27
1.06
0.08
APU_014
20
6/1
DPF Cold APU 1
17.59
0.98
1.56
0.18
19.15
1.16
0.08
APU_015
10
30/1
Cold Test 1
75.74
3.89
9.65
0.58
85.39
4.47
0.11
APU_016
10
30/1
Cold Test 2
73.83
3.79
9.61
0.58
83.44
4.37
0.12
APU_017
10
30/1
Cold Test 3
77.47
3.97
9.90
0.59
87.37
4.57
0.11
341
-------
Appendix L
Crankcase Emissions from Glider Trucks
The MOVES CrankcaseEmissionRatio table stores crankcaseEmissionRatios (ratio of crankcase to tailpipe
exhaust) by fuel type, source type and model year. MOVES applies the crankcase emission ratios to tailpipe
exhaust emissions (start exhaust, running exhaust, and extended idle exhaust) by fuel type, source type,
and model year. In the current MOVES structure, MOVES is unable to apply separate crankcase emission
ratios by regulatory class, including to distinguish crankcase emission ratios between glider trucks
(regulatory class ID 49) and other heavy-duty regulatory classes (HHD, MHD, LHD45, LH2b3).
Because glider trucks as defined by MOVES to use pre-2007 engines and emission control technologies, the
crankcase emissions should also use the pre-2007 crankcase emission ratios. Due to the higher tailpipe
exhaust emission rates for pre-2007 vehicles, the crankcase emission ratios are lower for THC, CO, and
NOx, but comparable for total PM2.5 emissions. (See Table 5-3, Table 5-8 and Table 5-16). The crankcase
emission ratios are quite different for the PM2.5 species, including elemental carbon, because the pre-2007
diesel trucks have much higher EC/PM fractions in their exhaust, whereas the EC/PM fraction for crankcase
emissions is low.
In developing the crankcase emission ratios for MY 2007 and later heavy-duty trucks, we did not consider
the impact of glider trucks. Developing crankcase emission ratios in the current structure would have made
the crankcase ratios much more complex and quality assure. For example, rather than having constant
crankcase emissions for model year groups and across source types, the emission ratios would need to
vary for each model year and source type according to the population of glider trucks. Also, for some use
cases of MOVES, it may not be desired to have the nationally representative population of glider trucks,
incorporated into the crankcase emission ratios.
This appendix shows the impact of this choice by comparing glider crankcase emissions calculated by
MOVES3 and a "Sensitivity Case," glider crankcase emissions calculated by post-processing the MOVES3
output to apply pre-2007 HD diesel crankcase ratios to the glider emissions. Table K-l displays the percent
differences between MOVES3 and the "Sensitivity Case." As expected MOVES3 overestimates the
emissions from gliders for THC, CO,and NOx for most cases. The total PM2.5 values are quite similar,
however the elemental carbon (EC) compoment of PM2.5 is overestimated substantially.
Table K-l also displays the percent difference for crankcase emissions aggregated across all heavy-duty
source types. As expected, the overestimation of crankcase emissions is much less when considering the
other regulatory classes, limited to 30% overestimation for THC and NOx, and differences of less than one
percent for CO and PM2.5- When running, start and all other exhaust processes are considered for heavy-
duty diesel trucks the impact is less than 3% for THC emissions, and lessthan 0.2% for CO, NOx, and PM2.5
342
-------
Table L-l. Percent Differences Between M0VES3 and Sensitivity Case Applying pre-2007 Crankcase Ratios to Glider
Emissions by Process and Regulatory Classes
Year
Process and
Regulatory Class
% Differences between MOVES3 and Sensitivity Case
THC
CO
NOx
PM2.5
EC
2015
Crankcase: gliders
548%
160%
1478%
-4%
4979%
2020
550%
15%
1506%
0%
5316%
2030
554%
-19%
1521%
0%
5446%
2045
557%
-26%
1525%
-2%
5456%
2015
Crankcase: all HD
diesel regulatory
classes
6%
1%
18%
0%
139%
2020
18%
0%
22%
0%
500%
2030
29%
-1%
19%
0%
833%
2045
24%
-1%
15%
-1%
667%
2015
All exhaust
processes: all HD
diesel regulatory
classes
0.4%
0.0%
0.1%
0%
1%
2020
1.5%
0.0%
0.1%
0%
3%
2030
2.7%
0.0%
0.2%
0%
12%
2045
2.7%
0.0%
0.2%
0%
18%
Table K-l shows that Elemental Carbon (EC) is the pollutant most impacted by the way we account for the
pre-2007 crankcase ratios for gliders. In current and historic years, the impact is less than 3%, but for
future years, MOVES is overestimating EC emissions by up to 18% in CY 2045.
Table K-2 displays the EC/PM fraction between the MOVES3 and Sensitivity Case for different groups of
processes and regulatory classes. As shown, MOVES3 estimates much higher EC/PM fractions for crankcase
emissions from gliders (66% to 71%) than is expected in the sensitivity case (1%). When aggregated to all
HD regulatory classes and exhaust emission processes, this causes the EC/PM fraction to be slightly higher
in MOVES3 than in the sensitivity case (1% higher in 2015 to 6% higher in 2045).
343
-------
Table L-2. EC/PM Frations from MOVES3 and Sensitivity Case Applying pre-2007 Crankcase Ratios to Glider
Emissions by Process and Regulatory Classes
Year
Process and
Regulatory Class
EC/PM
MOVES 3
Sensitivity Case
2015
HD diesel
crankcase: non-
gliders
2%
2%
2020
2%
2%
2030
5%
5%
2045
9%
9%
2015
HD diesel
crankcase:
gliders
66%
1%
2020
67%
1%
2030
69%
1%
2045
71%
1%
2015
HD diesel
crankcase: all
regulatory
classes
4%
2%
2020
11%
2%
2030
33%
4%
2045
40%
5%
2015
HD diesel all
exhaust
processes: all
regulatory
classes
61%
60%
2020
58%
57%
2030
48%
43%
2045
37%
31%
344
-------
8 References
1 USEPA (2020). Speciation of Total Organic Gas and Particulate Matter Emissions from Onroad Vehicles in MOVES3.
EPA-420-R-20-021. Office of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml.
