-------
Table of Contents
Page
List of Appendices iii
List of Tables v
1.0 INTRODUCTION 1
2.0 GLOSSARY OF KEY TERMINOLOGY 3
3.0 PROPOSED I/M PERFORMANCE STANDARDS 5
3.1 Enhanced I/M Performance Standard 5
3.1.1 Low Option 5
3.1.2 Medium Option 6
3.1.3 High Option 6
3.2 Recommended Enhanced I/M Program Design 6
3.4 Basic I/M Performance Standard 7
4.0 EMISSION REDUCTIONS FROM I/M PROGRAMS 8
4.1 Recent I/M Test Programs (Maryland and Indiana) 8
4.2 FTP HC/CO Correlation Comparison Between the IM240 and
the Second-chance 2500 rpm/Idle Test 12
4.2.1 I/M Test Assessment Criteria Overview 15
4.2.2 Detailed Discussion of Correlation and Test
Assessment 19
4.2.3 Avoiding Errors of Commission 20
4.3 Approval of Alternative Tests 27
4.4 Transient Testing Fast-Pass/Fast-Fail Strategies 29
4.5 Estimating I/M Testing Credits for MOBILE4.1 31
4.5.1 Tech4.1 Background and Assumptions 31
4.5.2 Evaporative and Running Loss Modeling, and the
Effectiveness of Purge/Pressure Testing 35
4.5.3 Benefits of IM240 NOx Inspections 36
5.0 REGULATORY IMPACT ANALYSIS - ESTIMATING COST AND COST
EFFECTIVENESS 45
Draft -i- 2/26/92
-------
5.1 Cost of Conventional I/M Testing 45
5.1.1 Inspection and Administration Costs 45
5.2 Estimated Cost of High-Tech I/M Testing 47
5.2.1 General Methodology
AQ
5.2.2 Equipment Needs and Costs *°
5.2.3 Cost to Upgrade Centralized Networks 50
5.2.4 Cost to Upgrade Decentralized Programs 56
5.3 Repair Costs 58
5.3.1 HC and CO Exhaust Repair Costs and Methodology 58
5.3.2 NOx Repair Costs and Methodology 60
5.3.3 Evaporative System Repair Costs and Methodology 62
5.4 Fuel Economy Benefits 64
5.4.1 Fuel Economy Benefits of Evaporative System
Repairs 64
5.4.2 Fuel Economy Benefits of IM240 Repairs 65
5.4.3 Fuel Economy Benefit for the 2500 rpm/Idle Test 66
5.5 Recurring Failure and Repair Rates 68
5.6 Method for Estimating Cost Effectiveness of I/M Programs?!
5.6.1 Inspection Costs 72
5.6.2 Repair Costs 72
5.6.3 Fuel Economy Cost Benefits 73
6.0 REGULATORY IMPACT ANALYSIS - COSTS AND BENEFITS OF ENHANCED
I/M OPTIONS 75
6.1 Emission Reduction Benefits 75
6.2 Cost Effectiveness Estimates 77
6.2.1 Assumptions and Inputs for the Program Options 77
6.2.3 Cost-Effectiveness Calculations 78
6.2.4 National Cost of Choosing Low Option I/M 80
6.3 National Costs and Benefits 81
6.3.1 Emission Reductions 81
6.3.2 Economic Costs to Motorists 83
6.4 Motorist Inconvenience Costs 85
7.0 REGULATORY FLEXIBILITY ANALYSIS 86
Draft -ii- 2/26/92
-------
7.1 Regulatory Flexibility Act Requirements 86
7.1.1 The Universe of Affected Entities 87
7.2 Types of Economic Impacts of Concern 88
7.3 Changes in Repair Activity 89
7.3.2 Repair Activity in Future I/M Programs 90
7.4 Changes in Emission Testing Activity in I/M Areas 91
7.4.1 The Existing Market in Centralized and
Decentralized Programs 91
7.4.2 Future Market in Enhanced I/M Programs 98
7.4.3 Centralized Programs 98
7.4.4 Decentralized Programs 98
7.4.5 Impact on Jobs in Decentralized Programs 103
7.4.6 National Impact on Jobs 107
7.5 Mitigating the Impact of Enhanced I/M on Existing
Stations 109
8.0 ONBOARD DIAGNOSTICS AND ON-ROAD TESTING 112
8.1 Onboard Diagnostics, Interim Provisions 112
8.2 On-road Testing, Interim Provisions 112
List of Appendices
Appendix
A Cost Effectiveness Model Version 4.1 Source Code Listing
B Tech4 Model Version 4.1 Source Code Listing
C Evaporative Emissions and Running Loss Emission Factor
Derivation
D Purge and Pressure Test Effectiveness Figures and Spreadsheet
E Regression Analyses and Scatter Plots for Fuel Injected 1983
and Later Vehicles
F MOBILE4.1 Technology Distribution and Emission Group Rates and
Emission Levels
G Exhaust Short Test Accuracy
H Data Analyses for Appendix G
I Evaporative System Purge and Pressure Diagrams
Draft -iii- 2/26/92
-------
J Evaporative System Failures and Repairs
K MOBILE4.1 Performance Standard Analyses, By Option
L Identifying Excess Emitters with a Remote Sensing Device
M Model Year Failure Rates by Test Type
Draft ~iv- 2/26/92
-------
List of Tables
Page
4-1 IM240 Selection Standards for Stratified FTP Recruitment 9
4-2 Weighting Factors for Correcting Recruitment Biases 10
4-3 FTP, Lane IM240, and "Tank Fuel" IM240 Results 21
4-4 Vehicle 1674 NOx, FE Results Versus Dynamometer Settings 22
4-5 Two-Ways-To-Pass Criteria Applied to Lane IM240 TBI Error-of-
Commission Vehicles 24
4-6 Variables Affecting CO Emissions 26
4-7 IM240 Bag-1 Fast-Pass/Fast-Fail Analysis 30
4-8 Short Test Identification Rates 34
4-9 Short Test Repair Effectiveness 35
4-10 Lane IM240 Based Emission Factor Levels with IM240 NOx
Cutpoints 39
4-11 Side Effects of I/M on NOx Emissions 41
4-12 Lane IM240 Based Emission Factors with IM240 Cutpoints 43
5-1 I/M Program Inspection Fees 46
5-2 Equipment Costs for New Tests 49
5-3 Expendables for New Tests 49
5-4 Peak Period Throughput Rates in Independent I/M Programs 50
5-5 Current Program Costs 53
5-6 Costs to Add Pressure Testing to Centralized Programs 54
5-7 Costs to Add Proposed Tests to Centralized Programs 55
5-8 Inspection Volumes in Licensed Inspection Stations 56
5-9 Costs to Conduct Pressure, Purge, and Transient Testing in
Decentralized Programs 58
5-10 Average Cost of Repairing Emission Control Components 60
5-11 NOx Repair Costs 62
5-12 Average Repair Costs and Fuel Economy Benefits 63
5-13 Zero Improvement Vehicle Sample Size Adjustments 66
5-14 Adjusted Zero FE Benefit Vehicle Sample Size 68
5-15 Exhaust Test Failure Rates 69
5-16 Default Inspection Costs in CEM4.1 72
5-17 Default Repair Cost in CEM4.1 73
5-18 Fuel Economy Benefits in CEM4.1 74
6-1 MOBILE4.1 Inputs for Enhanced I/M Performance Standard
Options 76
6-2 Benefits of I/M Programs Options 77
6-3 Total Annual Program Cost 78
6-4 Cost per Ton Allocating All Costs to VOC 79
6-5 VOC Cost per Ton Accounting for NOx and CO Benefit 80
6-6 Total Cost and Benefits of I/M Options 81
6-7 Excess Cost of Choosing Low Option I/M 81
6-8 National Benefits of I/M 83
6-9 Program Costs and Economic Benefits 85
6-10 Costs of the Biennial High Option including Inconvenience 86
7-1 Affected Businesses 88
7-2 Repair Expenses in Enhanced I/M Programs 90
7-3 Number of Inspection Stations by State 92
7-4 Inspection Stations by Category 93
7-5 Inspection Station Volumes and Incomes 94
Draft -v- 2/26/92
-------
7-6 Average Inspection Station Revenues, Costs, and Profits 96
7-7 Inspection Volumes in California 97
7-8 Station Revenues and Profits by Volume 98
7-9 Assumed Station Distributions 10~
7-10 Revenues and Profits for Low and Medium Volume Stations 102
7-11 Numbers of Inspectors per Station by State 104
7-12 Estimated Inspection FTE 105
1 (] *7
7-13 Summary of FTE Gains and Losses •"•" '
7-14 Impact on Jobs of I/M Proposal 108
Draft -vi- 2/26/92
-------
1.0 INTRODUCTION
Despite having the best vehicle control program in the world,
many areas in the United States continue to measure unhealthful
levels of air pollution, approximately half of which can be
attributed to motor vehicles. As a result, in addition to tighter
standards on new vehicles and their fuels, the Clean Air Act
Amendments of 1990 (Act) require the implementation of vehicle
Inspection and Maintenance (I/M) programs in areas that have been
designated as nonattainment for ozone or carbon monoxide. A total
of 162 such areas currently exist in the United States, 52 of
which do not presently operate I/M programs. Depending upon the
severity of the nonattainment problem, these areas will have to
implement either a basic I/M program (required in areas with
moderate ozone nonattainment, and in marginal areas with existing
I/M programs) or an enhanced I/M program (required in most
serious, severe, and extreme ozone areas, as well as most carbon
monoxide (CO) areas registering levels greater than 12.7 parts per
million (ppm)). Seventy-four of the 162 nonattainment areas
currently designated will require the implementation of an
enhanced I/M program.
The Environmental Protection Agency (EPA) has had oversight
and policy development responsibility for I/M programs since the
passage of the Clean Air Act in 1970, which included I/M as an
option for improving air quality. The first such I/M program in
the United States was begun in New Jersey in 1974, and the program
elements which made up this program's design (i.e., a centralized,
annual, idle test of all light-duty gasoline vehicles, with no
waivers or tampering checks) still constitute those design
features upon which the basic I/M performance standard is based.
However, many advances have been made in vehicle technology since
the time of that first I/M program, and while the idle test in use
in many current programs works well enough when it comes to
detecting emission problems in older, low-tech vehicles, its
effectiveness as a testing strategy rapidly drops off as we begin
testing newer, more sophisticated, computer-controlled vehicles.
High-tech vehicles need high-tech testing which more closely
simulates real-world driving conditions and the sort of test to
which vehicles are originally certified - a loaded, transient
test, which requires driving the vehicle through a prescribed
pattern of accelerations and decelerations on a dynamometer.
Much has also been learned since 1974 about the many ways
vehicles contribute to the problem of air pollution. Previously,
it was thought that the majority of the air pollution problem
attributable to mobile sources was the result of exhaust
emissions; it is now understood that emissions in the form of
evaporative and running losses are also major contributors. The
gasoline evaporating in the tank of a vehicle and escaping into
the environment is as much a source of volatile organic compound
(VOC) emissions as are the exhaust gases emitted from the
Draft -1- 2/26/92
-------
tailpipe. Vapor recovery and recirculation mechanisms have been
installed on vehicles since 1971, but these systems can
deteriorate with time and are often rendered useless as a result
of wear, tampering, and design defects. Cost effective tests have
been developed to detect evaporative system failures of this sort,
including the evaporative system purge test and the evaporative
system pressure test.
Under the terms of the Clean Air Act Amendments of 1990, EPA
is required to establish minimum performance standards for I/M
programs. The Act further specifies that the standard for
enhanced I/M shall be based upon a program that employs an annual
cycle of automated emissions analysis, performed at a centralized
site, and enforced through the denial of registration. EPA is
currently in the process of developing and formalizing these
standards in the form of a rulemaking.
In the past, the model program used to establish the
performance standard assumed a basic program along the lines -of
the original New Jersey program - a standard which remains
essentially unchanged for basic I/M programs. For the enhanced
I/M performance standard, however, EPA is evaluating potential
low, medium, and high options, the last of which would be based on
loaded, transient testing, in conjunction with evaporative system
purge and pressure tests (see section 3.0 for a more detailed
discussion of these potential options). Using EPA's MOBILE4.1
computer model, a high-tech I/M program, such as the one proposed
by EPA's high option, is expected to achieve emission reductions
from mobile sources on the order of approximately 31% for ozone-
forming hydrocarbons (HC) and 34% for CO (compared to a 5% HC and
16% CO emission reductions from the basic I/M performance standard
program design).
Given the potentially significant economic impact of this
decision, it is necessary to assess the costs and benefits of
enhanced I/M performance standards. This report provides the
technical background information supporting EPA's cost and benefit
projections.
In assessing the costs and benefits of enhanced I/M, we will
detail the findings of recent research and development on test
procedures and vehicle emissions, the basis for the computer
models used to establish emission benefits and program cost
effectiveness, the differences in cost effectiveness among
programs based upon network and test types, as well as projections
of the average per vehicle cost for inspection and repairs, and
the cost offset of the fuel economy benefit achieved by making
such repairs. The full program listings for the computer models
used to assess the costs and benefits of I/M program design
options, as well as other graphic and tabular support data, are
attached to this report as appendices.
Draft -2- 2/26/92
-------
2.0 GLOSSARY OF KEY TERMINOLOGY
Throughout this report several key terms will be used with
which the reader may not be immediately familiar. To facilitate a
better understanding of the issues involved, the following
glossary is provided.
"Concentration" Versus "Mass Emissions" Tests; Mass emissions
tests provide a much better indication of vehicle emission levels
than concentration tests. A concentration reading of 200 parts
per million (ppm) HC from a subcompact car and the same 200 ppm
reading from a large truck (which is entirely possible) suggest
that the two vehicles pollute equally. However, this is
incorrect. The truck will have a much higher volume of exhaust.
So, over a given one-mile drive, the subcompact car may only emit
50 cubic feet of exhaust gases, whereas the truck may emit 500
cubic feet. With both vehicles emitting 200 ppm HC over the mile,
the total amount of HC emitted by the truck will be 10 times
greater than the amount emitted by the small car. A mass
emissions test allows the total emissions per mile to be measured;
a concentration test does not. All currently approved I/M tests
are concentration tests. The Federal Test Procedure and the IM240
test, however, are mass emissions tests.
Error-of-Commission (Ec): On the basis of an emissions test, the
false failure of a vehicle as "dirty" (i.e., emitting high enough
that repair and a retest are required) when the vehicle, in fact,
meets EPA new car standards, based upon the Federal Test Procedure
(see definition below) . Usually, HC and CO EC'S are defined
without regard to NOX emissions, and vice versa.
Error-of-Omission; To falsely pass as clean a vehicle which, in
fact, exceeds EPA new car standards, based upon the results of the
Federal Test Procedure.
Federal Test Procedure (FTP): The Federal Test Procedure (FTP) is
a mass emissions test created to determine whether prototype
vehicles comply with EPA standards, thus allowing production
vehicles to be certified for sale in the United States. The FTP
has become the '"golden standard" for determining vehicle emission
levels, so it is also used to determine the emission levels of
win-use" vehicles. The FTP is too costly to use for I/M because
vehicles must be maintained in a closely controlled environment
for over 13 hours. The FTP is based on a 20 minute trip, driven
once when the engine is cold, and again when it is hot.
High-Tech Vehicles: Vehicles with computerized control of the
engine and emission control system, especially 1983 or newer
vehicles employing fuel injection (either port fuel injection or
throttle-body injection) as opposed to carburetion as a fuel
metering methodology.
Draft -3- 2/26/92
-------
Idle Test: A concentration-type emission test to measure the
percentage of CO and ppm HC in the exhaust stream of a gasoline-
powered vehicle operating at idle. The nondispersive infrared
detector (NDIR) equipment normally used gives a less accurate
measure of HC than does the flame ionization detector (FID)
equipment used in the FTP and IM240 tests.
IM240 Exhaust Test! A mass emissions (as opposed to
concentration), transient short test run on an inertial and power-
absorbing dynamometer using a 240 second driving cycle loosely
based upon the LA-4 cycle used in the FTP. EPA has usually
divided the driving cycle into 2 parts or "bags" with separate
emissions determinations, but other treatments are possible.
Unlike the idle test which is conducted at a single speed and
expresses emissions in terms of percentages and ppm, the ^IM240 is
conducted at a range of accelerations and decelerations and
provides emissions measurements in terms of grams per mile (gpm).
The IM240 has proved particularly effective in accurately
identifying high emitting, newer technology vehicles.
Multiple Independent Supplier Test-Only Network; A hybrid program
design with elements of both centralized and decentralized
networks in which multiple participants are licensed to perform
I/M testing (as opposed to a single contractor). To establish
equivalency with traditional centralized programs and to avoid the
decentralized discount incorporated in EPA's MOBILE model,
participants must operate test-only facilities and are barred from
making repairs, selling replacement parts, making referrals, or
otherwise engaging in activities that would violate the intention
of the test-only requirement (i.e., the avoidance of conflict-of-
interest) .
Pressure Test: A test whereby inert gas is injected into a
vehicle's evaporative system to establish the system's integrity
by indicating the presence of a leak(s) or by confirming the
system's ability to hold pressure.
Purge Test: A test to determine whether a vehicle's evaporative
emissions system recycles the gasoline vapors adsorbed on the
charcoal in the canister (i.e., whether or not the canister purges
vapors to the engine to be combusted). To provide representative
operation and opportunity for the purge control system to
demonstrate its proper working order, the purge test is conducted
on a dynamometer using the same 240-second transient driving cycle
as the IM240 exhaust gas test.
250. 0 rpm/Idle Test,; A two-speed, steady-state concentration-type
test in which emissions are sampled at both idle and 2500 rpm. TO
be considered a pass, a vehicle must pass at both speeds. The
two-speed test has a better identification rate for high emitting
vehicles than does the standard idle test.
Draft -4- 2/26/92
-------
3.0 PROPOSED I/M PERFORMANCE STANDARDS
3 .1 Enhanced I/M Performance Standard
Under the Act, EPA is required to establish a performance
standard for enhanced I/M programs including, at a minimum,
centralized, annual, automated emission testing of light-duty
vehicles and trucks, including a tampering check for emission
control devices, a misfueling check, and provisions for including
on-road emission testing and inspection of onboard diagnostic
devices (OBD). The performance standard is defined by completely
specifying the design of a model or benchmark I/M program. While
enhanced I/M programs need not match the performance standard's
model program element by element, such programs must be designed
and implemented to meet or exceed the minimum emission reductions
achieved by the performance standard. Any deviations from the
performance standard's program design that may lead to emission
reduction losses must be made up by strengthening other aspects of
the program. For example, while the Act constrains the
performance standard for enhanced I/M programs to be based on an
annual program, it is clear that a biennial program is more cost
effective and results in relatively minor emission reduction
losses over those achieved by an annual program. The emission
reduction losses resulting from a decision to test vehicles
biennially as opposed to annually can be made up, for example, by
extending transient exhaust testing and purge testing to cover
earlier model years than those specified in the performance
standard. This specific example will be discussed in more detail
in section 3.2 of this report, "Recommended Enhanced I/M Program
Design."
In its draft guidance document issued in April 1991, EPA
proposed four potential enhanced I/M performance standards for
comment. More recently, in discussion with interested parties and
for purposes of this document, these four options have been
reduced to three: a low, a medium, and a high option. Although
EPA's proposed performance standard most closely resembles the
high option proposed in EPA's April 1991 draft guidance, the
Agency is still soliciting public comment on the low and medium
options. The low, medium, and high options take an incremental
approach to advanced technology testing. These options are
presented below. Computer modeling runs for these options are
included in Appendix K, while a detailed listing of program
elements is provided in Table 6-1.
3.1.1 Low Option
The Low Option is similar to the better programs currently
operating, and, by the year 2000, under conditions of temperature,
fleet composition, etc. that are reasonably typical of an average
nonattainment area, is projected to yield a 10% reduction in
Draft -5- 2/26/92
-------
highway vehicle VOCs and a 25% reduction in CO over a non-I/M
scenario. This option includes annual, centralized idle testing
of model year 1968 and later light-duty vehicles and trucks to
8,500 Gross Vehicle Weight Rating (GVWR), including a visual
inspection for the catalyst and fuel inlet restrictor on 1981 and
newer vehicles. The option further assumes a pre-1981 failure
rate of 20%, a waiver rate of 1%, and an overall compliance rate
of 98%.
3.1.2 Medium Option
The Medium Option is identical to the above Low Option with
the exception that it includes pressure testing of the evaporative
system on 1971 and newer vehicles. This option is estimated to
yield a 20% reduction in VOCs and a 25% reduction in CO over a
non-I/M scenario (note that pressure testing is a VOC strategy
that yields no CO benefit, making the low and medium options
identical for CO purposes).
3.1.3 High Option
The High Option includes a transient IM240 exhaust test
incorporating NOX cutpoints, and purge testing of the evaporative
control system of 1986 and later vehicles (using cutpoints of 0.8
gpm HC, 15 gpm CO, and 2.0 gpm NOX for emissions during the the
entire 240-second test, but with an additional opportunity to pass
for HC and CO by demonstrating emissions less than 0.5 gpm and 12
gpm, respectively, during the final 147 seconds). Two-speed
testing is to be performed on 1981-1985 model year vehicles (using
cutpoints of 1.2% CO, 6% CO2, and 220 ppm HC) while idle testing
is to be used on pre-1981 vehicles. Idle test cutpoints for older
vehicles must yield a 20% failure rate. The performance standard
also includes visual inspection of the catalyst and fuel inlet
restrictor on all 1984 and later vehicles (as opposed to 1981 in
the low option) and evaporative system integrity (pressure)
testing of 1983 and later vehicles. The High Option is estimated
to yield a 28% reduction in VOCs, a 31% reduction in CO, and a 9%
reduction in NOX over a non-I/M scenario.
3 .2 Recommended Enhanced I/M Program Design
The Act requires EPA to establish a performance standard
based on an annual test program. States, however, are free to
perform a biennial program under such a requirement if a
demonstration can be made that such a program (in combination with
other features) would be equally effective. This demonstration is
made using EPA's mobile source emission model which includes
biennial and annual program credits. For example, using the
current version of the model, MOBILE4.1, a biennial high option
can achieve the same VOC reduction achieved by the annual high
option performance standard by doing transient/purge testing on
Draft -6- 2/26/92
-------
1984 and later vehicles and pressure testing on 1971 and later
vehicles. Given the added convenience and cost effectiveness of a
biennial program, EPA recommends that states adopt the biennial
high option since it can achieve reductions equal to that of the
proposed annual model program which defines the performance
standard. In addition, EPA recommends that initial testing of new
vehicles be delayed until the vehicle is two or three years old,
as the percentage of high emitting vehicles among newer cars is
relatively small.
3.4 Basic I/M Performance Standard
The basic I/M performance standard is based upon the program
design of the original New Jersey program and remains essentially
unchanged as a result of EPA's proposed action. The basic I/M
performance standard is estimated to yield a 5% reduction in
mobile source VOC emissions and a 16% reduction in CO. The
performance standard includes annual, centralized idle testing of
model year 1968 and later light duty vehicles. The pre-1981
failure rate is assumed to be 20%, with 0% waivers and 100%
compliance. Unlike the low option for enhanced I/M, the basic I/M
performance standard does not include testing of light-duty
trucks; neither does it include visual inspections of any emission
control components.
Draft -7- 2/26/92
-------
4.0 EMISSION REDUCTIONS FROM I/M PROGRAMS
4.1 Recent I/M Test Proci;rajns (Maryland and. Indiana)
The data used by EPA to assess the benefits of high-tech I/M
testing concepts, including the IM2401' and evaporative system
purge and pressure testing, have been obtained as a result of two
special testing programs performed under contract to EPA. The
'first testing program - an IM240 transient test pilot study - was
conducted as part of a cooperative project with the State of
Maryland in 1989, and utilized one of the state's I/M stations for
testing and recruiting vehicles. This was the first attempt to
perform transient emissions tests on consumer vehicles in a high
throughput system. A more extensive program is currently being
run in Indiana. The Maryland pilot study began testing in August
1989, and continued through December of that year, testing a total
of approximately 600 vehicles for an average of approximately 120
vehicles per month. The larger-scale Indiana program began
testing in February 1990. As of November 1, 1991, approximately
8,300 vehicles had been tested as part of the Indiana program,
with an average of approximately 120 vehicles per week. As such,
the database produced by this test program is the largest of its
kind ever assembled to assess I/M testing.
The Indiana testing contracts include two test facilities, a
laboratory in New Carlisle (a few miles west of South Bend) , and
an I/M station in Hammond. The laboratory is owned by Automotive
Testing Laboratories, Inc. (ATL), a contractor to EPA, and the I/M
station is owned by the Indiana Vocational-Technical College,
which operates the I/M program for the State of Indiana. The I/M
station includes four lanes, with ATL running one of the four.
EPA has three separate testing contracts in Indiana that
utilize the two facilities: Emission factor (EF), I/M, and running
loss testing. Reformulated fuels testing is being performed under
the EF contract. The three contracts use vehicles that are
selected at the I/M station. The selection criteria for follow-up
laboratory testing include model year, fuel metering type, and
results from the following tests: The IM240, canister purge flow
measurement, and evaporative control system pressure tests.
The goal at the I/M station originally was to test a random
sample of 1976 and newer light duty vehicles. On May 15, 1991
the recruitment goal changed to randomly sample 1983 and newer
vehicles, to increase the number of fuel-injected vehicles
represented in the database. This change was made to reflect the
fact that fuel injection is rapidly replacing carburetion as the
1 Pidgeon, W. and Dobie, N., «The IM240 Transient I/M Dynamometer Driving
Schedule and The Composite I/M Test Procedure," U.S. EPA Technical Report
Number EPA-AA-TSS-91-1, January 1991.
Draft -8- 2/26/92
-------
preferred fuel-metering method for new vehicles, and the
percentage of carbureted vehicles in the in-use fleet will become
insignificant in the future.
Choosing cars for further laboratory testing is driven by the
overriding importance of testing and assessing emissions from -
and the impact of repair on - dirty in-use vehicles. A random
sample of vehicles visiting the I/M station would result in the
contractor recruiting mostly clean vehicles, given that the
majority of excess emissions comes from a relatively small
percentage of vehicles known as high to super emitters. To avoid
the problem and cost of evaluating a majority of vehicles that
will ultimately be assessed as clean, a stratified recruitment
plan is employed to deliberately over-recruit dirty cars, based on
the results of IM240, purge and pressure tests. Actually, two
recruitment and lab testing programs operate simultaneously. In
one, a nominally 50/50 mix of IM240-clean and IM240-dirty vehicles
is recruited for FTP exhaust testing. In actual practice, more
clean cars than dirty have been recruited rather than allow lab
testing slots to be idle while waiting for a dirty car to be
recruited. The Hammond I/M lane vehicles were categorized as
clean or dirty using the IM240 standards listed in Table 4-1. In
the other lab-testing recruitment effort, a sample even more
heavily weighted toward purge and pressure test failures is
recruited for evaporative and running loss emissions testing.
Table 4-1
IM240 Selection Standards for Stratified FTP Recruitment
Selection Standards
(grams per mile)
Model Years HC CO
1986+ * >1.10 >15.0
1983-85 >1.20 >16.0
* The 1986+ standards were set to be more stringent than 1983-
1985 standards to improve recruitment of high emitters and
to balance the failure rates between model year groups.
The FTP database that results from EPA's recruitment targets
must be corrected to represent the clean/dirty vehicle ratio in
the in-use fleet to correctly determine excess emission
identification rates (IDR), error-of-commission rates (Ec) and
failure rates (all important criteria for assessing the overall
effectiveness of I/M testing strategies) . The database was
corrected using the weighting factors presented in Table 4-2.
Draft -9- 2/26/92
-------
Table 4-2
13.
Fuel
Metering
System
PFI
g=-^Vf-t-tl--L-Llv
Lane
IM240
Results
Clean
Total
a cj.^i-^j-1-
Lane
Count
1505
97
1602
' J-^JJ- W +
Lab
Sample
Count
55
19
74
Weighting
Factor
27.36
5.11
# Lab Veh
Passing
FTP
23
1
24
# Normals
Failing
FTP
24
2
26
Clean
Dirty
1555
166
73
35
21.30
4.74
25
4
32
6
TBI
Total 1721 108 29 38
Weighting factors are used as follows: If the 19 dirty
vehicles that received FTP tests in the Port Fuel Injected (PFI)
vehicle sample had excess HC emissions which totaled 100 gpm, the
database would be corrected in this case by multiplying 100 by the
5.11 weighting factor, resulting in a corrected excess emission
rate of 511 gpm for the dirty vehicles (excess emissions are those
FTP-measured emissions that exceed the certification emission
standards for the vehicle under consideration; an I/M test's
identification rate for excess emissions represents one of the
important criteria for assessing an I/M test's effectiveness, as
detailed in section 4.2 of this report) . In comparison, the
excess emissions of the IM240 clean vehicles have to be multiplied
by 27.36 to make their excess emissions representative. The total
simulated excess emissions are the sum of the simulated excess
emissions from the clean and dirty vehicles in the I/M lane
sample. The number of vehicles tested was similarly adjusted with
the factors for the purpose of calculating failure rates. The
large sample of 55 clean cars in this sample provides confidence
in conclusions about a test's relative tendency to avoid failing
clean cars.
Appendix H provides additional information on adjustments to
make the FTP database representative of the Hammond lane fleet's
ratio of clean and dirty vehicles. Appendix H also includes
tables that allow a comparison of cutpoint effects on IDR, I/M
failure rates, EC rates, and I/M failure rates for FTP-passing
vehicles.
At the Hammond I/M station, in addition to the IM240
technicians perform the official Indiana I/M test (2500 rpm/ldle)
and an additional second-chance 2500 rpm/ldle test1 for those that
fail the first chance test. Vehicles that require a second-chance
test first receive 3 minutes of preconditioning. The combination
Draft -10- 2/26/92
-------
of this ^enhanced" steady-state testing, along with the IM240 and
purge/pressure tests allows for direct comparison of these
alternative I/M procedures. section 4.2 of this report provides a
more detailed discussion of the results of comparing the degree to
which the IM240 and the second-chance 2500 rpm/Idle test correlate
with the FTP.
In addition to assessing the IM240 for correlation with the
FTP, several other issues are addressed as part of the Hammond
study. Since dirty vehicles are repaired at the lab, the repair
effectiveness can be evaluated. The running loss tests allow EPA
to characterize the air quality impact of vehicles failing
pressure and purge tests and the effectiveness of repairing these
vehicles. The transient short test developed by the Colorado
Department of Health (CDH-226) as well as a variety of steady-
state tests are performed at the lab and can be evaluated as
potential I/M tests. Additionally, one of the IM240s performed at
the lab was restricted to inertia weight settings of 2,500 pounds
or 3,500 pounds. This restriction allowed EPA to evaluate the FTP
correlation effect of a more economical dynamometer (with fewer
inertia weight settings). We found that an inertia weight range
of 2,000 to 5,500 pounds using four inertia wheels (500, 1,000,
and 2,000 pounds with a fixed wheel of 2,000 pounds) is worth the
moderate additional cost.
The evidence displayed in section 4.2 (see below) and
Appendices E and G of this report graphically and quantitatively
shows the advantage of the high-tech IM240 test for the sample of
vehicles tested in Indiana in 1990 and 1991. The actual
calculations of the exhaust emission reductions of the several
short tests are more detailed in order to best reflect the actual
characteristics of the fleet as it ages and changes in technology
mix. A computer model called Tech4.1 is used to calculate
technology- and age-specific adjustment factors that represent the
effect of I/M programs of different types (the so-called "I/M
credit"), and these factors are built into the mobile source
emissions model MOBILE4.1. Section 4.3 of this document contains
details on the Tech4.1 model, while Appendix B lists the source
code for the model.
Finally, the Indiana testing program has revealed the true
seriousness of evaporative emission control system malfunctions
that develop during real world operation. Previous EPA testing
programs (i.e., those conducted during the last 10 years or so)
that did not make use of an operating I/M lane to screen and
recruit vehicles for more thorough laboratory testing have focused
mostly on vehicles that were about 5 years old or younger, in
order to most quickly obtain information on the latest generation
of new technology vehicles. When special efforts were made to
recruit high mileage vehicles, they tended to be vehicles that had
accumulated unusually high mileage for their age, for example
vehicles from owners with long commutes or who used their vehicles
for business during the day. EPA staff have been concerned for
Draft -11- 2/26/92
-------
some time that testing such vehicles was not giving a true picture
of evaporative emission problems, which may develop more as a
function of passing time than of miles driven; for example,
deterioration of rubber and plastic components would be more time-
than mileage-based. Also, the recruitment practices in the test
programs prior to the Indiana I/M lane program relied on owner
response to letters and phone calls. There has been concern that
this resulted in a different sample of vehicles, probably a sample
biased towards better maintenance condition than would be found if
owners could be solicited face-to-face, as they are in the Hammond
study (where the level of motorist participation has been
sufficiently high to ameliorate these concerns) . These
differences in study design explain why the Indiana program has
produced results very different from previous estimates of in-use
evaporative emissions. EPA's interest in the high-tech
evaporative purge and pressure tests has been in response to these
findings.
Because of the extensive detail of the evaporative emissions
findings from Indiana, the results of the testing are presented in
Appendix C, rather than illustrated with figures and tables here.
Briefly stated, the Indiana program showed that by 13 years of
age, nearly one-half of all vehicles will experience an
evaporative system failure that renders the control system
virtually ineffective, causing evaporative and running loss
emissions to increase by factors of up to 10 times. Nearly all of
these failures can be detected by the combination of the pressure
and purge tests. Use of only one of these tests finds at least
some of the problem vehicles. The problems can be repaired, and
vehicles will then pass a re-inspection using the pressure and/or
purge test. Appropriate repairs reduce emissions back to normal
levels. Of course, the purge and pressure tests cannot overcome
the limited control capacity designed into vehicles by their
manufacturers, so under certain conditions of temperature and fuel
volatility, both passing and repaired vehicles will fail to meet
the certification emission standard.
4.2 FTP HC/CQ Correlation Comparison Between the IM24Q and the
Second-chance 2500 rpm/Idle Test
This section focuses on the comparison of the IM240 transient
test (using cutpoints of 0.8 gpm HC and 15 gpm CO for the results
over the full 240 seconds, with a provision that a vehicle also
may pass by having emissions during the last 147 seconds of the
test less than or equal to 0.5 gpm HC and 12 gpm CO - see section
4.2.3 for a more detailed explanation of the two-ways-to-pass
criteria) to EPA's currently recommended second-chance 2500
rpm/Idle test procedure2, and details the evaluation criteria upon
2 Tierney, E., Herzog, E. and Snapp, L. "Recommended I/M Short Test
Procedures For the 1990s: Six Alternatives", U.S. EPA Technical Report
Number EPA-AA-TSS-90-3, January 1991.
Draft -12- 2/26/92
-------
which the comparison is based. This comparison shows how an I/M
program based on one of the better currently used (non-
dynamometer) I/M tests (second-chance 2500 rpm/Idle) can be
improved upon by changing to the IM240 test, which has a much
better classical correlation with the FTP than the idle or 2500
rpm/Idle test for matched pollutants (see the regression analyses
including R-squared values and scatter plots in Appendix E for an
illustration of this better correlation).
For the sake of the correlation analysis illustrated in
Appendix E, only 1983 and newer vehicles equipped with fuel
injection were considered3. The vehicles in this sample received
both the second-chance 2500 rpm/Idle test and the IM240 at the
Hammond test site. At this time, most I/M programs have not
adopted second-chance testing and the test algorithms recommended
in EPA's Alternative Test Procedure report, which calls for an
immediate second-chance test for vehicles that initially fail the
emission standards. Under the recommended procedures, vehicles
are preconditioned in a non-loaded state for three minutes at 2500
rpm prior to the second test. Second-chance testing was devised
to reduce, to the extent possible, the problem of falsely failing
vehicles. For the purposes of this comparison and to enable
analyses of the effectiveness of more stringent standards, second-
chance tests were performed on 1983 and newer fuel-injected
vehicles if their emissions exceeded 100 ppm HC or 0.5% CO on
their initial 2500 rpm/Idle tests. Note that these standards are
substantially tighter than the standards of 220 ppm HC and 1.2% CO
used in nearly all I/M programs on 1981 and later vehicles.
One of the central concerns in developing a new I/M short
test was to devise a test that would pass vehicles that would pass
the FTP and fail those that would fail the FTP. With that in
mind, the IM240 was devised by truncating, splicing, and otherwise
augmenting the first two hills of the FTP driving cycle. One of
the goals of the pilot program was to assess how well the IM240
correlates with the FTP. Since performing the FTP in the Indiana
lane was not a practical alternative, both IM240s and FTPs were
conducted in the lab after the vehicles were recruited in the I/M
lane. The lab results of the IM240 and the FTP showed excellent
correlation. One can conclude that the IM240 is an excellent
measurement of the true emissions of the vehicle at the time and
place it is performed, given the fuel being used at the time.
Comparing lab FTP and lane IM240 results is problematic for
several reasons, but still shows good correlation. Since the lab
tests are performed at a different time from the lane IM240s,
intervening factors, such as intermittent problems or changes in
The emission reduction benefits presented in Section 6, however, do reflect
the application of the IM240 to carbureted vehicles as well as fuel-
injected vehicles; the comparisons of IDR, EC rate, and failure rate for
the various I/M tests presented in Appendices G and H also address
carbureted vehicles.
Draft -13- 2/26/92
-------
the vehicle, may affect the results. For example, exhaust systems
are repaired, when needed, prior to the lab tests. Another major
problem making lab and lane comparisons difficult is the fact that
FTP tests are all done on Indolene fuel4 while lane tests are done
on the fuel in the tank of the vehicle as received. Also, the
equipment used in the lanes measured a lower maximum emission
value than the lab equipment; for example, a car would have pegged
the lane instrument for hydrocarbons at 13 gpm while in the lab it
actually measured 25 gpm. Temperature and preconditioning at the
lane were also often different than at the lab. For these
reasons, lab/lane comparisons say less about the actual
performance of the test and more about the influence real world
differences make on vehicle emissions. Nevertheless, both sets of
comparisons are presented in Appendix E of this report.
One of the conclusions evident from the data collected as
part of the Hammond study is that for fuel injected vehicles in
particular, the high-tech IM240 test has a better correlation with
the FTP than the conventional idle or 2500 rpm/Idle test. This
section and Appendix E present some illustrations of this better
correlation.
For example, one indication of better correlation is
demonstrated by higher R-squared values from least-squares
regressions with FTP emissions as the dependent variable and short
test emissions as the independent variable. Statistics for these
regressions are given in the regression analyses tables in
Appendix E.
The better correlation of the IM240 test also can be seen
visually in the scatter plots of emissions results from vehicles
which received all four tests (Appendix E). Separate plots of FTP
versus short test results are included for each type of fuel
injection (whether PFI or Throttle Body Injection (TBI)),
pollutant (HC, CO, and NOX) , and each short test type (except for
idle and 2500 rpm/Idle for NOXf since representative in-use NOX
emissions cannot be measured on these tests). Because of the wide
range of the data, the graphs showing all the data contain a lump
of points near the origin. To allow examination of the
correlation for vehicles emitting in this range, an enlargement of
the data in this range is also provided for each of the graphs in
Appendix E.
The above two indications (R-squared values and scatter
plots) of better correlation do not directly enter the calculation
of the emission reduction advantages of the IM240. In an I/M
program, predicting the absolute level of a vehicle's FTP
4 Indolene is a special test fuel whose properties are held constant. This
is necessary because the normal changes in fuel properties of commercial
fuel can change a car's emissions results even if all of the other test
procedure variables and vehicle variables did not change between tests.
Draft -14- 2/26/92
-------
emissions is not as important as identifying a large majority of
the vehicles whose emissions are likely to be high enough to merit
repair (which are, themselves, a minority of the overall in-use
fleet) . Also, the short test should pass vehicles that are not
malfunctioning, in order to avoid impacting owners of vehicles
which have emissions low enough to not merit repair. The figures
in Appendix G, which are discussed in the following sections,
graphically demonstrate the differences between the second-chance
2500 rpm/Idle test and IM240 in regard to these objectives.
The I/M test must also do a good job of ensuring that
vehicles that have shown emission reductions from repairs large
enough to pass re-inspection on the short test have also achieved
sizeable FTP reductions. Better performance of one short test
versus another in identifying vehicles as generally clean or dirty
will also ensure that fewer vehicles can pass reinspection without
achieving real FTP reductions. Therefore, it is clear that the
IM240 test will be the better enforcer of good repairs. Analysis
of data from vehicles in Indiana that were repaired at the
laboratory and retested on both the FTP and IM240 shows that
reductions measured by the two tests are highly correlated, even
better than the correlation discussed above. Figures and
statistics to illustrate this are also included in Appendix E.
4.2.1 I/M Test Assessment Criteria Overview
In assessing the overall effectiveness of an I/M testing
procedure, it is important to determine the test's effectiveness
in measuring and determining a variety of factors, including the
IDR, the failure rate, the error-of-commission rate, the failure
rate among vehicles that pass FTP standards, and the failure rate
for so-called "normal emitters," which may fail an FTP standard
but are clean enough to make it an issue whether they will benefit
much from normal repair procedures. Each of these is discussed,
in turn, below. Section 4.2.2 provides a more detailed discussion
of the same topics.
4.2.1.1 Excess Emission Identification Rate (IDR)
EPA commonly uses the rate of excess emissions identified
during an I/M test to objectively and quantitatively compare I/M
test procedures. As mentioned earlier, excess emissions are those
FTP-measured emissions that exceed the certification emission
standards for the vehicle under consideration. For example, a
vehicle certified to the 0.41 gpm HC standard that failed the
second-chance 2500 rpm/Idle I/M test with an FTP result of 2.00
gpm, would have excess emissions equalling 1.59 gpm (i.e., 2.00 -
0.41 = 1.59) .
The excess emissions identification rate (IDR) equals the sum
of the excess emissions for the vehicles failing the I/M test
divided by the total excess emissions (because of imperfect
Draft -15- 2/26/92
-------
correlation between I/M tests and the FTP, some I/M passing
vehicles also have excess emissions which are used for calculating
the total excess emissions) . Thus, assuming an I/M area that
tests 1000 vehicles, 100 of which are emitting 1.59 gpm excess
emissions each, while the I/M test fails (identifies) 80 of the
excess emitting vehicles, the excess emission identification rate
can be calculated as follows:
80 failing vehicles * 1.59 gpm excess per vehicle ^ _
100 vehicles * 1.59 gpm excess per vehicle
As can be seen in Figures 1 and 4 in Appendix G, the IM240 using
two-mode criteria has been shown to identify more excess emissions
among the cars tested at the Indiana lane than the second-chance
2500 rpm/Idle test with current I/M program cutpoints.
4.2.1.2 Failure Rate
As the IDR increases, the opportunity to identify vehicles
for emission repairs also increases. However, this measure is not
sufficient for determining which is the more efficient and cost-
effective I/M test. Other criteria must also be addressed before
such an assessment can be made. One such criterion is the failure
rate, which is calculated by dividing the number of failing
vehicles by the number of vehicles tested. For example:
50 vehicles failed I/M „. , nn
-rrrr 7—: 7—,. ' * 100 = 5% I/M failure rate
1000 vehicles tested
The ideal I/M test is one that fails all of the dirtiest
vehicles while passing those below the FTP standard or close to
it, but still above it. The potential emission reduction benefit
decreases as emission levels from a vehicle approach the standard,
because the prospect for effective repair diminishes. Thus,
achieving a high IDR in conjunction with a low failure rate (as a
result of identifying fewer vehicles passing or close to the
standard) efficiently utilizes resources. As the figures in
Appendix G show, tightening the cutpoints on the idle test to
achieve IDRs comparable to the IM240's results in increasing the
failure rate well beyond that of the IM240. For example, for 1983
and newer, PFI vehicles, the failure rate rose from 12% to 38%
when second-chance, two-speed cutpoints were tightened to 100 ppm
for HC and 0.5% for CO, even though the two-speed test's IDRs for
HC and CO were only 77% and 82% respectively (compared to the
IM240's 82% and 85% IDRs for HC and CO, and its 14% failure rate).
The remaining figures in Appendix G illustrate a similar
relationship between IDR and failure rate for tighter two-speed
cutpoints for both TBI and carbureted vehicles. For a more
specific, model year breakdown of failure rates among the vehicles
in the Hammond lane sample, by test type, see the Appendix
M,."Model Year Failure Rates by Test Type."
Draft -16- 2/26/92
-------
4.2.1.3 Error-of-Commission (Ec) Rate
Properly functioning vehicles which pass FTP standards
sometimes fail the 2500 rpm/Idle test; these are referred to as
false failures or as errors of commission (Ecs). When error-of-
commission vehicles are sent to repair shops, no emission control
system malfunctions exist. Often, the repair shop finds that the
vehicle now passes the test without any changes. These false
failures waste resources, annoy vehicle owners, and may lead to
emissions increases as a result of unnecessary and possibly
detrimental "repairs." Motor vehicle manufacturers see this as a
significant problem, since it can contribute to customer
dissatisfaction and increased warranty costs. An I/M program
seeking larger emission reductions through more stringent emission
test standards may actually increase the number of false failures.
The error-of-commission rate is, therefore, an important measure
for evaluating the accuracy of I/M tests.
To see how an error-of-commission rate is calculated, assume
an I/M area which tests 1000 vehicles, of which 100 fail the I/M
test, although only 50 of those 100 failing vehicles also exceed
their FTP standard for HC or CO. The error-of-commission rate
equals the number of vehicles that fail the I/M test while passing
the FTP for HC and CO, divided by the total number of vehicles
which were I/M tested:
50 vehicles failed I/M but passed FTP HC and CO ^ _ *
1000 vehicles tested ~
*Error-of-commission
As the. error-of-commission rate decreases, vehicle owner
satisfaction and acceptance of the I/M program increases. Thus,
while it is relatively easy to improve the IDR by making the I/M
test standards more stringent, this "improvement" comes at the
cost of potential increases in the error-of-commission rate.
4.2.1.4 Failure Rate Among FTP-Passing Vehicles
The risk of failing an I/M test with a clean vehicle is not
expressed very clearly, however, by stating fleet error-of-
commission rates. Fleet rates tend to be very low, but the impact
on any individual motorist can be very significant. A more
informative statistic than error-of-commission rate is the failure
rate among all inspected vehicles which still pass their FTP
standard. This indicates the risk to the owner of having a clean
vehicle failed. For the IM240 using the two-ways-to-pass
criteria, the failure rate for vehicles which pass the FTP is zero
for the sample tested in Indiana (Appendix G). While the false
failure rate for the second-chance two-speed test is initially
comparable to the IM240 using the two-speed cutpoints in current
use, tightening these cutpoints to improve IDR has the effect of
Draft -17- 2/26/92
-------
increasing the false failure for the steady-state test. For
example, as illustrated in Figures I through 3 of Appendix G, _for
1983 and newer PFI vehicles, tightening the steady-state cutpoints
from 220 ppm EC and 1.2% CO (the cutpoints most commonly used in
current I/M programs) to 100 ppm HC and 0.5% CO has the effect of
increasing the test's false failure rate from 0% to 13% - this,
even though the two-speed test's IDRs for both HC and CO still
fall appreciably below that of the IM240. For 1983 and newer TBI
vehicles, the same tightening of cutpoints achieves HC and CO IDRs
for the steady-state test that actually exceed those of the IM240
by a percentage point or two, but this at the cost of a false
failure rate of 20% compared to no false failures for the IM240
(Figures 4-6, Appendix G).
Even when the two-ways-to-pass criteria are not used for the
IM240, the false failure rate for the vehicles in EPA's sample was
only 0.8%, representing a total of 5 EC vehicles - still much
lower than the false failure rate for the steady-state test with
comparable IDRs. Since even this number of full-test failures was
unexpected, given the IM240's similarity to the FTP and the
looseness of the 0.8/15 cutpoints compared to the 0.41/3.4 new car
standards, section 4.2.3 is included to discuss these false
failures in depth.
4.2.1.5 "Normal Emitter" Failure Rate
The IM240 failure rate for normal emitters will also be
lower. ""Normal" emitters are defined, for the purposes of this
discussion, as those vehicles that emit less than twice the FTP HC
standard and less than three times the FTP CO standard. Normal
emitters include those vehicles that pass the FTP. Repairs on
such vehicles usually do not produce large emission reductions (at
least short of catalyst replacement, which EPA generally avoids in
its emission repair evaluations due to cost and because testing
after a new catalyst is installed would not necessarily indicate
what emissions will be after the catalyst "wears in"), their
emissions are sometimes increased by inept repairs, and they
account for little of the total excess emissions (9% of excess HC
for 1983 and newer PFI vehicles [page 8, Appendix H], 6% excess HC
for 1983 and newer TBI vehicles [page 21, Appendix H] and 3%
excess HC for 1981 and newer carbureted vehicles [page 31,
Appendix H] ) . Therefore, normal emitters are not the most cost
effective to identify for repairs. These vehicles often lack
overt defects. Those that fall above one of the FTP standards
obviously have some problem, but may only have suffered catalyst
deterioration (which is difficult to diagnose) or may have been
either poorly designed or built in the first place. Thus, the
marginal costs of identifying and effectively repairing these
vehicles may not always be worth the marginal benefits that could
be expected.
Draft -18- 2/26/92
-------
4.2.2 Detailed Discussion of Correlation and Test Assessment
The following analysis shows that the IM240 test using the
two-ways-to-pass criteria is considerably more powerful as an I/M
test than the second-chance 2500 rpm/Idle test for all technology
type vehicles, but especially newer tech, fuel-injected vehicles.
The analysis presumes that the IM240 is implemented to achieve
higher IDRs. Given that rationale, IM240 standards of 0.8 gpm HC
and 15 gpm CO for the full test and 0.5 gpm HC and 12 gpm CO for
the last 147 seconds were selected for this analysis. These IM240
standards achieve IDRs that are significantly higher than for the
present second-chance 2500 rpm/Idle standards, while maintaining a
false failure rate of zero.
This discussion is limited to PFI vehicles, as this is the
most commonly used fuel metering system on new vehicles. Throttle
body injection, which is less sophisticated, may also be used on a
significant proportion of the future fleet, though less than for
PFI. Therefore, although analogous figures and tables are
included in Appendices G and H for both TBI and carbureted
vehicles, they are not formally discussed.
Figure 1 in Appendix G provides a comparison of the present
second-chance 2500 rpm/Idle test using current standards (220 ppm
HC and 1.2% CO) to the more effective, high-tech IM240 test using
the two-ways-to-pass criteria. Note the following:
• The FTP excess emissions identification rates are 17% higher
for HC and 12% higher for CO with the IM240 as compared to
the second-chance 2500 rpm/Idle test using the 1.2%/220 ppm
standards.
• Neither test failed FTP passing vehicles. (Without the two-
ways-to-pass criteria, however, 5 FTP-passing fuel-injected
vehicles would have failed the IM240 yielding an error-of-
commission rate of 0.8% using the weighting factors
(discussed in section 2.4.1) to correct the lab sample to
simulate the lane sample. These vehicles are discussed in
section 4.2.3.)
• The IM240 increases the failure rate to 14% from 12% for the
second-chance 2500 rpm/Idle test.
Figure 2 in Appendix G illustrates the power of the IM240
test compared to the 2500 rpm/Idle test using more stringent idle
standards, currently in use in California. I/M programs might
consider California idle standards because the emission reduction
from the program can be increased and the cost of implementation
is relatively small.
California uses standards of 1.0% CO and 100 ppm HC for the
idle mode, while using 1.2% CO and 220 ppm HC for the 2500 mode.
In Figure 2, only the stringency of the 2500 rpm/Idle test is
Draft -19- 2/26/92
-------
increased, while the IM240 standards are the same as those used in
Figure 1 (see Appendix G for both figures). Note the following:
• The IDRs are still 6% higher for HC and 3% higher for CO with
the IM240 as compared to the second-chance 2500 rpm/Idle test
with more stringent standards.
• The second-chance 2500 rpm/Idle test failure rate using
California standards is 34% compared to only 14% for the
IM240. So even with the IM240's higher IDRs, significantly
fewer vehicles will need to be repaired.
• Thirteen percent of the FTP-passing vehicles fail the second-
chance 2500 rpm/Idle test, while none fail the IM240.
Sending this many cars for unnecessary repairs, while also
identifying less excess emissions, wastes resources.
• The normal emitter failure rate is only 2.3% for the IM240
versus 22% for the second-chance 2500 rpm/Idle test. This
means that the vehicles identified for repairs by the IM240
are more likely to achieve significant emission reductions.
In Appendix G, Figure 3 compares the same IM240 standard to
the more stringent standards of 0.5% CO and 100 ppm HC for both
modes of the second-chance 2500 rpm/Idle test for PFI vehicles,
while Figures 4, 5, and 6 present data analogous to the first
three figures, but this time for TBI vehicles, and Figures 7-9
present this information for 1981 and newer carbureted vehicles.
However, second chance testing was only performed on 1983 and
newer vehicles, so Figures 7, 8, and 9 only include second chance
results for 1983 and newer vehicles, not for 1981 and 1982
vehicles.
4.2.3 Avoiding Errors of Commission
I/M test procedures and standards that cause low emitting
vehicles to fail I/M tests are obviously undesirable. Using full-
test IM240 standards of 0.8 gpm HC and 15 gpm CO (without
incorporating 0.5/12 cutpoints for the last 147 seconds) results
in five of the 182 vehicles that received FTP tests (1983 and
newer with fuel injection) failing when they should have passed.
EPA staff were curious to find that even this small number of FTP-
passing vehicles failed the IM240 0.8/15 standards since the IM240
driving schedule is taken from the FTP and is a hot start test at
the Indiana lane. This section will investigate why these
vehicles falsely failed the IM240 test, and in so doing will
explain how and why two-mode criteria were developed. Given that
attempting to repair properly functioning vehicles wastes
resources and adversely affects vehicle owners, I/M programs and
vehicle manufacturers, EPA regards this as an important issue.
The causes of IM240 false failures can be categorized into
those that may have occurred due to errors in performing the test
Draft -20- 2/26/92
-------
and those that occurred because of the interactions of vehicle
design, fuel properties and test conditions.
In addressing these errors of commission, we will concentrate
on the vehicles actually tested rather than the reweighted
database discussed elsewhere. Of the 74 PFI equipped 1983 and
newer vehicles receiving FTPs, 24 passed the FTP, while one of
these 24 failed the IM240 standards of 0.8/15. Of the 108 TBI
1983 and newer vehicles receiving FTPs, 29 passed the FTP and 4 of
these failed the IM240 0.8/15 gpm standards. Other identifying
data and test results for these vehicles are listed in Table 4-3.
Table 4-3
FTP, Lane IM240. and "Tank Fuel" IM24Q Results
(grams per mile)
Veh
t
1674
685
1607
1624
1663
Fuel
Sys
PFI
TBI
TBI
TBI
TBI
Model
Year
1989
1989
1988
1983
1986
Mfr.
Suzuki
Chry.
GM
GM
GM
Carbon Monoxide
Tank
Lane Fuel
IM240 IM240 FTP
15.3
16.3
16.1
19.1
23.3
1.7
2.7
1.2
11.0
3.4
3.1
2.4
2.5
2.2
0.5
Hydrocarbons
Tank
Lane Fuel
IM240 IM240 FTP
0.73
0.4
0.78
1.37
0.76
0.1 0.37
0.06 0.12
0.14 0.24
0.64 0.38
0.15 0.1
Oxides
Lane
IM240
0.8
1.3
0.9
1.4
0.7
of Nitrogen
Tank
Fuel
IM240 FTP
0.2
1.1
1.1
1.6
0.6
0.2
0.8
0.6
1.4
0.4
The first thing to note is that the lane IM240 HC and CO
emissions are considerably higher than for the two tests performed
at the laboratory. The "tank fuel" IM240 is the first test the
vehicles receive at the laboratory, usually using the same fuel
that was in the tank for the lane test. Since the vehicles are
stored outside prior to testing, each is preconditioned before
receiving the tank fuel IM240. The preconditioning consists of a
3 mile drive at 50 mph (on the road) immediately before the test.
The comparison of the lane and "tank fuel" tests strongly suggests
that it was some factor at the lane that resulted in these
vehicles being errors of commission, not some inherent
unsuitability of the IM240 test itself.
The following analysis of the test results suggests that
vehicle 1674, the only PFI error-of-commission vehicle, probably
failed as a result of testing errors at the lane. The four TBI
error-of-commission vehicles are likely the result of interactions
of vehicle design, fuel properties and test conditions.
4.2.3.1
Vehicle 1674
Draft
-21-
2/26/92
-------
One possible explanation for vehicle 1674's results on the
lane IM240 test is that it was run at too high an inertia weight
(IW) setting on the dynamometer. This is important because the
fuel metering system enrichens under heavy loads, causing CO
emissions to increase, which could explain the inconsistently high
CO result at the I/M lane. Also, NOX emissions typically increase
in conjunction with engine load, while fuel economy (FE)
decreases, suggesting one method for determining whether the
dynamometer settings were, in fact, inconsistent. Table 4-4
provides fuel economy and NOX results for vehicle 1674 for the
various tests conducted.
Table 4-4
Vehicle 1674 NOy, FE Results Versus Dynamometer Settings
NOX (gpm)
Lane
IM240
0.83
Fuel Economy (mpg) 29.61
IW Setting (Ibs.)
Road Load (hp)
2125
Tank Fuel
IM;
0.24
31.19
2125
indolene
IM24Q
0.12
33.90
2125
6.5
IW=2500#
IM24Q
0.18
32.43
2500
6.5
Note that the lane IM240 NOX is about 4 times greater than
any of the other NOX results, including the 2,500 pound inertia
weight test that was performed at the lab to simulate an
economical dynamometer with only two inertia weight settings
(i.e., 2,500 or 3,500 pounds). The lane IM240 shows poorer fuel
economy than for any of the other tests. Another indication
suggesting that vehicle 1674 represents a unique situation is the
fact that the other 4 vehicles which were errors of commission had
fairly consistent NOX emissions, (see Table 4-3) even though their
HC and CO emissions deviated considerably from lane to lab.
Another important consideration is the fact that the vehicle
tested at the lane just before vehicle 1674 used dynamometer
settings of 3,125 pounds and 7 horsepower (hp) . Vehicle 1674
should have been tested at 2,125 pounds and 7 hp. The
comparatively high NOX emissions and fuel consumption results in
conjunction with the consistent NOX results for the other 4
vehicles suggest that 1674 was, in fact, tested at the previous
vehicle's inertia weight setting of 3,125 pounds, 47% higher than
the 2,125 pound setting required.
The settings at the Hammond lane are selected manually, and
there is no feedback circuit to document the setting. The only
record of the setting is a handwritten data sheet completed by the
Draft
-22-
2/26/92
-------
lane technician, who may have recorded the correct dynamometer
setting, while failing to change the actual setting on the
dynamometer. Given the importance of using correct dynamometer
inertia weight settings, automatic selection of the IW settings is
essential to ensure that vehicles are tested correctly. EPA is
developing quality control procedures to insure correct settings
are chosen in I/M program lanes.
4.2.3.2 Interactions of Vehicle Design, Fuel Propertiesf and
Test Conditions
Three techniques were considered for preventing IM240 false
failures caused by the interaction of vehicle design, fuel
properties and test conditions.
1. Reduce the stringency of the pass/fail standard.
2. Allow second-chance testing for failing vehicles that are
close to the standard and precondition vehicles prior to the
second-chance IM240.
3. Since the IM240 is a 240-second test, use the latter part of
the test as a separate, second chance to pass, if the vehicle
fails over the entire test. For example, consider a vehicle
with a composite result of 0.7 gpm HC and 16 gpm CO, with
Bag-2 results of 0.4 and 11 gpm. Because it only slightly
exceeds the 15 gpm CO standard, the test software could be
designed to also consider the emissions in the 147-second
mode of the test (i.e., Bag 2). Vehicles pass if they pass
either the full-test outpoints or the Bag-2 outpoints.
Assuming Bag-2 outpoints of 0.5 gpm HC and 12 gpm CO, which
are significantly more stringent than the composite
standards, the vehicle in this example would pass because it
does not fail both sets of cutpoints. The logic supporting
the suitability of these pass/fail criteria will be discussed
later.
Each of the three strategies for avoiding false failures will be
further discussed below.
The first alternative, reducing the stringency of the
pass/fail standards, is unacceptable. Although this strategy
decreases errors of commission, it also unacceptably reduces IDR,
as shown in the tables in Appendix H. For example, page H-24
lists a variety of potential cutpoints of varying stringency, the
least stringent of which is 1.6/20 gpm. Three of the 4 error-of-
commission TBI vehicles were found to pass at this standard, but a
number of vehicles which were high emitters did also, lowering the
HC IDR from 89% to 61% and the CO IDR from 80% to 58%. These IDR
reductions would compromise the purpose of this high-tech test and
reduce the benefits from an I/M program.
Draft -23- 2/26/92
-------
Second-chance testing and the two-ways-to-pass criteria are
better ways of avoiding false failures. The theory behind this is
as follows. Assuming that the test was correctly performed in the
first place, the most likely reason that a properly functioning
vehicle would fail an IM240 is that the evaporative canister was
highly loaded with fuel vapors and that the vapors were being
purged into the engine during the test. This has been a
significant cause of false failures in existing I/M programs and
it has been shown that highly loaded canisters can cause both high
HC and CO emissions, even though the feedback fuel metering system
is functioning properly.
Since the canister is being purged during the IM240, the fuel
vapor concentration from the canister continually decreases during
IM240 operation. The decreasing fuel vapor concentration results
in decreasing HC and CO emissions. So, Bag-2 results should be
lower than the composite results, on a gram per mile basis. On
the other hand, if the vehicle is actually malfunctioning, Bag-2
emissions should remain high. For this reason, second chance
tests after preconditioning, as shown for current 2500 rpm/Idle
test, should be less influenced by canister purge.
Adding Bag-2 results to the pass/fail criteria completely
solves the error-of-commission problem for the vehicles in our
test sample. As shown in Table 4-5, all 4 of the TBI error-of-
commission vehicles will pass the lane IM240 if the two-ways-to-
pass criteria are applied such that a vehicle fails only by
exceeding both 0.8/15 gpm for the composite IM240 and Bag-2
cutpoints of 0.5/12 gpm.
Table 4-5
Two—Ways—To—Pass Criteria Applied to Lane IM240 TBI Error—of-
Commission Vehicles
(grams per mile)
Vehicle
Number
685
1607
1624
1663
Lane IM240
Bag-1
;
0
2
3
1
H£
.73
.01
.60
.94
£Q
29.
44.
53.
55.
1
4
6
0
0
0
0
0
Lane IM240
Bag-2
HC.
.28
.30
.47
.28
Lane IM240
Composite
£P_ HC.
11
4.
5.
10
.2
9
0
.5
0
0
1
0
.40
.78
.37
.76
m
16.31
16.09
19.08
23.28
Using two-ways-to-pass criteria has the added benefit of
reducing the normal emitter failure rate for TBI vehicles from
5.2% to 4.1%. At the same time, the overall TBI failure rate
Draft
-24-
2/26/92
-------
decreases from 22% to 21% without a perceptible reduction in the
IDRs which remain at 89% and 80% for HC and CO respectively.
Analogous benefits were observed for the PFI sample, where the
normal emitter failure rate for PFI vehicles decreases from 2.7%
to 2.3%, and the overall PFI failure rate decreases from 15% to
14% without a perceptible reduction in the IDRs which remain at
82% and 85% for HC and CO respectively.
All of the TBI error-of-commission vehicles passed 0.8/15 gpm
standards on the tank fuel IM240, suggesting that performing a
second-chance test after preconditioning would eliminate errors-
of-commission. Second-chance testing would increase the cost of
the testing process, however, so it may be best reserved as a last
resort. Given the efficacy of the two-ways-to-pass criteria, the
error-of-commission problem may be fully resolved. Nevertheless,
second-chance tests on select vehicles during high temperature
days in spring or fall would serve to insure no false failures.
Since such a procedure could be used on only the small fraction of
vehicles that are slightly above one or both of the standards, the
impact on test costs should be minimal.
The following discusses how vehicle design, fuel properties,
and test conditions can interact to cause false failures.
Considering these variables can facilitate designing algorithms
that determine when second-chance testing should be performed.
The success of utilizing the two-ways-to-pass criteria indicates
that second chance testing will not often be necessary. This
discussion shows that these 4 TBI EC vehicles were tested under
rather unique conditions and explains why second chance testing
may be useful for some cars.
HC emissions and especially CO emissions have both been shown
to vary with evaporative canister loading, so EPA investigated
variables that affect canister loading such as gasoline Reid Vapor
Pressure (RVP), ambient temperature, and the exhaust interaction
caused by canister purge flow (see Table 4-6).
While the specific fuel RVP for the lane IM240 is unknown,
the test dates do provide some information regarding RVP. The RVP
at service stations in the Hammond area was regulated by EPA to a
maximum 10.5 psi between June 1 and September 15. Three of the
error-of-commission vehicles were tested outside of this period.
ASTM guidelines allow RVP up to 13.5 psi in April and September,
and up to 15 psi in December. Since these are only guidelines,
they can be exceeded. Unseasonably warm test conditions can cause
high RVP fuels to vaporize at an excessive rate, causing high
canister loadings. If the RVP of the fuel in these error-of-
commission vehicles was 13.5 psi or higher, it is plausible that
these three vehicles had highly loaded canisters and/or high vapor
generation rates in their fuel tanks.
Draft -25- 2/26/92
-------
Table 4-6
Variables Affectincr CO Emissions
Vehicle
Number
685
1607
1624
1663
Canister
Purge
Volume
(liters)
35.7
45.7
79.7
84.9
Engine
Displacement
(liters)
2.3
4.5
5.7
4.3
I/M Lane
Temperature
(° F)
100
71
57
87
Lane
Test
9/6/90
10/11/90
12/10/90
4/6/91
Canister loading and fuel tank vapor generation, in turn,
generally increase with increasing ambient temperature. Vehicle
685 was the only error-of-commission vehicle that was tested
during the RVP-control-season, and the ambient temperature
recorded at the time of testing was 100°F, considerably above the
typical FTP test temperature of 75°F. The lane temperatures were
also unseasonably high for vehicles 1624 and 1663.
If the canister is highly loaded, CO is likely to increase as
a function of increasing purge flow. The median purge volume for
all vehicles tested at the lane was approximately 30 liters.
Vehicle 1663 had an unusually high purge volume of 85 liters and
was tested at an unusually high temperature (87°F) in April when
RVP is not regulated. Similarly, vehicle 1624 had a purge volume
of 80 liters and was tested at a lane temperature of 57°F in
December. If the outdoor temperature was in the thirties and the
fuel RVP was 15 psi, the warming of the fuel tank that would occur
while in the I/M station is likely to have led to a high canister
loading. High canister loading combined with the high purge rate
probably caused the high emissions. Table 4-5 shows that the Bag-
1 emissions were more than 5 times higher than the Bag-2
emissions, further indicating that a highly loaded canister was
sufficiently purged during the first mode to achieve relatively
clean Bag-2 results.
The evidence suggests that the fuel RVP and the conditions in
the Hammond station caused high CO emissions. Indiana is unique
in that existing buildings were adapted for I/M testing purposes
and, as in the case of the Hammond station, tend to be closed-
architecture structures like warehouses rather than open, airy
test stations. The likelihood of the kind of false failures seen
here should be much less in actual I/M programs for two reasons:
fuel RVP will be regulated to 9.0 psi during the summer months,
and the factors contributing to canister loading in the Hammond
lane are better controlled in typical I/M programs. The Hammond
IM240 lane residence time is 10-20 minutes for the research and
Draft
-26-
2/26/92
-------
development type testing; also, there is considerably less
ventilation (thus higher temperatures on sunny days) than in
typical I/M facilities. In designing future facilities for
enhanced I/M testing lanes it will be important to keep those
considerations in mind, as is already the case in most centralized
programs. Further, the number of vehicles receiving second-chance
tests can be minimized by developing algorithms that consider
canister purge rate, higher than normal ambient temperatures, and
probable fuel RVP in determining whether to allow second-chance
testing.
4.2.3.3 Error—of—Commission Summary
EPA views false failures as a significant problem and is
committed to investigating and utilizing strategies to prevent
their occurrence. The IM240 can help, because it allows
prevention strategies, such as two-ways-to-pass criteria, that are
not relevant to steady-state I/M procedures. The following
summarizes how errors of commission can be prevented in I/M
programs by using the IM240.
• There is strong evidence indicating that no PFI Ecs would have
occurred if the correct dynamometer inertia weight setting had
been used. Vehicle 1674's dynamometer inertia weight setting
was probably too high. EPA believes that automation and
quality control procedures will prevent such mistakes.
• All of the false failures in EPA's sample could have been
prevented by implementing pass/fail criteria that consider
Bag-2 results in addition to composite results. Under this
strategy, vehicles fail only if their composite emissions
exceed 0.8/15 and their Bag-2 emissions exceed 0.5/12 gpm.
• Second-chance testing after preconditioning on marginally
failing vehicles, especially those showing high purge rates
and tested at higher than normal temperatures, could be used
as a fail-safe mechanism to insure that no errors-of-
commission occur.
In conclusion, the data indicate that these false failures
were caused by unique circumstances rather than inherent test
procedure flaws or standards that are too stringent. Experience
with these vehicles suggests that the problem can be solved
through automation, quality control, two-ways-to-pass criteria,
and preconditioned second-chance tests on marginal failures.
4.3 Approval of Alternative Tests
Although the IM240, purge, and pressure tests represent EPA's
current trio of recommended high-tech tests, we do not rule out
the possibility of future, valid alternatives to these tests,
including fast-pass and fast-fail transient testing strategies
Draft -27- 2/26/92
-------
(see section 4.4, "Transient Testing Fast-Pass/Fast-Fail
Strategies"). States may seek approval of such strategies,
contingent upon the state's demonstrating to EPA's satisfaction
that such strategies are at least as effective as EPA's
recommended tests at identifying excess emissions while
maintaining a comparably low error-of-commission rate. As the
sheer number of analyses contained in this report can attest, EPA
does not promulgate new testing strategies capriciously. Before
proposing the IM240, purge, and pressure tests, EPA amassed a
compelling body of data on each through pilot programs conducted
in Maryland and Indiana (see section 4.1, "Recent I/M Test
Programs (Maryland and Indiana)" for further discussion of these
pilot studies) . Rigorous evaluations of each were conducted to
determine their effectiveness at identifying excess emissions
while maintaining low error-of-commission rates. Economic
analyses were also conducted to assess the cost effectiveness of
the tests, as no degree of technical excellence will justify a
testing strategy that is exorbitant in its overall cost. For
example, the FTP is the hallmark against which I/M testing
strategies are measured, but cannot itself be used as an I/M test,
given its cost.
Of the potential alternatives to EPA's recommended tests, the
one which has garnered the most attention is the suggestion by
some that steady-state loaded testing using a simple non-inertial
dynamometer be used to perform the purge check. EPA pursued
transient testing instead of steady-state because our best
engineering and technical judgement suggested that steady-state
testing as a mechanism for conducting the purge check would lead
to decreased emission reduction benefits, higher errors-of-
commission, and, ironically, higher overall costs per ton of
emission reductions produced. The rationale behind the assumption
that higher errors-of-commission rates would result is the fact
that purge strategies vary from vehicle to vehicle, and the
possibility of developing a steady-state test that successfully
addresses this variety is small to none. The result of failing to
address the full range of purge strategies is easy to predict:
Cars that should pass will fail, leading to unnecessary expense
and hardship for motorists, with no environmental benefit.
It has also been suggested by some that a loaded steady-state
test is an adequate alternative to transient emission testing.
One of the problems with this proposal arises from the requirement
under section 182(c)(3) of the Act that programs in enhanced I/M
areas achieve NOX benefits. EPA has found that NOX emission
testing (as opposed to visual inspection of emission control
devices) is essential for NO>: emission reductions. Further EPA
believes that a reliable steady-state test for NOx does not exist
and is not likely to be possible. Nevertheless, as indicated
earlier, EPA is open to demonstrations by states or their
representatives that proposed alternative testing strategies are
equal or superior to EPA's proposed tests in terms of identifying
excess emissions and keeping false failures to a minimum.
Draft -28-
2/26/92
-------
4 . 4 Transient Testing Fast—Pass/Fast-Fail Strategies
Among the alternative testing strategies that make
environmental and economic sense, the potential for fast-pass and
fast-fail transient testing ranks the highest. EPA is in the
process of looking at potential fast-pass and fast-fail
strategies, and preliminary results suggest that roughly 33% of
the vehicles tested could be fast passed or failed based upon
analysis of data gathered during the first 93 seconds of the IM240
(i.e., Bag 1) using separate fast-pass and fast-fail cutpoints.
In evaluating potential fast-fail criteria, EPA looked at a
sample of 4,158 1983 and newer vehicles tested at the Hammond
IM240 lane described in section 4.1, 1,033 (or 24.8%) of which
failed the IM240. 298 (or 28.8%) of the 1,033 vehicles that
failed would have failed within the first 93 seconds of the test
if Bag 1 cutpoints of 2.5 gpm HC, 50 gpm CO, and 5.0 gpm NOX were
used; there were no errors-of-commission. Although stricter Bag I
cutpoints could be used to increase the percentage of fast-failed
vehicles, the error-of-commission (Ec) rate would also rise. In
turn, when fast-pass Bag 1 cutpoints of 0.41/3.4/1.0 were used,
1,074 (or 34.4%) of the 3,125 vehicles that passed overall passed
within the first 93 seconds of the test. Seven additional false
passes were also recorded, resulting in an error-of-omission rate
of 0.7%. Tightening the fast-pass cutpoints to 0.25/1.5/1.0
eliminates the false passes but also reduces the fast-pass rate to
13.2%. Table 4-7 provides further details on the Bag 1 cutpoints
looked at in this analysis. While more development of fast-pass
and fast-fail criteria is needed, it is reasonable to conclude
that criteria can be developed to accurately pass and fail about
one third of all vehicles tested after only 93 seconds rather than
the full 240 seconds. Further, EPA will soon begin collecting
second-by-second IM240 data. This will allow the development of
algorithms that will permit especially clean cars to pass well
before 93 second, and others to pass after 93 seconds, but well
before 240 seconds. Once the algorithms are developed, only
vehicles that are close to the cutpoints are expected to continue
for the full 240 seconds to ensure that they are not falsely
failed.
Draft -29- 2/26/92
-------
Table 4-7
IM24Q Bag-1 Fast -Pa
-Fail Analysis
Fail
IM240
Total
1033
1033
1033
Fail
Fast-Fail
Total
1297
450
298
Fail
Both
902
445
298
Fail
Fast-Fail
Only
395
5
0
Fast-
Fail
^D rate
87.3%
43.1%
28.8%
EG Rate
30.5%
1.1%
0.0%
Pass
IM240
Total
3125
3152
3125
Pass
Fast-Pass
Total
2861
1081
413
Pass
Both
2730
1074
413
Pass
Fast-Pass
Only
131
7
0
Fast-
Pass
ID Rate
87.4%
34.4%
13.2%
False-
Pass
Rate
12.7%
0.7%
0-0%
Fast Fail
0.8/15/2.5
2.0/40/4.0
2.5/50/5.0
Fast Pass
0.8/15/2.5
0.41/3.4/1.0
0.25/1.5/1.0
Another area that EPA is investigating is the possibility
that the overall test time may be reduced. The IM240 is itself an
FTP-like short test based upon a modified and condensed driving
cycle that takes as its reference the LA-4 cycle used in the FTP.
EPA is currently investigating the possibility of further
abbreviating the test by comparing how well data from either of
the two hills of the IM240 driving cycle (i.e., Bag 1 and Bag 2)
taken separately correlate with the current two-mode IM240.
Preliminary results based upon a sample of 188 1983 and newer
fuel-injected vehicles which were recruited at the Indiana I/M
lane and subsequently retested under lab conditions (which
included each vehicle receiving an FTP) suggest that analysis of
Bag 2 (i.e., emissions sampled during the second hill of the IM240
driving cycle) may be about as good as the full IM240 when it
comes to identifying vehicles that would pass or fail on the basis
of the full test. Using Bag-2 cutpoints of 0.60/12 for HC and CO
respectively, and looking at Bag-2 results only, 90% of the excess
HC emissions and 84% of the excess CO emissions were identified,
with an EC rate of 0.7%, as compared to the full IM240 using the
0.8/15 cutpoints only (i.e., no Bag-2 cutpoints), which identified
82% and 85% of the excess HC and CO emission, respectively, with
an EC rate of 0.8%. These findings come with the caveat that they
are based upon a Bag 2 sample which followed the Bag 1 portion of
the driving cycle, meaning that Bag 2's high degree of correlation
with the IM240 may be the result of preconditioning occuring
during the Bag 1 phase. Even if such is, in fact, the case, the
prospect of a shorter overall test time still seems good since
adequate preconditioning for Bag 2 could probably be obtained in
less than 93 seconds by modifying Bag 1 to use a higher speed over
less time.
To determine whether or not preconditioning is a factor, EPA
plans to begin testing a sample of vehicles using what is, in
effect, a three bag test, beginning with the second hill of the
Draft
-30-
2/26/92
-------
IM240 driving cycle up front (hence no possibility for "Bag 1"
preconditioning) followed by a regular IM240. This study should
help EPA determine (1) whether or not preconditioning is a factor
in Bag 2's high degree of correlation with the full test and (2)
whether preconditioning would improve the correlation between Bag
1 and the full test. In addition, as mentioned above, EPA will
soon begin collecting second-by-second data, which will allow us
to determine whether or not there is some point in the testing
cycle by which time if vehicle X is emitting at a rate Y, it will
clearly pass or fail.
4.5 Estimating I/M Testing Credits for MQBILE4 1
As stated earlier, the data from the Indiana program were
analyzed and re-assembled in a manner which allows a comparison of
I/M program designs over a wide range of time frames and
conditions, rather than just for the particular sample of vehicles
tested in Indiana. This method for estimating the effect of I/M
program options on exhaust emissions (i.e., the I/M credit) is
fairly simple. Using the emission factor database, the fraction
of total vehicle FTP emissions which is identified by a particular
short test is determined for each of four strata of vehicles based
on FTP emission level. Using a subsample of vehicles which have
been repaired, the emission reductions attributable to these I/M-
triggered repairs is estimated for each strata. The Tech4.1 model
is used to calculate the emissions impact of a given short test by
reducing the total FTP emissions identified at each age by the
estimated emission reductions resulting from I/M repairs. When
the fleet average emission rates are recalculated by considering
the strata, the difference between the I/M and non-I/M case is
stored as an I/M credit for use in MOBILE4.1.
4.5.1 Tech4.1 Background and Assumptions
The Tech4.1 model divides the 1981 and newer light-duty
gasoline vehicle (LDGV) sample into several groups. The 1981 and
1982 model years are kept separate from the 1983+ model years. In
each model year group, the vehicles are divided by technology type
into closed-loop port fuel injection (PFI), closed-loop throttle-
body fuel injection (TBI), closed-loop carbureted (Garb) and all
(carbureted and fuel injected) open-loop (Oplp). Further, each of
these groups are divided into emission levels for Normal, High,
Very High and Super emitters. Table F-l in Appendix F provides
details on national fleet averages for passenger vehicle
distributions by model year and technology type; Tables F-2 and F-
3 provide data on emitter groups by model year group, technology
type, emission levels and rates, and mileage accumulation.
The model allows a separate IDE and repair effectiveness
estimate for each of these divisions of the data by I/M test type,
as illustrated in Table 4-8. It should be noted that the IDRs
listed in Table 4-8 for the traditional I/M tests (i.e., the idle
Draft -31- 2/26/92
-------
and 2500 rpm/Idle tests) are based upon historical emission factor
data gathered at EPA's Motor Vehicle Emission Lab (MVEL) in Ann
Arbor, Michigan, as well as elsewhere, and not at the Hammond,
Indiana test lane. The IDRs mentioned elsewhere in this report
(Appendices G and H, for example) were derived as part of the
Hammond study, and are not divided by emitter group, as is the
case in Table 4-8.
In practice, because of small sample sizes, several of the
divisions represented in Table 4-8 share information. _ In
particular, the small amount of steady-state Loaded/Idle testing
required that all vehicles without Loaded/Idle testing be assumed
to have the same short test result for Loaded/Idle testing as they
had for the 2500 rpm/Idle test for the purpose of determining the
IDR for the Loaded/Idle test.
For Super emitters (vehicles over 10 gpm HC or 150 gpm CO) ,
the IDR is the same for all technologies, but is separate for
1981-82 and 1983+ vehicles. Most 1981-82 vehicles are carbureted.
Most 1983+ vehicles are fuel injected. There are no Super open-
loop vehicles in the sample.
The two fuel injection groups in the 1981-82 grouping use the
same IDRs for Very High emitters (vehicles over 1.64 gpm HC or
13.6 gpm CO), High emitters (vehicles over 0.82 gpm HC or 10.2 gpm
CO) and Normals. In some cases, such as the High emitters, the
1983+ open-loop and carbureted technologies were combined.
Repair effectiveness (Table 4-9) was determined by dividing the
repaired sample by technology into PFI, TBI and Carb. Model year
grouping was not used. To be eligible for the repair effectiveness
analysis, a repaired vehicle must first fail the short test of
interest before repairs, and then after repairs, must pass the same
short test. Thus, different samples of repaired vehicles were used
for each short test. The sample was then ranked by before repair
emission level and divided into four equal-sized subgroups of
increasingly more severe emissions failure. The before and after
repair emission levels of each subgroup were then determined.
When plotted, before repair emission level versus after repair
emission level, these four emission failure points represent a
technology specific function used to determine repair effectiveness.
Generally, the vehicles with higher before repair emission levels
get larger absolute emission reductions from repairs, but do not
reach as clean a level after repairs as vehicles which began with a
milder degree of emission failure. Before repair emission levels of
High, Very High and Super emitters in many cases will fall between
the calculated points, and so had their after repair emission levels
determined by interpolation. Before repair emission levels lower
than the lowest point were interpolated between the low point and
zero. Before repair emission levels above the highest point were
assumed to be the same as the highest point.
Draft -32- 2/26/92
-------
Since few of the repaired vehicles had Loaded/Idle or IM240
testing data, it was assumed that vehicles repaired using a
Loaded/Idle test and the IM240 test would use the same before and
after repair curve as the 2500 rpm/Idle testing. EPA is being
conservative in assuming that vehicles failing the Loaded/Idle test
or the IM240 test, after repair, will have the same after repair
emission level as we estimate for the 2500 rpm/Idle test vehicles.
However, since the failure rates of vehicles in the high emitter
groups are larger for the Loaded/Idle test and the IM240 transient
test than for the 2500 rpm/Idle test, the total emission reduction
due to repairs will be larger.
As an example, the zero mile HC emission level of Very High
emitters for 1983+ PFI vehicles is 2.019 gpm and their slope is
taken to be the same slope as the Normals (i.e., 0.0115 gpm/10,000
miles) (see Table F-2) . At 5 years old, the average mileage of
these vehicles will be 60,829 miles. The non-I/M emission level is
therefore:
2.019 + .0115*6.0829 = 2.089 gpm
Assuming a 2500 rpm/Idle test is done, the HC IDR (see Table
4-8) for this group is 0.6187, or nearly 62% of the total emissions
from these vehicles is identified by failing vehicles using the 2500
rpm/Idle test. Table 4-9 shows the results of a data analysis
indicating the predicted average after repair levels given the
before repair emission level. The series of points in the table are
used to predict the after repair emission levels for all emitter
groupings, only dependent on the average before repair emission
level for that group. The before repair emission level falls
between the two emission levels 1.9846 and 3.9314. The after repair
levels for these emissions are 0.59231 and 1.0271 respectively.
Interpolating, the after repair level for the 2.089 gpm before
repair emission level is:
0.59231+((2.089-1.9846)/(3.9314-1.9846))*(1.0271-0.59231)=0.6153
Therefore the after repair HC emission level for 5 year old,
1983+ PFI vehicles tested on the 2500 rpm/Idle test is:
%
0.6187*0.6153 + (1-0.6187)*2.089 - 1.1772 gpm
Comparing the I/M and non-I/M cases indicates the "I/M benefit"
among Very High emitters.
(2.089-1.1772)72.089 = 43.6%
Draft -33- 2/26/92
-------
Table 4-8
SuDer Emitters
Test
Idle Test
Idle Test
2500/Idle
2500/Idle
Load/ Idle
Load/ Idle
IM240
IM240
Idle Test
Idle Test
2500/Idle
2500/Idle
Load/ Idle
Load/Idle
IM240
IM240
Model
Years
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
PFI
0.
0.
0.
0.
0.
0.
1.
1.
0.
0.
0.
0.
0.
0.
0.
0.
fld
6048
8978
6523
8978
6523
8978
0000
0000
2736
5676
2736
6187
2736
6187
8920
8800
PFI
CQ
0.6968
0.9656
0.8577
0.9656
0.8577
0.9656
1.0000
1.0000
Very
0.3231
0.6129
0.3231
0.7465
0.3231
0.7465
0.9460
0.9400
TBI
HC.
0.6048
0.8978
0.6523
0.8978
0.6523
0.8978
1.0000
1.0000
TBI
0.
0.
0.
0-
0.
0-
1.
1.
£Q
6968
9656
8577
9656
8577
9656
0000
0000
Garb
HG
0.6048
0.8978
0.6523
0.8978
0.6523
0.8978
1.0000
1.0000
Garb
£0.
0.6968
0.9656
0.8577
0.9656
0.8577
0.9656
1.0000
1.0000
Oplp
HC
0.0000
0.0000
0-0000
0.0000
0.0000
0.0000
0.0000
0.0000
Oplp
£Q
0.0000
0.0000
0.0000
0-0000
0-0000
0.0000
0.0000
0.0000
Hicrh Emitters
0.2736
0.2651
0.2736
0.3616
0.2736
0.3904
0.8770
0.8600
0.
0.
0.
0.
0.
0.
0.
0.
3231
2695
3231
4206
3231
4337
8750
8600
0.3858
0.3640
0.4789
0.5684
0.5476
0.5684
0.8760
0.9400
0.4108
0.3180
0.5331
0.6832
0.6037
0.6832
0.8680
0.8300
0.4568
0.3640
0.6197
0.5684
0.6197
0.5684
0.8760
0.9400
0.5194
0.3180
0.6162
0.6832
0.6162
0.6832
0.8680
0.8300
High Emitters
Idle Test
Idle Test
2500/Idle
2500/Idle
Load/ Idle
Load/ Idle
IM240
IM240
Idle Test
Idle Test
2500/Idle
2500/Idle
Load/Idle
Load/ Idle
IM240
IM240
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
81-82
83+
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0506
2507
0506
3436
0506
3866
0930
1300
0556
0360
0556
0575
0556
0907
0450
0500
0.1135
0.2208
0.1135
0.3501
0.1135
0.3937
0-0600
0.0800
Nor
0.0774
0.0414
0.0774
0,0694
0.0774
0.1023
0.0560
0.0600
0.0506
0.0336
0.0506
0.1924
0.0506
0.1924
0.5080
0.5100
znaJL EmJL
0.0139
0.0425
0.0139
0.0476
0.0139
0.0712
0.0970
0.1000
0.
0.
0.
0.
0.
0.
0.
0.
1135
0613
1135
1532
1135
1532
4190
4200
0.0563
0-0694
0.0898
0.0694
0.0910
0.0694
0.1820
0.1800
0.0492
0.0415
0.0834
0.0415
0.0896
0-0415
0.2060
0.2200
0.2274
0.0694
0.2274
0.0694
0.2274
0-0694
0.1820
0.1800
0.1522
0.0415
0.1522
0.0415
0.1522
0.0415
0.2060
0.2200
tters
0.
0-
0.
0.
0.
0.
0.
0.
0139
0436
0139
0514
0139
0739
0750
0800
0.0188
0.0023
0.0371
0-0140
0.0371
0.0140
0.1340
0.2400
0-0204
0.0078
0.0427
0.0156
0-0427
0.0156
0.1200
0.2100
0.0093
0.0023
0-0201
0.0065
0.0201
0-0231
0.1340
0.2400
0.0131
0.0078
0.0317
0.0208
0.0317
0.0403
0.1200
0.2100
Identification Rate (IDR) is the fraction of the total sample emissions
from vehicles failing the short test.
Draft
-34-
2/26/92
-------
Table 4-9
Short Test Repair Effectiveness
PFI/TBI Carb/Oplp
Before After Before After
Repair Repair Repair Repair
Idle Test HC
Idle Test HC
Idle Test HC
Idle Test HC
Idle Test
Idle Test
Idle Test
Idle Test
2500 rpm/Idle* HC?