November 2020. https://www.epa.gov/moves/moves-technical-reports.
2 USEPA (2020). Air Toxic Emissions from Onroad Vehicles in MOVES3. EPA-420-R-20-022. Office of Transportation
and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2020.
https://www.epa.gov/moves/moves-technical-reports.
3 USEPA (2015). Greenhouse Gas and Energy Consumption Rates for Onroad Vehicles: Updates for MOVES2014.
EPA-420-R-15-003. Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental
Protection Agency. Ann Arbor, MI. October, 2015. https://www.epa.gov/moves/moves-technical-reports.
4 USEPA (2012). Use of Data from "Development of Emission Rates for the MOVES Model," Sierra Research, March 3,
2010. Assessment and Standards Division. Office of Transportation and Air Quality. Ann Arbor, Ml. April 2012.5 USEPA
(2020). Evaporative Emissions from Onroad Vehicles in MOVES3. EPA-420-R-20-012. Office of Transportation and Air
Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2020. https://www.epa.gov/moves/moves-
technical-reports.
6 US EPA. Heavy-Duty Highway Compression-Ignition Engines and Urban Buses—Exhaust Emission Standards.
http://www.epa.gov/otaq/standards/heavv-duty/hdci-exhaust.htm. Accessed February 12, 2015.
7 40 CFR§ 86.091(2).
8 USEPA (2022). Exhaust Emission Rates for Light-Duty Onroad Vehicles in MOVES3. EPA-420-R-22-031. Office of
Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2022.
https://www.epa.gov/moves/moves-onroad-technical-reports
9 USEPA (2020). Population and Activity of Onroad Vehicles in MOVES3. EPA-420-R-20-023. Office of Transportation
and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2020.
https://www.epa.gov/moves/moves-technical-reports.
10 USEPA (2020). Brake and Tire Wear Emissions from Onroad Vehicles in MOVES3. EPA-420-R-20-014. Office of
Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2020.
https://www.epa.gov/moves/moves-technical-reports.
11 USEPA (2015). Exhaust Emission Rates for Heavy-Duty On-road Vehicles in MOVES2014. EPA-420-R-15-015a.
Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental Protection Agency.
Ann Arbor, Ml. November, 2015. https://www.epa.gov/moves/moves-technical-reports.
12 USEPA (2016), Memorandum to the Docket "Updates to MOVES for Emissions Analysis of Greenhouse Gas
Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles - Phase 2 FRM" Docket
No. EPA-HQ-OAR-2016. July, 2016.13 USEPA (2015). U.S. Environmental Protection Agency Peer Review Handbook.
EPA/100/B-15/001. Prepared for the U.S. Environmental Protection Agency under the direction of the EPA Peer
Review Advisory Group. Washington, D.C. 20460. October 2015. https://www.epa.gov/sites/production/files/2020-
08/documents/epa_peer_review_handbook_4th_edition.pdf.
345
-------
14 USEPA (2017). Exhaust Emission Rates for Heavy-Duty On-road Vehicles in MOVES201X - Draft Report. Draft report
and peer-review documents. Record ID 328830. EPA Science Inventory. September 2017.
https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryld=328830.
15 USEPA (2019). Exhaust Emission Rates of Heavy-Duty Onroad Vehicles in MOVES_CTI_NPRM - Draft Report. Draft
report and peer-review documents. Record ID 347135. EPA Science Inventory. December 2019.
https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryld=347135.
16 USEPA (2020). Exhaust Emission Rates of Heavy-Duty Onroad Vehicles in MOVES3 - Draft Report. Draft report and
peer-review documents. Record ID 347138. EPA Science Inventory. July 2020.
https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryld=347138.
17. USEPA (1998), Caterpillar, Inc., Detroit Diesel Corporation, Mack Trucks, Inc., Navistar International Transportation
Corporation, Renault Vehicules Industriels, s.a., and Volvo Truck Corporation Diesel Engines Settlement. October
22,1998. http://cfpub.epa.gov/enforcement/cases/.
18 Jack, Jason A. U.S. Army Aberdeen Test Center Support of Heavy Duty Diesel Engine Emissions Testing. U.S. Army
Aberdeen Test Center CSTE-DTC-AT-SL-E, Aberdeen Proving Ground, Maryland.
(http://www.epa.gov/ttn/chief/conference/eil5/sessionl/jack.pdf).
19 McClement, Dennis. Reformatting On-Road In-Use Heavy-Duty Emissions Test Data. Sierra Research, Sacramento,
CA. April 2008.
20 Gautam, Mridul, Nigel N. Clark, Gregory Thompson, Daniel K. Carder, and Donald W. Lyons. Evaluation of Mobile
Monitoring Technologies for Heavy-duty Diesel-Powered Vehicle Emissions. Dept. Mechanical and Aerospace
Engineering, College of Engineering and Mineral Resources, West Virginia University, Morgantown, WV.
21 Gautam, Mridul, Nigel N. Clark, Gregory Thompson, Daniel K. Carder, and Donald W. Lyons. Development of In-use
Testing Procedures for Heavy-Duty Diesel-Powered Vehicle Emissions. Dept. Mechanical and Aerospace Engineering,
College of Engineering and Mineral Resources, West Virginia University, Morgantown, WV.
22 Gautam, M., et al. Evaluation of In-Use Heavy-Duty Vehicle Emissions Using the Mobile Emissions Measurement
System (MEMS) for Engine Model Years 2001 to 2003 : Final Reports. Present to engine manufacturers to fulfill
testing requirements documented in Phases III and IV of the Heavy Duty Diesel Engine consent decree. Dept.
Mechanical and Aerospace Engineering, College of Engineering and Mineral Resources, West Virginia University,
Morgantown, WV. 2002 & 2007.