2500 rpm/Idle* HC
2500 rpm/Idle* HC
2500 rpm/Idle* HC
2500 rpm/Idle* CO
2500 rpm/Idle* CO
2500 rpm/Idle* CO
2500 rpm/Idle* CO
0.7400
1.9223
3.9023
14.2820
0.4108
0.6062
1.0769
1.3808
0.9677
2.0226
3.1063
8.5543
0.6224
1.1894
1.3254
1.5286
CO
CO
CO
CO
9.2708
28.0310
90.0380
190.6600
4.9900
9.4669
12.1480
20.6200
10.4870
29.5500
53.5200
134.7500
9.8624
12.9690
17.4340
18.2810
0.8267
1.9846
3.9314
14.2820
10.3340
35.5180
104.5000
190.6600
0.4075
0.5923
1.0271 ,
1.3808 ''
0.9303
1.9431
2.9862
8.2523
4.8950 i 10.6220
9.8631 29.0530
11.9250 54.2820
20.6200 136.9700
J
0.5764
1.0349
1.1413
1.4141
9.2808
12.4890
13.1900
13.5960
Also used for Loaded/Idle and IM240 repair effects.
In the Tech4.1 model, the technologies and emission
categories are combined before an average I/M benefit for the
model year is calculated.
4.5.2 Evaporative and Running Loss Modeling, and the
Effectiveness of Purge/Pressure Testing
A large part of the additional emission reduction available
through the use of high-tech I/M tests is the result of the
evaporative and running loss emission reductions achieved by the
repair of vehicles which fail the new evaporative system pressure
and purge tests. The effectiveness of evaporative system pressure
and purge checks in reducing the rate of pressure and purge
problems was calculated assuming that programs with these checks
would detect 100% of all problems detected by the EPA checks run
in the Hammond I/M program. This assumes that the program will
use methods similar to the procedures used in Indiana. Although
all of the pressure and purge problems are assumed to be detected,
since some problems will re-occur with time, the average rate of
problems over the inspection cycle will not be zero.
Draft
-35-
2/26/92
-------
For purposes of determination of program effectiveness, the
combined evaporative system pressure and purge failure rates from
over 2,400 vehicles tested in Indiana were used. The resulting
effectiveness estimates were then used for application of pressure
checks, purge checks and combined pressure and purge checks in the
MOBILE4.1 model.
The average reduction in the rate of failure is calculated by
determining the rate of failure at the midpoint between two
vehicle ages. The effect of inspection can be visualized by
plotting the non-program rate over age with the calculated before
and after repairs failure rate estimates assuming inspection (see
figure in Appendix D) . At each age, vehicles due for inspection
are checked and necessary repairs made. Between inspections, the
rate of failures increases until the vehicles are due for
inspection again. The slope of this failure rate line between
inspections is assumed to be equal to the slope of the non-program
line for that vehicle age. This creates a rising and falling
pattern of rates resembling a saw blade. The average reduction in
rates is then the average value of the "saw teeth" compared with
the non-program case.
With an inspection program, at age zero, when the calendar
year equals the model year, no vehicles are yet one year old and
due for inspection; therefore, no reductions are made. Assuming
an annual inspection, at age one, 25% of the model year is one
year old or older. Therefore, the rate at one year is reduced by
25% to reflect repairs on the vehicles due for inspection. By the
second year, all vehicles are inspected each year and the after
repair rate is always zero. The failure rate after a check is
always zero, since the detection rate is 100%. Therefore, the
midpoint failure rate is half the number of failures that occur in
that year, once inspections begin. In the biennial case, vehicles
are inspected every other year and the rate of failures
accumulates in the years between inspections.
This method was used in a computer spreadsheet to calculate
the reduction in failures from evaporative system pressure and
purge checks used in the MOBILE4.1 model. The spreadsheet is
shown in Appendix D with and without formulas. The spreadsheet
originally contained errors which resulted in the benefits used in
the MOBILE4.1 model to be smaller than the estimates reported in
this document (which are based upon the corrected spreadsheet)
The version of MOBILE4.1 released to the public does not yet
reflect these changes, although they will be incorporated into the
next MOBILE release.
4.5.3 Benefits of IM24Q NOx Inspections
None of the existing I/M program models or the MOBILE4 1
model itself are designed to estimate the effect of NOX emission
inspection as part of an I/M program. Therefore, to estimate the
Draft -36-
2/26/92
-------
effect of an IM240-based NOX inspection, a simple model was
developed.
A sample of over 3,200 1983 and newer model year vehicles,
tested in Hammond, Indiana using the IM240 test procedure, was
analyzed. The sample was divided into three technology groups:
multi-point fuel injection vehicles, throttle-body fuel injection
vehicles and carbureted vehicles. Two NOX cutpoint cases were
examined for each technology, one with a 10% failure rate and one
with a 20% failure rate.
Using an emission correlation mapping between IM240 NOX
measurements and NOX measured on the FTP, an FTP NOX emission
level was estimated for each vehicle in the sample. A linear
least-square regression was run for estimated FTP NOX emissions
versus mileage for each technology for two model year groups: 1983
through 1985 model year vehicles and 1986 and newer model year
vehicles. The regressions were then run again excluding vehicles
which fail the IM240 NOX inspection first using the 10% failure
rate cutpoints and then the 20% failure rate cutpoints. The
exclusion of the higher NOX emitters was intended to represent
their deletion from the fleet through repairs.
Using the technology mix used in MOBILE4.1, the regressions
were weighted together to produce emission factor zero mile levels
and deterioration rates for each model year from 1983 through
1992. The difference in the emission levels between the cases
with and without NOX failures removed is assumed to be the benefit
from the IM240 NOX emission test with only NOx related repairs
performed. Results are shown in Table 4-10.
Since it is expected that most NOX emission testing will be
done along with testing for HC and CO emissions, the side effect
of HC and CO repairs on NOX emissions should also be accounted
for. This effect is ignored in the standard MOBILE4.1 model.
Typically NOX emissions will increase, on average, when HC an CO
emission repairs are performed. The extent of this NOX emission
disbenefit was determined by calculating average NOX emission
levels corresponding to the Normal, High, Very High and Super
HC/CO emitter categories used in the MOBILE4.1 Tech4 model.
Using the post-repair emission levels of the same vehicles
used to calculate the after repair emission levels for HC and CO
emissions, the NOX emission levels of these vehicles after repairs
were determined. These NO>: emission levels are not the result of
NOX related repairs, but a by-product of HC and CO emission
repairs. Using the standard HC/CO 2500 rpm/Idle test IDRs along
with the repair effects on NOX and the NOX emission rates by
emitter group in the Tech4 model, the effect of NOX disbenefits
was determined for each age of each model year. See Table 4-11.
Draft -37- 2/26/92
-------
The NOX disbenefits, as a percent change, are applied to the
emission levels estimated from the regression equations at each
age. The resulting NOX emission levels by age are regressed
versus mileage for each model year to give the final emission
factor equation for NOX. Comparing the emission factor results of
the baseline case with the cases with 10% or 20% NOX emission
testing failure rates was done to estimate the benefits, in tons,
of the IM240 NOX emission test. Results are shown in Table 4-12.
For example, at age 5 and mean mileage of 60,829 miles, the W20%
fail" IM240 NOX cutpoints will reduce 1992 model year NOX from
0.887 to 0.710 gpm, a reduction of 20%.
The final emission factors were used as alternate input to
the MOBILE4.1 model and, in combination with the CEM4.1 model,
used to calculate the tons of NOX emission benefit from use of
IM240 NOX cutpoints. These benefits were used in applying the
cost credit. It should be noted that since both the cases with
and without the IM240 NOX inspection cutpoints should include the
disbenefits of EC/CO repairs, the disbenefits do not effect the
calculation of incremental NOX reduction from IM240 cutpoints.
For simplicity and consistency, therefore, the disbenefits were
not applied to the I/M scenarios involving only HC/CO cutpoints.
Draft -38- 2/26/92
-------
Table 4-10
Lane IM24Q Based Emission Factor Levels with IM24Q NOX Outpoints
Age 0123456789 10 11 12
Miles 0 1.3118 2.6058 3.8298 4.9876 6.0829 7.119 8.0991 9.0262 9.9031 10.7326 11.5172 12.2594
Year Base
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.146
1.048
0.983
0.608
0.593
0.561
0.570
0.550
0.547
0.547
1.078
1.036
0.970
0.600
0.591
0.559
0.556
0.529
0.525
0.523
0.985
0.955
0.911
0.591
0.582
0.551
0.550
0.525
0.521
0.520
1.197
1.115
1.049
0.683
0.669
0.633
0.639
0.614
0.610
0.610
1.092
1.051
0.984
0.650
0.640
0.603
0.600
0.569
0.564
0.562
0.992
0.962
0.917
0.629
0.619
0.586
0.585
0.558
0.553
0.552
1.248
1.181
1.114
0.758
0.744
0.705
0.707
0.677
0.672
0.671
1.106
1.066
0.998
0.699
0.688
0.647
0.643
0.608
0.602
0.600
0.999
0.968
0.922
0.666
0.656
0.621
0.619
0.590
0.585
0.584
1.296
1.243
1.175
0.828
0.815
0.773
0.772
0.737
0.731
0.729
1.120
1.080
1.011
0.746
0.734
0.688
0.684
0.645
0.639
0.636
1.006
0.974
0.927
0.702
0.691
0.654
0.652
0.621
0.616
0.614
1.341
1.303
1.233
0.895
0.882
0.837
0.833
0.794
0.787
0.784
1.132
1.093
1.023
0.790
0.777
0.727
0.723
0.681
0.673
0.670
1.012
0.979
0.932
0.735
0.724
0.685
0.683
0.650
0.644
0.642
1.384
1.359
1.288
0.958
0.946
0.897
0.890
0.847
0.840
0.837
1.144
1.105
1.034
0.831
0.817
0.764
0.759
0.714
0.706
0.703
1.018
0.984
0.937
0.766
0.755
0.715
0.712
0.677
0.671
0.669
1.425
1.412
1.340
1.017
1.006
0.954
0.945
0.898
0.889
0.886
10% Fail
1.155
1.117
1.045
0.871
0.856
0.800
0.794
0.745
0.737
0.733
20% Fail
1.023
0.989
0.941
0.796
0.785
0.743
0.739
0.703
0.697
0.695
1.463
1.462
1.389
1.074
1.063
1.009
0.997
0.946
0.936
0.932
1.166
1.128
1.056
0.908
0.893
0.833
0.827
0.775
0.766
0.762
1.028
0.994
0.945
0.825
0.813
0.769
0.765
0.728
0.721
0.719
1.499
1.509
1.436
1.127
1.117
1.060
1.046
0.991
0.981
0.977
1.176
1.139
1.065
0.943
0.927
0.864
0.858
0.803
0.793
0.790
1.033
0.999
0.949
0.851
0.839
0.794
0.790
0.751
0.744
0.741
1.533
1.554
1.480
1.177
1.168
1.108
1.092
1.034
1.023
1.018
1.185
1.149
1.074
0.977
0.960
0.894
0.887
0.830
0.820
0.816
1.038
1.003
0.953
0.877
0.864
0.818
0.813
0.773
0.766
0.763
1.566
1.596
1.521
1.225
1.216
1.154
1.136
1.075
1.063
1.058
1.194
1.158
1.083
1.008
0.991
0.922
0.914
0.855
0.844
0.840
1.042
1.007
0.956
0.901
0.888
0.840
0.835
0.793
0.786
0.783
1.597
1.636
1.561
1.270
1.262
1.198
1.177
1.113
1.101
1.095
1.203
1.167
1.091
1.038
1.020
0.948
0.941
0.879
0.868
0.864
1.046
1.011
0.959
0.923
0.910
0.861
0.856
0.813
0.805
0.803
1.626
1.674
1.598
1.313
1.305
1.239
1.216
1.149
1.136
1.130
1.211
1.176
1.099
1.067
1.047
0.973
0.965
0.901
0.890
0.885
1.050
1.014
0.962
0.945
0.932
0.881
0.875
0.832
0.824
0.821
Draft
-39-
2/26/92
-------
Table 4-10
- continued -
13 14 15 16 17 18 19 20 21 22 23 24 25 Age Regression
12.9615 13.6257 14.254 14.8483 15.4104 15.9421 16.4451 16.9209 17.3712 17.7969 18.1997 18.5806 18.941 Miles ZML DET
Base
1.653
1.710
1.633
1.353
1.345
1.278
1.253
1.183
1.170
1.164
1.679
1.744
1.666
1.392
1.384
1.314
1.288
1.216
1.202
1.195
1.704
1.776
1.698
1.428
1.420
1.349
1.321
1.247
1.232
1.225
1.727
1.807
1.728
1.462
1.455
1.382
1.353
1.276
1.261
1.254
1.749
1.835
1.756
1.494
1.488
1.413
1.382
1.303
1.288
1.280
1.770
1.863
1.782
1.525
1.519
1.442
1.410
1.329
1.313
1.306
1.790
1.888
1.808
1.554
1.548
1.470
1.437
1.354
1.338
1.329
1.808
1.913
1.831
1.581
1.575
1.497
1.462
1.377
1.360
1.352
1.826
1.936
1.854
1.607
1.602
1.521
1.486
1.399
1.382
1.374
1.842
1.957
1.875
1.632
1.626
1.545
1.508
1.420
1.402
1.394
1.858
1.978
1.896
1.655
1.650
1.567
1.530
1.439
1.422
1.413
1.873
1.998
1.915
1.677
1.672
1.588
1.550
1.458
1.440
1.431
1.887
2.016
1.933
1.698
1.693
1.608
1.569
1.476
1.457
1.448
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.146
1.048
0.983
0.608
0.593
0.561
0.570
0.550
0.547
0.547
0.0391
0.0511
0.0501
0.0575
0.0581
0.0553
0.0527
0.0489
0.0481
0.0476
10% Fail
1.218
1.184
1.107
1.093
1.074
0.997
0.989
0.923
0.911
0.906
1.226
1.191
1.114
1.119
1.098
1.020
1.011
0.943
0.931
0.926
1.232
1.198
1.120
1.143
1.122
1.041
1.032
0.962
0.949
0.944
1.239
1.205
1.126
1.165
1.144
1.061
1.052
0.980
0.967
0.962
1.245
1.211
1.132
1.187
1.165
1.080
1.071
0.997
0.984
0.979
1.251
1.218
1.138
1.207
1.185
1.098
1.088
1.013
1.000
0.994
1.256
1.223
1.143
1.226
1.203
1.115
1.105
1.028
1.015
1.009
1.261
1.229
1.148
1.244
1.221
1.131
1.121
1.043
1.029
1.023
1.266
1.234
1.153
1.261
1.238
1.146
1.136
1.056
1.042
1.037
1.271
1.239
1.157
1.277
1.254
1.161
1.150
1.069
1.055
1.049
1.275
1.243
1.162
1.293
1.269
1.174
1.164
1.081
1.067
1.061
1.279
1.248
1.166
1.307
1.283
1.187
1.176
1.093
1.078
1.072
1.283
1.252
1.169
1.321
1.296
1.199
1.188
1.104
1.089
1.083
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.078
1.036
0.970
0.600
0.591
0.559
0.556
0.529
0.525
0.523
0.0108
0.0114
0.0105
0.0381
0.0372
0.0338
0.0334
0.0303
0.0298
0.0296
20% Fail
1.054
1.018
0.965
0.965
0.952
0.900
0.894
0.849
0.841
0.838
1.058
1.021
0.968
0.984
0.971
0.918
0.912
0.866
0.858
0.854
1.061
1.024
0.971
1.002
0.988
0.935
0.928
0.882
0.873
0.870
1.064
1.027
0.973
1.019
1.005
0.951
0.944
0.896
0.888
0.885
1.067
1.029
0.976
1.035
1.021
0.966
0.959
0.910
0.902
0.898
1.070
1.032
0.978
1.051
1.037
0.980
0.973
0.924
0.915
0.911
1.073
1.034
0.980
1.065
1.051
0.994
0.987
0.936
0.927
0.924
1.075
1.037
0.982
1.079
1.065
1.006
0.999
0.948
0.939
0.935
1.077
1.039
0.984
1.092
1.077
1.019
1.011
0.959
0.950
0.947
1.080
1.041
0.986
1.104
1.090
1.030
1.022
0.970
0.961
0.957
1.082
1.043
0.987
1.116
1.101
1.041
1.033
0.980
0.971
0.967
1.084
1.045
0.989
1.127
1.112
1.051
1.043
0.990
0.980
0.976
1.086
1.046
0.990
1.137
1.122
1.061
1.053
0.999
0.989
0.985
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
0.985
0.955
0.911
0.591
0.582
0.551
0.550
0.525
0.521
0.520
0.0053
0.0048
0.0042
0.0288
0.0285
0.0269
0.0265
0.0250
0.0247
0.0246
Draft
-40-
2/26/92
-------
Table 4-11
Side Effects of I/M on NOx Emissions
(Disbenefit of HC/CO repairs)
Age 012 3 4 56789 10 11 12
Miles 0 1.3118 2.6058 3.8298 4.9876 6.0829 7.119 8.0991 9.0262 9.9031 10.7326 11.5172 12.2594
Year
Base
1983 0.61
1984 0.63
1985 0.63
1986 0.51
1987 0.50
1988 0.47
1989 0.47
1990 0.46
1991 0.45
1992 0.45
1983 0.61
1984 0.63
1985 0.63
1986 0.51
1987 0.50
1988 0.47
1989 0.47
1990 0.46
1991 0.45
1992 0.45
0.67
0.69
0.69
0.56
0.55
0.52
0.52
0.50
0.50
0.50
0.69
0.71
0.70
0.58
0.56
0.53
0.54
0.52
0.52
0.52
0.74
0.75
0.74
0.62
0.60
0.56
0.57
0.55
0.55
0.55
0.76
0.77
0.76
0.64
0.61
0.58
0.59
0.57
0.57
0.57
0.80
0.81
0.79
0.67
0.64
0.60
0.62
0.59
0.59
0.59
0.82
0.83
0.81
0.69
0.66
0.63
0.64
0.62
0.62
0.62
0.86
0.87
0.84
0.72
0.69
0.64
0.66
0.64
0.63
0.63
0.88
0.88
0.86
0.74
0.71
0.67
0.69
0.66
0.66
0.66
0.91
0.92
0.89
0.76
0.73
0.68
0.70
0.68
0.67
0.67
0.93
0.93
0.90
0.78
0.75
0.71
0.73
0.70
0.70
0.70
0.98
0.99
0.96
0.81
0.77
0.72
0.74
0.71
0.71
0.71
Idle
0.99
1.00
0.97
0.83
0.80
0.75
0.77
0.74
0.74
0.74
1.04
1.05
1.02
0.85
0.81
0.75
0.78
0.75
0.75
0.75
1.05
1.06
1.04
0.87
0.84
0.78
0.81
0.78
0.78
0.78
1.10
1.11
1.09
0.89
0.84
0.78
0.81
0.78
0.78
0.78
1.10
1.12
1.09
0.91
0.87
0.82
0.85
0.82
0.82
0.82
1.15
1.17
1.14
0.92
0.88
0.81
0.85
0.82
0.81
0.82
1.15
1.17
1.15
0.95
0.91
0.85
0.88
0.85
0.85
0.85
1.20
1.22
1.20
0.96
0.91
0.84
0.88
0.85
0.85
0.85
1.20
1.22
1.20
0.98
0.94
0.88
0.91
0.88
0.88
0.88
1.25
1.27
1.25
0.99
0.94
0.87
0.91
0.88
0.87
0.88
1.25
1.27
1.25
1.02
0.97
0.90
0.94
0.91
0.91
0.91
1.30
1.32
1.30
1.02
0.97
0.89
0.94
0.90
0.90
0.90
1.29
1.31
1.30
1.05
1.00
0.93
0.97
0.94
0.94
0.94
Two Speed
1983 0.61
1984 0.63
1985 0.63
1986 0.51
1987 0.50
1988 0.47
1989 0.47
1990 0.46
1991 0.45
1992 0.45
0.69
0.71
0.71
0.58
0.57
0.54
0.55
0.53
0.53
0.53
0.76
0.78
0.76
0.64
0.62
0.59
0.60
0.58
0.58
0.58
0.83
0.83
0.82
0.70
0.67
0.63
0.65
0.63
0.63
0.63
0.88
0.88
0.86
0.75
0.72
0.68
0.70
0.67
0.67
0.67
0.93
0.93
0.91
0.79
0.76
0.72
0.74
0.72
0.71
0.72
1.00
1.00
0.97
0.84
0.81
0.76
0.78
0.76
0.76
0.76
1.05
1.06
1.04
0.88
0.85
0.80
0.82
0.80
0.80
0.80
1.10
1.12
1.10
0.92
0.89
0.83
0.86
0.84
0.83
0.84
1.15
1.17
1.15
0.96
0.92
0.87
0.90
0.87
0.87
0.87
1.20
1.22
1.20
1.00
0.96
0.90
0.93
0.90
0.90
0.90
1.25
1.27
1.25
1.03
0.99
0.93
0.96
0.93
0.93
0.93
1.29
1.31
1.30
1.06
1.02
0.95
0.99
0.96
0.96
0.96
Draft
-41-
2/26/92
-------
Table 4-11
- continued -
13 14 15 16 17 18 19 20 21 22 23 24 25 Model Regression
12.9615 13.6257 14.254 14.8483 15.4104 15.9421 16.4451 16.9209 17.3712 17.7969 18.1997 18.5806 18.941 Year ZML DBF
Base
1.34
1.36
1.35
1.05
1.00
0.92
0.96
0.93
0.93
0.93
1.33
1.36
1.34
1.07
1.02
0.95
0.99
0.96
0.96
0.96
1.38
1.40
1.39
1.08
1.02
0.94
0.99
0.95
0.95
0.95
1.37
1.39
1.38
1.10
1.05
0.97
1.02
0.99
0.98
0.99
1.41
1.44
1.41
1.10
1.04
0.96
1.01
0.98
0.97
0.98
1.40
1.43
1.41
1.13
1.07
0.99
1.04
1.01
1.01
1.01
1.44
1.46
1.44
1.13
1.07
0.98
1.03
1.00
1.00
1.00
1.42
1.45
1.42
1.15
1.09
1.01
1.06
1.03
1.03
1.03
1.47
1.49
1.45
1.15
1.09
1.00
1.05
1.02
1.02
1.02
1.44
1.47
1.44
1.17
1.11
1.03
1.08
1.05
1.04
1.05
1.49
1.51
1.47
1.17
1.11
1.02
1.07
1.04
1.03
1.04
1.46
1.49
1.46
1.19
1.13
1.05
1.10
1.06
1.06
1.07
1.51
1.53
1.48
1.19
1.13
1.04
1.09
1.05
1.05
1.06
1.48
1.50
1.47
1.21
1.15
1.07
1.12
1.08
1.08
1.08
1.53
1.54
1.50
1.21
1.15
1.05
1.11
1.07
1.07
1.07
Idle
1.50
1.52
1.48
1.23
1.17
1.08
1.13
1.10
1.09
1.10
1.55
1.56
1.51
1.23
1.16
1.07
1.13
1.09
1.09
1.09
1.52
1.53
1.49
1.25
1.18
1.10
1.15
1.11
1.11
1.11
1.56
1.58
1.53
1.25
1.18
1.08
1.14
1.10
1.10
1.11
1.53
1.55
1.50
1.26
1.20
1.11
1.16
1.12
1.12
1.13
1.58
1.59
1.54
1.26
1.19
1.10
1.16
1.12
1.12
1.12
1.54
1.56
1.51
1.28
1.21
1.12
1.17
1.14
1.14
1.14
1.60
1.60
1.55
1.28
1.21
1.11
1.17
1.13
1.13
1.13
1.56
1.57
1.52
1.29
1.22
1.13
1.19
1.15
1.15
1.15
1.61
1.62
1.56
1.29
1.22
1.12
1.18
1.14
1.14
1.15
1.57
1.58
1.53
1.30
1.24
1.14
1.20
1.16
1.16
1.16
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
0.601 0.0550
0.617 0.0549
0.614 0.0527
0.511 0.0415
0.497 0.0383
0.470 0.0345
0.475 0.0375
0.455 0.0364
0.453 0.0365
0.452 0.0367
0.628 0.0519
0.640 0.0524
0.637 0.0505
0.530 0.0414
0.516 0.0385
0.491 0.0350
0.497 0.0377
0.479 0.0367
0.477 0.0367
0.476 0.0369
Two Speed
1.33
1.35
1.34
1.09
1.04
0.98
1.01
0.98
0.98
0.98
1.36
1.39
1.38
1.12
1.07
1.00
1.04
1.01
1.00
1.01
1.39
1.42
1.40
1.14
1.09
1.02
1.06
1.03
1.03
1.03
1.42
1.45
1.42
1.17
1.11
1.04
1.08
1.05
1.05
1.05
1.44
1.46
1.44
1.19
1.13
1.06
1.10
1.07
1.07
1.07
1.46
1.48
1.45
1.21
1.15
1.08
1.12
1.09
1.08
1.09
1.47
1.50
1.46
1.23
1.17
1.09
1.14
1.10
1.10
1.10
1.49
1.51
1.48
1.24
1.19
1.11
1.15
1.12
1.11
1.12
1.51
1.53
1.49
1.26
1.20
1.12
1.17
1.13
1.13
1.13
1.52
1.54
1.50
1.28
1.22
1.13
1.18
1.14
1.14
1.15
1.53
1.55
1.51
1.29
1.23
1.14
1.19
1.16
1.15
1.16
1.55
1.56
1.52
1.30
1.24
1.16
1.21
1.17
1.17
1.17
1.56
1.57
1.53
1.32
1.25
1.17
1.22
1.18
1.18
1.18
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
0.637 0.0509
0.645 0.0516
0.642 0.0499
0.535 0.0421
0.521 0.0395
0.497 0.0362
0.503 0.0385
0.487 0.0374
0.485 0.0374
0.485 0.0376
Draft
-42-
2/26/92
-------
Table 4-12
Lane IM24Q Based. Emission Factors with. IM24Q Outpoints
with Disbenefits of HC/CQ Repairs Included
Age 0 1 2 3 4 5 6 7 8 9 10 11 12
Miles 0 1.3118 2.6058 3.8298 4.9876 6.0829 7.119 8.0991 9.0262 9.9031 10.7326 11.5172 12.2594
Year
Base
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.146
1.048
0.983
0.608
0.593
0.561
0.570
0.550
0.547
0.547
1.078
1.036
0.970
0.600
0.591
0.559
0.556
0.529
0.525
0.523
0.985
0.955
0.911
0.591
0.582
0.551
0.550
0.525
0.521
0.520
1.235
1.145
1.081
0.707
0.695
0.662
0.667
0.646
0.642
0.641
1.126
1.080
1.014
0.673
0.664
0.630
0.627
0.599
0.593
0.591
1.023
0.988
0.945
0.651
0.643
0.613
0.611
0.587
0.582
0.580
1.288
1.214
1.147
0.787
0.775
0.740
0.742
0.714
0.710
0.709
1.142
1.095
1.027
0.726
0.717
0.679
0.675
0.642
0.636
0.634
1.032
0.995
0.950
0.692
0.684
0.652
0.650
0.623
0.618
0.617
1.336
1.271
1.208
0.861
0.851
0.814
0.812
0.781
0.775
0.774
1.154
1.104
1.039
0.776
0.766
0.725
0.720
0.684
0.676
0.675
1.037
0.995
0.953
0.730
0.721
0.689
0.686
0.658
0.652
0.651
1.374
1.325
1.260
0.930
0.922
0.884
0.877
0.841
0.835
0.833
1.160
1.112
1.045
0.821
0.812
0.768
0.761
0.721
0.715
0.712
1.037
0.996
0.952
0.764
0.757
0.724
0.719
0.689
0.684
0.682
1.413
1.378
1.313
0.995
0.989
0.949
0.939
0.899
0.891
0.887
1.168
1.121
1.054
0.864
0.854
0.808
0.801
0.757
0.749
0.746
1.039
0.998
0.955
0.797
0.790
0.756
0.750
0.718
0.712
0.710
1.448
1.427
1.361
1.059
1.053
1.013
0.999
0.956
0.947
0.943
10% Fail
1.174
1.130
1.062
0.907
0.896
0.849
0.839
0.793
0.784
0.781
20% Fail
1.040
1.001
0.956
0.829
0.822
0.788
0.781
0.749
0.742
0.739
1.480
1.473
1.406
1.119
1.114
1.072
1.054
1.006
0.998
0.993
1.179
1.137
1.068
0.947
0.936
0.885
0.874
0.825
0.816
0.812
1.040
1.002
0.956
0.860
0.852
0.817
0.809
0.774
0.768
0.766
1.509
1.517
1.448
1.175
1.172
1.128
1.106
1.056
1.045
1.040
1.183
1.145
1.074
0.984
0.973
0.919
0.907
0.855
0.845
0.841
1.040
1.004
0.957
0.888
0.881
0.845
0.835
0.800
0.792
0.790
1.536
1.558
1.486
1.227
1.226
1.181
1.155
1.101
1.091
1.084
1.187
1.152
1.079
1.018
1.008
0.952
0.938
0.884
0.874
0.869
1.040
1.005
0.957
0.914
0.908
0.871
0.860
0.823
0.816
0.813
1.565
1.595
1.526
1.275
1.278
1.230
1.200
1.143
1.131
1.125
1.193
1.157
1.087
1.050
1.041
0.982
0.967
0.909
0.898
0.894
1.041
1.006
0.959
0.937
0.933
0.895
0.883
0.844
0.836
0.833
1.590
1.631
1.561
1.323
1.325
1.275
1.243
1.183
1.170
1.164
1.198
1.164
1.091
1.081
1.071
1.009
0.993
0.934
0.922
0.918
1.042
1.007
0.959
0.961
0.956
0.916
0.904
0.864
0.856
0.853
1.614
1.666
1.594
1.365
1.368
1.319
1.283
1.219
1.206
1.199
1.202
1.169
1.097
1.108
1.098
1.036
1.018
0.956
0.944
0.939
1.043
1.009
0.960
0.982
0.977
0.938
0.923
0.882
0.874
0.871
Draft
-43-
2/26/92
-------
Table 4-12
- continued -
13 14 15 16 17 18 19 20 21 22 23 24 25 Model Regression
12.9615 13.6257 14.254 14.8483 154104 15.9421 164451 16.9209 17.3712 17.7969 18.1997 18.5806 18.941 Year ZML DBF
Base
1.637 1.660
1.699 1.728
1.626 1.655
1.405 1.443
1.409 1.448
1.357 1.395
1.320 1.354
1.252 1.285
1.240 1.269
1.231 1.262
1.680
1.757
1.685
1.478
1.484
1.429
1.387
1.314
1.299
1.290
1.698
1.783
1.711
1.510
1.519
1.462
1.417
1.342
1.327
1.318
1.716
1.807
1.735
1.541
1.549
1.493
1.445
1.367
1.351
1.342
1.733
1.832
1.761
1.571
1.579
1.522
1.471
1.392
1.376
1.366
1.748
1.854
1.783
1.597
1.607
1.547
1.496
1.414
1.396
1.387
1.763
1.875
1.803
1.623
1.632
1.570
1.519
1.435
1.416
1.406
1.779
1.895
1.823
1.645
1.655
1.594
1.539
1.454
1.435
1.425
1.791
1.913
1.842
1.667
1.679
1.616
1.560
1.473
1.453
1.444
1.803
1.931
1.860
1.689
1.699
1.634
1.578
1.488
1.470
1.460
1.816
1.948
1.878
1.707
1.719
1.653
1.596
1.506
1.486
1.475
1.825
1.962
1.893
1.726
1.738
1.671
1.612
1.520
1.501
1.490
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.196
1.087
1.022
0.638
0.623
0.594
0.605
0.589
0.587
0.587
0.0338
0.0468
0.0463
0.0584
0.0599
0.0580
0.0543
0.0503
0.0494
0.0488
10% Fail
1.207 1.211
1.176 1.180
1.102 1.106
1.135 1.160
1.124 1.149
1.059 1.082
1.041 1.062
0.976 0.996
0.965 0.982
0.959 0.977
1.215
1.185
1.111
1.183
1.172
1.102
1.083
1.014
1.001
0.995
1.218
1.189
1.115
1.203
1.194
1.122
1.102
1.031
1.017
1.011
1.221
1.193
1.119
1.224
1.213
1.141
1.119
1.046
1.032
1.026
1.225
1.197
1.124
1.243
1.232
1.158
1.135
1.061
1.047
1.040
1.227
1.201
1.128
1.260
1.249
1.173
1.151
1.074
1.059
1.053
1.230
1.205
1.131
1.277
1.265
1.187
1.164
1.086
1.071
1.064
1.234
1.208
1.134
1.291
1.279
1.201
1.176
1.098
1.082
1.076
1.235
1.210
1.137
1.305
1.294
1.214
1.189
1.109
1.093
1.087
1.237
1.214
1.140
1.319
1.307
1.225
1.201
1.118
1.103
1.096
1.240
1.216
1.143
1.331
1.319
1.235
1.211
1.129
1.113
1.106
1.241
1.218
1.145
1.343
1.331
1.246
1.221
1.137
1.121
1.114
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.118
1.067
1.001
0.628
0.619
0.590
0.588
0.564
0.560
0.558
0.0068
0.0082
0.0077
0.0385
0.0384
0.0355
0.0343
0.0311
0.0305
0.0302
20% Fail
1.044 1.045
1.011 1.011
0.961 0.962
1.002 1.021
0.997 1.015
0.956 0.974
0.942 0.958
0.899 0.915
0.891 0.905
0.887 0.902
1.046
1.012
0.963
1.038
1.033
0.990
0.974
0.929
0.921
0.916
1.046
1.013
0.964
1.053
1.050
1.006
0.989
0.943
0.934
0.930
1.047
1.014
0.964
1.068
1.064
1.021
1.003
0.955
0.946
0.942
1.048
1.015
0.966
1.082
1.078
1.034
1.015
0.967
0.958
0.954
1.048
1.015
0.967
1.095
1.091
1.045
1.027
0.978
0.968
0.964
1.048
1.017
0.967
1.107
1.103
1.056
1.038
0.988
0.978
0.973
1.050
1.017
0.967
1.118
1.114
1.067
1.047
0.997
0.987
0.982
1.049
1.017
0.968
1.128
1.125
1.078
1.057
1.006
0.996
0.992
1.050
1.018
0.969
1.139
1.134
1.085
1.066
1.013
1.004
0.999
1.051
1.019
0.970
1.147
1.143
1.094
1.074
1.022
1.011
1.006
1.050
1.019
0.970
1.157
1.153
1.102
1.082
1.028
1.018
1.013
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1.021
0.982
0.939
0.617
0.608
0.581
0.581
0.559
0.555
0.554
0.0018
0.0021
0.0017
0.0291
0.0294
0.0283
0.0272
0.0256
0.0252
0.0250
Draft
-44-
2/26/92
-------
5.0 REGULATORY IMPACT ANALYSIS - ESTIMATING COST AND COST
EFFECTIVENESS
5.1 Cost of Conventional I/M Testing
EPA has collected and analyzed cost data from all operating
I/M programs that could provide the information. EPA has analyzed
per vehicle costs in I/M programs based upon four basic pieces of
information: The I/M program agency budget, number of initial
tests, the fee for each test, and the portion returned to the
state or local government. This discussion will deal with three
aspects of I/M cost: Inspection costs, oversight costs, and repair
costs. In addition, it should be noted that where programs are
referred to as either "centralized" or "decentralized," this
indicates only whether or not the functions of testing and
repairing are separated and performed by different entities (i.e.,
in centralized networks - which can include multiple independent
supplier test-only networks - the test and repair functions are
separate, whereas in decentralized networks, these functions are
typically performed at a single facility.
5.1.1 Inspection and Administration Costs
Inspection fees are set in one of three ways: By a bid
process for a contract to supply inspection services, by
legislation or regulation establishing a maximum fee, or by market
forces. Ideally the fee is scaled to cover the cost of providing
the inspection, cover the fee to the state for oversight and
management, and to provide a reasonable profit to the operator
(except in government-run programs).
This ideal is not always met in actual I/M programs. In some
programs the inspection fee does not include a share for the
state's oversight costs, so these must be derived from the general
fund, with the result that oversight efforts are often
significantly underfunded. In many decentralized programs the
maximum fee is set below the actual cost (with profit) for the
test, so providers must make up for that cost by providing other
goods and services.
The economies of scale and efficiency of operation in high
volume test-only inspection networks enable motorists in these
programs to enjoy lower average inspection fees than in low volume
decentralized programs. Based upon 1989 I/M audits (which
collected information from all I/M programs), and taking into
account both inspection and oversight costs, decentralized
programs using computerized analyzers have the highest costs,
averaging about $17.70 per vehicle; centralized contractor-run
programs average $8.42 per vehicle (recently gathered 1990 data
show slightly different numbers, although these have not greatly
Draft -45- 2/26/92
-------
affected the overall averages). Table 5-1 shows the estimated
cost of the I/M program on a per vehicle basis, including
inspection and oversight costs.
Table 5-1
I/M Program Inspection Fees
Decentralized Programs Centralized Programs
Program Cost Per Vehicle Program Cost Per Vehicle
Anchorage $32.00 Arizona $6.00
Fairbanks $29.00 Connecticut $10.00
California $48.39 Florida $10.00
Colorado $11.20 Illinois $8.07
Georgia $10.68 Louisville $6.00
Massachusetts $17.18 Maryland $8.53
Michigan $10.87 Minnesota $8.00
Missouri $9.00 Nashville $6.00
North Carolina $15.40 Washington $9.00
New Mexico $16.00 Wisconsin $8.73
Nevada $21.26
New York $19.92
Pennsylvania $9.01
Dallas $17.25
El Paso $17.25
Davis County $9.00
Utah County $9.71
Salt Lake City $11.49
Virginia $13.50
In a centralized, contractor-run program, the contractor
bears the cost of acquiring land, constructing and equipping
inspection facilities, hiring and training staff, collecting and
processing data, and doing the routine testing work. The state's
role in this case is to make sure the contractor meets its
obligations and to study the outcomes of the program to make sure
it is meeting the goal.
In a decentralized program, individual firms and small
businesses are licensed to perform the inspection. In this case,
the state takes primary responsibility for many of the day to day
functions, such as data collection and processing, public
information, enforcement, inspector training, and so on. The fact
that inspections are performed by many business entities instead
of one, and that there are more inspection sites means that state
oversight and program evaluation activities need to be more
intensive in this type of program and, as a result are
Significantly more costly. '
Costs of quality assurance measures vary among programs
depending upon the comprehensiveness of the quality assurance
program and are not well documented in most state programs EPA
Draft -46- 2/26/92
-------
has recommended that the portion of the inspection fee returned to
the state be dedicated to program oversight, and that it be scaled
to cover the cost of carrying out an effective quality assurance
program. EPA audits have found that decentralized programs
typically have inadequate quality assurance programs, often due to
lack of funds, but also due to insufficient legal authority or
willingness.
5.2 Estimated Cost of High-Tech I/M Testing
5.2.1 General Methodology
EPA's estimates of the costs of high-tech test procedures are
driven by a number of assumptions. The first is that the new
tests would be performed in currently existing inspection networks
(i.e., centralized- and decentralized-based programs). Therefore,
the starting point for the cost estimates are the current
inspection costs in I/M programs as they are reported to EPA.
The second assumption is that adding the new tests will
increase inspection costs in programs that are now efficiently
designed and operated. In programs that are not now well
designed, current costs are likely to be higher than necessary and
the cost increase less if efficiency improvements are made
simultaneously. In order to perform the high-tech tests new
equipment will have to be acquired and additional inspector time
will be required for some test procedures. The amount of the cost
increase will be determined to a large degree by the costs of
acquiring new equipment and the impact of the longer test on
throughput in a high volume operation. Average test volume in
decentralized programs is low enough to easily absorb the
additional test time involved (although at a cost in labor time).
Equipment costs are analyzed in terms of the additional cost to
equip each inspection site (i.e., each inspection lane in
centralized inspection networks, and each licensed inspection
station in decentralized networks).
By focusing on the inspection lane or station as the basic
unit of analysis, the resulting cost estimates are equally
applicable in large programs, with many subject vehicles and
inspection sites, or small programs, with few subject vehicles and
inspection sites. Previous EPA analyses of costs in I/M programs
have found that the major determinants of inspection costs are
test volume and the level of sophistication of the inspection
equipment. Costs of operating programs were not found to be
measurably affected by the size of the program (for further
information the reader may refer to EPA's report entitled, "I/M
Network Type: Effects on Emission Reductions, Cost, and
Convenience"). Figures on inspection volumes at inspection
stations and lanes are available from I/M program operating data.
This information enables the equipment cost per vehicle and the
Draft -47- 2/26/92
-------
additional staff cost per vehicle to be calculated for each test
procedure.
The equipment cost figures presented in this paper are based
on the costs of the equipment EPA believes is best suited for
high-tech testing. They are current prices quoted by
manufacturers, and do not reflect what the per unit prices might
be if this equipment were purchased in volume. Staff costs are
based on prevailing wage rates for inspectors in both types of
programs as reported in conversations with state I/M program
personnel. For centralized networks, site costs and management
overhead are back calculated from current inspection costs. For
decentralized networks, it is assumed that longer test times could
be absorbed with no increase in sites. The current average volume
in decentralized stations is 1,025 vehicles per year (between 3
and 4 vehicles per day, depending upon the number of days per year
the station is open). Consequently, increasing the length of the
test, to the degree that the new procedures would, is not expected
to impact the number of inspections that can be performed.
5.2.2 Equipment Needs and Costs
A pressure metering system, composed of a cylinder of
nitrogen gas with a regulator, and hoses connecting the tank to a
pressure meter, and to the vehicle's evaporative system is needed
to perform evaporative system pressure testing. Hardware to
interface the metering system with a computerized analyzer is also
needed and is included in the cost estimate. Purge testing can be
performed by adding a flow sensor with a computer interface, a
dynamometer, and a Video Driver's Aid. With the further addition
of a Constant Volume Sampler (CVS) and a flame ionization detector
(FID) for HC analysis, two non-dispersive infrared (NDIR)
analyzers for CO and carbon monoxide (CO2) , and a chemiluminescent
(CI) analyzer for NOX, transient testing can be performed.
The analyzers used for the transient test are laboratory
grade equipment. They are designed to higher accuracy and
repeatability specifications than the NDIR analyzers used to
perform the current I/M tests. Table 5-2 shows the estimated cost
of equipment for conducting high-tech tests. This quality of
technology is essential for accurate instantaneous measurements of
low concentration mass emission levels.
Draft -48- 2/26/92
-------
Test
Pressure
Purge
Transient
Table 5-2
Equipment Costs for New Tests
Equipment price
Metering System $600
Flow Sensor $500
Dynamometer $45,000
Video Drivers Aid $3,000
CVS & Analyzers $95f OOP
TOTAL $144,100
The figures in Table 5-2 do not include the costs of
expendable materials. Nitrogen gas is used up in performing the
pressure test. Additionally, the FID burns hydrogen fuel.
Calibration gases are needed for each of the analyzers used in the
transient test. Because the analyzers used in the transient test
are designed to more stringent specifications than the analyzers
currently used in the field, bi-blends, gaseous mixtures composed
of one interest gas in a diluent (usually nitrogen) are used to
calibrate them. Multi-blend gases, such as are typically used to
calibrate current I/M equipment, are not suitable. Current
estimates for expendables are shown in Table 5-3. The replacement
intervals are estimated based on the usage rates observed in the
EPA Indiana pilot program and typical inspection volumes as
presented later in this section. Calculations of per vehicle
equipment costs presented throughout this report include per
vehicle costs of these expendables as well.
Table 5-3
ExDendables for New Tests
Test
Pressure
Material
N2 Gas
Replacement Interval
Cost Centralized Decentralized
$30
250 tests
250 tests
Transient
H2 Fuel $60
HC Cal Gas $60
CO Cal Gas $60
C02 Cal Gas $60
2 months
2 months
2 months
2 months
I year
1 year
1 year
1 year
Staff costs have been found to vary between centralized and
decentralized programs, as does the effect on the number of sites
Draft
-49-
2/26/92
-------
in the network infrastructure. Therefore, the following sections
are devoted to separate cost analyses for each network type.
5.2.3 Cost, to Upgrade Centralized Networks
5.2.3.1 Basic Assumptions
The starting point in this analysis is the current average
per vehicle inspection cost in centralized programs. A figure of
$8.50 was used based upon data from operating programs. This
figure includes the cost of one or more retests and network
oversight costs. The key variables to consider in estimating the
costs in centralized networks are throughput, equipment, and staff
costs. Data on these variables were obtained by contacting
program managers in a number of these programs, and by surveying
program contracts and Requests for Proposal.
Throughput refers to the number of vehicles per hour that can
be tested in a lane. The higher the throughput rate, the greater
the number of vehicles over which costs are spread, and the lower
the per vehicle cost. EPA contacted program managers and
consulted the contracts in a number of centralized programs to
determine peak period throughput rates in the different systems.
Rates were as reported in Table 5-4.
Table 5-4
Peak Period Throughput Rates in Independent I/M Programs
Program Vehicles Tested per Hour
Arizona 20
Connecticut 25-30
Illinois 25
Maryland 25-35
Wisconsin 25-30
On the basis of this information, 25 vehicles per hour was
assumed to represent the typical peak period throughput rate or
design capacity in centralized I/M programs. During off-peak
hours and days, throughput is lower since there is not a constant
stream of arriving vehicles. Conversations with individuals in
the centralized inspection service industry indicate that
inspectors start at minimum wage or slightly higher, that by the
end of the first year they earn $5.50 to $6 per hour, and that
they generally stay with the job for one to three years. Thus, $6
per hour was used to estimate the average inspector's hourly wage
Draft -50- 2/26/92
-------
Estimates of the costs of adding pressure testing, purge
testing, and transient tailpipe testing were derived by taking the
current costs for the new equipment to perform the new tests,
dividing it by the number of inspections expected to be performed
in the lane over a five year period and adding it to the current
$8.50 per vehicle cost, with a further adjustment for the impact
of test time on throughput, and thus on the number of sites and
site costs. The same is done to estimate additional personnel
costs associated with adding the new tests. When independent
programs were surveyed to determine the length of a typical
contract, it was discovered that Illinois, Florida, and Minnesota
all have five year contracts, Arizona has a seven year contract,
and the program in the State of Washington is operating under a
three year contract, resulting in an average contract length of
five years among the five programs surveyed. Five years was
therefore chosen as the typical contract length.
The number of inspections expected to be performed over the
five year contract period was derived by calculating the total
number of hours of lane operation, estimating the average number
of vehicles per lane and multiplying the two. A lane is assumed
to operate for 60 hours a week (lane operation times were found to
vary from 54 to 64 hours per week), 52 weeks a year for five years
for a total of 15,600 hours. Lanes are assumed to have a peak
throughput capacity of 25 vehicles per hour. Modern centralized
inspection networks are designed so that they can accommodate peak
demand periods with all lanes operating at this throughput rate.
Networks are usually designed so that average throughput is 50-65%
of peak capacity or 13-15 vehicles per hour. When operating for
15,600 hours over the life of a contract, a centralized inspection
lane is estimated to perform a total of 195,000 inspections, or
about 39,000 per year.
5.2.3.2 The Effect of Changing Throughput
The addition of evaporative system pressure testing to a
centralized program would result in a slight decrease in the
throughput capacity. The addition of purge and transient testing,
along with pressure testing, would result in a further decrease.
Assuming the same test frequency (i.e., annual or biennial)
the reduced throughput rate means that the number of lanes needed
to test a given number of vehicles would increase accordingly, as
would the size of the network infrastructure needed to support the
test program. The result is an increase in the cost per vehicle.
Actual consumer cost depends on the test frequency; EPA would
encourage states to adopt biennial programs to reduce the costs
and imposition of the program. Less frequent testing only
slightly reduces the emission reduction benefits while cutting
test costs almost in half.
One way to estimate the cost would be to simulate an actual
network of stations and lanes in a given city. One could attempt
Draft -51- 2/26/92
-------
to assess land costs, building costs, staff and equipment costs,
costs for all necessary support systems, and other cost factors.
However, this approach would be very time consuming and would rely
on information which is proprietary to the private contractors
that operate the programs and is, therefore, unavailable.
Instead, the cost of the increased number of lanes and stations is
derived by analyzing current costs and subtracting out equipment,
direct personnel, and state agency oversight costs. The remainder
is adjusted by the change in throughput'in the new system. Then,
new estimates of equipment, personnel and oversight costs are
added back in to obtain the estimated total cost.
As discussed previously, the typical high volume station can
test 25 vehicles per hour, performing (in most cases) a test
consisting of 30 seconds of high speed preconditioning or testing,
followed by 30 seconds of idle testing. In addition, a short time
is spent getting the vehicle into position and preparing it for
testing. This leads to a two to three minute test time on
average, depending upon what short test is performed. EPA
recently issued alternative test procedures for steady-state tests
that reduce various problems associated with those tests,
especially false failures, but at a cost of longer average per
test time.
Current costs were estimated by contacting operating program
personnel, equipment vendors and contractors. The most
sophisticated equipment installation (i.e., the equipment for
loaded steady-state testing) was used to estimate current
equipment costs.
The cost to acquire and install a single curve dynamometer
and an analyzer in existing networks is about $40,000 or 210 per
vehicle using the basic test volume assumptions. As indicated
previously, a staff person is assumed to earn $6.00 per hour.
When this figure is multiplied by 15,600 total contract hours and
divided by 195,000 vehicles, direct staff costs are estimated at
48* per vehicle. Existing centralized networks typically have two
staff per lane. Thus, total staff costs work out to 96* per
vehicle. In this analysis a figure of $1.25 is used to estimate
the amount of the state retainer. This reflects EPA's best
estimate of the per vehicle expense for a good state quality
assurance program in a centralized network. Equipment, staff, and
state costs add up to $2.42 per vehicle. Subtracting this amount
from the current average of $8.50 leaves $6.08 in infrastructure
costs and other overhead expenses including employee benefits and
employer taxes as shown in Table 5-5. This amount is then
factored by the change in the throughput rate and the equipment,
oversight, and staff costs for the new tests are then added.
Draft -52- 2/26/92
-------
Table 5-5
Current Program Costs
Per Vehicle Total Cost Less
Increments Cost Increments
Current $8.50
Equipment $0.21 $8.29
Staff $0.96 $7.33
State Retainer $1.25 $6.08
5.2.3.3 Evaporative System Pressure Testing
Most centralized programs use a two position test queue;
emission test are done in one position while emission control
devices are checked in the other, along with other functions such
as fee collection. In this type of system the throughput rate is
determined by the length of time required to perform the longest
step in the sequence, not by length of the entire test sequence.
With computer automation and refined pass/fail logic, the pressure
test could be performed by an extra technician while the emission
control device check and other functions are performed. In such a
system the test would take between two and three minutes to
perform. As a result, adding the pressure test is expected to
reduce peak throughput to about 20 vehicles per hour. To
accommodate peak demand periods and maintain short wait times, a
design throughput rate of half of capacity is assumed, for a
typical throughput rate of 10 vehicles per hour. Assuming the
same number of hours of lane operation, the total number of tests
per lane is estimated to be 156,000 over the five year contract
period.
Calculations of the additional costs of adding pressure
testing are detailed in Table 5-6. These are based upon the costs
of equipment and expendables as shown in Tables 5-2 and 5-3 and
staff costs and throughput assumptions detailed in the previous
paragraphs. Staff costs are derived by assuming two staff per
lane at a pay rate of $6.00 per hour for a total of 15,600 hours
in the contract period, and dividing by 156,000 vehicles tested
during that period. Equipment costs per lane are assumed to be
$40,600, also divided by 156,000 vehicles. This works out to 25*
per vehicle. Based on the cost and usage rates shown in Table 5-
3, nitrogen gas for the pressure test costs an additional 13* per
vehicle. Taken together, the pressure test would be expected to
add about $2 to the cost of an efficient, high-volume test system.
Draft -53- 2/26/92
-------
Table 5-6
Costs to Add Pressure Testing to Centralized Programs
Running Total
Increments Per Vehicle Cost Cost per Vehicle
Adjust for Throughput $6.08 * 12.5/10 $7.61
Staff $1.20 $8.81
Oversight $1.25 $10.05
Equipment and Nitrogen gas $0.38 $10.44
5.2.3.4 The Transient/Purge Short Test
The transient/purge test is a longer test procedure than the
ones currently used in most I/M programs and is the longest single
procedure in the whole inspection process. Thus, it is the
determining factor in lane throughput and will therefore influence
the number of test sites required.
The transient test takes a maximum of four minutes to
perform. An additional minute is assumed to prepare the vehicle
for testing, for a maximum total of five minutes. EPA is in the
process of looking at potential fast-pass and fast-fail
strategies, and preliminary results suggest that roughly 33% of
the vehicles tested could be fast passed or failed based upon
analysis of data gathered during the first 93 seconds of the IM240
(i.e., Bag 1) using separate fast-pass and fast-fail cutpoints.
Hence, EPA estimates that the average total test time could be
shortened to at least four minutes per vehicle. This translates
into a throughput capacity of 15 vehicles per hour. To
accommodate peak demand periods and maintain short wait times, a
design throughput rate of half of capacity is assumed, for a
typical throughput rate of 7.5 vehicles per hour. Assuming the
same number of hours of lane operation as previously, the total
number of tests per lane in a transient lane is estimated to be
117,000 over the five year contract period.
State quality assurance program costs would increase given
the complexity and diversity of the test system; an estimate of an
additional 50* is used here but the amount could vary depending
upon the intensity of the oversight function the state chooses
Staff costs per vehicle are calculated using the same assumptions
for wages and hours of operation as shown in Table 5-5; however
the cost is spread over 117,000 tests over the life of the
contract rather than 195,000. The result is staff costs of 80*
per staff per vehicle. Four staff per lane are assumed (it may be
Draft -54-
2/26/92
-------
that, in a well configured lane, three would suffice) to perform
the tests. The additional tasks performed by inspectors in
conducting the new tests - i.e., disconnecting vapor lines and
connecting them to analytical equipment for the evaporative tests
and driving the vehicle through the transient driving cycle - do
not require that inspectors have higher levels of skill than they
do presently. Rather, these tasks can be performed by comparably
skilled individuals trained to these specific tasks. Total staff
costs work out to $3.20 per vehicle. Equipment costs for each
test procedure are derived by taking the equipment costs from
Table 5-2 and calculating the costs of five years worth of
expendables using the figures in Table 5-3 and dividing by
117,000. The resulting costs estimates are shown in Table 5-7.
Table 5-7 shows the result of factoring the figure of $6.08, from
Table 5-5, by the change in the throughput rate and adding in the
equipment, staff, and state costs associated with the new test
procedures. The figure of $6.08 is multiplied by 12.5/7.5, i.e.,
the ratio of the design throughput rate in the current program to
the design throughput rate in a program conducting pressure purge
and transient testing.
Table 5-7
Costs to Add Proposed Tests to Centralized Programs
Increments
Adjust for Throughput
Staff
Oversight
Pressure Test
Purge Test
Transient Test
Per Vehicle Cost
$6.08 * 12.5/7.5
$3.20
$1.75
$0.13
$0.41
$0.87
Running Total
Cost per Vehicle
$10.14
$13.34
$15.09
$15.22
$15.64
$16.50
Thus, the cost of adding the new tests to centralized
networks is found to be about double the current average cost.
The cost of centralized test systems has been dropping in the past
few years as a result of competitive pressures and efficiency
improvements. These factors may drive down the costs of the new
tests as well, especially as they relate to equipment costs.
Given that conservative assumptions were made regarding two
additional personnel, equipment costs of $144,000 per lane, and
low throughput rates, the cost estimate presented here can be
fairly viewed as a worst case assumption. As discussed earlier,
the important issue is the quality of the test, not the frequency,
Draft
-55-
2/26/92
-------
so doing these tests on a biennial basis would offset the
increased per test cost.
5.2.4 Cost to Upgrade Decentralized Programs
5.2.4.1 Basic Assumptions
A methodology similar to that described above is used to
estimate the costs of upgrading decentralized programs. Equipment
and labor costs are key variables as they were in determining
costs for centralized programs. However, total test volume rather
than throughput and test time are the critical factors affecting
cost. Licensed inspection stations at present only perform, on
the average, about 1,025 inspections per year, as shown in Table
5-8 (note that this number is a station-weighted average). This
analysis is conducted based on the assumption that the number of
inspections per station per year on average will not change (i.e.,
time and data available did not permit construction of a supply
curve for testing services).
Table 5-8
Inspection Volumes in Licensed Inspection Stations
Program Vehicles per Year Vehicles per Station
California 6,180,093 799
Colorado 1,655,897 1,104
Dallas/Ft. Worth 1,948,333 1,624
El Paso 278,540 1,161
Georgia 1,118,448 1,729
Houston 1,482,349 1,348
Louisiana 145,175 1,037
Massachusetts 3,700,000 1,321
Nevada 523,098 1,260
New Hampshire 137,137 564
New York 4,605,158 1,071
Pennsylvania 3,202,450 834
Rhode Island 650,000 684
Virginia 481,305 1,301
Weighted Average 1,025
Two conclusions can be drawn from this. The first is that
adoption of the new tests is not likely to affect the total number
of inspections the average station can perform. The second is
that costs will be spread over a smaller number of vehicles than
in the case of high volume centralized stations. The incremental
cost per vehicle for each new test will therefore be larger in
this type of inspection network.
Draft -56-
2/26/92
-------
The current average cost of vehicle inspection in
decentralized programs is about $17.70 (again, the derivation of
this figure can be found in EPA's technical information document,
"I/M Network Type: Effects on Emission Reductions, Cost, and
Convenience"). Note that this figure may substantially
underestimate actual costs since most states limit the inspection
fee that a station can charge. In many cases, this fee is likely
to be below cost; stations presumably obtain sufficient revenue to
stay in business by providing other services, which may include
engine repair. The costs for adding high-tech tests are derived
by adding the incremental costs for each test type to this amount.
Incremental costs are again estimated by dividing the cost of the
equipment and the staff costs by the number of inspections
expected to be performed in a year, or 1,025 tests.
Equipment costs are spread over the useful life of the
equipment. While a piece of equipment's useful life can vary
considerably in actual practice, a five year equipment life is
assumed.
While large businesses, such as dealerships, are often able
to afford to purchase current analyzer equipment outright, the
smaller gas stations and garages often have to finance these
purchases. The higher cost of the equipment needed to perform
purge and transient testing ($144,000 for the dynamometer, CVS,
analyzers, etc., as opposed to $12,000 to $15,000 for the most
sophisticated of the current NDIR-based analyzers) makes it even
more likely that these purchases will have to be financed for most
inspection stations. The costs of the CVS and analyzers for
transient testing and the dynamometer are amortized over five
years at 12% interest in the analysis in this report. The costs
of the other, less expensive pieces of equipment are not
amortized, but are assumed to be purchased outright and the cost
spread out over five years worth of inspections, or 5,125 tests.
Program personnel in decentralized programs were contacted to
determine inspector wage rates. In many cases, inspectors are
professional mechanics earning about $25 per hour. However, most
states do not require inspectors to be mechanics, and inspections
may be performed by less skilled individuals who typically earn $6
or $7 per hour. The prevalence of different wage rates among
inspectors is unknown. Therefore, EPA assumed an average wage of
$15 per hour for this analysis. An overhead rate of 40% is
assumed, for a total labor cost of $21 an hour.
5.2.4.2 The Pressure Test
The pressure test can be performed while the inspector is
conducting the anti-tampering inspection. Thus, virtually no
extra labor time is added. The incremental cost for this test is
the cost of the equipment including the nitrogen gas, which, using
the costs from Tables 5-2 and 5-3 and spreading them over 5,125
tests, yields a cost of 23$ per vehicle. Adding this to the
current average cost of $17.70 yields a total cost of $17.93.
Draft " -57- 2/26/92
-------
5.2.4.3 The Transient/Purge Short Test
The transient/purge test is assumed to require five minutes
of the inspector's time. With labor costs at $21 per hour, as
described above, this works out to $1.75 per vehicle. Equipment
costs are taken from Table 5-2, except that the costs for the CVS,
analyzers, and dynamometer are amortized over the five year period
as described previously. This brings the total costs for these
pieces of equipment over the five year period to $126,793 for the
CVS and analyzer system, and $60,060 for the dynamometer. These
costs are divided by 5,125 inspections. The costs of expendables
from Table 5-3 are added in according to the usage rates assumed
for decentralized programs.
The incremental costs for these test procedures, and the
total cost with pressure testing included, are detailed in Table
5-9.
Table 5-9
Costs to Conduct Pressurer Purge,, and Transient Testing in
Decentralized Procrrams
Test Procedure Incremental Cost Total
Pressure $0.23 $17.93
Labor $1.75 $19.68
Purge (Equipment only) $12.40 $32.08
Transient (Equipment only) $24.97 $57.05
5.3 Repair Costs
5.3.1 HC and CO Exhaust Repair Costs and Methodology
The repair costs for HC and CO exhaust emission repairs are
split into two elements. One addresses the repair costs due to
failure of a tailpipe test, such as the 2500 rpm/Idle idle test or
the loaded transient test. The other element addresses the repair
costs of correcting tampering identified as a result of the visual
inspection for the presence and connection of emission control
components such as the catalyst (also known as "ATP failures")
Draft -58- 2/26/92
-------
5.3.1.1 Tailpipe Emission Test Failures
Based on current information from I/M programs which collect
repair cost information, the average cost to repair a 1981 or
newer vehicle failing the 2500 rpm/Idle test is approximately $75,
including parts and labor. For example, 1989 repair data from the
Louisville, Kentucky I/M program shows the average cost to be $54
for all model year vehicles if only commercial repairs are
included. The overall average cost drops to $42 per repaired
vehicle if the cost of self repairs (repairs performed by the
individual vehicle owner) are also included5. In addition, the $75
average repair cost figure is further supported by the findings
from an I/M repair study conducted in California which showed the
average repair cost to be $72 for 1980 and later model year
vehicles6. In this study, 500 vehicles that failed the California
I/M test were recruited, tested, and repaired at independent
commercial garages to pass I/M. Finally, a study of repair costs
conducted by the Oregon I/M program in 1985 and 1986 found the
average repair cost to be about $50 per failure.7
The average cost to repair a vehicle which fails both the
IM240 and the 2500 rpm/Idle test is also assumed to be $75. This
figure is based on the fact that these cars are likely to receive
on average the same types of repairs as are received by vehicles
failing only the 2500 rpm/Idle test. For the vehicles which fail
only the IM240 emission test, the average repair cost is assumed
to be $150, or twice as much. This higher repair cost accounts
for the additional and more thorough diagnosis needed to identify
the causes of the IM240 failures. In addition, it allows for the
possibility of more costly engine parts being required to repair
the IM240 failures. Therefore, blending the $75 cost of repairing
combined IM240 and 2500 rpm/Idle failures with the $150 cost of
repairing IM240-only failures, and assuming (based on observations
in Indiana) that there are slightly more 2500 rpm/Idle/IM240
failures than IM240-only failures, yields an average cost of $120.
5.3.1.2 Emission Control Inspection Failures
The average cost (separated by model year group) to repair
emission control components identified as needing repair or
replacement by a visual inspection are shown in Table 5-10.
These costs were estimated several years ago, based on
average retail parts and labor costs. For example, the average
air pump repair cost reflects the cost of replacing a broken air
5 "1989 Annual Report Vehicle Exhaust Testing Program Jefferson County,
Kentucky", April, 1990
6 "I/M Evaluation Program Series II", Summary from the California Air
Resources Board's I/M Evaluation Program, October 25, 1991.
7 Jasper, W. P. "A Discussion of Reported Maintenance and Repair Expenses
in an I/M Program", SAE Paper 861547
Draft -59- 2/26/92
-------
pump belt or reconnecting an air or vacuum line. This cost was
based on the assumption that most air pump tampering ^or
malmaintenance will focus on disabling the unit by disconnecting
the belt or line rather than removing the entire unit. If this is
the case, then the repairs will be relatively simple. The average
catalyst replacement cost was based on the retail cost of an
aftermarket converter. The misfueled catalyst replacement
reflected the cost of the converter plus an additional amount to
replace the poisoned oxygen sensor. The evaporative system repair
is the average cost of reconnecting a vapor or vacuum line after a
visual inspection of the system. The PCV and gas cap repairs are
the average cost of replacing these components.
Table 5-10
Averacre Cost of Reoairina Emission Control Components
Component Pre-81 1981+
Air Pump $15 $15
Catalyst Replacement $150 $165
Misfueled Catalyst Replace $175 $190
Evaporative System Repair $5 $5
PCV System Repair $5 $5
Gas Cap Replacement $5 $5
Repair of intentional tampering failures will contribute
relatively little to the overall cost of repairing I/M-failed
vehicles in the 1990s, due to decreasing tampering rates. The
estimated costs per vehicle, therefore, were not revisited.
5.3.2 NO3E Repair Coats and Methodology
Repair costs for NOX reduction, and the supporting analysis
are discussed separately from the HC and CO repair cost analysis
because repairs targeted to reduce HC and CO emissions often have
no effect on NOX emissions. Moreover, the Indiana data showed that
the HC/CO failures and the NOX failures were essentially separate
sets of vehicles8. For example, many vehicles requiring repairs to
correct high HC or CO emissions frequently have fairly low NOX
emissions, and consequently do not require NOX repairs.
Furthermore, for those vehicles which are high NOX emitters, the
November 1991, EPA memorandum from E. Glover to C. Harvey, "Average
Repair Costs and Benefits from Repairing High NOX Emitters."
Draft -60- 2/26/92
-------
most common repair is to the EGR system, and this often has little
impact on HC or CO emissions. In other words, the vehicles with
excessive HC and CO emissions usually need different types of
repairs than those with excessive NOX emissions. Thus, their
repair costs were analyzed separately.
The data used to calculate the average cost and benefit of
performing vehicle NOX repairs was collected in the on-going EPA
Emission Factor test program at the EPA's Motor Vehicle Emission
Laboratory (MVEL) in Ann Arbor, Michigan, as well as at the ATL
facility in South Bend, Indiana. In this program, large numbers
of in-use vehicles were recruited for testing and repair to better
characterize the emissions of the fleet. However, for the
analysis of NOX costs, the overall database was restricted to 1983
and later model year vehicles which had received an FTP test both
before and after repair, and had been tested in the last few
years. As a result, data from 169 1983+ model year vehicles with
repair data were obtained.
Most of the 169 vehicles were high emitters of HC or CO, and
had repairs aimed at those pollutants, since EPA had given the
testing contractor instructions to focus on HC and CO emissions.
In order to more accurately characterize the cost of effective NOX
repairs, criteria were used to further select vehicles which
clearly had high NOX emissions before repair, but had achieved
lower NOX emissions as the result of the repair. These criteria
were: Before repair FTP emissions had to exceed 2.0 gpm NOX, and
after repair FTP emissions could not exceed 1.25 gpm. As a result
of these criteria, 10 cars out of 169 were selected, and 9 were
used in the final cost analysis. Examining the individual vehicle
repairs of these 9 vehicles (see Table 5-11) shows that all of
them needed EGR repairs to lower the NOX emissions to levels which
could meet the criteria. On 6 of these vehicles, the EGR was
replaced, while on the other 3 the EGR passage was cleaned, or the
delay valve was replaced.
The tenth car (683), a Chevrolet Chevette, was removed from
the cost analysis because the repair it received was not targeted
toward NOX reduction. Instead, NOX emissions decreased primarily
due to an ineffective EC/CO repair, which caused the engine to go
to a rich air/fuel mixture as evidenced by a very large CO
emission increase (10 to 30 gpm).
The repair costs of the 9 individual vehicles as well as the
overall averages are shown in Table 5-11. For example, the price
of the repair parts averaged $44, using Mitchell's Summer
Collision Estimating Guide. The labor cost averaged $34, based on
0.68 hours at $50 per hour. These labor hours were determined
using Mitchell's 1991 Mechanic's Labor Estimating Gjiisie.. In
addition, each car was assumed to require 0.5 hours of diagnostic
time at the labor rate of $50/hour for an average cost of $25.
Summing these costs puts the total average cost of an effective NOX
Draft -61- 2/26/92
-------
repair at $103. For input into subsequent cost effectiveness
models this overall cost was rounded to $100.
Table 5-11
Repair Costs
Veh
861
94
803
1095
23
545
1131
1657
41
683
MX
87
86
86
89
87
84
86
83
85
85
Make
MERC
FORD
CHRY
PONT
FORD
CADI
DOD6
CHEV
CHEV
CHEV
Model
COUGAR
THUNDERBIRD
NEW YORKER
GRAND PRIX
TAUR0S
SEVILLE
W150
CELEBRITY
S-10
CHEVETTE
NOx Repair
Description
Replaced EGR Valve
Install EGR Vacuum
Line
Replaced EGR Valve
Clean EGR Passage
Replaced EGR Valve
Assembly
Clean EGR Passage
Clean EGR Passage
Replaced EGR Delay
Valve
Replaced EGR Valve
Clean EGR Passage
Replaced EGR Valve
O2 Sensor, Coolant
Labor
Hours
0
0
0
0
1
0
0
0
0
4
.80
.30
.80
.80
.00
.70
.30
.70
.70
.60
Labor
Cost
$40
$15
$40
$40
$50
$35
$15
$35
$35
$230
Parts
Cost
Retail
$42
$0
$41
$161
$0
$0
$22
$65
$67
$39
Diag-
nostic
Cost
$25
$25
$25
$25
$25
$25
$25
$25
$25
$25
Total
Cost
$107
$40
$106
$226
$75
$60
$62
$125
$127
$294
Temperature Sensor,
Rebuilt Carburetor
AVERAGE with #683
AVERAGE without #683
1.07
0.68
$53
$34
$43
$44
$25
$25
$122
$103
5.3.3
Evaporative System Repair Costs and Methodology
The repair and cost data used to calculate the average
evaporative system repair costs and subsequent fuel economy
improvements were collected during an EPA running loss test
program conducted at ATL during the Spring of 1991 in which
failing vehicles were repaired and retested. All comparisons were
done with data obtained from running loss tests at 95° F using a
9.0 RVP emission test fuel, and 3 consecutive LA-4 test cycles
(the first LA-4 being a cold start).
The cost-benefit calculation was based upon a sample of 25
vehicles which failed either the I/M purge or pressure test in
this test program, and for which evaporative system repair cost
information was available.9 Only 24 vehicles (vehicle 1667 was not
available) were used to calculate the average fuel economy cost
July 26, 1991, EPA memorandum from E. Glover to C. Harvey, "Average
Repair Costs and Benefits from Repairing Purge and Pressure Failures "
Draft
-62-
2/26/92
-------
Test
Pressure
Purge
M
Total Fuel
Economy
Improvement
7.87 gpm 6.1%
8.26 gpm 5.7%
Total
Fuel
Economy
Savings
gal /mi
0.0034
0.0035
Average
Parts
Cost
$15.03
$21.89
Average
Labor
Hour
0.45
0.96
savings resulting from evaporative system repair. The results are
shown in Table 5-12.
Table 5-12
Average Repair Costs and Fuel Economv Benefits
Average
Total
Cost*
$37.76
$70.10
* Labor costs are computed from labor time using a labor rate (including
California) of $50 per hour
The evaporative repair costs, excluding gas caps, are based
on parts costs as invoiced by ATL. If the cost of a repair part
for a particular vehicle was not available, then the average cost
from the other vehicles which also received that repair was used.
For example, in the analysis, the value of $29.46, obtained from
vehicle (1563) was used as an estimate of purge solenoid
replacement cost on two other vehicles (1525 and 1552) which
received that repair, but did not have invoiced repair costs. The
ATL invoiced gas cap replacement cost was available on only two
vehicles (1532 and 1542) . For the other vehicles which required
this repair, the gas cap cost was based on auto dealer retail
prices for an OEM part. Typically, the gas cap OEM retail price
was around $7. In addition, repair parts such as evaporative
hoses, or inexpensive in-line tees were assumed to cost nothing,
except as overhead in the labor cost of fixing them.
The time of repair is generally based on individual
diagnostic and repair durations provided by ATL. Typically, they
include both the time to diagnose the problem and replace or
reattach the parts. For example, vehicle 1548 required 6 hours of
diagnosis to discover the cause of the purge problem and replace
the defective part. Most of the time was spent in diagnosis,
though this length was unusual since most diagnoses and repairs
were completed in a half an hour or less.
In some cases, actual labor times were not available to
diagnose or replace a particular part. In these cases estimates
were made regarding the duration of a typical repair. For
example, gas cap replacement (including diagnosis) duration was
not usually itemized and, therefore, was estimated to be 15
minutes. In other cases, repair times from similar repairs on
other cars were used. However, for the sake of clarity, both the
Draft -63- 2/26/92
-------
parts and labor cost basis of each vehicle's repair are noted in
Table 2 of reference 8.
5.4 Fuel Economy Benefits
5.4.1 Fuel Economy Benefits of Evaporative System Repairs
The analysis of the data shows a substantial fuel economy
benefit under the 95° F test conditions as the result of
evaporative system repair. This fuel economy benefit is
attributed to two factors. The first is increased performance and
efficiency of the vehicle's engine following an evaporative repair
such as reconnection of a vacuum line or a TVS repair. This
increase in efficiency was directly measured by the CVS equipment,
and it was found that the measured fuel economy increased by an
average 3.2% for vehicles failing the pressure test, and an
average 2.8% for vehicles failing the purge test. The fuel
economy improvement is calculated by dividing the fuel consumption
reduction by the total fuel consumption, as illustrated in the
following calculations:
4.13 grams fuel/mile/128.3 grams fuel/mile =3.2% Pressure failures
4.05 grams fuel/mile/143.75 grams fuel/mile =2.8% Purge failures
The second factor involved in the fuel economy benefit
calculation is the utilization of the captured HC vapor which
would have otherwise been lost as running loss emissions. In a
properly designed closed-loop vehicle the engine should
effectively substitute these vapors for liquid tank fuel, and
reduce the vehicle's real fuel consumption. These vapor fuel
flows from the engine and the evaporative canister are not
measured during the running loss test.
Since actual fuel flow data were not measured, it was assumed
that 100% of the captured running loss emissions (i.e., the
difference between before and after repair levels) can be
effectively utilized as fuel. This assumption may be slightly
high given the fact that on average exhaust CO emissions increased
somewhat as the result of evaporative repairs, indicating that
some of the extra fuel was not fully combusted. However, such an
error (i.e., using an 'R' factor of 1.0) is probably small, and
its effect should not be large considering that the running loss
reductions are not large in comparison to total vehicle fuel
consumption.
The running loss vapors from pressure failures were converted
to_liquid fuel, using an R Factor of 1.0, the standard density of
Emission Test Fuel, and a carbon weight factor of 0.83 for the
fuel.
3.74 gpm running loss CH2.33 * R Factor =3.74 gpm liquid fuel (CHi 85)
Draft -64-
2/26/92
-------
3.74 g C/mi * (1cm3/ 0.745 g Fuel) * (1 g Fuel / 0.83 g C) *
(1.0 liter/ 1000 cm3) * (1.0 gal/3.79 liter) = 0.0016 gal/mi
The analogous running loss vapors from purge failures were
converted to liquid fuel, and the percentage fuel economy
improvement was calculated in a similar manner.
4.21 gpm running loss CH2.33 * R Factor =4.21 gpm liquid fuel (CH1.85)
4.21 g C/mi * (1 cm3/ 0.745 g Fuel) * (1 g Fuel / 0.83 g C) *
(1.0 liter / 1000 cm3) * (1.0 gal/3.79 liter) = 0.0018 gal/mi
The measured fuel economy improvements from better engine
operation were combined with the measured running loss reductions
to produce evaporative repair fuel economy benefits of 6.1% for
pressure failures and 5.7% for purge failures. Averaging these
together produced an overall fuel economy benefit from evaporative
repair of 5.9%.
5.4.2 Fuel Economy Benefits of IM240 Repairs
The fuel economy benefit for repairing a vehicle that has
been identified as failing the 0.8 gpm HC outpoint or the 15 gpm
CO outpoint on the IM240 test has been estimated as an increase of
12.6% in overall fuel economy, after repairs. This compares to an
8.0% fuel economy benefit realized by identifying and repairing
vehicles using the 2500 rpm/Idle test as a yardstick. These
percentages are derived from data gathered from the IM240 test
site in Hammond, Indiana, and are based upon an average difference
in fuel economy before and after repairs.
The 12.6% fuel economy benefit assessed for identifying and
repairing vehicles on the basis of the IM240 test lane results is
based upon two groups of 1983 and newer vehicles recruited at the
Hammond test site. The first group included those vehicles that
failed the emissions cutpoints of 0.8 gpm for HC and/or 15.0 gpm
for CO, which were subsequently FTP-tested, repaired and retested
at the ATL facility in Indiana (a total of 42 vehicles) . The
second group consisted of those vehicles that failed the emissions
cutpoints, were FTP-tested, but were not repaired (a total of 10
vehicles). Unrepaired vehicles were assumed to represent a fuel
economy benefit of zero, with the net effect that the overall fuel
economy benefit calculation is conservative.
The 10 IM240-failed vehicles mentioned above were not
repaired and retested because the original design of the testing
program sought to conserve testing slots by applying a criteria
that only vehicles with an FTP result twice the certification
standards for the vehicle would receive repairs and be retested.
These unrepaired vehicles were included in the analysis to
represent that fraction of vehicles (i.e., 19%, or 10 out of 52)
expected to fail the IM240 (in a future I/M program) but which
have only a marginal emissions problem and presumably only a
Draft -65- 2/26/92
-------
marginal fuel economy loss (if any), thus requiring only minimal
repairs which will not result in improved fuel economy. The
averaged fuel economy benefit represents a harmonic average of_the
FTP fuel economy before and after repairs for the 52 vehicle
sample group.
Table 5-13 shows that 17 vehicles which failed at the Hammond
lane were not repaired, which raises the issue of why only 10
vehicles were used to represent the wno improvement" vehicles.
The logic and assumptions were as follows. Of the 72 vehicles
that were repaired, only 42 (58.3%) had all the necessary data to
do the calculations. Assuming the same attrition rate (duetto
incomplete data) for the vehicles that did not receive repairs
(i.e., 58.3% of the 17 unrepaired vehicles) yields a total of 10
vehicles. The net effect of assuming this fraction of "zero
improvement vehicles" is a lower fuel economy benefit for the
IM240 (12.6% instead of a potential 15.7%).
Table 5-13
Zero Improvement Vehicle Sample Size Adjustments
Original # Remaining
of Vehs Description of Data Used and Removed Vehicles
98 1983+ Failed lane 0.8 & 15 & received FTPs 98
17 were less than twice standard & not repaired 81
9 were greater than twice the standard, but were not 72
repaired due to test schedule or cost (engine
rebuild or catalyst)
4 had no as-received IM240s at ATL (IM240-based fuel 68
economy benefits were initially evaluated, so this
test was required. In retrospect, they should have
been added back into the database for the FTP-based
FE improvement)
1 had no after-repair IM240 f!643 (to verify repair 67
success)
25 Failed after-repair IM240 (incomplete ATL repairs) 42
% of 72 repaired that can be included in analysis = 58.3%
58% of 17 <2 x standard & not repaired included as
zero improvement = 10
5.4.3 Fuel Economy Benefit for the 2500 rpm/Idle Te^-fr
The 8.0% fuel economy benefit assessed for identifying and
repairing vehicles on the basis of the 2500 rpm/Idle test is based
upon two groups of 1983 and newer vehicles recruited at the
Draft -66- 2/26/92
-------
Hammond test site. The first group consisted of 6 vehicles that
failed the 220 ppm HC and/or 1.2% CO cutpoints on their initial
2500 rpm/Idle I/M test, received an IM240 before and after
commercial repairs, and received passing scores on the retest.
The second group consisted of those vehicles that returned to the
Hammond lane after commercial repairs, but again failed the 2500
rpm/Idle test. These latter 6 retest failures are considered to
be the result of incomplete repairs, which would be corrected in
an enhanced I/M program.
The before and after IM240 fuel economy data was "corrected"
to reflect FTP fuel economy by employing a correction factor of
1.0925, reflecting the fact that, on average, FTP fuel economy
varies from IM240 measured fuel economy by 9.25%. This variance
reflects the fact that the IM240 and FTP are, after all, different
tests, using different driving cycles, etc. Still, the two tests
show a high degree of correlation, and, in the area of fuel
economy, the variance between the two tests is a relatively
constant difference of 9.25%. Therefore, multiplying IM240 fuel
economy readings by 1.0925 yields a reliable estimate of FTP fuel
economy.
After successful repairs (i.e., those resulting in a passing
retest), some marginal vehicles will fail to realize a noticeable
fuel economy improvement. Using a database of 48 cars, it was
determined that 4 of the vehicles that failed the 2500 rpm/Idle
I/M test were not repaired because their FTP emissions scores were
less than twice their certification standards, leading to the
conclusion that, had these vehicles been repaired, their fuel
economy benefit would be zero. These 4 vehicles represent 8.3% of
the 48 database vehicles for which all the necessary data was
available. Assuming that 8.3% of the 6 vehicles that were still
failing I/M would not get a fuel economy benefit after repairs
yields a figure of 0.48 vehicles that will show no noticeable fuel
economy benefit. Given that half of a vehicle cannot be added to
the database, each of the other 6 vehicles that did pass after
repairs were duplicated yielding twelve vehicles, and 1 vehicle
was added to represent the "no fuel economy improvement" case.
Adding the single "zero improvement vehicle" lowered the fuel
economy benefit of the 2500 rpm/Idle test from 8.6% to 8.0%.
Table 5-14 further details how these numbers were arrived at.
Draft -67- 2/26/92
-------
Table 5-14
Adjusted Zero FE Benefit Vehicle Sample Size
Original # of Remaining
Vehicles Description of Data Used and Removed Vehicles
312 1983+ Failed IN I/M at lane 312
256 not recruited to lab 56
8 missing data 48
44 dirty enough to expect an FE benefit 4
% that failed IN I/M but too clean for a FE 8.3%
benefit (4 of 48)
8% of 6 commercially repaired included as 0.48
zero improvement
While 6 vehicles may seem like a slim database, we did not
want to assume too low a fuel economy benefit for the conventional
2500 rpm/Idle test and risk overestimating the incremental benefit
of the IM240 test. A mid-1980s study with actual or simulated
commercial repairs of older technology 1981-83 vehicles showed
only a 3.5% improvement. This has not been shown to be applicable
to newer technology vehicles. We also did not want to claim too
much benefit. We did not rely on the ATL-performed repairs (as we
did for the fuel economy benefit when using IM240 outpoints)
because the ATL mechanics were instructed to repair all known
malfunctions that would likely affect FTP HC and CO. Therefore,
the emissions and fuel economy benefits would likely exceed what
would actually occur with real world repairs that stop as soon as
the 2500 rpm/Idle test cutpoints are met. In contrast, we judged
that because the IM240 is a mass emissions test that correlates
well with the FTP, real world repairs aimed at making vehicles
pass the fairly stringent IM240 cutpoints would not be so
different from those made by the ATL mechanics. The fact that 25
of the 67 ATL-repaired vehicles still failed the IM240 suggests
that ATL mechanics in general did not go too far.
5.5 Recurring Failure and Repair Rates and Fraction of Fleet
Affected by Fuel Economy Benefits
The rates at which vehicles recurrently fail tailpipe tests
and emission control inspections in an ongoing I/M program (i.e.,
the percentage of failing vehicles in a program that has been
established for a few years) are used within the Cost
Effectiveness Model (CEM) for determining repair costs. Fuel
economy credits for repairs resulting from tailpipe tests are
based on the hypothetical failure rates that would occur in the
Draft -68- 2/26/92
-------
first cycle of the I/M program if it were just starting. These
hypothetical rates in effect represent vehicles that have been and
remain affected by the I/M program that has in fact been
operating.
The exhaust test failure rates for calculation of repair
costs in CEM are in the form of a zero-mile failure rate and a
deterioration rate, such that the fraction of failing vehicles for
a given test type is calculated by multiplying the deterioration
rate by the average mileage and adding that result to the zero-
mile failure rate. Table 5-15 shows the zero-mile and
deterioration rates found in the BLOCK DATA section of the CEM
program listing in Appendix A.
Table 5-15
Exhaust Test Failure Rates
(fraction)
Test
Idle
2-Speed
Loaded
IM240
NOx
Zero—Mile
0.00
0.00
0.0252
0.00
0.032936
Deterioration
(per 10K miles)
0.01
0.01
0.01190
0.0373
0.0084805
Type
(recurring)
(recurring)
(recurring)
(first-cycle)
(recurring)
These numbers are based on regressions of emission test data
from the IM240 lane in Indiana. In 1990 and 1991, Indiana had
just revitalized its moribund I/M program and hence can be
considered to represent a hypothetical I/M program in its first
cycle of inspections) . For the IM240 the first-cycle HC/CO
failure rate per 10,000 miles was 0.0373 at an average of 50,000
miles observed among 3,436 model year 1983 and newer cars in
Indiana. The above recurring rates include adjustment of the
first-cycle rates by a factor of 1/1.87 (e.g., 0.01
0.0187/1.87). This adjustment factor is the recurring initial
failure ratio for idle testing, derived by comparing the Indiana
failure rates with failure rates from other operating I/M programs
with longer histories.
The recurring zero-mile rate used by the model for the IM240
is half of the first-cycle deterioration rate (0.0373/2 =
0.01995). The recurring deterioration rate used by the model for
the IM240 is half of 1/1.87 times the first-cycle failure rate.
This method represents a 50-50 compromise between the following
two assumptions, either of which would be reasonably plausible:
Draft
-69-
2/26/92
-------
(a) The IM240 test will require vehicle repairs sufficient to
return the emission control systems to like-new condition thus
yielding a constant failure rate equal to the rate found for the
first 10,000 miles of operation (0.0373), and (b) IM240 repairs
will deteriorate similarly to idle and 2-speed test repairs, which
would yield a deterioration rate of 0.0373/1.87 = 0.01995).
These failure rates assume cutpoints of 1.2% CO and 220 ppm
HC for the idle and 2-speed tests, and 0.8/15 gpm for the IM240
test. For NOx, separate cutpoints of 1.69 for PFI, 2.50 for TBI,
and 3.99 gpm for carbureted vehicles are used resulting in an
overall nominal failure rate of about 10% on the IM240.
In the case of ATP emission control component inspections CEM
calculates recurring repair rates for the first year a vehicle is
inspected from the difference in tampering rates given by
MOBILE4.1 for the no-program case and the with-ATP case. There is
also a small residual repair rate assumed for latter years, with a
very minor cost impact.
In the case of purge and pressure test failures MOBILE4.1
uses a lookup table which has different malfunction rates for each
vehicle age up to 13 years, and older vehicles are assigned the
rates of the 13 year old vehicles. The malfunction rates range
from roughly 4% to 33% for purge or pressure malfunctions, and 8%
to 50% for the combination of purge and pressure malfunctions.
This lookup table can be found as the EFFECT array at the
beginning of the FAIL function in the CEM program listing in
Appendix A. After appropriately weighting together these purge
and pressure failure rates, MOBILE4.1 uses them in its calculation
of evaporative and running loss emission factors in the absence of
an evaporative I/M program. These malfunction rates would become
the first-cycle failure rates for a new I/M program rather than
recurring failure rates.
CEM assumes these same initial failure rates in determining
fuel economy benefits of purge and pressure tests, since the fuel
economy effect of an I/M program in a given year depends on the
difference between the number of failures that would exist in a
no-program case and the near-zero number present with the I/M
program in operation. The fuel economy benefit calculation using
these failure rates is described in section 5.4.
To determine purge and pressure repair costs, CEM requires
recurring failure rates corresponding to an ongoing I/M program
wherein the failure rate would be lower than the initial failure
rate observed in Indiana's first cycle and used in MOBILE4.1 and
in the fuel economy benefit calculation to represent the no-
program case. The recurring purge and pressure failure rates used
for this purpose are:
Draft -70- 2/26/92
-------
Recurring Purge test failure rate: 3.0%
Recurring Pressure test failure rate: 2.5%
Recurring total Purge/Pressure failure rate: 5.0%
The exact use of these rates can be seen in the FAIL function of
the CEM program listing in Appendix A.
These recurring purge and pressure test failure rates were
derived from the initial rates of MOBILE4.1. As an example, the
5% total failure rate is based on roughly a 50% failure rate for
ten year old vehicles indicating that roughly 5% went bad each
year on average. For an analysis that did not treat age
explicitly this was an assumption that could be used for all ages,
and would definitely not underestimate costs, since much of the
rise to the 50% failure rate happens at higher mileages when there
are fewer cars still in use.
5.6 Method for Estimating Cost Effectiveness of I/M Programs
The cost of an I/M program is determined by summing the
estimated inspection fee costs, the estimated repair costs, and
the negative cost of estimated fuel economy benefits (gallons of
fuel saved * $/gallon). The emission benefits of an I/M program
are determined by subtracting the estimated emissions with the
program from the emissions with no I/M program. CEM does the
emissions calculation by making multiple runs of MOBILE4.1 and
manipulating the results of the various runs. Since MOBILE4.1
does not include the necessary cost components, CEM itself
calculates costs by combining the previously discussed information
on per vehicle costs and fuel economy benefits with the estimates
of failure rates.
Since MOBILE4.1 calculates the emission levels, tampering
rates, and misfueling rates for January 1st of each calendar year,
CEM performs two two consecutive sets of MOBILE4.1 runs and
interpolates between them to get an annual average emission rate
which is then converted into a ton per year value using the fleet
vehicle miles travelled (VMT) data contained in MOBILE4.1. In
order to separate out costs and benefits associated with various
portions of an I/M program, two intermediate MOBILE4.1 runs are
done between the full program and no-program runs. Therefore,
each CEM run performs a total of eight MOBILE4.1 runs as follows.
Draft -71- 2/26/92
-------
1) Full I/M & ATP program (as requested)
2) Run 1 minus any ATP and evap testing
3) Run 2 minus any tailpipe I/M, but with tampering
deterrence effect of I/M
4) Baseline, no program benefits at all)
5) Run 1 for next calendar year
6) Run 2 for next calendar year
7) Run 3 for next calendar year
8) Run 4 for next calendar year
5.6.1 Inspection Costs
Inspection costs are determined by multiplying user-input
inspection costs by the number of vehicles adjusted for compliance
rate (percentage of vehicles that fail to get inspected) .
Separate costs are input for tailpipe emission tests, emission
control checks, purge test, and pressure test. If a program calls
for biennial rather than annual inspections, the inspection costs
per year are divided in half. All default costs are found in the
SETUP routine of the GEM program listing in Appendix A. Default
inspection costs are shown in Table 5-16. Note that the cost of
performing the purge test overlaps many of the costs associated
with transient testing, including the cost of a dynamometer, video
driver's aid (VDA), and the throughput adjustment associated with
the longer test time. If purge testing is assumed, the
incremental cost of including the transient test is relatively
minor, including the cost of a constant volume sampler (CVS) and
the analyzers necessary to perform mass emissions testing.
Table 5-16
Default Inspection Costs in CEM4.1
Test Cost Comments
Steady-state Tailpipe Test j $10
Emission Control Checks
25C-1.75
Pressure Test I 69C
Purge Test
Transient Emission Test
5.6.2 Repair Costs
$6.53
$12 if biennial
Depends on checks done
Includes dyno, adjusted thruput
67C |Increment over purge cost
Calculating total repair costs is performed similarly to the
inspection costs, except that the costs are only applied to the
percentage of vehicles estimated to fail a given I/M test. it is
Draft -72- 2/26/92
-------
further adjusted for the percentage of vehicles that do not get
repaired because they require repairs costing more than the
applicable cost waiver limit. Default repair costs are as
follows.
Table 5-17
Default Repair Cost in CEM4.1
Failure Triggering Repair Pre—81 81+
Idle or 2500 rpm/Idle Test $50 $75
Transient Test (IM240) N/A $150
Air Pump $15 $15
Catalyst $150 $165
Misfueled Catalyst Cost $175 $190
Evaporative System $5 $5
PCV System $5 $5
Gas Cap $5 $5
Purge Test $70 $70
Pressure Test $38 $38
N0x Estimated $10°
In the case of transient exhaust testing, the fraction of
failing vehicles that would have failed a 2500 rpm/Idle test is
assigned the repair cost for the 2500 rpm/Idle test, while the
remainder is assigned the higher transient test repair cost.
5.6.3 Fuel Economy Cost Benefits
Fuel economy benefits are based on cumulative repairs made to
vehicles that fail an I/M tailpipe test and/or an evaporative
system pressure test. As described in section 5.5, the repair
rate used is the first-cycle failure rate corresponding to
inspection of vehicles that have not previously been subject to an
I/M program. The percentage improvement in fuel economy depends
on the type of test that was failed. The following benefits are
from the BLOCK DATA section of the CEM program listing.
Draft -73- 2/26/92
-------
Table 5-18
Fuel Economy Benefits in CEM4.1
Test FE Benefit
2500 rpm/Idle (pre-81) 0.0%
2500 rpm/Idle (81+) 8.0%
IM240 (83+) 12.6%
Purge/Pressure 5.9%
The model converts these percent MPG benefits into dollar benefits
using the VMT information from MOBILE4.1, fleet average fuel
economies for appropriate model years from CEM and a user-input
gasoline cost from CEM, which defaults to $1.25 per gallon.
Draft -74-
2/26/92
-------
6.0 REGULATORY IMPACT ANALYSIS - COSTS AND BENEFITS OF ENHANCED
I/M OPTIONS
6.1 Emission Reduction Benefits
Grams per mile emission factors were calculated using
MOBILE4.1 for low, medium, and high option enhanced I/M
performance standards. The test procedure variables and the
constant inputs are detailed in Table 6-1. All options are based
on centralized testing of 1968 and later light duty vehicles and
light duty trucks with annual frequency, as required by section
182 (c) (3) (B) of the Clean Air Act Amendments of 1990. The major
differences in the benefits of the options are due to the type of
testing utilized, specifically, the adoption of evaporative system
pressure testing in the medium option and of transient and
evaporative system purge testing in the high option. Other inputs
reflect national default values assumed in MOBILE4.1.
Draft -75- 2/26/92
-------
Table 6-1
>r Enhanced I/M
Low
20%
1968+
none
none •
none
none
1%
98%
central
annual
LDV/LDT1/LDT2
1981+
Yes
Yes
No
No
No
No
No
500 feet
11.5
8.7
1992
72°F
92°F
87.5°F
20.6/27.3/20.6
no
no
19.6 mph
MOB4.1 default
Performance Staj
Medium
20%
1968 +
none
1971+
none
none
1%
98%
central
annual
LDV/LDT1/LDT2
1981+
Yes
Yes
No
No
No
No
No
500 feet
11.5
8.7
1992
72°F
92°F
87.5°F
20.6/27.3/20.6
no
no
19.6 mph
MOB4.1 default
rr^r'l options
High
20%
1968-1980
1981-1985
1983+
1986+
1986+
1%
98%
central
annual
LDV/ LDT1/LDT2
1984+
Yes
Yes
No
No
No
No
No
500 feet
11.5
8.7
1992
72°F
92°F
87.5°F
20.6/27.3/20.6
no
no
19.6 mph
MOB4.1 default
Inputs
Pre-1981 Stringency
Idle
2500 rpm/Idle
Pressure
Purge
Transient
tWaiver Rate
tCompliance Rate
*Network Type
*Test Frequency
*Vehicle Coverage
ATP MY coverage
Catalyst
Fuel Inlet
Air Pump
Tailpipe Lead Test
Evap Disablement
PCV Disablement
Gas Cap
Altitude
Period 1 RVP
Period 2 RVP
Period 2 Start Year
Minimum Temperature
Maximum Temperature
Ambient Temperature
Operating Mode
Onboard Controls
Stage II Control
Vehicle Speeds
VMT Mix
t These percentages may not be realistic for some programs, in which case the
program will have to be "over designed" to make up the performance loss.
* Clean Air Act Amendments require these inputs as elements of the
performance standard.
The gram per mile emission factors for each option and the
emission reduction benefit as a percentage of the no-I/M case in
the calendar year 2000 are shown in Table 6-2. The no-I/M factors
were calculated assuming the same RVP, ambient temperatures
maximum and minimum temperatures, operating modes, altitude,
vehicle speeds, and VMT mix variables as assumed for the enhanced
I/M options. Stage II and on-board vapor recovery system effects
were not modeled in either the no-I/M or enhanced I/M options
case.
Draft
-76-
2/26/92
-------
Emission benefits from basic I/M (the current performance
standard) and from the biennial high option program (which EPA
recommends) are also shown. Note that the proposed enhanced I/M
performance standard listed below in Table 6-2 is an annual
program, as required by the Act. Note further that emission
reductions are expressed as a percentage of total highway mobile
source emissions. Many other mobile source programs are described
based on light-duty vehicles; doing so here would show a much
higher percent benefit.
Table 6-2
Benefits of I/M Programs Options*
VOC Emission Effects CO Emission Effects
Scenario
Base - No I/M
Basic I/M
Low Option
Medium Option
Biennial High Option
Proposed Enhanced
Performance Standard
Emission
Factor
(gpm)
2.084
1.971
1.870
1.661
1.495
1.503
Percent Emission
Reduction Factor
(gpm)
Percent
Reduction
5.4%
10.3%
20.3%
28.3%
27.9%
11.874
10.021
8.927
8.927
8.223
8.230
15.6%
24.8%
24.8%
30.7%
30.7%
* Total Highway Mobile Source Emissions in 2000
The results shown in Table 6-2 are our best estimates at this
time but our test programs and data analyses are continuing and we
anticipate refining the numbers as time goes on.
6.2 Cost Effectiveness Estimates
6.2.1
AssumDtions and Inputs for the Program Potions
EPA's estimates of the cost-effectiveness of the three
scenarios are based upon modeling with MOBILE4.1 and CEM4.1 with
assessments done for calendar year 2000, These are compared with
a modeling scenario in which no I/M program is assumed.
The low, medium, and high options are the same as those
described in Table 6-1. The assumed cost for an I/M inspection,
including a visual check of emission control devices, is $8.50.
The incremental cost of adding the evaporative system pressure
Draft
-77-
2/26/92
-------
test is $1.94. The incremental costs of adding the purge and
transient tests are $5.19 and $0.87, respectively. As indicated
in section 5.6.1, the cost of the purge test includes the cost of
a dynamometer and VDA, and also reflects a throughput ^adjustment
to accommodate the longer test; adding transient testing to the
purge test requires the addition of a CVS and the necessary
emissions analyzers. In addition, gasoline is assumed to cost
$1.25 per gallon. The average repair costs shown in Table 5-17
were assumed. It should be further noted that the incremental
costs of adding purge and transient testing to a decentralized
network ($12.40 and $24.97, respectively) are larger than ^in a
centralized network because of the assumption these additional
costs will be spread out over a smaller test volume (i.e., it is
assumed that the average number of vehicles tested per station in
a decentralized network will not change).
6.2.3 Cost-Effectiveness Calculations
Total annual program costs per million vehicles, as
calculated by CEM4.1, are presented in Table 6-3, including
inspection costs, repair cost and fuel economy benefits, shown on
an annual basis. Note that the total cost of I/M increases as the
option becomes more stringent; also note that the total cost (on a
per million vehicle basis) of a biennial high option is less than
both the annual low option and the basic I/M program. These
results make it clear that biennial testing should be a top
priority.
Table 6-3
Total Annual Program Cost
Scenario Cost
Basic I/M $6,412,000
Low Option $9,885,000
Medium Option $11,628,000
High Option $11,390,000
Biennial High Option $5,429,000
The next step is to calculate cost-effectiveness ratios, or
the annual cost per ton of emission reductions. For areas that
are required to do enhanced I/M due to ozone nonattainment (the
majority of enhanced I/M areas), the ratios could be calculated by
dividing the annual program costs, from Table 6-3, and dividing
them by the annual tons of hydrocarbon reductions. The results
are shown in. Table 6-4. Unlike the total costs in Table 6-3, the
cost per ton decreases with program stringency. This is because a
Draft -78- 2/26/92
-------
major part of the cost is the inspection and the small marginal
cost of doing a more effective test is overwhelmed by the large
marginal benefit. This is a critical factor to keep in mind when
choosing among various different ozone control strategies.
Table 6-4
Cost per Ton Allocating All Costs to VOC
Scenario Costs per Ton
Basic I/M $5,410
Low Option $4,404
Medium Option $2,621
High Option (Annual) $1,694
Biennial High Option $879
Since the I/M program yields CO benefits as well as VOC
benefits and some areas need reductions in both, it makes sense to
split the cost among pollutants. High-tech I/M can also obtain
significant NOX benefits and many ozone areas may need NOX control
as well to bring ozone levels into compliance with EPA standards.
To estimate the cost of only the VOC portion of the I/M benefit,
one can assess what the cost would have been to obtain the CO and
NOX reductions by other strategies. If all the program costs were
allocated to NOX reductions (which only occur in the high option
program) , then the cost per ton for the annual high-tech program
would be $6,298 per ton and for the biennial high-tech program
$3,267 per ton of NOX benefit. Alternative costs for NOX
reductions are estimated using cost per ton figures to obtain
stationary source NOX reductions through the use of more efficient
burners, estimated at $300 per ton. Allocating all of the program
costs to CO yields a cost per ton of about $143 for the biennial
high option. Costs for other control programs range from roughly
$100-225 (without fuel economy benefits) for cold temperature CO
standards. Oxygenated fuels programs range from about $200-400
per ton. A conservative, alternative cost per ton figure of $125
was chosen for this analysis. These alternative cost per ton
figures are then multiplied by the annual ton reductions
attributable to the various program scenarios. Other assumptions
about the cost of alternate CO or NOX programs would change the
cost remaining to allocate to VOC. Higher costs would leave less
to assign to VOC and vice-versa.
Since CO reductions are not needed in all areas, and only
about 44% of the vehicles that will be subject to enhanced I/M are
in CO areas, costs are not assigned in all areas. This is done by
Draft -79- 2/26/92
-------
reducing the tons of emission reduction to 44% of full benefit and
using that result to calculate the alternative cost per ton.
The results are shown in Table 6-5. As expected the costs
are lower in all cases, and the biennial high option program is
about $461 per ton.
Table 6-5
VOC Cost per Ton Accounting for NOS and CO Benefit
Scenario Cost Per Ton
Basic I/M $4,518
Low Option $3,655
Medium Option $2,242
High Option (Annual) $1,271
Biennial High Option $461
6.2.4 National Cost of Choosing Low Option I/M
The Clean Air Act requires nonattainment areas to meet
specific milestones of 15% reduction in VOC emissions by 1996 and
a 3% reduction per year thereafter. There are two ways for states
to achieve these goals: impose additional controls on stationary
sources (i.e., those beyond RACT requirements) or additional
controls on mobile sources. The question is: What is the cost of
doing a less stringent I/M program and getting additional
reductions from stationary sources instead?
A low performance standard for I/M means fewer tons of VOC
reductions than a high option program, as shown in Table 6-6. The
low option system, even when implemented in a centralized network,
costs more per ton than the high-tech approach. Thus, if states
choose to implement a low option program there is a direct cost to
the nation because of the higher expense. In addition to the
direct cost, there is also an indirect cost. As more and more
controls are imposed on stationary sources, the law of diminishing
returns would predict that the cost per ton will rise. It is
estimated that the cost of these marginal controls will likely
exceed $5,000 per ton.
Draft -80-
2/26/92
-------
Table 6-6
Total Cost and Benefits of I/M Options
Per Million Vehicles Tons Total Cost
High Option 6f724 $8,544,000
Centralized Low Option 2,245 $8,204,000
Decentralized Low Option 2,245 $17,062,000
To estimate the total cost of implementing a low option
program it was assumed that of the 56 million vehicles subject to
enhanced I/M 42 million vehicles would be in a decentralized
system and 14 million would be centralized. This reflects the
current mix of programs in the affected areas. It was also
assumed that each ton not obtained from I/M would be gotten from
stationary source controls at $5000 per ton. The results are
shown in Table 6-7. The extra direct cost of the low option would
be about $353 million while the indirect cost of the more
expensive stationary source controls amounts to about $1,254
million, for a total of about $1.6 billion in excess cost.
Table 6-7
Excess Cost of Choosing Low Option I/M
Low Option
Vehicles
millions
High Option 56
Centralized 14
Low Option Decentralized 42
Total
Stationary Cost
Total
6.3 National Costs
Low Option 56
High - Low
@ $5000/ton
Excess Cost
and Benefits
Benefits
tons
376,529
31,426
94,279
125,705
250,824
Cost
millions
$479
$115
$717
$832
$353
$1,254
$1,607
6.3.1 Emission Reductions
Estimates of the total costs and emission reduction benefits
of current and future I/M programs were obtained using CEM4.1.
Because average costs and effectiveness vary between centralized
Draft -81- 2/26/92
-------
and decentralized programs10 the costs and reductions were modeled
differently for each program type. The MOBILE4.1 output showing
the scenarios used are in Appendix K. Vehicle population figures
are needed in order to calculate total costs and emission
reductions. Because figures obtained from the states vary in
reliability, estimates were derived based upon Census data for
each area.
As shown in Table 6-8 below, current I/M programs obtain
estimated total annual emission reductions of 116,000 tons of VOC
and 1,566,000 tons of CO. Implementation of the high option
requirements of this proposed action on a biennial basis would
yield estimated annual emission reductions of 384,000 tons of VOC
and 2,345,000 tons of CO from enhanced I/M programs, and 36,000
tons of VOC and 500,000 tons of CO from basic programs. Enhanced
I/M programs would also reduce NOX emissions. The transient test
with NOX cutpoints designed to fail 10% to 20% of the vehicles
would yield estimated NOX reductions of 9% relative to emission
levels with no program in place.
10Tierney, E.,J. "I/M Network Type: Effects on Emission Reductions Cost
and Convenience," U.S. EPA Technical Information Document, number EPA-AA-
TSS-I/M-89-2, January 1991
Draft -82- 2/26/92
-------
Table 6-8
National Benefits of I/M
(tons of emissions reduced annually)
VQC £Q
Reductions from Continuing I/M Unchanged
Centralized Areas 55,540 775,228
Decentralized Areas 60f476 791f167
Current Total 116,016 1,566,395
Expected Reductions from Proposal
Enhanced Areas 384,130 2,345,278
Basic Areas
Centralized 23,289 326,290
Decentralized 12,996 174,186
Basic Total 36.285 5QQ.476
Total Future Benefits 420.415 2,845,754
Thus, enhanced I/M and improvements to existing and new I/M
programs will result in national emission reductions substantially
greater than current I/M programs.
6.3.2 Economic Costs to Motorists
EPA has developed estimates of inspection and repair costs in
a "high-tech" I/M program. The derivation of these estimates is
detailed in section 5.0. A conventional steady-state I/M test
including ATP currently costs about $8.50 per vehicle on average
in a centralized program, and $17.70 per vehicle on average in a
decentralized program. A complete high-tech test, including
transient, purge, and pressure testing, is expected to cost
approximately $17 per vehicle in an efficiently run high-volume
centralized program. In a program where 1984 and later vehicles
received the high-tech test, and older vehicles received a steady-
state test and ATP, and the inspection were performed biennially,
the estimated annual per vehicle cost would be about $9. The cost
is sensitive to whether test equipment and personnel face a steady
stream of vehicles or have idle periods. Therefore the cost would
be somewhat higher in a test-only multi-participant system if the
Draft -83- 2/26/92
-------
inspection network had more excess capacity than a typical
centralized program. Test-only stations may also not be as
proficient in testing each vehicle quickly, adding somewhat to
costs.
The overall average repair cost for transient failures is
estimated to be $120. Average repair costs for pressure and purge
test failures are estimated to be $38 and $70, respectively.
Repairs for NOX failures are estimated to cost approximately $100
per vehicle. Data from the Hammond test program indicate that it
would be very rare for one vehicle to need all three of these
repair costs.
These repairs have been found to produce fuel economy
benefits that will at least partially offset the cost of repairs.
Fuel economy improvements of 6.1% for pressure test failures and
5.7% for purge test failures were observed. Vehicles that failed
the transient short test at the proposed cutpoints were found to
enjoy a fuel economy improvement of 12.6% as a result of repairs.
Fuel economy improvements persist beyond the year of the test.
Currently, there are an estimated 63,550,000 vehicles subject
to I/M nationwide. Of these, 23,574,000 are in centralized
programs and 39,976,000 are in decentralized programs (see
Appendix K) . Inspection fees currently total an estimated $747
million annually, $182 million in centralized programs, and $565
million in decentralized programs. Repair costs are estimated at
$392 million, $140 million in centralized programs, and $252
million in decentralized programs. Current fuel economy benefits
are estimated at $245 million, $92 million in centralized
programs, and $153 million in decentralized programs.
As shown in Table 6-9 below, estimates using EPA's cost
effectiveness model show that total inspection costs in the year
2000 in enhanced I/M programs accounting for growth in the size of
the vehicle fleet are expected to be $451 million, with repairs
totaling $710 million assuming that programs are biennial. Fuel
economy benefits are expected to total $825 million, with $617
million attributable to the tailpipe emissions test and $208
million due to the functional evaporative tests.
In basic I/M programs, total annual inspection costs in the
year 2000 are estimated at $162 million, and repair costs are
expected to be approximately $113 million.
Thus, despite significant increases in repair expenditures as
a result of the program, the switch to biennial testing and the
improved fuel economy benefits from programs will result in a
lower national annual cost of the inspection program.
Draft -84- 2/26/92
-------
Table 6-9
Program Costs and Economic Benefits
(millions of dollars)
Emission
Emission Test Evap
Test Evap Fuel Fuel
Test Repair Repair Economy Economy Net
Cost Cost Cost Savings Savings Cost*
Costs and. Economic Benefits of Continuin I/M Unchaned
Central
Decentral
Total
$182
$565
$747
Expected Costs and
$140
$252
$392
Economic
na
na
Benefits
($92)
($153)
($245)
na
na
$230
$664
$894
From Proposal
Enhanced $451 $489 $221 ($617) ($208) $336
Basic
Central $67 $60 na ($39) na $88
Decentral $95 $53 na. ($31) na $117
Total $162 $113 ($70) $205
Grand $613 $602 $221 ($687) ($208) $541
Total
* Net cost is derived by adding inspection and repair costs and subtracting
fuel economy benefits .
6 • 4 Motorist Inconvenience Costs
There is an additional cost factor associated with I/M, the
cost of the time spent by vehicle owners in complying with the
inspection requirement. This cost was estimated by assuming that
motorists' leisure time is worth about $20 per hour. The amount
of time spent getting an inspection can vary considerably as well
and very little data on this subject is available. For the
purpose of this analysis, it was assumed that motorists typically
spend roughly 45 minutes travelling to the test site, getting
tested, and returning in an efficiently designed high volume test
program.
EPA calculated the cost effectiveness of the biennial high
option with this additional cost included. Table 6-10 below shows
Draft -85- 2/26/92
-------
the estimated total program cost per million vehicles, the cost
per ton with all costs allocated to VOC reduction, and the
adjusted cost per ton of VOC with costs allocated among pollutants
as discussed previously.
Table 6-10
Costs of the Biennial High Option inr.lndina Inconvenience
Total Cost $12,254,000
Cost per Ton
All costs to VOC $1,983
Cost per Ton
Adjusted VOC Cost $1,566
Comparing these figures with those in Tables 6-4 and 6-5 show
that the biennial high option, with motorist inconvenience costs
included, is still more cost effective than the low and medium
options without those costs considered.
7 . 0 REGULATORY FLEXIBILITY ANALYSIS
7.1 Regulatory Flexibility Act Requirements
The Regulatory Flexibility Act recognizes three kinds of
small entities and defines them as follows:
• Small business - any business which is independently owned
and operated and not dominant in its field as defined by
Small Business Administration regulations under section 3 of
the Small Business Act.
• Small organization - any not-for-profit enterprise that is
independently owned and operated and not dominant in its
field (e.g., private hospitals and educational institutions).
• Small governmental jurisdiction - any government of a
district with a population of less than 50,000.
Small governmental jurisdictions, as defined above, are
exempted from the requirements of this regulation. There are no
private non-profit organizations involved in the operation of I/M
programs. Consequently this analysis will be limited to the
affects on certain small businesses, namely providers of
inspection and repair services and of inspection equipment.
Draft -86- 2/26/92
-------
There is a significant impact on small entities whenever the
following criteria are satisfied:
• Annual compliance costs (annualized capital, operating,
reporting, etc.) increase total costs of production for small
entities for the relevant process or product by more than 5%
• Compliance costs as a percent of sales for small entities are
at least 10% higher than compliance costs as a percent of
sales for large entities
• Capital costs of compliance represent a significant portion
of capital available to small entities, considering internal
cash flow plus external financing capabilities
• The requirements of the regulation are likely to result in
closures of small entities
The enhanced I/M performance standard contained in the
proposed action includes new "high-tech" test procedures for newer
vehicles and enables states to obtain significantly higher
emission reductions from their I/M programs than they have
previously. This performance standard will affect different types
of businesses differently. Test providers will need to invest in
new equipment. Repair providers will be repairing more vehicles
for more types of inspection failures. The enhanced performance
standard will also affect different types of inspection networks
differently.
7.1.1 The Universe of Affected Entities
The Regulatory Flexibility Act's definition of "small
business" is based on the Small Business Administration's (SBA)
definitions. These are listed in 13 CFR Part 121 by Standard
Industrial Code (SIC) categories. The types of businesses that
have either been licensed to perform inspections or have been
involved in I/M in some other way, such as by selling inspection
equipment, and their SIC categories are listed in Table 7-1, along
with the size cutoffs used by SBA to define small business for
each. Size cutoffs are defined either in terms of number of
employees or gross annual revenue, expressed in millions of
dollars.
Draft -87- 2/26/92
-------
Table 7-1
Affected Businesses
Sic Description Cutoff
5013 Automotive Part and Supply Wholesalers 100
(i.e., auto engine testing equipment, employees
electrical)
5511 Motor Vehicle Dealers (New and Used) $11.5 M
5521 Motor Vehicle Dealers (Used) $11.5 M
5531 Auto and Home Supply Stores $3.5 M
5541 Gasoline Service Stations $4.5 M
7531 Top and Body Repair Shops $3.5 M
7534 Tire Retreading and Repair Shops $7.0 M
7535 Paint Shops $3.5 M
7538 General Automotive Repair Shops $3.5 M
7539 Auto Repair, Not Elsewhere Classified, $3.5 M
(e.g., radiator shops muffler shops,
transmission shops, etc.)
7549 Automotive Services, Except Repair and $3.5 M
Car Washes (e.g., diagnostic centers,
inspection centers, towing etc.)
Note that although all analyzer manufacturers are "affected,"
the size cutoff of 100 employees prevents them from meeting the
definition of "small business."
7.2 Types of Economic Impacts of Concern
This analysis looks at the types of impacts that inspection
and repair providers in existing programs will experience as a
result of the requirements of the proposed action. Since the
requirements for basic I/M programs will remain essentially the
same as the current I/M requirements, significant impacts are not
expected in these programs. Hence, this analysis will focus on
existing I/M programs that will have to become enhanced. This
analysis assumes that the enhanced program implemented will
reflect the high option on the basis that this would represent a
"worst case" scenario (i.e., that with the greatest economic
impact potential).
Draft -88- 2/26/92
-------
7.3 Changes in Repair Activity
The repair industry in enhanced areas that currently have I/M
programs will enjoy a significant increase in repair revenues.
The repair industry consists of motor vehicle dealers (SICs 5511
and 5521), general automotive repair shops (SIC 7538) and some
gasoline service stations (SIC 5541).
7.3.1 Repair Activity in Current I/M Programs
Reliable data do not exist on the number of repair facilities
in I/M program areas that do I/M repairs. However, repair
revenues that accrue to the industry as a whole can be estimated
using vehicle population data. EPA estimates that there are 64
million vehicles in current I/M program areas, 24 million of which
are in areas with centralized programs. Of these, an estimated 15
million are in areas that will become enhanced. There are an
estimated 40 million vehicles in decentralized programs. Of
these, about 33 million are in areas that must implement enhanced
I/M.
Repair cost information is generally not collected by the
states except when a motorist applies for a waiver. However, as
described in section 5.6, estimates of total repair costs can be
made using CEM4.1. EPA estimates that $392 million worth of
repair business would be generated by current I/M programs in the
year 2000 if these programs continued unchanged, $302 million in
areas that will go enhanced. Of this latter figure, an estimated
$89 million would be performed in areas that currently operate
centralized programs and $213 million in areas with decentralized
programs.
Draft -89- 2/26/92
-------
7.3.2 Repair Activity in Future I/M Programs
The transient test, with its superior ability to identify
excess emissions, is expected to generate more repairs than the
steady-state tests, while the purge and pressure tests will enable
I/M programs to identify excess evaporative emissions for the
first time. Estimates using CEM4.1 indicate that an additional
$100 million in annual repair business will be generated in areas
that currently operate centralized programs, and an additional
$212 million in areas that currently operate decentralized
programs as a result of the requirements proposed in this action.
The additional emission repairs identified by the transient test
are expected to generate an additional $41 million in areas that
currently have centralized programs and $79 million in areas that
currently have decentralized programs. The addition of purge and
pressure testing is expected to generate an additional $59 million
in areas that currently have centralized programs, and $132
million in areas that currently have decentralized programs. Thus
the repair industry in these areas is estimated to receive an
additional $312 million, and a total of $613 million annually as a
result of the proposed action, as summarized in Table 7-2.
Table 7-2
Repair Expenses in Enhanced I/M Programs
(millions of dollars)
Centralized Decentralized All Programs
Current $89 $213 $302
Additional
Transient Repairs $41 $79 $120
Evaporative Repairs $59 $132 $191
Total New $1QQ $211 $311
Total $189 $424 $613
The $311 million in extra repair expenditures is estimated to
comprise about 40% parts cost and the remainder for labor, profit,
and overhead. The automotive parts industry estimates that 20,000
jobs are created for every $1 billion spent on parts. Hence, the
additional parts demand ($125 million) will create 750 jobs in
parts manufacturing as well as additional business for retailers
and distributors, and is likely to create more jobs for clerks and
delivery employees. The remaining 60% is estimated to comprise
about 50% profit and overhead at the repair shop and 50% labor.
Hence, mechanics will earn an additional $93 million over all
program areas. At an average pay rate of $25 per hour this
Draft -90- 2/26/92
-------
translates into 1,800 full time equivalents (FTE) over all program
areas.
Firms that pursue this repair business may need to upgrade
repair technician skills and obtain additional diagnostic and
other equipment to perform effective repairs on new technology
vehicles. Inspection stations in decentralized programs, as well
as many repair shops in centralized programs, possess emission
analyzers. These will be useful in testing those vehicles still
subject to steady-state tests and may be used to diagnose vehicles
failing the transient test and to assess repair success. BAR90
analyzers, in particular, are designed to function as a platform
for a variety of engine diagnostic functions and to download OBD
fault codes.
7.4 Changes in Emission Testing Activity in I/M Areas
7.4.1 The Existing Market in Centralized and Decentralized
Programs
A number of different types of entities are involved in
providing inspections. The centralized programs in the States of
New Jersey, Delaware, Oregon, and Indiana are operated by the
state, those in the cities of Memphis, Tennessee, and Washington,
D.C. are operated by the local government. These programs cover
approximately 6 million vehicles. All of these programs except
Oregon and Memphis will be subject to the enhanced I/M
requirement. Therefore, 5 million vehicles in government operated
programs will be covered by this requirement. The remaining 18
million vehicles are in programs operated by private contractors
(SIC 7549), of which 10 million vehicles are in areas covered by
the enhanced I/M requirement. Both the government agencies, and
the private contractors exceed the cutoffs for small entities.
Inspection providers in decentralized programs fall into all
SIC categories in Table 7-1 except 5013 - Automotive Part and
Supply Wholesalers. However, the prevalence of the different
categories among licensed inspection stations varies. The total
number of inspection stations in decentralized areas covered by
the enhanced I/M requirement are listed in Table 7-3.
Draft -91- 2/26/92
-------
Table 7-3
Number of Inspection Stations by State.
State Stations
California 8,752
Colorado 1,500
Georgia 647
Houston 1,100
Louisiana 140
Massachusetts 2,800
Nevada 415
New Hampshire 243
New York 4,300
Pennsylvania 3,838
Rhode Island 950
Virginia 370
Total 25,055
Data on the distribution of inspection stations among the
different categories are not collected by most states, neither is
data on the number of stations that fall below the cutoffs for
small entities listed in Table 7-1. However, listings of
inspection stations were obtained from California and Pennsylvania
and stations were broken down into the following categories:
Service Stations, gas stations that also perform repairs (5541);
Dealerships (5511 and 5521); Independent Repair Shops (7538); Non-
Engine Repair Shops, such as tire shops, body shops, or
transmission shops (7531, 7534, 7535, and 7539); Retailers (5531);
and Test Only Stations (7549). The California data is based on an
analysis of the entire station population. The Pennsylvania data
is based on an analysis of a 10% random sample of licensed
stations.
Draft -92- 2/26/92
-------
Table 7-4
Inspection Stations bv Cateaorv
Station Type
Service Stations
Dealerships
Independent Repair Shops
Non-Engine Repair Shops
Retailers
Test Only Stations
Total
Station Type
Service Stations
Dealerships
Independent Repair Shops
Non-Engine Repair Shops
Retailers
Test Only Stations
Total
California
Number
2,183
1,361
3,272
734
276
131
7978
Pennsylvania
Number
124
95
67
46
16
0
348
• T-ryrr' J
Percentage
27
17
41
9
3
2
Percentage
36
27
19
13
5
0
Information on number of subject vehicles in each I/M
program, and the inspection fee and the portion of the fee
returned to the state in each program is readily available. EPA
also gathers data on the number of licensed stations in
decentralized programs. With this information, inspection station
revenue in decentralized programs can be estimated. These
estimates for programs in enhanced I/M areas are presented in
Table 7-5.
Draft
-93-
2/26/92
-------
Table 7-5
Inspection
Program
California X1
Colorado
Georgia
Houston13
Louisiana13
Massachusetts
Nevada
New Hampshire
New Yorkt
Pennsylvania
Rhode Island
Virginia
lojtai
Averages weighted
by # of stations
* Simple averages
Stations
8,752
1,500
647
1,100
140
2,800
415
243
4,300
3,838
950
220.
25055
2,088*
(i.e., non-
Station Volumes and Incomes
Vehicles Vehicles
per Year /Station
6,426,636
1,655,897
1,118,448
1,482,349
145,175
3,700,000
523,098
137,137
4,605,158
3,202,450
650,000
481.305
24r127P653
2,010,638*
•weighted)
734
1,104
1,729
1,348
1,037
1,321
1,260
564
1,071
834
684
1.301
963
Fee
$48. 3912
$9.00
$10.00
$11.25
$10.00
$15.00
$16.00
$14.00
$17.00
$8.48
$12.00
$12.50
$15.39
State
Share
$6.00
$1.50
$0.50
$3.50
$5.25
$2.50
$3.00
$1.25
$1.25
$0.48
-0-
$1.10
$3.35
Net
Revenue
$31,127
$8,279
$16,422
$10,444
$4,926
$16,518
$16,386
$7,195
$16,868
$6,675
$8,211
$14r829
$18,914
11 BAR 90 analyzers are used in these programs. All others currently use BAR
84 except Houston, Louisiana, and Rhode Island.
12 This figure was supplied to EPA by the State in October of 1991 and
represents an estimate based upon data from calendar year 1990. In its
Third Report to the Legislature (December 1991), the I/M Review Committee
reported an average cost per inspection of $36.23. This number is based
upon a. survey conducted in September 1991, and includes only the cost of
the inspection (not the $6 fee for the certificate). The resulting figure
of $42.23 suggests that, at least during September 1991, the average fee
charged to motorists may have dipped slightly.
13
Current I/M inspection is anti-tampering only. Station, vehicle, and
income data may change with the addition of tailpipe emissions testing'.
Draft
-94-
2/26/92
-------
The costs incurred by inspection stations are driven by a
number of factors. Labor (i.e., the amount of time required to
perform the inspection and the inspector's hourly wage) appears to
be the largest component of cost. The cost of the analyzer is the
second largest component. PC-based (BAR 90) analyzers are the
latest generation of analyzers used in decentralized programs.
Their cost can vary from $13,000 to $20,000. The most common
price appears to be approximately $15,000 each. A number of
service station based programs in areas required to implement
enhanced I/M are currently using BAR 84 analyzers. These cost
approximately $5,000 each. Many stations in the older BAR 84
programs have paid off the cost of their analyzers, which in turn
decreases their annual inspection expenses. Analyzer service
contracts and calibration gas add lesser increments to the total
cost.
Estimates were made of the typical costs incurred by
inspection stations, net profits were estimated and the results
presented in Table 7-6. While large businesses may be able to
afford to purchase current analyzer equipment outright, the
smaller entities, with which this analysis is concerned, often
have to finance these purchases. Analyzers are assumed to be
purchased and paid off over a five year period at a 12% rate of
interest. Conversations with program personnel in decentralized
programs indicated that inspectors are paid about $15 per hour.
Overhead (employers taxes, benefits, etc.) is assumed to be 40%,
for a total labor cost of $21 per hour.
Some cost factors are subject to regional variability. Local
data, as reported by state program officials and EPA Regional
offices, is used for such parameters as number of vehicles per
station per year, average length of test, and cost of service
contracts. Labor and equipment costs are estimated as described
previously. In programs where the equipment specification is more
than five years old, the analyzers are assumed to be paid off.
This, in turn, increases the stations' profits. The results are
listed in Table 7-6.
Draft -95- 2/26/92
-------
Table 7-6
Averaae Inspection Station_Revenues. Costs, and Profits
State
California11
Colorado
Georgia
Houston13
Louisiana13
Massachusetts14
Nevada
New Hampshire
New York11
Pennsylvania14
Rhode Island14
Virginia
Average
Average W/Q CA
Average w/o CA & NY
Vehicles
/Station
734
1,104
1,729
1,348
1,037
1,321
1,260
564
1,071
834
684
If301
963
1,086
1,091
Fee
$48.39
$9.00
$10-00
$11.25
$10.00
$15.00
$16.00
$14.00
$17.00
$8.50
$12.00
$13.50
$15.39
$12.39
$11.93
Net
Revenue
$31,127
$8,279
$16,422
$10,444
$4,926
$16,518
$16,386
$7,195
$16,868
$6,675
$8,211
$14,829
$18,914
$12r357
$10,741
Annual
Cost
$11,899
$5,202
$9,320
$7,075
$5,444
$13,498
$7,681
$4,257
$20,268
$2,811
$2,653
$5P546
$10,818
$10,238
$6,645
Net Profit
$19,228
$3,078
$7,102
$3,369
($518)
$3,020
$8,705
$2,938
($3,400)
$3,864
$5,557
$9r283
$8.096
$2,120
$4,097
This analysis revealed anomalies in the California and New
York programs relative to the others. California has a much
higher average fee than the other programs, and estimated average
profit is nearly twice that of the next highest program. The
estimate for New York reflects an unusually long test duration
(see Table 7-11) and shows the average station operating at a
loss; this estimate is supported by reports that station operators
have sued the state to be allowed to charge a higher fee.
Therefore, average revenues and profits were also calculated with
data from those states omitted.
These figures, based on the average inspection volumes for
each state, show that inspection services, by themselves, do not
14
Due to the age of the state analyzer specification, analyzer costs are
assumed to be paid off in stations in these programs.
Draft
-96-
2/26/92
-------
yield significant profit to the average inspection station. While
the average profit is low, the amount of revenue and profit can
vary a great deal among inspection stations since inspection
volumes vary considerably as well. The best available data on
station volumes was obtained from the California program. The
data covers a three month time period and is shown in Table 7-7.
Table 7-7
Inspection Volumes in California
Tests
0
1-100
101-200
201-300
301-400
401-500
501+
Total
Total Active
Stations
1,958
1,156
1,676
1,178
754
469
1.571
8.752
6,794
% Total
22
13
19
13
9
5
18.
% Active Stations
NA
17
25
17
11
7
23.
EPA analyzed revenues and profits for inspection stations at
different volumes; the results are presented in Table 7-8.
Revenues, costs and profits are calculated as in Tables 7-5 and 7-
6. California has a market based inspection fee, (i.e., stations
charge what the market will bear, since the state does not
regulate the fee) . Conversations with California program
officials indicate that higher volume stations charge lower fees
than the average. The fees assumed for 1,200 and 2,000 inspection
per year cases are based on figures suggested by the state.
Draft
-97-
2/26/92
-------
Table 7-8
Station Revenues and Profits bv Volume
Veh/Qtr
0
100
300
500
Veh/Year
0
400
1200
2000
Fee
$48.39
$48.39
$42.00
$32.00
Net Revenue
$0
$16,956
$43,200
$52,000
Annual Cost
$5,474
$8,974
$15,974
$22,974
Net Profit
($5,474)
$7,982
$27,226
$29,026
These figures indicate that inspections can be profitable if
volume is high, however, relatively few stations have high
inspection volumes. Based on the data in Table 7-7, 22% of the
licensed stations perform no inspections and therefore are losing
money invested in equipment, licensing, and training (only
equipment costs are estimated here). An additional 32% perform
800 inspections per year or less, and therefore appear to be
earning only a modest level of profit. 22% perform from 800 to
1,600 inspections per year, and an additional 23% perform more
than 1,600 inspections per year. Profitability is higher in these
latter two categories.
7.4.2 Future Market in Enhanced I/M Programs
Test providers will be required to invest in new equipment
for that portion of the subject vehicle fleet that will undergo
transient, purge, and pressure testing. The total cost to re-
equip an existing inspection site to perform the new tests is
estimated at about $144,000. EPA based this estimate on
conversations with equipment manufacturers over the past year;
more recent information indicates that a lower figure is likely.
7.4.3 Centralized Programs
As indicated in section 5.0, throughput rates would be lower
in centralized lanes performing transient, purge, and pressure
testing than in inspection lanes performing the current test
procedures. Since programs will be able to switch from an annual
inspection frequency to biennial at the same time they implement
the high-tech tests, EPA does not anticipate that a significant
number of new inspection lanes will need to be built in
centralized programs in order to satisfy the proposed requirements
and maintain waiting times at minimal levels.
7.4.4 Decentralized Programs
Enhanced areas that currently have decentralized programs
will have two options in meeting the requirements of the proposed
Draft -98- 2/26/92
-------
action: they can institute either a multi-participant test-only
network, or a single operator centralized system.
If a program were to switch to a multi-participant, test-only
system, stations that currently participate in the test and repair
network would have a choice between concentrating on inspections,
and becoming test-only stations, or concentrating on repairs.
That choice would likely be driven by the station's current
inspection volume and the degree to which its prospective income
is expected to be derived from inspection as opposed to repair and
other services. This analysis utilizes the simplifying assumption
that stations that perform a large volume of inspections, and that
currently derive more income from inspection than from repair or
other services, would be likely to become test-only stations. By
the same reasoning, stations that are more oriented toward repair
would focus on the additional repair business generated by the
inspections conducted elsewhere.
Data correlating average inspection volume with station type
are not available. However, survey data of motorists in I/M
programs point to the fact that stations that currently focus on
repair work and that do a steady volume of repairs are often
unable to make facilities available to provide inspections
promptly on request15. 27% of motorists in decentralized programs
reported being asked to bring their vehicles back for testing
another time. 20% reported having to take their vehicles to more
than one station to obtain a test. Nearly one out of three had to
leave their vehicles for inspection. On the average, the vehicles
had to be left for five hours. These data suggest that a focus on
repair leads to reduced opportunities to perform inspections and
probably to lower inspection volumes as a result.
The converse appears also to be true. Stations that are
readily able to provide inspections are often either unable, or
simply have not chosen to perform repairs. 53% of motorists
reported taking their vehicle to another station, other than the
one where the inspection was performed, for repairs.
Based on the data from Pennsylvania and California, the
following distribution of station types is assumed for this
analysis:
15 "Attitudes and Opinions Regarding Vehicle Emission Testing,
Research. September, 1991
Draft ' -99- 2/26/92
-------
Table 7-9
Assumed Station Distributions
Station Type Percentage
Service Stations 32
Dealerships 22
Independent Repair Shops 30
Non-Engine Repair Shops 11
Retailers 4
Test Only Stations 1
Some stations, such as dealerships and independent repair
shops, would be likely to concentrate on I/M repairs since their
business already has a decided orientation toward engine repairs.
Together, these constitute 52% of the assumed station population.
Because of their focus on repair, it is likely that these stations
tend to have lower inspection volumes, as discussed above, and
some of them are likely to be among the 22% of stations that
report no testing activity. For the purposes of this analysis, it
is assumed that half of the inactive inspection stations are in
this repair-oriented group.
These repair oriented stations will likely get the majority,
though not all, of the additional repair business estimated
previously at $211 million among all decentralized programs. If
these stations ultimately get 85% of this business (allowing for
15% of the repair stations to come from other categories, mainly
service stations) it will amount to annual revenues of roughly
$13,000 per year. This would offset inspection losses of $10,000
to $12,000 per year (Table 7-6).
The stations that have higher inspection volumes than average
are likely to be deriving a substantial portion of their current
profit from the inspection business and relatively little or none
from repair. Based on the California data, it is assumed that the
23% of the stations that have inspection volumes of approximately
200% of the program average or more would be likely to opt to
become test-only stations. Test-only stations, in those
decentralized programs where they exist, would, of course, be in
this group.
Some stations in this high volume group may be repair-
oriented stations, such as dealerships, independent repair shops,
and some service stations, and may prefer to opt out of the
inspection business for more profitable repair business. This
would create opportunities for other businesses to enter the test-
Draft -100- 2/26/92
-------
only market, including stations whose current inspection volume is
somewhat lower.
Current repair revenues in decentralized enhanced programs
are estimated at $213 million. If this 23% segment of the
stations had been getting 23% of this business (based on the
foregoing discussion, they have probably been getting less), then
they are giving up current annual revenues of $8,500 each in order
to pursue the inspection market.
The remaining 25% that do not have a clear orientation toward
engine repair, and that do not perform a high volume of
inspections, are a mix of service stations, whose business is a
mix of gasoline sales and, in some cases, engine repairs including
I/M repairs on some portion of the vehicles they test; non-engine
repair shops, such as tire shops, muffler shops, transmission
•shops, etc.; and retailers. Members of this group are assumed to
make up the other half of the 22% of stations that do no
inspections. These stations would not be adversely affected by
this rulemaking since they are currently deriving no income from
the inspection business.
This leaves 14% of the population of licensed inspection
stations that do not have a clear orientation toward engine repair
and derive some income from inspections. Since they are not high
volume stations, stations in this group do not derive high profits
from inspections on the average. Table 7-10 shows the projected
current revenues and profits for these stations assuming that they
are evenly distributed among the four low to medium groups in
Table 7-7 (those doing 1 to 400 inspections per quarter), assuming
that all stations charge the average fee of $48.39. Note also
that the numbers of inspections in each category represent the
mid-points of the ranges presented in Table 7-7. The column
entitled "% Avg Profit" shows the estimated profit for each
category as a percentage of the program average profit for
California in Table 7-6.
Given that the average profit in California is almost double
that for the next most profitable program, the profits calculated
based on California data were adjusted to reflect projected
national profits for stations with inspection volumes ranging from
about 25% to 200% of the average for the program. The national
average profits are based on the figure of $4,097 obtained as the
average net profit without data from California and New York.
Draft -101- 2/26/92
-------
Table 7-10
Revenues and Profits for Lowand Medium Volume Stations
Veh/Qtr
50
150
250
350
% Avg
Vol.
27
82
136
191
% Total
Stations
3.36
4.90
3.36
2.38
Net
Revenue
$8,478
$25,434
$42,390
$59,346
Net
Profit
$1,254
$14,710
$28,166
$41,622
Avg. Prof it
6.5
76.5
146.0
216.0
Profit Based
on Nat ' 1 Avg
$266
$3,134
$5,982
$8,849
The first two categories, representing 8% of the total number
of stations, appear to earn 77% of the program average profit or
less. The two higher volume categories, representing roughly 6%
of the total station population, derive substantial profits from
the inspection business (these estimates are based on data from
California which has the most profitable inspection program;
profits in other states probably do not increase with increasing
test volume as steeply as this analysis suggests, while revenues,
on the other hand, do increase in direct proportion to volume).
Data on the relative contribution of inspection revenue, compared
to other types of business are not available. Some of these
stations may be service stations that are currently doing a
profitable business in engine repairs, and would continue to do
so. Others, such as the 2.38% earning an estimated 216% of the
average profit might still opt into the test-only business where a
high volume station has opted out, as discussed previously.
Others, such as the non-engine repair shops and the retailers have
primary lines of business unrelated to I/M.
However, it may be that some of those stations earning 200%
or more of the average revenue would be unable to recoup this loss
any other way, and would be forced to close. The average revenue
loss for these stations would be $37,828 nationally, and $21,482
outside California and New York. It may also be that some of the
stations in the lower profit categories are so marginally
profitable that loss of inspection business would result in
closure as well. If 10% of this group of stations without clear
I/M related alternatives (14% of the total) were to close it would
amount to a total of roughly 350 stations nationwide.
If a single contractor centralized program were instituted in
an area where a decentralized program is currently operating the
option to pursue the test-only business would not be available to
the 23% of the station population that would be likely to pursue
it. Based on the foregoing analysis, these stations have current
inspection volumes of 200% or more of the program average, and may
have average profits of roughly 220% or more of the program
Draft
-102-
2/26/92
-------
average. Members of this group without profitable alternatives
would also face the risk of closure.
The likelihood of closure would depend upon the fraction of
income derived from inspections. Data on this is not available.
Since many of these stations have other lines of business, such as
gasoline sales, auto parts sales, or various types of vehicle
repair and servicing, the loss of business will not necessarily
mean closure. The fraction of these stations that would be unable
to recoup this loss and face closure is difficult to estimate
given the paucity of data. However, if, as before, 10% of these
stations were to close as a result of a switch to a single-
contractor centralized system, as well as 10% of the 14% of
stations identified previously as being at risk, then 927 stations
might close nationwide if all decentralized programs in enhanced
I/M areas switched to centralized, single-contractor systems. If
the areas containing half of the current inspection stations were
to switch to single-contractor, centralized systems, then
potential closures would number about 464.
The most severely impacted would be the test-only stations,
which in California comprise 2% of the test stations. Given that
they have no other lines of business to compensate for the loss of
inspection revenue, these stations would almost certainly close if
the area were to switch to a centralized, single-contractor
system, unless these stations were able to win the contract (some
of these businesses have indicated to EPA they they would try to
do so) .
7.4.5 Impact on Jobs in Decentralized Programs
Table 7-11 shows the number of inspectors in each program,
and the average number of inspectors per station for all
decentralized enhanced programs except Rhode Island, for which
data on the number of inspectors is unavailable. The national
weighted average number of inspectors per station excludes the
highest and lowest averages in the set, those from New York
(program officials in this state have indicated that the total
number of licensed inspectors is likely to include individuals no
longer working as inspectors) and Massachusetts.
Draft -103- 2/26/92
-------
Table 7-11
1* mitkyc J. JJ l^jj
Stations
8,752
1,500
647
1,100
140
2,800
415
243
4,300
3,838
370
ted Averac
1- J.J.H-'y^^ ^^r*.^ ff
Inspectors
18,000
2,930
2,845
2,645
513
1,208
1,249
933
21,640
19,221
1*114
re
Average
2.06
1.95
4.40
2.40
3.66
0.43
3.01
3.84
5.03
5.01
3. 01
2.05
Time per Test
25
5
10
15
15
25
10
5
40
3
5
20
California
Colorado
Georgia
Houston
Louisiana
Massachusetts
Nevada
New Hampshire
New York
Pennsylvania
yj.rcfi.nia
Average station volumes are low (Tables 7-5 and 7-6) - about
four per day. Given that there are, on the average, two
inspectors per station, and that the average inspection takes
twenty minutes to perform, it follows that the average inspector
spends 40 minutes per day performing inspections. This works out
to 0.08 of an FTE (i.e., inspections take about three hours and
twenty minutes out of a forty-hour work week). Hence, inspectors
are generally individuals employed primarily for other jobs (in
most cases as mechanics) who spend a small amount of their time on
inspections. Communications with program officials in these
states and EPA's experience in auditing these programs support
this conclusion. Table 7-12 shows the estimated total number of
FTE devoted to inspections in the different station categories
developed in this analysis, using the volume assumptions developed
previously.
Draft
-104-
2/26/92
-------
Table 7-12
Estimated Inspection FTE
Station Type %_ Number Tests/Day FTE
Repair Oriented 52% 13,029 3 1,612
Inspection Oriented 23% 5,763 8 1,902
No Inspections 11% 2,756 0 0
Remainder 14% 3f 508 4. 579
Total 4,093
In most cases, the time spent on inspections could be easily
re-oriented toward other tasks if inspection business were to
cease, however, some stations might experience some contractions
as a result of losing inspection business, and some might close,
as estimated previously. For the sake of analysis, all FTEs
currently devoted to inspections in decentralized enhanced
programs, as shown in Table 7-12, are counted as lost. Estimates
are also made of additional FTEs lost as a result of potential
station closures.
If a multiple independent supplier program were instituted,
it was estimated that 10% of the 14% of stations that have some
inspection business, and are not clearly positioned to pursue
either the inspection or repair markets, might potentially close.
Assuming that these stations have two FTEs in addition to
inspector FTEs, total job losses would amount to an additional 700
FTEs.
In the event of a switch to a single-contractor centralized
system, 10% of the 23% of stations that would otherwise have
pursued the test-only option would also be at risk of closing.
Potential closures are estimated to total 927. Average non-
inspection FTE per station in this case is assumed to be 2.5 since
some larger stations would be included in the risk group. In this
case, losses could total an additional 2,318 FTEs.
New jobs would be created by the test-only program, and the
increased repair business that would offset these potential losses
to the small business community and to labor.
EPA estimates that in a high volume enhanced I/M lane,
testing an average of 7.5 vehicles per hour, 3-4 inspectors would
be needed per lane instead of the 1-2 typically employed in
current high volume systems. Using an industry estimate of 267
FTE per million vehicles, and assuming a 20% retest rate, 5,340
FTEs are required to test the 33 million vehicles in currently
decentralized programs on a biennial basis (this estimate is based
Draft -105- 2/26/92
-------
on the assumptions and methodology developed in section 5.2 of
this report, "Estimated Cost of High-Tech I/M Testing").
In a multiple independent supplier system volume would likely
be lower. This analysis estimates that 4,200 inspections per
year, or about 16 per day would be likely. Therefore, two or
three inspectors per lane would be adequate. If two inspectors
per lane were employed, 11,525 FTEs would be created if all
current decentralized areas adopted a multiple independent
supplier system.
Additional jobs that would be created in the repair sector
were estimated previously in this analysis. Approximately 1,217
mechanic FTEs, and 506 FTEs in auto parts manufacturing would be
created, in addition to clerical, delivery and other support
personnel. The results are summarized in Table 7-13.
Some new inspection facilities would be constructed whether
programs adopted multiple independent supplier networks or single
contractor networks, also creating jobs. FTE estimates are based
on an industry estimate that construction of an inspection station
requires 4.79 man years of construction and 5.1 man years of
subcontracting. An average station is assumed to have 2.4 lanes.
The number of lanes required to inspect the fleet is based on the
assumptions of biennial inspections and a 20% retest rate. FTE
calculations are based on the assumption that total effort, i.e.,
modification of existing structures in those areas adopting
multiple independent supplier programs and construction of new
facilities in those areas adopting single-contractor programs, is
equal to that needed to construct lanes for half of the vehicles
in decentralized enhanced areas. The results are summarized in
Table 7-13.
Draft -106-
2/26/92
-------
Table 7-13
Summary of FTE Gains and. Losses
(in currently decentralized areas required to do enhanced I/M)
Losses
Gains
Current Inspection FTE 4,093
Station Closures
Multiple Independent 700
Contractor 2,318
New Inspector FTE
Multiple Independent
Contractor
New Repair FTE
Mechanic
Parts Manufacture
Construction
11,252
5,340
1,217
506
587
Net Gain
Multiple Independent
Contractor
7.4.6 National Impact on Jobs
8,769
1,239
EPA has estimated the total FTE in current I/M programs and
the projected changes in FTE nationwide as a result of the
proposed changes. These are summarized in Table 7-14. Note that
Table 7-14 includes areas which will be starting enhanced or
basic programs from scratch, while earlier tallies included only
areas already operating I/M programs.
Draft
-107-
2/26/92
-------
Table 7-14
Impact on Jobs of I/M Proposal
Current Test and Repair Jobs
Inspector Jobs FTE
Decentralized Programs 6,600
Centralized Programs 2,500
Repair Jobs
Decentralized Programs 800
Centralized Programs 1,500
Total Current Jobs 11,400
Future Test and Repair Jobs
Enhanced I/M Programs
Inspector Jobs
Multiple Independent Supplier
Single Contractor
Inspector Job Subtotal
Repair Jobs
Basic I/M Programs
Inspector Jobs
Repair Jobs
Total Future Inspection and Repair Jobs
Other Job Gains
10,500
2.700
2,700 - 10,500
5,500
2,700
700
11,600 - 19,400
Parts Manufacturing
1,034
Draft
-108-
2/26/92
-------
Construction 1,800
Small Business Services 800
Total Net Gain in Jobs 3,800 - 11,600
Small Business Services are estimated by assuming 15
additional FTEs per urbanized area. The 800 FTEs presented in the
table represent the jobs generated in the 52 urbanized areas that
do not have I/M programs now, but will be implementing them as a
result of the proposed action.
Whether programs adopt a multiple independent supplier
network or a single-contractor centralized one there will be
shifts in job opportunities with some net gain in either case.
Hence, the shift to high-tech enhanced I/M may cause significant
shifts in both business and job opportunities. Small businesses
that currently do both inspections and repairs in decentralized
I/M programs will have to choose between the two. Significant new
opportunities will exist in these areas for small businesses to
continue to participate. EPA believes there are ways states can
help test stations make the transition to an enhanced I/M program.
7.5 Mitigating the Impact of Enhanced I/M on Existing Stations
Three potential approaches to helping test stations make the
transition are presented here. The first approach would provide
direct assistance to stations that might be adversely affected by
the transition to a high-tech system. The second would be to
design the enhanced program to include transitional mechanisms to
soften the impacts of the new system. The third would be for
states to establish programs to assist stations and inspectors
through retraining and retooling programs. The previous section
discussed various strategies to assist repair technicians in the
retest process, including free retests and priority access to
retest lanes, as well as diagnostic and repair assistance.
In some states that are currently decentralized and will have
to implement enhanced I/M, analyzers have been in use for 10 years
or more and are fully amortized. In states that upgraded to BAR90
equipment (California and New York), the equipment was purchased
since 1990, and has years of useful life left. A number of other
states upgraded their equipment to BAR84 in the period from 1987
to 1990. Stations in these areas are likely to still be paying
for their equipment (see the footnote to Table 7-6). One means by
which the state could provide direct assistance to current test
stations would be to set up some type of state-supported analyzer
buy-back program for stations that were no longer going to
participate in either the test or repair business, possibly using
funds obtained from inspection fees. BAR90 analyzers would be
needed in the repair business both for diagnostic and repair work
Draft -109- 2/26/92
-------
as well as to check whether repairs on old technology vehicles
were effective. BAR90 analyzers could also be used to test older
technology vehicles in test-only stations. This concept would
allow stations that were planning to leave the I/M business to
recover all or part of their capital investment for equipment that
could not be used for diagnostics and repair. Such a buy-back
program might allow a fairer transition to test-only status.
A related strategy would be for EPA, the states, and industry
to support the development of new and improved uses for BAR90
analyzers so that current as well as future analyzer owners can
use this technology more effectively in the repair process. In
particular, it was California's intent in developing the BAR90
specification for the computer in the analyzer, which is an IBM
386 DOS-based system, to become a platform for vehicle diagnosis
and repair. EPA, the states, and industry could potentially
provide technical and financial support to speed the development
of such software. This would not only make better use of the
equipment in the field but would serve as an excellent mechanism
for providing critical technical assistance and training to the
repair community.
A second strategy to mitigate the impacts is to design
transitional features into the program. One approach would be to
allow test and repair shops to continue to do testing on vehicles
not subject to the transient/purge test for some transitional
period (note that EPA's recommended enhanced program would require
biennial, transient/purge tests on 1984 and later model year
vehicles, and biennial steady-state tests on older vehicles). EPA
is proposing to permit a phase-out of the decentralized test-and-
repair portion of the program such that all vehicles would be
inspected in test-only stations starting January 1, 1996. This
would allow these decentralized stations to continue to obtain
revenue to recover the investment made in testing equipment and
would allow additional time to plan other strategies to replace
the income to be lost from testing.
A related approach is to allow vehicles that have failed
initial inspections in test-only stations to be retested in
existing test and repair stations using conventional test
techniques during the first inspection cycle. This would allow
those stations to attract customers, conduct testing and perform
repairs, with the added benefit of sparing the customer from
returning to the test-only station for the retest.
A third strategy would be to provide targeted assistance to
stations to assure they were able to provide high-tech repair
services. This would require pre-program start-up training to
bring repair technicians in these stations up to speed on the
high-tech tests, vehicle diagnosis, and engine repair. It might
mean tuition grants or other financial assistance. This dovetails
with stronger repair technician training programs which EPA
envisions as being part of future I/M requirements, but differs in
Draft -110- 2/26/92
-------
terras of funding, timing, and intensity. This approach might also
include financial assistance to stations can for the purchase of
equipment to perform sophisticated diagnosis and repair on new
technology vehicles or to upgrade tools and equipment for more
sophisticated diagnosis and repair.
EPA does not consider this to be an exhaustive list of
possible strategies. The Agency expects to receive comments from
the states and affected industries on additional ways to mitigate
the impacts of upgrading to an enhanced program on current test
and repair stations in decentralized programs.
Draft -111- 2/26/92
-------
8.0 ONBOARD DIAGNOSTICS AND ON-ROAD TESTING
8 .1 Onboard Diagnosticsr Interim Provisions
EPA is required to issue onboard diagnostic (OBD) regulations
by May 15, 1992, while I/M programs will begin OBD checks two
years after the regulation has been issued. OBD checks are not
currently a part of EPA's performance standard and no credit has
been assessed for such checks in the MOBILE4.1 model/ such will be
determined after formal issuance of OBD regulations. For the
purpose of this cost-benefit analysis, the impact of OBD has not
been addressed. The impact of OBD will be relatively minor up
until the attainment deadline for serious areas, in November 1999.
EPA will certainly revisit the issue once OBD regulations are
final and as their implementation clarifies the potential of this
strategy in an I/M setting.
8 . 2 On—road Testing,. Interim Provisions
Section 182(c)(3)(B)(i) of the Act requires EPA to establish
a performance standard for enhanced I/M "including on-road
emission testing." The Act does not specify how programs or EPA
are to address the "on-road testing" requirement, and neither is
on-road testing defined within the Act itself. While potentially
a fruitful supplemental testing strategy, it is clear from the
legislative history of the 1990 Amendments that on-road testing
was not viewed as a potential replacement for I/M programs, as has
been suggested by some. Under the section addressing enhanced I/M
programs, the legislative history notes:
On-road emission testing is to be a part of the emission
testing system, but is to be a complement to testing
otherwise required since on-road testing is not intended
to replace such testing. On-road emission testing may
not be practical in every season or for every vehicle,
and is not required. However, it should play some role
in the state program. It is the Committee's intention
that states should take into consideration that the
results of on-road emission testing, when used, have not
been shown to be consistent with Federal emission
testing procedures. [Emphasis added]
In its current proposal for implementing this requirement,
EPA has specified that on-road testing be defined as "the
measurement of HC, CO, NOX, and/or C02 emissions on any road or
roadside in the nonattainment area or the I/M program, " and that
it be required in enhanced programs and an option for basic I/M
areas. Minimally, the on-road testing effort must evaluate the
emission performance of at least 0.5% of the subject fleet each
year. EPA believes that the on-road testing requirement can be
fulfilled by a range of approaches, including, but not limited to:
Draft -112-
2/26/92
-------
remote sensing devices (RSD) , random road-side pull-overs using
tailpipe tests and emission control device checks, or road-side
pull-overs of vehicles with high RSD readings, as well as through
the use of portable analyzers that can be placed on the vehicle
prior to on-road driving.
Of the above approaches, RSD has gained the most public
attention and has generated considerable interest. The objective
of RSD is to remotely measure the concentration of emissions from
vehicles as .they are operated on public roads, and in this aim,
RSD fully meets the definition of an on-road testing strategy. In
its current version, RSD works by focusing a beam, or, in some
cases, multiple beams, of infrared light across the roadway into
an infrared detector. The concentration of certain pollutants in
the exhaust stream are then determined by measuring the amount of
infrared light absorbed at specific wavelengths as it passes
through the exhaust in much the same way that astronomers study
stellar atmospheres by analyzing specific portions of a star's
spectrum. The analysis is tied to a vehicle through the use of a
video camera which records the vehicle's license plate as it
passes through the beam(s).
Given its non-intrusive nature and potentially high
throughput capabilities, RSD warranted further investigation. EPA
has conducted a preliminary analysis of RSD (see Appendix L,
"Identifying Excess Emitters with a Remote Sensing Device: A
Preliminary Analysis") that investigated the comparability of the
results obtained to those in the 2500 rpm/Idle test. EPA found
that, under controlled conditions and using stringent cutpoints,
RSD's performance in measuring CO emissions was comparable to the
2500 rpm/Idle test. Since then, other researchers, such as the
California Air Resources Board (CARB), have found that the
accuracy of the device for measuring HC emissions, while less
accurate than for CO, is within a practical range for roadside
monitoring. For example, CARB researchers recently reported to
the CARB I/M Review Committee16 that the device, under highly
controlled operating conditions, yielded results that compared to
calibrated on-board measurements as follows: the remote sensors
accurately measured CO within ± 5% and HC within ± 15% of the
instrumented vehicle measurements, respectively. EPA, however,
knows of no current RSD methodology for detecting and measuring
NOX emissions, although developmental work is being done in this
area. EPA encourages the states to be innovative in fulfilling
the on-road testing requirement.
There have been and continue to be a number of efforts in the
area of RSD evaluation, including those at the University of
16 D. Lawson, J. Gunderson, "In-Use Emission Study and High Emitter Phase,"
Presentation to I/M Review Committee, Sacramento, California, January 29,
1992.
Draft -113- 2/26/92
-------
Denver, where the first RSD testing strategies were developed.
The bibliography17 of research in this area continues to grow.
Currently, it is difficult for EPA to project a standard
"emission credit" for on-road testing for the purpose of
performance standard modeling. Hence, for the purpose of this
cost-benefit analysis, the impact of on-road testing is not
addressed. Nonetheless, emission reduction credits will be
assessed for on-road testing efforts once additional experience is
gained in the actual use of various on-road testing strategies,
including RSD technology. Under EPA's current proposal, on-road
testing programs required by the Act "shall provide information
about the emission performance of in-use vehicles, by measuring
on-road emissions through the use of remote sensing devices or
roadside pullovers including tailpipe emission testing. The
program shall collect, analyze and report on-road testing data" as
part of the state's annual report to EPA. EPA shall use this
data, in conjunction with data gathered as part of the Agency's
on-going investigation of these testing strategies, to develop
testing protocols and guidance.
17 In addition to the sources referenced in Appendix L, the following works
have contribute to the body of information concerning RSD.
1. D.R. Lawson, P.J. Groblicki, et. al., "Emissions for In-use Motor
Vehicles in Los Angeles: A Pilot Study of Remote Sensing and the Inspection
and Maintenance Program," Journal of the Air Waste Management Association,
40(8): 1096 (1990)
2. R.D. Stevens and S.H. Cadle, "Remote Sensing of Carbon Monoxide
Emissions," Journal of the Air Waste Management Association, 40(1):39
(1990)
3. G.A. Bishop, D.H. Stedman, et. al., "IR Long-Path Photometry, A Remote
Sensing Tool for Automobile Emissions," Analytical Chemistry, 61, 671A-677A
(1989)
4. D.H. Stedman and G.A. Bishop, "Evaluation of a Remote Sensor for Mobile
Sources CO Emissions," Report to the Environmental Protection Agency, EPA-
600-S4-90-032.
5. D.H. Stedman, G.A. Bishop, et. al., On-Road CO Remote Sensing in the Los
Angeles Basin. Final Report on Contract No. A932-189, California Resources
Board, Research Division, Sacramento, 1991.
6. D.H. Stedman and G.A. Bishop. An Analysis of On-Road Remote Sensing as
a Tool for Automobile Emissions Control, ILENR/RE-AQ-90/05, Final Report to
Illinois Department of Energy and Natural Resources, Springfield, IL, 1990.
Draft ' -114- 2/26/92
-------
Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Appendix F
Appendix G
Appendix H
Appendix I
Appendix J
Appendix K
Appendix L
Cost Effectiveness Model Version 4.I Source Code
Listing
Tech4 Model Version 4.1 Source Code Listing
Evaporative Emissions and Running Loss Emission Factor
Derivation
Purge and Pressure Test Effectiveness Figures and
Spreadsheet
Regression Analyses and Scatter Plots for Fuel
Injected 1983 and Later Vehicles
MOBILE4.1 Technology Distribution and Emission Group
Rates and Emission Levels
Exhaust Short Test Accuracy: IM240 vs. Second-chance
2500 rpm/Idle Test
Data Analyses for Appendix G: 1983 and Newer PFI, TBI,
and Carbureted Vehicles
Evaporative System Purge and Pressure Diagrams
Evaporative System Failures and Repairs
MOBILE4.1 Performance Standard Analyses, By Option
Identifying Excess Emitters with a Remote Sensing
Device: A Preliminary Analysis
Appendix M Model Year Failure Rates by Test Type
Draft
-115-
2/26/92
-------
APPENDIX A
COST EFFECTIVENESS MODEL VERSION 4.1 SOURCE CODE LISTING
-------
Appendix A
c
C #$*$*$*$* CEM4.1 #$#$#$#$#
C
C The Coat-Effectiveness Version of MOBILE4.1
C
C
C
C Re-Revise repair coat adjustment for recurring IM240. Dec 10, 1991
C Fix problem in PCLEFT Tech 1 & 2 credits.
C
C OEM version 4.IE Dec 4, 1991
C Trial change to CPFUEL, adding IMFLAG to coverage checks.
C Trial init of CSIZE in RESET.
C Changed Frg/Pra repair EF benefit from 7.4% to 5.9%
C Install new PPEFF array in FAIL routine for revised
C detection rates.
C Change block FAIL to FAILRT to avoid confusion with
C Function FAIL().
C CEM version now based on November 1991 MOBILE4.1,
C
C CEM version 4.ID November 19, 1991
C Revise repair cost adjustment for recurring IM240
C from .0199/10K miles to .0373 constant.
C Block Data from .0199 to .0373, and remove corresponding FE
C 1.87 adjustment.
C Still needs revisions for November M4.1:
C multiple runs and speed correction factors, etc.
C
C CEM version 4.1C November 6, 1991
C Re-installed 1.87 FE benefit adjustment factor for non-111240
C (as it has been for IM240)
C
C CEM version 4.IB October 30, 1991
C Revised FAIL to skip over program reductions when doing
C recurring case for repair costs.
C Still has excess code for multiple evaluation years
C but can not be used.
C October 18, 1991
C
C CEM Version 4.1A, 10/17/91, is the last version prior to
C cutting out the excess evaluation year capability (JCYDX > 2).
C It includes much unused code and variables, but is fully M4.1
C compatible plus a couple little fixes (HDV handling in FAIL routine).
C It includes input file list capability and 'interactive' I/O file
C specification, but needs revision for November release of M4.1.
C
C
C
C ****** Release MOBILE4.1 ****** November 4, 1991
C
C
C Release MOBILE4.1 is derived from 102 (June 12, 1990) MOBILE4.
C
C See Dnn.T-NOTES for a chronologically ordered summary of the coding changes
C performed up through the current In-House Version. See Dnn.T-UGBASE for the
C In-House MOBILE4 User Guide Supplement for Version Dnn. All In-House
C Version files have the Dnn prefix. The backup tape is *M4.1*, mounted by
C Dnn.C-TAPE. All files are on ST99.
C
C Version 00 was opened 02/28/91 and closed 03/04/91.
C Version 01 was opened 03/04/91 and closed 04/09/91.
C Version 02 was opened 04/11/91 and closed 04/30/91.
C Version 03 was opened 05/01/91 and closed 05/15/91.
C Version 04 was opened 05/16/91 and closed 05/24/91.
C Version 05 was opened 05/28/91 and closed 06/05/91.
C Version 06 was opened 06/06/91 and closed 06/21/91.
C Version 07 was opened 06/26/91 and closed 07/29/91.
C Version 07 was rereleased as Corrected M41 on 08/28/91.
C Version 07 was rereleased as FINAL M41 on 11/04/91.
C
C
C MOBILE4 is the fourth version of the FORTRAN program implementation of the
C Mobile Source Emission Model. It updates and replaces MOBILES. The model
C estimates HC, CO and NOx exhaust and HC evaporative emission factors by
C motor vehicles. For information on using the program and on the
C differences between MOBILES and MOBILE4, consult the User's Guide to
C MOBILES, EPA 460/3-84-002, and the User's Guide to MOBILE4,
C EPA-AA-TEB-89-01. Otherwise questions should be directed to:
C
-------
Appendix A
C Office of Mobile Sources
C Emission Control Technology Division
C Test and Evaluation Branch
C 2565 Plymouth Road
C Ann Arbor, Mich. 48105
C Telephones 313-668-4462
C FTS 374-8462
C
C
C Source Code Documentation
c
C
C The source code itself is commented to assist the users who require a
C more detailed understanding of the program than provided in the Guides.
C This section provides the following information:
C
C 1) Execution summary
C 2) Source code structure
C 3) Source code comments structure
C a) Subroutines and functions header format
C b) BLOCK DATA subprograms header format
C c) In-line comments
C 4) Subscript dictionary
C 5) Subroutine / function parameter dictionary
C
C
C 1) Execution summary
C
C To run the program, attach the input file to logical I/O Unit 5 and the
C output file to Logical I/O Unit 6. If supplying optional alternate I/M
C credits, attach the credits file to Logical I/O Unit 4. Refer to the
C User's Guide to MOBILE4 for more detail on setting up runs and check
C your operating system documentation for the system commands for compiling
C and executing the program. There is discussion of running the program on
C the IBM PC AT under DOS in the User's Guide.
C
C For information on the source code itself, refer to the Programmer's Guide
C to MOBILE4, EPA-AA-TEB-89-01.
C
C
C 2) Source code structure
C
C This section describes the layout of the code plus the numbering and
C naming conventions.
C
C The source code for the subprograms is grouped as follows:
C
C program header
C MAIM
C MOBILE subroutine
C QUITER subroutine
C input subroutines
C computation subroutines S functions
C pointer functions
C output subroutines
C BLOCK DATA subprograms
C
C Within each grouping and nesting level, the routines usually appear in
C calling order.
C
C The User's Guide listing of the code has each subprogram starting at an
C integral multiple of 1000, simplifying the construction of the Table
C of Contents. That listing also has the comments in mixed case, to
C improve readability. The MOBILE4 tape being mailed out to the user
C community has this mixed case version in File 1. An upper case only
C version, with hyphens replacing underscores and blanks replacing plus
C carriage controls, is in File 2, for use by installations not able to
C process these features. For example, EPCDIC PN character set is upper
C case only.
C
C Statement number ranges are defined for usage as follows:
C
C 1 - 9 - not used
C 10 - 98 - branching
C 99 - branch to last RETURN / RETURN1 / STOP before END
C 100 - 199 - READ formats
C 200 - 299 - WRITE prompt and output formats
C 300 - 499 - WRITE error and warning messages
C 500+ - not used
C
-------
Appendix A
C Statement numbers are local. For example, within each subprogram,
C branch numbers start at 10, READ formats start at 100, etc.
C
C Given FORTRAN'S 6 character maximum on subprogram name lengths, useful
C naming conventions are hard to come by. There is also the limitation
C posed by not wanting to change arbritrarily all the subprogram names of
C the predecessor program, MOBILE3. Hence, the only conventions are:
C
C a) Input driver sections all have the suffix 'SEC'.
C b) Pointer functions all begin with 'I' (to get default INTEGER typing).
C c) Pointer functions all contain/end with 'PT' or 'PTR'.
C d) Output subroutines all have the prefix 'OUT'.
C
C Otherwise, new subprogram names were picked to indicate the purpose of
C the code.
C
C Note: some features during the course of the model development were
C subsequently removed or "turned off". The latter case would occur when
C the prospect of restoring the feature for in-house MOBILE4 or for MOBILES
C seemed likely. Commented out code to this end is done with a 'CC1 .
C
C
C 3) Source code comments structure
C
C Each subprogram contains a documentation header, set up in a standard
C format as described below. All variables and arrays used in the code
C are defined in the comments. The subroutine / function headers define
C local parameters and local variables / arrays. The BLOCK DATA headers
C describe the COMMON storage areas. Array subscripts and subprogram
C parameters are specified in dictionaries following this section. The
C intent is to allow the user to scan the source code on-line for
C definitions, rather than have to leaf through hard-copy documentation
C such as the Proramtiter'a Guide to MOBILE4.
C
C a) Subroutines and functions header format
C
C
C C
C C
C C
C C Called by
C C
C C Calls
C C
C C Input on call:
C C
C C parameter list:
C C common block(s):
C C //
C C
C C .
C C
C C Output on return:
C C
C C parameter list:
C C common block(s):
C C //
C C .
C C .
C C
C C Local array subscripts:
C c
C C (<#>,...) - (, ...)
C C .
C C .
C C
C C Local variable / array dictionary.
C C
C C Name Type Description
C C
C C
C C
C C
C C
C C Notes:
C C
C C
C C
C C
C
C
-------
Appendix A
C Notes on format:
C
C A local variable is one not passed in via the parameter list
C or a common block. It can be used in the argument list of a
C subprogram call made by the defining subprogram.
C
C Also, if a section of the header is not needed, its format is
C omitted. For example, if no subprograms are called, then the
C 'C Calls ' line is not included. "Notes" are always included.
C
C Finally, 'Output on return1 includes any variable / array whose
C contents can be modified by the subprogram. This usually will not
C include data that can be changed by subprograms called by the
C subprogram, aince the comment headers on those subprograms will
C indicate what is being passed back up.
C
C b) BLOCK DATA subprogram header format
C
C BLOCK DATA
C C
C C BLOCK DATA subprogram <#>
C C
C C
C C
C C Common block array subscripts:
C C
C C (<#>,...) - (,...)
C C .
C C .
C C
C C Common block dictionary:
C C
C C Name Type Description
C C
C C //:
C C
C C
C C
C C
C C Local array subscripts:
C C
C C (<#>,...) - (,...)
C C .
C C .
C C
C C Local array dictionary:
C C
C C Name Type Description
C C
C C
C C
C C
C C
C COMMON // ...
C
C
C
C C
C DIMENSION (or type) (subscript(s))
C
C
C C
C EQUIVALENCE (,)
C '.
C
C C
C C //: 'a labels, purpose or data
c c or on the structure of the initializing DATA
C C statement(a)>
C C
C DATA /...
C
C
C
C .../
C C
C C //: ...
C
C
-------
Appendix A
c
c END
c
C Notes on format:
C
C Local arrays are defined and then equivalenced to common block arrays
C that are too large be intialized in 1 DATA statement. Too large -
0 would require greater than 19 continuation lines, the FORTRAN limit.
C
C The local arrays dictionary is often compressed by providing a
C generic definition of the conventions used to construct and name the
C set of local arrays to be equivalenced to a given common array.
C
C "//:" prefix is included only if a large number of common
C blocks are being initialized in a single BLOCK DATA (see BD 23).
C
C
C c) In-line statements
C
C In-line comments are intended to describe the algorithm being used
C by the code. Where MOBILE4 is identical to MOBILES in operation,
C the MOBILES comments have usually been left intact. Where the
C method has been updated or the whole section / subprogram is new,
C the in-line comments describe the MOBILE4 methodology.
C
C Particular attention is paid to the use of indices and pointers.
C Some of the ambiguities of MOBILES, such as the use of IDX vs JDX,
C have been cleared up. The usage of new keys and pointers is specified
C in detail where they occur.
C occur.
C
C
C 4) Subscript dictionary
C
C Subscript variables used to index arrays in MOBILE4 are globally
C defined, not in the sense that the values are available to all
C subprograms, but rather that the same label is used for the same
C subscript function throughout the code. IF is the 3 pollutants index
C for all of MOBILE4, although its instances are usually as a local
C variable. As array subscripts, all these variables must be of type
C INTEGER. The default INTEGER typing syntax rule is employed by having
C all subscripts begin with a letter from the range I through N.
C
C Name Range Description
C IAD 1-5 index into ID dimension of ATR for ADJFAC equation
C
C IADJ 1-2 index into CSIZE adjustment vector ADJ7AC
C
C IAY 1-2 anti-tampering program start year flag:
C 1 - in year range up thru start year
C 2 - in year range after start year
C
C IB 1-3 bag number:
C 1 - cold start
C 2 - hot stabilized
C 3 - hot start
C
C IBER 1-3 terms of user supplied new emission rate equation:
C 1 - zero mile level (intercept)
C 2 - deterioration rate (slope)
C 3 - deterioration rate (slope), LDGV 81+ HC/CO 50K+
C
C 1C 1-7 overlap effects categories computed direct from base
C tamper rates; each case is the set intersection of
C the indicated disablement types:
C 1 - AorE * CATS
C 2 - AorE * NCKS
C 3 - AorE * TNKS
C 4 - AIRS * CATS * NCKS
C 5 - AIRS * CATS * TNKS
C 6 - CATS * NCKS
C 7 - CATS * TNKS
C AorE - AIRS if HC or CO
C » EGRS if NOx
C
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
8-12 no overlap effects categories computed from the 7
overlap cases:
8 - AIRS
9 - CATS
10 - NCKS
11 - INKS
12 -
ICAT 1-3 catalyst technology configurations:
1 - ox catalyst / no catalyst
2-3 way catalyst
3-3 way + ox catalyst
ICH 1-20 character string array position index:
1 - 1st 4/8 characters (CHARACTER*4/8)
2 - 2nd 4/8 characters
also serves to subscript any other CHARACTER type array.
ICOEF2 1-6 coefficient number of 2nd degree polynomial
(Oth (intercept) to 2nd decree term coefficients)
ZCOEF5 1-6 coefficient number of 5th degree polynomial
(Oth (intercept) to 5th decree term coefficients)
ICOL 1-3 index into EFF's column dimension (BD 24 defines abbrs):
1 - PREV CAT ; 2 - PREV ; 3 - SOBS
ICOLD 1-2 index into cold CO standard case of MYCOL fi CCSTD
ICR 1-2 Replace I/M credits inclusion vector NUDATA's array
identifier subscript:
1 - (CR12HC £ CR12CO) - Technologies 1-2
2 - CRED4V - Technologies 4+ LDGV
ICS 1-NT secondary index: selects pointers to category
sizes to be summed up to get SUB (in C8TO11) and
DIV (in A8TO11)
IC11 1-11 positive tampering effects categories 1-11; 12 is the
no effects case.
IC7 1-7 overlap tampering effects categories 1-7
ID 1-9 disablement types (1-5,9 for bag, 6-8 for evap):
1 *• AIRS " air pump disabled
2 - CATS - catalyst removed
3 - NCKS - misfueled - filler neck
4 - INKS - miafueled - tank (all but filler neck)
5 - EGRS - E6R disabled
6 — EVAP ™ evaporative system tampered - cannister only
7 - PCVS - PCV disabled
8 - CAPS - evaporative system tampered - cannister and
gas (fuel inlet) cap
9 - MISF - misfueled - all sources - NCKS + TNKS
IDBAG 1-5 bag exhaust disablement types (see ID)
IDPB 1-3 misfueling related disablement types:
1 - NCKS - misfueled - filler neck
2 - TNKS - misfueled - tank (all but filler neck)
3 - MISF - miafueled - all sources - NCKS + TNKS
IDU 1-5 index for Diurnal (DU) indicating full, 8AM-11AM,
10AM-3PM, 8AM-2PM, and multiple
IDUSER 1-8 ID cases £ order of cases presented to user in TAMFLG
(TAMZML £ TAMDR) S ATPFLG (DISTYP) options. IDUSER
differs as follows: 4 = MISF. GETTAM derives TNKS =
MISF - NCKS for TAMDR £ TAMZML and stores the read and
computed values in ID order. ATPAER does not need TNKS
£ uses DISTYP in the read in IDUSER order.
IDX 1-25 model year (my) window index:
1 = 25+ my age group cell «> tamper effect is
always 0.0 => this case skipped in TAMPER (only)
2 - 24th my cell
3 - 23rd my cell
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
IEG
IEK
IELST
IEQU
IERNEW
IEVEQ
IEVF
IEVX
IE1ST
IFAC
IFDS
IFG
IFL
IFUEL
IG
IGCSF
IGD
IGER
IGIDL
1-7
1-3
4-7
1-20
1-20
1-9
1-4
1-32
1-3
1-100
1-3
1-13
1-2
1-3
1-20
1-3
1-3
1-20
1-15
24 = 2nd my cell
25 - 1st my cell - calendar year (ICY)
tampering effects group:
1 - air pump disabled, air pump only equipped
2 =» air pump disabled, air pump/catalyst equipped
3 » catalyst removed
4 = miafueled catalyst
5 - EGR disabled
6 - EGR disabled / catalyst removed
7 - EGR disabled / misfueled
catalyst emission effects groups: IEK - IEG - 1
last non-zero tampering effects group size, given IFG
pointer to idle equation to be zeroed out due to entry
of new exhaust ef equation by user
pointer to location in default ef rates & model years
arrays where new ef equation's parameters are to be
stored (IERNEW m 0 -> no space for new equation)
HS and DU equation term coefficient - see definitions
for HSEQ in BD 09 and DUEQ in BD 11
index for gram evap rates EVF, GREVP and VGREVP:
1-hot soak
2=in-use diurnal
3— multiple diurnal
4—crankcaae
model year (my) window index for evaporative data input.
i.e. to the lEVXth myg 'a rate(s) for a given NO set
first non-zero tampering effects group size, given IFG
index into new (user supplied) emission rates arrays;
the user can enter up to 100 replacement cases .
fuel delivery system type:
1 » carbureted 2 — fuel injected OR
1 - CARbureted 2 = TBI 3 = FFI
flag names for prompts (listed by common block used) :
/FIAGS1/ /FLAGS2/ /FLAGS3/ /FLAGS4/
1 - TAMFIiG 4 - MYMRFG 8 - ATPFLG 12 - PRTFLG
2 - SPDFLG 5 - NEWFLG 9 - TEMFLG 13 - IDLFLG
3 - VMFLAG 6 - IMFLAG 10 - RLFLAG 14 - NMHFLG
7 - ALHFLG 11 - OUTFMT 15 - HCFLAG
The 14th flag (or 0 - PROMPT) is read without a prompt.
myg bounds: 1 - First year 2 - Last year
fuel type: 1 - Gasoline, 2 - Ether Blend, 3 - Alcohol
Blend
model year group (my?) pointer:
- null string,C,CAT,D,E,F,L,LM,S,T,U,W,l,2,3,4
IG links a data array to its corresponding myg index
array. These data arrays all have my as a subscript
plus have cases of contiguous subsets of years with
the same data point. Core space is then saved by
storing each point only once per my subset and then
constructing an index array to enable the program to
find the location IG of the data point for a given
my.
model year group index in the array CSF1ST:
1 - pre-1981 model years
2 = 1981-1983 model years
3 - 1984+ model years
MYG index into 3rd dimension of TAMZML & TAMDR
1 - pre-1981 2 - 1981-83 3 - 1984+ for LDGV/T
1 - 1981+ for HDGV
myg pointer into basic emission rate arrays from IERPTR
myg pointer into idle emission rate arrays from IDLPTR
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
IG50 l-i:
IGTL 1-4
IH 1-3
IHG 1-2
IHDSAL 1-2
IBRD 1-6
ILU 1-2
IM 1-2
IMDL 1-2
INO 1-2
IF 1-3
IFAR 1-5
IPARTL 1-3
IPG 1-2
IPIM 1-2
IP1 1-3
IP2 1-3
IF50 1-2
IQG 1-6
myg pointer into the 50K+ deterioration rate arrays
ERB50K S ERU50K (the arrays differ in lookup algorithm)
running loss HC: emission rate MYG pointer
non-exhaust HC emission case:
1 =• hot soak evaporative
2 = diurnal evaporative
3 ** crankcaae
non-exhaust HC technology / tampering case:
1 - evaporative
2 » crankcase
HDGV sales fraction case
1 - HDGV weighing < 14000 Ibs
2 - HDGV weighing >- 14000 Ibs
evaporative temperature types:
1 « hot soak
2 - running loss
3 - diurnal, 8 AM
4 - diurnal, 10-11 AM
5 - diurnal, 2 FM
6 — resting loss
bounds of a value range (ex.: tags of nonzero
category sizes):
1 ** lower bound
2 - upper bound
I/M program flag:
1 - I/M program not in effect for MY x IV x ICY case
1 - I/M program is in effect for MY x XV x ICY case
coefficients for trips per day model equation:
1 - intercept
2 - slope
index into NOYES character string vector:
1 - No 2 - Yes
pollutants:
1 = HC = hydrocarbon
2 «• CO " carbon monoxide
3 - NOx — oxides of nitrogen
parameters identifying where and when to apply user
supplied new emission rates:
1 ™ region
2 «• vehicle type 4 « first my of range covered
3 - pollutant 5 - last my of range covered
index for Fartial Diurnal (PARD) indicating
8AM-11AM, 10AM-3FM and 8AM-2PM
pollutants grouped for tampering effects & SK2 scfs:
1 - HC and CO
2 - NOX
pollutants affected by I/M programs:
1 - HC
2 - CO
lower bound of output IP loop (first pollutant to
have results printed)
upper bound of output IF loop (last pollutant to
have results printed)
pollutant types affected by 1981+ LDGV 50K+ change in
deterioration rate
emission control equipment group; 2 more groups
computed from the stored 6 groups data:
1 *• air pump only
2 = air pump / catalyst
3 =• catalyst only
4 = EGR only
5 = EGR / 3-way catalyst
6 - no EGR / 3-way catalyst
-------
Appendix A
c
C IR 1-2 region code:
C IR haa same value range aa IREJN, but is not necessarily
C the scenario field's value
C 1 - low altitude
C 2 - high altitude
C
C IREC 1-999 input record number, on a given READ
C
C IRON 1-15 index into EFF's row dimension - the 8 ID cases split
C into 15 subcases (see BD 24 definition of EFF)
C
C IRT 1-4 temperature cases in RVP cf calculations, the values
C referenced depending on the (sub)array. See /RVFEX1/.