23 "Control of Emissions of Air Pollution From New Motor Vehicles: In-Use Testing for Heavy-Duty Diesel Engines and
Vehicles", 70 FR 34594, June 2005.
24 USEPA. (2019). "Manufacturer-Run In-Use Testing Program Data for Heavy-Duty Diesel Engines." Retrieved
September 17, 2020, from https://www.epa.gov/compliance-and-fuel-economy-data/manufacturer-run-use-testing-
program-data-heavy-duty-diesel-3.
25 Sandhu, Gurdas; Sonntag, Darrell; Sanchez, James. 2018. Identifying Areas of High NOx Operation in Heavy-Duty
Vehicles, 28th CRC Real-World Emissions Workshop, March 18-21, 2018, Garden Grove, California, USA
26 Data Collection of Drayage Trucks in Houston-Galveston Port Area Draft Report. EP-C-06-080. May 27, 2011
27 Bradley, Ron. "Technology Roadmap for the 21st Century Truck Program." U.S. Department of Energy: Energy
Efficiency and Renewable Energy, Washington, D.C., December 2000.
346
-------
28 Rakha, Hesham and Ivana Lucic. Variable Power Vehicle Dynamics Model for Estimating Truck Accelerations. Page 6.
http://filebox.vt.edu/users/hrakha/Publications/Variable%20Power%20Truck%20Acceleration%20-
%20Ver%202.0.pdf
29 National Renewable Energy Laboratory. Development of LNG-Powered heavy-Duty Trucks in Commercial Hauling.
NREL/SR-540-25154, Golden, CO, December 1998.
30 Goodyear. "Factors Affecting Truck Fuel Economy - Section 9" Page 5.
http://www.Roodyear.com/truck/pdf/radialretserv/Retread_S9_V.pdf.
31 Ramsay, Euan and Jonathan Bunker. Acceleration of Multi-Combination Vehicles in Urban Arterial Traffic
Corridors. PhD dissertation, Queensland University of Technology. August 2003, Page 11.
http://eprints.qut.edu.au/archive/00002359/01/RS&ETechForum2003_Ramsay &Bunker_2.pdf
32 Society of Automotive Engineers. Commercial Truck and Bus SAE Recommended Procedure for Vehicle
Performance Prediction and Charting. SAE J2188. Revised October 2003.
33 Bradley, Ron. "Technology Roadmap for the 21st Century Truck Program." U.S. Department of Energy: Energy
Efficiency and Renewable Energy, Washington, D.C., December 2000. Page 32.
34 Pritchard, Ewan G. D. and Richard R. Johnson. Hybrid Electric School Bus Preliminary Technical Feasibility Report.
Advanced Energy Corporation and Department of Mechanical Engineering, North Carolina State University, Raleigh,
NC, September 14, 2004. Page 25.
35 Hedrick, J.K. and A. Ni. Vehicle Modeling and Verification of CNG-Powered Transit Buses. California PATH Working
Paper UCB-ITS-PWP-2004-3. California Partners for Advanced Transit and Highways (PATH), Institute of
Transportation Studies, University of California, Berkeley. February 2004. Page 21.
http://repositories.cdlib.org/cgi/viewcontent.cgi?article=1169&context=its/path
36 Motor Industry Research Association (MIRA). News from MIRA-Automotive Engineering Specialists. Hybrid Theory:
Hybrid Vehicle engineering for economy, the environment, and customer delight. Nuneaton, Warwickshire, UK. Issue
2, Spring 2007.
37 Choi, D., J. Koupal and M. Church (2012). Analysis of Recent Heavy-Duty Vehicle Emission Test Programs. MOVES
Review Workgroup, Ann Arbor, Ml. September 25, 2012 https://www.epa.gov/sites/production/files/2016-
06/documents/02-hd-em ission-vehicle-test-programs.pdf.
38 USEPA (2015). Exhaust Emission Rates for Heavy-Duty On-road Vehicles in MOVES2014. EPA-420-R-15-015a.
Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental Protection Agency.
Ann Arbor, Ml. November, 2015. https://www.epa.gov/moves/moves-technical-reports.
39 USEPA (2002). Update of Heavy-Duty Emission Levels (Model Years 1988-2004) for Use in MOBILE6. EPA-420-R-02-
018. July 2002. http://www.epa.gov/otaq/models/mobile6/r02018.pdf. (Table 17)
40 USEPA Office of Transportation and Air Quality. Heavy Duty Diesel Engine Consent Decree Low NOx Rebuild
Program Summary. Washington, D.C. http://www.epa.gov/compliance/resources/cases/civil/caa/diesel/progress.pdf.
41 USEPA (2014). Tier 3 Vehicle Emission and Fuel Standards Program. Regulatory Impact Analysis. EPA-420-R-14-004.
February 2014. http://www.epa.gov/otaq/tier3.htm
42 Clark, Nigel et al. California Heavy Heavy-Duty Diesel Truck Emissions Characterization for Program E-55/59. West
Virginia University Research Corporation. Morgantown, WV. November 2005.
347
-------
43 Hsu, Y., and Mullen, M. 2007. Compilation of Diesel Emissions Speciation Data. Prepared by E. H. Pechan and
Associates for the Coordinating Research Council. CRC Contract No. E-75, October, 2007. Available at
www.crcao.org.
44 Nam, Ed and Robert Giannelli. Fuel Consumption Modeling of Conventional and Advanced Technology Vehicles in
the Physical Emission Rate Estimator (PERE). EPA420-P-05-001. USEPA Office of Transportation and Air Quality,
Assessment and Standards Division, Ann Arbor, Ml. (http://www.epa.gov/otaq/models/ngm/420p05001.pdf).
February 2005.
45 Kinsey, J. S., et al. (2006). Evaluation of methods for the determination of diesel-generated fine particulate matter:
Physical characterization results. Journal of Aerosol Science, 37 (1), 63-87. DOI:
http://dx.doi.Org/10.1016/j.jaerosci.2005.03.007.