C
C IRQ1 1-3 RVP cf equation parameters set 1 (RVP > 9.0 adj.):
C RADJCF-(bl+b2*RVP)/b3
C
C IRQ2 1-2 RVP cf equation parameters set 2 (T interpolation)
C 1 - initial temperature (T range Ib)
C 2 - temperature difference (interval length)
G
C IRQ3 1-3 RVP & high T combined cf formula coefficients:
C TCF (IB)-EXP (al* (RVP-9) +a2* (T-75)+a3* (RVP-9) * (T-75))
C
C IRVP 1-2/4 types of non-certification RVP: 1 » Base, 2 - In-use
C 3 =" Uncontrolled refueling loss
C 4 - Hot soak (weathered)
C
C IS 1-4 overlap tampering effects category sizes 8-11
C ( IS - 1C - 7 ); used in A8TO11.
C
C ISC 1-5 underscore string repetition counter - for example,
C ISC - 3 -> repeat the string 3 times
C
C ISP 1-21 my group pointer into speed correction equation array;
C found by ISPPTR; indexes SKI & SCUNA1.
C
C IS2a 1-2/3 number of coefficients in a new scf equation:
C IS2C - 2 (A/X+B); IS2N - 3 (A+B*X+C*X**2)
C
C IT 1-9 pointer to temperature correction equation
C coefficients; found by ITCPTR; indexes TTA/4/7COF
C arrays.
C
C ITECH 1-15 emission control technology class wrt I/M impact;
C f(MY,IVLD) aa determined by IMPTR; indexes I/M credits
C arrays CR12HC/CO £ CRD4VA/B. ITECHA/B/C are in BD8S34.
C
C ITEM 1-2 NAMTEM index: 1 - 'mini', 2 - 'maxi' prefixes
C
C ITER 1-4 unspecified string iterations counter; same usage as
C ISC, only with any string
C
C ITL 1-13 running loss HC: defined trip lengths for 3 cycle types:
C 1-6 - HTCC 7-12 - IA-4 13 - HFET
C
C ITLR 1-4 running loss HC: RVPs at which measured rl HC available
C
C ITLT 1-4 running loss HC: temperatures at which measured rl HC
C emission rates are available
C
C ITY 1-3 type tampering rates wrt I/MxIGD table heading substring
C pointer
C
C IUDI 1-5 index into Uncontrolled Diurnal rate UDI:
C 1 - standard FTP conditions
C 2 - in-use conditions
C 3 - 8AM-11AM conditions
C 4 - 10AM-3PM conditions
C 5 m 8AM-2PM conditions
C
C IV 1-8 vehicle types evaluated by MOBILE4:
C 1 - LDGV - light duty (Id) gasoline fueled vehicle
C 2 - LDGT1 - Id gasoline fueled truck, <- 6000 Iba.
C 3 = LD6T2 » Id gasoline fueled truck, 6001-8500 Ibs.
C 4 =• HDGV => heavy duty gasoline fueled vehicle
C 5 » LDDV - light duty diesel fueled vehicle
C 6 - LDDT - light duty diesel fueled truck
C 7 - HDDV - heavy duty diesel fueled vehicle
C 8 - MC «• motorcycle
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
1-6
vehicle type subgroups; Index correction factor arrays;
- LM, TEMP, A, B or LDG, defined as:
IVEFF
IVF1
IVF2
IVF3
IVIM
IVHD
IVP
IVREV
IVTAM
TV9
12ND
JCS
JDX
1-2
1-6
1-9
1-10
1-4
1-2
1-3
1-4
1-4
1-9
1-4
1-7
25-1
IVLM
1 - LDGV
2 - LDGT1
3 - LDGT2
4 - LDDV
5 - LDDT
o *• MC
indexes :
MYGSP1
BFRCOF
IVTEMP
1
2
3
4
5
=
m
at
9
=.
LDGV
LDGT1
LDGT2
HDGV
MC
MYGTCF
IVA
1 - LDGV
2 - LDGT1
3 - LDGT2
4 - MC
SCADJ1
ISP GBP
IVB
1
2
3
4
- HDGV
- LDDV
- LDDT
- HDDV
GPBSCO
IVLDG
1
2
3
- LDGV
- LDGT1
= LDGT2
TTFC
JG
1-20
ALHRET
See subroutines BIGCFX £ BIGALH and function ISPPTR.
vehicle class groupings for Stage II vrs percent
efficency in reducing HC refueling losses:
1 - LDGV, LDGT1 £ LDGT2 2 = HDGV
vehicle class types for MOBILE1 numeric output
format (OUTFMT - 1):
1 - LDV 3 - LDT2 5 • HDD
2 - LDT1 4 - HDG 6 - MC
vehicle class types for MOBILES numeric output
format (OUTFMT - 2):
1 - LDGV 4 - LDGT 7 - LDDT
2 - LDGT1 5 - HDGV 8 - HDDV
3 - LDGT2 6 - LDDV 9 - MC
vehicle class types for MOBILE4 descriptive output
format (OUTFMT - 3): same as IVF2 plus 10 = All Vehicles
vehicle class types that can be included in MOBILE4
inspection / maintenance (I/M) programs:
1 - LDGV 2 - LDGT1 3 = LDGT2 4 - HDGV
vehicle class types for heavy duty only (used for
units conversion of user entered hd zml s dr):
1 - HDGV 2 - HDDV
vehicle types whose ef's can be modified by correction
factors for extra load, trailer towing S/or CO offset:
1 - LDGV 2 - LDGT1 3 - LDGT2
vehicle types covered by the revised idle EF algorithm:
1 - LDGV 2 - LDGT1 3 - LDGT2 4 - HDGV (HCSCO)
IVI - IVREV, except 3 & 4 -> 3 (HDGV uses LDGT2 data)
vehicle types impacted by tampering:
1 - LDGV 2 - LDGT1 3 - LDGT2 4 - HDGV
vehicle type categories for output arrays EFEVAP,
EFIDLE, S EFFTP- First 8 are identical to TV'a.
9 - LDGT
• LDGT1 + LDGT2, each weighted by its vmt share
secondary index (pointer into a pointer vector)
pointer to category size to be summed into DIV
(in A8TO11) or (* TGSUSE) into EGS (in SETEGS)
model year window index - age of vehicle relative to
ICY + 1. In detail:
25 » 25+ my age group cell •> tamper effect is
always 0.0 -> this case skipped in TAMPER (only)
24 - 24th my cell
23 - 23rd my cell
2 » 2nd my cell
1 - 1st my cell = calendar year (ICY)
model year group pointer - IG or 21 - IG, depending on
whether looking for match or unused cell, respectively,
in MYGERB. Also used (as is KG) to distinguish (from IG)
myg dimension size differences in BD comments.
10
-------
Appendix A
c
c
c
G
C
C
C
C
C
C
C
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
JOU
JPGD
JSTD
1-3
1,5
1-6
JO
JV
KSTD
KDX
LAY
LIM
1-9
1-2
1-2
1-21
1-2
1-2
output unit key:
call I is assigned IOUREP's value
cell 2 is assigned lOUERR's value
cell 3 is assigned lOUASK's value
pointer to base tamper rate to be used in
computing CSIZE(1-3,IAY). Depends on IPG.
emissions standard model year group for gas fueled IV.
The group key is the same as ZSTD, but JSTD is set
internally, instead of being passed in via EVMAIN.
(The ISTD definition is in the Parameter Dictionary.)
input / output (i/o) device unit numbers
vapor pressure calculation loop S vectors index (CALUDI)
mapping of JSTD used to index R2ALT:
KSTD - 1, unless JSTD = 4 -> KSTD = 2
model year window index - ICY - 1979, unless result is
< 1 (-> - 1) or > 21 ( -> - 21); used by HDDMYM
same as IAY, but set for DISCAL call only.
- IM, except - 1 when there is an I/M kink, but it is
after LDXSY & the PREV segment is being evaluated:
ZM IMKINK
LKINK
LIM
Comments
1
2
2
2
1
1
2
2
1
1
1
2
I
2
1
2
P S S: no I/M
P S Si all I/M
P: no I/M; S: kinked I/M
P: kinked I/M; S: all I/M
LIMIT
LINES
LJDX
LKINK
NCKIAY
1-2
1-9
25-1
1-2
1-3
index for range check on a variable's value assignment:
1 " lower bound 2 •» upper bound
number of lines in user entered data set title
model year relative to LAPSY - JDX - LDXSY -I- 1;
used with CUMMIL to get mileage of "previous" segment
same as IMKINK, except IMDXSY must be within (JDX,LDXSY)
index into ECK - ICOL index
5) Subroutine / function parameter dictionary
Variables /arrays occurring in 1 or more of the subprogram parameter
lists are defined in this section, unless the variable is a subscript,
in which case look it up in 4) 's subscript dictionary. The motivation
for defining all the parameter lists in one place is to have only one
instance of a definition in the code plus to encourage only one use
for a mnemonic in the code plus the combined lists contain only 18
names after subscripts have been removed. Whether scanning on-line or
off a listing, the lookup will be quick.
Parameter array subscripts
CSF0SE(7) - CSFOSE ( IC7 )
CSF1ST(7,2,3) - CSF1ST ( IC7, IPG, IGCSF )
S2LEFT(4) - S2LEFT ( IVTAM )
Name
Type Description
CSFOSE R category size factors used * AIR to calculate CSIZE
CSF1ST R category size factor for 1st calculation of CSIZE
FLGERR I number of flag values found to be out of range
IBEFSW 0-2 switch identifying routine calling BEF S result returned
0 =• BEFIDL => always return uncorrected basic non-idle ef
1 - BEFIDL m return uncorrected basic ef
2 « EFCALX or HCCALX - return basic ef corrected for
operating mode, temperature, RVP (if high temp),
I/M, oxy fuel and tampering
ICASE I flags the my groups array to be searched.
ICY I calendar year for which emission factors are calculated
INERR I cumulative count of data entry errors
IOUOUT I unit number on which title is printed
11
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
IOUSUM
IVALUE
IY
IY1
IY2
MESSAG
MY
HERR
NREC
RVALUE
S2LEFT
VMLDGT
VMTAGE
Evap Study
Dame
EVRATE
ISTD
ITRSTL
IVGAS
IVTL
KEYEQ
KEYEVP
RIVAL
I sum of acceptable changed i/o unit numbers
I entered integer value causing error / warning message
I year to test and adjust to 4-digit year
I lower 4-digit bound for IY
X upper 4-digit bound for IY
I message code sent by calling subprogram
I model year
I number of error if IY is out of bounds
X number of records to be skipped on next READ
R entered real value causing error / warning message
R See local dictionary definition in REFUEL header.
R sum of vmt for LDGT1 and LDGT2 (- X23 in MOBILE2)
R vehicle cumulative mileage
XI Parameter List Dictionary for CCEVRT block of routines:
Range Description
R dependent variable value from evaporative component
equation (s) calculation:
KEYEVP EVRATE
1 — > hot soak rate
2 -> diurnal rate
1-7 emissions standard model year group for gas fueled IV:
IVGAS value myg
1 1 pre-1971
2 1971
3 1972-1977
4 1978-1980
5 1981+ carbureted (KEYEQ-1 or 2)
6 1981+ TBI KEYEQ-1 or 2
7 1981+ PFI KEYEQ-1 or 2
2 1 pre-1971
2 1971
3 1972-1977
4 1978-1980
5 1981+ carbureted (KEYEQ-1 or 2)
6 1981+ TBI KEYEQ-1 or 2
7 1981+ PFI KEYEQ-1 or 2
3 1 pre-1979
2 1979-1980
3 1981+ carbureted (KEYEQ-1 or 2)
4 1981+ TBI KEYEQ-1 or 2
5 1981+ PFI KEYEQ-1 or 2
4 1 pre-1985
2 1985+ (KEYEQ-1 or 2)
5 1 pre-1978
2 1978-1979
3 1980-81
4 1982-84
5 1985+
1-2 technology type for resting losses calculation
1— fuel injected with open bottom canister
2-fuel injected with closed canister
(currently carbureted-fuel injected, open)
1-5 gasoline fueled vehicle types (— IVTEMP) :
1 - LDGV 2 - LDGT1 3 - LDGT2 4 - HDGV 5 - MC
IVGAS - IV, except MC (5 vs 8); IVEVAP is an alt name.
1-4 running loss HC: vehicle classes with rl HC emissions
1-3 HS/DU equation group selection key:
1 — passing
2 — failed purge
3 — failed pressure
1-2 Evaporative component calculation selection key
1 = hot soak 2 - diurnal
R independent variable value in evaporative component
equation (a) calculation:
KEYEVP RIVAL
1 -> hot soak -> fuel RVP
2 — > diurnal — > diurnal Index
***********************************************************
12
-------
Appendix A
c
C 1C, IDX, IFDS, IG, IM, IP, IP1, IP2, IR, IV and JSTD are parameters
C defined in the subscript dictionary.
C
C Notes:
C
C **** HOT IN USE **** - the subroutine/function is not called by the current
C version of MOBILE4. Release MOBILE4 (final version), for example, does not
C use WTLROL or OUTTAM.
C
C
13
-------
Appendix A
c
c MAIN
c
C Dummy main program for MOBILE4.1.
C Required to allow proper RUN entry to MOBILE subroutine.
C
C Called by: none
C
C Calls MOBILE
C
C The following variables are passed in argument lists or common blocks
C
C argument lists: none
C common blocks:
C /SAVE01/ JCALL
C
C Output on return:
C
C parameter list: none
C
C Local variable dictionary: none
C
C Reset JCALL (initialized to 1 by BD12) to indicate stand-alone
C
C COMMON /SAVE01/ JCALL,VA001(15,3,8,2),VA002(15,3,8,2)
C JCALL-0
C CALL MOBILE(INERR, *99)
C STOP
C 99 STOP 1
C END
C
C *****************************************************************
C
14
-------
Appendix A
SUBROUTINE MOBILE(INERR, *)
C
C Called by user's program (example: Air Quality Analysis System).
C
C Calls BDSAVE, CONSEC, EFCALX, ONESEC, OUTDT5, OUTNEW, OUTPUT,
C PARSEC, QUITER, REINIT, RESTOR and REGMOD.
C
C The following variables are passed in argument lists or common blocks and
C in MOBILE:
C
C argument lists: ICY,INERR
C common blocks:
C /FLAQS3/ OUTFMT
C /IOUCOM/ IOUERR
C /REGION/ IREJN
C /SAVE01/ JCALL
C
C Output on return:
C
C parameter list: INERR
C
C Local variable dictionary:
C
C Name Type Description
C ICYOLD I calendar year from previous pass of scenario loop
C IROLD I region from previous pass of scenario loop
C ISAV93 I flag for 1993 disclaimer message (QUITER #130)
C 0 « message has not yet been given
C 1 — message has been given: not again!
C LEAST1 I flag: 0 - OUTPUT has not yet been called.
C 1 - OUTPUT has been called at least 1 time.
C
C Notes:
C
C MAIN was converted to subroutine MOBILE in Version 01 of development MOBILE4.1
C The /BYMYCd/ CBs were added for Version 02.
C
C
CHARACTER*26 BYVTN
CHARACTER*!? YFM
CHARACTER* 8 NAMFLG, NAMMMR
CHARACTER* 8 USNAME
CHARACTER*? PN,U7
CHARACTER* 8 NAMRVP
CHARACTER* 4 NAMTEM
CHARACTER* 4 NO YES, COMMA, PERIOD
CHARACTER* 4 IMNAME, PRO JID , SCNAME
CHARACTER*3 PSEC
CHARACTER*! COLON, VB
INTEGER PROMPT, TAMFLG, SPDFLG, VMFLAG, OXYFLG, DSFLAG
INTEGER ALHFLG
INTEGER ATPFLG, TPDFLG, RLFLAG, TEMFLG, OUTFMT, OUTPRF
INTEGER PRTFLG, HCFLAG, COLDFG
INTEGER ATPPGM,ATPFQT,DISTYP
INTEGER ADDCSW
REAL JULMYR,JANMYR
REAL LHDVM, MHDVM, HHDVM, LHDRG, MHDRG, HHDRG
C
COMMON /ALURAC/ ACEQIP(8,3,2),ACCF(2,3,2,3,2)
COMMON /ALUHIN/ AC,XLOAD (3) , TRAILR(3) ,ABSHUM, DB,HB
COMMON /ALUHLT/ XLCF(3,6,3,2),TMCF(3,6,3,2)
COMMON /ALUOUT/ ALHRET(25,4,3)
COMMON /ALUPAC/ MYGACE (8, 3, 2) ,MYGACC (2, 3, 2)
COMMON /ALUPLT/ MYGXLCJ6,3,2),MYGTWC(6,3,2)
COMMON /ATPAR1/ LAPSY,LAPlST,LAPLST,LVTFLG(4)
COMMON /ATPAR2/ ATPPGM,ATPFQT, CRATP, DISTYP (8)
COMMON /ATPAR3/ AIR(2) ,CAT (2) ,ECK (3) , INK (2) , EGR(2)
COMMON /ATPAR4/ CAN(2),PCV(2),CAP(2),EFF(2, 3,15)
COMMON /BASEQl/ ERBZML(23,3,8,2),ERBDR<23,3,8,2),ERB50K(12,2, 3, 2)
COMMON /BASEQ2 / MYGERB (23,3,8,2), MAXERB
COMMON /BASEQ3/ NEWVEH, NEMMYF, NEWMYL, ZMLNEW, DRNEW, KINKS
COMMON /BASEQ4/NEWPAR(5,100) ,BERNEW (3,100) ,NEWFIT (100) ,NEWCT,MAXCT
COMMON /BASEQ5/ ERUZML(12,3,8,2),ERUDR(12,3,8,2),ERU50K(12, 2, 3, 2)
COMMON /BASEQ6/ MYGERU (12, 2, 3, 8, 2) , MAXERU, NUMERU (3, 8, 2) ,KEYER, IGER
COMMON /BASEQ7/ IPMOD(3)
COMMON /BASEQ8/ IHDVMT(23,2),HDVZML(23,2) ,HDVBDR(23, 2)
COMMON /BYMYC1/ IVPICK (8) , IMPICK, IDPICK,BYCUM (25, 8)
COMMON /BYMYC2/ BYBEF4(3,25,8),BYTAM(3,25,4),BYFER(3,25,8)
COMMON /BYMYC3/BYEVAP(25,8),BYRUNL(25,4),BYREFL(25,4),BYRSTL(25, 8)
COMMON /BYMYC4/ BYVTN (8) , YFM,PN (10) , U7,PSEC (3) , VB
15
-------
Appendix A
COMMON /BYMYC5/ BYIMCR(2, 25, 4) ,BYIMRE (2, 25, 4) ,REDSUM<2, 4)
COMMON /CEVCOM/ USRTPD(8),USRMPD(8)
COMMON /CEVBMY/ BMYMPD (25, 8) ,BMYTPD (25, 8) ,BMTPDF (25, 8)
COMMON /CITCIN/ UDI(5),IUDI,PFRATE(3)
COMMON /CITPAR/ SCNAME(4)
COMMON /CITRV1/ RVPBAS,RVPIUS,RVPICY, I0SESY,RVPUWX
COMMON /CITRV2/ RVPHS,RVPX,RVP090,RVP115
COMMON /CRACOM/ MAXCRA, CRAVAL(3, 8, 2) ,MYGCRA<3, 8, 2)
COMMON /CUMCOM/ CUMMIL(25,8)
COMMON /EGSCAL/ AER(11,11,2,2),TGS(11,6,4),EGS(7,2)
COMMON /EMECAL/ EM3W(3,3, 3) ,EMOX (3,3, 3) ,EMEGR<4, 3,3) ,EMI (7, 3, 3)
COMMON /EVADC1/ DUEQ(3,2,3) ,EFDU(3,2,5,5)
COMMON /EVADC2/ HIDO(3,2),R2D0(2,2)
COMMON /EVAHIA/ HIHS(3,2),HIADJ
COMMON /EVAHS1/ HSEQ<3,2,9),EFHS(3,2,5,5)
COMMON /EVAPAR/ IFDS, I3TDC, ISTDT,ISTDP,FINJ(2)
COMMON /EVAPGR/ EVP (4) , GREVP (4, 9) ,VGREVP (4) ,PARD (3) ,GRPD (3, 9)
COMMON /EVAEHD/ HDSAL, HDWGT (2) , IWMAP (5)
COMMON /EVASTD/ MYGSTD(5,S),MAXSTD,KSTD81(4,3)
COMMON /EVPDAT/ EVPAERJ3,2),EVPTGS(3,2,4),CCEMI(7,4)
COMMON /FID/ MAXFID,MYFID(16,5),TOGCF(16,5,3),VOCCF(16,5,3)
COMMON /FLAGS1/ PROMPT, TAMFLG, SPDFLG,VMFLAG,OXYFLG,DSFLAG
COMMON /FLAGS2/ MYMRFG,NEWFLG, IMFLAG, ALHFLG
COMMON /FLAGS3/ ATPFLG, TPDFLG,RLFLAG,LOCFLG, TEMFLG, OOTFMT
COMMON /FLAGS4/ PRTFLG, IDLFLG, NMHFLG,HCFLAG,COLDFG
COMMON /GSFCOM/ MAXGSF,GSFRAC(22,8,2),MYGSF(22,8,2)
COMMON /HDCCOM/ HDCFAC (37,2), LOWYR, MAXCIG
COMMON /HDDMAR/ C2BDVM (25) , LHDVM (25) , MHDVM (25) , HHDVM (25)
COMMON /HDDREG/ C2BDRG(21),LHDRG(21),MHDRG(21),HHDRG(21)
COMMON /IDLEQ1/ ZMLIDL(15,3,8,2),DR1DL(15,3,8,2)
COMMON /IDLEQ2/ MYGIDL(15,3,8,2),MAXIDL
COMMON /IDIBQ3/ UTIDOF(14,3,2)
COMMON /ICR4VA/ CRD4VA(24, 2,12,3)
COMMON /ICR4VB/ CRD4VB(24,2,12,3)
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER(2) ,CRIM
COMMON /IMPAR2/ ILDT (4) , ITEST, NUDATA (2) , NLIM, IMNAME (20, 9)
COMMON /IMPAR3/ IMDX1, IMDX2, IMCASE, IMVEH, IM
COMMON /IMPAR5/ MYGIM2 (3,3) , LBIM4P,CRHDGV(2, 2) ,DISCNT (3)
COMMON /IMPAR6/ IFREQ,INTYP
COMMON /IM12HC/ CR12HC(19,20,5, 2)
COMMON /IM12CO/ CR12CO(19,20,5,2)
COMMON /IM240P/ DSIZE (25, 4) , IM24YR, IPRGYR, IPRSYR
COMMON /INJECT/ TBI (13, 4) ,PFI (13, 4) ,MAXFIY,MINFIY
COMMON /IOUCOM/ IOUIMD, IOUGEN, IOUREP, IOUERR, IOUASK
COMMON /IVPCOM/ IVPTRL(8) , IVPTRT (8) , IVPTRA(8) , IVPTRB(8)
COMMON /LOOKUP/ IVTAM, IQG, IPG, JPGD, IHG, IGCSF
COMMON /MAXIMA/ MAXVEH, MAXLTW, MAXPOL, MAXREG,MAXYRS
COMMON /MPDCOM/ MAXMPD , VALMPD (2,8), MYGMPD (2,8)
COMMON /MYCODE/ MY,IDX, JDX,LDXSY,LMYRVT, XAY, IMDXSY, IMKINK
COMMON /MYRCAL/ XMYM(25,8),JANMYR(2S,8),TF(25,8),TFMYM(25,8)
COMMON /MYRSAV/ AMAR (25, 8) , JULMYR (25, 8) , NEWCUM
COMMON /MYOB1/ MYTGS (11, 6, 4) ,MYEGR(4, 4) ,MYVTS (3, 2, 4)
COMMON /MYUB2/ MYCCEI(7,4)
COMMON /NAMESl/ NAMFLG (19) , NAMMMR (2, 3, 8) , NAMRVP (4) , NAMTEM (2)
COMMON /NMETH1/ ZDMTH1(2,3,2, 7,3,2),ZDMTH2(2,7,2,2),ZDMTH3(7,3,2)
COMMON /NMETH2/ MYGMTH(7,8,2),MTHMYG
COMMON /OFFSET/ OFFCO(25,3),OFFMTH(25,8)
COMMON /OMTCOM/ OMTCF(25,3,8),OMTTAM(25,3,4),OMTCFF(25,3,8)
COMMON /OPMOD1/ BFRCOF(8,8,3,6),BFSTEP(8,12,3,3),BFR50K(4,12,3)
COMMON /OPMOD2/ MY3TEP,MAXBFR,MYGBFR(8, 3, 6)
COMMON /OPMOD3/ BFMILE,KINKBF,BFGT50
COMMON /OXY1/ SHRMKT(3),OXYCNt(2),IGASHW
COMMON /OXY2/ NFUEL, MXOXMY, MAXOXY, OXYVAL (25,12), OXYBF (25,12) , OXYCF
COMMON /PROJEC/ PROJID(20)
COMMON /QUITXQ/ N1QUIT
COMMON /REGION/ FEET(2),IREJN,ALT,INITPR
COMMON /REGISF/ GSFVCT(8),USRGSF(25,2)
COMMON /RESTJL1/ EFFTP (3, 9) ,EFEXH (9) ,EFEVAP (9) ,EFREFL (9) ,EFRUNL (9)
COMMON /RESUL2/ RLGGAL (9) ,RSTGPH (9) , EFRSTL (9) ,EFIDLE (3, 9)
COMMON /RESUL3/ VFTP (3) , VEXH, VEVAP, VLOSS, VRUNLS, VRSTLS, VIDLE (3)
COMMON /RLCOMl/ ROADFE(32,4),DISPL,SPILL,OBED,OBES,OBDF(4)
COMMON /RLCOM2/ IOBMY,IVOB(4),IS2SY,NPHASE,S2EFF(4)
COMMON /RLCOM3/ RLRATE(25,4)
COMMON /RONLS1/ TLEMI (6, 3, 4, 4, 4, 4) , TLVMT (6) , TLVMTU (6) ,MAXTL
COMMON /RONLS2/ TLFAIL(6,4,4)
COMMON /RONLS3/ MYGRUL (4, 4) ,RULRVP (4) ,ROLTEM (4) ,RULSFD (3) ,MAXROL
COMMON /RESTLS/ PTGRST(13,5,2),RESTA(2),RESTS
COMMON /RVPEX1/ R9G1(3,2),RG23(3,3,3,2),ILL(2)
COMMON /RVPEX2/ MYGRVP(4,4),RVPCF(25,3,4)
COMMON /RVPNAT/ RVPLIM(2),RVPDIU(2),PFUL
COMMON /SCENE1/ SPD(8),PCCN,PCHC,PCCC
16
-------
Appendix A
COMMON /SCENE2/ FCC,FCN,FHC,FHN
COMMON /SPEED1/ SKI(6,18,3),SK2C(2,14,3,3),SK4C(2,14,3,3)
COMMON /SPEED2/ MYGSP1 (18,2,3) ,MYGSMC (18, 2,3) ,MYGSP2 (3,3)
COMMON /SPEEDS/ SADJ,SCUNA1(18,3,8),SCADJl(3,4),ISPGRP(3,4)
COMMON /SPEED4/ MAXSP1,LB1STS(3),LBLAST,IVB,IVA,SCUNAO(18,3, 8,2)
COMMON /SPEEDS/ MYHSPC(2,3),MAXHSP(3)
COMMON /SPEEDS/ SALHCF(25,3,8),SCFIDL(25,4),SCIADJ(25,3,8)
COMMON /SPEED7/ GPBSCO(3,3,4)
COMMON /SIZCAL/ ATR(5,2),TGSUSE,CSIZE(12,2),CSAE(12)
COMMON /SIZPAR/ NTERMS(4),INDXCS(5,4)
COMMON /STRING/ NOYES(2),COMMA,PERIOD, COLON
COMMON /TAMEQ1/ TAMZML(9, 4,3, 2) ,TAMDR<9, 4,3,2) ,TAMA50 (9, 4,3, 2)
COMMON /TAMEQ2/ MYGTAM(3,4),FSOK
COMMON /TAMEQ3/ COMCAP,CUMALL,caMNIM,CUMIM,QT50KA,GT50KN,GT50KI
COMMON /TAMEQ4/ BTR(9,2)
COMMON /TAMID1/ EMI3W(3,3,3),EMIOX(3,3,3),EMIDL(3,4,3)
COMMON /TAMID2/ MYIDLE(3,3),IEK
COMMON /TAMID3/ OFFIDL(3,25,4)
COMMON /TAMOUT/ TAMBAG (3,3, 25, 4), THS (2,25, 4) , TDU (25, 4), TCC (25, 4)
COMMON /TAMOO2/ TOB(25,4),GCONLY(25,4)
COMMON /TAMPB1/ LPOD,PBBTR(3,2,25, 4)
COMMON /TEMPC1/TTFC (2, 3, 9, 3) , TTFD (3, 3, 3), TT4 (2, 6,3) , TT8 (2,3, 4,3)
COMMON /TEMPC2/ MAXTCF,MYGTCF(9,3,5),MYGTFD(3),ISHIFT
COMMON /TEMPC3/ MDLOHI (3) , TDIFF (3) ,CHFRAC (3) , TCF (3)
COMMON /TEMPC4/ MYGCOO(3) ,LOWCO,COOTFC(3) ,COOTFD (3)
COMMON /TEMPC5/ NCC,MYCOL(2),CCSTD(3,2)
COMMON /TEMPC6/ CSUB (3,2), CM0L(3, 2) ,PIVPCT (3,3, 2)
COMMON /TEMPC7/ ICOLD,ADDCSW,MULCSW
COMMON /TEMPS/ AMBT, TEMMIN, TEMMAX, TEMEXH (3) , TEMEVP (6)
COMMON /TPDCOM/ TPDCO(2,8)
COMMON /USDATA/ USNAME(4,4),NUSD,IUSD(4)
COMMON /VMXCOM/ REGMIX(8),TFNORM(8),VMTMIX(8), VCOUNT(39,8)
COMMON /YEARS4/ IY1941,IY1960,IY2020
C
COMMON /SAVE01/ JCALL,VA001(15,3,8,2),VA002(15,3, 8, 2)
COMMON /SAVE02/ VA003(24,2,12,3),VA004(24,2,12,3)
COMMON /SAVE03/ VA005 (19, 20,5,2) ,VA006 (19,20,5,2)
COMMON /SAVE04/ VA007(25,8),VA008(25, 8)
COMMON /SAVE05/ VA009(9,4,3,2),VA010(9, 4,3,2)
COMMON /SAVE06/ VA011(8)
C
C
C The GEM Post-Processor COMMON BLOCKS
C
COMMON /COST02/ IFEVYR, ILEVYR
COMMON /COST07/ CSTIM,GASCST,IMSTRT,ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST27/ IGRFLG, IATPLP, IDIFF,OUTPRF
C Add JPRGYR & JPRSYR, 5/29/91
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, OPRSYR
C
COMMON /COSTXX/ JMYR1,JMYR2
C
COMMON /ZONAME/ M4IN,M4OUT,M4ZMC,C4IN
C
C CHARACTER*20 M4IN,M40UT
C
C CALL GETCL(M4IN)
C
ISAV93-0
IF(JCALL.EQ.O) GOTO 10
C
C Subroutine call, aetup run.
C
CALL REINIT
C
IF(JCALL.EQ.l) CALL BDSAVE
C
IF(JCALL.EQ.2) CALL RESTOR
JCALL-2
C
C... Set for pc file i/o
C
C OPEN(ONIT-5,FILE-'M4INPOT',STATUS-'OLD')
CC OPEN (UNIT-5, FILE-M4IN, STATUS-' OLD ')
C OPEN (UNIT-6, FILE-'M4OUTPUT' , STATUS- 'NEW ,
C * CARRIAGE CONTROL-"FORTRAN")
C
C Initialize non-COMMON variables (INERR t local variables).
C
10 INERR-0
ICYOLD—1
17
-------
Appendix A
IROLD=»-1
PRECY=0
LEAST1=0
C
C Control and One-Time Data Sections:
C
C Get the execution control flags and parameters, and then use them to read
C in data, if any, that is read in once and applied to all scenarios. Note:
C alternate RETURN1 is the fatal read error exit: execution stops.
C
CALL CONSEC(*99,*999)
C
C Set up for an ATP only run.
C
IATPLP - 1
IF(IMFLAG.EQ.l .AND. ATPFLG.EQ.2) IATPLP - 2
C
CALL ONESEC(INERR,*99)
C
C Store original input settings
C (ISTRT=-IMSTRT set in ONESEC)
C
ISTRT- IMSTRT
JIM - IMFLAG
JATP - AIPFLG
JERGYR - IPRGYR
JPRSYR - IPRSYR
JMYR1 - MODYR1
JMYR2 - MODYR2
C
GOTO 15
C
C Enter scenario loop.
C
ENTRY MOBNXT(INERR, *)
C
15 CALL PPINIT
C
C
C Note that if INERR > 0 coming out of CONSEC/ONESEC, then the scenarios will
C be read and processed, but EFCALX/OUTPOT will not be called. The user will
C only get error diagnostics out of the run.
C
C 20 CALL PARSEC (ICY, INERR, *98)
20 CALL PARSEC (ICY, INERR, *999)
C
C Display status on screen...
C
WRITE (0,100) ICY,PROJID
100 FORMATC Start CEM4.1 Run for Evaluation Year: ',I4,/,
* 1X,20A4 )
C
IF(ICY.LT.ICYIM .AND. ICY.LT.LAPSY) GOTO 20
C
C Removed old CALL CHKSEN...
C
IFEVYR " ICY
C
C later add direct input of f of consec yrs (IDIFF) to ilevyr
C
ILEVYR - IFEVYR + IDIFF - 1
C IDIFF - ILEVYR - IFEVYR + 1
C
C... later can change 1 to some com input number of consecutive yrs if want.
C
22 JCYDX - ICY - IFEVYR + 1
C WRITE (8, 102) JCYDX, ICY, IFEVYR, ILEVYR
102 FORMAT(IX,'jcydx,icy,ifevyr,ilevyr:',416)
C
C Enter program scenario loop.
C
DO 25 JSCEN-1,4
C
ISCEN - 4 - JSCEN + 1
C
C If ATP-Only case, skip I/M MOBILE4 runs
C
IF(IATPLP.EQ.2 .AND. JSCEN.GE.3) GOTO 26
C
CALL RESET
C
18
-------
Appendix A
C 50+ errors is (arbritrary) run limit •»> atop execution. Otherwise,
C one or more input processing errors -> abort current scenario S then do
C input processing for error diagnostics on the remaining scenarios. INERR
C is never reset to 0, but instead accumulates until the run ends or the
C cutoff value (50) is reached.
C
IF(INERR.GT.50) GOTO 30
C IF(INERR.GT.O) GOTO 20
IF(INERR.GT.O) GOTO 30
C
IF(ICYOLD.NE.ICY.OR.IROLD.NE.IREJN) CALL REGMOD (ICY, INERR)
IF(INERR.GT.50) GOTO 30
C IF(INERR.GT.O) GOTO 20
IF(INERR.GT.O) GOTO 30
ICYOLD-ICY
IROLD-IREJN
C
C Warning 130 - for ICY'a 1993+ MOBILE4.1's disclaimer
C
IF(ISAV93.EQ.O.AND.ICY.GE.1993) CALL QUITER(0.,0,120,INERR)
IF(ICY.GE.1993) ISAV93-1
C
CALL EFCALX (ICY, INERR)
C. . .
C IF(INERR.GT.50) GOTO 30
C IF(INERR.GT.O) GOTO 20
IF(INERR.GT.O) GOTO 30
C. . .
CALL EFSAVE(ISCEN, JCYDX)
CALL SAVER (ICY)
IF(OUTPRF.EQ.l) GOTO 25
IF(OUTPRF.EQ.2 .AND. JCYDX. LE. IDIFF .AND.
* JSCEN.EQ.l) CALL OUTPUT (ICY)
IF(OUTPRF.EQ.3 .AND. JCYDX.LE. IDIFF .AND.
* (JSCEN.EQ.l.OR.JSCEH.EQ.4)) CALL OUTPUT(ICY)
IFJOUTPRF.EQ.4 .AND. JCYDX. LE. IDIFF) CALL OUTPUT(ICY)
IF(OUTPRF.EQ.S) CALL OUTPUT(ICY)
C
25 CONTINUE
C
LEAST1-1
C GOTO 20
GOTO 28
26 CALL ATPSAV(ICY)
C
28 IF(JCYDX.LE.IDIFF) THEN
ICY - ICY+1
GOTO 22
ENDIF
GOTO 99
C
C GOTO 20
C
30 CALL QUITER(0.0,0,28,INERR)
C
C If user is supplying replacement basic FTP e£ parameter records, always
C print the records contents and disposition table, even when input errors
C have prevented the normal output routines from being called. The only
C exception is when a READ error / end-of-file occurs in CONSEC or ONESEC.
C
C Thia section can be expanded to print other optional input.
C
98 IF(LEAST1.EQ.O.AND.NEWFLG.EQ.2) CALL OUTNEW
C
99 RETURN
C
C End-of-file on 1st read of call -> input for call missing entirely.
C Calling program can branch to attach a new input file or terminate
C calls to MOBILE.
C
999 RETURN 1
C End MOBILE
END
19
-------
Appendix A
SUBROUTINE TAMPER(ICY)
C
C TAMPER computes tampering effects in grama/mile on bag 1, 2 & 3 HC, CO,
C & NOX and on hotsoak, diurnal S crankcase HC.
C
C Called by EFCALX.
C
C Calls BAGEME, DISATP, EFFGRP, EMIRAT, EVPEME, FINDPB, and IMCHEK.
C
C Input on call:
C
C parameter list: ICY
C common blocks:
C /ATPAR1/ LAPSY,LAP1ST,LAPLST,LVTFLG
C /FLAGS4/ PRTFLG
C /IMPAR1/ ICYIM
C /IMPAR3/ IM
C /MAXIMA/ MAXYRS
C /TAMPB1/ LPOD
C
C Output on return:
C
C common blocks:
C /LOOKUP/ IVTAM
C /MYCODE/ MY, IDX, JDX, LDXSY, LMYRVT, IMDXSY, IMKINK
C
C Local variable / array dictionary:
C
C Name Type Description
C ™~~™™» —«»« V —>««««*««MMBWM — «>«_~~~M.—««••««.•_« ~~ •• — ••••••.™™™™.»™™»™™™.«™
C LDX1ST I JDX order pointer to first model year covered by ATP
C LDXLST I JDX order pointer to last model year covered by ATP
C LMYR I flag relating model year to ATP parameters
C 1 •• model year not covered by an ATP
C 2 = JDX is covered by an ATP (but IVTAM may not be,
C hence, check for inclusion to set LMYRVT)
C
C Notes:
C
C TAMPER was changed for MOBILE4.1v6 to call FINDPB.
C
C... ATPFLG.EQ.l check added, and ATPFLG declared aa INT.
C. . .
INTEGER PRTFLG, HCFLAG, COLDFG, ATPFLG
C
COMMON /ATPAR1/ LAPSY, LAP1ST, LAPLST, LVTFLG (4)
C. . .
COMMON /FLAGS3/ ATPFLG, TPDFLG,RLFLAG, LOCFLG, TEMFLG, OUTFMT
C
COMMON /FLAGS4 / PRTFLG, IDLFLG, NMHFLG, HCFLAG, COLDFG
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER (2) ,CRIM
COMMON /IMPAR3/ IMDX1, IMDX2, INCASE, IMVEH, IM
COMMON /LOOKUP/ IVTAM,IQG,IPG,JPGD,IHG,IGCSF
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG, MAXYRS
COMMON /MYCODE/ MY, IDX, JDX, LDXSY, LMYRVT, IAY, IMDXSY, IMKINK
COMMON /TAMPB1/ LPOD,PBBTR(3,2,25,4)
C. . .
COMMON /COST07/ CSTIM, GASCST, IMSTRT, ISCEN,MOVIM,MOVATP,REPIM(4)
C
C Relate start year of anti-tampering program (ATP) and 1st and last model
C years covered to calendar year. Put them in JDX order. For example:
C
C ICY - LAPSY -> LDXSY <* 1
C ICY - LAPSY + 1 -> LDXSY - 2
C
C
C ICY - LAPSY + 24 -> LDXSY - 25
C
C JDX order is used because these values are used to access the JDX ordered
C CUMMIL. Note that LDXSY <- 0 => ICY < LAPSY => no ATP was in effect during
C any part of the calendar year's 25 model year window.
C
C The ATP has to be in effect at least 1 year prior to ICY for it to affect
C ICY's emissions. The mileage accumulated through LAPSY all goes to the no
C program case.
C
C Only calculate these values once per scenario => compute them here before
C the tamper loops begin.
C
LDXSY-ICY-LAPSY+1
LDXIST-ICY-LAPIST+I
20
-------
Appendix A
LDXLST-ICY-LAPLST+l
C
C Decide whether to use "Hon-I/M" (IM-1) or "I/M" (IM-2) TAMZML and TAMDR:
C
CALL IMCHEK(1,ICY,IVTAM,IDX)
C
C Relate I/M start year to calendar year, using JDX ordering. This pointer
C will be used within IDX loop to set kinked tampering rate curves flag.
C
IMDXSY-ICY- ICYIM+1
C
C For each tampering vehicle type and each model year in the 24 year window
C from ICY-23 thru ICY (25+ year case is always zeroed), compute the bag and
C evaporative tampering offsets.
C
DO 40 IVTAM-1,4
C
CALL IMCHEK(2,ICY, IVTAM,IDX)
C
DO 30 IDX-2,MAXYRS
C
C MY pointer info using IDX & ICY may be done here, before sub calls.
C
C Identify this loop's model year for use in ITAMPT lookups.
C
MY-ICY- 1, increases with age) my index for use in
C apportioning mileage weights.
C
JDX-MAXYRS-IDX+1
C
C Reset (turn off) the ATP my range switch. Turn on if there is an ATP
C program, it starts before the calendar year, the model year JDX is in the
C affected range and the vehicle type has been selected by the user for
C coverage.
C
C LMYRVT - 1 -> switch is off. ATP not a factor in my JDX'a tampering
C LMYRVT - 2 =»> switch is ON. ATP is a factor in my JDX's tampering
C
C Note: ATPFLG - 1 -> LAPSY - 2020 => LDXSY < 1 => 1st 2 criteria can be
C checked using LDXSY. (...Except for GEM, which changes ATPFLG
C without changing LAPSY or LDXSY accordingly. So we need to include
C the ATPFLG check for use with CEM.)
C
LMYR-1
LMYRVT-1
C
IF (LDXSY.GT.l.AND.LDX1ST.GE. JDX.AND. JDX.GE.LDXLST) LMYR-2
C
C IF (LDXSY.GT.l.AND.ATPFLG.GT.l.AND.LDX1ST.GE. JDX.AND. JDX.GE.LDXLST)
C * LMYR-2
C
C either add next line or use revised long IF line above for CEM...
C. . .
IF (ATPFLG. EQ. 1) LMYR-1
C
IF(LMYR.EQ.2.AND.LVTFLG(IVTAM) .EQ.2) LMYRVT-2
C
CALL IMCHEK(3,ICY,IVTAM, IDX)
C
C Set kinked disablement rate curves flag.
C
IMKINK-1
IF (IM. EQ. 2. AND. JDX. GT. IMDXSY) IMKINK-2
C
C Figure the base tampering rates for both exhaust and non-exhauat emissions.
C
CALL DISATP
C
C If ICY is beyond LPOD, use the tampering rates as of LPOD for the misfueling
C categories (ID-3,4,9) for vehicles in ICY's fleet produced thru LPOD.
C
IF(ICY.GT.LPOD) CALL FINDPB(ICY)
C
C The next 3 subroutines compute the tampering bag emission additives.
C
CALL EFFGRP
CALL EMIRAT
CALL BAGEME
C
21
-------
Appendix A
C The non-exhaust HC emission additives calculation procedure is much
C simpler because there are no overlaps. EVPEME does all the work.
C
IF(PRTFLG.EQ.1.0R.PRTFLG.EQ.4) CALL EVPEME
C
30 CONTINUE
40 CONTINUE
C
RETURN
C End TAMPER
END
22
-------
Appendix A
SUBROUTINE EFFGRP
c
C EFFGRP builds the tampering effects group
C rates from DISATP are modified to account
C types, overlap factors, anti-tampering pro
C technology group sizes. The output rates
C corresponding to the significant emission
C
C Called by TAMPER.
C
C Calls JVDJ12, ATPEFF, A8TO11, C1TO7, C8TO12
C
C Input on call:
C
C common blocks:
C /EGSCAL/ TGS
C /LOOKUP/ IVTAM
C /MYCODE/ MY,tJDX,LDXSY,LMYRVT
C /SIZCAL/ CSIZE
C
C Output on return:
C
C common blocks:
C /EGSCAL/ EGS
C /LOOKUP/ IQG, IPG, JPGD, IGCSF
"C /MYCODE/ IAY
C /SIZCAL/ TGSUSE,CSAE
C
C Local array subscripts:
C
C CSF1ST(7,2,3) - CSF1ST ( IC7, IPG, IGCSF
C
C Local variable / array dictionary:
C
C Name Type Description
C ,_™_~_-____-___~_~~______________
C CSF1ST R category size factor for 1st
C as - =• AorE for
C
C Notes :
C
C EFFGRP was modified for MOBILE4.1 v5 to ex
C its rates. Tampering effects are stored i
C
C
sizes array. The tampering
for technology vs tampering
gram effectiveness factors and
are divided into effects groups
impact tampering groups .
, ITAMPT, SETEGS and TECHO.
)
calculation of CSIZE, expressed
.c. £req> / .
IC7-1-3, CATS for IC7-4-7
pand CSF1ST to 3 MYGs fi updates
n array BSIZE for CEM4.1.
C
c
c
c
c
c
c
DIMENSION CSF1ST(7,2,3)
COMMON /EGSCAL/ AER (11, 11, 2, 2) , TGS (11, 6, 4) , EGS (7, 2)
COMMON /LOOKUP/ IVTAM,IQG,IPG,JPGD,IHG,IGCSF
COMMON /MYCODE/ MY, IDX, JDX, LDXSY, LMYRVT, IAY, IMDXSY, IMKINK
COMMON /SIZCAL/ ATR(5, 2) , TGSUSE, CSIZE (12, 2) ,CSAE (12)
COMMON /COST32/ BSIZE(4,25,2,6,17,2),TRAT(9)
Continuation code is IPG - H (HC £ CO) £ N (NOx)
MYG 1 applies to all MY for HD6V; the indicated ranges are for LDGV/T.
DATA CSF1ST/
CSIZE
4
123
MYG 1 (pre-1981)
H .0708, .1420, .1538, .5529, .0000, .1513,
N .0948, .1209, .0033, .3059, .0039, .3843,
MYG 2 (1981-83)
H .1421, .0519, .0364, .2840, .0406, .1917, .1469,
N .0698, .0233, .0116, .2727, .0260, .2338,
MYG 3 (1984+)
H .1739, .0750, .0938, .1011, .0000, .0470,
N .1212, .0000, .0000, .0000, .0000, .0588,
.2500,
.0118,
.0130,
.0556,
.OOOO/
Zero out tampering effects group sizes (EGS) array.
DO 20 IPG-1,2
DO 10 IEG-1,7
EGS(IEG,IPG)-0.0
10 CONTINUE
20 CONTINUE
23
-------
Appendix A
C Add in each equipment (technology) group's before and after anti-tampering
C program (ATP) atart year contribution to each EGS category for each
C pollutant group.
C
DO 80 IPG-1,2
C
C Set ATR pointer for 1st 3 sizes: HC/CO -> uae AIRS, NOX -> use EGRS.
C
IP(IPG.EQ.l) JPGD-1
IF(IPG.EQ.2) OPGD-5
C
DO 70 IQG-1,6
C
C Lookup technology group size to be used on this pass.
C
IGl-ITAMPT(l)
TGSUSE-TGS (IG1, IQG, IVTAM)
C
C Check how technology type £ size affect tampering rates.
C
CALL TECHO(*70)
C
C HC S CO effects category sizes are identical -> only 2 pollutant passes.
C
DO 65 IAY-1,2
C
C Skip to end of loop if no mileage accumulated (hence no tampering)
C for the pre/post ATP year case being checked.
C
IF(IAY.EQ.1.AND.LMYRVT.EQ.2.AND.LDXSY.GT.JDX) GOTO 64
IF(IAY.EQ.2.AND.LMYRVT.EQ.l) GOTO 64
C
IGCSF-ITAMPT(6)
CALL C1TO7(CSF1ST(1,IPG,IGCSF))
CALL C8TO12
C
C The first calculation pass can produce negative sizes for the no
C overlap (8-11) and the no effects (12) categories. Correct these
C by calling A8TO11 and ADJ12, respectively.
C
DO 30 IC=8,11
IF(CSIZE(IC, IAY) .LT.O) CALL A8TO11 (1C)
30 CONTINUE
IF(CSIZE(12,IAY).LT.O) CALL ADJ12 (CSF1ST)
C
C "Subsequent" (to ATP start year) pass + have split mileage case =>
C "subsequent" category size = overall size - "previous" size
C
C I.e., in the split mileage case, the program computes the previous to
C ATP start year and overall values up to this point. After this next
C loop assigns the subsequent values as residuals, the program works
C directly with the previous and subsequent cases in the calls to ATPEFF
C and SETEGS.
C
IF(IAY.EQ.1.0R.LMYRVT.EQ.1.OR.LDXSY.GT..TOX) GOTO 50
DO 40 IC-1,12
CSIZE (1C, 2) -CSIZE (1C, 2) -CSIZE (1C, 1)
40 CONTINUE
C
C If ATP covers this MY, apply effectiveness factors to the category sizes.
C
50 IF(LMYRVT.EQ.2) CALL ATPEFF(*60)
C
C Otherwise just shift lAY's sizes to CSAE without modification.
C
DO 55 IC-1,12
CSAE (1C) -CSIZE (1C, IAY)
55 CONTINUE
C
C Set the effects group sizes.
C
60 CALL SETEGS
C. . .
64 DO 61 IC-1,12
BSIZE (IVTAM, JDX,IPG,IQG, 1C, IAY) - CSAE (1C)
61 CONTINUE
65 CONTINUE
70 CONTINUE
80 CONTINUE
RETURN
END
24
-------
Appendix A
SUBROUTINE EVPEME
C
C EVPEME serve a the same functions for evaporative (evap - hot soak + diurnal)
C and crankcase (cc) HC as EFFGRP, EMIRAT and BAGEME do for exhaust IF,
C producing ha S du HC tampering ratea £ cc HC offset. The task is simplified
C into one subroutine because there is no overlap categorization or
C weighting. The appropriate numbers are looked up and multiplied
C together with only anti-tampering effectiveness reductions (if any) applied.
C
C Called by TAMPER.
C
C Calls ITAMPTrOTCALC.
C
C Input on call:
C
C common blocks:
C /EVPDAT/ EVPAER,EVPTGS,CCEMI
C /LOOKUP/ IVTAM
C /MYCODE/ IDX, JDX,LDXSY,LMYRVT
C /TAMEQ4/ BTR
C
C Output on return:
C
C common blocks:
C /IM240P/ DSIZE
C /LOOKUP/ IHG
C /TAMOUT/ THS,TDU,TCC
C /TAMOU2/ TOB,GCONLY
C
C Local variable dictionary:
C
C Name Type Description
u
C
C
C
C
C
C
C
C
C
C
C
OTR
OTRCAR
OTRINJ
PCTEQP
Notes :
EVPEME
R
R
R
R
overall (previous
carbureted hot
fuel injected
percentage of
given IG3
modified for MOBILE4.
tampering
rates in DSIZE.
+ subsequent my ranges) tampering rate
soak affecting overall
hot
soak affecting
IVTAM fleet
1 to
store
tampering
overall
equipped with
diurnal
and
rate
tampering rate
IHG technology
,
crankcaae
INTEGER ATPFLG, TPDFLG, RLFLAG, TEMFLG, OUTFMT
C
COMMON /EVPDAT/ EVPAER (3, 2) , EVPTGS (3, 2, 4) , CCEMI (7, 4)
COMMON /FLAGS3/ ATPFLG, TPDFLG, RLFLAG, LOCFLG, TEMFLG, OUTFMT
COMMON /IM240P/ DSIZE (25, 4) , IM24YR, IPRGYR, IPRSYR
COMMON /LOOKUP/ IVTAM, IQG, IPG, JPGD, IHG, IGCSF
COMMON /MYCODE/ MY, IDX, JDX,LDXSY, LMYRVT, IAY,IMDXSY, IMKINK
COMMON /TAMEQ4/ BTR(9,2)
COMMON /TAMOUT/ TAMBAG(3, 3, 25, 4) , THS (2, 25, 4) ,TDU (25, 4) , TCC (25, 4)
COMMON /TAMOU2/ TOB (25, 4) , GCONLY (25, 4)
COMMON /COST32/ BSIZE (4,25,2, 6,17,2) ,TRAT (9)
C
C For MOBILE4, crankcase (IH » 3) processing does not change.
C Diurnal processing only changes in its selection of the new evaporative
C cannister + gas cap tampering (ID - 8) instead of the previously used
C evaporative cannister only (ID = 6) . Since aside from data point selection,
C the procedures for calculating the diurnal and crankcase offsets are the
C same, a loop structure remains the convenient way to program the procedure
C Hot soak, on the other hand, has become more complicated and has therefore
C been pulled from the old IH loop and treated separately, after the diurnal
C and crankcaae offsets have been computed.
C
C MOBILE4 defers HS and DU tampering offset calculation to CCEVRT,
C returning only the % equipped * overall tampering rates instead.
C
C The Diurnal and Crankcase tampering rates are stored in BSIZE for CEM4.1.
C
C Store the non-ATP rates.
C
DSIZE (IDX, IVTAM) -BTR (8, 1) +BTR(8, 2)
C
DO 10 IH-2,3
C
25
-------
Appendix A
C There are only two evaporative cases, not three, when indexing the
C anti-tampering effectiveness rates, evaporative technology group sizes,
C and the model year groups for the group sizes and the evaporative emission
C impact rates: for these arrays, the hot soak and diurnal numbers are
C identical. For convenience we use IHG instead of IH in these arrays.
C
IHG-IH-1
C
C Assign the tampering index (see paragraph preceding DO statement).
C
ID-9-ZH6
C
C Get the overall tampering rate. As in DISATP, there are 3 cases:
C
C (1) No ATP or its start year is after ICY or IDX is not in affected my
C range: use the "previous" base rate without applying the
C effectiveness rates.
C
IF(LMYRVT.EQ.l) OTR-BTR(ID, 1)
C
C (2) There is an ATP and its start year is before IDX and IDX is in the
C affected my range: use the "subsequent" rate * the effectiveness
C rate.
C
IF (LMYRVT. EQ. 2. AND. LDXSY. GT. JDX. AND. ID. EQ. 7 )
* OTR-OTCALC(ID,2,IHG,2)
IF(LMYRVT.EQ.2.AND.LDXSY.GT.JDX.AND.ID.EQ.8)
* OTR-OTCALC(ID-2,2,IHG,2)+OTCALC(ID,2,3,2)-OTCALC(ID-2,2,3,2)
C
C (3) There is an ATP and its start year is at or after IDX but before ICY
C and IDX is in the affected my range: sum the rates * their
C corresponding effectiveness rates. Note that in this case, BTR holds
C the "previous" and overall rates -> subtract 1st from 2nd to get the
C "subsequent" rates.
C
IF(LMYRVT.EQ.2.AND.LDXSY.LE.JDX.AND.ID.EQ.7)
* OTR-OTCALC (ID, 1, IHG, 1) +OTCALC (ID, 2, IHG, 2) -OTCALC (ID, 1, IHG, 2)
IF (LMYRVT.EQ. 2. AND. LDXSY. LE. JDX. AND. ID.EQ. 8)
* OTR-OTCALC(ID-2,1,IHG,1)+OTCALC(ID,1,3,1)-OTCALC(ID-2,1,3,1)
* -K>TCALC(ID-2,2,IHG,2)+OTCALC (ID, 2, 3,2)-OTCALC(ID-2,2,3,2)
* -(OTCALC(ID-2,1,IHG,2)+OTCALC(ID,1,3,2)-OTCALC(ID-2,1,3,2))
C. . .
C Save for CEM via EBSIZE...
C
CALL EBSIZE(ID,OTR)
C
C Lookup model year group pointers.
C
IG3-ITAMPT(3)
C
IF(IH.EQ.3) IG4-ITAMPT(4)
C
C Lookup and apply the percentage equipped and emission impact factors.
C
PCTEQP-EVPTGS(IG3,IHG,IVTAM)
IF(IH.EQ.2) TDO(IDX,IVTAM)-OTR*PCTEQP
IF(IH.EQ.3) TCC(IDX,IVTAM)-OTR*PCTEQP*CCEMI(IG4,IVTAM)
10 CONTINUE
C
C Now calculate the hot soak offset. This case is complicated by both the
C usage of 2 tampering types (evaporative cannister only and evaporative
C cannister plus gas caps) and the weighting together of fuel injected and
C carbureted emission impact rates.
C
IH-1
IHG-IH
C
C Assign the carbureted tampering index • evaporative canniater only, since
C carbureted hot soak HC is not affected by gas cap tampering. The fuel
C injected index is ID+2 - 8 - cannister + gaa caps.
C
ID-6
C
C Get the overall tampering rates. As with diurnal and crankcase, there are
C 3 cases, but now 2 rates are computed.
C
C (1) No ATP or ita start year is after ICY or IDX is not in affected my
C range: use the "previous" base rate without applying the
C effectiveness ratea.
C
IF(LMYRVT.NE.l) GOTO 20
26
-------
Appendix A
OTRCAR-BTR (ID, 1)
OTRINJ-BTR(ID+2,1)
GOTO 40
C
C (2) There is an ATP and its start year is before IDX and IDX is in the
C affected my range: use the "subsequent" rate * the effectiveness
C rate.
C
20 IF(LDXSY.LE.JDX) GOTO 30
OTRCAR-OTCALC (ID, 2, IHG, 2)
OTRINJ-OTCALC(ID,2,IHG,2)+OTCALC(ID+2,2,3,2)-OTCALC(ID,2,3,2)
GOTO 40
C
C (3) There is an ATP and its start year is at or after IDX but before ICY
C and IDX is in the affected my range: sum the rates * their
C corresponding effectiveness rates. Note that in this case, BTR holds
C the "previous" and overall rates -> subtract 1st from 2nd to get the
C "subsequent" rates.
C
30 OTRCAR-OTCALC(ID,1,IHG,1)+OTCALC(ID,2,IHG,2)-OTCALC(ID,1,IHG,2)
OTRINJ-OTCALC (ID, 1, IHG, 1) +OTCALC (ID+2, 1, 3,1) -OTCALC (ID, 1, 3,1)
* +OTCALC (ID, 2, IHG, 2) +OTCALC (ID+2,2,3, 2) -OTCALC (ID, 2,3,2)
* -(OTCALC(ID,1,IHG,2)+OTCALC(ID+2,1,3,2)-OTCALC(ID,1,3,2))
C
C Save the unweighted (by PCTEQP) Evaporative Cannister (EC) Only tampering
C rate for figuring the On Board (OB) EC tampering impact on refueling losses.
C Lookup model year group pointer for the percentage equipped table.
C
40 TOB(IDX,IVTAM)-OTRCAR
IG3-ITAMPT(3)
C
C Lookup the percentage equipped with canisters factor and apply it to
C the otr's to get the tampered percentage of carbureted and fuel injected
C cases for hot soak.
C
PCTEQP=EVPTGS (IG3, IHG, IVTAM)
THS (1, IDX, IVTAM) -PCTEQP*OTRCAR
THS (2, IDX, IVTAM) =PCTEQP*OTRINJ
C
C Store the gas cap only tampering values for running loss weighting.
C
GCONLY (IDX, IVTAM) =OTRINJ-OTRCAR
IF (GCONLY (IDX, IVTAM) .LT.0.0) GCONLY (IDX, IVTAM)-0. 0
C
RETURN
C End EVPEME
END
27
-------
Appendix A
FUNCTION FAIL (MYR, KDX, IV, IK)
C
C FAIL handles the effect of Purge and Pressure failures as well as
C evap system tampering, returning a fraction for purge/pressure.
C Revised for CEM to save P/P failure rates for an in-place
C inspection program (i.e., fail rates lower than for M4.1).
C
C Called by HCCALX.
C
C Calls ENFORC, ITAMPT.
C
C Input on call:
C
C parameter list: MYR,KDX,IV,IK,IG3
C common blocks:
C /EVPDAT/ PCTEQP
C /FLAGS3/ ATPFLG
C /IMPAR1/ GRIM
C /IMPAR2/ ILDT
C /IMPAR6/ INTYP
C /IM240P/ DSIZE,IPRGYR,IPRSYR
C /LOOKUP/ IVTAM
C /MAXIMA/ MAXYRS
C /MYCODE/ MY
C /TAMOUT/ TDU
C
C Output on return:
C
C function: FAIL
C
C Notes.
C
C FAIL was added for MOBILE4.1.
C
C... for CEM...
C PPFAIL Block Recurring (lower) repair rates for Purge, Press,
C Purge-fPress for each MY S IV for Repair Costs.
C PPFIXR Block Purge/Pressure Fix rates = Full non-program fail
C rate minus w/program failure rate for FE benefits.
C ISRECR Local Flag "Is Recurring" to loop again with lower
C P/P fail rates.
C
C
C
INTEGER ATPFLG, TPDFLG, RLFLAG, TEMFLG, OUTFMT
C
COMMON /EVPDAT/ EVPAER(3, 2) ,EVPTGS (3, 2, 4) ,CCEMI (7, 4)
COMMON /FLAGS3/ ATPFLG, TPDFLG, RLFLAG, LOCFLG, TEMFLG, OUTFMT
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER(2) ,GRIM
COMMON /IMPAR2/ ILDT (4) , ITEST,NUDATA(2) ,NLIM, IMNAME (20, 9)
COMMON /IMPAR6/ IFREQ, INTYP
COMMON /IM240P/ DSIZE (25, 4) , IM24YR,IPRGYR, IPRSYR
COMMON /LOOKUP/ IVTAM, IQG, IPS, JPGD, IHG, IGCSF
COMMON /MAXIMA/ MAXVEH,MAXLTW, MAXPOL, MAXREG,MAXYRS
COMMON /MYCODE/ MY, IDX, JDX,LDXSY,LMYRVT, IAY, IMDXSY, IMKINK
COMMON /TAMOUT/ TAMBAG(3,3,25,4),THS(2,25,4),TDU(25,4),TCC(25,4)
C. . .
COMMON /COST07/ CSTIM,GASCST,IMSTRT,ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /COST47/ PPFAIL(25,3,10,4),PPFIXR(25,10,4)
C
DIMENSION RATE(3,13),OVLP(5),OVLR(5),PPEFF(25,2)
C
C Revised Purge/Pressure Failure Rate calc'a, 3/26/91
DATA RATE / .043, .043, .080,
.043, .043, .080,
.043, .043, .080,
.060, .053, .096,
.060, .053, .103,
.069, .053, .120,
.094, .053, .150,
.094, .106, .188,
.142, .152, .258,
* .214, .171, .323,
* .224, .236, .389,
* .229, .336, .442,
* .229, .336, .451 /
C
DATA FACTOR,DEFF / 0.50, 0.50 /
C
28
-------
Appendix A
c
C Purge/Pressure Check Effectiveness
C
C Purge/Pressure Check Effectiveness, Annual, Biennial
C
DATA PPEFF /
* .000,.238,.853,.910,.836,.789,.765,.732,.798,.837,
* . 903, . 967, . 992, . 981, . 967, . 958, . 922, . 928, . 927, . 941,
* .972,.992,.981,.967,.958,
* .000,.238,.853,.910,.779,.789,.612,.732,.582,.837,
* . 733, . 967, . 945, . 973, . 947, . 925, . 849, . 850, . 701, . 853,
* .705,.972,.945,.973,.947 /
C
C Initialize variables
C
FAIL=0.0
PRGR-0.0
PRSR-0.0
PPRT-0.0
DO 10 1=1,5
OVLP(I)=0.0
OVLR(I)-0.0
10 CONTINUE
C. . .
ISRECR - 1
C
C For diesel S motorcycles, return 0.0 for FAIL
C
IF(IV.GT.4) RETURN
C
C Determine percent of vehicles equipped with evap systems
C (Use KDX and MYR to avoid problems with common block MYCODE)
C
MY-MYR
IDX-KDX
C
IHG-1
IVTAM-IV
IG3-ITAMPT(3)
PCTEQP-EVPTGS(IG3,1,IV)
C
C Determine overall Pressure and Purge failure rates
C
JDX= (MAXYRS+1) -IDX
JDXX — JDX
IF(JDXX.GT.13) JDXX=13
PRGR-RATE (1, JDXX) *PCTEQP
PRSR-RATE (2, JDXX) *PCTEQP
PPRT-RATE(3,JDXX)*PCTEQP
C
TRAT-DSIZE (IDX, IV) *PCTEQP
C. . .
C This is where we jump in for the recurring rate loop...
C
15 SUM-PRGR+PRSR
IF(PPRT.GT.SUM) PPRT-SUM
C
C Determine proportion of overall in each overlap
C
OVLP ( 4) -TRAT*FACTOR
OVLP (5) -TRAT* (1. 0-FACTOR)
OVLP (1) =PPRT-PRGR-OVLP (4)
IF(OVLP(1) .LT.0.0) OVLP(1)=0.0
OVLP (2) -PRSR-OVLP (1) -OVLP (4) -OVLP (5)
IF(OVLP (2) .LT.0.0) THEN
OVLP(2)-0.0
OVLP (5) -PRGR- (PPRT-PRSR)
OVLP (4) -OVLP (4) +TRAT* (1. 0-FACTOR) -OVLP (5)
ENDIF
OVLP (3) -PRGR-OVLP (2) -OVLP (5)
IF(OVLP (3) .LT.0.0) OVLP(3)-0.0
C
C Initialize rates in overlaps
C
DO 20 1-1,5
OVLR(I)-OVLP(I)
20 CONTINUE
C
C... 10/30/91 change to skip over program reductions with
C recurring repairs for ongoing program (already part of
C recurring RATEs)
29
-------
Appendix A
c
IF(ISRECR.EQ.2) GOTO 40
G
C Adjust rates to reflect programs
C
EADJ-1.0 - ENFORC(CRIM,1)*PPEFF(JDX,IFREQ)
ADJ-l.O-EADJ
C
C Purge Check Alone
C
IF(MY.GE.IPRGTTR.AND.ILDT(IV) .EQ.2) THEN
IF(INTYP.EQ.l) THEN
OVLR(2)-EADJ*OVLP(2)
OVLR(3)-EADJ*OVLP(3)
OVLR (5) -EADJ*OVLP (5)
ENDIF
IF(INTYP.GT.l) THEN
OVLR(2)-DEFF*ADJ*OVLP(2)+EADJ*OVLP(2)
OVLR (3) -DEFF*ADJ*OVLP (3) +EADJ*OVLP (3)
OVLR(S)=DEFF*ADJ*OVLP(5)+EADJ*OVLP(5)
ENDIF
OVLR(1)-OVLP (1)+OVLP (2) -OVLR(2)
OVLR(4)-OVLP (4)+OVLP (5) -OVLR(5)
ENDIF
C
C Pressure Check Alone
C
IF(Mr.GE.IFRSYR.AND.ILDT(IV) .EQ.2) THEN
IF -OVLP (3) +OVLP (2) -OVLR (2) H-OVLP < S) -OVLR (5)
ENDIF
C
C Combined Purge/Pressure Checks
C
IF(INTTP.NE.3.
* AND.MY.GE.IPRSYR.
* AND.MY.GE.IPRGYR.
* AND.ILDT(IV).EQ.2) THEN
IF(INTYP.EQ.l) THEN
OVLR(1)-EADJ*OVLP(1)
OVLR (2) -EADJ*OVLP (2)
OVLR (3)-EADJ*OVLP (3)
OVLR ( 4)-EAD J*OVLP (4)
OVLR(5)-EADJ*OVLP (5)
ENDIF
IF(INTYP.GT.l) THEN
OVLR (1) -DEFF*ADJ*OVLP (1) +EADJ*OVLP (1)
OVLR(2)-DEFF*ADO*OVLP(2)+EADJ*OVLP(2)
OVLR (3 ) -DEFF*AD J*OVLP ( 3 ) +EAD J*OVLP (3)
OVLR (4) -DEFF*AD J*OVLP (4 ) +EAD J*OVLP ( 4)
OVLR (5) -DEFF*AD J*OVLP (5 ) +EAD J*OVLP ( 5)
ENDIF
ENDIF
C
C Anti-tampering program effects
C
IF(ATPFLG.EQ.2.AND.ILDT(IV) .EQ.2) THEN
IF(MY.LT.IPRSYR.OR.ILDT(IV) .NE.2) THEN
OVLR (4) -TDO (IDX, IV) *FACTOR
OVLR (5) -TDO (IDX, IV) * (1. 0-FACTOR)
ENDIF
ENDIF
C
C Return adjusted overall Pressure/Purge rate
C
C IK — 1 : Purge only effects
C 2 : Pressure effects
C
IF(IK.EQ.l) FAIL-OVLR<3)
IF (IK. EQ.2) FAIL-OVLR(1)+OVLR(2)+OVLR(4)+OVLR(5)
30
-------
Appendix A
c...
C Only need CEM info for full ATP run...
C
IF(ISCEN.NE.4) GOTO 99
C
C
C Save difference between no pgm S with pgm total failure rate
C for FE benefit calcs (PPFEB) in PPFIXR...
C
IF(ISCEN.EQ.4) PPFIXR(IDX, JCYDX, IV) -
* OVLP(1)+OVLP(2)+OVLP(3)+OVLP(4)+OVLP(5)
* -OVLR(1)-OVLR(2)-OVLR(3)-OVLR(4)-OVLR(5)
C
C Force lower "recurring" failure rates for in-place program...
C
PRGR - 0.03
PRSR - 0.025
PPRT - 0.05
ISRECR - 2
C
C Go back and recalc overlaps for recurring failures...
C
GOTO 15
C
C Adjusted overall Pressure/Purge rate
C
C Store Purge fi Pressure Repair rates for later repair cost calcs
C
C OVLR(l) " Press fail only (no tampering)
C OVLR(2) - Press fi Purge fail (no tampering)
C OVLRJ3) - Purge fail only (no tampering)
C OVLRJ4) - Press fail due to tampering
C OVLR(5) - Purge £ Press fail due to tampering
C
C Save recurring Purge/Press failure £ fix rates from loop with ATP..
C
C PPFAIL(x,lrx,x) - Entire PURGE failure rate
C PPFAILJx,2,x,x) = Entire PRESSURE failure rate
C PPFAILJx,3,x,x) - Total PURGE fi PRESSURE failure rate
C i.e., overlap would need to be computed by:
C PPFAIL(l) + PPFAIL(2) - PPFAIL(3)
C
40 IF(ISCEN.EQ.4 .AND. IV.LE.4) THEN
PPFAIL(IDX,1, JCYDX, IV) - OVLP (2) + OVLP (3)
* + OVLP(5)
PPFAIL(IDX,2,JCYDX,IV) - OVLP(l) + OVLP (2)
* + OVLP (4) + OVLP (5)
PPFAIL(IDX,3,JCYDX,IV) - OVLP(l) + OVLP (2)
* + OVLP (3) + OVLP (4)
* + OVLP(5)
ENDIF
C
99 RETURN
C End FAIL
END
31
-------
Appendix A
FUNCTION PCLEFT(MY,ICY,IP,IV)
C
C PCLEFT determines the basic emission factor multiplicative adjustment
C ( - percentage left of BEF - PCLEFT ) for the effects of an inspection /
C maintenance (I/M) program, after checking whether or not I/M applies to the
C factor being computed by the calling function.
C
C Called by BEF.
C
C Calls IMPTR.
C
C Input on call:
C
C parameter list: MY,ICY,IP,IV
C common blocks:
C /FLAGS2/ IMFLAG
C /ICR4VA/ CRD4VA
C /ICR4VB/ CRD4VB
C /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2,WAIVER,CRIM
C /IMPAR2/ ILDT,ITEST
C /IMPAR5/ CRHDGV, DISCNT
C /IMPAR6/ IFREQ,INTYP
C /IM12HC/ CR12HC
C /IM12CO/ CR12CO
C /IM240P/ IM24YR
C
C Output on return:
C
C function: PCLEFT
C
C Local variable / array dictionary:
C
C Name Type Description
C
C AGE1ST I age of the vehicle at first inspection
C BY I benefit year for technology 1 or 2 vehicle
C IBY I benefit year for technology 4 plus vehicle
C IREM I remainder •> stringency - greatest multiple of 10 < stringency
C ISTRN I stringency index into technology 1 or 2 credits array
C ITESTL I local version of ITEST used for IM240 check
C REM R IREM converted to REAL value
C WAIV R waiver rate used on this call, depends on MY
C
C Notes:
C
C Non-TECH4+ vehicles use 19th value for 20-24 benefit year.
C TECH4+ vehicles have arrays expanded to MAXYRS-1.
C PCLEFT was modified to include the IM240 check.
C
C
CHARACTER* 4 IMNAME
INTEGER ALHFLG
C
COMMON /FLAGS2/ MYMRFG, NEWFLG, IMFLAG, ALHFLG
COMMON /IM12HC/ CR12HC(19,20,5,2)
COMMON /IM12CO/ CR12CO(19,20,5,2)
COMMON /IM240P/ DSIZE (25, 4) ,IM24YR,IPRGYR,IPRSYR
COMMON /ICR4VA/ CRD4VA(24,2,12,3)
COMMON /ICR4VB/ CRD4VB (24,2,12,3)
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER (2) , CRIM
COMMON /IMPAR2/ ILDT (4) , ITEST,NUDATA(2) , NLIM, IMNAME (20, 9)
COMMON /IMPAR5/ MYGIM2 (3, 3) , LBIM4P, CRHDGV(2, 2) , DISCNT (3)
COMMON /IMPAR6/ IFREQ,INTYP
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG,MAXYRS
C
COMMON /COST07/ CSTIM,GASCST, IMSTRT, ISCEN,MOVIM,MOVATP,REPIM (4)
C
INTEGER BY,AGE1ST
C
C Initialize PCLEFT to "no reduction" value - 1.0.
C
PCLEFT-1.0
C
C If factor being calculated by BEF is not covered by I/M, then RETURN.
C
IF(IMFLAG.EQ.1.OR.IP.EQ.3.OR.IV.GT.4) GOTO 99
C. . .
IF(ILDT(IV).EQ.l.OR.
* ICY.EQ.MY.OR.ICY.LE.IMSTRT.OR.
* MY.LT.MODYR1.OR.MY.GT.MODYR2) GOTO 99
32
-------
Appendix A
C Assign waiver rates
C Control WAIV (waver rates) for INTYP (I/M program type). INTYP 2-
C I/M Decentralized/Computerized, £ 3- Decentralized/Manual, if INTYP is
G Hot a 1- Centralized I/M and the WAIV is LE 50%, WAIV is set to zero.
C
IF(MY.LE.1980) WAIV»WAIVER(1)
IF(MY.GT.1980) WAIV-WAIVER(2)
IF(WAIV.LE.0.50.AND.INTYP.GT.1) WAIV-0.0
C
IF(IV.EQ.4) GOTO 50
C
C Selecting I/M reduction for LDGV or LDGT. Several parameters must be set.
C
C Determine technology by model year and vehicle type.
C
ITECH-IMPTR(MY,IV)
C
C Find the benefit year:
C Limit value according to technology type.
C
C BY-ICY-ICYIM
C... IF(MY.GT.ICYIM) BY-ICY-MY
BY-ICY-IMSTRT
IF(MY.GT.IMSTRT) BY-ICY-MY
IF(ITECH.LE.3.AND.BY.GT.19) BY-19
IF(ITECH.GE.4.AND.BY.GT.MAXYRS-1) BY-MAXYRS-1
C
C Find the age of the vehicle at first inspection.
C Limit value according to technology type.
C
AGE1ST-1
C. . .
IF(MY.LT.IMSTRT) AGE1ST-IMSTRT-MY+1
C
IF(ITECH.LE.3) THEN
IF(AGE1ST.GT.19) AGE1ST-19
IF(BY+AGE1ST.GT.19) BY-20-AGE1ST
ENDIF
C
C For now, continue MOBILES policy of using technology 2 credits for ITECH-3.
C
IF(ITECH.EQ.3) ITECH*>2
C
C Branch on technology type: TECH 1 & 2 form 1 group, TECH 4+ another.
C
IF(ITECH.GE.4) GOTO 40
C
C Select correct I/M credits for TECH 1 fi 2. The same credits array is used
C for LDGV and LDGT, but the I TECH mygs differ, so that the same MY may yield
C a different credit for LDGV than for LDGT.
C
C Interpolate between 10,20,30,40 £ 30% stringency.
C
IKEM-ISTRIH-(ISTRIN/10)*10
REM-IREM*.1
ISTRH-(ISTRIN-IREM)/10
C
IF (IFREQ. EQ. 1) THEN
IF(IP.EQ.l) THEN
PCRED-CR12HC (BY,AGE1ST, ISTRN, ITECH)
IF(ISTRN.LT.S.AND.IREM.GT.O) PCRED-
* (CR12HC (BY,AGE1ST, ISTRN+1, ITECH) -PCKED) *REM+PCRED
ELSE IF(IP.EQ.2) THEN
PCRED-CR12CO(BY,AGE1ST,ISTRS,ITECH)
IF(ISTRN.LT.S.AND.IREM.GT.O) PCRED-
* (CR12CO (BY, AGE1ST, ISTRN+1, ITECH) -PCRED) *REM+PCRED
ENDIF
C
ELSE IF(IFREQ.EQ.2) THEN
IF(IP.EQ.l) THEN
PCRED-CR12HC (20-BY, 21-AGE1ST, ISTRN, ITECH)
IF(ISTRN.LT.5.AND.IREM.GT.O) PCRED-
* (CR12HC (20-BY, 21-AGE1ST, ISTRN+1, ITECH) -PCRED) *REM+PCRED
ELSE IF(IP.EQ.2) THEN
PCRED-CR12CO(20-BY,21-AGE1ST,ISTRN,ITECH)
IF(ISTRN.LT.5.AND.IREM.GT.O) PCRED-
* (CR12CO (20-BY, 21-AGE1ST, ISTRN+1, ITECH) -PCRED) *REM+PCRED
ENDIF
C
ENDIF
C
33
-------
Appendix A
GOTO 60
c
C Select I/M credits for TECH 4+ vehiclea. For TECH 4+, there are separate
C credits arrays, one for annual and one for biennial inspection programs.
C Use the Loaded/Idle Test for IM240 check.
C
40 IBY-BY+AGE1ST-1
C
ITESTL-ITEST
IF(M¥.GE.IM24YR.AND.MY.GE.1981) ITESTL-3
C
IF (IFREQ. EQ. 1) PCRED-CRD4VA (IBY, IP , ITECH-3 , ITESTL)
IF(IFREQ.EQ.2) PCRED-CRD4VB(IBY,IP,ITECH-3,ITESTL)
GOTO 60
C
C Assign HDGV I/M credit.
C
50 PCRED-CHHDGV(IP,IFREQ)
C
C Set I/M benefit adjusted for waivers and enforcement and discount
C
60 PCLEFT-1. 0- (PCRED* (1. 0-WAIV) *ENFORC (GRIM, 1) *DISCNT (INTYP) )
C
99 RETURN
C End PCLEFT
END
34
-------
Appendix A
c
c
c.
c
c
c.
c
c.
c
G
C
c
c
c
c
c
c
c
c
c
c
c
c
c
MAIN
.Main Driver for
Model .
modified MOBILE4/Cost Effectiveness
.Calls M4MAIN, SETUP, INTWDW, FLEET, TBASE, LOOP, SUMWDW and PRINT
.Variable
Name
IATPLP
INERR
ISIZE1
ISIZE2
PPFAIL
Dictionary:
Source
Block
Par am
Par am
Far am
Block
Description
Flag indicating whether an ATP only scenario
is to be run.
MOBILE 4 error counter in subroutine QUITER
Variable containing either the size of the
floating model year window for the I/M coverage
or the earliest model year inspected in I/M.
Variable containing either the size of the
floating model year window for the ATP coverage
or the earliest model year inspected in an ATP.
Failure rates for Purge, Press, Purge+Press for
each model year and each IVGAS
REAL JTJLMYR
INTEGER PROMPT
CHARACTER*20 M4IN,M4OUT
CHARACTER*8 USNAME
CHARACTER* 4 NOTES , COMMA, PERIOD
CHARACTER*! COLON
COMMON /COST01/ ATPFEE
COMMON /COST02/ IFEVYR, ILEVYR
COMMON /COST03/ STRNGY,RMAXF,REPEAT,FFACT
COMMON /COST04/ TAMPER(25,6,4,10,4)
COMMON /COST05/ REPLCE(25,6,10,4)
COMMON /COST06/ STOAER
COMMON /COST07/ CSTIM,GASCST, IMSTRT,ISCEN,MOVTM,MOVATP,REPIM(4)
COMMON /COST08/ FAILRT(2,25,39,4)
COMMON /COST09/ DRSTRN(5),ZMSTRN(5)
COMMON /COST10/ FEBNFT(25,10,4),FECNMr(20,4)
COMMON /COST11/ PCNTSVJ4)
COMMON /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
COMMON /COST13/ CIMFEE(25,10,4),CIMREP(25,10,4)
COMMON /COST14/ TONRED(25,2,10,4),COST(25,10,4)
COMMON /COST15/ TOTCST(2),TOTRED(2),TCOST
COMMON /COST16/ DATPFE,DIMFEE,DIMREP,0ATPRE,AATPFE,AIMFEE
COMMON /COST17/ DETER(25,2,10,4),TDETER(25,2,10, 4)
COMMON /COST18/ AIMREP,AATPKE, CFACTR,CATPFE (25, 10, 4)
COMMON /COST19/ ICOV1,ICOV2,INDVL1,INDVL2
COMMON /COST20/ TD0CT(25,2,3,10,4),TEVAP(25,10,4)
COMMON /COST21/ CATPRE(25,10,4),FOLBEN(25,20,4)
COMMON /COST22/ ATCOST(2,8)
COMMON /COST23/ VCNT(39,8),GSDSCT(8,10),HDDCnM(25,10)
COMMON /COST24/ ATEQP(20,5,4)
COMMON /COST25/ FRLOSS(25,4,4,10),FREF(25,4,4,10)
COMMON /COST26/ FRSTLS(25,8,4,10),YRSTLS(10),BRSTLS
COMMON /COST27/ IGRFLG, IATPLP, IDIFF,OUTPRF
COMMON /COST28/ JATP,JIM,ISTRT,JCYDX,JPRGYR,JPRSYR
COMMON /COST30/ EVAP (25, 8, 4,10) ,FER(3,25, 8, 4,10)
COMMON /COST31/ 3AER(11,11,2,2),SEVP(2,2),ZAER(11,11, 2,2)
COMMON /COST32/ BSIZE(4,25,2,6,17,2),TRAT(9)
COMMON /COST33/ EMCAL4(7,3,3),EVPD1(3,2,4),EVPD2(7, 4)
COMMON /COST34/ TCOONT,BYEAR,VGRORT
COMMON /COST35/ TCST,TBEN(2),ADFBEN,ADIM(2),ADAT(2),ADET(2)
COMMON /COST36/ YTONS(2,10),BTONS(2),BIDTON(2) ,AEVAP
COMMON /COST37/ YRCOST(10),YRBEN<10,2),YCOUNT(39),ACOUNT
COMMON /COST38/ STOCK(12,8),CLDIST(8)
COMMON /COST40/ YEVTON (10) ,BEVTON, YEREF (10) ,BEREF
COMMON /COST41/ YERLOS(10),BERLOS
COMMON /COST42/ ECAL02(10,6,4),ECAL03(1,2),ECAL04(20, 4)
COMMON /COST43/ EMCAL1(3,3,3),EMCAL2(3,3,3),EMCAL3(4, 3, 3)
COMMON /COST44/ LOOK1 (5) ,MYCDl (8), TAM3Q1 (6)
COMMON /COST45/ TAMD1(8,3,3),TAMD2(8,3,3),TAMD3(8,4,3)
COMMON /COST46/ REDIM(25,2,10,4),REDATP(25,2,10,4)
COMMON /COST47/ PPFAIL(25,3,10,4),PPFIXR(25,10,4)
COMMON /COST48/ PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
COMMON /COST49/ PPFEB(25,10,4),APFBEN
COMMON /COSTXX/ JMYR1,JMYR2
35
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
COMMON /IONAME/ M4IN,M4OUT,M4IMC,C4IN
C
c
c
c
c
c
t
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
Common block with names other than COST?? originate in
:he MOBILE4 section of this program.
/ALUHIN/ AC,XLOAD{3) ,TRAILR<3) , ABSHUM, DB, WB
/ATOAR1/ LAPSY,LAP1ST,LAPLST,LVTFLG(4)
/ATPAR2/ ATPPGM,ATPFQT,CRATP,DISTYP (8)
/CITPAR/ SCNAME(4)
/CITRV1/ RVPBAS, RVPIUS, RVEICY, IUSESY, RVPUWX
/CUMCOM/ CUMMIL(25,8)
/FLAGS1/ PROMPT, TAMFLG, SPDFLG, VMFLAG, OXYFLG, DSFLAG
/FLAGS2/ MTMRFG, NEWFLG, IMFLAG, ALHFLG
/FLAGS3/ ATPFLG, TPDFLG, RLFLAG, LOCFLG, TEMFLG, OUTFMT
/IMPARl/ ICYIM,ISTRIN,MODYR1,MODYR2,WAIVER(2) ,CRIM
/IMPAR2/ ILDT(4),ITEST,NUDATA(2),NLIM, IMNAME(20,9)
/IMPAR6/ IFREQ,INTYP
/IM240P/ DSZZE (25, 4) , IM24YR, IPRGYR, IPRSYR
/IOUCOM/ IOUIMD, IOCGEN, IOOREP , IOUERR, IOUASK
/LOOKUP/ IVTAM, IQG, IPG, OPGD , IHG, IGCSF
/MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG,MAXYRS
/MYCODE/ MY, IDX, JDX, LDXSY, LMYRVT, I AY, IMDXSY, IMKINK
/MYRSAV/ AMAR(25,8),JULMYR(25,8),NEWCOM
/PROJEC/ PROJID<20)
/REGION/ FEET (2) , IRE JN, ALT, INITPR
/SAVE01/ JCALL,VA001 (15, 3, 8, 2) , VA002 (15, 3, 8, 2)
/SCENE1/ SPD(8),PCCN,PCHC,PCCC
/STRING/ NOYES (2) , COMMA, PERIOD, COLON
/TAMEQ4/ BTR(9,2)
/TEMPS/ AMBT, TEMMIN, TEMMAX, TEMEXH (3) ,TEMEVP(6)
/USDATA/ USNAME (4,4), NUSD , IUSD ( 4 )
COMMON /VMXCOM/ REGMIX (8) , TFNORM(8) , VMTMIX (8) ,VCOUNT (39, 8)
COMMON /YEARS4/ IY1941,IY1960,IY2020
DRIVER (as of 08/07/91):
DRIVER demonstrates how to call the subroutine version of
MOBILE4.1
Calls MOBILE.
Local variable / array dictionary:
Name Type Description
INTER
M4CALL
C
I
flag for interactive mode ('Y') vs. batch mode ('N')
loop index for MOBILE4.1 call loop
PROGRAM DRIVER
IFPRMT IsFilePrompt:
1 - M4INPUT...C4OOTPOT
2 - Prompt for each of 4 I/O files for each run
3 - All I/O files for all runs listed in one given filena
LOGICAL ISFILE,ISFLST
CHARACTER*20 C4IN, C4OOT, M4IMC, TMPFIL, M4OOT1, C4OUT1
CHARACTER*! INTER
IOUGEN-5
IOUGEN«0
IOUREP-6
IOUASK-0
IOUERR-6
M4IMC - 'M4IMC'
M4OUT1 = ' '
C40UT1 = ' '
WRITE(IOUASK,300)
300 FORMAT(' ****** CEM4.1 DRIVER for MOBILE4.1 *****'/)
CALL GETCL(M4IN)
36
-------
Appendix A
IF(M4IN.EQ.' ') THEN
IFPRMT - 1
M4IN - 'M4INPUT'
M4OUT - 'M4OUTPUT'
C4IN - 'C4INPUT'
C40UT - 'C4OUTPUT'
ELSE IF(M4IN.EQ.' /'.OR.M4IN.EQ.'/'
* .OR.M4IN.EQ.'?'.OR.M4IN.EQ. ' ?') THEN
C Filename request mode. . .