46 USEPA (2009). Development of Emission Rates for Heavy-Duty Vehicles in the Motor Vehicle Emissions Simulator
(Draft MOVES2009). EPA-420-P-09-005. Office of Transportation and Air Quality. US Environmental Protection
Agency. Ann Arbor, Ml. August, 2009. http://www.epa.gov/otaq/models/moves/techdocs/420p09005.pdf.
47 USEPA Office of Transportation and Air Quality. Update Heavy-Duty Engine Emission Conversion Factors for
MOBILE6: Analysis of BSFCs and Calculation of Heavy-Duty Engine Emission Conversion Factors. EPA420-R-02-005,
M6.HDE.004. Assessment and Standards Division, Ann Arbor, Ml.
(http://www.epa.gov/otaq/models/mobile6/r02005.pdf).
48 Khalek, Imad, Thomas L Bougher and Patrick M. Merritt. Phase 1 of the Advanced Collaborative Emissions Study
(ACES). SwRI Project No. 03.13062. Southwest Research Institute, San Antonio, TX; Coordinating Research Council
(CRC), Alpharetta, GA; Health Effects Institute, Boston, MA. June 2009.
http://www.crcao.org/reports/recentstudies2009/ACES%20Phase%201/ACES%20Phasel%20Final%20Report%2015J
UN2009.pdf
49 US EPA. 2012. Black Carbon Report to Congress. EPA-450/R-12-001. March 2012.
http://www.epa.gov/blackcarbon/2012report/fullreport.pdf.
50 Quiros, D. C., et al. (2016). Real-World Emissions from Modern Heavy-Duty Diesel, Natural Gas, and Hybrid Diesel
Trucks Operating Along Major California Freight Corridors. Emission Control Science and Technology, 2 (3), 156-172.
DOI: 10.1007/s40825-016-0044-0.
51 Dixit, P., et al. (2017). Differences between emissions measured in urban driving and certification testing of heavy-
duty diesel engines. Atmospheric Environment, 166, 276-285. DOI:
http://dx.doi.Org/10.1016/j.atmosenv.2017.06.037.
52 Thiruvengadam, A., et al. (2015). Emission Rates of Regulated Pollutants from Current Technology Heavy-Duty
Diesel and Natural Gas Goods Movement Vehicles. Environ Sci Technol, 49 (8), 5236-5244. DOI:
10.1021/acs.est.5b00943.
53 40 CFR § 86.1816-18. Table 1- Fully Phased-in Tier 3 HDV Exhaust Emission Standards.
54 Graboski, Michael S., Robert L. McCormick, Janet Yanowitz, and Lisa Ryan. Heavy-Duty Diesel Vehicle Testing for the
Northern Front Range Air Quality Study. Colorado Institute for Fuels and High-Altitude Engine Research, Colorado
School of Mines, Golden, Colorado. Prepared for Colorado State University. February 1998.
348
-------
55 Energy and Environmental Analysis, Inc. Documentation and Analysis of Heavy-Duty Diesel Vehicle Emission Test
Data. Prepared for New York Department of Environmental Conservation, December 2000.
56 40 CFR § 1065.260
57 40 CFR § 1065.250
58 USEPA (2012). Updates to the Greenhouse Gas and Energy Consumption Rates in MOVES2010a. EPA-420-R-12-025.
Office of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. August, 2012.
http://www.epa.gov/otaq/models/moves/documents/420rl2025.pdf.59 USEPA (2014). Greenhouse Gas Emissions
from a Typical Passenger Vehicle. EPA-420-F-14-040a. Office of Transportation and Air Quality. US Environmental
Protection Agency. Ann Arbor, Ml. May 2014. http://www.epa.gov/otaq/climate/documents/420fl4040a.pdf.
60 USEPA (2011). Greenhouse Gas Emission Standards and Fuel Efficiency Standards for Medium- and Heavy-Duty
Engines and Vehicles (76 FR 57106, September 15, 2011)
6140 Code of Federal Register Volume 76 at 57216 and 57236, September 15, 2011.
62 Greenhouse Gas Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles -
Phase 2 Final Regulatory Impact Analysis, Chapter 5. EPA-420-R-16-900. August 2016
63 USEPA (2012). 2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average
Fuel Economy Standards (77 FR No. 199, October 15, 2012)
64 USEPA (2022). Emission Adjustments for Temperature, Humidity, Air Conditioning, and Inspection and Maintenance
for Onroad Vehicles in MOVES3. EPA-420-R-22-030. Assessment and Standards Division. Office of Transportation and
Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. November 2022.
https://www.epa.gov/moves/moves-onroad-technical-reports
65 Calcagno, James A. Evaluation of Heavy-Duty Diesel Vehicle Emissions During Cold-Start and Steady-State Idling
Conditions and Reduction of Emissions from a Truck-Stop Electrification Program. PhD Dissertation, Department of
Civil and Environmental Engineering, University of Tennessee, Knoxville, December 2005.
66 40 CFR Part 86, Appendix I (f) (2)
67 EMFAC 2000 Section 6.7 START CORRECTION FACTORS. 4/20/2000. On-Road Emissions Model Methodology
Documentation, http://www.arb.ca.gov/msei/onroad/doctable_test.htm
68 Glover, E.; Carey, P. Determination of Start Emissions as a Function of Mileage and Soak Time for 1981-1993 Model-
year Light-Duty Vehicles. EPA420-R-01-058 (M6.STE.003). USEPA Office of Transportation and Air Quality, Ann Arbor,
Ml. November, 2001.
69 USEPA (2017). Heavy-Duty Chassis Start Emissions Testing. Office of Transportation and Air Quality. US
Environmental Protection Agency, Ann Arbor, Ml. July 2017.
70 USEPA (2017). On-Road Light-Duty and Heavy-Duty Vehicle Start Emissions Testing. Office of Transportation and Air
Quality. US Environmental Protection Agency, Ann Arbor, Ml. July 2017
71 USEPA (2005). Energy and Emissions Inputs. EPA-420-P-05-003. Office of Transportation and Air Quality. US
Environmental Protection Agency. Ann Arbor, Ml. March, 2005. http://www.epa.gov/otaq/models/moves/moves-
reports.htm.