IFPRMT - 2
ELSE
C File list mode...
INQUIRE(FILE - M4IN,EXIST - ISFLST)
IF(ISFLST) THEN
IFPRMT - 3
OPEN(UNIT-3, FILE-M4IN, STATUS-'OLD')
ELSE
WRITE(0,200) M4IN
200 FORMAT(' File list not found: ',A20)
ENDIF
ENDIF
C
M4CALL-0
C
C ******************************************************
C BEGIN Driver LOOP To run each input/output file aet...
C ******************************************************
C
50 IF(IFPRMT.EQ.3) THEN
READ(3,220,ERR-98,END-99) M4IN,M4OUT,C4IN,C4OUT,M4IMC
220 FORMAT (5 (A20/) )
ELSE IF(IFPRMT.EQ.2) THEN
230 FORMAT(A20)
WRITE(IOUASK,240)
240 FORMAT(' ','Pleaae enter the MOBILE4.1 Input filename',
* ' (Return Quits):')
READ(IOUASK,230,END-99) M4IN
IF(M4IN.EQ.' ') GOTO 99
WRITE(IOUASK,250)
250 FORMAT('S1,'Pleaae enter the MOBILE4.1 Output filename:')
READ (IOUASK, 230,END-99) M4OUT
WRITE(IOUASK,260)
260 FORMAT('fi1,'Pleaae enter the CEM4.1 Input filename:')
READ(IOUASK,230,END-99) C4IN
WRITE(IOUASK,270)
270 FORMAT('S','Pleas* enter the CEM4.1 Output filename:')
READ (IOUASK, 230, END-99) C4OUT
WRITE(IOUASK,280)
280 FORMAT('S','Enter the MOBILE4.1 Credita filename (if any):')
READ(IOUASK,230,END-99) M4IMC
ENDIF
IF(M4IMC.EQ.' ') M4IMC - 'M4IMC'
WRITE(0,390) M4IN,M4OUT,C4IN,C4OUT,M4IMC
390 FORMAT(' I/O Filea: ',5(/« ',A20)/ )
C
C Now check exiatence of chosen M41 fi CEM41 input filea...
C
TMPFIL - M4IN
INQUIRE (FILE - TMPFIL, EXIST - ISFILE)
IF(ISFILE) THEN
TMPFIL » C4IN
INQUIRE(FILE - TMPFIL,EXIST - ISFILE)
C IF(ISFILE) THEN
C TMPFIL - M4IMC
C INQUIRE(FILE - TMPFIL,EXIST - ISFILE)
C ENDIF
ENDIF
IF(ISFILE) THEN
CC
OPEN(UNIT-IOUGEN, FILE-M4IN, STATUS-'OLD')
OPEN(UNIT-7, FILE-C4IN, STATUS-'OLD')
C
C Append if different output files are not given
C
OPEN (UNIT-IOUREP , FILE=M4OUT, STATUS-' UNKNOWN' ,
* CARRIAGE CONTROL-"FORTRAN", POSITION-"APPEND")
OPEN(UNIT-8, FILE-C4OUT, STATUS-'UNKNOWN',
* CARRIAGE CONTROL-"FORTRAN", POSITION="APPEND")
37
-------
Appendix A
ELSE
WRITE(0,290) TMEFIL
290 FORMAT(' File not found: ',A20)
GOTO 99
ENDIF
C
M4CALL-M4CALL+1
C
C ***********************************
C
C Get Coat Effective Inputs
C
CALL GETCEI
C M4MAIN ia the modified MOBILE4 driver. It automatically
C runs MOBILE4 four times for each evaluation date.
C The runs are:
C
C 1. An I/M and ATE run
C 2. An I/M run
C 3. An I/M Oeterence run
C 4. A no I/M and no ATP program run (baseline)
C
ICALL - 1
10 JCALL - 1
IF(ICALL.EQ.l) CALL MOBILE (INERR, * 98)
CC
IF(ICALL.GT.l .AND. ICALL.LT.SO) CALL MOBNXT (INERR, * 98)
IF (ICALL. GE. 50 .OR. INERR. GT.O) GOTO 98
ICALL - ICALL + 1
C CALL MOBILE(INERR,*99)
C
WRITE(0,310) M4CALL,INERR
310 FORMAT(/' ','Run #',13,' INERR -',I2/>
C
C
C Calculate or assign the default coat and benefit assumptions...
C
CALL SETUP(ISIZE1,ISIZE2,INERR)
C
C Initialize all cost and benefit arrays to zero...
C
CALL INTWDW
C
C
C Normalize the vehicle fleet to the first evaluation year requested
C by the user. The default fleet is a total nationwide fleet baaed on
C in-use vehicle counts done by MVMA...
C
CALL FLEET
C
C Calculate the emissions in tons for a given fleet with no
C control program in place...
C
CALL TBASE
C
C
C Compute fuel economy benefits for scenarios with I/M...
C
IF(IATPLP.NE.2) CALL CFFUEL
C
C
C Calculate costa and benefits for each model year and vehicle
C type in each calendar year...
C
C
CALL LOOP(ISIZE1,ISIZE2)
38
-------
Appendix A
c
C SUMWDW auma the coat and benefits over all calendar
C yoara.
C
CALL SUMWDW
C
C Output the programs resulta.
C
CALL PRINT(ISIZE1,ISIZE2)
C
GOTO 10
C
C Close any no longer needed filea...
C
98 CLOSE(5)
CLOSE (6)
CLOSE(7)
CLOSE(8)
C CLOSE(4)
C
IF(IFPRMT.GT.l) GOTO 50
C
99 WRITE(0,320)
320 FORMAT(' ','DRIVER calla completed.')
C
STOP
C End Main Driver
END
39
-------
Appendix A
SUBROUTINE PPINIT
c
C Initializes PPFIXR and PPFAIL arrays to zero, needed for MacFortran
C
C Called by MOBILE.
C
C Calls none.
C
C Input on call:
C
C common blocks:
C
C /BYMTC2/ BYBEF4,BYTAM,BYFER
C /BYMrC3/BYEVAP,BYRUNL,BYREFL,BYRSTL
C
C
C Output on return:
C
C common blocks:
C
C /COST47/ PPFAIL(25,3,10,4),PPFIXR(25,10,4)
C
C Notes:
C
C PPINIT was added for CEM4.1 9/27/91 for better MacFortran use
C
C
COMMON /COST47/ PPFAIL(25,3,10, 4) ,PPFIXR(25,10, 4)
C
C Init PPFIXR, PPFAIL
C
DO 50 13-1,4
DO 50 12-1,10
DO 50 11-1,25
PPFIXR(II,12,13) - 0.0
DO 50 14-1,3
PPFAIL(II,14,12,13) - 0.0
50 CONTINUE
C
99 RETURN
C End PPINIT
END
40
-------
Appendix A
SUBROUTINE EFSAVE(ISCEN,JCYDX)
C
C EFSAVE stores the emission factors for each CE aanario and
C evaluation year.
C
C Called by MOBILE.
C
C Calls none.
C
C Input on call:
C
C parameter liat: ISCEN,JCYDX
C common blocks:
C
C /BYMYC2/ BYBEF4,BYTAM,BYFER
C /BYMYC3/BYEVAP,BYRUNL,BYREFL,BYRSTL
C
C
C Output on return:
C
C common blocks:
C
C /COST25/ FRLOSS,FREF
C /COST26/ FRSTLS
C /COST30/ EVAP, FER
C
C Notes:
C
C EFSAVE was added for CEM4.1 VERSION 2
C
C
C
COMMON /BYMYC2/ BYBEF4 (3, 25, 8) ,BYTAM (3,25, 4) ,BYFER<3,25, 8)
COMMON /BYMYC3/BYEVAP (25, 8) ,BYRUNL(25, 4) ,BYREFL(25, 4) ,BYRSTL (25, 8)
COMMON /COST25/ FRLOSS(25,4,4,10),FREF(25,4,4,10)
COMMON /COST26/ FRSTLS(25,8,4,10),YRSTLS(10),BRSTLS
COMMON /COST30/ EVAP(25,8,4,10),FER(3,25,8,4,10)
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG,MAXYRS
COMMON /SFEED6/ SALHCF(25,3,8),SCFIDL(25,4),SCIADJ(25,3,8)
C
C
DO 10 IDX-1, MAXYRS
DO 10 IV=1, MAXVEH
DO 20 IP-1,MAXPOL
JDX - MAXYRS + i - IDX
C
FER (IP, IDX, IV, ISCEN, JCYDX) - BYBEF4 (IP, JDX, IV)
IF(IV.LE.4)
* FER(IP, IDX, IV, ISCEN, JCYDX)»FER(IP, IDX, IV, ISCEN, JCYDX) +
* BYTAM(IP, JDX,IV)
FER (IP , IDX, IV, ISCEN, JCYDX) -
* FER (IP, IDX, IV, ISCEN, JCYDX) *SALHCF (IDX, IP , IV)
C
20 CONTINUE
C
EVAP (IDX, IV, ISCEN, JCYDX) - BYEVAP (JDX, IV)
FRSTLS (IDX, IV, ISCEN, JCYDX) - BYRSTL (JDX, IV)
C
10 CONTINUE
C
C
DO 30 IDX-1,MAXYRS
DO 30 IV-1,4
JDX - MAXYRS + 1 - IDX
C
FRLOSS (IDX, IV, ISCEN, JCYDX) - BYRUNL (JDX, IV)
FREF(IDX,IV, ISCEN, JCYDX) - BYREFL (JDX, IV)
C
30 CONTINUE
RETURN
C End EFSAVE
END
41
-------
Appendix A
SUBROUTINE SAVER (ICY)
c
C. . SAVER saves the Emission Factors and Tampering Rates generated
C . . from each of the
C. .variables.
C
C. .Called by M4MAIN
C
C. .Calls None
C
C. .Array Subscript a:
C
C. .BSIZE<4,25,2, 6,15
C. .TAMPER (25, 6, 4, 10,
C. .FER(3,25,8)
C. .TRAT(7)
C
four (or two) MOBILE 4 runs in the appropriate
,2) - BSIZE(IV,IDX,IPG,IQG,IC,IAY)
4) - TAMPER (MYDX, IP, IISCEN, ICYDX, IV)
- FER (IP, IDX, IV)
- TRAT(IEG)
C. .Variable Dictionary:
C
C. . Name Source
C BSIZE Block
C
C EFALL Block
C FER Block
C
C ICY Par am
C ICYDX Local
C IDX Local
C IF Local
C ISCEN Par am
C IV Local
C JDX Local
C
C MY Local
C MYDX Local
C TAMPER Block
C TRAT Block
C
C
C
Description
Tampering rates in each overlap section from
MOBILE4
Emission factors from MOBILE4 .
MOBILE4 array variable which originally contains
the emission factors.
Calendar year.
Absolute calendar year 1975-1.
Forward model year index (ascending model years) .
Pollutant type index (1:HC 2: CO 3:NOx).
Control program index.
Vehicle type index.
Backward model year index (descending model
years) .
Model Year.
Absolute model year index (1950-1) .
Tampering rate factors from MOBILE 4.
Intermediate array which contains the tampering
rates .
C
c
c
c
c
c
c
c
COMMON /COST04/ TAMPER(25,6,4,10,4)
COMMON /COST07/ CSTIM, GASCST, IMSTRT,ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST28/ JATP, JIM,ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /COST32/ BSIZE(4,25,2,6,17,2),TRAT(9)
COMMON /COST40/ YEVTON (10) ,BEVTON, YEREF (10) ,BEREF
COMMON /COST41/ YERLOS (10) , BERLOS
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG, MAXYRS
DO 40 IDX - 1,MAXYRS
JDX - MAXYRS + 1 - IDX
ICYDX - ICY - 1974
ICYDX - ICY - 1981
MYDX - MAXYRS + ICYDX - JDX
MY - MYDX + 1950
DO 30 IV-1,4
TRAT (1) -BSIZE (IV, JDX, 1,1
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
IF(TRAT(1) .EQ.0.0)
* TRAT (1)-BSIZE (IV, JDX, 1,
* + BSIZE (IV, JDX,
* + BSIZE (IV, JDX,
* + BSIZE (IV, JDX,
* + BSIZE (IV, JDX,
* + BSIZE (IV, JDX,
TRAT (2) -BSIZE (IV, JDX, 1, 3
* + BSIZE (IV, JDX, 1
TRAT (3) -BSIZE (IV, JDX, 1, 3
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
* + BSIZE (IV, JDX, 1
,1,1)+BSIZE(IV,JDX,1,1,1,2)
,1,2,1) + BSIZE(IV,JDX,1,1,2,2)
,1,3,1) + BSIZE(IV,JDX,1,1,3,2)
,1,4,1) + BSIZE (IV, JDX, 1,1, 4, 2)
,1,5,1) + BSIZE(IV,JDX,1,1,5,2)
,1,8,1) + BSIZE (IV, JDX, 1,1,8,2)
2,1,1)+BSIZE(IV,JDX,1,2,1,2)
1,2,2,1) + BSIZE (IV, JDX, 1,2, 2, 2)
1,2,3,1) + BSIZE (IV, JDX, 1,2, 3, 2)
1,2,4,1) + BSIZE(IV,JDX,1,2,4,2)
1,2,5,1) + BSIZE (IV, JDX, 1,2,5,2)
1,2,8,1) + BSIZE (IV, JDX, 1,2,8,2)
,1,1)+BSIZE(IV, JDX, 1,3,1,2)
,3,9,1) 4- BSIZE (IV, JDX, 1,3,9,2)
, 2,1)+BSIZE(IV, JDX, 1,3,2, 2)
,3,4,1) + BSIZE (IV, JDX, 1,3, 4, 2)
,3,6,1) + BSIZE(IV, JDX,1,3,6,2)
,3,10,1) +BSIZE(IV, JDX,1,3,10,2)
42
-------
C
C
C
C
C
C
C
TRAT(4)-BSIZE(IV,JDX,1,
* + BSIZE(IV,JDX,
* + BSIZB (IV, JDX,
" + BSIZE(IV,JDX,
TRAT(5)-BSIZE(IV,JDX,2,
* + BSIZE (IV, JDX,
* + BSIZE(IV,JDX,
» + BSIZE(IV,JDX,
* + BSIZE(IV,JDX,
* + BSIZE (IV, JDX,
TRAT(6) - BSIZE(IV,JDX,
TRAT (7) • BSIZE (IV, JDX,
TRAT(8) - BSIZE(IV,JDX,
Appendix A
3,3,1)+BSIZE(IV,JDX,1,3,3,2)
1,3,5,1) + BSIZE(IV,JDX,1,3,5,2)
1,3,7,1) + BSIZE(IV,JDX,1,3,7,2)
1,3,11,1) +BSIZE(IV,JDX,1,3,11,2)
4,1,1)+BSIZE(IV,JDX,2,4,1,2)
2,4,2,1) + BSIZE (IV, JDX, 2, 4,2,2)
2,4,3,1) + BSIZE(IV,JDX,2,4,3,2)
2,4,4,1) + BSIZE(IV, JDX, 2,4,4,2)
2,4,5,1) + BSIZE(IV,JDX,2,4,5,2)
2,4,8,1) + BSIZE (IV, JDX, 2, 4,8,2)
1,1,13,1)
1,1,14,1)
1,1,15,1)
C
C
Store £l«at variables for current IM program caae
TAMPER/2nd subscript: 1-air, 2-cat miaaing/no misfueling,
3-mi»fueled 4-pcv, 5»evap, 6=gas cap
TAMEER(IDX,1,ISCEN,JCYDX,IV)-THAT(1)
TAMPER(IDX,2,ISCEN,JCYDX,TV)=TRAT(2)
TAMPER(IDX, 3, ISCEM, JCYDX, IV) -THAT (3) +TRAT (4)
TAMPER (IDX, 4 , ISCEN, JCYDX, IV) -TRAT (7 )
TAMPER(IDX,5,ISCEN,JCYDX,IV)-TRAT(6)
TAMPER (IDX, 6, ISCEN, JCYDX, IV) -TRAT (8)
30 CONTINUE
40 CONTINUE
99 RETURN
End SAVER
END
43
-------
Appendix A
c
c.
c
c
c
c
c.
c
c.
c
c.
c
c.
c
c.
c
c.
c
c
c
G
c
c
c
c
c
c
G
c
c
c
c
c
c
c
c
c
c
c
c
c
c
SUBROUTINE ATPSAV(ICY)
.ATPSAV ia called if an ATP only program is to be modeled. It
allows M4MAIN to be run only twice and atorea the EFALL and
TAMPER array elementa for the no control program situation and
the ATP only program.
.Called by M4MAIN
.Calls None
.Array Subscript:
.TAMPER(25,6,4,10,4)
.Varialbe Dictionary:'
- TAMPER (IDX, IVTAM, IISCEN, JCYDX, IV)
Name
Source Deacription
ICY Param Calendar year.
ICYDX Local Absolute calendar year index 1975—1.
IDX Local Forward model year index (ascending model
yeara) .
IISCEN Local Mobile3 run index(1:No program, 2:Deterrence,
3:I/M only, 4:I/M;ATP).
IP Local Pollutant type index (1:HC 2:CO,3:NOx).
IV Local Vehicle type index.
IVTAM Local Emission control device tampering type index
(1:AIR,2:CAT,3:MISF,4:EVAP 5:PCV).
JDX Local Backward model year index (descending model
yeara).
MY Local Model year
MYDX Local Absolute model year index (1950=1).
TAMPER Block Tampering rate factora from MOBILE4.
COMMON /COST04/ TAMPER(25,6, 4,10,4)
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /COST30/ EVAP(25,8,4,10),FER(3,25,8,4,10)
COMMON /COST25/ FRLOSS(25,4,4,10),FREF(25,4,4,10)
COMMON /COST26/ FRSTLS(25,8,4,10),YRSTLS(10),BRSTLS
COMMON /COST40/ YEVTON(10),BEVTON,YEREF(10),BEREF
COMMON /COST41/ YERLOS (10) , BERLOS
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREG, MAXYRS
DO 40 IDX - 1,MAXYRS
JDX = MAXYRS + 1 - IDX
ICYDX - ICY - 1974
ICYDX - ICY - 1981
MYDX - MAXYRS + ICYDX - JDX
MY - MYDX + 1950
DO 40 IV - 1,8
IF(TV .GT. 4) GOTO 20
DO 10 IVTAM - 1,5
DO 10 IISCEN "1,2
TAMPER (IDX, IVTAM, IISCEN, JCYDX, IV) -
* TAMPER (IDX, IVTAM, 3, JCYDX, IV)
10 CONTINUE
20 DO 30 IISCEN -1,2
DO 25 IP - 1,2
FER (IP, IDX, IV, IISCEN, JCYDX) -FER (IP, IDX, IV, 3, JCYDX)
25 CONTINUE
EVAP (IDX, IV, IISCEN, JCYDX) -EVAP (IDX, IV, 3, JCYDX)
FREF (IDX, IV, IISCEN, JCYDX) -FREF (IDX, IV, 3, JCYDX)
FRLOSS (IDX, IV, IISCEN, JCYDX) -FRLOSS (IDX, IV, 3, JCYDX)
FRSTLS (IDX, IV, IISCEN, JCYDX) -FRSTLS (IDX, IV, 3, JCYDX)
30 CONTINUE
40 CONTINUE
99 RETURN
End ATPSAV
END
44
-------
Appendix A
SUBROUTINE RESET
C
C..RESET resets the IMFLAG and ATPFLG for each MOBILE4 call in M4MAIN.
C..NOW also resets IPRGYR & IPRSYR
C
C..Called by M4MAIN
C
C..Calls None
C
C..Array Subscripta:
C
C. .AER(11,11,2,2) - AER(I4,I3,I2,I1)
C..BSIZE(4,25,2,6,15,2) - BSIZE (11,12,13,14,15,16)
C. .SAER(11,11,2,2) - SAER(I4,I3,I2,I1)
C..SEVP(2,2) - SEVP(2,2)
C. .VA023(11,11,2,2) - VA023(I4,I3,I2,I1)
C..VAO24(2,2) - VAO24(I2,I1)
C
C..Variable Dictionary:
C
C.. Name Source Description
C
C AER Block The ATP effectiveness rates for non-evaporative
C exhaust emission tampering.
C ATPFLG Block The anti-tampering flag
-------
Appendix A
MODYRl = JMYRl
MODYR2 - JMYR2
C Now set per scenario
IF(ISCEN.EQ.4) THEN
ATPFLG = JATP
IF (JIM.EQ. 2) IMSTRT - ISTRT
IPRGYR - JPRGYR
IPRSYR - JPRSYR
ELSE IF(ISCEN.EQ.3) THEN
IF(JIM.EQ.2) IMSTRT = ISTRT
CC
C ELSE IF(ISCEN.EQ.2) THEN I No changesB needed.
C MODYRl - 2020
C MODYR2 - 2020
CC
ELSE IF(ISCEN.EQ.l) THEN
C IMFLAG - 1
MODVR1 » 2020
MODYR2 - 2020
ENDIF
C
99 RETURN
C End RESET
END
46
-------
Appendix A
SUBROUTINE SETUP(ISIZE1,ISIZE2,INERR)
C
C..SETUP defines moat of the default coat and benefit assumptions
C..used in the model.
C
C..Called by MAIN
C
C..Calls QUITER
C
C..Variable Dictionary
C
C.. Name Source Description
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
ATCOST Block Cost of Replacing emission control devices as a
result of failing a tampering inspection. The
default assumptions are: ATCOST(TECH,TYPE)
TYPE =• 1 : Air Pump System
- 2 : Catalyst Removal
• 3 : Misfueled Catalyst
» 4 : Evaporative Control System
= 5 : FCV System
— 6 : Gas Cap
ATPFEE Block Anti-tampering inspection fee
BYEAR Block Fleet size and growth base year default - 1986
CSTIH Block Inspection Maintenance I/M inspection fee
FFACT Block Fraction of annual fuel economy benefit achieved
by biennial I/M (based on HC emission reduction)
GASCST Block Cost of one gallon of unleaded gasoline
IAFEFG Block Flag indicating whether user is inputting
optional ATP fees and costs or using the defaults
1 = defaults 2 = user inputs
IAFXFG Block Flag indicating whether the ATP has a fixed or
floating model year coverage
lAIiDV Local Variable indicating which vehicle classes are
to be inspected by ATP. Must be the same as
those inspected by I/M.
IATPLP Block Flag indicating whether an ATP only scenario ia
to be run
IFREQ Block Flag indicating whether the I/M program ia
annual or biennial
IGRFLG Block Flag indicating whether user ia inputting the
vehicle fleet distribution and the fleet growth
rate or is using the built-in defaults
ILDT Block Parameter data from MOBILE4 indicating the
number of vehicle classes inspected by I/M
IMCSFG Block Flag indicating whether user ia inputting
optional I/M repair coat and fees or ia using
the defaults
IMFXFG Block Flag indicating whether the I/M program has a
fixed or floating model year coverage
INERR Param MOBILE4 error counter in subroutine QUITER
ISIZE1 Block Variable containing either the aize of the
floating model year window for I/M coverage or
the earliest model year inspected in an I/M
program
ISIZE2 Block Variable containing either the size of the
floating modely year window for I/M coverage or
the earliest model year inspected in an I/M
program
DISTYP Block Array of flags indicating which emission
control components are to be inspected by the ATP
TVTYP Local Vehicle class index for all eight vehicle classes
ATPFQT Block Flag indicating whether ATP program is annual
or biennial
LAP1ST Block Variable containing the oldest model year
covered by an ATP if a fixed model year program
is run
MODYR1 Block Variable containing the oldest model year
covered by I/M if a fixed model year program ia
run
MOVATP Block Number of model years covered in an ATP if a
floating model year program is run
MOVIM Block Number of model years covered in I/M if a.
floating model year program is run
REPEAT Block Fraction of I/M failures which fail in the
subsequent year also
REPIM Block Array containing the I/M repair costs
REPIM(l) - Old Tech ; REPIM(2) - New Tech
REPIM(3) - IM240 ; REPIM(4) = NOx (IM240)
RMAXF Block Maximum I/M failure rate that 1981 and later
vehicles are allowed
47
-------
Appendix A
C TCOUNT Block Vehicle fleet size in base year
C VGRORT Block Array containing the growth rates for each
C vehicle class
C
INTEGER BYEAR,ATPFLG,PROMPT,ATPFQT,DISTYP
C
C COMMON BLOCK DEFINITIONS
C
COMMON /COST01/ ATPFEE
COMMON /COST02/ IFEVYR, ILEVYR
COMMON /COST03/ STRNGY,RMAXF,REPEAT, FFACT
COMMON /COST07/ CSTIM,GASCST,IMSTRT, ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
COMMON /COST22/ ATCOST(2,8)
COMMON /COST27/ IGRFLG, IATPLP, IDIFF,OUTPRF
COMMON /COST34/ TCOUNT,BYEAR, VGRORT
COMMON /COST48/ PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
C
COMMON /COSTXX/ JMYR1,JMYR2
C
COMMON /ATPAR1/ LAPSY,LAP1ST,LAPLST,LVTFLG(4)
COMMON /FLAGSl/ PROMPT, TAMFLG, SPDFLG, VMFLAG, OXYFLG, DSFLAG
COMMON /FLAGS2/ MYMRFG,NEMFLG,IMFLAG,ALHFLG
COMMON /ATPAR2/ ATPPGM, ATPFQT, CRATP, DISTYP (8)
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2,WAIVER<2) ,CRIM
COMMON /IMPAR6/ IFREQ,INTYP
C
CC
MODYR1 - JMYR1
MODYR2 - JMYR2
CC
C I/M program defaults
C
RMAXF - .25
REPEAT - .20
C
C.. FFACX was 0.70 in CEM4, changed 8/14/91
C
FFACT - .90
C
C Default I/M coat assumptions:
C
IF(IMCSFG.EQ.2) GOTO 10
IF(IFREQ.EQ.l) CSTIM - 10.00
IF(IFREQ.EQ.2) CSTIM - 12.00
GASCST - 1.25
REPIM(l) - 50.00
REPIM(2) - 75.00
REPIM(3) - 150.00
REPIM(4) - 100.00
C
C Default Purge/Pressure Inspection costs...
C
PRGFEE - 6.53
PRSFEE - 0.69
C TRNFEE is just increment for IM240 over Purge insp fee. 6/24/91
TRNFEE - 0.67
C
C Default ATP cost assumptions:
C
10 IF
-------
Appendix A
ATPFEE ia the inspection coat for the ATP .
The default assumes any underhood inapection will cost
$1.00 regrardleaa of the number of items inspected, a
lead deposit teat will coat an additional $0.50, and an exterior
vehicle inapection of either catalyst, fuel inlet reatrictor,
or gaa cap will coat an additional $0.25. Max cost of $1.75.
20 IF(IAFEFG.EQ.2)
GOTO 30
ATPFEE -0.0
IF(
*
*
*
*
IF(
*
*
*
IF(
DISTYP(l)
DISTYP(S)
DISTYPJ6)
DISTYP(7)
DISTYP(2)
DISTYP(3)
DISTYP (8)
DISTYP(4)
.EQ.
.EQ.
.EQ.
.EQ.
.EQ.
.EQ.
.EQ.
.EQ.
2
2
2
2 )
2
2
2 )
2 )
.OR.
.OR.
.OR.
.OR.
-OR.
C
C
C
C
C
C
C
C
C
C
C
ATPFEE - 1.00
ATPFEE - ATPFEE + 0.25
ATPFEE = ATPFEE + 0.50
The default fleet distribution and growth rates by vehicle type.
30 IF(IGRFL6.EQ.2) GOTO 45
. . default growth rate used to be 2.0
VGRORT - 0.0
BYEAR
TCOONT
1986
1000000.
This block determines whether a fixed or floating
program ia to be run and transfers the model year
window aize or program start date.
IMFXFG =• 1
IAFXFG - 1
IMFXFG - 2
IAFXFG - 2
Moving Model Year Coverage
Fixed Model Year Coverage
45 IF (IMFXFG. EQ.l) ISIZE1 = MOVIM
IF ( IAFXFG. EQ.l) ISIZE2 - MOVATP
IF (IMFXFG. EQ.2) ISIZE1 - MODYR1
IF (IAFXFG. EQ.2) ISIZE2 - LAP1ST
99 RETURN
End SETUP
END
49
-------
Appendix A
SUBROUTINE INTWDW
C
C.
C
C.
C
C.
C
C.
C.
C.
C.
C.
C.
C.
C.
C.
C.
C.
C.
C.
C
C
C
C.
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
.INTWDW initializes window scenario array elements to zero
.Called by MAIN
C
C
C
C
.Array Subscripta
.CATPFE(25,10, 4)
.CATPRE (25,10, 4)
.CIMFEE(25,10,4)
.CIMREP (25,10, 4)
.COST(25,10,4)
.DETER(25,2,10,4)
.FULBEN(25,10,4)
.REDATP(25,2,10,4)
.REDIM(25,2,10,4)
.TDETER<25,2,10,4)
.TONRED(25,2,10,4)
.TOTCST(2)
.TOTRED (2)
.Name Type
CATPFE (IDX, JCYDX, IV)
CATPRE (IDX, JCYDX, IV)
CIMFEE (IDX, JCYDX, IV)
CIMREP (IDX, JCYDX, IV)
COST(IDX,JCYDX,IV)
DETER (IDX, IP, JCYDX, IV)
FULBEN (IDX, JCYDX, IV)
REDATP (IDX, IP, JCYDX, IV)
REDIM (IDX, IP, JCYDX, IV)
TDETER (IDX, IP , JCYDX, IV)
TONRED (IDX, IP, JCYDX, IV)
TOTCST(IP)
TOTRED(IP)
Dictionary
CATPFE Block Array which holds and sums the ATP fees for each
model year, calendar year and vehicle type
CATPRE Block Array which holds and sums the ATP repair costs
for each model year, calendar year and vehicle type
CIMFEE Block Array which holds and sums the I/M fees for each
model year, calendar year and vehicle type
CIMREP Block Array which holds and sums the I/M repair coats
for each model year, calendar year and vehicle type
COST Block Array which holds and sums the total cost
for each model year, calendar year and vehicle type
DETER Block Array which holds and sums the deterence benefit
for each model year, calendar year and vehicle type
FULBEN Block Array which holda and sums the fuel economy benefit
for each model year, calendar year and vehicle type
ICYDX Local Calendar year index 1 - 1975
IP Local Pollutant index 1 - HC 2 - CO
IV Local Vehicle class index
MYDX Local Model year index 1 - 1950
REDATP Block Array which holds and sums the ATP benefit for each
model year, calendar year and vehicle type
REDIM Block Array which holds and sums the I/M benefit for each
model year, calendar year and vehicle type
TCOST Block Array which passes the total coat of control programs
TDETER Block Array which passes the total deterence benefit
TONRED Block Array which passes the total benefit
TOTCST Block Array which passes the total cost of control programs
COMMON /COST10/ FEBNFT(25,10,4),FECNMY(20,4)
COMMON /COST13/ CIMFEE(25,10,4),CIMREP(25,10,4)
COMMON /COST14/ TONRED(25,2,10,4),COST(25,10,4)
COMMON /COST15/ TOTCST(2),TOTRED(2),TCOST
COMMON /COST16/ DATPFE, DIMFEE, DIMREP, DATPRE, AATPFE, AIMFEE
COMMON /COST17/ DETER(25,2,10,4),TDETER(25,2,10,4)
COMMON /COST18 / AIMREP , AATPRE , CFACTR, CATPFE (25,10,4)
COMMON /COST21/ CATPRE(25,10,4),FULBEN(25,20,4)
COMMON /COST20/ TDUCT(25,2,3,10,4),TEVAP(25,10,4)
COMMON /COST46/ REDIM(25,2,10,4),REDATP(25,2,10,4)
COMMON /COST47/ PPFAIL(25,3,10,4),PPFIXR(25,10,4)
COMMON /COST49/ PPFEB(25,10,4),APFBEN
COMMON /COST50/ TRLOSS(25,4,3),TREVAP(25,4,3)
COMMON /COST50/ TRLOSS(25,4,3),TREVAP(25,4,3),TRSTLS(25,4,3)
COMMON /MAXIMA/ MAXVEH, MAXLTW, MAXPOL, MAXREG, MAXYRS
TCOST-0.0
DO 10 IP=1,2
TOTCST(IP)=0.0
TOTRED(IP)-0.0
10 CONTINUE
50
-------
Appendix A
DO 20 IV-1,4
DO 20 JCYDX-1,10
DO 20 IDX - 1,MAXYRS
COST(IDX,JCTDX,IV)-0.0
CIMFEE(IDX,JCYDX,IV)-0.0
CATPFE(IDX,JCYDX,IV)-0.0
CIMREP(IDX,JCYDX,IV)-0.0
CATPRE(IDX,JCYDX,IV)-0.0
POLBEN(IDX,JCYDX,IV)-0.0
FEBNFT (IDX, JCYDX, IV) -0 . 0
PPFEB(IDX,JCYDX,IV)-0.0
TEVAP(IDX,JCYDX,IV)-0.0
DO 20 IP-1,2
TOMBED(IDX,IP,JCYDX,IV)-0.0
BEDIM(IDX,IP,JCYDX,IV) -0.0
REDATP(IDX,IP,JCYDX,IV)-0.0
DETER(IDX,IP,JCYDX,IV) -0.0
TDETER(IDX,IP,JCYDX,IV)-0.0
DO 20 IMC-1,3
TDOCT(IDX,IP,IMC,JCYDX,IV)-0.0
C TRSTLS(IDX,IV,IMC)-0.0
TREVAP(IDX,IV,IMC)-0.0
TRLOS8(IDX,IV,IMC)-0.0
20 CONTINUE
C
RETURN
C End INTHDW
END
51
-------
Appendix A
c
c
c.
c.
c.
c
c.
c
c.
c
c.
c.
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
SUBROUTINE FLEET
.FLEET determines the number of vehicles in each vehicle
.class in each calendar year. FLEET also sums the total
.number of vehicles in each calendar year.
.Called by MAIN
.Array Subscripts
.STOCK(12,8)
.VCNT (46, 8)
.VGRORT(8)
.Variable Dictionary
- STOCK(JBY,IV)
- VCNT(IY,IV)
- VGRORT(IV)
Name
Source Description
C
C
C
C
C
C
C
C
ACOUNT Faram
BCOUNT
BYEAR
CLDIST
ICYDX
rv
JEXP
NCOUNT
TCOUNT
VCNT
VGRORT
YCOUNT
Local
Par am
Block
Local
Local
Local
Block
Block
Block
Block
Block
The average number of vehicles over all the
calendar years selected in the scenario records.
Sum of MOBILE VCOUNT for 1st eval year
The fleet size base year
Vehicle claas registration distribution
The absolute calendar year index 1 - 1975
Vehicle class index
Exponent index
Desired vehicle fleet size in 1st evaluation year
Vehicle fleet size in base year
Number of vehicles of each type in each
calendar year
Array containing the fleet growth rates for each
vehicle claas
The total number of vehicles in a given calendar
year
INTEGER BYEAR
COMMON /COST02/ IFEVYR, ILEVYR
COMMON /COST23/ VCNT(39,8),GSDSCT(8,10),HDDCUM(25,10)
COMMON /COST34/ TCOUNT,BYEAR,VGRORT
COMMON /COST37/ YRCOST(10),YRBEN(10,2),YCOUNT<39),ACOUNT
COMMON /REGISF/ GSFVCT(8),USRGSF(25,2)
COMMON /VMXCOM/ REGMIX (8) , TFNORM (8) , VMTMIX (8) , VCOUNT (39, 8)
DIMENSION OLDCNT (39)
Determine base year index
IBY - BYEAR - 1981
Adjust YCOUNT from given base year to current IFEVYR...
Leave TCOUNT as the base year count
and use YCOUNT as the new count for current eval year
DO 20 ICYDX - 1,39
YCOUNT(ICYDX) - TCOUNT
IF(VGRORT.NE.0.0 .AND. ICYDX.NE.IBY) THEN
IF(ICYDX.LT.IBY) THEN
JEXP - IBY - ICYDX
YCOUNT (ICYDX) - TCOUNT / ((1.0+ (VGRORT/100 .) ) ** JEXP)
ELSE IF (ICYDX.GT.IBY) THEN
JEXP = ICYDX - IBY
YCOUNT(ICYDX) - TCOUNT * ((1.0+(VGRORT/100.))**JEXP)
ENDIF
ENDIF
OLDCNT(ICYDX) - 0.0
DO 10 IV-1,8
VCNT(ICYDX,IV) - VCOUNT(ICYDX,IV) * GSFVCT(IV)
OLDCNT (ICYDX) - OLDCNT (ICYDX) + VCNT (ICYDX, IV)
10 CONTINUE
52
-------
Appendix A
G normalize VCNT by Desired YCOUNT/(Old VCOUNT from M.41)
C
DO 50 ICYDX - 1,39
ADJCNT - YCOONT(ICYDX)/OLDCHT(ICYDX)
DO 40 IV - 1,8
VCNT (ICYDX, IV) - VCHT (ICYDX, TV) *ADJCNT
40 CONTINUE
50 CONTINUE
C
C Sum up YCOUNT and ACOUNT...
C
105 IFICY - IFEVYR - 1981
ILICY = ILEVYR - 1981
C
ACOUNT-0.0
DO 120 ICYDX=IFICY,ILICY
ACOUNT - ACOUNT + YCOUNT(ICYDX)
120 CONTINUE
C
ACOUNT = ACOUNT / ( ILEVYR - IFEVYR 4- 1 )
C
999 RETURN
C End FLEET
END
53
-------
Appendix A
SUBROUTINE TEASE
c
C..TBASE calculate the non-program emission levels for each
C . . calendar year .
C
C. .Called by MAIN
C
C . . Calla CAHIL
C
C. .Array Subscripts :
C
C. .YTONS(2,10)
C. . JULMYR(25,8)
C. .BTONS(2)
C. .VCNT(39,8)
C
- YTONS (IP, JCYDX)
- JULMYR (JDX, IV)
- BTONS (IP)
- VCNT( ICYDX, IV)
C. .Variable Dictionary:
C
C. . Name Source
C ACCMIL Param
C
C BTONS Block
C
C EFALL Block
C ICYDX Local
C IDX Local
C IFEVYR Local
C IFICY Local
C ILEVYR Local
C ILICY Local
C IP Local
C IV Param
C JDX Param
C
C JULMYR Block
C
C MY Local
C MYDX Local
C NUMCY Local
C VCHT Block
C
C VMT Param
C YTONS Block
C
C
Description
The average accumulated mileage in a one year
period.
Average benefits over all evaluated calendar
years .
Emission factors from MOBILE4.
Absolute Calendar year index (1975-1) .
Forward model year index (ascending model years)
First calendar year evaluated.
First evaluated calendar year index (1975-1) .
Last calendar year evaluated.
Last evaluated calendar year index (1975 -1) .
Pollutant type index (1:HC, 2: CO, 3:NOx) .
Vehicle type index.
Backward model year index (descending model
year) .
Fraction of vehicles in each model year from
MOBILE4 .
Model year.
Absolute model year index (1955—1) .
The number o£ calendar years being evaluated.
Total number of vehicles in a vehilce class in a
calendar year.
Total vehicle miles traveled in a calendar year.
Total benefits in each evaluted calendar year.
REAL JULMYR
COMMON /COST02/ IFEVYR,ILEVYR
COMMON /COST04/ TAMPER(25,6,4,10,4)
COMMON /COST23/ VCNT'(39, 8) ,SSDSCT (8,10) ,HDDCUM(25,10)
COMMON /COST25/ FRLOSS(25,4,4,10),FREF(25,4,4,10)
C. . .
COMMON /COST26/ FRSTLS(25,8,4,10),YRSTLS(10),BRSTLS
COMMON /COST27/ IGRFLG,IATPLP, IDIFF, OUTPRF
C
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, CPR6YR, JPRSYR
C Add AEVAP 6/3/91
COMMON /COST36/ YTONS(2,10),BTONS(2),BIDTON(2),AEVAP
COMMON /COST30/ EVAP(25,8,4,10),FER(3,25,8,4,10)
COMMON /COST40/ YEVTON(10),BEVTON,YEREF(10),BEREF
COMMON /COST41/ YERLOS(10),BERLOS
COMMON /CUMCOM/ CUMMIL(25,8)
COMMON /MYRSAV/ AMAR(25, 8) , JULMYR(25, 8) ,NEWCUM
COMMON /MAXIMA/ MAXVEH,MAXLTW,MAXPOL,MAXREGrMAXYRS
COMMON /REGISF/ GSFVCT (8) , USRGSF (25, 2)
C
DIMENSION YHCFTP(10)
C
C IFICY = IFEVYR - 1974
IFICY - IFEVYR - 1981
C IDIFF - ILEVYR - IFEVYR + 1
C
C Compute the number of calendar years analyzed.
C
C
DO 5 IPP - 1,2
BTORS(IPP) =0.0
BIDTON(IPP) =0.0
5 CONTINUE
54
-------
Appendix A
BEVTON - o.o
BEREF =0.0
BERIiOS -0.0
C. . .
BRSTLS =0.0
BFTP — 0.0
C
DO 30 JCYDX - 1,IDIFF
C
C ICYDX - (IFEVYR-1974) + JCYDX - 1
ICYDX - (IFEVYR-1981) + JCYDX - 1
C
YTONS(1,JCYDX) - 0.0
YTONS(2,JCYDX) - 0.0
YEVTON (JCYDX) - 0.0
YEREF (JCYDX) - 0.0
YERLOS(JCYDX) =0.0
YRSTLS(JCYDX) - 0.0
C... add 8/12/91...
YHCFTP(JCYDX) =0.0
C
DO 20 IV = 1,8
C. . .
DO 10 IDX =• 1,MAXYRS
JDX = MAXYRS + 1 - IDX
C MYDX = MAXYRS + ICYDX - JDX
C MY MYDX + 1950
C
C Determine the vehicle miles traveled (VMT)
C VCNT now already has GSFVCT incorporated into it...
C
CALL CALMIL(ODX,IV,ACCMIL,VMT)
C
VADJ = VCNT (ICYDX, IV) *JOLMYR(JDX, IV) *1.1025E-6*.001*VMT
C
TMP-VADJ* (EVAP (IDX, IV, 1, JCYDX) +EVAP (IDX, IV, 1, JCYDX+1) ) /2. 0
YEVTON (JCYDX) = YEVTON (JCYDX) + IMP
C * (EVAP (IDX, IV, 1, JCYDX) +EVAP (IDX, IV, 1, JCYDX+1) ) /2 *
C * VADJ
BEVTON = BEVTON + IMP
C * (EVAP (IDX, IV, 1, JCYDX) +EVAP (IDX, IV, 1, JCYDX+1) ) /2 *
C * VADJ
C. . .
TMP=VADJ* (FRSTLS (IDX, IV, 1, JCYDX) +FRSTLS (IDX, IV, 1, JCYDX+1) ) /2 . 0
YRSTLS (JCYDX) - YRSTLS (JCYDX) + IMP
C
BRSTLS = BRSTLS + TMP
C
TMP-VADJ* (FER (1, IDX, IV, 1, JCYDX) +FER(1, IDX, IV, 1, JCYDX+1) ) /2 . 0
YHCFTP (JCYDX) = YHCFTP (JCYDX) + TMP
C
BFTP = BFTP + TMP
C
TMP-VADJ*(FER(2,IDX,IV,1,JCYDX)+FER(2,IDX,IV,1,JCYDX+1))/2.0
YTONS(2, JCYDX)- YTONS(2,JCYDX) + TMP
C
BTONS(2) - BTONS(2) + TMP
C
IF(IV.LE.4) THEN
TMP-VADJ* (FREF (IDX, IV, 1, JCYDX) +FREF (IDX, IV, 1, JCYDX+1)) /2 . 0
YEREF(JCYDX) - YEREF (JCYDX) + TMP
C
BEREF - BEREF + TMP
C
TMP-VADJ* (FRLOSS (IDX, IV, 1, JCYDX) +FRLOSS (IDX, IV, 1, JCYDX+1) ) /2 . 0
YERLOS (JCYDX) - YERLOS (JCYDX) + TMP
C
BERLOS = BERLOS + TMP
C
ENDIF
C
10 CONTINUE
C
20 CONTINUE
C. . .
YTONS(1,JCYDX) - YHCFTP(JCYDX)
* + YEVTON (JCYDX) + YEREF (JCYDX)
* + YERLOS(JCYDX) + YRSTLS(JCYDX)
C
30 CONTINUE
C
55
-------
Appendix A
BTONS(l) = BFTP + BEVTON + BEREF + BERLOS + BRSTLS
DO 40 1=1,2
BTONS(I) - (BTONS(I) / IDIFF)
40 CONTINUE
BEVTON = BEVTON / IDIPF
BEREF = BEREF / XDIFF
BERLOS = BERLOS / IDIFF
BRSTLS - BRSTLS / IDIFF
BFTP - BFTP / IDIFF
C
C
99 RETURN
C End TBASE
END
56
-------
Appendix A
c
c.,
c.,
c.,
c.,
c.
c.
c.
c.
c
c.
c.
c.
c.
c.
c
c.
c
c.
c
c.
c
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
SUBROUTINE CPFUEL
CPFUEL computes the fuel economy benefits due to car
repairs. This routine is called only if an I/M
program is selected. If this ia the case, this routine
calculates the fuel savings accumulated from I/M
repairs over the life of the I/M program. It is assumed
that a given vehicle receives only one fuel economy
benefit over its entire life and that once received
this benefit ia permanent.
Because this subroutine may have the largest calendar
year coverage (the ICYPX looping begins at the I/M
start year not the first scenario year (IFEVYR)), the
I/M failure rate calculation ia performed in this
, subroutine.
Called by MAIN
.Calls CALMIL
-Array Subscripts:
DRSTRN(3)
,FAILRT(25,10,4)
,FEBNFT(25,10,4)
.FECNMY(20,4)
.JULMYR(25,8)
,NEWSTR(4,100)
,PCNTSV(4)
,PPFEB(25,10,4)
.WAIVER (2)
, ZMSTRN(3)
.Variable Dictionary
- DRSTRN (ITEST)
- FAILRT(IDX,JCYDX,IV)
- FEBNFT(IDX,ICYDX,IV)
- FECNMY (IDX, IV)
- JULMYR(IDX,IV)
- NEWSTR(4, IM)
- PCNTSV(ITECH+IM24+P/P)
- PPFEB(IDX,ICYDX,IV)
- WAIVER (ITECH)
- ZMSTRN(ITEST)
Name
Source Description
ACCMIL Param Accumulated mileage at a given age for each
vehicle class
DRSTRN Block Growth in the I/M failure as a function of
mileage for new tech vehicles
FAIL Block Calculated I/M failure rate (1-HC/CO, 2-NOx)
FAIL1 Local I/M failure rate for next year
FAIL2 Local I/M failure rate for current year among cars
which were OK after last year's repairs
FEBNFT Block Final fuel economy benefit for exhaust repairs
FECARS Local Number of cars for which fuel economy credits
should be claimed
FECNMY Block Fuel economy of each vehicle type for each
model year in miles per gallon
FFACT Block Fraction of annual fuel economy benefit achieved
by biennial I/M (based on HC emission reduction)
GASCST Block Cost of one gallon of unleaded gasoline
ICUTS Block I/M outpoint index from MOBILE4
ICYDX Local Calendar year index 1 - 1975
ICYDX2 Local Subsequent calendar year index ICYDX2-ICYDX + 1
ICY Local Calendar year
IDX Local Model year index (forward counter 1 - 25)
IFEVYR Block First calendar year evaluated
IFICY Local First calendar year index
IFREQ Block Flag indicating whether the I/M program is
annual or biennial
ILEVYR Block Last calendar year evaluated
ILICY Local Last calendar year index
IMICY Local Calendar year index in which I/M starts.
Only used if I/M atart precedes first calendar
evaluated.
ISTRIN Block Number of input stringency groups
ISTRT Block I/M start year
ITECH Block Technology group 1 = old tech , 2 - new tech
Old - myr 1951-1980 LDGV , myr 1951-1983 LDT's
New - myr 1981+ LDGV , myr 1984+ LDT's
ITEST Block I/M test type from MOBILE4
IV Local Vehicle class index
JDX Local Model year index (backwards counter 20 - 1)
MY Local Model year
MYDX Local Model year index
NEWSTR Block Array containing information read from the
stringency records.
NEWSTR(1,IM) 1 = Vehicle class
57
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
NEWSTR(2,IM) 2 - First model year in group
NEWSTR(3,IM) 3 - Last model year in group
NEWSTR(4,IM) 4 - Group's stringency
NMYF Local Firat model year in group (from NEWSTR(2,IM)
NMYLL Local Last model year in group (from NEWSTR(3,IM)
NRPEAT Local Cars which are not part of the REPEAT failure
category
PCNTSV Block Average fuel savings by technology group
PPCARS Local Number of cars for which Purge/Pressure repair
fuel economy credits should be claimed
PPFEB Block Fuel economy benefit for Purge/Press repairs
PREVCY Block Values used in calculating the previous
calendar year's fuel economy cars. Used to
calculate next year's cars
PREVNF Block Cars repaired in previous years which have not
yet failed again
REPEAT Block Fraction of I/M failures which fail in the
subsequent year also.
RMAXF Block Maximum failure rate that 1981 and later
vehicles are allowed
VMT Param Vehicle miles travelled in a given year
WAIVER Block Array containing the waiver rates for old
and new technology vehicles
ZMSTRN Block Failure rate of new technology vehicles at
zero miles.
NOTES: EVPFEB added 7/11/91 could be done away with and the
results simply added in with FEBNFT if we didn't want
to see it separately in the printout.
9/8/91 Add NOx to FAIL array FAILRT(IPOL,ODX,ICYDX,IV), where
IPOL: HC,CO-1, NOx-2
COMMON /COST01/ ATPFEE
COMMON /COST02/ IFEVYR,ILEVYR
COMMON /COST03/ STRNGY,RMAXF,REPEAT,FFACT
COMMON /COST05/ REPLCE (25, 6,10, 4)
COMMON /COST07/ CSTIM, GASCST, IMSTRT, ISCEN, MOVTM,MOVATP, REPIM (4)
COMMON /COST08/ FAILRT(2,25,39,4)
COMMON /COSTO 9/ DRSTRN(5 ) ,ZMSTRN(5)
COMMON /COST10/ FEBNFT(25,10,4),FECNMY(20,4)
COMMON /COST11/ PCNTSV(4)
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, OPRGYR, JFRSYR
COMMON /COST47/ PPFAIL(25,3,10,4),PPFIXR(25,10,4)
COMMON /COST49/ PPFEB(25,10,4),APFBEN
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER(2) ,CRIM
COMMON /IMPAR2/ ILDT (4) , ITEST,NUDATA(2) ,NLIM, IMNAME (20, 9)
COMMON /IMPAR6/ IFREQ,INTYP
COMMON /IM240P/ DSIZE(25,4),IM24YR,IPRGYR,IPRSYR
COMMON /MAXIMA/ MAXVEH, MAXLTW, MAXPOL,MAXREG,MAXYRS
IFICY - IFEVYR - 1981
ILICY - ILEVYR - 1981
IMICY - ISTRT - 1981
The IMICY allows this subroutine to loop for calendar
earlier than the first calendar evaluated (IFEVYR.) .
ISICY - IMICY
IF((ILICY-IMICY).GT.19) ISICY
ILICY - 19
DO 52 IV-1,4
IF(ILDT(IV) .NE. 2) GOTO 52
DO 51 ICYDX-ISICY, ILICY
JCYDX-ICYDX-IFICY+1
ICYDX2-ICYDX+1
ICY - 1981 + ICYDX
DO 50 IDX = 1,MAXYRS
JDX = MAXYRS + 1 - IDX
MYDX - MAXYRS + ICYDX - JDX
MY - MYDX + 1957
Get vehicle miles traveled (VMT) that year
CALL CALMIL(JDX,IV,ACCMIL,VMT)
Determine technology group
ITECH-1
58
-------
Appendix A
IF(IV.EQ.LAND.MY.GE.1981) ITECH=2
IF(IV.GT.LAND.MY.GE.1984) ITECH-2
C
C Compute IM failure rate
C
IF(ITECH .EQ. 2) GOTO 20
C
IF(MY.GE.MODYR1.AND.MY.LE.MODYR2.AND.IHFLAG.GT.1)
* FAILRT(1,IDX, ICYDX, IV) - ISTRIN/100.0
C
C Add separate IM240 failure rates, 6/6/91, S adjusted for time
C by factor of 0.01/0.0187... 0.0199 put into DRSTRN(4) 9/9/91
C 0.4/15 t 0.0750 ... 0.0401
C 0.8/15 : 0.0373 ... 0.0199
C Revise 12/10/91, to use half of flat zero mile
C and half of sloped DR.
C 0.8/30 : 0.0324 ... 0.0173
C 1.6/30 t 0.0148 ... 0.0079
C
20 IF(ITECH .EQ. 2) THEN
IF (MY .GE. IM24YR) THEN
C FAILRT (1, IDX, ICYDX, IV) - ZMSTRN(4)
C * + DRSTRN (4) *ACCMIL*0. 0001
C 12/10/91...
FAILRT(1,IDX,ICYDX,IV) -
* DRSTRtt(4)/2.0 + (DRSTRN(4) / (2.0*1. 87))*ACCMIL*0.0001
FAILRT(2,IDX, ICYDX, IV) - ZMSTRN(S)
* + DRSTRN (5) *ACCMIL*0. 0001
ELSE
FAILRT(1,IDX,ICYDX, IV) -
* ZMSTRN (ITEST) +(DRSTRN(ITEST) * (ACCMIL*. 0001) )
FAILRT (2, IDX, ICYDX, IV) - 0.0
ENDIF
IF (FAILRT (1, IDX, ICYDX, IV) . GT .RMAXF)
* FAILRT (1, IDX, ICYDX, IV) -RMAXF
FAILRT(2,IDX,ICYDX,IV) - 0.0
ENDIF
C
C Add check for negative failure rate, 6/6/91 ...
C
IF (FAILRT (1, IDX, ICYDX, IV) .LT. 0.0) FAILRT (1, IDX, ICYDX, IV) =0.0
C
C FAIL1 is used only for FE benefit calcs
C
IF(ITECH.EQ.2 .AND. MY.GE.IM24YR) THEN
FAIL1 - ZMSTRN(4) + DRSTRN (4) *ACCMIL*0 . 0001
ELSE
FAIL1 - FAILRT(1,IDX,ICYDX,IV)*1.87
ENDIF
C
C Removed all code regarding NRPEAT, FAIL2, PREVCY £ PREVNF since
C we've been using different FE calc method since 6/19/91, 12/10/91...
C
C Calculate the number of cars to get fuel economy benefit...
C
FECARS - FAIL1*ENFORC(CRIM,1) * (1. 0-WAIVER (ITECH) )
IF (FECARS. GT. 1. ) FECARS-1. 0
C
C Calculate annual fuel economy benefits for a vehicle in
C a given model year.
C
IDXP1-MY-1967
IF(IDXPl.LE.O) IDXP1=1
IF(IDXP1.GE.20) IDXP1=20
C
C Calculate fuel economy benefit in each calendar year.
C
IF(ICYDX.GE.IFICY) THEN
IF(ITECH.EQ.2 .AND. MY.GE.IM24YR) THEN
FEBNFT (IDX, JCYDX, IV) =
* (PCNTSV (3) *GASCST*VMT/FECNMY (IDXP1, IV) ) *FECARS
ELSE
FEBNFT(IDX,JCYDX,IV) =
* (PCNTSV (ITECH) *GASCST*VMT/FECNMY (IDXP1, IV) ) *FECARS
ENDIF
C ENDIF
C
59
-------
Appendix A
C Add FE benefit calc for Purge/Pressure Repairs 7/11/91...
C PPFIXR already includea ENFORC(CRIM,1), but not WAIVER
C since no waivers are expected for P/P repairs.
C
IF(MY.GE. JPRSYR .OR. MY. GE. JPRGYR) THEN
C PECARS - PPFIXR (IDX, JCYDX, IV)
PFFEB(IDX,JCYDX,IV) -
* PCNTSV(4)*GASCST*VMT/FECNMY(IDXP1,IV)*
* PSFIXR (IDX, JCYDX, IV)
EHDIF
ENDIF
C
C Adjust annual FE benefit if the inspection is biennial.
C
IF
-------
Appendix A
SUBROUTINE CALMIL (JDX, XV, ACCMIL, VMT)
C
C..CALMIL calculates the mileage statistics for each model year in
C..each vehicle class in each calendar year.
C
C..Called by CMPUTE,CTFUEL,SMCOST,TBASE
C
C..Calls None
C
C..Array Subscripts:
C
C..CUMMIL(25,8) - CUMMIL(JDX,IV)
C
C..Variable Dictionary:
C
C.. Name Source Description
C ~—— •• •»»—™~ _WB«.»«._~«.MWWa»_~.u__.H>»~_«»~H.MM _.».»«__ —— _~—««
C ACCMIL Param The average accumulated mileage in a one year
C period.
C CUMMIL Block The average accumulated mileage at a given
C vehicle age.
C IV Param Vehicle type index.
C JDX Param Backward model year index (decending model
C- years).
C VMT Param Total vehicle miles traveled in a calendar
C year.
C
C
COMMON /COST23/ VCNT(39,8),GSDSCT(8,10),HDDCUM(2S, 10)
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /CUMCOM/ CUMMIL(25,8)
COMMON /MAXIMA/ MAXVEH,MAXLTW, MAXPOL,MAXREG, MAXYRS
C
IF(JDX.NE.25) ACCMIL- (CUMMIL (JDX, IV) +CUMMIL( JDX+1, IV) ) /2. 0
IF (JDX.EQ. 25) ACCMIL = CUMMIL (JDX, IV)
C
IF(JDX.NE.25) VMT - CUMMIL (JDX+1, IV) - CUMMIL( JDX, IV)
IFJJDX.EQ.25) VMT - CUMMIL(JDX,IV) - CUMMIL(JDX-1,IV)
C
C
RETURN
C End CALMIL
END
61
-------
Appendix A
SUBROUTINE LOOP(ISIZE1,ISIZE2)
c
c.
c.
c.
c.
c
c.
c
c.
c
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
.LOOP does the vehicle type, calendar year, model year and
. pol lut ant 1 ooping
Inside theae loopa the subroutines
.COVER, CMPUTE, CMPREP, POLL, REDUC, and SMCOST are
.called.
.Called by MAIN
.Calls COVER, CMPUTE, POLL, CMPREP and SMCOST
.Array Subacriptea
.Variable Dictionary
. Name Source
IATPLP Block
ICOV1 Local
ICOV2 Local
ICY Local
ICYDX Local
IDX Local
IFEVYR Block
IFICY Local
ILEVYR Block
ILICY Local
INDVLl Local
INDVL2 Local
IP Local
ISIZE1 Par am
ISIZE2 Par am
ISTRT Block
IV Local
JDX Local
LAPSY Block
m Local
MYDX Local
INTEGER PROMPT
Description
Flag indicating whether an ATP only scenario is
to be run
Variable indicating whether a given calendar year
is covered by I/M
Variable indicating whether a given calendar year
is covered by an ATP
Calendar year
Calendar year index
Model year index (forward 1-25)
First calendar year evaluated
Index of first calendar year evaluated
Last calendar year evaluated
Index of last calendar year evaluated
Variable indicating whether a given model year
is covered by I/M
Variable indicating whether a given model year
is covered by an ATP
Pollutant index
Variable containing either the size of the
floating model year window for I/M coverage or
the earliest model year inspected in an I/M
program.
Variable containing either the size of the
floating model year window for ATP coverage
or the earliest model year inspected in an ATP .
I/M program start year
Vehicle class index
Model year index (backwards counter 20 - 1)
ATP start year
Model year
Model year index
C
c
c
c
c
COMMON /COST02/ IFEVYR, ILEVYR
COMMON /COST07/ CSTIM,<3ASCST, IMSTRT, ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /IMPAR1/ ICYIM, ISTRIN,MODYR1,MODYR2, WAIVER(2) , CRIM
COMMON /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
COMMON /COST19/ ICOV1, ICOV2, INDVLl, INDVL2
COMMON /COST27/ IGRFLG, IATPLP, IDIFF,OUTPRF
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /ATPARl/ LAPSY,LAP1ST,LAPLST,LVTFLG(4)
COMMON /FLAGS1/ PROMPT, TAMFLG, SPDFLG, VMFLAG, OXYFLG, DSFLAG
COMMON /MAXIMA/ MAXVEH, MAXLTW, MAXPOL, MAXREG, MAXYRS
IDIFF - ILEVYR - IFEVYR + 1
IFICY - IFEVYR - 1974
ILICY - ILEVYR - 1974
IFICY - IFEVYR - 1981
ILICY - ILEVYR - 1981
DO 22 IV - 1,4
DO 21 JCYDX - 1,IDIFF
C ICYDX = (IFEVYR-1974) + JCYDX - 1
C ICY - ICYDX + 1974
ICYDX - (IFEVYR-1981) + JCYDX - 1
ICY - ICYDX + 1981
ICOV1 - 0
ICOV2 = 0
IF(IATPLP .EQ. 2) ISTRT - 2020
62
-------
Appendix A
c
IF(ICY.LT.ISTRT .AMD. ICY.LT.LAPSY) GOTO 21
IF(ICY.GE.ISTRT) ICOV1 - 1
IF(ICY.GE.LAPSY) ICOV2 - 1
C
DO 20 IDX - 1,MAXYRS
JDX - MRXYRS + 1 - IDX
MYDX - MAXYRS + ICYDX -JDX
C MY - MYDX + 1950
MY - MYDX + 1957
C
IHDVL1 - 0
IHDVL2 - 0
C
C Determine the vehicles covered by an inspection program.
C
CALL COVER (MYDX, ICY, IV, ISIZE1, ISIZE2)
C
C Compute emission reduction benefits and
C failure rates.
C
DO 10 IP - 1,2
CALL CMEUTE (IV, ICYDX, MYDX, IP, IDX, JDX, JCYDX)
CALL POLL (MYDX, IP, IV, IDX, JCYDX)
10 CONTINUE
C
C Compute overall anti-tampering failure/replacement
C rates.
C
CALL CMPKEP (IV, ICY, MYDX, IDX, JDX, JCYDX)
C
CALL SMCOST (MYDX, ICYDX, IV, JDX, IDX)
C
20 CONTINUE
21 CONTINUE
22 CONTINUE
C
99 RETURN
C End LOOP
END
63
-------
Appendix A
c
c
c
c
c
c
c
c
SUBROUTINE COVER (MYDX, ICY, IV, ISIZE1, ISIZE2)
C
c.
c.
c
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
.COVER determines
. is covered by an
whether a given model year or calendar year
I/M or ATP program.
.Called by Loop
.Variable
. Name
IAFXFG
IATPLP
ICOV1
ICOV2
ICY
ICYDX
IMFXFG
INDVL1
INDVL2
ISIZE1
ISIZE2
ISTRT
LAPSY
MY
MYDX
Dictionary
Source
Block
Block
Local
Local
Local
Local
Block
Local
Local
Par am
Far am
Block
Block
Local
Local
Deacription
Flag determining whether the ATP has a fixed or
floating model year coverage
Flag indicating whether an ATP only scenario is
to be run
Variable indicating whether a given calendar year
is covered by I/M
Variable indicating whether a given calendar year
is covered by an ATP
Calendar year
Calendar year index
Flag indicating whether the I/M program has a
fixed or floating model year coverage
Variable indicating whether a given model year
is covered by I/M
Variable indicating whether a given model year
is covered by an ATP
Variable containing either the size of the
floating model year window for I/M coverage or
the earliest model year inspected in an I/M
program.
Variable containing either the size of the
floating model year window for ATP coverage
or the earliest model year inspected in an ATP .
I/M program start year
ATP start year
Model year
Model year index
INTEGER PROMPT
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
/COST05/
/COST07/
/COST12/
/COST15/
/COST19/
/COST27/
/COST28/
/ATPAR1/
/FLAGSl/
/IMPAR2/
REPLCE(25,6,10,4)
CSTIM, GASCST, IMSTRT, ISCEN, MOVIM, MOVATP , REPIM (4)
IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
TOTCST(2),TOTRED(2),TCOST
ICOV1, ICOV2, INDVL1, INDVL2
IGRFLG, IATPLP, IDIFF, OUTPRF
JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
LAPSY,LAP1ST,LAPLST,LVTFLG(4)
PROMPT, TAMFLG, SPDFLG, VMFLAG, OXYFLG, DSFLAG
ILDT (4) , ITEST, NUDATA (2) , NLIM, IMNAME (20, 9)
MY - MYDX + 19S7
IF(IATPLP.EQ.2)
IF ((IMFXFG.EQ.l)
IF((IMFXFG.EQ.2)
GOTO 30
GOTO 30
.AND. (ICOV1.EQ.1)) GOTO 10
.AND. (ICOVl.EQ.l)) GOTO 20
Seta coverage for floating I/M programs.
10 IF (MY.GT.ICY) INDVL1 = 1
IF((MY.LE.ICY) .AND. (MY.GT.(ICY-ISIZE1))) INDVL1=2
IF(MY.EQ.(ICY-ISIZEl)) INDVL1=3
IF«MY.EQ. (ICY-ISIZE1-1) ) .AND. ( (ICY-ISTRT) .GT. 1) )
* INDVL1 = 4
IF((MY.EQ.(ICY-ISIZE1-2)) .AND. ((ICY-ISTRT).GT.2))
* INDVL1 = 5
IF(MY .LE. (ICY-ISIZE1-3) ) INDVL1 - 6
GOTO 30
Sets coverage for fixed I/M programs.
20 IF(MY.GT.ICY)
IF(MY.LE.ICY .AND.
(MY.GE.ISIZE1))
30 IF(IArPLP.EQ.2)
IF(IAFXFG.EQ.l .AND. ICOV2.EQ.1) GOTO 40
IF(IAFXFG.EQ.2 .AND. ICOV2.EQ.1) GOTO SO
GOTO 98
INDVL1
INDVL1
INDVL1
64
-------
Appendix A
c
C Sets coverage for floating ATPa.
C
40 IF (MT. GT. ICY) INDVL2 - 1
IF((MY.LE.ICY) .AND. (MY.GT.(ICY-ISIZE2))) INDVL2-2
IF(MT.EQ.(ICY-ISIZE2)) INDVL2=3
IF((MY.EQ. (ICY-ISIZE2-1) ) .AND. ( (ICY-LAPSY) . GT. 1) )
* INDVL2 = 4
IF((MY.EQ. (ICY-ISIZE2-2)) .AND. <(ICY-LAESY).GT.2))
* INDVL2 - 5
IF(MY .LE. (ICY-ISIZE2-3)) INDVL2 - 6
GOTO 98
C
C Seta coverage for fixed ATPa.
C
50 IF (MY.GT. ICY) INDVL2 - 1
IF (MY. LE. ICY .AND. (MY.GE. ISIZE2) ) INDVL2 - 2
C
98 IF(ILDT(IV) .EQ.l) INDVL1-1
IF(LVTFLG(IV).EQ.l) INDVL2-1
C
99 RETURN
C End COVER
END
65
-------
Appendix A
SUBROUTINE CMPUTE(IV,ICYDX,MYDX,IP,IDX,JDX,JCYDX)
c
c.
c.
c.
c
c.
c
c.
c
c.
c.
c.
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
. CMPUTE computes
the overall emission reduction
.benefits by computing the differences in the MOBILE 4
. EFALL arraya.
.Called by LOOP
.Array Subscripts
. JULMYR(25,8)
.TDUCT (25, 2, 3, 10,
.VCNT (46, 8)
.WAIVER (2)
- JULMYR (JDX, IV)
4) - TDUCT (25, 2, 3, 10, 4)
- VCNT (ICYDX, IV)
- WAIVER (ITECH)
.Variable dictionary
. Name Source
ICY Local
ICYDX Local
IETYP Local
IMC Local
IMCAS1 Local
IMCAS2 Local
IF Local
I TECH Local
IV Local
JDX Local
JULMYR Block
tfl Local
MYDX Local
Description
Calendar year
Calendar year index
Control program reduction type index
Control program reduction type index
Control program reduction type counter
Second control program reduction type counter
Pollutant index 1 - HC 2 = CO
Technology type index 1 - old tech 2 - new
Vehicle class index
Model year index (backwards 20 - 1)
Distribution of vehicle fleet by model year,
One distribution per vehicle class
Model year
Model year index
tech
by calendar year by vehicle class as a result of
RDUCT1 Local
RDUCT2 Local
RDUCT Local
TDUCT Block
VCNT Block
VMT Par am
WAIVER Block
a control program
Emissions of no control program or lesser
control program
Emissions of greater control program
Difference in emissions between RDUCT1 and
RDUCT2 RDUCT = RDUCT1 - RDUCT2
Array containing final emission factors.
VMT weighting and fleet weighting have been
Number of vehicles of each type in each
calendar year
Vehicle miles travelled in a given calendar
Array containing waiver rates
done
year
C
c
c
c
c
c
c
c
c
c
c
REAL JULMYR
COMMON /COST01/ ATPFEE
COMMON /COST04/ TAMPER(25,6,4,10,4)
COMMON /COST05/ REPLCE(25,6,10,4)
COMMON /COST07/ CSTIM, GASCST, IMSTRT,ISCEN,MOVIM,MOVATP,REPIM (4)
COMMON /COST20/ TDUCT(25,2,3,10,4),TEVAP(25,10,4)
COMMON /COST23/ VCNT(39,8),GSDSCT(8,10),HDDCUM(25,10)
COMMON /COST25/ FRLOSS(25,4,4,10),FREF(25,4,4,10)
COMMON /COST26/ FRSTLS(25,8,4,10),YRSTLS(10),BRSTLS
COMMON /COST50/ TRLOSS(25,4,3),TREVAP(25,4,3)
COMMON /COST50/ TRLOSS(25,4, 3),TREVAP(25,4,3),TRSTLS(25,4,3)
COMMON /COST30/ EVAP(25,8,4,10),FER(3,25,8,4,10)
COMMON /COST40/ YEVTON(10),BEVTON,YEREF(10),BEREF
COMMON /COST41/ YERLOS(10),BERLOS
COMMON /IMPAR1/ ICYIM, ISTRIN, MODYR1, MODYR2, WAIVER (2) , CRIM
COMMON /MYRCAL/ XMYM(25,8),JANMYR(25,8),IF(25,8),TFMYM(25,8)
COMMON /MYRSAV/ AMAR(25, 8) , JULMYR (25, 8) ,NEWCUM
COMMON /REGISF/ GSFVCT(8),USRGSF(25,2)
JCYDX2 = JCYDX + 1
ICY - 1981 4- ICYDX
MY - MYDX + 1957
There are three cases (IMC) for which emission reductions
are calculated:
IMC
1 : I/M tampering deterrence effect.
66
-------
Appendix A
C IMC - 2 : The effect o£ I/M repairs.
C IMC = 3 : The effect of ATP repairs.
C
DO 30 IMC-1,3
IMCAS1-IMC
IMCAS2-IMC+1
C
IF(IDX.HE.1) GOTO 10
C
C Calculate basic emission reductions without losses.
C for last model year index in given window,
C set mid-year emissions equal to Jan 1 emissions,
C otherwise interpolate between the Jan 1st emissions
C from one calendar year to the next for the given
C model year.
C
RDUCT1-FER (IP, IDX, IV, IMCAS1, JCYDX)
RDUCT2-FER (IP, IDX, IV, IMCAS2, JCYDX)
C
REVAPl-EVAP (IDX, IV, IMCAS1, JCYDX)
REVAP2-EVAP (IDX, IV, IMCAS2, JCYDX)
C
RLOSS1-FRLOSS (IDX, IV, IMCASl, JCYDX)
RLOSS2-FRLOSS (IDX, IV, IMCAS2, JCYDX)
C. . .
C RSTLS1-FRSTLS (IDX, IV, IMCASl, JCYDX)
C RSTLS2-FRSTLS (IDX, IV, IMCAS2, JCYDX)
C
GOTO 20
C
C The January 1st EFALL value for a given calendar
C year is averaged with the subsequent year's EFALL
C value to arrive at a July 1st emission value.
C
10 RDUCT1 - (FER(IP,IDX,IV,IMCAS1, JCYDX) +
* FER(IP,IDX,IV,IMCAS1,JCYDX2))/2.0
RDOCT2 - (FER(IP,IDX,IV,IMCAS2,JCYDX) +
* FER(IP,IDX,IV,IMCAS2,JCYDX2))/2.0
C
REVAP1 - (EVAP (IDX, IV, IMCASl, JCYDX) +
* EVAP (IDX, IV, IMCASl, JCYDX2))/2.0
REVAP2 - (EVAP (IDX, IV, IMCAS2, JCYDX) +
* EVAP(IDX,IV,IMCAS2,JCYDX2))/2.0
C
RLOSS1 - (FRLOSS(IDX,IV, IMCASl, JCYDX) +
* FRLOSS (IDX, IV, IMCASl, JCYDX2))/2.0
RLOSS2 - (FRLOSS (IDX, IV, IMCAS2, JCYDX) +
* FRLOSS (IDX, IV, IMCAS2, JCYDX2))/2.0
C
C. . .
C RSTLS1 - (FRSTLS(IDX,IV, IMCASl, JCYDX) +
C * FRSTLS (IDX, IV, IMCASl, JCYDX2))/2.0
C RSTLS2 - (FRSTLS (IDX, IV,IMCAS2, JCYDX) +
C * FRSTLS (IDX, IV, IMCAS2, JCYDX2) )/2.0
C
20 RDUCT - RDOCT1 - RDUCT2
C
C. . .
REVAP - (REVAP1-REVAP2) + (RLOSS1-RLOSS2)
C REVAP - (REVAP1-REVAP2) + (RLOSS1-RLOSS2) + (RSTLS1-RSTLS2)
C
C Get the vehicle miles traveled (VMT)
C
CALL CALMIL(JDX,IV,ACCMIL,VMT)
C
C Change emission reductions to WEIGHT (grams) reductions.
C On a per vehicle basis, not a fleet basis.
C
C
C TDUCT(IDX,IP,IMC, JCYDX, IV)
C represents the I/M or ATP emission's benefit (tonnage) per
C car per year and multiplies it by the appropriate fleet
C size. The units are then converted from tons to grams
C when calculating TDUCT.
C
C Save these 2 separately for debugging...
C
VADJ - VCNT (ICYDX, IV) * JULMYR (JDX, IV) * 1.1025E-6 *
* VMT !* GSFVCT(IV) iGSDSCT (IV, JCYDX)
C
TREVAP(IDX, IV, IMC) = (REVAP1-REVAP2) * VADJ
67
-------
Appendix A
TRLOSS(IDX,IV,IMC) = (RLOSS1-RLOSS2) * VADJ
C
IF(IP.EQ.l) TEVAP(IDX, JCYDX, IV) - REVAP * VADJ
C
IF(IP.EQ.2) REVAP = 0.0
TDOCT(IDX,IP,IMC,JCYDX,IV) - (RDOCT+REVAP) * VADJ
C
30 CONTINUE
C
99 RETURN
C End CMPUTE
END
68
-------
Appendix A
SUBROUTINE POLL (MYDX, IP, IV, IDX, JCYDX)
c
c.
c.
c.
c.
c
c.
c
c.
c
c.
c
c.
c.
c.
c.
c.
c.
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
.POLL trana'fera the
.which are specific
contents from the array TDUCT into other arrays
to I/M, ATP or DETERRENCE. These new arrays are
.subsequently summed by model year, calendar year, pollutant and
.vehicle class in subroutine SUMWDW.
.Called by LOOP
•Calls None
.Array Subscripts:
.CRDTAT(3,2)
.CRDTIM(3,2)
.DETER (25, 2, 10, 4)
.REDATP (25, 2, 10, 4)
.REDIM(25,2,10,4)
.TDUCT(25,2,3,10,4)
. TONRED(25,2,10, 4)
- CRDTAT (IAYRDX, IP)
- CRDTIM(IMYRDX,IP)
- DETER (IDX, IP, JCYDX, IV)
- REDATP (IDX, IP, JCYDX, IV)
- REDIM (IDX, IP, JCYDX, IV)
- TDUCT (IDX, IP, 3, JCYDX, IV)
- TONRED(IDX, IP, JCYDX, IV)
.Variable Dictionary:
. Dame Source
CRDTAT Block
CRDTIM Block
DETER Block
IAYRDX Local
IMYRDX Local
INDVL1 Local
INDVL2 Local
IP Par am
IV Par am
MY Local
MYDX Par am
REDATP Block
REDIM Block
TDUCT Block
TOHRED Block
TONSAT Block
TONSDE Local
TONSIM Local
Description
Fraction of ATP effectiveness for vehicles which
are no longer inspected.
Fraction of I/M effectiveness for vehicles which
no longer inspected.
Array which contains the pollutant reductions for
each model year, calendar year and vehicle type
due to tampering deterence.
Index used in CRDTAT array. Relates INDVL2 to
the model years which are no longer inspected.
Index used in CRDTIM array. Relates INDVL1 to the
model years which are no longer inspected.
Flag determining whether a given vehicle class
with a given model year in a calendar year is
covered by I/M.
Flag determining whether a given vehicle class
with a given model year in a calendar year is
covered by ATP •
Pollutant type index (ascending model years) .
Vehicle type index.
Model year .
Abolute model year index (1950=1) .
Array which contains the pollutant reduction for
each model year, calendar year and vehicle type
due to ATP.
Array which contains the pollutant reductions
for each model year, calendar year and vehicle
type due to ATP.
The emission factor after it has been multiplied
by the fleet size and converted to units of tons.
Array which contains the pollutant reductions for
each model year calendar year and vehicle type
due to all control programs.
Intermediate variable holding the pollutant
reductions due to the ATP.
Intermediate variable holding the pollutant
redutions due to the deterence.
Intermediate variable holding the pollutant
reductions due to the I/M program.
C
c
DIMENSION CRDTAT (3,2), CRDTIM (3,2)
COMMON /COST01/ ATPFEE
COMMON /COST14/ TONRED(25,2,10,4),COST(25,10, 4)
COMMON /COST16/ DATPFE, DIMFEE, DIMREP, DATPRE, AATPFE, AIMFEE
COMMON /COST18/ AIMREP, AATPRE, CFACTR, CATPFE (25,10, 4)
COMMON /COST13/ CIMFEE(25,10,4),CIMREP(25,10, 4)
COMMON /COST21/ CATPRE(25,10,4),FULBEN(25,20,4)
COMMON /COST46/ REDIM(25,2,10,4),REDATP(25, 2,10, 4)
COMMON /COST17/ DETER(25,2,10,4),TDETER<25,2,10, 4)
COMMON /COST19/ ICOV1,ICOV2,INDVL1,INDVL2
COMMON /COST20/ TDUCT(25,2,3,10,4),TEVAP(25,10,4)
COMMON /COST50/ TRLOSS(25,4,3),TREVAP(25, 4, 3)
COMMON /COST50/ TRLOSS(25,4,3),TREVAP(25,4,3),TRSTLS(25,4,3)
69
-------
Appendix A
COMMON /ATPARl/ LAPSY,LAP1ST,LAPLST,LVTFLG(4)
COMMON /ATPAR2/ ATPPGM,ATPFQT,CRATP, DISTYP (8)
C
DATA CRDTIM/.50,.25,.125, .50,.25,.125/
DATA CRDTAT/.75,.50,.25, .75,.50,.25 /
C
C MY - MYDX + 1950
MY - MYDX + 1957
C ICY - ICYDX + 1974
C
TONSDE >0.0
TONSIM -0.0
TONSAT -0.0
C
C Aaaign benefita for vehicle covered by * program.
C
IF(INDVL1.EQ.2) TONSDE - TDUCT(IDX,IP,1,JCYDX,IV)
IF(INDVL1.EQ.2> TONSIM - TDUCT(IDX, IP, 2, JCYDX, XV)
IF(INDVL2.EQ.2) TONSAT - TDUCT(IDX,IP,3,JCYDX,IV)
C
C
IMYRDX - INDVL1 - 2
IAYRDX - INDVL2 - 2
C
IF(INDVLl.GE.3.AND.INDVL1.LT.6)
* TONSDE-TDUCT (IDX, I?, 1, JCYDX, IV) *CRDTIM(IMYRDX, IP)
IF (INDVL1. GE. 3 .AND. INDVLl. LT. 6)
* TONSIM-TDUCT(IDX,IP,2,JCYDX,IV)*CRDTIM(IMYRDX,IP)
IF (INDVL2 . GE. 3 . AND. INDVL2 . LT. 6)
* TONSAT-TDUCT (IDX, IP, 3, JCYDX, IV) *CRDTAT (IAYRDX, IP)
C
IF (INDVLl. GE. 3. AND.INDVL1.LT.6) THEN
TEVAP(IDX,JCYDX,IV) - TEVAP(IDX,JCYDX,IV)*CRDTAT(IMYRDX,IP)
TREVAP (IDX, JCYDX, IV) - TREVAP (IDX, JCYDX, TV) *CRDTAT (IMYRDX, IP)
TRLOSS (IDX, JCYDX, IV) - TRLOSS (IDX, JCYDX, IV) *CRDTAT (IMYRDX, IP)
ENDIF
IF (INDVLl. EQ.l .OR. INDVLl. GE. 6) THEN
TEVAP (IDX, JCYDX, IV) - 0.0
IF(IV.LE.4) THEN
DO 20 IMC-1,3
TREVAP(IDX,IV,IMC) -0.0
TRLOSS (IDX, IV, IMC) -0.0
20 CONTINUE
ENDIF
ENDIF
C
C Add IM tona to ATP tons to gut total.
C
TONRED (IDX, IP, JCYDX, IV) -TONSIM+TONSAT+TONSDE
REDIM (IDX, IP, JCYDX, IV) -TONSIM
REDATP (IDX, IP, JCYDX, IV) -TONSAT
DETER (IDX, IP, JCYDX, IV) -TONSDE
C
RETURN
C End POLL
END
70
-------
Appendix A
SUBROUTINE CMPREP (IV, ICY, MYDX, IDX, JDX, JCYDX)
c
c.
c.
c.
c.
c
c
c
c.
c
c.
c.
c
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
.CMPREP uses the differences in MOBILE 4 I/M,ATP, and no
. control program tampering rates and compute a ATP
.failure/replacements for five different tampering
. situations .
Called by LOOP
.Array Subscript
.REPLCE (25, 5, 10, 4) - REPLCE (IDX,IT, JCYDX,IV)
.TAMPER (25, 6, 4, 10, 4) - TAMPER(MYDX, IT, 3, ICYDX,IV)
.Variable dictionary:
. Name Source Description
ICY Local Calendar year
ICYDX Local Calendar year index
ICYDX2 Local Subsequent calendar year index ICYDX2 = ICYDX+1
IT Local Emission control tampering component index
IT2 Local Tampering component index for DISTYP
DISTYP Block Array of flags indicating which emission
control components are to be inspected by
ATP
IV Local Vehicle class index
LAPSY Block ATP start year
MY Local Model year
MYDX Local Model year index
PBFAC Local Plumbtesmo efficiency factor
RETAIL Local Percentage of vehicles which refail the
the
tampering inspection the following year after
being fixed the previous year
REPLCE Block The difference in tampering rates as a result
of a control program
STOAER Block Condition determining if the Plumbtesmo
efficiency is applied
TAMPER Block The MOBILE4 tampering rates
INTEGER DISTYP
DIMENSION REFAIL (6)
COMMON /COST01/ ATPFEE
COMMON /COST04/ TAMPER(25, 6, 4,10, 4)
COMMON /COST05/ REPLCE (25, 6,10, 4)
COMMON /COST06/ STOAER
COMMON /COST40/ YEVTON (10) ,BEVTON, YEREF (10) ,BEREF
COMMON /COST41/ YERLOS (10) , BERLOS
COMMON /ATPAR1/ LAPSY, LAP1ST,LAPLST,LVTFLG (4)
COMMON /ATPAR2/ ATPP6M,ATPFQT,CRATP, DISTYP (8)
C
c
c
c
c
c
c
c
c
DATA PBFAC / 1.061 /
DATA REFAIL / O.OS, 0.05, 0.10, 0.30, 0.30, 0.30 /
JCYDX2 - JCYDX + 1
IDX2 - IDX - 1
MY - MYDX + 1957
DO 30 IT-1,6
Compute initial year ATP
failure/replacement by substracting non-ATP (IM only)
from IM+ATP.
IF (ICY .GT. LAPSY) GOTO 10
NOTE: the tampering rate for the last model year in each
calendar year is always zero.
REPLCE (IDX, IT, JCYDX, IV) « 0
IF(IDX .EQ. 2) REPLCE(IDX,IT,JCYDX,IV) «
* (TAMPER(IDX, IT, 3, JCYDX, IV)
* - TAMPER(IDX,IT,4, JCYDX,IV) ) *ENFORC (CRATP, 2)
IF (IDX .GT. 2) REPLCE (IDX, IT, JCYDX, IV) =
* (TAMPER (IDX, IT, 3, JCYDX, IV)
* - TAMPER(IDX2, IT, 4, JCYDX2, IV) ) *ENFORC (CRATP, 2)
GOTO 20
71
-------
Appendix A
C Calculate the number of replacements in subsequent years and
C adjust replacements to account for enforcement problems.
C
10 REPLCE(IDX,IT,JCYDX,IV) -
* TAMPER(IDX, IT, 3, JCYDX, IV) *REFAIL(IT) *ENFORC (CRATP, 2)
C
C If a given emission control component is not inspected,
C set the replacement rate to zero.
C
20 IT2 - IT
IF(IT .GT. 3) IT2 = IT + 2
IF(DISTYP(IT2) .NE. 2) REPLCE (IDX, IT, JCYDX, IV) - 0.0
C
C Replacement rates cannot be less than zero
C
IF(REPLCE(IDX,IT,JCYDX,IV).LT.0.)
* REPLCE(IDX,IT,JCYDX,IV) =0,0
C
C 30 CONTINUE
C
C Adjust misfueling replacements to account for lead
C deposit testing failures.
C
IF(STOAER .LT. 1.0 .AND. IT .EQ. 3)
* REPLCE (IDX, 3, JCYDX, IV) -REPLCE (IDX, 3, JCYDX, IV) *PBFAC
C
30 CONTINUE
C
99 RETURN
C End CMPREP
END
72
-------
Appendix A
SUBROUTINE SMCOST (MYDX, ICYDX, IV, JDX, IDX>
C
C. . Calculates and sums the I/M and ATP inspection and repair coats
C
C. .Called by LOOP
C
C. .Calls CAMIL
C
C. .Array Subscripts
C
C. .ATCOST (2, 6)
C. .ATEQP (25, 4, 4)
C. .CATPFE(25,10,4)
C. .CATPRE (25, 10, 4)
C. .CIMFEE (25, 10, 4)
C. .CIMREP (25, 10, 4)
C.. COST (25, 10, 4)
C. .FAILRT(25,20,4)
C. .FEBNFT (25, 20, 4)
C. .FULBEN (25, 20, 4)
C. . JULMYR(25,8)
C. .REPLCE(25,5,10,4)
C. .REFIH(3)
C. .VCNT(46,8)
C
- ATCOST (ITECH, ITAMS)
- ATEQP (MYDX,ITAM4, IV)
- CATPFE (MYDX, ICYDX, IV)
- CATPRE (IDX, JCYDX, IV)
- CIMFEE (IDX, JCYDX, IV)
- CIMREP (IDX, JCYDX, IV)
- COST (IDX, JCYDX, IV)
- FAILRT (IDX, JCYDX, IV)
- FEBNFT (IDX, ICYDX, IV)
- FULBEN (IDX, JCYDX, IV)
- JULMYR(JDX,IV)
- REPLCE (IDX, ITAMS , JCYDX, IV)
- REPIM (ITECH)
- VCNT (ICYDX, IV)
C. .Variable Dictionary
C
C Name Source
C ACCMIL Faram
C
C ATCOST Block
C
C
C
C
C
C
C
C
C
C
C ATEQP Block
C
C
C ATPFEE Block
C CATPFE Block
C
C CATPRE Block
C
C
C CIMFEE Block
C
C CIMREP Block
C
C
C CNSPEC Local
C
C COST Block
C
C CREPAT Local
C CREPIM Local
C CSTIM Block
C FAIL Block
C FAIL2S Local
C FEBNFT Block
C FULBEN Block
C
C
C ICY Local
C ICYDX Local
C INDVL1 Local
C
C INDVL2 Local
C
C ITAM4 Block
C ITAM5 Block
C ITECH Block
C
C
C IV Local
Description
The average accumulated mileage in a one year
period
Cost of replacing emission control devices as a
result of failing a tampering inspection. The
default assumptions are: ATCOST (TECH, TYPE)
TYPE - 1 Air Pump System
- 2 Catalyst Removal
= 3 Misfueled Catalyst
* 4 Evaporative Control System
- 5 PCV System
» 6 Gas Cap
• 7 Purge Test
•* 8 Pressure Test
Array containing the fraction of each model year
in each vehicle class equipped with each
tampering component .
Anti-tampering inspection fee
Array which holds and sums the ATP fees for each
model year, calendar year and vehicle type
Array which holds and sums the ATP repair coats
for each model year, calendar year and vehicle
type
Array which holds and sums the I/M fees for each
model year, calendar year and vehicle type
Array which holds and sums the I/M repair costs
for each model year, calendar year and vehicle
type
Variable which sums and holds the I/M and the
ATP fees
Array which holds and sums the total coat for
each model year, calendar year and vehicle type
Holds the ATP repair coat
Holds the I/M repair coat
I/M inspection fee
Calculated I/M failure rate
Failure rate for 2 speed teat only
Final fuel economy benefit (before summing)
Array which holds and sums the fuel economy
benefit for each model year, calendar year and
vehicle type
Calendar year
Calendar year index
Variable indicating whether a given model year is
covered by I/M
Variable indicating whether a given model year is
covered by ATP
Emission control tampering component index
Emission control tampering component index
Technology group 1 - old tech , 2 = new tech
Old - myr 1951-1980 LDGV , myr 1951-1983 LDT's
New - myr 1981+ LDGV , myr 1984+ LDT's
Vehicle class index
73
-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
Appendix A
JDX Local Model year index (backwards 20-1)
JULMYR Block Fraction of vehicles in each model year from
MOBILE4
MY Local Model year
MYDX Local Model year index
REPLCE Block The difference in tampering rates as a result of
control programs. See ATCOST above for elements
REPIM Block Array containing the I/M repair costs
REPIM(1) - Old Tech ; REPIM(2) - New Tech
REPIM(3) - IM240 ; REPIM(4) - NOx (IM240)
VCNT Block Array of the total number of vehicles in given
calendar years
REAL JULMYR
INTEGER BYEAR,ATPFLG,PROMPT,ATPFQT,DISTYP
COMMON /COST01/
COMMON /COST03/
COMMON /COSTOS/
COMMON /COST07/
COMMON /COST08/
COMMON /COST09/
COMMON /COST10/
COMMON /COST12/
COMMON /COST13/
COMMON /COST14/
COMMON /COST15/
COMMON /COST16/
COMMON /COST18/
COMMON /COST19/
COMMON /COST21/
COMMON /COST22/
COMMON /COST23/
COMMON /COST23/
COMMON /COST24/
COMMON /COST28/
add PPFAIL 6/19/91,
COMMON /COST47/
add Purge/Pressure
COMMON /COST48/
COMMON /COST49/
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
/ATPAR2/
/IMPAR1/
/IMPAR2/
/IMPAR6/
/MYRSAV/
/IM240P/
/REGISF/
/YEARS4/
ATPFEE
STRNGY , RMAXF , REPEAT , FFACT
REPLCE(25,6,10,4)
CSTIM, GASCST, IMSTRT, ISCEN, MOVTM, MOVATP, REPIM (4)
FAILRT(2,2S,39,4)
DRSTRN (5) , ZMSTRN (5)
FEBNFT (25,10, 4), FECNMY (20, 4)
IMFXFG,IAFXFG,IMCSFG,IAFEFG,IALDV
CIMFEE(25,10,4),CIMREP(25,10,4)
TONRED(25,2,10,4>,COST(25,10,4)
TOTCST(2),TOTRED(2),TCOST
DATPFE , DIMFEE , DIMREP , DATPRE, AATPFE , AIMFEE
AIMREP, AATPRE,CFACTR,CATPFE (25,10, 4)
ICOV1,ICOV2,INDVL1,INDVL2
CATPRE(25,10,4),FULBEN(25,20,4)
ATCOST(2,8)
VCNT(46,8)
VCNT(39,8),GSDSCT(8,10),HDDCUM(25,10)
ATEQP(20,5,4)
JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
and PPFEB,PPFIXR,APFBEN 7/11/91 ...
PPFAIL(25,3,10,4),PPFIXR(25,10,4)
inspection fees, 6/21/91 ...
PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
PPFEB(25,10, 4) ,APFBEN
ATPPGM,ATPFQT, CRATP, DISTYP (8)
ICYIM, ISTRIN, MODYR1,MODYR2, WAIVER (2) , CRIM
ILDT (4) , ITEST, NUDATA(2) , NLIM, IMRAME (20, 9)
IFREQ,INTYP
AMAR(25,8),JULMYR(25,8),NEWCUM
DSIZE (25, 4) , IM24YR, IPRGYR, IPRSYR
GSFVCT < 8 ) ,USRGSF(25,2)
IY1941,IY1960,IY2020
MY - MYDX + 1957
ICY - ICYDX + 1974
ICY - ICYDX + 1981
CNSPEC -0.0
CREPIM - 0.0
CREPAT - 0.0
to allow addition of Purge/Pressure inspection costs, 6/21/91...
CATPRE(IDX, JCYDX, IV) - 0.0
CEVFEE (IDX, JCYDX, IV) =0.0
Determine vehicle technology group.
ITECH-1
IF(IV. EQ. LAND. MY. GE. 1981) ITECH-2
IF(IV.GT.LAND.MY.GE.1984) ITECH-2
Get average miles accumulated during the calendar
year, for the given model, and aet I50K index.
I50K-1
CALL CALMIL(JDX, IV, ACCMIL, VMT)
IF(ACCMIL.GT.50000.0) I50K-2
Skip cost calculation if vehicle is not inspected.
IF(INDVL1.NE.2 .AND. INDVL2. NE. 2 .AND. MY.LT.JPRGYR .AND.
* MY.LT.JPRSYR) GOTO 30
74
-------
Appendix A
IF(INDVL1.NE.2.AND.INDVL2.NE.2) GOTO 5
C
C Inspection coat is the sum of the I/M fee plus the ATP
C fee.
C
FEE - CST1M
G
C TRNFEE is incremental transient inspection cost over purge
C inspection coat, PRGFEE
C. . .
IF(MY.GE.IM24YR) THEN
FEE - FEE + TRNFEE
IF (IM24YR.LT. JPRGYR .AMD. MY. LT. JPRGYR) FEE - FEE + FRGFEE
ENDIF
IF(IFREQ.EQ.2) FEE-FEE/2.0
IF(INDVL1.EQ.2)
* CTMFEE (IDX, JCYDX, IV) =
* VCNT (ICYDX, IV)* JOLMYR( JDX, IV) *FEE
C
C FEE=ATPFEE
C IF(ATPFQT.EQ.2) FEE-FEE/2.0
IF(INDVL2.EQ.2) THEN
FEE - ATPFEE
IF(ATPFQT.EQ.2) FEE - FEE/2.0
CATPFE (IDX, JCYDX, IV) -
* VCNT (ICYDX, IV) *JULMYR (JDX, IV) *FEE
ENDIF
C
C Add Purge S Preaaure Inapection coat calcs, 6/21/91 ...
C uaing PRGFEE, PRSFEE, and CEVFEE
C
5 IF(IV.LE.4 .AND. (MY.GE. JPRGYR .OR. MY.GE. JPRSYR) ) THEN
C IF(IV.LE.4) THEN
IF (MY. GE. JPRGYR) THEN
FEE - PRGFEE
IF(IFREQ.EQ.2) FEE - FEE/2.0
CEVFEE (IDX, JCYDX, IV) - CEVFEE (IDX, JCYDX, IV) +
* VCNT (ICYDX, IV)* JULMYR( JDX, IV) *FEE
ENDIF
IF (MY. GE. JPRSYR) THEN
FEE = PRSFEE
IF(IFREQ.EQ.2) FEE - FEE/2.0
CEVFEE (IDX, JCYDX, IV) = CEVFEE (IDX, JCYDX, IV) +
* VCNT (ICYDX, IV) *JOLMYR( JDX, IV) *FEE
ENDIF
C ENDIF
ENDIF
C CNSPEC - CIMFEE (IDX, JCYDX, IV) + CATPFE (IDX, JCYDX, TV)
CNSPEC - CIMFEE (IDX, JCYDX, IV) + CATPFE (IDX, JCYDX, IV)
* -1- CEVFEE (IDX, JCYDX, IV)
C
8 IF(INDVL1.NE.2) GOTO 10
C
C IM repair coat.
C For IM240 failures, uae 2-apd repair coat for that fraction of
C IM240 failures that would have failed the 2-speed teat.
C
IF(MY.GE.IM24YR) THEN
FAIL2S - ZMSTRN(2) + (DRSTRN(2)*ACCMIX.*0.0001)
IF(FAIL2S.GT.RMAXF) FAIL2S - RMAXF
IF(FAIL2S.LT.O.O) FAIL2S - 0.0
IF (FAIL2S . LE. FAILRT (1, IDX, ICYDX, IV) ) THEN
CREPIM - FAIL2S*REPIM(2) +
* ((FAILRT(1,IDX,ICYDX,IV) - FAIL2S) * REPIM(3))
ELSE
CREPIM - FAIL2S*REPIM(2)
ENDIF
C
C Add NOx repair costs...
C
CREPIM - CREPIM + FAILRT (2, IDX, ICYDX, IV) *REPIM (4)
C
ELSE
IF((MY.LE.1980) .OR. (MY.LE.1983 .AND. IV.GE.2))
* CREPIM-FAILRT (1, IDX, ICYDX, IV) *REPIM (1)
IF((MY.GE.1984) .OR. (MY.GE.1981 .AND. IV.EQ.l))
* CREPIM-FAILRT (1, IDX, ICYDX, IV) *REPIM(2)
ENDIF
C
CREPIM - CREPIM* (GRIM/100) * VCNT (ICYDX, IV) *JULMYR( JDX, IV)
CIMREP (IDX, JCYDX, IV) -CREPIM
75
-------
Appendix A
c
10 IP(INDVL2.HE.2> GOTO 21
C
C AT? repair coat.
C
DO 20 ITAM5-1,6
ITAM4=ITAM5
IF(ITAM4.GT.2) ITAM4-ITAM4-1
C
IDXP2 - MY - 1966
IF(IDXP2.LE.O) IDXP2-1
IF(IDXP2.GE.20) IDXP2-20
CREPAT - CREPAT+(ATEQP(IDXP2,ITAM4,IV)*
* REPLCE (IDX, ITAMS, JCYDX, IV) *ATCOST (ITECH, ITAM5) *
* (CRATP/100.))
C
20 CONTINUE
C
21 IF ( (MY. IiT. JPRGYR .AND. MY.LT. JPRSYR) .OR. IV.GT.4) GOTO 25
C
C Fix Purge, Press repair costs including separate Prg+Prs cost
C in new block PPFAIL, 6/21/91
C
IF (MY.GE. JPRGYR) CREPAT - CREPAT4- (PPFAIL (IDX, 1, JCYDX, IV) *
* (ATCOST (ITECH, 7))* (CRATP/100.))
IF (MY. GE. JPRSYR) CREPAT - CREPAT+ (PPFAIL (IDX, 2, JCYDX, TV) *
* (ATCOST(ITECH,8))*(CRATP/100.))
C
25 CREPAT - CREPAT*VCNT(ICYDX,IV)*JULMYR(JDX,IV)
CATPRE (IDX, JCYDX, IV) - CREPAT
C
C Sum inspection, I/M and ATP costs to get total.
C Note: If fuel economy benefits are larger than the program
C costs, then a negative cost is assumed for that model year.
C
30 COST(IDX,JCYDX,IV) - CNSPEC + CREPIM + CREPAT
IF (COST (IDX, JCYDX, IV) . GT. 0. 0 . AND. INDVL1. EQ. 2)
* COST (IDX, JCYDX, IV) -
* COST (IDX, JCYDX, IV) - ( FEBNFT (IDX, JCYDX, IV) *
* VCNT(ICYDX,IV)*JULMYR(JDX,IV) )
C
IF(INDVL1.EQ.2)
* FOLBEN (IDX, JCYDX, IV) - FEBNFT (IDX, JCYDX, IV) *
* VCNT(ICYDX,IV)*JULMYR(JDX,IV)
C
C Add Purge/Press FE benefit summation...
C
IF((MY.GE. JPRGYR .OR. MY. GE. JPRSYR) .AND. IV.LE.4) THEN
PPFEB (IDX, JCYDX, IV) - PPFEB(IDX, JCYDX, IV) *
* VCNT(ICYDX, IV)*JULMYR(JDX,IV)
C
IF (COST (IDX, JCYDX, IV) . GT. 0 . 0 . AND. INDVL1. EQ. 2 )
* COST(IDX,JCYDX,IV) -
* COST (IDX, JCYDX, IV) - PPFEB (IDX, JCYDX, XV)
ENDIF
C
C
99 RETURN
C End SMCOST
END
76
-------
Appendix A
SUBROUTINE SUMWDW
c
C.. SUMWDW auma emission reductions and coats across model years
C . . for each calendar
C
C. .Array Subscripts
C
C. .ADAT(2)
C. .ADET (2)
C. .ADIM(2)
C. .CATPFE (25, 10, 4)
C. .CATPRE (25, 10, 4)
C. .CIMFEE (25, 10, 4)
C. .CIMREP (25, 10, 4)
C. .COST (25, 10, 4)
C. .DETER(25,2,10,4)
C. .FULBEN(25,20,4)
C. .REDATP (25, 2, 10, 4)
C. .REDIM(25,2,10,4)
C. .TBEN(2)
C. .TONRED (25, 2, 10, 4)
C. .YRBEN(10,2)
C. .YRCOST(IO)
C
year, and store in appropriate variable
- ADAT (IP)
- ADET (IP)
- ADIM (IP)
- CATPFE (MYDX, JCYDX, IV)
- CATPRE (MYDX, JCYDX, IV)
- CIMFEE (IDX, JCYDX, IV)
- CIMREP (IDX, JCYDX, IV)
- COST (IDX, JCYDX, IV)
- DETER (IDX, IP, JCYDX, IV)
- FULBEN (MYDX, JCYDX, IV)
- REDATP (IDX, IP, JCYDX, IV)
- REDIM(IDX,IP, JCYDX, IV)
- TBEN(IP)
- TONRED (MYDX, IP, JCYDX, IV)
- YRBEN( JCYDX, IP)
- YRCOST (JCYDX)
C. .Variable Dictionary
C
C Name Source
C AATFFE Block
C AATPRE Block
C ADAT Block
C
C ADET Block
C
C ADFBEN Block
C
C ADIM Block
C AEVAP Block
C AIMFEE Block
C AIMREP Block
C APFBEN Block
C CATPFE Block
C
C CATPRE Block
C
C
C CIMFEE Block
C
C CIMREP Block
C
C
C COST Block
C
C DETER Block
C
C
C FULBEN Block
C
C
C ICY Local
C JCYDX Local
C IDX Local
C IFEVYR Block
C IFICY Local
C ILEVYR Block
C ILICY Local
C IP Local
C IV Local
C JDX Local
C MY Local
C MYDX Local
C NUMCY Local
C
C REDATP Block
C
C REDIM Block
C
C THEN Block
C
C TCST Block
C TONRED Block
Description
Final sum of the ATP inspection fees
Final sum of the ATP repair costs
Array containing the final sum of the ATP
benefits
Array containing the final sum of the I/M
deterence benefits
Containa the final sum of the fuel
economy benefits
Array containing the final sum of the I/M benefit
Final sum of the Evap+Runloas benefits in tons.
Final sum of the I/M inspection fees
Final sum of the I/M repair coats
Final sum of Purge/Pressure Fuel economy benefits
Array which holda and auma the ATP fees for each
model year, calendar year and vehicle type
Array which holda and auma the ATP repair costs
for each model year, calendar year and vehicle
type
Array which holda and auma the I/M fees for each
model year, calendar year and vehicle type
Array which holda and sums the I/M repair coats
for each model year, calendar year and vehicle
type
Array which holda and sums the total coat for
each model year, calendar year and vehicle type
Array which containa the pollutant reductions for
each model year, calendar year and vehicle type
due to tamering deterence.
Array which holds and auma the fuel economy
benefit for each model year, calendar year and
vehicle type
Calendar year
Calendar year index
Model year index (forward 1-25)
First calendar year evaluated
Index of firat calendar year evaluated
Laat calendar year evaluated
Index of last calendar year evaluated
Pollutant index (1 - HC ; 2 = CO)
Vehicle class index
Model year index (backwards 20-1)
Model year
Model year index
Number of calendar years which are being
evaluated (number of scenario records - 1)
Array which holds and sums the ATP benefit for
each model year, calendar year and vehicle type
Array which holda and sums the I/M benefit for
each model year, calendar year and vehicle type
Array containing the final total benefit from
I/M deterence, I/M and ATP
Final total cost of all control programs
Array which passea the total benefit
77
-------
Appendix A
c
c
c
c
c
c
c
c
c
c
c
YRBEN Block Array which contains the individual calendar
year benefits
YRCOST Block Array which contains the individual calendar
year costs
REAL JULMYR
COMMON /COST02/
COMMON /COST14/
COMMON /COST16/
COMMON /COST17/
COMMON /COST18/
COMMON /COST13/
COMMON /COST21/
COMMON /COST20/
COMMON /COST27/
COMMON /COST35/
COMMON /COST36/
COMMON /COST37/
COMMON /COST46/
COMMON /COST47/
COMMON /COST48/
COMMON /COST49/
COMMON /COST50/
COMMON /COS-ISO/
COMMON /COST51/
COMMON /MAXIMA/
IFEVYR, ILEVYR
TOURED (25, 2, 10, 4) , COST (25, 10, 4)
DATPFE, DIMFEE, DIMREP, DATPRE, AATPFE, AIMFEE
DETER(25,2,10,4),TDETER(25,2,10,4)
AIMREP, AATPRE, CFACTR, CATPFE (25, 10, 4)
CIMFEE (25, 10, 4) ,CIMREP (25,10, 4)
CATPRE (25, 10, 4) ,FOLBEN (25, 20, 4)
TDOCT (25, 2, 3, 10, 4) , TEVAP (25, 10, 4)
IGRFLG, IATPLP , IDIFF, OUTPRF
TCST, THEN (2) , ADFBEN, ADIM (2) ,ADAT(2) ,ADET(2)
YTONS (2, 10) ,BTONS (2) , BIDTON (2) , AEVAP
YRCOST (10) , YRBEN (10, 2) , YCOUNT (39) , ACOUNT
REDIM(25,2,10,4),REDATP(25,2,10,4)
PPFAIL(25,3,10,4),PPFIXR(25,10,4)
PR6FEE,PRSFEE,CEVFEE(25,10,4) , TRNFEE
PPFEB (25, 10, 4) ,APFBEN
TRLOSS (25, 4, 3) , TREVAP (25, 4, 3)
TRLOSS (25, 4, 3) , TREVAP (25, 4, 3) , TRSTLS (25, 4, 3)
ARLOSS , AREVAP , ARS TLS
MAXVEH, MAXLTW, MAXPOL, MAXRE6, MAXYRS
IFICY - IFEVYR - 1981
ILICY - ILEVYR - 1981
C
C
c
c
c
c
c
Initialize summing
TCST - 0.0
AIMFEE - 0.0
AIMREP - 0.0
AATPRE - 0.0
AATPFE - 0.0
ADFBEN =0.0
AFFBEN - 0.0
AEVAF - 0.0
ARLOSS - 0.0
AREVAP -0.0
DO 10 IP - 1,2
TBEN(IP) - 0.0
ADIM(IP) - 0.0
ADAT(IP) - 0.0
ADET(XP) -0.0
10 CONTINUE
variables
over vehicle type, model year and calendar v
C
c
c
c
DO 60 JCYDX-1, IDIFF
YRCOST(JCYDX) - 0.0
DO 12 IP - 1,2
YRBEN(JCYDX,IP) - 0.0
12 CONTINUE
DO 50 17-1,4
ICYDX - (IFEVYR-1981) •»• JCYDX - 1
ICY - 1981 + ICYDX
DO 40 IDX - 1,MAXYRS
JDX - MAXYRS + 1 - IDX
MYDX - MAXYRS + ICYDX - JDX
MY - MYDX + 1957
TCST - TCST + COST(IDX,JCYDX,IV)
AIMFEE = AIMFEE+CIMFEE(IDX, JCYDX, IV) +CEVFEE (IDX, JCYDX, IV)
AIMREP » AIMREP + CIMREP (IDX, JCYDX, IV)
AATPRE m AATPRE + CATPRE (IDX, JCYDX, IV)
AATPFE » AATPFE + CATPFE (IDX, JCYDX, IV)
ADFBEN - ADFBEN + FULBEN (IDX, JCYDX, IV)
APFBEN - APFBEN + PPFEB (IDX, JCYDX, IV)
Sum cost over vehicle class and model year.
As of 10/7/91 this includes PP FE benefit
78
-------
Appendix A
c
c
c
c
c
YRCOST (JCYDX) - YRCOST (JCYDX) + COST (IDX, JCYDX, IV)
Sum benefits over vehicle claaa, model year and
calendar year for each pollutant. Alao sum over
vehicle claaa and model year only.
DO 20 IP-1,2
TBEN(IP) - TBEN(IP)
ADIM(IP) - ADIM(IP)
AD AT (IP) = ADAT (IP)
ADET(IP) = ADET(IP)
+ TONRED(IDX,IP,JCYDX,IV)
+ REDIM(IDX,IP,JCYDX,IV)
+ REDATP (IDX, IP, JCYDX, IV)
DETER(IDX,IP,JCYDX,IV)
20
YRBEN(JCYDX,IP) - YRBEN(JCYDX,IP)+TONRED(IDX,IP,JCYDX,IV)
CONTINUE
C
C
C
Sum evaporative emission benefita
AEVAP - AEVAP + TEVAP (IDX, JCYDX, IV)
15
DO 15 IMC=1,3
AREVAP - AREVAP + TREVAP (IDX, IV, IMC)
ARLOSS - ARLOSS + TRLOSS (IDX, IV, IMC)
CONTINUE
C
C
C
C
40 CONTINUE
50 CONTINUE
60 CONTINUE
Compute annual averages changing to thousands of tons and
thousands of dollars.
TBEN(l)
TBEN(2)
TCST
AATPFE
AIMFEE
AIMREP
AATPRE
ADFBEN
APFBEN
AEVAP
ADIM(l)
ADIM(2)
ADAT(l)
ADAT(2)
ADET(l)
ADET(2)
ARLOSS >
AREVAP *
TBEN(l) * .001 / IDIFF
TBEN(2) * .001 / IDIFF
TCST * .001 / IDIFF
AATPFE *
AIMFEE *
AIMREP *
AATPRE *
ADFBEN *
APFBEN *
AEVAP *
ADIM(l)
ADIM(2)
ADAT (1)
ADAT (2)
ADET(l)
ADET (2)
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
* .001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
.001 / IDIFF
ARLOSS * .001 / IDIFF
AREVAP * .001 / IDIFF
C
C
DO 30 JCYDX - 1,IDIFF
YRCOST (JCYDX) - YRCOST (JCYDX) * .001
YRBEN(JCYDX,!) - YRBEN(JCYDX,1) * .001
YRBEN(JCYDX, 2) - YRBEN (JCYDX, 2) * .001
30 CONTINUE
RETURN
End SUMHDW
END
79
-------
Appendix A
BLOCK DATA
C
C
C
C
C
C
C
C
C
G
C
C
C
C
C
C
C
COMMON /COST09/ DRSTRN(5),ZMSTRN(5>
COMMON /COST10/ FEBNFT(25,10,4),FECNMY(20,4)
COMMON /COSTll/ PCNTSV(4)
COMMON /COST24/ ATEQP(20,5,4)
COMMON /COST38/ STOCK(12,8),CLDIST(8)
ZMSTRN(ITEST,ICUT) is the zero-mile intercept used to calculate the
failure rates for 1981 and newer vehicles. Idle, 2-Speed, Loaded,
IM240(.8/15),NOx(10% fail-1.69 MPFI & 2.50 TBI fi 3.99 Garb)
DATA ZMSTRK / 0.0,0.0,0.0252,0.0,0.032936/
DRSTRN(ITEST,ICUT) is the rate per 10k miles used to calculate the
failure rates for 1981 and newer vehicles. Idle, 2-speed, Loaded,
IM240,NOx
Revise IM240 back to initial rate 11/18/91
DATA DRSTRN / 0.01,0.01,0.01190,0.0373,0.0084805/
FECNMY is the miles-per-gallon average for each model year
by vehicle class. (MY 1968 - 1987+)
DATA FECNMY/
1 13.9, 13.9, 13.2, 13.1, 12.9, 12.6, 13.9, 14.9, 15.6, 16.7,
1 18.5, 19.6, 21.8, 23.3, 24.6, 26.0, 27.4, 28.8, 30.2, 31.6,
2 10.6, 10.6, 10.4, 10.2, 9.9, 9.6, 12.0, 12.6, 13.8, 14.3,
2 15.2, 16.3, 18.1, 18.4, 18.9, 19.5, 20.2, 21.1, 22.0, 22.9,
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
3 7.9, 7.9, 7.7, 7.4, 7.0, 6.9, 8.8, 9.7, 9.4,
3 9.8, 11.5, 13.3, 13.8, 14.3, 14.9, 15.4, 16.0, 16.6,
4 7.9, 7.9, 7.7, 7.4, 7.0, 6.9, 8.8, 9.7, 9.4,
4 9.8, 11.5, 13.3, 13.8, 14.3, 14.9, 15.4, 16.0, 16.6,
FCNTSV is the average fuel savings by technology group.
Pre-81 2spd/idle, 81+ 2spd/idle, 83+ IM240, Purge/Press
DATA PCNTSV/ 0.0, 0.080, 0.126, 0.059 /
ATEQP determines the fraction of each model year in each
class equipped with each tampering component .
ATEQP CODE:
1-air
2-cat
3"pcv
4-evap
5»gas cap
DATA ATEQP/
LDGV
1 .00, .05, .05, .05, .10, .30, .30, .45, .40, .30,
1 .30, .30, .65, .85, .70, .60, .60, .40, .40, .30,
2 .00, .00, .00, .00, .00, .00, .00, .80, .85, .85,
2 .90, .90, .95,1.00,1.00,1.00,1.00,1.00,1.00,1.00,
3 .00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,
3 1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,
4 .00, .00, .00, .00,1.00,1.00,1.00,1.00,1.00,1.00,
4 1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,
5 20*1.00,
LDGT1
1 .00, .05, .05, .05, .10, .30, .30, .40, .40, .30,
1 .30, .50, .50, .50, .50, .50, .50, .50, .50, .50,
2 .00, .00, .00, .00, .00, .00, .00, .70, .80, .75,
2 .75, .80, .80,1.00,1.00,1.00,1.00,1.00,1.00,1.00,
3 .00, 19*1.00,
4 .00, 19*1.00, 20*1.00,
9.6,
17.2,
9.6,
17. 2/
Repair
vehicle
LDGT2 IVTAM4-1 IVTAM4-2 IVTAM4-3 IVTAM4-4
* 11*.00,9*.50, 11*.00,9*1.(
* 20*1.00,
C HDGV IVTAM4=1
IVTAM4=2
* 11*.00,9*.50, 11*.00,9*1.
* 20*1.OO/
END
IVTAM4=3
00, .00,19*1.00,
3*.00,8*.05,9*1.00,
IVTAM4=4
3*.00,8*.05,9*1.00,
80
-------
Appendix A
SUBROUTINE PRINT(1SIZE1,ISIZE2)
c
c.
c
c
c.
c
c.
c
c.
c
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c.
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
.PRINT lists the
overall emission benefits and costs far
the indicated calendar years combined. No discounting'"
.Called by IAIN
.Calls PRNIM,PRNATP,PRNCST,PRNBEN
.Array Sub script a
.ADAT (2)
.ADET(2)
.ADIM(2)
.BTONS(2)
.IVPRT(4)
.KCOV(2,2>
.KCOVST(2,2)
.KFREQ<2,2)
.KTYPE(4,2)
.LVTFLG(4)
.»OYES(2)
.TBEN(2)
.TTYPE(4,3)
.V6RORT(8)
.VNAME(4)
.WAIVER (2)
.YCOUNT(46)
.YRBEN(10,2)
.YRCOST(IO)
.YTONS(2,10)
- ADAT (IP)
- ADET(IP)
- ADIM(IP)
- BTONS (IP)
- IVPRT(NCOVER),IVPRT(I2ND)
- KCOV(ICH,IAFXFG)
- KCOVST(ICH,IAFXFG)
- KFREQ (ICH, IFREQ)
- KTYPE(ICH,ATPPGM)
- LVTFLO(IVTAM)
- NOYES (IMTFLS)
- TBEN(IP)
- TTYPE(ICH,ATPPGM)
- VGRORT(KI)
- VNAME (XVPTR (NCOVER) )
- WAIVER (ITECH)
- YCOUNT(IY)
- YRBEN(IY,IP)
- YRCOST(IY)
- YTONS(2,IY)
•Variable dictionary
Name Source
AATPFE Block
AATPRE Block
ACODNT Block
ADAT Block
AOET Block
ADFBEN Block
ADIH Block
AIMFEE Block
AIMREP Block
Al'FFEE Block
BTONS Block
BYEAR Block
CLDIST Block
COMMA Local
CSTIM Block
CUTCO Local
CUTHC Local
GASCST Block
SHORT BLock
IAFXFG Block
IATPLP Block
ICH Local
ICO Local
ICUTS Block
ICYIM Block
ICY2 Block
IPEVYR Block
IFREQ Block
IFICY Local
ILEVYR Block
ILICY Local
IMFXFG Block
IMTFIiG Block
IOOREP Block
ISIZE1 Block
Description
Final sum of the ATP inspection fees
Final sum of the ATP repair costs
Total vehicle count over all scenario years
Array containing the final sum of the ATP
benefits
Array containing the final sum of the I/M
deterence benefits
Contains the final sum of the fuel
economy benefits
Array containing the final sum of the I/M benefit
Final sum of the I/M inspection fees
Final sum of the I/M repair costs
Anti-tampering inspection fee
Array containing the number of tons of pollutant
without a control program
The fleet size base year
Vehicle class registration distribution
Variable holding the character ' , '
I/M inspection fee
Variable holding character data pertaining to CO
Variable holding character data pertaining to HC
The cost of one gallon of unleaded gasoline
The vehicle fleet growth rate by vehicle class
converted into percent for printing
Flag indicating whether ATP has fixed or floating
model year coverage
Flag indicating whether an ATP only scenario is
to be run
DO loop counter
DO loop counter
I/M outpoint index from MOBILE 4
Tampering deterence start year
Calendar year
First calendar year evaluated
Flag indicating whether the I/M program is
annual or biennial
Index of first calendar year evaluated
Last calendar year evaluated
Index of last calendar year evaluated
Flag indicating whether the I/M program has a
fixed or floating model year coverage
Mechanic Training flag
Output Device number
Variable containing either the size of the
floating model year window for I/M coverage or
81
-------
C
c
c
c
c
c.
c
c
c
c
c
Appendix A
the earliest model year inspected in an I/M
program
Variable containing either the size of the
floating model year window for I/M coverage or
the earliest model year inspected in an I/M
program
Number of input stringency groups
I/M start year
DO loop counter calling subroutine ATPOUT
Array of flags indicating which emission
control components are to be inspected by the
ATP
I/M test type from MOBILE4
I/M program type 1 = Centralized
2 «• Decentralized
DO loop index
Flag indicating whether ATP program is annual
or biennial
Variable holding the character data :
'floating' and 'fixed'
Array holding a string of character data
Array holding the character data :
'annual' and 'biennial'
DO loop index
Array holding the character data :
'centralized' and 'decentralized'
ATP start year
Counter variable 1-4
Mechanic training program character string
Variable holding the project ID character
string
Array containing the final total benefit from
I/M deterence, I/M and ATP
Vehicle fleet size in base year
Final total cost of all control programs
Variable holding the character data 'idle'
Array containing the growth rates for each
vehicle class
Array holding character data
Array containing the waiver rates for old
and new technology vehicles
Variable which contains the old tech waiver rate
Variable which contains the new tech waiver rate
The fleet count in a particular calendar year.
Total benefits in each evaluted calendar year.
Array which contains the individual calendar
year benefits
Array which contains the individual calendar
year costs
INTEGER ALHFLG,ATPFLG,TPDFLG,RLFLAG,OUTFMT
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /PROJEC/ PROJID (20)
DIMENSION IDATE(3),ITIME(3)
just use hh:mm:sec of time w/o hundreths of second
CHARACTER* 4 PROJID
CHARACTER*8 CDATE,CTIME
IREP - 3
MTS date/time...
CALL ANSITM(ITIME)
WRITE (IREP, 100) PROJID, IDATE (2) , IDATE (3) , IDATE (1) ,
* ITIME(1),ITIME(2),ITIME(3)
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
ISIZE2
ISTRIN
ISTRT
IT
DISTYP
ITEST
INTYP
IVTAM
ATPFQT
KCOV
KCOVST
KFREQ
KII
KTYPE
LAPSY
NCOVER
NOTES
PROJID
THEN
TCOUNT
TCST
TTYPE
VGRORT
VNAME
WAIVER
MAIV1
MAIV2
YCOUNT
YTONS
YRBEN
YRCOST
Block
Block
Block
Local
Block
Block
Block
Local
Block
Local
Local
Local
Local
Local
Block
Local
Block
Block
Block
Block
Block
Local
Block
Local
Block
Local
Local
Block
Block
Block
Block
C 100 FORMAT('ICost Effect. Model (CEM4.1)
* IX,'
-------
Appendix A
C MacFortran date/time...
C
C CALL DATE(IDATE(2),IDATE(3),IDATE(1))
C CALL TIME(ISECS)
C ITIME(l) - ISECS/3600
C ISECS - ISECS - (ITIME(1)*3600)
C ITIME<2) - ISECS/60
C ITIME(3) - ISECS - ITIME(2)*60
C WRITE(IREF,100) EROJID,IDATE(2),IDATE(3),IDATE<1),
C * ITIME(1),ITIME(2),ITIME(3)
C 100 FORMAT('ICoBt Effect. Model (CEM4.1): ',20A4,
C * 1X,'
-------
Appendix A
SUBROUTINE PRNIM(IREP, ISIZE1)
C
C. .PRNIM lists the
C
C. .Called by PRINT
C
C. .Array Subscript a
C
C. .ILDT(4)
C. .IMSTR(4,20)
C. .IVPTR(4)
C. .KCOV(2,2)
C. .KCOVST(2,2)
C. .KFREQ(2,2)
C. .KTYPE(4,2)
C. .NEWSTR(4,100)
C. .TTYPE(4,3)
C. .VNAME(4)
C. .WAIVER (2)
C
I/M program parameters
- ILDT(IVTAM)
- IMSTR (IV, NMY)
- IVPTR (NCOVER) , IVPTR (I2ND)
- KCOV (ICH, IAFXFG)
- KCOVST (ICH, IAFXFG)
- KFREQ (ICH, IFREQ)
- KTYPE (ICH,ATPPGM)
- NEWSTR (IT, ISTRN)
- TTYPE (ICH, ITEST)
- VNAME (IVPTR (NCOVER))
- WAIVER (ITECH)
C. .Variable dictionary
C
C Name Source
C COMMA Local
C GRIM Block
C I Local
C ICH Local
C ICYIM Block
C IFREQ Block
C
C IMFXFG Block
C
C IMSTR Local
C INTYP Block
C
C IREP Local
C ISIZE1 Block
C
C
C
C ISTRIN Block
C ISTRT Block
C ITEST Block
C IV Local
C
C
C
C KCOV Local
C
C KCOVST Local
C KFREQ Local
C
C KTYPE Local
C
C NEWSTR Block
C
C
C
C
C NCOVER Local
C TTYPE Local
C VNAME Local
C WAIVER Block
C
C WAIV1 Local
C WAIV2 Local
C
Description
Variable holding the character ' , '
I/M program compliance rate
DO loop counter
DO loop counter
Tampering deterence start year
Flag indicating whether the I/M program is
annual or biennial
Flag indicating whether the I/M program has a
fixed or floating model year coverage
Stringency by model year and vehicle type
I/M program type 1 = Centralized
2 « Decentralized
Output device indicator
Variable containing either the size of the
floating model year window for I/M coverage or
the earliest model year inspected in an I/M
program
Number of input stringency groups
I/M start year
I/M test type from MOBILE4
Vehicle type 1 - LDGV
2 - LDGT1
3 - LDGT2
4 - HDGV
Variable holding the character data :
'floating1 and 'fixed1
Array holding a string of character data
Array holding the character data :
1 annual ' and ' biennial '
Array holding the character data :
'centralized' and 'decentralized1
Contains stringency information
1 : Contains vehicle class code
2 : Contains first model year covered
3 : Contains last model year covered
4 : Contains stringency value
Counter variable 1-4
Variable holding the I/M teat type character data
Array holding vehicle class character data
Array containing the waiver rates for old
and new technology vehicles
Variable which contains the old tech waiver rate
Variable which contains the new tech waiver rate
CHARACTER*8 VNAME (4)
CHARACTER*4 KTYPE (4, 2) ,KCOV(2,2) ,KCOVST (8, 2)
CHARACTER*4 TTYPE(4,3),KFREQ(2,2)
CHARACTER* 4 NOYES, COMMA, PERIOD
CHARACTER*! COLON
COMMON /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
COMMON /COST28/ JATP, JIM, ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /IM240P/ DSIZE(25,4),IM24YR,IPRGYR,IPRSYR
COMMON /IMPAR1/ ICYIM, ISTRIN, MODYR1, MODYR2, WAIVER (2) ,CRIM
COMMON /IMPAR2/ ILDT (4) , ITEST,NODATA(2) ,NLIM, IMNAME (20, 9)
COMMON /IMPAR6/ IFREQ, INTYP
COMMON /STRING/ NOYES(2),COMMA,PERIOD,COLON
84
-------
Appendix A
c
DIMENSION IVPTR (4)
DIMENSION IMSTR(4,20)
C
DATA KCOV /
* 'Floa','ting',
* 'Fixe','d ' /
DATA KCOVST /
* 'Numb','er o','f mo','del ','year','a in',' win','dow:',
* 'Olde','at m','odel1,' yea','r co-,'vere','d: ',' ' /
DATA KFREQ /
* 'Annu','al ',
* "Bien1,'nial' /
DATA KTYPE /
* 'Cent','rali','zed ',' ',
* 'Dece','ntra','lize','d ' /
DATA TTYPE/'Idle',' ',' ',' ',
2 '2500',' rpm',' / I','die ',
3 'Load','ed /',' Idl','e '/
DATA VNAME/'LDGV ','LDGT1 ' , 'LDGT2 •,'HDGV ' /
C
C
I1-MODYR1-1900
C WRITE(IREP,100) ISTRT,(1,1-11,83,1)
WRITE(IREP,100) ISTRT,ISTRIN
100 FORMAT('0',55X,'I/M Program Selected:'/
* 130('_'),/,
* ' I/M start year (January 1) : ',I4,25X,
* 'Stringency:',IX,12 )
C
I2-83-I1+1
DO 20 IV-1,4
DO 20 NMY-1,12
IMY-MODYR1+NMY-1
IMSTR(IV,NMY)-0
IF (IMY. GE. MODYR1. AND. IMY. LE. MODYR2)
* IMSTR(IV,NMY)-ISTRIlf
10 CONTINUE
20 CONTINUE
I3-80-I1+1
WRITE(IREP,101) ICYIM
101 FORMAT(' ',' Tampering deterrence start: ',I4,30X)
C
IF(IMFXFQ.EQ.l) WRITE (IREP, 102) (KCOVST (ICH, IMFXFG) , ICH-1, 8) ,
* ISIZE1,
* (KCOV(ICH,IMFXFG),ICH-1, 2),
* (IMSTR(2,I),1-1,12,1)
102 FORMAT(4X,8A4,IX,14,2X,2A4,20X,'LDGT1 ',16(12,IX))
IF (IMFXFG.EQ. 2) WRITE (IREP, 103) (KCOVST (ICH, IMFXFG) , ICH-1, 8) ,
* ISIZE1,MODYR2,
* (KCOV(ICH, IMFXFG) , ICH-1, 2)
C * (IMSTR(2,I), 1-1,12,1)
103 FORMAT(4X,8A4,IX,14,'-',14,2X,2A4,1SX)
C
C
WRITE(IREP,104) (KFREQ(ICH,IFREQ),ICH-1,2)
C * (IMSTR(3,1), 1-1,12,1)
104 FORMAT(' ',' I/M inspection frequency: ',2A4,26X)
C * 'LDGT2 ',16(12,IX))
C
C
WRITE(IREP,105) (KTYPE(ICH,INTYP),ICH-1,4)
105 FORMAT(' ',' I/M program type: ',4A4,8X)
C
C
WRITE(IREP,106) (TTYPE(ICH,ITEST),ICH-1,4) , IM24YR
106 FORMAT(' ',' 1981 fi later MYR teat type: ',4A4,
* T75,'IM240 test model year coverage: ',14,'+')
C
C
NCOVER-0
DO 30 IV-1,4
IF(ILDT(IV).EQ.l) GOTO 30
NCOVER-MCOVER+1
IVPTR (NCOVER) -IV
30 CONTINUE
C
IF (NCOVER.EQ.l) WRITE (IREP, 107) VNAME (IVPTR (1) ), JPRGYR
107 FORMAT(' ',' Vehicle types covered: ',A4,
* T75,'Purge check model year coverage: ',14,'+')
NCOMMA-NCOVER-1
85
-------
Appendix A
IF(NCOVER.EQ.2)
* WRITE(IREP,108) VNAME (IVPTR(l)) , COMMA,
* VNAME (IVPTR (HCOVER) ) , JPRGYR
108 FORMAT(' ',' Vehicle types covered: ',7X,A4,A2,A5,
* T75,'Purge check model year coverage: ',14,' + ')
IF (NCOVER.EQ. 3)
* WRITE (IREP, 109) (VNAME (IVPTR(I) ) ,COMMA, I-1,HCOMMA) ,
* VNAME (IVPTR (NCOVER) ) , JPRGYR
109 FORMAT(' ',' Vehicle types covered: ',7X,A4,A2,A5,A2,A5,
* T75,'Purge check model year coverage: ',14,'+')
IF(NCOVER.EQ.4)
* WRITE (IREP, 110) (VNAME(IVPTR(I)),COMMA,I-1,NCOMMA),
* VNAME (IVPTR (NCOVER) ) , JPRGYR
110 FORMAT(' ',' Vehicle types covered: ',7X,A4,A2,2(A5,A2),A5,
* T7S,'Purge check model year coverage: ',14,'+')
C
WAIV1 - WAIVER (1) * 100.
WAIV2 = WAIVER(2) * 100.
WRITE (IREP, 120) WAIV1, JPRSYR,WAIV2, GRIM
120 FORMAT( ' I/M waiver rate (pre-1981): ',F5.!,'%',
* T75,'Pressure check model year coverage: ',14,'+',/,
* ' I/M waiver rate (post-1981): ',F5.1,'%',/,
* ' I/M compliance rate: ',F5.1,'%')
C
999 RETURN
END
86
-------
c
c
Appendix A
SUBROUTINE PRNATP (IRJEP, ISIZE2)
INTEGER ATPPGM,ATPFQT,DISTYP
CHARACTER*8 VNAME(4)
CHARACTER*4 KTYPE(4, 2) ,KCOV(2, 2) ,KCOVST (8,2)
CHARACTER*4 TINSP(3,8)
CHARACTER*4 TTYPE (4,3) ,KFREQ (2,2)
CHARACTER*4 NOYES,COMMA,PERIOD
CHARACTER*! COLON
COMMON /COST12/ IMFXFG,IAFXFG,IMCSFG,IAFEFG,IALDV
COMMON /ATPAR1/ LAPSY,LAP1ST,LAPLST, LVTFLG (4)
COMMON /ATPAR2/ ATPPGM,ATPFQT,CRATP,DISTYP (8)
COMMON /STRING/ NOTES (2) , COMMA,PERIOD, COLON
DIMENSION IVPTR(4) ,ITPTR(8)
DATA KCOV /
* 'Floa','ting',
* 'Fixe','d ' /
DATA KCOVST /
* 'Numb','er o','f mo
* 'Olde','at m1,'odel','
DATA KFREQ /
* 'Annu','al ',
* 'Bien','nial' /
DATA KTYPE /
* 'Cent','rail','zed ', '
* 'Decs','ntra','lize1, 'd
del ','ye-ax1,'• in',1 win','dow: ',
yea','r oo','vere1,'d: ',' ' /
C
c
DATA
*
*
*
*
*
*
*
DATA
TINSP/'Air '
'Cata1
'Fuel'
'Plum'
'EGR '
'Evap1
'PCV '
'Gaa '
VNAME/ 'LDGV
'Pump*
'lyat'
' Inl1
'btea'
'Syst'
• sya'
1 Syat '
•Cap '
', 'LD
et
mo
em
tern
em
/
GT1 '
1LDGT2
'HDGV
C
C
C
C
C
C
C
C
WRITE (IREP, 101) LAPSY,CRATP
101 FORMAT('0', SOX,'Anti-Tampering Program Selected:',/,
* 130C '),/,
* T ATP atart year (January 1): ',I4,25X,
* ' ATP compliance rate: ',F5.1,'%')
WRITE (IREP, 102) (KFREQ (ICH,ATPFQT) ,ICH-1,2) ,
* (KTYPE(ICH,ATPPGM),ICH-1,4)
102 FORMAT(' ',' ATP inapection frequency: ',2A4,21X,
* ' ',' ATP program type: ',4A4)
IF (IAFXFG.EQ.l) WRITE (IREP, 103) (KCOVST(ICH,IAFXFG),ICH-1,8),
* 1SIZE2
103 FORMAT(' ',' ',8A4,4X,I4,2X,2A4)
IF(IAFXFG.EQ.2) WRITE(IREP,104) (KCOVST(ICH,IAFXFG),ICH-1,8),
* ISIZE2,LAPLST,
* (KCOV (ICH, IAFXFG) , ICH-1, 2)
104 FORMAT(' ',' ',8A4,IX,14,'-',14,2X,2A4)
NCOVER-0
DO 10 TV-1,4
IP(LVTFLG(IV) .EQ.l) GOTO 10
NCOVER-NCOVER+1
IVPTR(NCOVER)-IV
10 CONTINUE
IF (NCOVER. EQ.l) WRITE (IREP, 105) VNAME (IVPTR (1) )
105 FORMAT(' ',' Vehicle types covered: ',A4,A4)
NCOMMA=NCOVER-1
IF(NCOVER.GT.l)
* WRITE (IREP, 106) (VNAME (IVPTR (I) ), COMMA, 1=1, NCOMMA) ,
* VNAME (IVPTR (NCOVER) )
106 FORMAT(' ',' Vehicle typea covered: ',7X,3(A5,A2),A5)
NCOVER-0
87
-------
Appendix A
DO 20 IT=1,8
IF(DISTYP(IT).EQ.l) GOTO 20
NCOVER=NCOVER+1
ITPTR (NCOVER) =IT
20 CONTINUE
C
IF(NCOVER.EQ.l) WRITE(IREP,107) (TINSE
-------
Appendix A
SUBROUTINE PRNCST(IREP)
C
INTEGER BYEAR
INTEGER PROMPT, TAMFLG, SPDFLG,VMFLAG
C
COMMON /VMXCOM/ REGMIX(8) , TFNORM(8) ,VMTMIX(8) ,VCOUNT (39, 8)
C
COMMON /COST01/ ATPFEE
COMMON /COST07/ CSTIM,GASCST,IMSTRT,ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST22/ ATCOST(2,8)
COMMON /COST27/ IGRFLG,1ATPLP,IDIFF,OUTPRF
COMMON /COST28/ JATP, JIM,ISTRT, JCYDX, JPRGYR, JPRSYR
COMMON /COST34/ TCOUNT, BYEAR, VGRORT
COMMON /COST38/ STOCK(12,8),CLDIST(8)
COMMON /FLAGS1/ PROMPT,TAMFLG,SPDFLG,VMFLAG,OXYFLG,DSFLAG
COMMON /COST48/ PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
C
DIMENSION IVPTR(4),DIST(8)
C
DO 10 1-1,8
DIST(I) - REGMIX(I)*100.
10 CONTINUE
C
IF (JIM. EQ. 2 .AND. JATP.EQ.2) THEN
C
WRITE(IREP,101) CSTIM, (DIST (KJ) ,KJ-1, 8),
* ATPFEE, VGRORT
101 FORMAT (
* '0',11X,'Cost Assumptions',31X,'Flaet Assumptions',5X,
* 'LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC ',/,
* 43(' '),12X,75('_'),/,
* ' I/M inspection cost: (per inap) $',F6.2,15X,
* ' Vehicle percentages: ',8(FS.1,1X>,/
* ' ATP inapection coat: (per insp) $',F6.2,15X,
* ' Growth rate: ',F4.1,' % per year'
* )
C
ELSE IF(JIM.EQ.l .AND. JATP.EQ.2) THEN
C
C WRITE (IREP, 104) ATPFEE,(VGRORT(KI),KI-1,8),
C * (DIST (KJ), KJ-1,8)
WRITE (IREP, 104) ATPFEE, (DIST (KJ) ,KJ-1, 8) , VGRORT
104 FORMAT(
* '0',11X,'Coat Assumptions',31X,'Fleet Assumptions', 5X,
* ' LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC' , /,
* 43(' '),12X,75('_'),/,
* ' ATP Tnapection coat: (per insp) $',F6.2,15X,
* ' Vehicle percentages: ',8(F5.1,IX),/,
* ' ',6X,15X,
* ' Growth rates ',F4.1,' % per year' )
C
ELSE IF(JIM.EQ.2 .AND. JATP.EQ.l) THEN
C
WRITE (IREP, 105) CSTIM, (DIST (KJ) ,KJ-1, 8) , VGRORT
105 FORMAT(
* '0',11X,'Coat Assumptions',31X,'Fleet Assumptions', 5X,
* 'LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC ',/,
* 43(' -),12X,75(' '),/,
* ' I/M Tnapection coat: (per inap) $',F6.2,15X,
* ' Vehicle percentagea: ',8(F5.1,1X),/,
* ' ',6X,15X,
* ' Growth rates ',F4.1,' % per year' )
C
END IF
C
IF (JPRGYR.NE.2020 .OR. JPRSYR.NE.2020)
* WRITE(IREP,110) PRGFEE,PRSFEE,TRNFEE
110 FORMAT(' Purge inapect cost (per insp) : $',F6.2,/,
* ' Pressure inap cost (per insp) : $',F6.2,/,
* ' IM240 insp incremnt over Purge: $',F6.2)
C
IF(IATPLP.EQ.l) WRITE(IREP,102) GASCST,TCOONT,BYEAR
102 FORMAT( ' Avg. Gasoline Coat(per gallon): $',F6.2,1SX,
* ' Fleet aize is ',F9.0,' vehicles in base year ',14)
C
C
IF(IATPLP.EQ.2) WRITE (IREP, 103) TCOUNT,BYEAR
103 FORMAT(57X,
* ' Fleet aize is ',F9.0,' vehicles in base year ',14)
C
C
89
-------
Appendix A
IF ( JIM.EQ.2 .OR. JATP.EQ.2 .OR. JPRGYR.LT.2020 .OR.
* JPRSYR.LT.2020 ) WRITE(IREP,106)
106 FORMAT(
* '0',5X,' Repair Coata',12X,'Pre-81',4X,'81+',/,
* 46('_')
* )
C
C
IF (JIM.EQ.2) WRITE(IREP,107)(REPIM(KI),KI-1,4)
107 FORMATC Avg. I/M repair coat : $ ',F4.0,4X,F4.0,
* ' ( IM240 repair - $ ',F4.0,' NOx repair- $',F4.0,' )'
* )
C
C
IF (JATP.EQ.2) WRITE (IREP, 108) ( (ATCOST (KI,KJ) ,KI=1,2) ,KJ-1, 6)
108 FORMAT (
Air Pump repair coat : $ ',F4.0,4X,F4.0,/,
Catalyat replacement coat: $ ',F4.0,4X,F4.0,/,
Miafueled catalyat coat : $ ',F4.0,4X,F4.0,/,
Evap. ayatem repair coat : $ ',F4.0,4X,F4.0,/,
PCV ayatem repair coat : $ ',F4.0,4X,F4.0,/,
Gas cap repair coat : $ ',F4.0,4X,F4.
*
C
IF (JPRGYR.LT.2020 .OR. JPRSYR.LT.2020)
* WRITE(IREP,109)((ATCOST(KI,KJ),KI-1,2),KJ-7,8)
109 FORMAT(
* ' Purge repair coat : $ ',F4.0,4X,F4.0,/,
* ' Preaaure repair coat : $ ',F4.0,4X,F4.0
C * ' ( Purge+Preaa repair ™ Purge repair )'
* )
C
C
999 RETURN
END
90
-------
Appendix A
SUBROUTINE PRNBEN(IREP)
C
INTEGER BYEAR, ATPFLG, PROMPT, ATPFQT, ATPPGM, DISTYP
REAL EVDOL,NETCST,NETBEN
C
CHARACTER*8 VNAME(4)
CHARACTERS KTYPE(4,2),KCOV(2,2),KCOVST(8,2)
CHARACTER*4 TTYPE (4, 3) ,KFREQ (2, 2)
C
COMMON /COST02/ IFEVYR,ILEVYR
COMMON /COST16/ DATPFE, DIMFEE, DIMREP,DATPRE, AATPFE, AIMFEE
COMMON /COST18/ AIMREP,AATPRE,CFACTR,CATPFE(25,10, 4)
C.. Add resting loaaea to output, 8/9/91
COMMON /COST26/ FRSTLS (25, 8, 4, 10) , YRSTLS (10) ,BRSTLS
COMMON /COST27/ IGRFLG,IATPLP, IDIFF, OUTPRF
COMMON /COST28/ JATP,JIM, ISTRT, JCYDX, OPRGYR, JPRSYR
COMMON /COST35/ TCST,TBEN(2) ,ADFBEN,ADIM(2) ,ADAT (2) ,ADET (2)
C Add AEVAF 6/3/91
COMMON /COST36/ YTONS(2,10),BTONS(2),BIDTON(2),AEVAP
COMMON /COST37/ YRCOST(10),YRBEN(10,2),YCOUNT(39),ACOUST
COMMON /COST40/ YEVTON (10) ,BEVTON, YEREF (10) ,BEREF
COMMON /COST41/ YERLOS(10),BERLOS
COMMON /COST49/ PPFEB(25,10,4),APFBEN
C
COMMON /COST51/ ARLOSS , AREVAP , ARSTLS
C
C Write heading
C
WRITE (IREE, 101) IFEVYR, ILEVYR
101 FORMAT(' ',/,
* 8X,'Benefita (1000 tona/yr) and Coats (1000 $/yr) ',
* 'are averaged over the calendar yeara ',14,' thru ',14,'.',/,
* 130('_') )
C
C IFICY = IFEVYR - 1974
C ILICY - ILEVYR - 1974
IFICY - IFEVYR - 1981
ILICY - ILEVYR - 1981
C IDIFF - ILEVYR - IFEVYR + 1
C
C...Separate out exhaust HC from total voc
C
BEXTON - BTONS (1) -BEVTON-BEREF-BERLOS-BRSTLS
C
C Write more headings
C
WRITE(IREP,103)
103 FORMAT(
* 35X,'Gal.',4X,' Program Benefits and Coata ',
* 4X,' Fleet ',9X,'Emissions Without Program Elements ',/,
* 43X,28(' '),17X,50(' •),/,
* 35X,'Year'74X, ' VOC CO Coat ', 4X, ' Size ',
* 4X, ' ExhVOC CO Evap HC Refuel RnLoaa Rat Loo ',/,
* 35X,4(' '),4X,2(7(' '),2X),10('_'),4X,9('_'),
* 2X,6(2X,7T'_')) )
C
C Write benefita for each calendar year
C
IYY - IFICY - 1
DO 10 IY - 1,IDIFF
IYY - IYY + 1
ICY2 - IY + IFEVYR - 1
YEXTON - YTONS (1, IY)-YEVTON (IY)-YEREF
-------
Appendix A
c
c
c
IF(JATP.EQ.l) THEN
ATPBEN = ADAT(l)
NETBEN - TBEN(l) 4- AEVAP
ENDIF
NETCST = TCST - APFBEH
EXHBEN - TBEN(l) - AEVAP
Write total benefit breakdown and total benefits
WRITE (IREP,102) ATPBEN, ADAT (2) , AATPFE,
102 FORMAT(
AEVAP, APFBEN,
ADIM(l) ,ADIM(2) ,AIMFEE,
AATPRE , AIMREP , ADFBEN ,
ADET(l) ,ADET(2) ,
NETBEN,TEEN(2),TCST,ACOUNT,
BEXTON,BTONS (2) , BEVTON,BEREF, BERLOS,BRSTLS
43X,2(7('_'),2X),10('_'),/,
Average Annual ATS Benefit/Fee
Evap/RunLoaa HC s MEG Benefit
Average Annual I/M Benefit/Fee
Avg Ann. ATF + P/P Repair Cost
Average Annual I/M Repair Coat
Average Annual Fuel Savings
Tampering Deterrence
,12X,2(F7.3,2X),F10.0,/,
,12X,F7.3,10X,'(',F10.0,')',/
,12X,F7.3,1X,F8.3,2X,F10.0,/,
,12X,18X,F10.0,/,
,12X,18X,F10.0,/,
,12X,17X,'(',F10.0,')',/,
,12X,2(F7.3,2X),/,
43X,2(7('_'),2X),10('_'),4X,9(' '),2X,6(2X,7('_')),/,
Evap RnLoaa ExhHC',5X,'Annual Average
*
*
* F7.3,IX,F8.3,2X,F10.0,4X,F9.0,2X,2X,F7.3,1X,F8.3,4<2X,F7.3)
* )
C
C Calculate and write total benefits in kilograms per day.
C
KBEHHC - <(TBE»(1)/.365)*2000.)/2.2046
KBENCO - ((TBEN(2)/.365)*2000.)/2.2046
C
WRITE(IREP,105) AREVAP,ARLOSS,EXHBEN,KBENHC,KBENCO,BTONS(1)
105 FORMAT(3(F7.3,1X),17X,'( ',17,19,' kilograms per day )',6X,
* '( ',F7.3,' Total VOC ) ')
C
999 RETURN
END
92
-------
Appendix A
SUBROUTINE EBSIZE (ID,OTR)
C
C
C Calla Function OTCALC
C
C The Evap, Gas Cap and Crankcaae tampering rates are stored
C in BSIZE.
C
C NHG-13 - Evap Canister
C NHG-14 - Crankcaae
C NHG-15 - Gas Cap
C
COMMON /LOOKUP/ IVTAM, IQG, IPG, JPGD,IHG,IGCSF
COMMON /MYCODE/ MY,IDX, JDX,LDXSY,LMYRVT,IAY, IMDXSY, IMKINK
COMMON /TAMEQ4/ BTR(9,2)
COMMON /COST32/ BSIZE(4,25,2,6,17,2),TRAT(9)
C
C For MOBILE4, crankcaae (IH - 3) processing does not change.
C PCV Case
C
IF(ID.EQ.7) THEN
BSIZE(IVTAM,JDX,1,1,14,1) - OTR
C
C EVAP Canister & GAS Cap Cases
C
C NOTE: OTCALC(i,j,k,l)
C
C i - 8 — EVAP + GAS Cap tampering rate
C 6 — EVAP only tampering rate
C j - 1 - A Vehicle's tampering rate previous to ATP start
C 2 " Previous + Subsequent (Overall) tampering rate
C k - 3 - ATP effectiveness code for GAS Cap
C 1 - ATP effectiveness code for EVAP
C 1 - 1 •» Previous tampering effectiveness code
C 2 - Subsequent tampering effectiveness code
C
ELSE IF(ID.EQ.S) THEN
C
C No ATP Case
C
IF(LMYRVT.EQ.l) THEN
BSIZE(IVTAM,JDX,1,1,13,1) - BTR(6,1)
BSIZE (IVTAM,JDX, 1,1, 15,1) - BTR(8,1) - BTR(6,1)
C
C ATP Starts before the vehicle is built
C
ELSE IF(LMYRVT.EQ.2 .AND. LDXSY.GT. JDX) THEN
BSIZE
-------
Appendix A
SUBROUTINE PRNLAP(IREP)
c
c
c
c
G
C
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
PRNLAP writes scenario parameters
Called by PRINT.
common blocks :
/ALUHIN/
/CITRV1/
/FLAGS2/
/FLAGS3/
/FLAGS 4 /
/MAXIMA/
/REGION/
/RESUL1/
/RESUL2/
/RESUL3/
/SCENE!/
/STRING/
/TEMPS/
/USDATA/
/VMXCOM/
AC,XLOAD,TRAILR,ABSHUM,DB,WB
RVPICY
IMFLAG, ALHFLG
ATPFLG
PRTFLG, IDLFLG, NMHFLG, HCFLAG
MAXVEH
IRE JN, ALT
EFFTP , EFEXH, EFEVAP , EFLOSS , EFRUNL
EFIDLE
VFTP , VEXH , VEVAP , VLOS S , VRUNLS , VTDLE
SPD, PCCN, PCHC, PCCC
NOTES , COMMA, PERIOD
TEMEVP , TEMEXH
USNAME, NUSD, IUSD
VMTMIX
Local array subacripta:
NAMEVP (4)
NAMNMH (2)
NAMPOL(3)
NAMREG (3)
NAMVEH (10)
- NAMEVP ( IEVB )
- NAMNMH ( NMHFLG )
- NAMPOL ( IP )
- NAMREG ( IREON )
- NAMVEH ( IVP3 )
Local variable / array dictionary:
Name Type Description
IP1 I
IP2 I
see parameter dictionary
see parameter dictionary
NAMNMH C*8 prefix indicating type of HC (total vs non-methane)
NAMPOL C*4 pollutant names
NAMREG C*4 region names
NAMVEH C*8 vehicle class names
NWITHC I
number of substrings to be written with a comma appended
USC8 C*8 8 character underscore
C
c
c
INTEGER ALHFLG
INTEGER ATPFLG, TPDFLG, RLFLAG , TEMFLG , OUTFMT
INTEGER PRTFLG,HCFLAG
CHARACTER* 8 USNAME
CHARACTER* 4 NOTES, COMMA, PERIOD
CHARACTER*! COLON
CHARACTER*! ASTMCL, ACLASS
CHARACTER* 4 SCNAME
COMMON /ALUHIN/ AC, XLOAD (3) , TRAILR (3) , ABSHUM, DB, WB
COMMON /CITPAR/ SCNAME(4)
COMMON /CITRV1/ RVPBAS,RVPIUS,RVPICY, IUSESY,RVPUWX
COMMON /FLAGS2/ MYMRFG,NEWFLG, IMFLAG, ALHFLG
COMMON /FLAGS3/ ATPFLG,TPDFLG,RLFLAG,LOCFLG,TEMFLG,OUTFMT
COMMON /MAXIMA/ MAXVEH, MAXLTW, MAXPOL, MAXREG, MAXYRS
COMMON /REGION/ FEET(2),IREJN,ALT,INITPR
COMMON /SCENE!/ SPD(8),PCCN,PCHC,PCCC
COMMON /STRING/ NOYES(2),COMMA,PERIOD,COLON
COMMON /TEMPS/ AMBT, TEMMIN, TEMMAX, TEMEXH (3) , TEMEVP (6)
COMMON /USDATA/ USNAME(4,4),NUSD,IUSD(4)
CHARACTER*9 NAMVEH (8) , NAMNMH (2) ,USC8, NAMEVP (4)
CHARACTER* 4 NAMREG (3 ) , NAMPOL (3)
DATA NAMVEH/
LDGV'
LDGT1'
LDGT2'
HDGV1
LDDV
LDDT'
* HDDV
* MC ' /
DATA NAMREG/'Low ','High','Mid '/
94
-------
Appendix A
DATA uses/' '/
c
DATA NAMPOL/1 HC:',1 CO:','HOX:'/
C
DATA NAMNMH/'Total ','Non-Math'/
C
DATA NAMEVP/'Hot Soak ','Diurnal ','Multiple ','Crankcaae'/
C
WRITE(IREP,100)
100 FORMAT(SIX,'Local Parameters Selected: ',/,130(' ') )
C ~
IF(NUSD.EQ.O) GOTO 10
IF(NUSD.EQ.l) WRITE (IREP, 200) (USNAME (ICH, IUSD(1)) ,ICH-1, 4).
* PERIOD
NWITHC-NUSD-1
IF(NWITHC.GT.O) WRITE(IREP,200)
* ( (USNAME (ICH, IUSD (ICT) ) , ICH-1, 4) , COMMA, ICT-1, NWITHC) ,
* (USNAME (ICH, IUSD (NUSD) ) ,ICH-1, 4) ,PERIOD
200 FORMAT(' ','User supplied',4A8,T46,A1,4A8,T78,A1,4A8,T110,A1/
* ' ',T1S,4A8,T46,A1,/)
C
10 WRITE (IREP,215) SCNAME,TEMEXH,NAMREG(IREJN)
215 FORMAT(
* T3,4A4,
* T38,'Ambient Temp:',F5.1,2(' /',F5.1),' (F) ',
* 185,'Region: ',A4)
C
WRITE (IREP, 220) PCCN,PCHC,PCCC,ALT
220 FORMAT)
* T36,'Operating Mode:',F5.1,2(' /',F5.1),
* T83,'Altitude:',F6.0,' Ft.')
C
WRITE(IREP,300)
* RVPBAS,TEMMIN,RVPIUS,IUSESY,TEMMAX
300 FORMAT('0',
* T5, ' Base RVP: ',F4.1,
* T79, 'Minimum Temp: ',F4.0,' (F)1,/,
* T5 ,' In-uae RVP: ',F4.1,
* T35,'In-use Start Tr: ',14,
* T79,'Maximum T«mp: ',F4.0,' (F)')
C
IF(ALHFLG.GT.l) WRITE(IREP,225)
* ABSHUM,
* AC,DB,WB
225 FORMAT(
*' ',T19,'Absolute Humidity:',F6.2,
* T48,'AC (DB / WB):',F5.1,' (',F5.1,' /',F5.1,')')
C
WRITE(IREP,230)
* NAMVEH,(USC8,ITER-1,8),
* SPD
230 FORMAT(
*'0',' Vehicle Type:',7X,8(2X,A8),/,
* ' ',8X,8(2X,A8),/,
*' ',' Veh. Speeds:',7X,8F10.1)
C
IF(ALHFLG.ST.l) WRITE(IREP,235)
* XLOAD,
* TRAILR
235 FORMAT(
*' ',' Extra Load:',F11.3,2F10.3/
*' ','Trlr in Tow:',F11.3,2F10.3)
C
RETURN
END
95
-------
Appendix A
SUBROUTINE GETCEI
c
C GETCEI reads in the Coat Effective Inputs particular
C to the Coat Effective model CEM4.1.
C
C Called by MAIN.
C
C Calls TRUEST.
C
C Input on call:
C
C
C Output on return:
C
C common blocks:
C
C /COST01/ ATPFEE
C /COST07/ CSTIM,GASCST,IMSTRT,ISCEN,MOVIM,MOVATP,REPIM(4)
C /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
C /COST22/ ATCOST(2,8)
C /COST27/ IGRFLG,IATPLP,IDIFF,OUTPRF
C /COST34/ TCOUNT,BYEAR,VGRORT,NCOUNT
C /COST38/ STOCK(12,8),CLDIST(8)
C /COST48/ PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
C /YEARS4/ IY1941,IY1960,IY2020
C
C Notes:
C
C GETCEI was added for CEM4.1 VERSION 2
C
C
INTEGER BYEAR, OUTPRF
C
..COMMON /COST01/ ATPFEE
COMMON /COST07/ CSTIM,GASCST, IMSTRT, ISCEN,MOVIM,MOVATP,REPIM(4)
COMMON /COST12/ IMFXFG, IAFXFG, IMCSFG, IAFEFG, IALDV
COMMON /COST22/ ATCOST(2,8)
COMMON /COST27/ IGRFLG, IATPLP, IDIFF, OUTPRF
COMMON /COST34/ TCOUNT,BYEAR,VGRORT
COMMON /COST48/ PRGFEE,PRSFEE,CEVFEE(25,10,4),TRNFEE
COMMON /YEARS4/ IY1941, IY1960, IY2020
C
C
READ(7,*)
READ (7, 700) IMFXFG,MOVIM, IMSTRT, IMCSFG, IDIFF, OUTPRF
IDIFF - 1
IF(OUTPRF.LT.1.0R.OUTPRF.GT.5) OUTPRF - 3
IF (IMCSFG.EQ. 2) THEN
READ (7,701) CSTIM,REPIM(1) ,REPIM(2) ,GASCST,REPIM(3) ,PRSFEE,
* PRGFEE,TRNFEE,REPIM(4)
ENDIF
C
READ (7, 702) IAFXFG, MOVATP, IAFEFG
IF (IAFEFG.EQ. 2) THEN
READ<7,703) ATPFEE, ((ATCOST(I,J),1-1,2), J-l,8)
ENDIF
C
READ(7,704) IGRFLG
IF (IGRFLG.EQ. 2) THEN
READ (7, 705) VGRORT
READ<7,707) TCOUNT,BYEAR
ENDIF
C
700 FORMAT(II,2(IX,12),3(IX,II))
701 FORMAT(FS.2,2(1X,F6.2) , 1X,F4. 2,1X,F6.2, 3 (1X,F5.2) ,1X,F6.2)
702 FORMAT(II,IX,12,IX,II)
703 FORMAT(F5.2,IX,16(F4.0,1X))
704 FORMAT(II)
705 FORMAT(F4.1)
706 FORMAT(8(F5.3,1X»
707 FORMAT(F9.0,IX,14)
C
C Call YRTEST and change 2 digit years to 4 digit
C
CALL YRTEST (IMSTRT, 32, IY1960, IY2020, INERR)
C
RETURN
C
END
96
-------
APPENDIX B
TECH4 MODEL VERSION 4.1 SOURCE CODE LISTING
-------
B-l
Appendix B : Tech 4.1 Model Source Code
CC
CC..MOBILE4.1 I/M Credit Model for 1981 and newer LDGV
CC With Oxygenated Fuel Effects
CC (May 14, 1991)
CC With Bag Fractions
CC (July 1991)
CC With Calculated Repair Effects
CC (July 20, 1991)
CC
CC..Program Main
CC
CC..COMMON Blocks and DIMENSION Statements
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE, ICUT, ITST
COMMON /DAT02/ AMIL(25) ,ODOM(25),TMILE(25),WGT(25)
COMMON /DAT03/ ENOX(4,2,2)
COMMON /DAT04/ FAIL(25,4,2)
COMMON /DAT05/ OXYE(2,3,4,3),OXYGON
COMMON /DAT06/ FRAC(4,12)
COMMON /DAT07/ ESO(2,4,4,2),EHO(2,4,4, 2)
COMMON /DAT08/ DN(3,4,4,2)
COMMON /DAT09/ ZMIL(2,4,4,2),CWO(2,4,25,4,2),CIMW(2, 4,25, 4, 2, 3)
COMMON /DAT10/ EWO(3,4,25,12),EIMW(2,4,25,12,3),EZM(2, 4,12)
COMMON /DAT11/ CREDIT(2,25,12,3, 4)
COMMON /DAT12/ ZML(3,4,12),ZML1(3,4,12),ZML2(3,4,12)
COMMON /DAT13/ BFZML1(3,4,12),BFDET1(3,4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3, 4,12)
COMMON /DAT15/ CWOA(2,25,4,2),EWOA(3,25,12),EZMA(2,12)
COMMON /DAT16/ XSIDR(2,4,2,3),XHIDR(2, 4,2, 3)
COMMON /DAT17/ RSUP (2, 4, 2, 3) , RHIG (2, 4, 2, 3)
COMMON /DAT18/ SIZE(25,4,2,4)
COMMON /DAT19/ GV(4,2),GH(4,2),GS(4,2), BH(4, 2) ,BV(4, 2)
COMMON /DAT20/ EMO(2,4,4,2),ENO(3,4, 4, 2)
COMMON /DAT21/ BFDET2 (3,4,12),BFZML2(3,4,12)
COMMON /DAT22/ RMAR(2,4,2,3),RNOR(2, 4, 2, 3)
COMMON /DAT24/ XMIDR(2,4,2,3),XNIDR(2,4,2, 3)
COMMON /DAT25/ EXPO (2,4,3,2),DOTS(4,2,4, 2,2)
CC
C OPEN(1,FILE-'M5SIZE')
C OPEN(2,FILE-•M5TEST')
C OPEN(3,FILE-'M5MYRf)
C OPEN(4,FILE-'M5REG1)
OPEN(13,FILE-'T5V7OXY.CSV')
OPEN(14,FILE-'AAAAA')
CC
CC
CC..Calculate the mileage accumulated in each one year interval.
CC
AMIL(l) - ODOM(l)
CC
DO 10 IAGE-2,25
AMIL(IAGE) - ODOM(IAGE) - ODOM(IAGE-l)
10 CONTINUE
CC
CC..Model Year Grouping
CC..ISTD - 1 : 1981,1982 Model Year.s
CC..ISTD - 2 : 1983 and newer Model Years
CC
DO 600 ISTD=1,2
CC
-------
B-2
Appendix B : Tech 4.1 Model Source Code
CC..ITECH indicates the technology type used in the vehicles.
CC
CC..ITECH - 1 : Closed-Loop, Multi-Point Fuel Injected
CC..ITECH - 2 : Closed-Loop, Throttle-Body Fuel-Injected
CC..ITECH - 3 : Closed-Loop, Carbureted
CC..ITECH - 4 : Open-Loop, Any Carbureted or Fuel Injected
CC
DO 600 ITECH-1,4
CC
CC..Vehicle age in years
CC
DO 600 IAGE-1,25
CC
CALL EGROUP
CC
CC..FTP Bag ( 1-FTP; 2-BAG1; 3-BAG2; 4-BAG3 )
CC
DO 600 IBAG-1,4
CC
CC..Pollutant ( 1:HC, 2:CO )
CC
DO 600 IP-1,2
CC
CALL EMIT
CALL IMEMIT
CC
600 CONTINUE
CC
CALL MYRSUB
CALL REGR
CALL JAN1
CC
OPEN(7,FILE-'T5V7EF.DAT')
OPEN(8,FILE-'T5V7ANN.IMC')
OPEN(9,FILE-'T5V7BI.IMC')
OPEN(10,FILE-'T5V7BF.DAT')
CC
CALL OUTPUT
CC
C CLOSE(1)
C CLOSE(2)
C CLOSE(3)
C CLOSE(4)
CLOSE(7)
CLOSE(8)
CLOSE(9)
C CLOSE(10)
CLOSE(13)
CLOSE(14)
CC
STOP
END
-------
B-3
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE EGROUP
CC
CC..This routine predicts the number of vehicles in each emission
CC..level category by technology and age.
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT, ITST
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25),WGT(25)
COMMON /DAT04/ FAIL(25,4,2)
COMMON /DAT18/ SIZE (25,4,2,4)
COMMON /DAT19/ GV(4,2),GH(4,2),GS(4,2),BH(4, 2) ,BV(4, 2)
CC
CC..Calculate the number of "HIGH" emitting vehicles
CC
SIZE(IAGE,ITECH,ISTD,2) - GH(ITECH,ISTD) * ODOM(IAGE)
CC
CC.."BH" is the change in the rate of occurrance of "HIGH"
CC..emitting vehicles assumed to occur at 50,000 miles.
CC
IF(IAGE.GT.l .AND. ODOM(IAGE-l).GT.5.0)
* SIZE(IAGE,ITECH,ISTD,2) - SIZE(IAGE-1,ITECH,ISTD,2)
* + BH(ITECH,ISTD)*GH(ITECH,ISTD)*AMIL(IAGE)
IF(SIZE(IAGE,ITECH,ISTD,2).GT.1.00) SIZE(IAGE,ITECH,ISTD,2) - 1.00
IF(SIZE(IAGE,ITECH,ISTD,2).LT.0.00) SIZE(IAGE,ITECH,ISTD,2) - 0.00
CC
CC..Calculate the number of "VERY HIGH" emitting vehicles
CC
SIZE(IAGE,ITECH,ISTD,3) - GV(ITECH,ISTD) * ODOM(IAGE)
CC
CC.."BV" is the change in the rate of occurrance of "VERY HIGH"
CC..emitting vehicles assumed to occur at 50,000 miles.
CC
IF(IAGE.GT.l .AND. ODOM(IAGE-l).GT.5.0)
* SIZE(IAGE,ITECH,ISTD,3) =• SIZE(IAGE-1,ITECH,ISTD,3)
* + BV(ITECH,ISTD)*GV(ITECH, ISTD)*AMIL(IAGE)
IF(SIZE(IAGE,ITECH,ISTD,3).GT.1.00) SIZE(IAGE,ITECH,ISTD,3) - 1.00
IF(SIZE(IAGE,ITECH, ISTD, 3) .LT.O.O(X) SIZE (IAGE, ITECH, ISTD, 3) = 0.00
CC
CC..Calculates the number of "SUPER" emitting vehicles
CC
SIZE(IAGE,ITECH,ISTD,4) - GS(ITECH,ISTD) * ODOM(IAGE)
IF(SIZE(IAGE,ITECH,ISTD,4).GT.1.00) SIZE(IAGE,ITECH,ISTD, 4) - 1.00
IF(SIZE(IAGE,ITECH,ISTD,4).LT.0.00) SIZE(IAGE,ITECH,ISTD, 4) = 0.00
CC
CC..Check for logical sizes
CC
CHECK - SIZE(IAGE,ITECH,ISTD,3)
* + SIZE(IAGE,ITECH,ISTD,4)
IF(CHECK.GT.l.O)
* SIZE(IAGE,ITECH,ISTD,3) - 1.0 - SIZE(IAGE,ITECH,ISTD,4)
CC
CHECK - SIZE(IAGE,ITECH,ISTD,2)
* + SIZE(IAGE,ITECH,ISTD,3)
* + SIZE(IAGE,ITECH,ISTD,4)
IF(CHECK.GT.1.0)
* SIZE(IAGE,ITECH,ISTD,2) - 1.0 - SIZE(IAGE,ITECH,ISTD, 3)
* - SIZE(IAGE,ITECH,ISTD,4)
CC
CC..Calculate the number of "NORMAL" emitting vehicles
CC
SIZE(IAGE,ITECH,ISTD,1) -1.0
-------
B-4
Appendix B : Tech 4.1 Model Source Code
* - SIZE(IAGB,ITECH,ISTD,2)
* - SIZE(IAGE,ITECH,ISTD,3)
* - SIZE(IAGE,ITECH,ISTD,4)
C WRITE(1,701) IAGE,ITECH,ISTD,(SIZE(IAGE,ITECH,ISTD,J),J-1,4)
C 701 FORMAT(313,4F8.4)
CC
999 RETURN
END
-------
B-5
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE EMIT
CC
CC..This routine combines the emission levels of each emission
CC..category based on the predicted category size.
CC
COMMON /DAT01/ MYR, ISTD, ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25) ,WGT(25)
COMMON /DAT07/ ESO(2,4,4,2),EHO(2, 4, 4, 2)
COMMON /DAT08/ DN<3,4,4,2)
COMMON /DAT09/ ZMIL(2,4,4,2),CWO<2,4,25,4, 2) ,CIMW(2, 4,25, 4,2, 3)
COMMON /DAT15/ CWOA(2,25,4,2),EWOA(3,25,12),EZMA(2,12)
COMMON /DAT18/ SIZE(25,4,2,4)
COMMON /DAT19/ GV(4,2),GH(4,2),GS(4,2),BH(4,2) ,BV(4, 2)
COMMON /DAT20/ EMO (2,4,4,2),ENO(3,4, 4,2)
CC
IF(IAGE.GT.l) GOTO 10
CC
CC..Emission levels at zero mileage point
CC
ZMIL(IP, IBAG,ITECH,ISTD) - END(IP,IBAG,ITECH,ISTD)
CC
CC..Emission levels by age
CC
10 ES - ESO(IP,IBAG,ITECH,ISTD)
EH - EHO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP,IBAG,ITECH,ISTD)*ODOM(IAGE) )
EM - EMO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP,IBAG,ITECH,ISTD)*ODOM(IAGE) )
EN - ENO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP,IBAG,ITECH,ISTD)*ODOM(IAGE) )
CC
CC Adjust emissions for effects of Oxygenated Fuels
CC
ENA=EN*OXY(IP,ITECH,EN)
EMA-EM*OXY(IP,ITECH,EM)
EHA-EH*OXY(IP,ITECH, EH)
ESA-ES*OXY(IP,ITECH, ES)
CC
CC..Calculate the base (without I/M) composite emission levels by age
CC
CWO(IP,IBAG,IAGE,ITECH,ISTD) -
* SIZE(IAGE,ITECH,ISTD,1) * EN
* + SIZE(IAGE,ITECH,ISTD,2) * EM
* + SIZE(IAGE,ITECH,ISTD,3) * EH
* + SIZE(IAGE,ITECH,ISTD,4) * ES
IF(IBAG.EQ.l)
* CWOA(IP,IAGE,ITECH,ISTD) -
* SIZE(IAGE,ITECH,ISTD,1) * ENA
* + SIZE(IAGE,ITECH,ISTD,2) * EMA
* + SIZE(IAGE,ITECH,ISTD,3) * EHA
* + SIZE(IAGE,ITECH,ISTD,4) * ESA
CC
C IF(IP.EQ.LAND.IBAG.EQ.l)
C * WRITE(2,772) IAGE,ITECH,ISTD,EN,EM, EH,ES
C 772 FORMAT(314,4F7.3)
CC
999 RETURN
END
-------
B-6
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE IMEMIT
CC
CC..This routine combines the emission levels of each emission
CC..category based on the predicted catagory size.
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25),WGT(25)
COMMON /DAT07/ ESO(2,4,4,2),EHO(2,4,4,2)
COMMON /DAT08/ DN(3,4,4,2)
COMMON /DAT09/ ZMIL(2,4,4,2),CWO(2,4,25, 4, 2) ,CIMW(2, 4, 25,4,2, 3)
COMMON /DAT16/ XSIDR(2,4,2,3),XHIDR(2,4,2,3)
COMMON /DAT17/ RSUP(2, 4, 2, 3), RHIG(2, 4, 2, 3)
COMMON /DAT18/ SIZE(25,4,2,4)
COMMON /DAT20/ EMO(2,4,4,2),END(3,4,4,2)
COMMON /DAT22/ RMAR(2, 4, 2, 3) ,RNOR(2, 4, 2, 3)
COMMON /DAT24/ XMIDR(2,4,2,3),XNIDR(2,4,2,3)
COMMON /DAT25/ EXPO(2,4,3,2),DOTS(4,2,4,2,2)
CC
DIMENSION STD(2)
DATA STD / .41, 3.4 /
CC
CC..Non-I/M emission levels
CC
ES2 - ESO(IP,IBAG,ITECH,ISTD)
EH2 - EHO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP, IBAG, ITECH, ISTD) *ODOM(IAGE) )
EM2 - EMO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP,IBAG,ITECH,ISTD)*ODOM(IAGE) )
EN2 - ENO(IP,IBAG,ITECH,ISTD)
* + ( DN(IP,IBAG,ITECH,ISTD)*ODOM(IAGE) )
CC
CC..For each test type ITEST - 1 : Idle Test
CC.. ITEST - 2 : IM240 with 1.6/30 HC/CO Cutpoints
CC.. ITEST - 3 : IM240 with 0.8/15 HC/CO Cutpoints
CC
DO 10 ITEST-1,3
CC
CC..The emissions of vehicles passing the short test are combined
CC..with the estimated emission levels of vehicles which are repaired.
CC
CC DOTS(ITECH,A/B,IDOT,IP,ITST)
CC
ITST-ITEST
IF(ITEST.EQ.3) ITST-2
CC
C IF(ES2.GT.O) REPS - EXPO(IP,ITECH,ITEST,1)
C * * EXP(-EXPO(IP,ITECH,ITEST,2)/ES2)
CALL CONECT(ES2,REPS)
IF(REPS.LT.STD(IP) .AND. ES2.GT.STD(IP)) REPS-STD(IP)
IF(REPS.GT.ES2) REPS-ES2
IF(ES2.LT.STD(IP)) REPS-ES2
EIMS =• (XSIDR(IP, ITECH, ISTD, ITEST) *REPS) +
* ((1 - XSIDR(IP,ITECH,ISTD,ITEST))*ES2)
CC
C REPH - EXPO(IP,ITECH,ITEST,1)
C * * EXP(-EXPO(IP,ITECH,ITEST, 2)/EH2)
CALL CONECT(EH2,REPH)
IF(REPH.LT.STD(IP) .AND. EH2.GT.STD(IP)) REPH-STD(IP)
IF(REPH.GT.EH2) REPH-EH2
IF(EH2.LT.STD(IP)) REPH-EH2
-------
B-7
Appendix B : Tech 4.1 Model Source Code
EIMH - (XHIDR(IP,ITECH,ISTD,ITEST)*REPH) +
* ((1 - XHIDR(IP,ITECH,ISTD,ITEST))*EH2)
CC
C REPM - EXPO(IP,ITECH,ITEST,1)
C * * EXP(-EXPO(IP,ITECH,ITEST,2)/EM2)
CALL CONECT(EM2,REPM)
IF(REPM.LT.STD(IP) .AND. EM2.GT.STD(IP)) REPM-STD(IP)
IF(REPM.GT.EM2) REPM=EM2
IF(EM2.LT.STD(IP)) REPM-EM2
EIMM = (XMIDR(IP,ITECH,ISTD,ITEST)*REPM) +
* (d - XMIDR(IP, ITECH,ISTD, ITEST) )*EM2)
CC
C REPN - EXPO(IP,ITECH,ITEST,1)
C * * EXP(-EXPO(IP,ITECH,ITEST,2)/EN2)
CALL CONECT(EN2,REPN)
IF(REPN.LT.STD(IP) .AND. EN2.GT.STD(IP)) REPN-STD(IP)
IF(REPN.GT.EN2) REPN-EN2
IF(EN2.LT.STD(IP)) REPN-EN2
EIMN - (XNIDR(IP,ITECH,ISTD,ITEST)*REPN) +
* ((1 - XNIDR(IP,ITECH,ISTD,ITEST))*EN2)
CC
PS=-XSIDR(IP, ITECH, ISTD, ITEST)
PH-XHIDR(IP,ITECH,ISTD,ITEST)
PM-XMIDR(IP,ITECH,ISTD,ITEST)
PN-XNIDR(IP,ITECH,ISTD,ITEST)
IF(IBAG.EQ.l .AND. IAGE.EQ.1 .AND. IP.EQ.l)
* WRITE(14,881) ISTD,ITECH,IP,ITEST,IAGE,REPS,REPH,REPM,REPN,
* ES2,EH2,EM2,EN2,EIMS,EIMH,EIMM,EIMN,PS,PH,PM,PN
881 FORMAT(5I3,12F7.2,4F7.3)
CC
CC..Emission levels by age and by test
CC..Calculate the base (with I/M) composite emission levels by age
CC
CIMW(IP,IBAG,IAGE,ITECH,ISTD,ITEST) -
* SIZE(IAGE,ITECH,ISTD,1) * EIMN
* + SIZE(IAGE,ITECH,ISTD,2) * EIMM
* + SIZE(IAGE,ITECH,ISTD,3) * EIMH
* + SIZE(IAGE,ITECH,ISTD,4) * EIMS
CC
C IF(ITEST.EQ.3) THEN
C CHK1-CIMW(IP,IBAG,IAGE,ITECH,ISTD,ITEST)
C CHK2-CIMW(IP,IBAG,IAGE,ITECH,ISTD,ITEST-1)
C IF(CHK1.GT.CHK2) THEN
C CIMW(IP,IBAG,IAGE,ITECH,ISTD,ITEST)-CHK2
C WRITE (14, 882) ISTD, ITECH, IP, ITEST, IAGE, REPS, REPH, REPM, REPN,
C * ES2,EH2,EM2,EN2,EIMS,EIMH,EIMM,EIMN,PS,PH,PM,PN
C 882 FORMAT(5I3,12F7.3,4F7.3)
C ENDIF
C ENDIF
CC
CC
10 CONTINUE
CC
999 RETURN
END
-------
B-8
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE MYRSUB
CC
CC..This section combines the technologies into
CC..model year emission levels.
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25),WGT(25)
COMMON /DAT06/ FRAC(4,12)
COMMON /DAT08/ DN(3,4,4,2)
COMMON /DAT09/ ZMIL(2,4,4,2),CWO(2,4,25,4,2),CIMW<2,4,25,4,2,3)
COMMON /DAT10/ EWO(3,4,25,12),EIMW<2,4,25,12,3),EZM<2,4,12)
COMMON /DAT12/ ZML (3, 4,12), ZML1 (3, 4,12) , ZML2 (3, 4,12)
COMMON /DAT15/ CWOA(2,25,4, 2),EWOA(3,25,12),EZMA(2,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
COMMON /DAT20/ EMO(2,4,4,2),END(3,4,4,2)
CC
DIMENSION EFFECT(12,25)
CC
ZERO=0.0
CC
CC..Loop by MYR, CO standard, technology, age, bag, & pollutant
CC
CC..The ITEST loops only for the I/M composite emission arrays
CC
CC
DO 300 MYR-1,12
ISTD-1
IF(MYR.GE.3) ISTD-2
DO 300 IP-1,2
DO 300 IBAG-1,4
DO 300 ITECH-1,4
CC
CC..Zero mile emission levels by model year
CC
EZM(IP,IBAG,MYR) - EZM(IP,IBAG,MYR)
* + FRAC(ITECH,MYR) * ZMIL(IP,IBAG,ITECH,ISTD)
CC
CC Oxygenated Fuel Effect
CC
IF(IBAG.EQ.l) EZMA(IP,MYR)-EZMA(IP,MYR)
* + FRAG(ITECH,MYR)*ZMIL(IP,IBAG,ITECH,ISTD)
* * OXY(IP, ITECH, ZMIL(IP,IBAG,ITECH,ISTD))
CC
DO 300 IAGE-1,25
CC
CC..Calculates the emission levels for
CC..January 1st dates from the emission levels by age.