349
-------
72 Keel-Blackmon, K., S. Curran and M. V. Lapsa (2016). Summary of OEM Idling Recommendations from Vehicle Owner
s Manuals. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Fuels, Engines and Emissions
Research Center; National Transportation Research Center (NTRC).
https://info.ornl.gov/sites/publications/Files/Pub61263.pdf.
73McCormick, Robert, M. S. Graboski, T. L. Alleman , J. Yanowitz. Idle Emissions from Heavy-Duty Diesel and Natural
Gas Vehicles at High Altitude. Journal of the Air and Waste Management Association, 50(ll):1992-8. Revised May 3,
2000.
74Lim, Han. Study of Exhaust Emissions from Idling Heavy-duty Diesel Trucks and Commercially Available Idle-Reducing
Devices. EPA420-R-02-025. US EPA Office of Transportation and Air Quality, Certification and Compliance Division.
October 2002.
75 Irick, David K. and Bob Wilson. NOx Emissions and Fuel Consumption ofHDDVs during Extended Idle. University of
Tennessee, IdleAire Technologies Inc. In: Proceedings, Coordinated Research Council 12th Annual On-Road Vehicle
Emission Workshop, San Diego, California, April 15-17, 2002.
76Lambert, Douglas, et al. Roadside Emissions Study: Preliminary Results for Stationary and On-Road Testing of Diesel
Trucks in Tulare, California. California Environmental Protection Agency. Air Resources Board. Mobile Source
Operations Division, Clean Air Technologies International, Inc May 15, 2002.
77Gautam, Mridual and Nigel N. Clark. Heavy-duty Vehicle Chassis Dynamometer Testing for Emissions Inventory, Air
Quality Modeling, Source Apportionment and Air Toxics Emissions Inventory. Phase I Interim Report, CRC Project No.
E-55/E-59, West Virginia University Research Corporation, Morgantown, July 2002.
78National Cooperative Highway Research Program (NCHRP). Heavy-duty Vehicle Emissions. NCHRP Project 25-14,
Cambridge Systematics, Inc., Battelle Laboratories, Sierra Research and West Virginia University. October 2002.
79Tang, Shida and John Munn. Internal Report- Idle Emissions from Heavy-Duty Diesel Trucks in the New York
Metropolitan Area. New York State Dept of Environmental Conservation, November 9, 2001.
80Brodrick, Dwyer. Potential Benefits of Utilizing Fuel Cell Auxiliary Power Units in Lieu of Heavy-Duty Truck Engine
Idling. Paper UCD-ITS-REP-01-01. Institute of Transportation Studies, University of California, Davis, 2001.
81Storey, John M.E., John F. Thomas, Samuel A. Lewis, Sr., Thang Q. Dam, K. Edwards, Dean, Gerald L. DeVault, and
Dominic J. Retrossa. Particulate Matter and Aldehyde Emissions from Idling Heavy-Duty Diesel Trucks. SAE Paper
2003-01-0289. Society of Automotive Engineers, Warrendale, PA.
82 Keel-Blackmon, K., S. Curran and M. V. Lapsa (2016). Summary of OEM Idling Recommendations from Vehicle Owner
s Manuals. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Fuels, Engines and Emissions
Research Center; National Transportation Research Center (NTRC).
https://info.ornl.gov/sites/publications/Files/Pub61263.pdf.
83 Lutsey, N., Brodrick, C-J., Sperling, D., Oglesby, C., Transportation Research Record: Journal of the Transportation
Research Board, No 1880, TRB, National Research Council, Washington, D.C., 2004, pp. 28-38
84 Hoekzema, A. (2015). Modeling Truck Idling Emissions in Central Texas. 6800 Burleson Road, Building 310, Suite 165
Austin, Texas 78744, Capital Area Council of Governments.
350
-------
85 Khan, A. S., N. N. Clark, M. Gautam, W. S. Wayne, G. J. Thompson and D. W. Lyons (2009). Idle Emissions from
Medium Heavy-Duty Diesel and Gasoline Trucks. Journal of the Air & Waste Management Association, 59 (3), 354-
359.
86 Farzaneh, M., et al. (2014). TEXAS-SPECIFIC DRIVE CYCLES AND IDLE EMISSIONS RATES FOR USING WITH EPA'S
MOVES MODEL. FHWA/TX-14/0-6629-1. Texas A&M Transportation Institute. May, 2014.
http://tti.tamu.edu/documents/0-6629-l.pdf.
87 ARB (2015). EMFAC2014 Volume III - Technical Documentation. California Environmental Protection Agency, Air
Resources Board, Mobile Source Analysis Branch, Air Quality Planning & Science Division. May 12, 2015.
88 California Environmental Protection Agency. Heavy-Duty Engines and Vehicles Executive Orders. Motor Vehicle and
Engine Certification Program. Air Resources Board. Available at:
http://www.arb.ca.gOv/msprog/onroad/cert/cert.php#3
89 US EPA, Navistar Inc. Heavy-Duty Engine Recall. EPA-420-F-13-038, Ann Arbor, Ml: June 2013,
https://www3.epa.gov/otaq/cert/eng-recall/420fl3038.pdf.
90 Clark, N. and M. Gautam (2007). HEAVY-DUTY Vehicle Chassis Dynamometer Testing for Emissions Inventory, Air
Quality Modeling, Source Apportionment and Air Toxics Emissions Inventory. CRC Report. No. E55/59. Aug-07.
91 US EPA, Engine Certification Data. On-Highway Heavy Duty - Diesel and Gasoline (2012).
https://www3.epa.gOv/otaq/certdata.htm#oh
92 Khalek, I. A., M. G. Blanks, P. M. Merritt and B. Zielinska (2015). Regulated and unregulated emissions from modern
2010 emissions-compliant heavy-duty on-highway diesel engines. Journal of the Air & Waste Management
Association, 65 (8), 987-1001. DOI: 10.1080/10962247.2015.1051606.
93 Khan, A. S., et al. (2009). Idle Emissions from Medium Heavy-Duty Diesel and Gasoline Trucks. Journal of the Air &
Waste Management Association, 59 (3), 354-359. DOI: 10.3155/1047-3289.59.3.354.
94 Zietsman, J. and J. Johnson (2014). Auxiliary Power Unit Testing for SmartWay Idle Reduction Verification. DRAFT
FOR REVIEW. EP-ll-H-000527, Auxiliary Power Unit Testing for SmartWay Idle Reduction Verification. Texas A&M
Transportation Institute. August, 2014.
95 Frey, H. C. and P.-Y. Kuo (2009). Real-World Energy Use and Emission Rates for Idling Long-Haul Trucks and Selected
Idle Reduction Technologies. Journal of the Air & Waste Management Association, 59 (7), 857-864. DOI:
10.3155/1047-3289.59.7.857.
96 TTI (2012). Development of a NOx Verification Protocol and Actual Testing of Onboard Idle Reduction Technologies.
New Technology Research and Development Program. Texas Transportation Institute. Revised: January 2012.
97 Storey, J. M., et al. (2003). Particulate matter and aldehyde emissions from idling heavy-duty diesel trucks. SAE
Technical Paper.
98 See 40 CFR 1037.106(g).
99 US EPA, Nonroad Compression-Ignition Engines: Exhaust Emission Standards, EPA-420-B-16-022, March 2015.
https://www.epa.gov/sites/production/files/2016-03/documents/420bl6022.pdf
351
-------
100 US EPA. Memo to Docket: Updates to MOVES for Emissions Analysis of Greenhouse Gas Emissions and Fuel
Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles - Phase 2 FRM. August 8, 2016. Pages 33-35.
Docket EPA-HQ-OAR-2014-0827-2227.
101 US EPA, Frequently Asked Questions about Heavy-Duty "Glider Vehicles" and "Glider Kits", EPA-420-F-15-904, Ann
Arbor, Ml: July 2015, https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100MUVI.PDF
102 US EPA, Chassis Dynamometer Testing of Two Recent Model Year Heavy-Duty On-Highway Diesel Glider Vehicles,
EPA-HQ-OAR-2014-0827-2417, Ann Arbor, Ml, November 2017, https://www.regulations.gov/document?D=EPA-HQ-
OAR-2014-0827-2417
103 USEPA (2008). Mobile Source Observation Database (MSOD): User Guide and Reference. EPA420-B-08-017. Office
of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, Ml. December, 2008.
https://www.epa.gov/moves/mobile-source-observation-database-msod.
104 EPA. 1999. Tier 2 Vehicle & Gasoline Sulfur Program Final Rule, http://www.epa.gov/otaq/standards/light-
dutv/tier2stds.htm
105 USEPA (2017). Light-Duty Vehicles, Light-Duty Trucks, and Medium-Duty Passenger Vehicles: Tier 2 Exhaust
Emission Standards and Implementation Schedule EPA-420-B-17-028. Office of Transportation and Air Quality. US
Environmental Protection Agency. September 2017.
https://nepis.epa.gov/Exe/ZyPDF.cgi/P100SMQA.PDF?Dockey=P100SMQA.PDF.
106 Development of Emission Rates for Heavy-Duty Vehicles in the Motor Vehicle Emissions Simulator MOVES2010
(131 pp, EPA-420-B-12-049, August 2012)
107 EPA. 2001. "Control of Air Pollution From New Motor Vehicles: Heavy-Duty Engine and Vehicle Standards and
Highway Diesel Fuel Sulfur Control Requirements; Final Rule" http://www.epa.gov/otaq/standards/heavv-dutv/hdsi-
exhaust.htm108 USEPA (2015). Exhaust Emission Rates for Heavy-Duty On-road Vehicles in MOVES2014. EPA-420-R-
15-015a. Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental Protection
Agency. Ann Arbor, Ml. November, 2015. https://www.epa.gov/moves/moves-technical-reports.
109 USEPA. Final Rulemaking to Establish Greenhouse Gas Emission Standards and Fuel Efficiency Standards for
Medium- and Heavy-Duty Engines and Vehicles: Regulatory Impact Analysis. EPA-420-R-13-901, August 2011. Page 2-
30.
110 USEPA. Final Rulemaking to Establish Greenhouse Gas Emission Standards and Fuel Efficiency Standards for
Medium- and Heavy-Duty Engines and Vehicles - Phase 2: Regulatory Impact Analysis. EPA-420-R-16-900, August
2016. Page 5-14.
111 USEPA (2016). Heavy-Duty Highway Spark-Ignition Engines: Exhaust Emission Standards. EPA-420-B-16-019.
Office of Transportation and Air Quality. US Environmental Protection Agency. March 2016
https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P1000A01.pdf.
112 Toro, C., J. Warila, D. Sonntag, D. Choi and M. Beardsley (2019). Updates to "high-power" emission rates and start
deterioration for light-duty vehicles MOVES Review Workgroup, Ann Arbor, Ml. April 10, 2019
https://www.epa.gov/sites/production/files/2019-06/documents/03-updates-ld-emission-rates-start-deterioration-
2019-04-10.pdf.
352
-------
113 USEPA (2015). Exhaust Emission Rates for Light-Duty On-road Vehicles in MOVES2014. EPA-420-R-15-005.
Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental Protection Agency.
Ann Arbor, Ml. October, 2015. https://www.epa.gov/moves/moves-technical-reports.
114 40 CFR Part 86 Appendix 1(f)(1)
115 Boyce, B. 2014. Cummins Westport - Heavy Duty Natural Gas Engines for Trucks and Buses, presented at the
Southeast Alternative Fuels Conference & Expo, October 22, Raleigh, NC, USA.
116 Ayala, A., Gebel, M., Okamoto, R., Rieger, P. et al., "Oxidation Catalyst Effect on CNG Transit Bus Emissions,"
Society of Automotive Engineers, SAE Technical Paper 2003-01-1900, 2003. (http://papers.sae.org/2003-01-1900).