CC..Since model year introduction is on October 1st, this
CC..requires a 75%/25% staggering.
CC
EWO (IP, IBAG, IAGE,MYR) »
* EWO(IP,IBAG,IAGE,MYR)
* + FRAC(ITECH,MYR) * CWO(IP,IBAG,IAGE,ITECH,ISTD)
CC
CC Oxygenated Fuel Effect
CC
IF(IBAG.EQ.l)
* EWOA(IP,IAGE,MYR) =• EWOA(IP, IAGE,MYR)
* + FRAC(ITECH,MYR) * CWOA(IP,IAGE,ITECH,ISTD)
CC
-------
B-9
Appendix B : Tech 4.1 Model Source Code
DO 200 ITEST-1,3
CC
EIMW(IP,IBAG,IAGE,MYR, ITEST) =
* EIMW(IP,IBAG,IAGE,MYR,ITEST)
* + FRAC(ITECH,MYR) * CIMW(IP,IBAG,IAGE,ITECH,ISTD, ITEST)
CC
CC
200 CONTINUE
300 CONTINUE
CC
CC Write out the Oxygenated Fuels Effect
CC
DO 774 MYR-1,12
ISTD=1
IF(MYR.GE.3) ISTD-2
IAGE-0
CC
C WRITE(13,773) MYR, IAGE,ZERO,
C * E2M(1,1,MYR),EZMA(1,MYR),
C * EZM(2,1,MYR),EZMA(2,MYR)
CCCCCC * EZM(3,1,MYR),EZMA(3,MYR)
CC
DO 774 IAGE-1,25
DO 772 ITECH-1,4
EMIT-ENO(3,1,ITECH,ISTD)+DN(3,1,ITECH,ISTD)*ODOM(IAGE)
EWO(3,1,IAGE,MYR)-EWO(3,1,IAGE,MYR)
* + EMIT*FRAC(ITECH,MYR)
EWOA(3,IAGE,MYR)-EWOA(3,IAGE,MYR)
* + EMIT*FRAC(ITECH,MYR)*OXY(3,ITECH, EMIT)
772 CONTINUE
CC
EFFECT(MYR,IAGE) -
* (EWO(2,1,IAGE,MYR)-EWOA(2,IAGE,MYR))
* / EWO(2,1,IAGE,MYR)
CC
C WRITE(13,773) MYR,IAGE,ODOM(IAGE) ,
C * EWO (1, 1, IAGE,MYR) , EWOA (1, IAGE,MYR) ,
C * EWO (2,1, IAGE,MYR) , EWOA (2, IAGE,MYR) ,
C * EWO (3,1, IAGE,MYR) , EWOA (3, IAGE, MYR)
C 773 FORMAT(2(I4,','),F9.4,',',6(F8.3, ', '))
CC
774 CONTINUE
CC
DO 10 MYR-1,12
WRITE(13,775) (EWO(2,1,IAGE,MYR),IAGE-1, 25)
10 CONTINUE
DO 20 MYR-1,12
WRITE(13,776) (EFFECT(MYR,IAGE),IAGE=1,25)
20 CONTINUE
775 FORMAT(5X,f*',25(F7.2, ', '))
776 FORMAT(5X,'*',25(F4.3, ', '))
CC
999 RETURN
END
-------
B-10
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE REGR
CC
CC..This subroutine uses a weighted regression equation to
CC..linearize the emission level results for each model year.
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25),WGT(25)
COMMON /DAT03/ ENOX(4,2,2)
COMMON /DAT06/ FRAC(4,12)
COMMON /DAT07/ ESO(2,4,4,2),EHO(2,4,4,2)
COMMON /DAT08/ DN(3,4,4,2)
COMMON /DAT10/ EWO(3f4,25,12),EIMW(2,4,25,12,3),EZM(2,4,12)
COMMON /DAT12/ ZML (3, 4,12) , ZML1 (3, 4,12) , ZML2 (3, 4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
COMMON /DAT20/ EMO(2,4,4,2),ENO(3,4,4,2)
CC
DO 40 MYR-1,12
DO 40 IBAG=1,4
DO 40 IP-1,2
CC
SUMX - 0.0
SUMY - 0.0
SUMXY - 0-0
SUMXX - 0.0
CC
N - 5
CC
DO 10 IAGE-1,N
CC
IF(IAGE.EQ.l) EM - EZM(IP, IBAG,MYR)
IF(IAGE.GT.l) EM - EWO(IP,IBAG,IAGE-1,MYR)
CC
IF(IAGE.EQ.l) XM - 0.0
IF(IAGE.GT.l) XM - ODOM(IAGE-l)
CC
SUMX - SUMX + XM
SUMY - SUMY + EM
SUMXY - SUMXY + (XM*EM)
SUMXX - SUMXX + (XM**2)
CC
C IF(IP.EQ.LAND.IBAG.EQ.l)
C * WRITE(4,774) IAGE,MYR,EM,XM,SUMX,SUMY,SUMXY,SUMXX
C 774 FORMAT(214,6F8.4)
CC
10 CONTINUE
CC
SUM1 - N * SUMXY - SUMX * SUMY
SUM2 - N * SUMXX - SUMX**2
Dl - SUM1 / SUM2
Zl - (SUMY/N) - Dl * (SUMX/N)
CC
CC..Store the regression results
CC
ZML1(IP,IBAG,MYR) - Zl
DET1(IP,IBAG,MYR) - Dl
ZML2(IP,IBAG,MYR) - ZML1(IP,IBAG,MYR) + DET1(IP,IBAG,MYR)*5.0
CC
IF(ZML1(IP,IBAG,MYR).GE.0.0) GO TO 30
CC
CC..If the emission level at zero miles is less than zero,
-------
B-ll
Appendix B : Tech 4.1 Model Source Code
CC..then the regression is altered to intercept at zero
CC
ZML1(IP,IBAG,MYR) - 0.0
DET1(IP,IBAG,MYR) - SUMXY / SUMXX
ZML2(IP,IBAG,MYR) - ZML1 (IP, IBAG,MYR) + DET1 (IP, IBAG,MYR) *5. 0
CC
30 SUMX =0.0
SUMY =0.0
SUMXY =0-0
SUMXX =0.0
CC
Ml = 25 - N
M2 = N + 1
CC
DO 20 IAGE=M2,25
CC
IF(IAGE.EQ.l) EM - EZM(IP,IBAG,MYR)
IF(IAGE.GT.l) EM = EWO(IP,IBAG,IAGE-1,MYR)
CC
IF(IAGE.EQ.l) XM - 0.0
IF(IAGE.GT.l) XM = ODOM(IAGE-l)
CC
SUMX = SUMX + XM
SUMY = SUMY + EM
SUMXY = SUMXY + (XM*EM)
SUMXX = SUMXX + (XM**2)
CC
20 CONTINUE
CC
SUM1 - Ml * SUMXY - SUMX * SUMY
SUM2 - Ml * SUMXX - SUMX**2
Dl = SUM1 / SUM2
Zl - (SUMY/MI) - Dl * (SUMX/M1)
CC
CC..Store the regression results
CC
DET2(IP,IBAG,MYR) - Dl
CC
CC..Single Linear Regression
CC
SUMX =0.0
SUMY =0.0
SUMXY =0.0
SUMXX =0.0
C SUMW =0.0
CC
DO 60 IAGE-1,25
CC
IF(IAGE.EQ.l) EM = EZM(IP,IBAG,MYR)
IF(IAGE.GT.l) EM = EWO(IP,IBAG,IAGE-1,MYR)
CC
IF(IAGE.EQ.l) XM = 0.0
IF(IAGE.GT.l) XM = ODOM(IAGE-l)
CC
SUMX = SUMX + ( WGT(IAGE) * XM )
SUMY = SUMY + ( WGT(IAGE) ,* EM )
SUMXY = SUMXY + ( WGT(IAGE) * (XM*EM))
SUMXX = SUMXX + ( WGT(IAGE) * (XM**2))
CC
60 CONTINUE
-------
B-12
Appendix B : Tech 4.1 Model Source Code
CC
SUM1 - SUMXY - SUMX * SUMY
SUM2 - SUMXX - SUMX**2
Dl - SUM1 / SUM2
Zl - SUMY - Dl * SUMX
CC
CC..Store the regression results
CC
ZML (IP, IBAG,MYR) = 211
DET(IP,IBAG,MYR) - Dl
CC
IF(ZML(IP,IBAG,MYR).GE.0.0) GO TO 40
CC
CC..If the emission level at zero miles is less than zero,
CC..then the regression is altered to intercept at zero.
CC
ZML(IP,IBAG,MYR) - 0.0
DET (IP, IBAG,MYR) - SUMXY / SUMXX
CC
40 CONTINUE
CC
CC..Since the NOx emissions are not combined from emission level
CC..groups, the NOx emission factors can be calculated directly
CC..from the regressions.
CC
IP-3
CC
DO 50 MYR-1,12
ISTD-1
IF(MYR.GE.3) ISTD-2
DO 50 IBAG-1,4
DO 50 ITECH-1,4
CC
ZML(IP,IBAG,MYR) - ZML (IP, IBAG,MYR) +
* ENO(IP,IBAG,ITECH,ISTD)*FRAC(ITECH,MYR)
CC
DET(IP,IBAG,MYR) - DET(IP,IBAG,MYR) +
* DN(IP,IBAG,ITECH,ISTD)*FRAC(ITECH,MYR)
CC
ZML1
-------
B-13
Appendix B : Tech 4.1 Model Source Code
DET2(IP,1,MYR) =DET1(IP,1,MYR)
56 CONTINUE
CC
CC
CALL BAGF
CC
999 RETURN
END
-------
B-14
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE BAGF
CC
CC..This routine calculates the bag fractions for hot/cold starts
CC
CC..Last Updated : November 15, 1988
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT12/ ZML (3, 4, 12), ZML1 (3, 4,12), ZML2 (3, 4, 12)
COMMON /DAT13/ BFZML1(3,4,12),BFDET1(3,4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
COMMON /DAT21/ BFDET2(3,4,12),BFZML2(3,4,12)
CC
DIMENSION BFRAC(4)
CC
DATA BFRAC / 1.000, 0.206, 0-521, 0.273 /
CC
DO 20 IP-1,3
DO 20 MYR-1,12
CC
Z2 - 0.0
Z3 = 0.0
D2 - 0.0
D3 - 0.0
CC
CC..Sum up the bag regression coeffs weighted by the FTP bag fractions
CC
DO 10 IBAG-2,4
CC
Z2 - Z2 + ZML1(IP,IBAG,MYR) * BFRAC(IBAG)
Z3 » Z3 + ZML2(IP,IBAG,MYR) * BFRAC(IBAG)
D2 =- D2 + DET1(IP,IBAG,MYR) * BFRAC (IBAG)
D3 - D3 + DET2(IP,IBAG,MYR) * BFRAC(IBAG)
CC
10 CONTINUE
CC
CC..Set the combined FTP bag fraction to 1.00
CC
BFZML1(IP,1,MYR) - Z2 / Z2
BFZML2(IP,1,MYR) - Z3 / Z2
BFDET1(IP,1,MYR) - D2 / Z2
BFDET2(IP,1,MYR) - D3 / Z2
CC
CC..Divide each bag regression coeff by the weighted sum
CC
DO 20 IBAG-2,4
CC
BFZML1(IP,IBAG,MYR) - ZML1(IP,IBAG,MYR) / Z2
BFZML2(IP,IBAG,MYR) - ZML2(IP,IBAG,MYR) / Z2
BFDET1(IP,IBAG,MYR) - DET1(IP,IBAG,MYR) / Z2
BFDET2(IP,IBAG,MYR) - DET2(IP,IBAG,MYR) / Z2
CC
20 CONTINUE
CC
RETURN
END
-------
B-15
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE JAN1
CC
CC..This subroutine calculates the average emissions of each
CC..model year on January first. It creates the I/M credits
CC..and passes them to OUTPUT.
CC
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25) ,WGT(25)
COMMON /DAT10/ EWO(3,4,25,12),EIMW(2,4,25,12,3) ,EZM<2, 4,12)
COMMON /DAT11/ CREDIT(2,25,12,3, 4)
COMMON /DAT12/ ZML(3, 4, 12),ZML1(3,4,12),ZML2(3,4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
CC
DIMENSION ANSWNO(2,25,12),ANSWIM(2,25,12, 3, 3)
DIMENSION EPRED(2,4,25,12,3),PRED(2,4,25, 12, 3)
DIMENSION SLOPE(2,12,26),ZERO(2,12, 26)
CC
CC
IBAG = 1
DO 100 MYR - 1,12
DO 100 IP - 1,2
CC
CC..The deteriorations before and after the "KINK" are transferred
CC..to the array SLOPE for each vehicle age.
CC
DO 95 I =- 1,25
CC
IF(I.LE.4) SLOPE(IP,MYR,I) - DET1(IP,IBAG,MYR)
IF(I.GE.5) SLOPE(IP,MYR,I) - DET2(IP,IBAG,MYR)
CC
IF(I.LE.4) ZERO(IP,MYR, I) - ZML1(IP,IBAG,MYR)
IF(I.GE.S) ZERO(IP,MYR, I) - ZML2(IP,IBAG,MYR) -
* DET2(IP,IBAG,MYR)*5.0
CC
95 CONTINUE
CC
CC..Computes the NON-I/M emission level by age, by pollutant,
CC..by bag, and by myr.
CC
DO 100 IAGE - 1,24
CC
IF(IAGE .GT. 1) GOTO 33
CC
CC..Vehicle age is one.
CC
ANSWNO(IP,IAGE,MYR) -
*.75*(ZERO(IP,MYR,IAGE) + SLOPE(IP,MYR,IAGE)*.625*ODOM(IAGE) ) 4
*.25* (ZERO(IP,MYR,IAGE+1) + SLOPE(IP,MYR,IAGE+1)*
* (.125*(ODOM(IAGE+1)-ODOM(IAGE))+ODOM(IAGE)) )
GOTO 34
CC
CC..Vehicle age is greater than one.
CC
33 ANSWNO(IP,IAGE,MYR) -
*.75*(ZERO(IP,MYR,IAGE) + SLOPE(IP,MYR,IAGE) *
* (,625*(ODOM(IAGE)-ODOM(IAGE-1))+ODOM(IAGE-1)) ) +
*.25*(ZERO(IP,MYR,IAGE+1) + SLOPE(IP,MYR,IAGE+1) *
* (.125*(ODOM(IAGE+1)-ODOM(IAGE))+ODOM(IAGE)) )
CC
CC..Compute the I/M emission level by age, by pollutant,
CC..by bag, by myr and by test for IAGE - 1. The predicted emission
-------
B-16
Appendix B : Tech 4.1 Model Source Code
CO..level is from the regression equation and the actual model
CC..emission level points.
CC
34 DO 100 ITEST-1,3
CC
EPRED(IP,IBAG,IAGE,MYRrITEST) -
* 1 - ((EWO(IP,IBAG,IAGE,MYR) - EIMW(IP,IBAG,IAGE,MYR,ITEST)) /
* EWO(IP ,IBAG,IAGE,MYR))
CC
PRED(IP,IBAG,IAGE,MYR, ITEST) -
* (ZERO (IP,MYR,IAGE) + SLOPE(IP,MYR,IAGE)*ODOM(IAGE)) *
* EPRED(IP,IBAG,IAGE,MYR,ITEST)
CC
CC..Determine I/M credits for each inspection frequency.
CC
CC ITYP => 1 Annual
CC 2 Biennial 1 - 3 - 5 - etc
CC 3 Biennial 2 - 4 - 6 - etc
CC
DO 110 ITYP -1,3
CC
IF(ITYP.GE.2) GOTO 60
CC
CC..Annual I/M Credits
CC
IF(IAGE .GT. 1) GOTO 50
CC
ANSWIM(IP,IAGE,MYR,ITEST,ITYP) -
* (.75*(.625*SLOPE(IP,MYR,IAGE)*ODOM(IAGE) + ZERO(IP,MYR,IAGE)))
* + .25*(PRED(IP,IBAG,IAGE,MYR,ITEST) + SLOPE(IP,MYR,IAGE+1)*
* (.125*(ODOM(IAGE+1)-ODOM(IAGE))) )
GOTO 60
CC
50 ANSWIM(IP,IAGE,MYR,ITEST,ITYP) -
*.75*(PRED(IP,IBAG,IAGE-1,MYR,ITEST) +
* SLOPE(IP,MYR,IAGE) * (.625*(ODOM(IAGE)-ODOM(IAGE-1))) ) +
*.25*(PRED(IP,IBAG,IAGE,MYR,ITEST) +
* SLOPE(IP, MYR,IAGE+1) * (.125*(ODOM(IAGE+1)-ODOM(IAGE))) )
CC
GOTO 90
CC
CC..Biennial I/M Credits
CC
CC IMODE - 1 : Odd year
CC 2 : Even year
CC
60 IMODE - MOD(IAGE,2)
CC
CC..Biennial 1 - 3 - 5 - etc 1st Year Exception
CC. . same as no I/M first year
CC
IF(ITYP .EQ. 2 .AND. IAGE .EQ. 1)
* ANSWIM(IP,IAGE,MYR,ITEST,ITYP) =
* (.75*(.625*SLOPE(IP,MYR,IAGE)*ODOM(IAGE) + ZERO(IP, MYR, IAGE)))
* + .25*(PRED(IP,IBAG,IAGE,MYR,ITEST) + SLOPE(IP,MYR,IAGE+1)*
* (.125*(ODOM(IAGE+1)-OPOM(IAGE))) )
CC
CC..Biennial 2 - 4 - 6 - etc 1st Year Exception
CC. . same as no I/M first year
CC
-------
B-17
Appendix B : Tech 4.1 Model Source Code
IF(ITYP .EQ. 3 .AND. IAGE .EQ. 1)
* ANSWIM (IP,IAGE,MYR,ITEST,ITYP) -
* .75*(ZERO(IP,MIR,IAGE) + SLOPE(IP, MYR,IAGE)*.625*ODOM(IAGE)) +
* .25*(ZERO(IP,MYR,IAGE) + SLOPE(IP,MYR,IAGE+1)*
* (.125*(ODOM(IAGE+1)-ODOM(IAGE))+ODOM(IAGE)) )
CC
CC..Biennial 2 - 4 - 6 - etc 2nd Year Exception
CC
IF(ITYP .EQ. 3 .AND. IAGE .EQ. 2)
* ANSWIM(IP,IAGE,MYR,ITEST,ITYP) -
* .75*(ZERO(IP,MYR,IAGE) +
* SLOPE(IP,MYR,IAGE)*(ODOM(IAGE-l)) +
* SLOPE (IP, MYR, IAGE)* (.625* (ODOM(IAGE)-ODOM(IAGE-l) ) ) ) +
* .25*(PRED(IP,IBAG,IAGE,MYR,ITEST) +
* SLOPE (IP,MYR, IAGE+1)*. 125* (ODOM(IAGE+1) -ODOM(IAGE) ) )
CC
IFdAGE.EQ.l .OR. (ITYP.EQ.3 .AND. IAGE.EQ.2)) GOTO 90
CC
CC..The Principle Biennial Cases 1-3-5-etc and 2-4-6-etc
CC
CC..An Even Year for the 1-3-5 or An Odd Year for the 2-4-6
CC There is no I/M inspection that year
CC
IF((IMODE.EQ.l .AND. ITYP.EQ.3) .OR. (IMODE.EQ.O.AND.ITYP.EQ.2))
* ANSWIM(IP, IAGE,MYR, ITEST, ITYP) -
* .75*(PRED(IP,IBAG,IAGE-1,MYR,ITEST) +
* SLOPE(IP,MYR,IAGE)*(.625*(ODOM(IAGE)-ODOM(IAGE-1)))) +
* .25*(PRED(IP,IBAG,IAGE-1,MYR,ITEST) +
* SLOPE(IP,MYR,IAGE)*(ODOM(IAGE)-ODOM(IAGE-l)) +
* SLOPE(IP,MYR,IAGE+1)*.125*(ODOM(IAGE+1)-ODOM(IAGE)))
CC
CC..An Odd Year for the 1-3-5 or An Even Year for the 2-4-6
CC.. There is an I/M inspection that year
CC
IF((IMODE.EQ.O .AND. ITYP.EQ.3) .OR. (IMODE.EQ.LAND.ITYP.EQ.2))
* ANSWIM (IP, I AGE, MYR, ITEST, ITYP) -
* .75*(PRED(IP,IBAG,IAGE-2,MYR, ITEST) +
* SLOPE(IP,MYR,IAGE-1)*(ODOM(IAGE-1)-ODOM(IAGE-2)) +
* SLOPE(IP,MYR,IAGE)*(.625*(ODOM(IAGE)-ODOM(IAGE-l)))) +
* .25*(PRED(IP,IBAG,IAGE,MYR,ITEST) +
* SLOPE(IP,MYR, IAGE+1)*(.125*(ODOM(IAGE+1)-ODOM(IAGE))))
CC
CC..Combined 1-3-5 and 2-4-6 biennial cases
CC
90 CREDIT(IP,IAGE,MYR,ITEST,ITYP) -
* (ANSWNO (IP, IAGE,MYR) -ANSWIM (IP, IAGE,MYR, ITEST, ITYP) )
* / (ANSWNO(IP,IAGE,MYR))
CC
110 CONTINUE
CC
CC..Store resulting I/M credits
CC
CREDIT(IP,IAGE,MYR,ITEST,4) -
* (CREDIT(IP,IAGE,MYR, ITEST,2) + CREDIT(IP,IAGE,MYR,ITEST, 3))/2
CC
100 CONTINUE
CC
RETURN
END
-------
B-18
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE OUTPUT
CC
CC..Outputs results for emission factors, bag fractions and
CC..I/M credits.
CC
COMMON /DATO 5/ OXYE(2,3,4,3),OXYGON
COMMON /DAT11/ CREDIT(2, 25,12, 3, 4)
COMMON /DAT12/ ZML (3, 4, 12) , ZML1 (3, 4,12) , ZML2 (3, 4,12)
COMMON /DAT13/ BFZML1(3,4,12),BFDET1(3,4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
COMMON /DAT21/ BFDET2(3,4,12),BF2ML2(3,4,12)
CC
INTEGER IP,IBAG
CHARACTER*4 LABI(3) /' HC1,' CO',' NOX1/
CHARACTER*4 LAB2(5)/'FTP ',
* 'BAG1',
* 'BAG2',
* 'BAG3',
* 'BAGI'/
CC
NP = 3
Nl - 1
N3 - 3
CC
CC..Write out Emission Factors on Device #7
CC
WRITE(7,100) Nl
WRITE(7,600)
DO 20 IP=1,NP
WRITE(7, 500)
DO 20 MYR-1,12
NYR-1980+MYR
IBAG=1
T50 - ZML1(IP,IBAG,MYR) + 5.0*DET1(IP,IBAG,MYR)
T100 - T50 + 5.0*DET2(IP,IBAG,MYR)
WRITE(7,300) NYR,LAB2(IBAG),LABI(IP),
* ZML1(IP,IBAG,MYR),DET1(IP,IBAG,MYR), DET2(IP,IBAG,MYR),
* T50,T100,ZML(IP,IBAG,MYR),DET(IP,IBAG,MYR)
20 CONTINUE
CC
CC..Write out Annual I/M credits on Device #8
CC
WRITE(8,102) N3
DO 10 ITEST-1,3
DO 10 MYR-1,12
NYR-1980+MYR
DO 10 IP-1,2
WRITE(8,200) (CREDIT(IP,IAGE,MYR,ITEST,1),IAGE-1,24),
* NYR,LAB1(IP)
10 CONTINUE
CC
CC..Write out Biennial I/M Credits on Device #9
CC
WRITE(9,103) N3
DO 130 ITEST-1,3
DO 130 MYR-1,12
NYR-1980+MYR
DO 130 IP-1,2
WRITE(9,200) (CREDIT(IP,IAGE,MYR,ITEST,4),IAGE-1,24),
* NYR,LAB1(IP)
-------
B-19
Appendix B : Tech 4.1 Model Source Code
130 CONTINUE
CC
CC..Write out Bag Fractions on Device #10
CC
WRITE(10,101) Nl
DO 30 IP-1,3
WRITE(10, 500)
DO 30 MYR=1,12
NYR-1980+MYR
WRITE(10,400) NYR,LAB1(IP),
* (BFZML1 (IP, IBAG,MYR),BFDET1 (IP, IBAG,MYR) ,BFZML2 (IP, IBAG,MYR),
* BFDET2(IP,IBAG,MYR),IBAG-2,4),
* BFZML1 (IP, 1,MYR),BFDET1 (IP, 1,MYR), BFZML2 (IP, 1,MYR) ,
* BFDET2(IP,I,MYR)
30 CONTINUE
CC
100 FORMAT(II,/,' **',/,
*' ** MOBILE4.1 LDGV Emission Factors1,/,1 **')
101 FORMAT(II,/,/,
*' ** MOBILE4.1 LDGV Bag Fractions **',/,/,
*23X,'Bag l',20X,'Bag 2',25X,'Bag 3',25X,'FTP',/,
*9X, 4 (' '), /,
*9X,4(' ZML1 DET1 ZML2 DET2'),/,
*9X,4(' '))
102 FORMAT(II,/,' **',/,
*' ** MOBILE4.1 Annual I/M Credits',
*/,' **')
103 FORMAT(II,/,' **',/,
*' ** MOBILE4.1 Biennial I/M Credits',
*/,' **')
200 FORMAT(24F4.3,5X,I4,A4)
600 FORMAT (18X, ' ZML ', 3X, ' DET1 ', 3X, ' DET2 ' , 3X,
* ' Q 50k1,' @100k',3X,' ZML ',3X,' DET ')
300 FORMAT(
* IX,14,' : ',2A4,'-',
* F6.3,' + ',F6.3,3X,
* F6.3,3X,2F10.3,2(3X,F6.3))
400 FORMAT(1X,I4,A4,16F7.4)
500 FORMAT('-')
CC
RETURN
END
-------
B-20
Appendix B : Tech 4.1 Model Source Code
FUNCTION OXY(IP,ITECH,EMIT)
COMMON /DAT05/ OXYE(2,3,4,3),OXYGON
OXY-1- 0
IEG=0
IF(EMIT.LT.OXYE(1,1,ITECH,IP)) THEN
OXY-1.0-OXYCON*OXYE(2,1,ITECH,IP)/100.
GOTO 99
ENDIF
IF(EMIT.GE.OXYE(1,3,ITECH, IP)) THEN
OXY-1.0-OXYCON*OXYE(2,3,ITECH,IP)/100.
GOTO 99
ENDIF
IF(EMIT.LT.OXYE(1,2,ITECH,IP) .AND.
* EMIT.GE.OXYE(1,1,ITECH,IP)) IEG-2
IF(EMIT.LT.OXYE(1,3,ITECH,IP) .AND.
* EMIT.GE.OXYE(1,2,ITECH, IP)) IEG-3
Xl-OXYE(1,IEG-1,ITECH,IP)
X2-OXYE(1,IEG,ITECH,IP)
Yl-OXYE(2,IEG-1,ITECH,IP)
Y2-OXYE(2,IEG,ITECH, IP)
ODET-(Y2-Y1)/(X2-X1)
OZML-Y1-X1*ODET
OXY-1-(OZML+ODET*EMIT)*OXYCON/100.
99 RETURN
END
-------
B-21
Appendix B : Tech 4.1 Model Source Code
SUBROUTINE CONECT(BIN,OUT)
CC
CC DOTS(ITECH,A/BfIDOT,IP/ITEST)
CC
COMMON /DAT01/ MYR,ISTD,ITECH,IBAG,IP,IAGE,ICUT,ITST
COMMON /DAT25/ EXPO(2,4,3,2),DOTS(4,2,4,2,2)
CC
OUT-EIN
CC
IF(EIN.LE.DOTS(ITECH,1,1,IP,ITST)) THEN
SLP = DOTS(ITECH,2,1,IP,ITST)
* / DOTS(ITECH,1,1,IP,ITST)
OUT - SLP * BIN
ENDIF
CC
IF(BIN.GT.DOTS(ITECH,1,1,IP,ITST) .AND.
* EIN.LE.DOTS(ITECH,1,2,IP,ITST)) THEN
SLP = ( DOTS(ITECH,2,2,IP,ITST) -
* DOTS(ITECH,2,1,IP,ITST) )
* / ( DOTS(ITECH,1,2,IP,ITST) -
* DOTS(ITECH,1,1,IP,ITST) )
CPT = DOTS(ITECH,2,1,IP,ITST) -
* DOTS(ITECH,1,1,IP,ITST)*SLP
OUT - CPT + SLP*EIN
ENDIF
CC
IF(BIN.GT.DOTS(ITECH,1,2,IP,ITST) .AND.
* EIN.LE.DOTS(ITECH,1,3,IP,ITST)) THEN
SLP - ( DOTS (ITECH, 2, 3, IP, ITST) -
* DOTS(ITECH,2,2,IP,ITST) )
* / ( DOTS(ITECH,1,3,IP,ITST) -
* DOTS(ITECH,1,2,IP,ITST) )
CPT - DOTS(ITECH,2,2,IP,ITST) -
* DOTS(ITECH,1,2,IP,ITST)*SLP
OUT - CPT + SLP*EIN
ENDIF
CC
IF(BIN.GT.DOTS(ITECH,1,3,IP,ITST) .AND.
* EIN.LE.DOTS(ITECH,1,4,IP,ITST)) THEN
SLP = ( DOTS(ITECH,2,4,IP,ITST) -
* DOTS (ITECH, 2, 3, IP, ITST) )
* / ( DOTS(ITECH,1,4,IP,ITST) -
* DOTS(ITECH,1,3,IP,ITST) )
CPT - DOTS(ITECH,2,3,IP,ITST) -
* DOTS(ITECH,1,3,IP,ITST)*SLP
OUT = CPT + SLP*EIN
ENDIF
CC
IF(BIN.GT.DOTS(ITECH,1,4,IP,ITST)) THEN
OUT - DOTS(ITECH,2,4,IP,ITST)
ENDIF
CC
RETURN
END
-------
B-22
Appendix B : Tech 4.1 Model Source Code
CC
CC.
CC
CC
CC
CC
CC
CC
BLOCK DATA BD01
.This block data is used to initialize data arrays
COMMON /DAT10/ EWO<3,4,25,12),EIMW(2,4,25,12,3),EZM(2,4,12)
COMMON /DAT12/ ZML (3, 4,12) , ZML1 (3, 4,12), ZML2 (3, 4,12)
COMMON /DAT14/ DET(3,4,12),DET1(3,4,12),DET2(3,4,12)
COMMON /DAT15/ CWOA<2,25,4,2) ,EWOA(3, 25, 12) ,EZMA(2,12)
DATA EWO
DATA EIMW
DATA EZM
DATA ZML
DATA ZML1
DATA ZML2
DATA DET
DATA DET1
DATA DET2
DATA CWOA
DATA EWOA
DATA EZMA
END
3600*0.0
7200*0.0
96*0.0
144*0,
144*0.
144*0,
144*0.
144*0,
144*0.
400*0.0
900*0.0
24*0.0
-------
B-23
Appendix B : Tech 4.1 Model Source Code
BLOCK DATA BD02
CC
CC..Emission Level Data Block
CC
COMMON /DAT07/ ESO(2,4,4,2),EHO(2, 4, 4, 2)
COMMON /DAT19/ GV(4,2),GH(4,2),GS(4,2),BH<4,2),BV(4,2)
CC
CC
CC..Change in the rate of increase in the number of HIGH emitters
CC.. BH(ITECH,ISTD)
CC MPFI TBI Carb Oplp
DATA BH /13.8353,13.8353,0.9643,1.0000,
* 4.4709,15.2076,2.3200,1.0000 /
CC
CC..Change in the rate of increase in the number of VERY HIGH emitters
CC.. BV(ITECH,ISTD)
CC MPFI TBI Carb Oplp
DATA BV / 1.0000,1.0000,1-0000,1.0000,
* 1.0000,1.0000,1.0000,1.0000 /
CC
CC..Growth in the number of HIGHS per 10,000 miles
CC.. GH(ITECH,ISTD)
CC MPPI TBI Carb Oplp
DATA GH / .0065, .0065, .0270, .02704,
* .0090, .0040, .0140, .01298 /
CC
CC..Growth in the number of VERY HIGHS per 10,000 miles
CC.. GV(ITECH,ISTD)
CC MPFI TBI . Carb Oplp
DATA GV / .01933, .01198, .03762, .03530,
* .00840, .02012, .01487, .00433 /
CC
CC..Growth in the number of SUPERS per 10,000 miles
CC.. GS(ITECH,ISTD)
CC MPFI TBI Carb Oplp
DATA GS / .00257, .00257, .00257, .00000,
* .00194, .00194, .00194, .00000 /
CC
CC..Average emissions of SUPERS (from 37 EF & IM vehicles)
CC.. ESO(IP,IBAG,ITECH,ISTD)
CC
DATA ESO /
CC..1981,1982 model year vehicles
1 15.259,173.84, 18.376,154.87,16.344,193.62,10.813,150.52,
2 15.259,173.84, 18.376,154.87,16.344,193.62,10.813,150.52,
3 15.259,173.84, 18.376,154.87,16.344,193.62,10.813,150.52,
4 0.000, 0.00, 0.000, 0.00, 0.000, 0.00, 0.000, 0-00,
CC..1983 and newer model year vehicles
1 11.123,189.00, 7.6414,163.53,12.874,204.06,10.460,179.79,
2 18.083,183.59, 13.582,177.29,23.081,208.03,12.076,142.32,
3 12.870,246.85, 13.435,266.05,13.935,265.45,10.435,197.65,
4 0.000, 0.00, 0.000, 0.00, 0.000, 0.00, 0.000, 0.00 /
CC
CC..Emission Levels of VERY HIGH Emitters at zero miles
CC.. EHO(IP,IBAG,ITECH,ISTD)
CC
DATA EHO/
CC..1981,1982 model year vehicles
* 0.606, 18.049, 1.489, 25.589, 0.392, 15.029, 0.349, 18.063,
* 1.163, 22.059, 2.896, 64.029, 0.583, 7.181, 0.954, 18.611,
-------
B-24
Appendix B : Tech 4.1 Model Source Code
* 1.950, 33.396, 3.604, 47.486, 1.632, 32.110, 1.301, 25.174,
* 1.589, 27.650, 2.583, 45.178, 1,225, 21.845, 1.524, 25-403,
CC..1983 and newer vehicles - -._
* 2 019, 22.301, 5.041, 37.189, 1.742, 22.043, 1.369, 16.248,
* 2'.242 44 416 2.900 56.086, 2.267, 46.292, 1.705, 32.087,
* 2.002 36.130 4.320, 55.834, 1.464, 30.566, 1.272, 31.744
* 1.352, 34.021, 1-441, 40.120, 1.386, 34.668, 1.217, 28-081 /
CC
END
-------
B-25
Appendix B : Tech 4.1 Model Source Code
CC
CC.
CC
CC
CC
CC.
CC.
CC
CC.
CC.
CC
CC
CC.
CC.
CC
CC.
CC.
CC
CC.
CC.
CC
CC.
CC.
BLOCK DATA BD03
.Emission Level Data Block
COMMON /DAT03/ ENOX(4,2,2)
COMMON /DAT08/ DN(3,4,4,2)
COMMON /DAT20/ EMO(2,4,4,2),ENO(3,4,4,2)
.Emission level of HIGHs at zero mileage
EMO(IP,IBAG,ITECH,ISTD)
DATA EMO/
,1981,1982 model year
* 0.374, 3.879, 0
* 0.758, 6.843, 1
* 0.781, 8.496, 2
* 0-757, 8.142, 1
.1983 and newer model
* 0.997, 7.602, 1
* 0.762, 8.820, 1
* 0.703, 8.577, 1,
* 0.840, 8.386, 1
vehicles
954, 5.778,
540, 17.730,
027, 22.919,
494, 17.342,
year vehicles
933, 12.231,
288, 18.793,
740, 17.608,
461, 17.215,
0.272,
0.566,
0.404,
0.474,
0.712,
0.611,
0.376,
0.536,
4.591, 0.128, 0.992,
5.088, 0.530, 2.013,
4.073, 0.551, 5.963,
4.430, 0-735, 8.196,
6.344, 0.834, 6.538,
5.238, 0.651, 8.072,
4.252, 0.543, 9.925,
3.871, 0.951, 10.367 /
.Emission level of NORMAL emitting vehicles at zero mileage
END(IP,IBAG,ITECH,ISTD)
DATA SNO /
,1981,1982 model year vehicles
*.188, 1.548, .380,.722,5.515,.821,.
*.279, 3.422, .545,.789,9.081,.902,.
*.288, 3.067, .678,.752,9.350,1.203,
*.306, 3.368, .649,.632,8.501,.983,.
,1983 and newer model year vehicles
*.269, 2.598, .617,.831,7.073,.966,.
*.242, 2.953, .529,.619,7.074,.902,.
*.222, 2.209, .635,.586,7.006,1.021,
*.334, 4.093,.524,.594,10.555,.640,.
034,.466,.218,.
138,2.155,.451,
.127,.911,.491,
185,1.275,.512,
080,1.052,.451,
118,1.337,.395,
.108,.468,.446,
207,1.262,.478,
071,.578,.348,
.160,1.578,.460,
.241,2.394,.635,
.287,3.447,.655,
.202,2.163,.667,
.193,2.912,.504,
.164,1.890,.699,
.377,4.589,.5217
.Emission deterioration of NORMAL vehicles per 10,000 miles
DN(IP,IBAG,ITECH,ISTD)
DATA DN /
.1981,1982 model year
*.0450,.5807,.1280,
*.0556,.9654,.1345,
*.0161,.2520,.1414,
*.0156,.4582,.1497, .
*.0253,.3281,.0696,
*.0535,.8313,.0561,.
*.0230,.2877,.0440,
*.0654,1.0489,-0281,
,1983 and newer model
*.0115,.1554,.0163,
*.0137,.1382,.0248,.
*.0134,.1572,.0419,
*.0188,.1931,.0469,.
*.0199,.2282,.0593,
*.0429, .7472,.0555,.
vehicles
0395,.4866,.1112,.0487,.4784,.1557,
0119,.0323,.1264,.0245,.5045,.1626,
0181,.1893,.0657,.0183,.2174,.0876,
.0106,.0813,.0444,-0151,.1088,.0556,
year vehicles
0122,.1838,.0106,.0080,.1105,.0218,
0125,.1606,.0328,.0109,.1213,.0549,
0107, .0732,.0587,.0194,.1298, .0631,
-------
B-26
Appendix B : Tech 4.1 Model Source Code
*.0248,.2093,.0540,
*.0623,.4080,.0731,.0164,.1454,.0441, .0128, .1799, .0585/
CC
CC
CC..1983-85/1986+ split for NOx emissions
CC ENOX(ITECH,MYR SPLIT,ZML/DR)
CC
DATA ENOX/
CC..Zero Mile Levels
CC MPFI TBI Carb OPLP
* 0.689, 0.585, 0-692, 0.524,
* 0.539, 0.316, 0.478, 0.524,
CC..Deterioration Rates
CC MPFI TBI Carb OPLP
* 0.0185, 0.0470, 0.0558, 0.0540,
* 0.0185, 0.0470, 0.0558, 0-0540 /
CC
END
-------
B-27
Appendix B : Tech 4.1 Model Source Code
CC
CC.
CC
CC
CC
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
CC
CC
CC.
CC
BLOCK DATA EDO4
.I/M Identification Rate (IDR) effects block data
COMMON /DAT16/ XSIDR(2,4,2,3),XHIDR(2, 4,2, 3)
COMMON /DAT24/ XMIDR(2,4,2,3),XNIDR(2,4, 2, 3)
.The fraction of excess emissions identified by the short
.test for each emission level group.
.Modified Idle, 2spd, & Loaded (1.2/220) for total emissions
.rather than Excess emissions for Mobile 4.1, 7/19/91.
.Format: MPFI TBI GARB OpLoop
HC CO HC CO HC CO HC CO
. 81-81
. 83+
XSIDR(IP,ITECH,ISTD,ITEST)
DATA XSIDR/
.Idle test identification rate for SUPERS
1 .6048,.6968,.6048,.6968,.6048,.6968,.0, .0,
2 .8978,.9656,.8978,.9656,.8978,.9656,-0, .0,
.2500/Idle test identification rate for SUPERS
1 .6523,.8577,.6523,.8577,.6523,.8577,.0, .0,
2 .8978,.9656,.8978,.9656,.8978,.9656,.0, .0,
.Loaded/Idle test identification rate for SUPERS
1 .6523,.8577,.6523,.8577,.6523,.8577,.0, .0,
2 .8978,.9656,.8978,.9656,.8978,.9656, .0, .O/
XHIDR(IP,ITECH,ISTD,ITEST)
DATA XHIDR/
.Idle test identification rate for VERY HIGHs
1 .2736,.3231,.2736,.3231,.3858,.4108,.4568, .5194,
2 .5676,-6129,.2651,.2695,.3640,.3180, .3640, .3180,
.2500/Idle test identification rate for VERY HIGHs
1 .2736,.3231,.2736,.3231,.4789,.5331,-6197, .6162,
2 .6187,.7465,.3616,.4206,.5684,.6832,.5684,.6832,
.Loaded/Idle test identification rate for VERY HIGHs
1 ,2736,.3231,.2736,.3231,.5476,.6037,.6197, .6162,
2 .6187,.7465,.3904,.4337,.5684,.6832,.5684,-6832 /
XMIDR(IP,ITECH,ISTD,ITEST)
DATA XMIDR/
.Idle test identification rate for HIGHs
1 .0506,.1135,.0506,.1135,.0563,.0492,.2274,.1522,
2 .2507,.2208,.0336,.0613,.0694,.0415, .0694, .0415,
.2500/Idle test identification rate for HIGHs
1 .0506,.1135,.0506,.1135,.0898,.0834,.2274,.1522,
2 .3436,.3501,.1924,.1532,.0694,.0415, .0694, .0415,
.Loaded/Idle test identification rate for HIGHs
1 .0506,.1135,.0506,.1135,.0910,.0896,.2274,.1522,
2 .3866,.3937,.1924,.1532,.0694,.0415,.0694, .0415 /
XNIDR(IP,ITECH,ISTD,ITEST)
DATA XNIDR/
-------
B-28
Appendix B : Tech 4.1 Model Source Code
CC..Idle test identification rate for NORMALS
1 .0556, .0774, .0139,.0139,.0188,.0204,.0093,.0131,
2 .0360,.0414,.0425,.0436,.0023,.0078,.0023,.0078,
CC..2500/Idle test identification rate for NORMALS
1 .0556,.0774,.0139,.0139,.0371,.0427,.0201,.0317,
2 .0575,.0694,.0476,.0514,.0140,.0156,.0065,-0208,
CC..Loaded/Idle test identification rate for NORMALS
1 .0556,.0774, .0139, .0139,.0371,.0427,.0201,.0317,
2 .0907,.1023,.0712,.0739,.0140,.0156,.0231,.0403 /
CC
END
-------
B-29
Appendix B : Tech 4.1 Model Source Code
CC
CC.
CC
CC
CC
CC
CC
CC
CC
CC
CC
CC
CC
CC
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
CC
CC.
CC
BLOCK DATA BD05
. I/M Repair Effects block data
COMMON /DAT17/ RSUP(2,4,2,3),RHIG(2, 4, 2, 3)
COMMON /DAT22/ RMAR(2,4,2,3),RNOR(2, 4,2, 3)
COMMON /DAT25/ EXPO<2,4,3,2),DOTS(4,2,4, 2, 2)
DOTS(ITECH,A/B,IDOT,IP,ITST)
DATA DOTS /
*0.7400,0.7400,.96765,.96765, .41083,.41083,.62235,.62235,
*1.9223,1.9223,2-0226,2.0226, .60615,.60615,1.1894,1.1894,
*3.9023,3.9023,3.1063,3.1063, 1.0769,1.0769,1.3254,1.3254,
*14.282,14.282,8.5543,8.5543, 1.3808,1.3808,1.5286,1.5286,
*9.2708,9.2708,10-487,10. 487, 4.9900,4.9900,9.8624,9.8624,
*28.031,28.031,29.550,29.550, 9.4669,9.4669,12.969,12.969,
*90.038,90.038,53.520,53.520, 12.148,12.148,17.434,17.434,
*190-66,190.66,134.75,134.75, 20.620,20.620,18.281,18.281,
*.82667, .82667,.93026,.93026, .40750,.40750,.57641,.57641,
*1.9846,1.9846,1.9431,1.9431, .59231,.59231,1.0349,1.0349,
*3.9314,3.9314,2.9862,2.9862, 1.0271,1.0271,1.1413,1.1413,
*14.282,14.282,8.2523,8.2523, 1.3808,1.3808,1.4141,1.4141,
*10.334,10.334,10.622,10.622, 4.8950,4.8950,9.2808,9.2808,
*35.518,35.518,29.053,29.053, 9.8631,9.8631,12.489,12.489,
*104.50,104.50,54.282,54.282, 11.925,11.925,13.190,13.190,
*190.66,190.66,136.97,136.97, 20.620,20.620,13.596,13.596 /
EXPO(IP,ITECH,ITEST,A/B) Aexp(-B/x)-y
DATA EXPO /
* 0.6682, 6.1319, 0-7476, 8.4157, 1.3629, 13.4934,1.1487, 9.9503,
* 0.6736, 6.8360, 0.6839, 7.9328, 1.1585, 10.7263,1.2079, 9.1495,
0.6736, 6.8360,
0.2676, 3.4348,
0.6839, 7.9328, 1.1585, 10.7263,1.2079, 9.1495,
0.3419, 3.0945, 0.7034, 5.4373, 0.5682, 2,1526,
* 0.2911, 6.3200, 0.3262, 2.8702, 0.6309, 4.1224, 0.4449, 1.7352,
* 0.2911, 6.3200, 0.3262, 2.8702, 0.6309, 4.1224, 0.4449, 1.7352 /
.Emission level after repairs expressed as a fraction of
.the emission level before repairs.
RSUP(IP,ITECH,ISTD,ITEST) 4/11/91
DATA RSUP/
.2500/Idle Test
1 .958,
2 .958,
.2500/Idle Test
1 .958,
2 .958,
.2500/Idle Test
1 .958,
2 .958,
emission effect
.975, .968, .952
.975, .968, .952
emission effect
.975, .968, .952
.975, .968, .952
emission effect
.975, .968, .952
.975, .968, .952
from repairs for SUPERS
, .864, .930, .000, .000,
, .864, .930, .000, .000,
from repairs for SUPERS
, .864, .930, .000, .000,
, .864, .930, .000, .000,
from repairs for SUPERS
, .864, .930, .000, .000,
864, .930, -000, .000 /
RHIG(IP,ITECH,ISTD,ITEST)
-------
B-30
Appendix B : Tech 4.1 Model Source Code
CC.
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
CC
CC.
CC
CC.
CC.
CC.
DATA RHIG/
.2500/Idle Test
1 .609,
2 .609,
.2500/Idle Test
1 .609,
2 .609,
.2500/Idle Test
1 .609,
2 .609,
DATA RMAR/
.2500/Idle Test
1 .609,
2 .609,
.2500/Idle Test
1 .609,
2 .609,
.2500/Idle Test
1 .609,
2 .609,
DATA RNOR/
.2500/Idle Test
1 .375,
2 .375,
.2500/Idle Test
1 .375,
2 .375,
.2500/Idle Test
1 .375,
2 .375,
emission effect of
.772, .879, .909, ,
.772, .879, .909, ,
emission effect of
.772, .879, .909, ,
.772, .879, .909, ,
emission effect of
.772, .879, .909, .
.772, .879, .909, ,
repairs for
668, .760,
668, .760,
repairs for
668, .760,
668, .760,
repairs for
668, .760,
668, .760,
VERY
.627,
.627,
VERY
.627,
.627,
VERY
.627,
.627,
HIGHS
.741,
.741,
HIGHS
.741,
.741,
HIGHS
.741,
.741 /
RMAR(IP,ITECH,ISTD,ITEST)
(Same as for Very Highs)
emission effect of
.772, .879, .909,
.772, .879, .909,
emission effect of
.772, .879, .909,
.772, .879, .909,
emission effect of
.772, .879, .909,
.772, .879, .909,
repairs for
.668, .760,
,668, .760, .
repairs for
,668, .760, .
,668, .760, ,
repairs for
,668, .760, .
,668, .760, .
RNOR(IP,ITECH,ISTD,ITEST)
emission effect of
.508, .352, .390, ,
.508, .352, .390, ,
emission effect of
.508, .352, .390, ,
.508, .352, .390, ,
emission effect of
.508, .352, .390, ,
.508, .352, .390, ,
repairs for
263, .280, ,
263, .280, ,
repairs for
263, .280, .
263, .280, .
repairs for
263, .280, ,
263, .280, ,
HIGHS
627, .741,
627, .741,
HIGHS
627, .741,
627, .741,
HIGHS
627, .741,
627, .741 /
NORMALS
420, .603,
420, .603,
NORMALS
420, .603,
420, .603,
NORMALS
420, .603,
420, -603 /
END
-------
B-31
Appendix B : Tech 4.1 Model Source Code
CC
CC.
CC
CC
CC.
CC.
CC
CC
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC.
CC
CC.
CC
CC
CC.
CC
CC
c
c
c
BLOCK DATA BD06
.Fleet Description Block
COMMON /DAT02/ AMIL(25),ODOM(25),TMILE(25),WGT(25)
COMMON /DAT06/ FRAG(4,12)
.Technology Sales Fractions Projections
FRAC(ITECH,MYR)
DATA FRAC /
MPFI TBI Carb Oplp
.1981 Model Year
* .061, .029, .629, .281,
.1982 Model Year
* .062, .106, .506, .325,
.1983 Model Year
* .088, .183, .486, .244,
.1984 Model Year
* .110, .282, .551, .058,
.1985 Model Year
* .307, .208, .408, .077,
.1986 Model Year
* .400, .270, .325, .005,
.1987 Model Year
* .374, .365, .250, .010,
.1988 Model Year
* .492 ,-407, .101, .000,
.1989 Model Year
* .597, .275, .125, .003,
.1990 Model Year
* .761, .219, .018, .001,
.1991 Model Year
* .795, .202, .003, .000,
.1992 and Newer Model Years
* .815, .185, .000, .000 /
.Fleet January 1st VMT weighting factors
DATA WGT / 0-030, 0.120, 0.111, 0-099, 0-088,
* 0.078, 0-068, 0.060, 0.054, 0.048,
* 0-043, 0.038, 0.033, 0.028, 0.024,
* 0.020, 0.017, 0.013, 0.010, 0.007,
* 0.004, 0.002, 0.002, 0.001, 0.003 /
.Cumulative January 1st mileage accumulation by age
DATA ODOM/ 1.3118, 2.6058, 3.8298, 4.9876, 6.0829,
* 7.1190, 8.0991, 9.0262, 9.9031,10.7326,
* 11.5172,12.2594,12.9615,13.6257,14.2540,
* 14.8483,15.4104,15.9421,16.4451,16.9209,
* 17.3712,17.7969,18.1997,18.5806,18.9410 /
END
BLOCK DATA BD07
Oxygenated Fuels Effects Block
COMMON /DATO 5/ OXYE(2,3,4,3),OXYGON
-------
B-32
Appendix B : Tech 4.1 Model Source Code
DATA OXYGON / 1.0 /
C
C OXYE(XY,IEMIT,ITECH,IP)
C
DATA OXYE /
C HC
* 24*0.00,
C CO
* 3.50,3.46, 15.26,9.87, 95-79,11.44,
* 3.72,4.93, 15.26,9.87, 95.79,11.44,
* 4.18,6.77, 15.26,9.87, 95.79,11.44,
* 3.50,9.97, 15.26,9.97, 95.79, 9.97,
C NOx
* 24*0.00 /
C
END
-------
APPENDIX C
EVAPORATIVE EMISSIONS AND RUNNING LOSS
EMISSION FACTOR DERIVATION
-------
EVAPORATIVE HC EMISSIONS
1.0 Introduction
There have been more vehicles tested by EPA's Emission Factors
Program (EFP) since the issuance of MOBILE4. From the EFP test
facility located at Motor Vehicle Emissions Laboratory (MVEL) in
Ann Arbor, Michigan (EFP-MVEL), EPA continues recruiting in-use
vehicles and testing them with the certification test procedure —
a diurnal test, followed by exhaust emissions test on a Federal
Testing Procedure (FTP) cycle, and then a hot soak test. A
certification test fuel with fuel volatility level of 9.0 psi Reid
Vapor Pressure (RVP) is used. Two additional fuels with 10.4 and
11.7 psi RVPs were also used during FY84 through FY89 to
characterize the evaporative emissions effect due to different fuel
volatility levels that were commercially available for in-use
vehicles. Since FY90, the 9.0 psi RVP fuel has been the only
gasoline fuel used in the EFP-MVEL testing.
EPA added "Hammond Program" (EFP-Hammond) to the so-called
MVEL "traditional" EFP during FY90. A brief description of this
new program is given in section 1.1. Due to the addition of this
new test program, a different modeling approach is used for
MOBILE4.1's evaporative diurnal and hot soak emissions.
For example, in the MOBILE4 evaporative diurnal and hot soak
emissions model, there were seven elements of evaporative
emissions: standard level, emissions due to insufficient canister
purge, emissions due to excess fuel volatility, problem-free
emission levels, emissions due to malmaintenance and/or defect,
non-tampered emission levels, and emissions from tampering. The
problem-free emission rates were the sum of the first three
elements. The non-tampered emissions were the emissions from
problem-free vehicles plus the emissions from malmaintenance and/or
defect. Finally, the in-use evaporative diurnal and hot soak
emissions were estimated by accounting for emissions from both the
non-tampered and tampered fractions of the vehicle fleet.
However, in the MOBILE4.1 evaporative emissions model, there
are three different types of evaporative HC emissions: emissions
from pass vehicles, emissions from vehicles that failed pressure
test, and emissions from vehicles that failed purge test. These
three types of emissions are defined for both diurnal and hot soak
emissions in MOBILE4.1.
-1-
-------
1.1 Hammond Program
EPA's Hammond program was initiated in FY90. This program
utilizes the testing facilities for the State of Indiana's
Inspection and Maintenance (I/M) programs located at Hammond,
Indiana. While waiting for the required I/M tests, in-use vehicle
owners were asked to participate additional tests defined by EPA's
new EFP. One of the tests those I/M vehicles were asked to
participate was a transient cycle called "IM240," which is a
shortened version of the certification FTP cycle. All
participating vehicles were also tested for their evaporative
emissions control system to see if there exists either pressure
and/or purge failure.
The pressure test is performed by using a gauge to measure the
pressure between a vehicle's fuel tank rollover valve and
evaporative canister, by introducing nitrogen at 14" of water
pressure. Two minutes after the initiation of this test, if the
evaporative system holds less than 8" pressure, the vehicle is
considered to have pressure problem in its evaporative emissions
control system, thus failing the pressure test.
The purge test is performed by using a flow meter measuring
the purge air between a vehicle's canister and engine. The
cumulative purge air over the entire IM240 cycle is measured. If
the flow rates are less than 1.0 cubic liter, the vehicle's
evaporative emissions control system is considered to have
insufficient purge, thus failing the purge test.
As of January 30, 1991, 2,497 I/M vehicles were tested for
potential pressure and purge failures under the Hammond program.
The model years of these in-use vehicles ranged from 1976 to 1991.
Overall failure rates were: 12.7 percent for the pressure test,
11.5 percent for the purge test, and 21.1 percent for either.
These failure rates were analyzed and then characterized in terms
of vehicle's age, and are summarized in Table 1.
Some of the failed vehicles were also tested for their diurnal
and hot soak emission levels. The results seem to suggest that
emissions from vehicles that failed pressure test are similar to
the emissions from vehicles with no evaporative emissions control
system, and are somewhat higher than the emissions from vehicles
that failed purge test.
1.2 Modeling Approach
As mentioned in section 1.0, the MOBILE4.1 evaporative
emissions model include: emissions from pass vehicles, emissions
from vehicles that failed pressure test, and emissions from
vehicles that failed purge test.
-2-
-------
Test results obtained from EFP-MVEL were used to:
a) characterize the fuel volatility effect on emissions,
b) derive emission levels for pass vehicles, and,
c) describe the emissions effects due to other parameters,
such as fuel delivery system, vehicle type, etc.
Results obtained from EFP-Hammond program were used to:
a) define pass versus pressure and/or purge failure vehicles,
b) estimate the emission levels due to pressure and/or purge
failures, and,
c) see if the pressure/purge test results are applicable to
and/or are consistent with mechanics' visual diagnosis
check in EPA's EFP at MVEL.
In addition to test data, a theoretical vapor generating model
(based on Wade equation) was also used for the diurnal emissions
model to:
a) determine the emissions effect from a combination of fuel
volatility, temperature rise, fuel tank capacity, and
tank fill level,
b) describe the emissions from vehicles that failed either
pressure or purge test, and,
c) define the emission levels from vehicles that have two or
more consecutive no-driving days.
The "pass" vehicles are defined as vehicles that are not
tampered, nor failing pressure/purge tests. Therefore, in terms of
the MOBILE4 evaporative emissions categories, the pass vehicles
include vehicles that have no visible nor measurable problems in
their evaporative emissions control system components, or problems
that do not appear to directly relate to pressure leaks or purge
deficiency.
Table 2 provides a list of the criteria used in defining
pressure/purge failures as well as tampered vehicles from the
EFP-MVEL data base. As can be seen, "tampered" in the MOBILE4.1
model is now a subsample of the failure vehicles by definition. In
general, pressure failures indicate the potential existence of a
leakage on vehicle's evaporative emissions control system
components, which include: gas cap, filler neck, sending unit,
rollover valve, and vent hoses. Purge failures are usually caused
by an in-operative canister purge solenoid or valve, or
disconnected, missing, or damaged purge hoses.
-------
2.0 Diurnal Emissions
As in MOBILE4, the diurnal emissions (in unit of grams per one
hour test) are described as a function of Uncontrolled Diurnal
Index (UDI). This UDI is the uncontrolled diurnal emissions (in
grams), calculated from a combination of weathered fuel volatility
level (in psi RVP), fuel tank temperature rise (in °F), fuel tank
capacity (assumed a sales-weighted 16.2 gallon tank for LDGVs), and
percent tank fill (set as 54.57%), normalized at FTP conditions
(which include using 9.0 psi RVP fuel, 60 to 84°F temperature rise,
and 40% tank fill). The calculations of uncontrolled diurnal
emissions are based upon a theoretical vapor generating model (Wade
equation). For example, with 9.0 psi RVP fuel, 72 to 96°F
temperature rise and 40% tank fill, the UDI is calculated to be
1.7448. The followings are some examples of the calculated UDI
values, all at 60 to 84°F temperature rise and 40% tank fill, with
different fuel volatility levels:
Fuel Volatilities in psi RVP
9.0 10.5 11.5 11.7
UDI 1.0 1.4567 1.9581 2.0677
The UDIs are used to describe diurnal emissions for pass vehicles
as well as for vehicles that failed pressure/purge tests.
2.1 1981+ Light-Duty Gasoline-Powered Vehicles and Trucks
The MOBILE4.1 diurnal emissions for model years 1981 and later
light-duty gasoline-powered vehicles and trucks (LDGVs and LDGTs)
are based upon EPA's EFP test results. Table 3 summarizes the
emissions data obtained from EFP-MVEL in terms of sample sizes and
average emissions for various categories (such as pass/fail
vehicles, different fuel delivery systems and vehicle types).
Note that due to a lack of class 2 truck data in the EFP-MVEL
sample, the two classes of light-duty trucks (classes 1 and 2) are
combined as one LDGT group. Also, because of insufficient data,
two types of fuel injection fuel delivery systems (ported
fuel-injection [PFI] and throttle body fuel-injection [TBI]) are
also being combined as one fuel-injected (FINJ) group for the LDGTs.
2.1.1 Pass Vehicles
Diurnal emissions for pass vehicles were derived from EPA's
EFP-MVEL data. The derived equational form for model years 1981+
pass light-duty vehicles and trucks is:
-4-
-------
3 + b * UDI + C * UDI1
(2.1)
where:
Vehicle Type
/Fuel System
LDGV/CARB
LDGV/PFI
LDGV/TBI
LDGT/CARB
LDGT/FINJ
Regression Coefficient
-4.3304
•0.0879
-0.8233
•7.6803
•3.8537
5.9904
0.0
0.0
9.7703
4.9437
0.0
1.4179
2.0433
0.0
0.0
The coefficients from equation (2.1) were derived from EFP-MVEL
data, where the three UDI values corresponding to 9.0, 10.4, and
11.7 psi RVP fuel volatility levels were 1.0, 1.4567, and 2.0677,
respectively. The constant terms of equation (2.1) (denoted as "a"
above) were adjusted so that when UDI is 1.0 the predicted diurnal
emissions are equal to the average diurnal emissions at 9.0 psi RVP
from the EFP-MVEL sample.
For carbureted vehicles, the equation (2.1) is to be
extrapolated when the UDI values are greater than 2.0677. However,
when the UDI values are less than 1.0, new sets of coefficients
listed below are used. These coefficients were derived from a
two-step analysis: first, all carbureted vehicle data were used to
fit an exponential function, then, the predicted emissions for UDIs
at 1.0 and lower were used to fit a linear regression line. Again,
the constant terms were adjusted so that when UDI is 1.0 the
predicted diurnal emissions are equal to the average diurnal
emissions at 9.0 psi RVP from the EFP-MVEL sample. In addition, a
lower bound of the diurnal emissions is set to equal to the resting
loss emissions.
Vehicle Type
/Fuel System
LDGV/CARB
LDGT/CARB
Regression Coefficient
•0.2888
0.1412
1.9488
1.9488
0.0
0.0
For fuel-injected vehicles, equation (2.1) is to be
extrapolated when the UDI values are beyond the existing data
(i.e., UDI values are either less than 1.0 or greater than
2.0677). At low UDI values the calculated diurnal emissions may
become negative. A lower bound of the diurnal emissions is set to
equal to the resting loss emissions.
Graphic presentations of 1981+ LDGV/LDGT diurnal emissions as
a function of UDI for pass vehicles are shown in Figures 1 and 2.
-5-
-------
2.1.2 Failed Vehicles
To estimate the diurnal emission levels for vehicles that
failed either pressure or purge test, two steps of analyses were
used: 1) the theoretical vapor generating model, and, 2) data from
EFP-Hammond.
The theoretical vapor generating model was used to describe
the relationships between UDI and the emission levels from failed
vehicles (see Attachment). It is assumed that fuel delivery system
is not a significant parameter on failed diurnal emission levels,
although the emissions simulation was based on a late model-year
fuel-injection system. Uncontrolled diurnal emissions (in grams)
were calculated at various fuel volatility levels (from 7.5 to
11.0 psi RVP) for the temperature rise of 72 to 96°F. These
calculated uncontrolled diurnal emissions were used for the
pressure failure levels. The purge failure levels were the
calculated uncontrolled emissions subtracted by a backpurge
effect. The g/test diurnal emissions for failed LDGVs are
expressed as linear functions of UDI:
Dlraiied Pressure - 2.2002 + 7.5626 * UDI
Dlpalied Purge - -4.2897 -r 7.4563 * UDI
At UDI - 1.7448 (72 to 96°F temperature rise, 9.0 psi RVP fuel, and
40% tank fill), the diurnal emissions estimated from the above two
equations are 15.40 and 8.72 grams for vehicles that failed
pressure and purge test, respectively.
From EFP-Hammond data base, 30 LDGVs (as listed in Table 4)
were also tested for diurnal emissions under the conditions of 72
to 96°F temperature rise, 9 psi RVP fuel, and 40% tank fill (i.e.,
also with UDI « 1.7448). These vehicles were found to have failed
either pressure or purge test at the Hammond I/M Lane. Their
average diurnal emissions were:
G/Test
Category N Diurnal
Failed Pressure 12 27.73
Failed Purge 18 15.58
The average diurnal emission levels from EFP-Hammond were
compared with the estimated emissions from the theoretical vapor
generating model. The data versus model ratios of 1.801185
( = 27.73/15.40 for the pressure failure) and 1.786687
( = 15.58/8.72 for the purge failure) were the multiplicative
factors used to derive new diurnal emission equations for LDGVs
that failed either pressure or purge test:
-6-
-------
Dlpailed Pressure - 1.801185 * (2.2002 + 7.5626 * UDI)
= 3.9630 + 13.6216 * UDI (2.2)
Dlraiied Purge = 1.786687 * (-4.2897 + 7.4563 * UDI)
- -7.6643 + 13.3221 * UDI (2.3)
Note that equations (2.2) and (2.3) were derived based on a
sales-weighted LDGV fuel tank capacity of 16.2 gallons. The fuel
tank capacities of LDGTs are, in general, larger than the LDGVs.
Many of the LDGTs are also equipped with an auxiliary fuel tank of
a similiar capacity as the primary tank. Based on EPA's
Certification data base, the sales-weighted primary fuel tank
capacity for model year 1988 LDGTls was 19.0 gallons. When
including the auxiliary fuel tank, the sales-weighted capacity for
LDGTls became 20.8 gallons. Therefore, an average fuel tank
capacity ratio of 1.17 (19.0 gallon/16.2 gallon) is used to
estimate the diurnal emissions of failed LDGTs:
LDOTs
= 1.17 * DI F a i I e 4 LDGVs (2.4)
This approach implies that the diurnal emissions from failed
LDGTs are 117 percent higher than those from failed LDGVs. This
assumption is well-supported from EFP-MVEL test data. As can be
seen from Table 3, the average diurnal emissions from failed LDGTs
are 125 percent higher at 11.7 psi RVP fuel, and 138 percent higher
at 9.0 psi RVP fuel, than the average emissions from failed LDGVs.
Under FTP conditions, the diurnal emissions for vehicles that
failed either pressure or purge test are calculated to be:
Diurnal (G/Test)
Vehicle Failed Failed
Type Pressure Purge
LDGVs 17.58 5.66
LDGTs 20.57 6.62
In general, diurnal emission levels for vehicles that failed
either pressure or purge test at any other UDI values for 1981+
LDGVs and LDGTs can be calculated from equations (2.2) through
(2.4). Note that the pressure failure equation can be extrapolated
down to UDI « 0.0 without getting negative emission values. When
UDI is at 0.58 or lower, however, the purge failure equation is
predicting zero or negative diurnal emissions. A lower bound of
the failed purge diurnal emissions is also set to equal to the
resting loss emissions.
Graphic presentations of 1981+ LDGV/LDGT diurnal emissions as
a function of UDI for vehicles failed either pressure or purge test
are shown in Figures 3 and 4.
-7-
-------
2.1.3 Pre-1981 LDGV Diurnal Emission Rates
Pre-1981 model year LDGVs are predominately carbureted
vehicles. In this section, the diurnal emissions from pass
vehicles and from vehicles that failed either pressure or purge
test for pre-1981 LDGVs are determined. The following is a list of
methodologies used for deriving these emissions:
1) For pre-1971 no evaporative emission standard vehicles,
all three emissions are set to be the same as the MOBILE4
uncontrolled emission rates. However, if the MOBILE4 uncontrolled
emission rates are lower than the. estimated emission levels from
vehicles that failed pressure test, the new failed pressure
emission levels are used.
2) Emissions are calculated from equations derived for 1981+
LDGV group (e.g., for diurnal emissions, use equations (2.1)
through (2.4) defined in sections 2.1.1 and 2.1.2).
3) As the MOBILE4 emission rates came from the average values
of test data, it is assumed that data from various model years were
results from vehicles tested at about the same age by EPA's EFP.
The average mileages of the current 1981+ sample were:
VTYP CARS PFI TBI FIKJ
LDGV 55,057 42,115 48,890
LDGT 52,794 44,759
These average mileages fell into 4-5 year old category. The
fractions of the 4 year old LDGV fleet that failed either pressure
or purge test are (from Table 1) 0.053 and 0.060, respectively.
With the emission levels from vehicles that failed either pressure
or purge test being already defined, the following relationships
are established:
E » 0.887 * Epass + [0.053 * EFaile«l Pressure]
+ [0.06 * Efailed Purge]
Therefore, the emission levels for pass vehicles can be back
calculated from the above equation.
4) Assume the same MOBILE4 uncontrolled emission rates.
5) Use the emission ratio from:
(MOBILE4MYRi / MOBILE4HYR2)
= (MOBILE4.1MYRi / MOBILE4.1MYR2)
6) Use the emission ratio from:
(Failed PressureMYRi / Failed Pressure,,YRZ)
» (Failed PurgenYRi / Failed
-8-
-------
The MOBILE4 FTP diurnal emissions in g/test for pre-1981 LDGVs
at two different RVP fuels were:
MYR Group
Pre-1971
1971
1972-77
1978-80
M4 Diurnal (G/Testl
9.0 RVP 11.5 RVP
26.08
16.28
8.98
5.16
47.99
38.58
23.53
14.47
The derived MOBILE4.1 LDGV FTP diurnal emissions for the two RVP
fuels are:
FTP Diurnal Emissions (G/Test)
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Garb
Note that the 1981+ carbureted diurnal emissions were calculated
from equations (2.1) through (2.3). They were listed above for
comparison purposes only, since all pre-1981 LDGVs are of
carbureted fuel delivery system. As in MOBILE4, the pre-1981 FTP
LDGV diurnal emissions in g/test for any other RVP fuels are
calculated by interpolating between these two given emission
levels.
Using the assumption that in-use vehicles were tested at the
age of four, the MOBILE4.1 FTP diurnal emissions in g/test for
LDGVs at two different RVP fuels are:
at 9
.0 RVP
Failed
26
15
8
4
1
Pass
.08l
.033
.27s
.393
.662
Purge
26
26
9
5
5
.081
.081
.33a
.66*
.66*
Failed
Pressure
26
26
20
17
17
.08l
.08'
.39s
.582
.582
47
37
22
13
7
Pass
.991
.383
.80s
.243
.402
at
11.5
Failed
Purge
47
47
23
18
18
.991
.99l
.S33
.422
.422
RVP
Fai
led
Pressure
47.
47.
35.
30.
30.
99'
99l
454
642
642
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Garb
M4.1 Diurnal (G/Test)
9.0 RVP 11.5 RVP
26.08
16.28
8.98
5.16
2.74
47.99
38.58
23.53
14.47
9.29
-9-
-------
2.1.4 Pre-1981 LDGT1 Diurnal Emission Rates
In MOBILE4, as well as its earlier versions, pre-1981 LDGV
emission rates were used directly for pre-1981 LDGTls, since there
were insufficient test data from LDGTls to derive their own
emission rates and LDGTls have the same evaporative emission
standards as the LDGVs. The MOBILE4 FTP diurnal emissions in
g/test for pre-1981 LDGTls at two different RVP fuels were:
M4 Diurnal (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1971
1971
1972-77
1978-80
26.08
16.28
8.98
5.16
47.99
38.58
23.53
14.47
However, as discussed in section 2.1.2, LDGTls in general are
equipped with larger capacity of fuel tank than the LDGVs. From
model years 1981+ EF data, where the majority of truck data were
available, the average LDGT1 diurnal emissions are found to be
significantly higher than their LDGV counterpart (see Table 3).
Thus, it is reasonable for LDGTl's to have higher diurnal emissions
than the LDGVs, whether for pass vehicles, or for vehicles that
failed either pressure or purge test.
The MOBILE4.1 LDGT1 diurnal emission rates were estimated by
multiplying a factor of 1.17 (the ratio of the average LDGT1 and
LDGV fuel tank capacities, as discussed in section 2.1.2).
FTP Diurnal Emissions (G/Test)
Pass
at 9.0 RVP
Failed Failed
Purge Pressure
Pass
at 11.5 RVP
Failed Failed
Purge Pressure
30.51
17.59
9.68
5.14
2.092
30.51
30.51
10.92
6.62
6.62
30.51
30.51
23.86
20.57
20.57
56.15
43.73
26.68
15.49
11. 452
56.15
56.15
27.88
21.55
21.55
56.15
56.15
41.48
35.85
35.85
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Carb
Note that the 1981+ carbureted diurnal emissions were calculated
from equations (2.1) through (2.4). They were listed above for
comparison purpose only, since all pre-1981 LDGTls are of
carbureted fuel delivery system. As in MOBILE4, the pre-1981 FTP
LDGT1 diurnal emissions in g/test for any other RVP fuels are
calculated by interpolating between these two given emission levels.
-10-
-------
Using the assumption that in-use vehicles were tested at the
age of four, the MOBILE4.1 FTP diurnal emissions in a/test for
LDGTls at two different RVP fuels are:
M4.1 Diurnal (G/Testl
MYR Group 9.0 RVP 11.5 RVP
Pre-1971 30.51 56.15
1971 19.05 45.13
1972-77 10.51 27.54
1978-80 6.05 16.93
1981+ Garb 3.34 13.35
As can be seen, the new LDGT1 diurnal emissions for MOBILE4.1 are
slightly higher than the diurnal rates assumed for MOBILE4.
2.1.5 Pre-1981 LDGT2 Diurnal Emission Rates
In MOBILE4, the same post-1979 LDGT1 diurnal emissions were
assumed for LDGT2s, based on their similar evaporative emissions
standards and their similar fuel tank capacities. The diurnal
emissions for pre-1979 LDGT2s were assumed to be the same as the
pre-controlled HDGV diurnal emission levels, as there were no
separate vehicle category of LDGT2 before the model year of 1979.
The MOBILE4 FTP diurnal emissions in g/test for LDGT2s at two
different RVP fuels were:
M4 Diurnal (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1979 42.33 77.89
1979-80 5.16 14.47
Using the same methodology as in LDGTls (section 2.1.4), the
MOBILE4.1 FTP g/test diurnal emissions for LDGT2s at two different
RVP fuels are:
FTP Diurnal Emissions (g/test)
at 9.0 RVP at 11.5 RVP
Failed Failed Failed Failed
MYR Group Pass Purge Pressure Pass Purge Pressure
Pre-1979 42.33 42.33 42.33 77.89 77.89 77.89
1979-80 5.14 6.62 20.57 15.49 21.55 35.85
1981+ Garb 2.09* 6.62 20.57 11.452 21.55 35.85
-11-
-------
Note that the 1981+ carbureted diurnal emissions were calculated
from equations (2.1) through (2.4). They were listed above for
comparison purpose only, since all 1979-80 LDGT2s are of carbureted
fuel delivery system. As in MOBILE4, the pre-1981 FTP LDGT2
diurnal emissions in g/test for any other RVP fuels are calculated
by interpolating between these two given emission levels.
Using the assumption that in-use vehicles were tested at the
age of four, the MOBILE4.1 FTP diurnal emissions in g/test for
LDGT2s at two different RVP fuels are:
M4.1 Diurnal (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1979 42.33 77.89
1979-80 6.05 16.93
1981+ Garb 3.34 13.35
2.1.6 HDGV Diurnal Emission Rates
The MOBILE4 FTP diurnal emissions in g/test for HDGVs at two
different RVP fuels were:
M4 Diurnal (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1985 42.33 77.89
1985+ 4.68 17.94
The same algorithm as MOBILE4 HOGV evaporative emissions were
used to estimate model year 1985+ HDGV diurnal emission levels for
pass vehicles, as well as HDGVs that failed either pressure or
purge test. The bases used in the calculations were the 1981+
LDGT1 diurnal emission rates (i.e., 2.09/11.45 g/test for pass
vehicles, 20.57/35.85 g/test for vehicles that failed pressure
test, and 6.62/21.55 g/test for vehicles that failed purge test, at
9.0 and 11.5 psi RVP fuels, respectively). In general, the
following equation is used to estimate HDGV diurnal emission rates:
* 1.5 * 0.875]
i * 2.0 * 0.125]
where, the constant values of 1.5 and 2.0 were the ratios of the
evaporative emission standards between HDGVs and LDGTls. The
constant values of 0.875 and 0.125 were the estimated market shares
of HDGVs under each evaporative emissions standards.
-12-
-------
The MOBILE4.1 FTP g/test diurnal emissions for HDGVs at two
different RVP fuels were:
FTP Diurnal Emissions (g/test)
MYR Group
Pre-1985
1985+
As in MOBILE4, the FTP HDGV diurnal emissions in g/test for
any other RVP fuels are calculated by interpolating between these
two given emission levels. Assuming that in-use vehicles were
tested at the age of four, the MOBILE4.1 FTP diurnal emissions in
g/test for HDGVs at two different RVP fuels are:
M4.1 Diurnal (G/Test)
MYR Group 9.0 RVP 11.5 RVP
at
Pass
42.33
3.27
9.0 RVP
Failed Failed
Purge Pressure
42.33 42.33
10.34 32.14
Pass
77.89
17.89
at 11.5
Failed
Purge
77.89
33.67
RVP
Failed
Pressure
77.89
56.02
Pre-1985
1985+
42.33
5.22
77.89
20.86
2.1.7 Motorcycle Diurnal Emission Rates
There were no new data available for the motorcylce diurnal
emissions since the release of MOBILE4 model. Therefore, the
MOBILE4.1 diurnal emissions for motorcycles are the same as
those used for MOBILE4.
-13-
-------
2.2 Other Types of Diurnal Emissions
Three other types of diurnal emissions used in the
MOBILE4/MOBILE4.1 models are: in-use full diurnal emissions (FDI),
partial diurnal emissions (PDI), and multiple diurnal emissions
(MDI). The g/test diurnal emissions discussed in the previous
sections (2.1, 2.1.1 through 2.1.7) were all derived from one hour
heat build test results.
Under in-use conditions, however, vehicles may experience
diurnal emissions from a much slower temperature rise for a much
longer time frame, for example, for the entire period of six hours
when the ambient temperatures are increasing from the minimum to
the maximum. Sometimes one full diurnal may be interrupted (or cut
short) by the start of a trip during the ambient temperature rising
period so that vehicles may only experience a partial diurnal.
Vehicles may have multiple diurnal emissions when they have not
being driven for two or more consecutive days. Some vehicles may
have short but frequent trips or one long trip that they have no
diurnal emissions at all during one particular day. The fractions
of occurrences of these various types of diurnal emissions are
discussed in section 5.0. The levels of these various types of
diurnal emissions are discussed in the following sections.
2.2.1 In-Use G/Day Full Diurnal (FDI) Emissions
In MOBILE4, the in-use (6-hour) g/day full diurnal emissions
are the FTP one-hour emission rates (g/test) adjusted for 121 and
124 percent at 9.0 and 11.5 psi RVP fuels, respectively.
There has been very limited data available in this area since
the release of MOBILE4 model. The adjustment factors of 1.21 and
1.24 were derived from testing of one late model year carbureted
vehicle with its canister disconnected. It is unclear whether the
differences in emissions between a slower (six hours) temperature
rise and faster (one hour) heat build should be classified as
diurnal or resting loss emissions, as the levels seem to be more
similar to the latter. For these reasons, it is assumed that the
21 (or 24) percent emission increases due to the slower temperature
rise are the resting loss emissions and the in-use six-hour full
diurnal emission rates are the same as the one-hour FTP diurnal
emissions. Consequently, the MOBILE4 adjustment factors of 1.21
and 1.24 used for in-use full diurnal emission calculations are no
longer applicable in MOBILE4.1.
-14-
-------
2.2.2 Partial Diurnal (PDI) Emissions
Many of the MOBILE4 assumptions used for partial diurnals
remained unchanged for the MOBILE4.1 model, with one exception.
By definition, partial diurnals occur only when the vehicle's
engine has been turned-off for at least three hours, but less than
six hours, between the daily minimum and maximum temperatures when
the ambient temperatures are increasing (i.e., between 7 AM and
3 PM). The two temperatures used to calculate the partial diurnal
emissions are:
1) the ambient temperature at two hours (rather than the
one hour assumed in MOBILE4) after the end of a trip,
and,
2) the ambient temperature at the time when a new trip
starts.
The three types of partial diurnals and their characteristics are
described below:
Temperature Percent of
Description
8 AM - 11 AM
10 AM - 3 PM
8 AM - 2 PM
The temperatures at 10AM, 11AM, and 2PM are calculated by the
following equations:
Tio AM - Tmin + 0.53 * (Tmax - Tmin) (2.5)
Tii AM = Tmin + 0.71 * (Tmax - Tmin) (2.6)
T2 PM - Tmin + 0.94 * (Tmax - Tmin) (2.7)
where:
Tmin - the minimum temperature of the day, in °F
Tmax - the maximum temperature of the day, in °F.
The ambient temperature at 3PM is just the maximum ambient
temperature of the day.
Minimum
Tl o AM
Ti 2 PM
Tl o AM
Maximum
T 1 1 AM
Ta PM
Tj PM
Occurrence
32.36
7.09
3.69
-15-
-------
2.2.3 Multiple Diurnal (MDI) Emissions
In MOBILE4, due to a lack of data, the adjusted in-use diurnal
emissions from tampered vehicles were used for multiple diurnal
emission rates. It has been suggested that this assumption may not
be technically correct. With vehicles parked for two or more
consecutive days, their diurnal emission levels are dependent upon
many factors: ambient temperature rise, backpurge during the
cooling of the ambient temperature, canister capacity and
efficiency, fuel weathering, and tank fill level. When all these
parameters are being considered, the multiple diurnal emissions may
be lower (on a grams-per-day basis) than the diurnal emissions
under uncontrolled or tampered conditions.
Similar to the FTP one-hour (or in-use full) diurnal
emissions, the MOBILE4.1 multiple diurnal emissions model also
includes three elements: emissions from pass vehicles, emissions
from vehicles that failed pressure test, and emissions from
vehicles that failed purge test. Multiple diurnal emissions for
vehicles that failed either pressure or purge test are assumed to
be the same as their FTP one-hour diurnal counterparts. These
emissions were based on both the theoretical vapor generating model
and test data from EFP-Hammond (see discussions in section 2.1.2).
Multiple diurnal emission rates for pass vehicles were based
on the theoretical vapor generating model. Overall grams per day
emissions modeled from late model year fuel-injected technology
vehicles were simulated at various fuel volatility levels for one
to seven consecutive no-driving days. The temperature rise of 72
to 96°F and 40% tank fill level were assumed. Also, it is assumed
that backpurge occures in 50% of the vehicles experiencing multiple
diurnals.
To summarize results, Table 2 from Attachment has been copied
and presented at the top portion of Table 5. Ratios of emissions
from any day (day 2 through day 7) to emissions at day 1 for each
fuel volatility level were calculated and listed at the lower
portion of Table 5. These ratios, then, were weighted according to
the re-normalized fractions of occurrence for days 2 through 7.
Finally, a nonlinear regression line was fitted through the
weighted ratios. The equational form for the ratio as a function
of fuel volatility is:
CFMOI.P.SS - EXP (4.8491 - 0.33424 * RVP) (2.8)
Both the weighted and predicted ratios at each fuel volatility
level are given in Table 5. The multiple diurnal emissions for
pass vehicles are calculated by:
MDIpass - DIPass * CFMDi.Pasi (2.9)
-16-
-------
Using equations (2.8) and (2.9), the multiple diurnal
emissions for pass vehicles are sometimes calculated to be higher
than the emissions from vehicles that failed purge, especially for
pre-2.0 SHED evaporative emissions standard vehicles (e.g.,
pre-1981 LDGVs). One of the reasons is that the emissions
calculated to derive equation (2.8) were based on late model year
fuel-injected vehicle performance, while all pre-2.0 SHED
evaporative emissions standard vehicles are of carbureted
technology. Also, for the pre-2.0 SHED evaporative emissions
standard era, the emission levels for pass vehicles were similar to
the emission levels for vehicles that failed purge. Therefore, it
is reasonable to set the maximum levels of multiple diurnal
emissions for pass vehicles equal to the emissions from vehicles
that failed purge, i.e.,
MDIpass =MDIFalled Purge (2.10)
To summarize, multiple diurnal emissions for pass vehicles are
calculated by the following two methods:
1) use equations (2.8) and (2.9), or,
2) use equation (2.10), if the calculated emission levels
from equations (2.8) and (2.9) are greater than the
multiple diurnal emissions from vehicles that failed
purge.
These two methods are used for all LDGVs and LDGTs, pre-1981 model
years as well as 1981+ model years. As there is no multiple
diurnals assumed for HDGVs and MCs, these two methods are not
applicable to these two vehicle types.
-17-
-------
3.0 Hot Soak Emissions
As in MOBILE4, the hot soak emissions (in unit of grams per
one hour test) are described as two step functions of fuel
volatility in psi RVP and ambient temperature in °F. The hot soak
emissions are measured at FTP test conditions, i.e., with 9.0 psi
RVP fuel, 40% tank fill, and at the ambient temperature of 82°F.
In the model, hot soak emissions are first adjusted for fuel
volatility effect, then corrected for the temperature effect at
ambient temperatures 41°F and higher.
3.1 1981+ LDGVS and LDGTS
As in the MOBILE4.1 diurnal emissions model, the g/test hot
soak emissions are derived for pass vehicles, for vehicles that
failed pressure test, and for vehicles that failed purge test.
Table 3 also summarizes the average hot soak emissions from the
current EFP-MVEL data base.
3.1.1 Pass Vehicles
Hot soak emissions for pass vehicles were derived from EPA's
EFP-MVEL data. For model years 1981+ pass light-duty vehicles and
trucks, the derived equational forms for g/test hot soak emissions
are:
a + b * RVP
or,
HS p a s s
where:
Vehicle Type
/Fuel System
LDGV/CARB
LDGV/PFI
LDGV/TBI
LDGT/CARB
LDGT/FINJ
d * RVP + e * RVP
(3.1)
(3.2)
Regression Coefficient
0.14593 0.13823
0.46673 0.10297
0.198327 0.041297
0.16407
0.078327
0.13823
0.041297
-1.436657
18.437
-4.669710
-5.196600
-4.789710
0.69740
0.58219
0.0 0.034897
-4.2538 0.25072
0.58219 0.0
0.0
0.0
-18-
-------
Equation (3.1) is used when fuel volatility is less than
9.0 psi RVP, while equation (3.2) is used when fuel volatility is
equal to or high than 9.0 psi RVP. The coefficients from equation
(3.1) were regression results from data where the fuel RVP is
9.0 psi and from limited emissions data at lower fuel volatility
levels of 6.5 and 8.0 psi RVP. The coefficients from equation
(3.2) were derived from EFP-MVEL data, where the three fuel
volatility levels were 9.0, 10.5, and 11.7 psi RVP.
The constant terms of equations (3.1) and (3.2) (denoted as
"a" and "c" above) were adjusted so that at 9.0 psi RVP the
predicted hot soak emissions are equal to the average values at
9.0 psi RVP of the EFP-MVEL sample. Note that the lower bound of
fuel volatility for equation (3.1) is 5.0 psi RVP, and the lower
bound of hot soak emissions is set to equal to the resting loss
emissions.
Graphic presentations of 1981+ LDGV/LDGT hot soak emissions
for pass vehicles as a function of fuel volatility are shown in
Figures 5 and 6. There has been no new data available to develop
new hot soak temperature correction factors. Therefore, the
equations used to correct for temperature effect from MOBILE4 will
be used again in MOBILE4.1 for pass vehicles.
3.1.2 Failed Vehicles
During FY91, five vehicles were tested for hot soak emissions
with canister connected and disconnected at two levels of fuel
volatilities (9.0 and 10.5 psi RVP) and at three levels of ambient
temperatures (70, 82, and 95°F). These five vehicles were:
Model Year Make Fuel System
1979 Cutlass Carbureted
1983 Cutlass Carbureted
1983 Skylark TBI
1990 Taurus PFI
1990 Lumina PFI
Overall average hot soak emissions with canister disconnected
(or uncontrolled hot soak emissions) at various ambient
temperatures and fuel volatilities are summarized below:
Temperature Hot Soak (G/Test)
in °F 9.0 RVP 10.5 RVP
70.0 5.62 10.71
82.0 7.93 16.11
95.0 20.08 39.27
-19-
-------
Regression equation has been derived to describe the
relationships of the uncontrolled hot soak emissions, versus the
fuel volatility, and ambient temperature:
HSU
ncontrolled
EXP [2.2307 + 0.4443 * (RVP - 9.0)
+ 0.05114 * (TEMP - 82.0)]
where:
HS = uncontrolled hot soak emissions in g/test,
EXP » exponential function,
RVP = fuel volatility in psi, and
TEMP = ambient temperatures in °F.
Due to limited available data, it is assumed that the estimated
uncontrolled hot soak emissions are the emission levels for
vehicles that failed pressure test.
From EFP-Hammond data base, 18 of the 30 LDGVs (as listed in
Table 4) were also tested for hot soak emissions at 92°F ambient
temperature, and with 9.0 psi RVP fuel. These vehicles were found
to have failed purge test at the Hammond I/M Lane. The average hot
soak emissions were 10.56 g/test. Therefore, for vehicles that
failed purge test, the hot soak emissions can be estimated from the
above failed pressure hot soak emission equation but normalized at
the average emissions from Hammond data.