117 "Central Business District (CBD)," Emission Test Cycles. DieselNet. Last Updated: September 2000, Accessed: 10
July 2012. (http://www.dieselnet.com/standards/cycles/cbd.php).
118 DieselNet, (http://www.dieselnet.com/standards/cycles/cbd.php).
119 Hesterberg, T.; Lapin, C.; Bunn, W.; 2008. "A Comparison of Emissions from Vehicles Fueled with Diesel or
Compressed Natural Gas." Environ. Sci. Technol. 42(17): 6437-6445.
(http://pubs.acs.org/doi/abs/10.1021/esQ71718i).
120 Melendez, M.; Taylor, J.; Zuboy, J. et al. Emission Testing of Washington Metropolitan Area Transit Authority
(WMATA) Natural Gas and Diesel Transit Buses. Technical Report NREL/TP-540-36355, National Renewable Energy
Laboratory, Office of Energy Efficiency and Renewable Energy, Department of Energy. Golden, CO. December 2005.
(http://www.afdc.energy.gov/pdfs/36355.pdf).
121 LeTavec, C., Uihlein, J., Vertin, K., Chatterjee, S. et al., "Year-Long Evaluation of Trucks and Buses Equipped with
Passive Diesel Particulate Filters," Society of Automotive Engineers. SAE Technical Paper 2002-01-0433, 2002.
(http://papers.sae.org/2002-01-0433).
122 Ayala, A., Kado, N., Okamoto, R., Holmen, B. et al., "Diesel and CNG Heavy-duty Transit Bus Emissions over
Multiple Driving Schedules: Regulated Pollutants and Project Overview," Society of Automotive Engineers, SAE
Technical Paper 2002-01-1722, 2002. (http://papers.sae.org/2002-01-1722).
123 Lanni, T., Frank, B., Tang, S., Rosenblatt, D. et al., "Performance and Emissions Evaluation of Compressed Natural
Gas and Clean Diesel Buses at New York City's Metropolitan Transit Authority," Society of Automotive Engineers.,
SAE Technical Paper 2003-01 -0300, 2003. (http://papers.sae.org/2003-01-0300).
124 McKain, D., Clark, N, Balon, T., Moynihan, P. et al., "Characterization of Emissions from Flybrid-Electric and
Conventional Transit Buses," Society of Automotive Engineers, SAE Technical Paper 2000-01-2011, 2000.
(http://papers.sae.org/2000-01-2011).
125 Clark, N, Gautam, M., Lyons, D., Bata, R. et al., "Natural Gas and Diesel Transit Bus Emissions: Review and
Recent Data," Society of Automotive Engineers, SAE Technical Paper 973203, 1997. (http://papers.sae.org/973203).
126 McCormick, R., Graboski, M., Alleman, T., Flerring, A. et al., "In-Use Emissions from Natural Gas Fueled Fleavy-
Duty Vehicles," Society of Automotive Engineers, SAE Technical Paper 1999-01-1507, 1999.
(http://papers.sae.org/1999-01-1507).
127 Clark, et al., "Effects of Average Driving Cycle Speed on Lean-Burn Natural Gas Bus Emissions and Fuel Economy,"
SAE Technical Paper 2007-01-0054, 2007
128 "Heavy-Duty Highway Compression-Ignition Engines and Urban Buses -- Exhaust Emission Standards," Emission
Standards Reference Guide, USEPA. Last Updated: 5 July 2012, Accessed: 20 July 2012.
353
-------
129 US EPA OTAQ, 2012. (http://iaspub.epa.gov/otaqpub/pubsearch.jsp).
130 40 CFR 86.105-94. "Clean-fuel fleet emission standards for heavy-duty engines." Code of Federal Regulations.
131 Search for compliance documents in the "Heavy-Duty Highway Compression-Ignited Engines" category on EPA's
Transportation and Air Quality Document Index System (DIS), Last Updated: 10 August 2012, Accessed: 10 August
2012. (http://iaspub.epa.gov/otaqpub/pubsearch.jsp).
132 Hajbabaei, M., G. Karavalakis, K. C. Johnson, L Lee and T. D. Durbin (2013). Impact of natural gas fuel composition
on criteria, toxic, and particle emissions from transit buses equipped with lean burn and stoichiometric engines.
Energy, 62 (0), 425-434. DOI: http://dx.doi.Org/10.1016/j.energy.2013.09.040.
133 Yoon, S.; Collins, J.; Thiruvengadam, A.; Gautam, M.; Herner, J.; Ayala, A. Criteria pollutant and greenhouse gas
emissions from CNG transit buses equipped with three-way catalysts compared to leanburn engines and oxidation
catalyst technologies, Journal of the Air & Waste Management Association, 2013, 63:8, 926-933,
http://dx.doi.org/10.1080/10962247.2013.800170.
134 Jaaskelainen, H. Crankcase Ventilation. DieselNet Technology Guide. www.DieselNet.com. Copyright © Ecopoint
Inc. Revision 2012.12.
135 40 CFR 86.004-11. "Control of Emissions from New and In-Use Highway Vehicles and Engines." Code of Federal
Regulations.
136 Zielinska, B.; Campbell, D.; Lawson, D. R.; Ireson, R. G.; Weaver, C. S.; Hesterberg, T. W.; Larson, T.; Davey, M.; Liu,
L.-J. S. 2008. Detailed characterization and profiles of crankcase and diesel particulate matter exhaust emissions using
speciated organics Environ. Sci. Technol. 42(15): 5661-5666.
137 Ireson, R.G., Ondov, J. M., Zielinska, B.,. Weaver, C. S., Easter, M. D., Lawson, D. R., Hesterberg, T. W., Davey, M. E.,
Liu, L.-J. S. Measuring In-Cabin School Bus Tailpipe and Crankcase PM2.5: A New Dual Tracer Method, Journal of the
Air & Waste Management Association, 2011, 61:5, 494-503
138 Hill, L. B.; Zimmerman, N. J.; Gooch, J.; A Multi-City Investigation of the Effectiveness of Retrofit Emissions Controls
in Reducing Exposures to Particulate Matter in School Buses. January 2005. Clean Air Task Force.
139 40 CFR 86.007-11. "Control of Emissions from New and In-Use Highway Vehicles and Engines." Code of Federal
Regulations.
140 Khalek, I. A.; Bougher, T. L; Merrit, P. M.; Phase 1 of the Advanced Collaborative Emissions Study. CRC Report: ACES
Phase 1, June 2009.
141 Clark, N. McKain, D., Barnett, R., Wayne, S., Gautam, M., Thompson, G., Lyons, D. "Evaluation of Crankcase
Emissions Abatement Device," August 8, 2006. West Virginia University. Available at:
http://www.worldnci.com/files/Documents/WVA_FinalReport_200600808.pdf
142 Clark, N., Tatli, E., Barnett, R., Wayne, W. et al., "Characterization and Abatement of Diesel Crankcase Emissions,"
SAE Technical Paper 2006-01-3372, 2006, doi:10.4271/2006-01-3372.
143 Hare, C. T.; Baines, T. M.; Characterization of Diesel Crankcase Emissions. Society of Automotive Engineers, Off-
Highway Vehicle Meeting and Exhibition. MECA, Milwaukee. 1977.
144 Kalayci, Veli. "Spiracle™ Crankcase Filtration Systems: Technical Article" . Donaldson Company, Inc. January 2011.
Available at: www.donaldson.com/en/engine/support/datalibrary/053490.pdf
354
-------
145 Gerhardt, M. J., et al. (2020). Crankcase Emissions for MY 2007+ Heavy-Duty Diesel Trucks. MOVES Review
Workgroup, Ann Arbor, Ml. October 15, 2020. https://www.epa.gov/moves/moves-model-review-work-group.
146 McClenny, W. A. (2000). Recommended Methods for Ambient Air Monitoring of NO, N02, NOy, and Individual NOz
Species. EPA/600/R-01/005. National Exposure Research Laboratory, US EPA. September 2000.
147 Seinfeld, J. H. and S. N. Pandis (2012). Atmospheric chemistry and physics: from air pollution to climate change,
John Wiley & Sons.
148 Dunlea, E. J., et al. (2007). Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban
environment. Atmos. Chem. Phys., 7 (10), 2691-2704. DOI: 10.5194/acp-7-2691-2007.
149 Kurtenbach, R., et al. (2001). Investigations of emissions and heterogeneous formation of HONO in a road traffic
tunnel. Atmospheric Environment, 35 (20), 3385-3394. DOI: http://dx.doi.org/10.1016/S1352-2310(01)00138-8.
150 Wang, Y. J., A. DenBleyker, E. McDonald-Buller, D. Allen and K. M. Zhang (2011). Modeling the chemical evolution
of nitrogen oxides near roadways. Atmospheric Environment, 45 (1), 43-52. DOI:
http://dx.doi.Org/10.1016/j.atmosenv.2010.09.050.
151 USEPA (2009). Development of Emission Rates for Heavy-Duty Vehicles in the Motor Vehicle Emissions Simulator
(Draft MOVES2009). EPA-420-P-09-005. Office of Transportation and Air Quality. US Environmental Protection
Agency. Ann Arbor, Ml. August, 2009. http://www.epa.gov/otaq/models/moves/techdocs/420p09005.pdf.
152 U.S. Census Bureau. Vehicle Inventory and Use Survey. Service Sector Statistics Division, Transportation
Characteristics Branch, http://www.census.gov/svsd/www/vius/products.html.
153 Zhou, Lei. Revision of Heavy Heavy-Duty Diesel Truck Emission Factors and Speed Correction Factors. California Air
Resources Board, Sacramento. October 2006.
154 Illinois Environmental Protection Agency. Effectiveness of On-Board Diagnostic l/M Testing: Report to the General
Assembly, Response to Public Act 92-0682. Bureau of Air, Springfield, IL. September 2003. Page 21.
www.epa.state.il. us/air/publications/obd-report-final. pdf
155 Manufacturers of Emission Controls Association. Technology Details - Catalytic Converters - SCR System.
http://www.meca.org/technology/technology-details?id=5
156 Song, Qingwen, and Zhu, George. Model-based Closed-loop Control of Urea SCR Exhaust Aftertreatment System for
Diesel Engine. SAE 2002-01-287. Society of Automotive Engineers, Warrendale, PA.
157 Darlington, T., Dennis Kahlbaum and Gregory Thompson. On-Road NOx Emission Rates from 1994-2003 Heavy-
Duty Diesel Trucks. SAE 2008-01-1299. Society of Automotive Engineers, April 2008.
158 Preble, C. V., et al. (2015). Effects of Particle Filters and Selective Catalytic Reduction on Heavy-Duty Diesel
Drayage Truck Emissions at the Port of Oakland. Environ Sci Technol, 49 (14), 8864-8871. DOI:
10.1021/acs.est.5b01117.
159 Bishop, G. A., et al. (2015). On-road Heavy-duty Vehicle Emissions Monitoring System. Environ Sci Technol, 49 (3),
1639-1645. DOI: 10.1021/es505534e.
160 ARB (2015). Evaluation of Particulate Matter Filters in On-Road Heavy-Duty Diesel Vehicle Applications. California
Air Resources Board. May 8, 2015. http://www.arb.ca.gov/msprog/onrdiesel/documents/DPFEval.pdf.
161 40 C.F.R. §86.004-2. Revised as of Aug. 8, 2014.
162 US EPA, Chassis Dynamometer Testing of Two Recent Model Year Heavy-Duty On-Highway Diesel Glider Vehicles,
EPA-HQ-OAR-2014-0827-2417, Ann Arbor, Ml, November 2017, https://www.regulations.gov/document?D=EPA-HQ-
OAR-2014-0827-2417 355
------- |