The hot soak emissions equations for failed vehicles are:
HSFaile
-------
Note that these equations can be extrapolated beyond the
ambient temperature ranges of 70.0 and 95.0°F and fuel volatility
levels of 9.0 and 10.5 psi RVP. TO be consistent with the hot soak
emissions calculations for pass vehicles, lower limits are also set
at 5.0 psi RVP fuel volatility and 41°F ambient temperature. There
are also a lower bound of hot soak emissions being equal to the
resting loss emissions and an upper bound of 40.00 g/test of hot
soak emissions for LDGVs.
Equations (3.3) and (3.4) are also used to calculate hot soak
emissions for all other vehicle types (such as LDGTls, LDGT2s, and
HDGVs) that failed pressure or purge test.
Graphic presentations of 1981+ LDGV/LDGT hot soak emissions as
a function of fuel volatility at 82°F for vehicles that failed
either pressure or purge test are shown in Figure 7.
3.1.3 Pre-1981 LDGV Hot Soak Emission Rates
The same list of methodologies used to derive diurnal
emissions, as described in section 2.1.3, is also used for hot soak
emissions calculations. The MOBILE4 FTP hot soak emissions in
g/test for pre-1981 LDGVs at two different RVP fuels were:
MYR Group
Pre-1971
1971
1972-77
1978-80
M4 Hot Soak (G/Test)
9.0 RVP 11.5 RVP
14.67
10.91
8.27
2.46
22.45
16.15
12.32
4.30
The derived MOBILE4.1 LDGV FTP hot soak emissions for the two RVP
fuels are:
FTP Hot Soak Emissions (g/test)
MYR Group
Pass
at 9.0
Failed
Purge
RVP
Failed
Pressure
Pass
at 11.5
Failed
Purge
RVP
Failed
Pressure
Pre-1971
1971
1972-77
1978-80
1981+ Garb
14.671
10. 433
7.773
1.793
1.392
14.671
14.671
9.98s
6.332
6.332
14. 671
14.671
14. 67"
9.312
9.312
28.261
14. 613
11.10s
3.89s
3.18Z
28.261
28.261
19. 232
19. 232
19. 232
28. 261
28. 261
28.26'
28. 262
28. 262
-21-
-------
Note that the 1981+ carbureted hot soak rates were calculated from
equations (3.2) through (3.4). They were listed above for
comparison purpose only, as all pre-1981 LDGVs are of carbureted
fuel delivery system. As in MOBILE4, the pre-1981 FTP LDGV hot
soak emissions in g/test for any other RVP fuels are calculated by
interpolating between these two given emission levels.
Using the assumption that in-use vehicles were tested at the
age of four, the MOBILE4.1 FTP hot soak emissions in g/test for
LDGVs at two different RVP fuels are:
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Garb
M4.1 Hot Soak (G/Test)
9.0 RVP 11.5 RVP
14.67
10.91
8.27
2.46
1.91
28.26
16.15
12.50
6.10
5-47
3.1.4 Pre-1981 LDGT1 Hot Soak Emission Rates
As in MOBILE4, the pre-1981 LDGV hot soak emission rates are
used also for pre-1981 LDGTls because of their similar evaporative
emission standards. The MOBILE4 FTP hot soak emissions in g/test
for pre-1981 LDGTls at two different RVP fuels were:
MYR Group
Pre-1971
1971
1972-77
1978-80
M4 Hot Soak (G/Test)
9.0 RVP 11.5 RVP
14.67
10.91
8.27
2.46
22-45
16.15
12.32
4.30
The derived MOBILE4.1 LDGT1 FTP hot soak emissions for the two RVP
fuels are:
FTP Hot Soak Emissions (g/test)
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Garb
at 9.0 RVP
Pass
Failed
Purge
Failed
Pressure
at 11.5 RVP
Pass
Failed
Purge
Failed
Pressure
14.67
10.43
7.77
1.79
1.08
14.67
14.67
9.98
6.33
6.33
14.67
14.67
14.67
9.31
9.31
28.26
14.61
11.10
3.89
2.82
28.26
28.26
19.23
19.23
19.23
28.26
28.26
28.26
28.26
28.26
-22-
-------
Note that the 1981+ carbureted hot soak rates were calculated from
equations (3.2) through (3.4). They were listed above for
comparison purpose only, as all pre-1981 LDGTls are of carbureted
fuel delivery system. As in MOBILE4, the pre-1981 FTP LDGT1 hot
soak emissions in g/test for any other RVP fuels are calculated by
interpolating between these two given emission levels.
Assuming that in-use vehicles were tested at the age of four,
the MOBILE4.1 FTP hot soak emissions in g/test for LDGTls at two
different RVP fuels are:
MYR Group
Pre-1971
1971
1972-77
1978-80
1981+ Carb
M4.1 Hot Soak (G/Test)
9.0 RVP 11.5 RVP
14.67
10.91
8.27
2.46
1.83
28.26
16.15
12.50
6.10
5.15
3.1.5 Pre-1981 LDGT2 Hot Soak Emission Rates
As in MOBILE4, the same post-1979 LDGT1 hot soak emission
rates are assumed for LDGT2s, based on their similar evaporative
emission standards. The pre-1979 LDGT2 hot soak emission rates
were the same as the pre-controlled HDGV hot soak rates, as there
were no separate vehicle category of LDGT2s before the model year
of 1979. The MOBILE4 FTP hot soak emissions in g/test for pre-1981
LDGT2s at two different RVP fuels were:
MYR Group
Pre-1979
1979-80
M4 Hot Soak (G/Test)
9.0 RVP 11.5 RVP
18.08
2.46
27.66
4.30
The derived MOBILE4.1 LDGT2 FTP hot soak emissions for the two RVP
fuels are:
FTP Hot Soak Emissions (q/test)
MYR Group
Pre-1979
1979-80
1981+ Carb
at 9 . 0 RVP
Pass
Failed
Purge
Failed
Pressure
at 11.5
Failed
Pass Purge
RVP
Failed
Pressure
18.08 18.08
1.79 6.33
1.08 6.33
18.08
9.31
9.31
44.16
3.89
2.82
44.16
19.23
19.23
44.16
28.26
28.26
-23-
-------
As in MOBILE4, the pre-1981 FTP LDGT2 hot soak emissions in g/test
for any other RVP fuels are calculated by interpolating between
these two given emission levels.
Assuming that in-use vehicles were tested at the age of four,
the MOBILE4.1 FTP hot soak emissions in g/test for LDGT2s at two
different RVP fuels are:
M4.1 Hot Soak (g/test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1979 18.08 44.16
1979-80 2.46 6.10
1981+ Garb 1.83 5.15
3.1.6 HDGV Hot Soak Emission Rates
The MOBILE4 FTP hot soak emissions in g/test for HDGVs at two
different RVP fuels were:
M4 Hot Soak (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1985 18.08 27.66
1985+ 2.12 4.77
As in the diurnal emissions calculations, the same algorithm
from MOBILE4 HDGV evaporative emissions are used to estimate 1985+
hot soak emission rates for pass vehicles, as well as HDGVs that
failed either pressure or purge test. The bases used in the
calculations were the 1981+ LDGT1 hot soak emission rates
(1.08/2.82 g/test for pass vehicles, 9.31/28.26 g/test for vehicles
that failed pressure test, and 6.33/19.23 g/test for vehicles that
failed purge test, at 9.0 and 11.5 psi RVP fuels, respectively).
The new MOBILE4.1 FTP g/test hot soak emissions for HDGVs at
two different RVP fuels are:
FTP Hot Soak Emissions (g/test)
at
MYR Group
Pre-1985
1985 +
at 9.0 RVP
Failed Failed
Pass Purge Pressure
18.08 18.08 18.08
1.69 9.89 14.55
at 11.5 RVP
Failed Failed
Pass Purge Pressure
44.16 44.16 44.16
4.41 30.05 44.16
-24-
-------
As in MOBILE4, the FTP HDGV hot soak emissions in g/test for
any other RVP fuels are calculated by interpolating between these
two given emission levels.
Assuming that in-use vehicles were tested at the age of four,
the MOBILE4.1 FTP hot soak emissions in g/test for HDGVs at two
different RVP fuels are:
Hot Soak (G/Test)
MYR Group 9.0 RVP 11.5 RVP
Pre-1985 18.08 44.16
1985+ 2.86 8.06
3.1.7 Motorcycle Hot Soak Emission Rates
There were no new data available for motorcylces hot soak
emission rates since the release of MOBILE4 model. Therefore, the
MOBILE4.1 hot soak emissions for motorcycles are the same as those
used for MOBILE4.
-25-
-------
4.0 High Altitude
In MOBILE4 model, a set of altitude adjustment factors was
used to calculate the evaporative diurnal and hot soak emissions
for vehicles operating in high altitude region. These altitude
adjustment factors were based on either actual test data or from
their emissions standard ratio. For example, the MOBILE4 FTP
diurnal and hot soak emissions in g/test for high altitude LDGVs at
two different RVP fuels were:
MYR Group
Pre-1971
1971
1972-76
1977
1978-80
Diurnal (G/Test)
9.0 RVP 11.5 RVP
Hot Soak (G/Test)
9.0 RVP 11.5 RVP
33.90
21.16
17.15
8.98
13.36
62.39
50.15
44.93
23.53
36.85
19.07
14.18
17.15
8.27
6.37
29.18
20.99
20.96
12.32
11.15
Note that the emission rates for model years 1972-76 were actual
test results from EPA's EFP test facility in Denver, Colorado. The
emissions for model year 1977 were the same as the low altitude
rates (because of their same evaporative emissions standard). The
emission rates for model years 1978-80 were estimated by
multiplying a factor of 2.59 to the low altitude rates. Finally,
the emission rates for all other model year groups were calculated
by multiplying an adjustment factor of 1.30.
The same set of altitude adjustment factors is used again in
MOBILE4.1 and are summarized in Table 6. The only exception is
that, for model years 1978-80, the MOBILE4 adjustment factor of
2.59 has been changed to 1.30. These high altitude adjustment
factors are used for both diurnal and hot soak emissions of pass
vehicles, as well as vehicles that failed either pressure or purge
test. For example, the derived MOBILE4.1 high altitude LDGV
diurnal emissions at two RVP fuel levels are:
LDGV Diurnal Emissions (G/Test)
at 9.0 RVP
at 11.5 RVP
MYR Group
Pre-1971
1971
1972-76
1977
1978-80
1981-83 Garb
1984+ Garb
Pass
Failed
Purge
33.90
19.54
10.16
8.27
5.71
2.16
1.66
33.90
33.90
13.82
9.33
7.36
7.36
5.66
Failed
Pressure
33.90
33.90
27.46
20.39
22.85
22.85
17.58
Failed
Pass Purge
Failed
Pressure
62.39
48.59
33.13
22.80
17.21
9.62
7.40
62.39
62.39
45.06
23.83
23.95
23.95
18.42
62.39
62.39
56.15
35.45
39.83
39.83
30.64
-26-
-------
Similarly, the derived MOBILE4.1 high altitude LDGV
emissions at two RVP fuel levels are:
LDGV Hot Soak Emissions (G/Test)
hot soak
at 9.
0 RVP
Failed
MYR Group
Pre-1971
1971
1972-76
1977
1978-80
1981-83 Garb
1984+ Carb
Pass
19.
13.
10.
7.
2.
1.
1.
07
56
03
77
33
81
39
Pur
19
19
14
9
8
8
6
Fai
led
ge Pressure
.07
.07
.52
.98
.23
.23
.33
19
19
19
14
12
12
9
.07
.07
.07
.67
.10
.10
.31
at 11
.5 RVP
Failed
Pass
36.
18.
18.
11.
5.
4.
3.
74
99
87
10
06
13
18
Pur
36
36
27
19
25
25
19
ge
.74
.74
.30
.23
.00
.00
.23
Failed
Pressure
36.74
36.74
36.74
28.26
36.74
36.74
28.26
Assuming that in-use vehicles were tested at the age of four,
the MOBILE4.1 FTP diurnal and hot soak emissions in g/test for high
altitude LDGVs at two different RVP fuels are:
Diurnal (G/Test)
MYR Group
Pre-1971
1971
1972-76
1977
1978-80
1981-83 Carb
1984+ Carb
9.0 RVP
33.90
21.16
11.30
8.98
6.72
3.57
2.74
11.5 RVP
62.39
50.15
35.07
23.53
18.81
12.08
9.29
Hot Soak (G/Test)
9.0 RVP 11.5 RVP
19.07
14.18
10.91
8.27
3.20
2.74
2.11
36.74
20.99
20.32
12.50
7.94
7.11
5.47
-27-
-------
5.0 Trip Related Estimates
Most of the MOBILE4 trip-related estimates, such as trips per
day (TPD), miles per day (MPD), fractions of occurrences for FDI,
MDI, and PDI are also used in MOBILE4.1, with some modifications.
There is no new data available to derive new estimates. Therefore,
the data base where these estimates were derived from is the same
1979 GM-NPD survey data used in MOBILE4.
5.1 Trips Per Day and Miles Per Day
The trips per day and miles per day equations used for LDGVs
and LDGTs have the following forms:
TPD - 4.7187 - 0.058508 * AGE (5.1)
MPD - JAMAR / 365.0 (5.2)
where: TPD = estimated number of trips per day,
AGE - vehicle age in years,
MPD = estimated miles driven per day,
JAMAR » January 1 annual mileage accumulation rate, which is
also a function of vehicle age.
These equations were used in MOBILE4 to calculate the estimated
trips per day and miles per day for vehicles at ages 1 through 20.
In MOBILE4.1, equations (5.1) and (5.2) are used to obtain TPD and
MPD estimates for LDGVs and LDGTs for ages 1 through 25.
The TPD and MPD estimates for HDGVs and MCs remain unchanged
in MOBILE4.1:
HDGVs MCs
TPD 6.88 1.35
MPD 33.97 10.02
-28-
-------
5.2 Other Trip-Related Estimates — LDGVS and LDGTS
Percents of other trip-related estimates for LDGVs and LDGTs
used in MOBILE4 were:
Total
Description Percent Percent
Driving Days 76.15 76.15
Single Full Diurnal Days 22.37
Partial Diurnal Days 43.14
i) 8 AM to 11 AM 32.36%
ii) 10 AM to 3 PM 7.09%
iii) 8 AM to 2 PM 3.69%
3+ multiple Diurnal Days 3.89
No Diurnal Days 6.75
No Driving Days 23.85 23.85
Single Full Diurnal Days 11.59
2+ Multiple Diurnal Days 12.26
In MOBILE4.1, all the above estimates are expressed in terms
of vehicle's age. For example, the proportions of vehicles having
at least one trip on any given day for age 1 vehicles are higher
than those for age 2 vehicles, etc. The equational forms for
various types of driving days (in percent) are:
Driving Days - 87.07 - 1.72 * AGE (5.3)
Single Full Diurnal = 25.58 - 0.5 * AGE (5.4)
Partial Diurnal Days:
8AM-11AM Partial Diurnal - 37.0 - 0.73 * AGE (5.5)
10AM-3PM Partial Diurnal = 8.11 - 0.16 * AGE (5.6)
8AM-2PM Partial Diurnal - 4.22 - 0.08 * AGE (5.7)
3+ Multiple Diurnal = 4.45 - 0.09 * AGE (5.8)
No Diurnal Days - 7.72 - 0.15 * AGE
The equational forms for various types of driving days (in percent)
are:
No Driving Days - 12.93 + 1.72 * AGE
Single Full Diurnal Days - 6.31 + 0.84 * AGE (5.9)
2+ Multiple Diurnal Days - 6.62 + 0.88 * AGE (5.10)
Table 7 summarizes the results.
-29-
-------
5.5 Other Trip-Related Estimates for HDGVs
Percents of all trip-related estimates for HDGVs used in
MOBILE4 were:
Total
Description Percent Percent
Driving Days 27.10 27.10
No Driving Days 72.90 72.90
Single Full Diurnal Days 49.80
2 Multiple Diurnal Days 13.70
3+ Multiple Diurnal Days 9.40
Due to a lack of data, no equations as functions of vehicle's
age were derived. These estimates are used in MOBILE4.1 again.
-30-
-------
Table 1
Pressure/Purge Failure Rates
by Vehicle Age
Source: EPA's EFP-Hammond Data
Age of LDGV Failure Rates
Vehicle Purge Pressure Either
01 0.043 0.043 0.080
02 0.043 0.043 0.080
03 0.043 0.043 0.080
04 0.060 0.053 0.096
05 0.060 0.053 0.103
06 0.069 0.053 0.120
07 0.094 0.053 0.150
08 0.094 0.106 0.188
09 0.142 0.152 0.258
10 0.214 0.171 0.323
11 0.224 0.236 0.389
12 0.229 0.336 0-442
13+ 0.229 0.336 0.451
-31-
-------
Table 2
Criteria Used to Define Pressure and Purge Failures
Emission Factors Programs
Evaporative System Component Diagnosis
Purge Failures
Canister Purge Solenoid/Valve Missing
Disconnected or Bypassed
Leaks Vacuum
Sticking
Inoperative
Vacuum or Vent Lines Disconnected or Missing
Plugged or Damaged
Misrouted
Purge Hose disconnected or Missing
Split or Not sealed
Canister Purge TVS Stuck
ECM signal to purge solenoid None detected
Pressure Failures
Gas Cap Non OEM or Missing
Leaking
Sending Unit Gasket Leaking
Fuel Tank Rollover Valve leaking or Stuck
Fuel Tank Filler Neck Leaking
Both Pressure and Purge Failures
Canister Missing
Carburetor Bowl Vent Line Disconnected or Leaks
Fuel Line to Canister Disconnected
EFE Control Switch Missing
Bowl Vent Solenoid Disconnected
-32-
-------
Table 3
Date: April 26, 1991
Statistics on Evaporative Emissions
(Average G/Test Hot Soak and Diurnal)
Source: EPA's EFP-MVEL
Fuel
Delivery
System
All
Carbureted
PFI
TBI
Failed Purge
Carbureted
PFI
TBI
9.0 RVP
10.4 RVP
11.7 RVP
N
HS
DI
N
HS
DI
N
HS
DI
Light-Duty Gasoline-Powered Vehicles
386 1.99
347 0.70
315 0.82
Failed Pressure
Carbureted 37
PFI 12
TBI 9
(FPurge,FPressure)
All 115
Tampered
Carbureted
PFI
TBI
Pass
Carbureted
PFI
TBI
Others*
Carbureted
PFI
TBI
(Pass,Others)
Carbureted
PFI
TBI
2.65
1.90
1.75
158
69
95
3-43
2.09
1.42
5.52
4.80
3.90
31 5.32 5.11
13 4.27 6.93
13 3.26 7.19
3.15 7.44
3.22 10.84
2.77 9.21
17 7.14 8.76
2 16.21 7.70
2 3.27 13.52
17 6.14 9.76
3 9.16 9.36
2 1.25 11.77
3.85 7.22 43 6.86 9.51
7 7.77 10.28
1 0.29 13.79
2 11.49 10.52
258 1.21
299 0.41
233 0.47
1.38
1.03
0.93
2 16.00 22.70
fl-
o-
gs 1.89 3.74
61 1.36 3.66
79 0.85 3.32
53
22
58
2
1
1
.27
.19
.01
3
5
2
.02
.42
.38
27
3
12
3
0
4
.88
.51
.91
5
21
4
.80
.57
.77
311
321
291
1.39
0.46
0.57
1.66
1.33
1.22
122
64
91
2.33
1.32
1.39
4.20
4.50
3.51
380
254
274
4.17 9.59
3.84 6.85
2.99 8.91
30 10.25 15.19
9 21.07 12.63
10 13.69 21.71
37 5.10 15.93
7 10.96 25.27
7 13.92 24.42
100 9.97 17.24
7 9.05 19.36
1 0.20 21.44
1 36.43 28.28
253 3.25 7.25
219 2.97 5.64
206 1.77 7.68
53 3.79 11.85
18 3.25 10.78
50 3.71 8.86
306 3.34 8.05
237 2.99 6.03
256 2.15 7.91
*A11 vehicles that had some non-zero ECOMP codes but were not
defined as either purge, failure, pressure failure, or tampered.
-33-
-------
Table 3 (Continued)
Date: April 26, 1991
Statistics on Evaporative Emissions
(Average G/Test Hot Soak and Diurnal)
Source: EPA's EFP-MVEL
Fuel
Delivery 9.Q RVP 10.4 RVP 11.7 RVP
System N HS PI N HS PI N HS PI
Light-Puty Gasoline-Powered Trucks
All
Carbureted 170 1.53 2.93 6 5.35 10.01 199 4.13 14.02
Fuel-Injected 80 0.58 1.90 0 - - 42 1.32 8.64
Failed Purge
Carbureted 7 5.06 9.44 1 21.68 25.47 12 10.20 17.36
Fuel-Injected 1 0.19 0-62 0 - - 0 -
Failed Pressure
Carbureted 10 2.89 8.71 0 - 11 7.23 23.65
Fuel-Injected 2 0.40 22.42 0 1 6.81 54.35
(FPurge,FPres)
All 20 3.27 9.93 1 21.68 25.47 24 8.70 21.79
Tampered
Carbureted 4 8.94 8.41 0 - - 6 19.47 32.06
Fuel-Injected 1 11.10 23.32 0 1 26.94 53.62
Pass
Carbureted 129 1.00 1.69 4 2.31 3.63 128 2.19 10.54
Fuel-Injected 69 0.43 0.95 0 37 0.36 6.75
Others*
Carbureted 20 1.59 4.68 1 1.17 20.07 42 5.31 18.58
Fuel-Injected 7 0.62 2.50 0 - ' - 3 2.83 1.70
(Pass,Others)
Carbureted 149 1.08 2.09 5 2.08 6.92 170 2.96 12.52
Fuel-Injected 76 0.45 1.09 0 - - 40 0.55 6.37
*A11 vehicles that had some non-zero ECOMP codes but were not
defined as either purge, failure, pressure failure, or tampered.
-34-
-------
Table 4
Failed Pressure/Purge Test Results
92°F Hot Soak, 9 psi RVP Fuel
72-96°F Diurnal Heat Build, 40% Tank Fill
Source: EPA's EFP-Hammond
Vehicle #
576
640
660
665
724
443
736
1541
1635
1640
1530
1552
722
1542
740
1524
673
1526
1612
1532
743
1537
1544
1529
659
1533
1548
1555
730
1525
Fuel
Failed
Model Year System Hot Soak
1981
1981
1981
1981
1982
1983
1983
1983
1984
1984
1984
1984
1985
1985
1985
1985
1985
1985
1985
1986
1986
1986
1986
1987
1988
1988
1988
1988
1989
1989
Garb
PFI
Garb
Garb
Garb
TBI
Garb
TBI
PFI
PFI
TBI
TBI
PFI
PFI
Garb
Garb
TBI
TBI
TBI
PFI
TBI
TBI
TBI
PFI
PFI
PFI
PFI
TBI
PFI
PFI
N:
Average:
1.95
15.96
6.17
12.05
1.62
3.70
26.08
1.73
9.26
5.54
33.30
25.37
0.96
2.77
0.52
20.97
7.36
14.71
18
10.557
Purge Failed Pressure
Diurnal
13.32
11.93
2.68
13.36
20.08
21.92
25.55
19.30
22.85
13.01
22.22
14.40
2.80
21.34
0.65
34.61
15.44
4.90
18
15.576
Diurnal
76.22
77.92
24.76
17.55
10.85
17.01
16.38
11.38
14.47
5.41
33.45
27.37
12
27.727
-35-
-------
Table 5
Multiple Diurnal Emissions
Pass Vehicles
Source: Table 2 of Attachment
Fractions
of
Day Occurrence
1
2
3
4
5
6
7
0.341
0.078
0.037
0.021
0.014
0.008
0.003
2
3
4
5
6
7
0.484
0.230
0.130
0.087
0.050
0.019
7.5
RVP
8.0 8.5 8.7 9.0
RVP RVP RVP RVP
10.0
RVP
11.0
RVP
Calculated Emissions (in Grams)
0.2
1.6
2.6
3.3
3.8
4.3
4.7
0.4 0.7 0.8 1.1
2.4 3.4 3.8 4.7
3.6 5.0 5.7 6.8
4.5 6.2 6.9 8.2
5.3 7.0 7.9 9.2
5.8 7.7 8.6 10.0
6.3 8.2 9.0 10.5
2.5
8.7
11.9
13.8
15.0
15.8
16.4
5.3
15.5
20.0
22.3
23.6
24.3
24.7
Calculated Emission Ratios
(Relative to Day 1 Emissions)
8.0
13.0
16.5
19.0
21.5
23.5
6.0 4.9 4.8 4.3
9.0 7.1 7.1 6.2
11.3 8.9 8.6 7.5
13.3 10.0 9.9 8.4
14.5 11.0 10.8 9.1
15.8 11.7 11.3 9.5
3.5
4.8
5.5
6.0
6.3
6.6
2.9
3.8
4.2
4.5
4.6
4.7
Occurrence Weighted Ratio
12.2 8.6 6.8 6.7 5.8
4.5
3.5
10.4
Predicted Ratio*
8.8 7.4 7.0
6.3
4.5
3.2
*Predicted ratios are calculated from equation:
Ratio - EXP (4.8491 - 0.33424 * RVP)
-36-
-------
Table 6
High Altitude Adjustment Factors
Evaporative Emissions
Vehicle
Type
LDGVs
LDGTlS
Model Year
Group
Pre-1972
1972-76
1977
1978-81
1982-83
1984 +
Pre-1972
1972-76
1977
1978-81
1982+
Adjustment
Factor
1.30
*
1.00
1.30
1.30
1.00
1.30
*
1.00
1.30
1.30
LDGT2S
HDGVs
MCS
All
All
All
1.30
1.30
1.30
* Based on actual test data.
-37,-
-------
TABLE 7
Trip-Related Estimates (in Percent)
LDGVs and LDGTs
Source: 1979 GM-NPD Survey Data
Trip Days
Veh
Age
01
02
03
04
05
06
07
OS
09
10
11
12
13
14
15
16
17
IS
19
20
21
22
23
24
25 +
Trip
Days
85.35
83.63
81.91
80.19
78.47
76.75
75.03
73.31
71.59
69.87
69.87
68.15
66.43
64.71
62.99
61.27
59.55
57.83
56.11
54.39
50.95
49.23
47.51
45.79
44.07
Single
Full
DI
Days
25.08
24.58
24.08
23.58
23.08
22.58
22.08
21.58
21.08
20.58
20.08
19.58
19.08
18.58
18.08
17.58
17.08
16.58
16.08
15.58
15.08
14.58
14.08
13.58
13.08
Partial
Diurnal Days
SAM-
HAM
36.27
35.54
34.81
34.08
33.35
32.62
31.89
31.16
30.43
29.70
28.97
28.24
27.51
26.78
26.05
25.32
24.59
23.86
23.13
22.40
21.67
20.94
20.21
19.48
18.75
10AM
8AM
-3PM -2PM
7.95
7.79
7.63
7.47
7.31
7.15
6.99
6.83
6.67
6.51
6.35
6.19
6.03
5.87
5.71
5.55
5.39
5.23
5.07
4.91
4.75
4.59
4.43
4.27
4.11
4.14
4.06
3.98
3.90
3.82
3.74
3.66
3.58
3.50
3.42
3.34
3.26
3.18
3.10
3.02
2.94
2.86
2.78
2.70
2.62
2.54
2.46
2.38
2.30
2.22
3 +
MDI
Days
4.36
4.27
4.18
4.09
4.00
3.91
3.82
3.73
3.64
3.55
3.46
3.37
3.28
3.19
3.10
3.01
2.92
2.83
2.74
2.65
3.46
3.37
3.28
3.19
3.10
No
DI
Days
7.57
7.42
7.27
7.12
6.97
6.82
6.67
6.52
6.37
6.22
6.07
5.92
5.77
5.62
5.47
5.32
5.17
5.02
4.87
4.72
6.07
5.92
5.77
5.62
5.47
No
No
Trip
Da
14
16
18
19
21
23
24
26
28
30
31
33
35
37
38
40
42
43
45
47
31
33
35
37
38
ys
.65
.37
.09
.81
.53
.25
.97
.69
.41
.13
.85
.57
.29
.01
.73
.45
.17
.89
.61
.33
.85
.57
.29
.01
.73
Trip^ Days
Single
Full
DI
Days
7.15
7.99
8.83
9.67
10.51
11.35
12.19
13.03
13.87
14.71
15.55
16.39
17.23
18.07
18.91
19.75
20.59
21-43
22.27
23.11
15.55
16.39
17.23
18.07
18.91
2 +
MDI
Days
7.50
8.38
9.26
10.14
11.02
11.90
12.78
13.66
14.54
15.42
16.30
17.18
18.06
18.94
19.82
20.70
21.58
22.46
23.34
24.22
16.30
17.18
18.06
18.94
19.82
-38-
-------
ATTACHMENT
-39-
-------
Figure 1
1981+ Passing Vehicles LDGV Diurnal Emissions (g/test)
16 -r
i i"n—r~ i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7
UDI
10:34 AM 7/25/91
-------
Figure 2
1981+ Passing Vehicles LDGT Diurnal Emissions (g/test)
20 -r
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
UDI
2 2.1 2.2 2.3 2.4 2.5 2.6 2.7
10:38 AM 7/25/91
-------
Figure 3
1981+ Failed Vehicles LDGV Diurnal Emissions (g/test)
45 T
to
03 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 12 1-3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 2.7
10:31 AM 7/25/91
-------
Figure 4
1981+ Failed Vehicles LDGT Diurnal Emissions (g/test)
50 T
GO
Failed
Pressure
Failed
Purge
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
2.1 2.2 2.3 2.4 2.5 2.6 2.7
8:38 AM 6/25/91
-------
Figure 5
1981+ Passing Vehicles at 82°F LDGV Hot Soak Emissions (g/test)
3.5 T
5.5
6.5
7.5 8 8.5 9 9.5
Fuel Volatility (in psi RVP)
10 10.5 11 11.5 12
9:11 AM 6/25/91
-------
Figure 6
1981+ Passing Vehicles at 82°F LDGT Hot Soak Emissions (g/test)
3 T
en
H
o 2.5
t
S
o
a
k
E
m
i
s
s
i
o
n
s
2 -
1.5
0.5 -'
0
FINJ
CAR
5.5
6.5
7.5 8 8.5 9 9.5
Fuel Volatility (in psi RVP)
10
10.5
11
11.5
12
10:10 AM 6/25/91
-------
Figure 7
1981+ Failed Vehicles at 82°F LDGV & LDGT Hot Soak Emissions (g/test)
5.5
6.5
7.5 8 8.5 9 9.5
Fuel Volatility (in psi RVP)
10
10.5
11
11.5
12
9:14 AM 7/3/91
-------
RUNNING LOSS EMISSIONS
1.0 Introduction
Running loss emissions test program has been performed and
data collected by Automotive Testing Laboratories, Inc. (ATL)
located at South Bend, Indiana under EPA's Emission Factors Program
contract.
The test program was designed to test in-use vehicles with
three different driving cycles:
1) a low speed cycle (known as the New York City Cycle
[NYCC]) with an average speed of 7.1 mph,
2) a Federal Testing Procedure (FTP) LA-4 driving cycle
with an average speed of 19.6 mph, and,
3) a high speed cycle (or Highway Fuel Economy Test [HFETl)
with an average speed of 47.9 mph.
The duration of the running loss test is approximately one hour for
each driving cycle. Therefore, the NYC driving cycle is repeated
six times (6 bags), the two portions of the LA-4 cycle are repeated
three times (6 bags), and the HFET driving cycle is repeated five
times (5 bags). The distances, average speeds, and durations of
these three driving cycles are summarized in the following:
Distance Speed Duration
Driving in in in
Cycle Bags Miles mph Minutes
NYCC 1 to 6 1.18 7.1 10.0
LA-4 1,3,5 3.59 25.6 8.4
2,4,6 3.89 16.2 14.4
HFET 1 to 5 10.20 47.9 12.8
Under this test program, many vehicles were tested only on LA-4
cycles. Fewer vehicles were tested on HFET cycle.
The running loss emissions test program was designed to
collect data at four levels of fuel volatility (7.0, 9.0, 10.4,
11.7 psi in Reid Vapor Pressure [RVP]) and three levels of ambient
temperature (80, 95, and 105°F).
-47-
-------
Not all vehicles were tested for all combinations of fuel RVPs
and ambient temperatures, however. There is usually no testing at
extreme conditions, such as the combinations of high RVP fuel and
high ambient (11.7 psi/105°F), and low RVP fuel and low ambient
(7.0 psi/80°F), because of their less likely occurrences in the
real world. Also, if from a test vehicle the running loss emission
results are low (less than 0.5 grams) at certain fuel and
temperature combination (for example, 9.0 psi/95°F), it is assumed
that at the combinations of lower fuel volatilities and/or lower
ambient temperatures (i.e., 7.0 psi/95°F, 9.0 psi/80°F, and,
7.0 psi/80°F), this vehicle would have emissions at a similarly low
level. To save resources, the vehicle is not tested for the
combinations of lower fuel volatilities and lower ambient
temperatures. Further, there has been no testings on 11.7 psi RVP
fuel shortly after the issuance of MOBILE4 in 1989.
-48-
-------
2.0 True Vapor Pressure
In MOBILE4 model, when the test data is not available at
certain combinations of fuel volatility and ambient temperature,
the g/mi running loss emissions were estimated from a variable
called "True Vapor Pressure (TVP)." in MOBILE4.1 model, this TVP
is used to correlate with the running loss emissions from failed
vehicles. These TVPs by bag are expressed as functions of fuel
volatility and fuel tank temperature, and are calculated by the
following five steps of equations:
U7
V7
1.0223*RVP
66.401 - 12.718*U7
0.0018407*U7**4
[0.0357*RVP / (1.0 - 0.0368*RVP)]
1.3067*U7**2 - 0 . 077934*U7**3
TB i = Tmin + DTB
for i=l,2,...,6
CD7Bi = 262.0 / [(V7/6.0 + 560.0) - 0.0133] * (100.0 - TBi
+ V7
TVP
Bi -
14.697 - 0.5308*CD7Bi
0.000055631*CD7Bi**3 + 1.769*10
0 . 0077215*CD7B i **2
*CD7B1**4
for i =
-1,2.
where:
RVP
TBi
Tmin
DTBi
in °F,
TVP
B i =
fuel volatility in psi,
cumulative tank temperature for bag i
initial tank temperature in °F,
cumulative tank temperature rise for bag i
and,
calculated bag i TVP value..
in °F,
The TVP values were calculated for all combinations of fuel
volatilities (7.0, 9.0, 10.4, and 11.7 psi RVP) and tank
temperature profiles (with the initial tank temperatures at 80, 87,
95, and 105°F).
The tank temperature rises for each bag were estimated from a
44-car tank temperature data collected at Laredo, Texas during the
time frames of September 11 through October 23, 1989, by ATL under
EPA's contract. In this program, vehicles' tank temperatures were
measured and recorded while being driven under the LA-4 driving
cycle repeated for three times (thus, a total of six bags). The
initial tank temperatures were adjusted to be within + 2°F of the
ambient temperatures at the start of the test. An ambient
temperature of 95°F was used. The fuel tanks were filled at 40%
level with 9.0 psi RVP fuel.
-49-
-------
The initial and final tank temperatures for each bag from
these 44 vehicles were examined and an average tank temperature
rise for each bag was calculated. The resulting average tank
temperature rises for the six bags are: 5, 9, 4, 5, 2, and 3°F,
respectively. The cumulative tank temperature rises (5, 14, 18,
23, 25, and 28°F), then, were used in the above five equations to
calculate the corresponding TVP values. Therefore, the initial and
final tank temperatures for the four temperature profiles were:
Initial Tank Final Tank Temperature in °F
Temp in °F Bag 1 Bag 2 Bag 3 Bag 4
80.0
87.0
95.0
105.0
Table 1 summarizes the calculated TVP values for each bag under
each fuel volatility level and each initial ambient (tank)
temperature.
85.0
92.0
100.0
110.0
94.0
101.0
109.0
119.0
98.0
105.0
113.0
123.0
103.0
110.0
118.0
128.0
105.0
112.0
120.0
130.0
108.0
115.0
123.0
133.0
-50-
-------
3.0 Running Loss Emissions Data
Similar to EPA's Hammond program, all vehicles under the
running loss emissions test program since FY90 were also checked
for pressure and/or purge failures of their evaporative emissions
control system. Descriptions of these pressure/purge tests were
given in the Evaporative HC Emissions Chapter. Failure rates from
either pressure and/or purge test were described as a function of
vehicle's age. These failure rates are the same as those used for
the diurnal/hot soak emissions model.
Therefore, the MOBILE4.1 running loss emissions model include
three portions: emissions for pass vehicles, emissions for
vehicles that failed pressure test, and emissions for vehicles that
failed purge test.
For vehicles tested prior to the Hammond program, diagnoses
and comments from mechanics and their running loss emissions test
results at 9.0 psi/95°F were examined to see which category the
vehicles should be grouped into. For example, if diagnoses and
comments from mechanics indicated that there exists a leakage on
one of the evaporative emissions control system components (such as
gas cap, filler neck, sending unit, rollover valve, or vent hoses),
the vehicle is categorized as "failed pressure test". If there
exists an in-operative canister purge solenoid or valve, or
disconnected, missing, or damaged purge hoses, the vehicle is
grouped as "failed purge test." If there were no problems
detected, the vehicle is categorized as "pass vehicle."
Among those pre-Hammond vehicles that had problems detected in
their evaporative emissions control system components by the
mechanics' diagnoses and comments, some of them had relatively low
running loss emissions. Among those pre-Hammond vehicles that had
no visible problems detected in their evaporative emissions control
system components, some of them had very high running loss
emissions. Due to this lack of consistency, it was decided that
all LDGVs tested prior to Hammond program be excluded from
MOBILE4.1 model. For LDGTs, however, the sample sizes were very
limited, their running loss emissions test results were relatively
consistent, with no visible problems detected. For these reasons,
all LDGT data were used as pass vehicles for the LDGT fleet.
-51-
-------
3.1 Exception Vehicles
Among the data used for MOBILE4.1 analysis, there were two
vehicles with "unusual" test results. One of them is a 1986
Chevrolet Camaro, identified as vehicle #1532, which is equipped
with a 5.0L engine, ported fuel injection fuel delivery system, by
engine family G1G5.0V8NTA8, and with evaporative emissions control
system 6BO-1A. This vehicle, when tested at Hammond I/M lane,
passed pressure check but failed purge flow check. The other
vehicle is a 1986 Pontiac Sunbird, identified as vehicle #1578,
which is equipped with a 1.6L engine, throttle body fuel injection
fuel delivery system, by engine family G2G1.8V5TDG2, and with
evaporative emissions control system 6AO-2B. This vehicle was
found to have a leaking gas cap. It failed pressure check but
passed purge flow check.
Two things were observed on these two vehicles: their tank
temperature rises were extremely high, and their emission levels
were equally high under as-received as well as after repair
conditions. As discussed above (in section 2.0), the average tank
temperature rise from an initial ambient of 95°F is 28°F. The tank
temperature rises for these two vehicles, however, were more than
40°F. The cumulative temperature rises from the six bags were:
12, 22, 33, 40, 44, 47°F for the Camaro, and 9, 26, 32, 37, 41,
43°F for the Sunbird. The following table shows their running loss
emissions in total grams (the sum of 6 bags) tested at 9.0 psi/95°F
from LA-4 driving cycle:
Vehicle Total Emissions
Number Status (Grams)
1532 as-received (failed purge) 315.5
after reconnect purge line 330.8
after replace gas cap 319.9
1578 as-received (failed pressure) 263.2
after repair 325.5
Note that these repair/maintenance steps of reconnecting purge
line and replacing gas cap usually will reduce a major portion of
the evaporative running loss emissions for most vehicles. The
above table has illustrated that for these two vehicles, however,
the high levels of emissions may not be caused by either pressure
or purge problems in their evaporative emissions control system.
The repair work failed to decrease their emissions, even increased
the total grams slightly. Thus, their extremely high tank
temperature rises could be the main reason why these two vehicles
had such high running loss emissions. For these reasons, it was
decided that these two vehicles be treated as exceptions. Their
average emissions from as-received and after repair are to be
included in both the pass and failed vehicle categories.
-52-
-------
Market shares of these two engine families for model years
1983 through 1990 were obtained from EPA's Certification files.
The "Camaros" included all models of Camaro and Firebird with
either 5.0L or 5.7L engines. Their sales ranged from a low 0.46 to
a high 2.02 percent, with an average of 1.06 percent. The
"Sunbirds" were models under the names of Sunbird, Skyhawk, and
Firenza with either 1.8L or 2.0L engines. Their sales ranged from
0.48 to 2.01 percent, with an average of 1.28 percent. An
alternative was to derive an estimate based on the sample. As
there were a total of 60 LDGVs in the MOBILE4.1 running loss
emissions test program sample, weighting factors of 0.01667 each
are used to represent the Camaros and the Sunbirds in the fleet
(i.e., a weighting factor of 0.96666 is used for the "regular"
vehicles).
Table 2 summarizes the calculated TVP values for each bag
under each fuel volatility level and each initial ambient
temperature for these two exception vehicles. Table 3 shows the
running loss emissions (in cumulative grams) of these two vehicles
to be used for failed vehicles. Table 6 shows the running loss
emissions (in grams per mile) of these two vehicles to be used for
pass vehicles.
-53-
-------
3.2 Pass Vehicles
Among the "regular" vehicles, those that passed both pressure
and purge checks are categorized as the "pass" vehicles. For
example, there are 32 model year 1981+ pass LDGVs tested at
9.0 psi/95°F for the LA-4 driving cycle.
The running loss emissions for pass vehicles are expressed in
unit of grams per mile (g/mi) for each bag (bags 1 through 6) under
each driving cycle (NYCC, LA-4, and HFET). Note that there are
only five bag testing results from the HFET driving cycle. As the
emission levels from five bags under the combinations of 95°F and
fuel volatilities of 9.0 and 11.7 psi RVP are relatively stable,
the emissions from the sixth bag of the HFET driving cycle are
assumed to be the same as the emissions from the fifth bag.
Therefore, there are a total of 18 (6 bags * 3 speeds) emission
values for the combination of four fuel volatility levels (7.0,
9.0, 10.4, 11.7 psi RVP) and four ambient temperature profiles (80,
87, 95, 105°F).
The g/mi running loss emissions are derived by dividing the
total grams by the cumulative trip distances in miles. For
example, for bags 1 and 2:
RLsi,g/m\ — RLei , g / DBi ,m i
RLe z ,gxm i
= (RLB1)g + RLB2,g) / (DBi,mi + DBZ,mi)
where:
D = trip distance in miles.
The average g/mi running loss emissions for each bag under each
driving cycle for 1981+ "regular" pass LDGVs and 1981+ pass LDGTs
are given in Tables 7 and 9. Note that the sample sizes for the
combinations of fuel volatility and ambient temperature vary, due
to the limitation of the test data discussed previously in sections
1.0 and 3.0. When there is no actual test data available, the
sample size is denoted as zero ("0"), and the emission values are
estimated by linear interpolation/extrapolation. Table 8
summarizes the g/mi emissions used for MOBILE4.1 model 1981+ pass
LDGVs, including the "regular" pass vehicles and two exception
vehicles.
-54-
-------
3.3 Failed Vehicles
Due to the limited sample sizes, there is no differentiation
made between vehicles that failed pressure test and vehicles that
failed purge test. The criteria used to define vehicles that
failed either pressure or purge test are the same as those used for
hot soak/diurnal emissions model.
The running loss emissions for failed vehicles in the
MOBILE4.1 model are expressed in terms of cumulative grams by bag
for the combinations of fuel volatilities and ambient
temperatures. Based on their similar emission levels in cumulative
grams, results from the NYC and LA-4 cycles were combined. The
emissions are also used for HFET cycle. The emissions in
cumulative grams are to be converted onto grams per mile unit (and
adjusted for speed) in the MOBILE4.1 model, as to be discussed
later in section 4.0. Emissions for failed LDGVs are used also for
LDGTs, due to a lack of LDGT data. Further, failed emission levels
are also used for vehicles made during the pre-control era.
Table 4 summarizes the running loss emissions used for
"regular" failed vehicles. The actual test data by each bag were
used to fit equations as a function of True Vapor Pressure (as
discussed in section 2.0). The predicted emissions in cumulative
grams are used in MOBILE4.1 model for the failed emission levels.
Table 5 shows failed emissions used for LDGVs and LDGTs, including
the "regular" failed vehicles and two exception vehicles.
-55-
-------
4.0 Driving Characteristics
In MOBILE4, a VMT weighted urban type driving characteristics
was built into the running loss emissions model so that the
emission rates were independent of the vehicle speed. This set of
urban VMT weighting factors was derived from tabulated frequencies
by trip distance in miles and average speed in mph matching the
three driving cycle characteristics used in the running loss
emissions test program. The data sources were the 1979 GM-NPD
Survey Data, and an Operational Characteristics Study sponsored by
EPA in Columbus, Ohio.
In MOBILE4.1, however, a VMT weighted driving characteristics
in terms of trip duration alone is used in the model so that the
emission rates can be adjusted by any given vehicle speed. The
1979 GM-NPD Survey Data again are used. The frequency table was
based on trip durations in 10-minute intervals, and the deriving
VMT weighting factors are:
Trip Duration Corresponding Percent
in minutes Bag # VMT
10 and less 1 33.227
11-20 2 32.883
21-30 3 14.871
31-40 4 7.886
41-50 5 3.645
52 and up 6 7.488
Total: 100.0
This new set of percent VMT weighting factors is used to represent
each bag of the three driving cycles.
4.1 Pass Vehicles
For pass vehicles in the MOBILE4.1 model, the running loss
emissions are given for all six bags of the three driving cycles in
grams per mile unit, for each of the four fuel volatilities and
four ambient temperatures. The speed adjustment algorithm for pass
vehicles are described in the following.
At any given average vehicle speed between 7.1 and 47.9 mph,
the emissions are calculated by linear interpolation between the
emission rates at the two adjacent default vehicle speeds. For
example, emissions at 10.0 mph are to be calculated based on
emissions at 7.1 mph (NYCC) and emissions at 19.6 mph (LA-4), and
emissions at 30.0 mph are based on emissions at 19.6 mph (LA-4) and
emissions at 47.9 mph (HFET):
-56-
-------
ADJ = (SPD - SPD1) / (SPD2 - SPD1) (4.1)
where: ---- ----
ADJ = calculated speed adjustment factor,
SPD = user input speed in mph, which is between 7.0 and
47.9 mph,
SPD1 = average speed in mph for the default driving
cycle 1 (either NYCC or LA-4),
SPD2 = average speed in mph for the default driving
cycle 2 (either LA-4 or HFET) .
Then, with the calculated speed adjustment factor, the g/mi running
loss emissions are estimated by subtracting the resting loss
emissions, and weighted by the VMT weighting factors:
RESTL8/mi - RESTLg/hr / (6.0 * SPDm , , h P ) (4.2)
RL B i , S P D
where:
i.spDi + ADJ * (RLBi,SpD3 - RLBi,spDi)
- RESTLg/mi] * VMTBi for i=l,2,...,6 (4.3)
RESTL = resting loss emissions,
RLBi,spD = calculated g/mi bag i running loss emissions at
speed SPD,
RLBi,spDi = g/mi bag i running loss emissions for the default
driving cycle 1 (either NYCC or LA-4),
RLBi,spD2 = g/mi bag i running loss emissions for the default
driving cycle 2 (either LA-4 or HFET),
VMTB i = bag i VMT weighting factor.
For any average speed below 7.1 mph but greater than 2.5 mph,
the emissions are calculated from the emissions at 7.1 mph and an
adjustment factor calculated from the ratio of the two speeds (the
input speed divided by 7.1 mph). This same algorithm is applicable
to where the speed is above 47.9 mph but less than 65.0 mph.
Therefore, eguations (4.1) and (4.3) above are now reformulated as:
ADJ - SPD / SPD1
RL Si , S P D
= [RLBi,spDi + ADJ * RLBi,spDi - RESTLg/mi]
* VMTB i ~ for i = l,2, ... ,6
where:
SPD = user input speed in mph, which is either between 2.5
and 7.1 mph, or between 47.9 and 65.0 mph,
-57-
-------
SPD1 = average speed of 7.1 mph from NYC driving cycle if SPD
is less than --2..JD mph, or average speed of 47.9 mph from
HFET driving cycle if SPD is greater than 47.9 mph,
RLflt.spDi = g/mi bag i running loss emissions from NYC
driving cycle if SPD is below 7.1 mph, or those
from HFET driving cycle if SPD is above 47.9 mph.
Finally, the running loss emissions are the total of the g/mi
emissions over the six bags:
RL = RL B 1 , S P D + RL B Z , S P D + RL B 3 , S P D + RL B 4 , S P D
+ RLBS,SPD '+ RLB6,SPD (4.4)
4.2 Failed Vehicles
The running loss emissions for failed vehicles in the
MOBILE4.1 model are expressed in cumulative grams. For each
combination of fuel volatility and ambient temperature, there are
only six emission values, each representing the cumulative bag
emissions for all three driving cycles. The speed adjustment
algorithm for failed vehicles are based on the following two
assumptions:
1) the durations for each bag of each driving cycle are
approximately the same, (i.e., about 10 minutes),
2) the cumulative emissions in grams by bag under each
driving cycle from failed vehicles are the same.
For failed vehicles, the speed and VMT adjustment algorithm
includes converting the emissions from cumulative grams onto grams
per mile unit by the given speed, as well as subtracting off the
resting loss emissions and multiplying VMT weighting factor:
RESTLg/io min= RESTLg/nr / 6.0io mln/hr
RL8i,g/mi = [(RLBi,g - RESTL) * VMTBl] / DB1,mi
= (RLBi,g - RESTL) * VMTB ! * 6 / SPD
RLB2>g/mi = [(RLB2>g - RESTL) * VMTB2]
/ (DB 1 ,mi + DB2 ,mi )
= (RLB2>g - RESTL) * VMTB2 / (2 * Dai.mi)
= (RLB2)g - RESTL) * VMTB2 * 6 / (2 * SPD)
then,
RLBi>g/mi = (RLBi(g - RESTL) * VMTBi * 6 / (i * SPD)
for i = 1, 2, .. . , 6
-58-
-------
and,
RL SPD = RI» B 1 , g / m 1 + RL B 2 , g X m 1 + RL B 3
•f RLB4,gXmi + RLflS.gXmi + RLB6,gXmi
where:
RESTL - resting loss emissions,
RLBi,gxini = calculated g/mi bag i running loss emissions at
speed SPD,
i,g = bag i cumulative grams running loss emissions,
i = bag i VMT weighting factor,
DBi,mi = trip distance in miles for bag i, and,
SPD = user input speed in mph.
-59-
-------
5.0 Other Model Years and Other Vehicle Types
The running loss emissions data described in Tables 5, 7, and
9 are used in MOBILE4.1 for other model year groups and other
vehicle types according to the following mapping:
Data
1981+ Pass LDGV
1981+ Pass LDGT1
Pass/
Fail
Pass
Pass
Pass
Pass
Pass
Pass
Vehicle Type
LDGV
1972-77
1978-80
1981+
LDGT1
LDGT2
HDGV
1985+
1972-77
1978-80 1979-80
1981+ 1981+
1981+ Failed LDGV All Pre-1972 Pre-1972 Pre-1979 Pre-1985
Failed 1972+ 1972+ 1979+ 1985+
-60-
-------
Fuel Initial
RVP Ambient
in psi Temp °F
7.0
9.0
10.4
11.7
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Table 1
True Vapor Pressure
Regular Vehicles
True Vapor Pressure
Bag 1
5.58
6.42
7.50
9.04
7.28
8.32
9.64
11.54
8.63
9.82
11.34
13.50
9.76
11.07
12.74
15.11
Bag 2
6.67
7.64
8.88
10.65
8.63
9.82
11.34
13.50
10.19
11.55
13.27
15.72
11.47
12.96
14.86
17.54
Bag 3
7.21
8.24
9.56
11.44
9.30
10.56
12.17
14.45
10.95
12.39
14.21
16.80
12.31
13.89
15.89
18.71
Bag 4
7.94
9.04
10.46
12.48
10.19
11.54
13.27
15.71
11.96
13.50
15.46
18.22
13.42
15.11
17.26
20.27
Bag 5
8.24
9.38
10.84
12.92
10.56
11.95
13.73
16.24
12.39
13.97
15.98
18.82
13.89
15.63
17.83
20.92
Bag 6
8.72
9.91
11.44
13.60
11.14
12.60
14.45
17.06
13.05
14.70
16.80
19.74
14.61
16.43
18.71
21.93
-61-
-------
Table 2
True Vapor Pressure
Exception Vehicles
Fuel Initial
RVP Ambient
in psi Temp °F
7.0
9.0
10.4
11.7
7.0
9.0
10.4
11.7
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
True Vapor Pressure
Bag 1
6.42
7.35
8.55
10.28
8.32
9.47
10.94
13.04
9.82
11.14
12.82
15.21
11.07
12.52
14.37
16.97
6.05
6.94
8.09
9.73
7.86
8.96
10.37
12.38
9.30
10.56
12.17
14 .46
10.49
11.88
13.65
16.16
Bag 2
Camaro
7.79
8.88
10.28
12.27
10.00
11.34
13.04
15.45
11.75
13.27
15.21
17.93
13.19
14.86
16.97
19.95
Sunbird
8.40
9.56
11.04
13.14
10.75
12.17
13.96
16.51
12.60
14.21
16.25
19.12
14. 13
15.89
18.12
21.25
Bag 3
9.56
10.84
12.48
14.80
12.17
13.73
15.71
18.50
14.21
15.98
18.22
21.37
15.89
17.83
20.27
23.69
9.38
10.65
12.27
14.56
11.95
13.50
15.45
18.21
13.97
15.72
17.93
21.03
15.63
17.54
19.95
23.33
Bag 4
10.84
12.27
14.07
16.63
13.73
15.45
17.62
20.68
15.98
17.93
20.38
23.81
17.83
19.95
22.62
26.35
10.28
11.64
13.37
15.82
13.04
14.69
16.78
19.72
15.21
17.07
19 .43
22.74
16.97
19.02-
21.59
25.18
Bag 5
11.64
13.14
15.05
17.75
14.69
16.51
18.80
22.02
17.07
19.12
21.70
25.31
19.02
21.25
24.06
27.97
11.04
12.48
14.31
16.90
13.96
15.71
17.91
21.01
16.25
18.22
20.71
24.18
18.12
20.27
22.97
26.75
Bag 6
12.27
13.84
15.82
18.63
15.45
17.34
19.72
23.07
17.93
20.06
22.74
26.48
19.95
22.27
25.18
29.24
11.44
12.92
14.80
17.46
14.45
16.24
18.50
21.68
16.80
18.82
21.37
24.93
18.71
20.92
23.69
27.56
-62-
-------
Table 3
Cumulative Grams Running Loss Emissions
Exception Vehicles
Fuel
RVP
in psi
7.0
9.0
10.4
11.7
7.0
9.0
10.4
11.7
Ambient
Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Running Loss Emissions (Cumulative Grams)
Bag 1
Bag 2
Bag 3
Bag 4
Bag 5
Bag 6
Camaro
1.83
1.96
2.15
2.47
2.11
2.31
2.61
3.11
2.38
2.65
3.05
3.71
2.64
2.98
3.47
4.28
0.71
1.40
2.42
4.10
1.52
2.97
5.10
8.66
3.45
5-41
8.26
13.01
5.29
7.72
11.24
17.07
2.27
3.99
7.00
12.07
4.12
8.62
15.16
25.98
10.12
16.13
24.80
39.04
15.78
23.17
33.78
51.11
Sunbi
2.00
4.66
8.48
22.72
6.73
15.79
28.93
50.59
18.83
30.89
48.24
76.64
30.19
44.99
66.18
100.64
3.83
7.98
16.62
30.78
5.07
19.69
40.76
75.27
24.59
43.92
71.58
116.55
42.82
66.43
100.05
154.42
rd
4.24
5.43
12.92
34.93
17.02
21.95
51.91
139.24
26.83
58.18
128.15
283.75
55.95
113.31
221.83
441.95
8.06
11.20
32.01
82.27
33.30
45.99
130.51
378.61
59.71
148.49
347.42
789.42
142.28
305.45
614.02
818.00*
6.68
17.12
43.89
87.78
18.73
48.02
123.67
247.17
65.67
135.05
234.04
394.43
131.16
215.68
335.71
529.16
12.55
16.73
47.79
120.61
50.25
67.15
192.70
565.10
87.34
219.75
518.46
818.00*
210.54
455.53
818.00*
818.00*
10.05
25.76
66.05
132.10
28.42
72.88
185.66
369.14
99.29
202 .69
349 .77
587.29
196.95
322.60
500.51
719.06*
36.78
49.04
122.60
240.39
100.62
134.16
336.79
664.77
181.88
367.54
630.51
818.00*
357.42
582.20
818.00*
818.00*
16.20
41.53
94.39
178.09
41.96
107.60
241.92
460.06
139.12
262.23
437.11
719.06
255.44
404.86
616.16
719.06*
*An upper limit based on actual test results
-63-
-------
7.0
9.0
10.4
11.7
Table 4
Cumulative Grams Running Loss Emissions
Failed Vehicles
Fuel
RVP Ambient
in psi Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Running Loss Emissions (Cumulative Grams)
Bag 1
0.50
1.02
1.52
3.25
1.23
2.52
3.76
4.79
2.85
3.90
4.72
4.90
3.85
4.62
4.97
6.66
Bag 2
1.44
2.48
7.27
14.13
6.32
10.92
16.79
25.13
12.34
17.59
24.28
33.74
17.30
23.08
30.42
40.78
Bag 3
3.26
6.80
11.34
26.24
9.28
19.29
32.04
50.13
22.37
33.79
48.28
68.78
33.15
45.69
61.58
83.99
Bag 4
4.33
10.07
17.37
44.20
13.69
31.71
54.63
87.09
37.28
57.80
83.80
120.49
56.67
79.17
107.64
147.69
Bag 5
4.91
12.27
21.52
58.84
16.41
41.49
73.37
118.47
49.25
77.78
113.91
164.85
76.21
107.49
147.02
202.59
Bag 6
5.83
15.34
27.40
77.40
20.58
54.20
96.87
157.15
64.60
102.78
151.10
219.11
100.70
142.53
195.33
269.48
-64-
-------
7.0
9.0
10.4
11.7
Table 5
Cumulative Grams Running Loss Emissions
1981+ Failed LDGVs and LDGTs
Fuel
RVP Ambient
in psi Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Running Loss Emissions (Cumulative Grams)
Bag 1
0.53
1.04
1.55
3.25
1.25
2.52
3.76
4.83
2.85
3.90
4.75
5.02
3.85
4.64
5.05
6.79
Bag 2
1.46
2.54
7.29
14.24
6.29
10.96
16.97
25.57
12.41
17.79
24.69
34.54
17.49
23.45
31.07
41.95
Bag 3
3.29
6.80
11.45
26.46
9.34
19.34
32.52
52.04
22.48
34.37
50.00
73.16
33.69
47.16
64.89
91.13
Bag 4
4.43
10.21
18.06
45.56
14.10
32.22
57.05
94.62
38.13
60.60
90.70
136.21
59.34
85.22
119.88
165.22
Bag 5
5.12
12.57
22.70
61.09
17.17
42.44
77 .23
130.10
50.72
82.23
124.59
182.78
80.46
116.88
164.10
222.58
Bag 6
6.52
16.34
30.10
81.80
22.27
56.42
103.29
170.66
67.80
109.85
163.86
237.42
107.56
154.23
212.73
287.77
-65-
-------
Table 6
7.0
9.0
10.4
11.7
7.0
9.0
10.4
11.7
Grams per Mile Running Loss Emissions
Exception Vehicle: Camaro
Fuel
RVP Ambient
in psi Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Running Loss Emissions (Grams/Mile)
Bag 1
New
1.55
1.66
1.82
2.09
1.79
1.96
2.21
2.64
2.02
2.25
2.58
3.14
2.24
2.53
2.94
3.63
Bag 2
Bag 3
Bag 4
Bag 5
Bag 6
York City Cycle
0.96
1.69
2.97
5.11
1.75
3.65
6.42
11.01
4.29
6.83
10.51
16.54
6.69
9.82
14.31
21.66
LA-4 Driving
0.51
0.55
0.60
0.69
0.59
0.64
0.73
0.87
0.66
0.74
0.85
1.03
0.74
0.83
0.97
1.19
0.30
0.53
0.93
1.61
0.55
1.15
2.02
3.46
1.35
2.15
3.31
5.21
2.10
3.09
4.50
6.81
1.08
2.25
4.69
8.69
1.43
5.56
11.51
21.26
6.95
12.41
20.22
32.92
12.10
18.77
28.26
43.62
Cycle
0.35
0.72
1.50
2.78
0.46
1.78
3.68
6.79
2.22
3.96
6.45
10.51
3.86
5.99
9 .02
13.92
1.71
2.37
6.78
17.43
7.06
9.74
27.65
80.21
12.65
31.46
73.61
167.25
30.14
64.71
130.09
173.31
0.54
0.75
2.13
5.48
2.22
3.07
8.70
25.24
3.98
9.90
23.16
52.63
9.49
20.63
40.93
54.53
2.13
2.84
8.10
20.44
8.52
11.38
32.66
95.78
14.80
37.25
87.87
138.64*
35.68
77.21
138.64*
138.64*
0.68
0.90
2.57
6.49
2.70
3.61
10.37
30.40
4.70
11.82
27.89
44 .00*
11.33
24.50
44 .00*
44 .00*
5.19
6.93
17.32
33.95
14.21
18.95
47.57
93.89
25.69
51.91
89.06
115.54*
50.48
82.23
115.54*
115.54*
1.63
2.18
5.45
10.68
4.47
5.96
14.97
29.55
8.08
16.34
28.02
36.36*
15.89
25.88
36.36*
36.36*
-66-
-------
Table 6 (Continued)
Grams per Mile Running Loss Emissions
Exception Vehicle: Sunbird
Fuel
RVP
in psi
7.0
9.0
10.4
11.7
7.0
9.0
10.4
11.7
Ambient
Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Running Loss Emissions (Grams/Mile)
Bag 1
New
0.60
1.19
2.05
3.47
1.29
2.52
4.32
7.34
2.92
4.58
7.00
11.03
4.48
6.54
9.53
14.47
Bag 2
Bag 3
Bag 4
Bag 5
Bag 6
York City Cycle
0.85
1.97
3.59
9.63
2.85
6.69
12.26
21.44
7.98
13.09
20.44
32.47
12.79
19.06
28.04
1.20
1.53
3.65
9.87
4.81
6.20
14.66
39.33
7.58
16.44
36.20
80.16
15.81
32.01
62.66
42.64 124.84
LA-4 Driving
0.20
0.39
0.67
1.14
0.42
0.83
1.42
2.41
0.96
1.51
2.30
3.62
1.47
2.15
3.13
4.75
0.27
0.62
1.13
3.03
0.90
2.11
3.86
6.75
2.51
4.12
6.43
10.22
4.03
6.00
8.82
13.42
Cycle
0.38
0.49
1.17
3.15
1.53
1.98
4.68
12.56
2.42
5.25
11.56
25.59
5.05
10.22
20.00
39.85
1.42
3.63
9.30
18.60
3.97
10.17
26.20
52.37
13.91
28.61
49.58
83.57
27.79
45.69
71.13
112.11
0.45
1.14
2.93
5.85
1.25
3.20
8.24
16.48
4.38
9.00
15.60
26.30
8.74
14.38
22.38
35.28
1.70
4.37
11.19
22.39
4.82
12.35
31.47
62.57
16.83
34.35
59.28
99.54
33.38
54.68
84.83
133.28
0.54
1.39
3.55
7.11
1.53
3.92
9.99
19.86
5.34
10.90
18.81
31.59
10.59
17.35
26.92
42.30
2.29
5.87
13.33
25.15
5.93
15.20
34.17
64.98
19.65
37.04
61.74
101.56*
36.08
57.18
87.03
101.56*
0.72
1.85
4.20
7.92
1.86
4.78
10.75
20.45
6.18
11.65
19.43
31.96*
11.35
17.99
27.38
31.96*
-67-
-------
Table 7
Grams per Mile Running Loss Emissions
Regular Pass Vehicles
Fuel
RVP
psi
7.0
9.0
10.4
11.7
7.0
9.0
10.4
11.7
Ambient
Temp °F
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
80.0
87.0
95.0
105.0
Sample
Size
0
0
12
12
0
0
8
14
0
0
0
0
0
0
0
0
0
0
12
12
0
0
32
14
4
0
0
0
0
0
0
0
Running Loss Emissions (Grams/Mile)
Bag 1
New York
0.28
0.28
0.28
0.29
0.28
0.29
0.33
0.67
0.29
0.36
0.63
1.37
0.35
0.59
1.10
1.95
Bag 2
City
0.30
0.30
0.30
0.31
0.30
0.31
0.35
0.91
0.31
0.40
0.85
2.25
0.38
0.77
1.73
3.35
Bag 3
Cycle
0.31
0.31
0.31
0.32
0.31
0.32
0.37
1.72
0.32
0.50
1.58
6.09
0.45
1.39
4.40
9.65
Bag 4
0.31
0.31
0.31
0.33
0.31
0.32
0.38
2.80
0.32
0.61
2.55
13.71
0.53
2.20
9.53
22.60
Bag 5
0.30
0.30
0.30
0.32
0.30
0.31
0.53
3.90
0.31
0.85
3.55
22.93
0.74
3.08
15.63
38.43
Bag 6
0.30
0.30
0.30
0.32
0.30
0.31
0.63
4.87
0.31
1.04
4.45
29.25
0.73
3.85
19.88
49.17
LA-4 Driving Cycle
0.07
0.07
0.07
0.09
0.07
0.08
0.12
0.12
0.08
0.12
0.12
0.25
0.12
0.12
0.20
0.36
0.09
0.09
0.09
0.11
0.09
0.10
0.17
0.22
0.10
0.17
0.22
0.42
0.17
0.21
0.34
0.58
0.09
0.09
0.09
0.18
0.09
0.10
0.18
0.29
0.10
0.19
0.28
1.32
0.19
0.27
0.92
2. 16
0.09
0.09
0.09
0.11
0.09
0.10
0.21
0.34
0.10
0.22
0.33
4.23
0.22
0.31
2.74
7.40
0.09
0.09
0.09
0.11
0.09
0.10
0.23
0.65
0.10
0.27
0.61
6.28
0.26
0.55
4.12
10.86
0.09
0.09
0.09
0.11
0.09
0.10
0.30
1.00
0.10
0.36
0.92
7.58
0.34
0.82
5.05
12.96
-68-
-------
Table 7 (Continued)
Grams per Mile Running Loss Emissions
Regular Pass Vehicles
Fuel
RVP
psi
7.0
9.0
10.4
11.7
Ambient
Temp
80.
87.
95.
105.
80.
87.
95.
105.
80.
87.
95.
105.
80.
87.
95.
105.
op
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Sample
Size
0
0
0
0
0
0
7
0
0
0
0
0
0
0
7
0
Bag
Highway
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Running Loss
1
Fuel
02
02
02
06
02
02
02
09
02
02
08
09
02
05
09
09
Bag 2
Bag
Emissions
3
Bag 4
(Grams /Mile)
Bag 5
Bag 6
Economy Test
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.05
0.05
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
02
02
02
02
02
02
02
02
02
02
02
03
02
03
04
04
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
-69-
-------
Table 8
Grams per Mile Running Loss Emissions
1981+ Pass LDGVs
Fuel
RVP Ambient Running Loss Emissions (Grams/Mile)
in psi Temp °F
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
Bag 1
New
0.31
0.32
0.34
0.37
0.32
0.35
0.43
0.81
0.36
0.46
0.77
1.56
0.45
0.72
1.27
2.18
Bag 2
York City
0.32
0.35
0.40
0.55
0.37
0.47
0.65
1.42
0.50
0.72
1.34
2.99
0.70
1.23
2.38
Bag 3
Cycle
0.34
0.36
0.44
0.62
0.40
0.50
0.79
2.67
0.55
0.96
2.47
7.78
0.90
2.19
5.77
4.31 12.14
LA-4 Driving
0.08
0.08
0.09
0.12
0.08
0.10
0.15
0.17
0.11
0.15
0.17
0.32
0.15
0.17
0.26
0.44
0.10
0.11
0.12
0.18
0.11
0.15
0.26
0.38
0.17
0.27
0.37
0.66
0.27
0.35
0.55
0.90
Cycle
0.10
0.11
0.13
0.21
0.12
0.16
0.31
0.60
0.18
0.34
0.57
1.88
0.33
0.53
1.37
2.98
Bag 4
0.35
0.40
0.57
0.92
0.48
0.64
1.27
4.92
0.76
1.59
4.52
17.42
1.48
3.97
12.57
26.60
0.10
0.12
0.17
0.30
0.14
0.20
0.49
1.02
0 .24
0.53
0.96
5.40
0.51
0.88
3.70
8.65
Bag 5
0.35
0.41
0.61
1.02
0.51
0.70
1.58
6.41
0.83
2.02
5.89
26.14
1.87
5.18
18.83
41.68
0.11
0.13
0.19
0.33
0.16
0.22
0.56
1.47
0.27
0.64
1.37
7.33
0.61
1.23
5.17
11.94
Bag 6
0.41
0.50
0.80
1.29
0.63
0.87
1.97
7.36
1.06
2.49
6.82
31.89
2.15
6.05
22.59
55.39
0.13
0.15
0.25
0.42
0.19
0.28
0.72
1.80
0.34
0.82
1.68
8.47
0.79
1.53
5.94
13.74
-70-
-------
Table 8 (Continued)
Grams per Mile Running Loss Emissions
1981+ Pass LDGVs
Fuel
RVP Ambient Running Loss Emissions (Grams/Mile)
in psi Temp °F
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
Bag 1
Highway
0.02
0.02
0.02
0.06
0.02
0.02
0.02
0.09
0.02
0.02
0.08
0.09
0.02
0.05
0.09
0.09
Bag 2
Fuel
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.05
0.05
Bag 3
Economy
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.04
0.04
Bag 4
Test
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
Bag 5
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
Bag 6
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
-71-
-------
Table 9
Grams per Mile Running Loss Emissions
1981+ Pass LDGTs
Fuel
RVP Ambient Running Loss Emissions (Grams/Mile)
in psi Temp °F
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
Bag 1
New
0.45
0.49
0.55
0.37
0.61
0.60
0.56
0.81
0.62
0.71
0.80
1.56
0.66
0.77
0.89
2.18
Bag 2
York City
0.41
0.45
0.50
0.55
0.61
0.54
0.46
1.42
0.56
0.63
0.71
2.99
0.57
0.74
0.90
4.31
LA-4 Driving
0.12
0.13
0.09
0.23
0.12
0.10
0.10
0.25
0.15
0.16
0.18
0.18
0.17
0.16
0.16
0.23
0.14
0.14
0.12
0.25
0.12
0.10
0.10
0.18
0.15
0.15
0.16
0.16
0.12
0.16
0.16
0.16
Bag 3
Cycle
0.34
0.41
0.48
0.62
0.77
0.62
0.53
2.67
0.66
0.86
1.14
7.78
0.72
1.23
1.74
12.14
Cycle
0.11
0.11
0.13
0.22
0.11
0.09
0.09
0.15
0.13
0.14
0.16
0.16
0.10
0.19
0.23
0.21
Bag 4
0.29
0.37
0.45
0.92
0.85
0.67
0.61
4.92
0.74
1.07
1.57
17.42
0.82
1.96
3.09
26.60
0.07
0.08
0.17
0.23
0.12
0.09
0.09
0.28
0.14
0.18
0.24
0.26
0.09
0.41
0.53
0.58
Bag 5
0.29
0.39
0.43
1.02
0.99
0.78
0.71
6.41
0.88
1.34
2.08
26.14
1.11
2.78
4.40
41.68
0.06
0.07
0.19
0.26
0.11
0.09
0.09
0.35
0.14
0.20
0.28
0.31
0.08
0.57
0.74
0.91
Bag 6
0.32
0.44
0.42
1.29
1.21
0.93
0.78
7.36
1.05
1.66
2.67
31.89
1.66
3.80
5.93
55.39
0.05
0.07
0.25
0.32
0.11
0.11
0.11
0.51
0.16
0.24
0.39
0.43
0.11
0.95
1.07
1.72
-72-
-------
Table 9 (Continued)
Grams per Mile Running Loss Emissions
1981+ Pass LDGTs
Fuel
RVP Ambient Running Loss Emissions (Grams/Mile)
in psi Temp °F
7.0 80.0
87.0
95.0
105.0
9.0 80.0
87.0
95.0
105.0
10.4 80.0
87.0
95.0
105.0
11.7 80.0
87.0
95.0
105.0
Bag 1
Highway
0.02
0.02
0.02
0.06
0.02
0.02
0.02
0.09
0.02
0.02
0.08
0.09
0.02
0.05
0.09
0.09
Bag 2
Fuel
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.05
0.05
Bag 3
Economy
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.04
0.04
Bag 4
Test
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
Bag 5
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
Bag 6
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.03
0.02
0.03
0.03
0.03
-73-
-------
MOBILE4.1
Technology Distribution for Gasoline Light-Duty Trucks
Model
lea
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
Model
feK
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
Model
Year
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
CLLP
3way
Curb
0
57339
74447
109748
255529
167151
260447
260231
174645
7985
21900
CLLP
Sway
Cafe
0.0%
3.9%
3.8%
3.7%
82%
4.6%
7.4%
66%
37%
02%
0.4%
CLLP
Carb
1.8%
3.9%
13.1%
25.0%
25.8%
13.1%
12.8%
92%
7.7%
20%
1.5%
CLLP
Sway
MPFI
0
0
3438
65296
206437
861000
716266
1099548
1544412
1948880
2177487
CLLP
Sway
MEQ
0.0%
0.0%
02%
22%
6.6%
23.8%
20.4%
280%
33.1%
41.3%
43.5%
CLLP
MPFI
0.0%
0.0%
0.2%
22%
6.6%
23.8%
32.7%
41.6%
54.0%
60.1%
57.7%
CLLP
Sway
IBI
0
0
0
0
144973
328016
780032
1455516
1435056
1533017
1815964
CLLP
Sway
IBI
0.0%
0.0%
0.0%
00%
4.7%
9.1%
22.3%
37.0%
30.7%
32.5%
362%
CLLP
IBI
0.0%
0.0%
0.0%
0.0%
4.7%
13.6%
28.0%
442%
36.9%
36.7%
40.7%
Oplp
Sway
Cab
13589
0
0
20757
15454
17780
48058
15860
10998
9998
6000
Oplp
Sway
CMfe
0.9%
0.0%
0.0%
0.7%
05%
05%
1.4%
04%
02%
02%
0.1%
Oplp
Any
982%
96.1%
86.8%
72.8%
62.8%
495%
26.4%
5.0%
1.4%
12%
0.1%
Oplp
Sway
MPFI
0
0
0
0
0
0
0
0
0
0
0
Oplp
Sway
MEQ
0.0%
0.0%
0.0%
0.0%
00%
0.0%
0.0%
0.0%
0.0%
0.0%
00%
Oplp
Sway
HI
0
0
0
0
0
0
0
38829
0
0
0
Oplp
Sway
HI
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
1.0%
0.0%
00%
0.0%
Model
Xsss
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
CLLP
OxSway
Carb
26593
906
163194
631304
546263
308664
188653
101129
185103
87597
53055
CLLP
OxSway
Cert.
1.8%
0.1%
93%
213%
17.6%
85%
5.4%
2.6%
4.0%
1.9%
1.1%
Any
Carte
100.0%
100.0%
99.8%
97.8%
88.7%
62.7%
39.3%
132%
9.1%
32%
1.6%
CLLP
OxSway
MPFI
0
0
0
0
0
0
429429
534243
974850
884251
712782
CLLP
OxSway
MPFI
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
123%
13.6%
20.9%
18.7%
142%
Any
MPFI
0.0%
0.0%
02%
22%
6.6%
23.8%
32.7%
416%
54.0%
60.1%
57.7%
CLLP
OxSway
HI
0
0
0
0
368
163878
202385
283427
288728
200673
223736
CLLP
OxSway
HI
0.0%
0.0%
0.0%
0.0%
0.0%
45%
5.8%
72%
62%
43%
4.5%
Any
HI
0.0%
0.0%
0.0%
0.0%
4.7%
13.6%
28.0%
452%
36.9%
36.7%
40.7%
Oplp
OxSway
Carb
0
0
0
51116
66837
0
64044
69315
0
0
0
Oplp
OxSway
Carb
0.0%
0.0%
0.0%
1.7%
22%
0.0%
1.8%
1.8%
0.0%
0.0%
0.0%
Oplp
OxSway
MEQ
0
0
0
0
0
0
0
0
0
0
0
Oplp
OxSway
MEQ
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Model
feat
1981
1962
1983
1984
1985
1986
1987
1988
1989
1990
1991
Oplp
OxSway
HI
0
0
0
0
0
0
0
0
0
0
0
Oplp
OxSway
HI
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Any
Sway
0.9%
3.9%
4.0%
6.6%
20.0%
37.9%
515%
73.0%
67.8%
742%
803%
Oplp
Qxjd
1429888
1429253
1658258
2081626
1868467
1776740
814666
72566
53200
40500
0
Oplp
Oxid
943%
96.1%
84.1%
70.3%
602%
49.0%
232%
1.8%
1.1%
0.9%
0.0%
Any
OxSway
1.8%
0.1%
9.3%
23.1%
19.8%
13.0%
252%
25.1%
31.0%
24.9%
19.7%
Oplp
None
46007
0
51277
0
0
0
0
0
0
4969
0
Oplp
None
3.0%
0.0%
2.6%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.1%
0.0%
Any
fixjd
943%
96.1%
84.1%
703%
602%
49.0%
232%
1.8%
1.1%
0.9%
0.0%
AH vehicle counts taken from CAFE estimates, except 1989 and newer model years, which are from General Label predictions.
7/2/91
-------
APPENDIX D
PURGE AND PRESSURE TEST EFFECTIVENESS FIGURES AND SPREADSHEET
-------
o
a:
0.500
0.450
0.400
0.350
0.300
Sf 0.250
3
° 0.200
0.150
0.100
0.050
0.000
Purge-Pressure Detection Rates
Annual
A— W/Program
~D— No-Program
D D- D D D D D
-------
0.500
0.450
0.400
0.350
-g 0.300
ce:
gf 0.250
13
"L° 0.200
0.150
0.100
0.050 1
0.000
Purge-Pressure Detection Rates
Biennial
-A— W/Program
-D— No-Program
D
a
D—a
10
15
20
25
Age Index
-------
Revised Purge/Pressure Failure Rates for MOBILE4.1
ANNUAL
In- lane
Age No-program
Index fail rate
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
0.080
0.080
0.080
0.096
0.103
0.120
0.150
0.188
0.258
0.323
0.389
0.442
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
New
Detect
Rate
0.0800
0.0800
0.0600
0.0160
0.0070
0.0170
0.0300
0.0380
0.0700
0.0650
0.0660
0.0530
0.0250
0.0070
0.0170
0.0300
0.0380
0.0700
0.0650
0.0660
0.0530
0.0250
0.0070
0.0170
0.0300
0.0380
0.0412
Annual
After
0.0800
0.0600
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
Annual
average
0.0800
0.0600
0.0080
0.0035
0.0085
0.0150
0.0190
0.0350
0.0325
0.0330
0.0265
0.0125
0.0035
0.0085
0.0150
0.0190
0.0350
0.0325
0.0330
0.0265
0.0125
0.0035
0.0085
0.0150
0.0190
Averaqe
Annual
% benefit
0.000
0.238
0.853
0.910
0.836
0.789
0.765
0.732
0.798
0.837
0.903
0.967
0.992
0.981
0.967
0.958
0.922
0.928
0.927
0.941
0.972
0.992
0.981
0.967
0.958
0.908
BIENNIAL
No-program
Age in-lane
Index fail rate
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
0.080
0.080
0.080
0.096
0.103
0.120
0.150
0.188
0.258
0.323
0.389
0.442
0.451
0.451
0.451
0.451
0.451
0.451
-0.451
0.451
0.451
0.451
0.451
0.451
0.451
New Detect
1
0
Rate
0.0800
0.0800
0.0600
0.0160
0.0070
0.0240
0.0300
0.0680
0.0700
0.1350
0.0660
0.1330
0.0250
0.0250
0.0240
0.0240
0.0680
0.0680
0.1350
0.1350
0.1330
0.1330
0.0250
0.0250
0.0240
0.0240
0.0630
Bien.
After
0.0800
0.0600
0.0000
0.0000
0.0070
0.0000
0.0300
0.0000
0.0700
0.0000
0.0660
0.0000
0.0250
0.0000
0.0240
0.0000
0.0680
0.0000
0.1350
0.0000
0.1330
0.0000
0.0250
0.0000
0.0240
Bien.
averaqe
0.0800
0.0600
0.0080
0.0035
0.0155
0.0150
0.0490
0.0350
0.1025
0.0330
0.0995
0.0125
0.0250
0.0120
0.0240
0.0340
0.0680
0.0675
0.1350
0.0665
0.1330
0.0125
0.0250
0.0120
0.0240
Averaqe
Bien.
% benefit
0.000
0.238
0.853
0.910
0.779
0.789
0.612
0.732
0.582
0.837
0.733
0.967
0.945
0.973
0.947
0.925
0.849
0.850
0.701
0.853
0.705
0.972
0.945
0.973
0.947
0.843
-------
Revised Purge/Pressure Failure Rates for MOBILE4.1
ANNUAL
Age
Index
1
-14A5
-14A6
-14A7
-14A8
-14A9
-14A10
-14A11
-14A12
-14A13
-14A14
-14A15
-14A16
-14A17
-14A1B
-14A19
— 1 4ATI
— iTAiV
-14A21
-14A22
-14A23
-14A24
-14A25
-14A26
-14A27
-14A28
In-lane
No-program
fail rate
0.08
0.08
0.08
0.096
0.103
0.12
0.15
0.188
0.258
0.323
0.389
0.442
0.451
0.451
0.451
0.451
O»KI
• 431
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
Hew
Detect
Rate
-B5
-B5
-N6
-B8-B7
-B9-B8
-B10-B9
-B11-B10
-B12-B11
-B13-B12
-B14-B13
-B15-B144(B6-B5)
-B16-B154 (B7-B6)
-B17-B164(B8-B7)
-V9
-V10
-Vll
— tr| 9
"v±<
-VI 3
-VI 4
-V15
-VI 6
-VI 7
-VI 8
-VI 9
-V20
-V21
-AVERAGE (VS:V30)
Annual
After
-V5
-0.75*V6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Annual
average
-{W54V6)/2
-(H64V7)/2
- ** 1 f)
" \ H*. J. T V«i *. 1 / £
-(H224V23I/2
-(H234V24I/2
-(H244V25I/2
-(H254V26I/2
-(H264V27I/2
-(W274V28I/2
-(H264V29)/2
-(H294V301/2
Average
Annual
% benefit
-(5L6-X6|/((SL645L7| .21
-(SL7-X7|/((SL74$L8| '21
- (SL8-XB) / ( (SL84JL9I /2)
- (SL9-XSI / ( (SL94JLKM /Cl
- (SL10-X10) / ( (SL1045L11 ) i2
-(SL11-X11)/ ((SL1HSL12) .'2
-(SL12-X12) / ( (SL124SL13) 12
-(SL13-X13)/ (I5L134SL14I '2
-(SL14-X14)/ (1SL144SL15) /2
-(SL15-X1S)/ ((SL154SL16) '2
-(SL16-X16) / ( ($L164SL17) /2
-(5U7-X17) / ( (5L174$L1B) !2
-(SL18-X18)/ ((SL184SL1!-! i2
-(SL19-X19) / ( (SLlfuJLIfi) '2
- (5L20-X20) / ( (SL22
- (SL29-X29) / ( (SL294SL30) /2
-AVERAGE(V7:V29I
BIENNIAL
Age
Index
1
-14A59
-14A60
-14A61
-14A62
-14A63
-14A64
-14A65
-14A66
-14AS7
-14A68
-14A69
-14A70
-14A71
-1+A72
-14A73
-14A74
-14A75
-14A76
-14A77
-14A78
-14A79
-14A80
-14A81
-14A82
No-program
in-lane
fail rate
0.08
0.08
o.oa
0.096
0.103
0.12
0.15
0.168
0.258
0.323
0.389
0.442
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
0.451
New Detect
1
-1-V56
Rate
-B59
-B59
-H60
-B62-B61
-B63-B62
-V634(B64-B63I
-B65-B64
-V654(B66-B65|
-B67-B66
-V674 (B68-B67) 4 (B60-BS9)
- (B69-B68) 4 (B60-B59)
-V64(B70-B69)4(B61-B60)
-(B71-B70)4(B62-B61)
-W714(V63-H63)
-H724(V64-H64)
-H734(V65-H65)
-H744(V66-H66)
-H75+(V«7-H67)
-H764(V68-H68)
-W774 (V69-H69I
-H784 (V70-H70)
-H794(V71-H71)
-H804(V72-H72)
-H814(V73-W73)
-H824 (\TJ4-W74)
-H834(V75-N75)
-AVERAGE (V59:V84I
Bien.
After
-V59
-0 . 25*V$57«V604 (0 . 75*V«0)
-V61«V557
-H614(V$57« (V62-H61) )
-V63
-W624(V$57*(V64-H62)>
-VS5
-H644(V$57«(H824V83)/2
-(H834V84I/2
Average
Bien.
% benefit
-(SL59-X59)/(($Lf-<'4SLCii) /;
- ($L60-X60) / ( ($L6(i4$LCl) /2
-(SL61-X61)/((SL6145L62) /2
-(SL62-X62) / ( ($L624$L63) /2
- (SL63-X63) / ( ($L634$L64 ) /2
- ($L64-X64) / ( ($L644$L£5) 12
- (SL65-X65) / ( ($L654$ie£) 12
- (SL66-X66I / ( ($L664$L67) /2
- (SI.67-X67) / ( ($L674$16B) /2
- (5L68-X68) / ( ($L684$LC9) 12
- (SL69-X69) / ( ($L«94$L70) /2
- (SL70-X70) / ( (5L704SL71 ) /2
-(SL71-X71) /( (SL714$L72) /2
- (SL72-X72) / ( (SL724SL73) /2
- ($173-X73( / ( ($L734$L74) /2
- (SL74-X74) / ( ($L744$L75) /2
- ($L75-X75| / ( ($L754$L76) /2
- (5L76-X76) / ( (SL764SL77) /2
- (5L77-X77) / ( (5L774SL78) /2
- ($L78-X78| / ( ($17845179) /2
- {$L79-X79) / ( ($L794$L80) /2
- (SL80-X80) / ( (SL804SLB1 ) /2
-(SL81-X81)/((SI.814$l82)/2
- (5182-X82) / ( (SL824$183) /2
- (SLB3-X83) / ( (SL634SL84 ) /2
-AVERAGE (V61:Y83)
-------
APPENDIX E
REGRESSION ANALYSES AND SCATTER PLOTS FOR FUEL-INJECTED
1983 AND LATER VEHICLES
-------
Table of Contents
Appendix E
E-l IM240 Vehicles Tested to Date 11/14/91
PFI Vehicles
E-2 Selecting One Score From the 2500/Idle Test
E-3 Regression Analyses of FTP HC versus HC Emissions Over
IM240, Idle Test, and 2500/Idle Test for PFI Vehicles
E-4 HC Emissions - FTP vs Lane IM240
E-6 HC Emissions - FTP vs Lane 2500/Idle
E-8 HC Emissions - FTP vs Lane Idle
E-10 Regression Analyses of FTP CO versus CO Emissions
Over: IM240, Idle Test, and 2500/Idle Test for PFI
Vehicles
E-ll CO Emissions - FTP vs Lane IM240
E-13 CO Emissions - FTP vs Lane 2500/Idle
E-15 CO Emissions - FTP vs Lane Idle
E-17 Regression Analyses of FTP NOx versus IM240 NOx
Emissions
E-18 NOx Emissions - FTP vs Lane IM240
E-20 Vehicles Tested by ATL at Southbend, Indiana
E-21 HC Emissions - FTP vs Indolene-IM240
E-23 CO Emissions - FTP vs Indolene-IM240
E-25 NOx Emissions - FTP vs Indolene-IM240
E-27 EF Vehicles Tested at EPA's MVEL
E-28 HC Emissions - FTP vs Indolene-IM240
E-30 CO Emissions - FTP vs Indolene-IM240
E-32 NOx Emissions - FTP vs Indolene-IM240
E-34 Description of Repair Regressions Analysis
-------
E-35 Statistical Summaries for Variables Used in PFI Before
and After Regression Analyses
E-36 Regression Analyses for the Change in Emission
Constituents Before and After Repairs of the PFI
Vehicles for the FTP vs IM240-A
E-37 EC Emissions - AFTP vs AIM240-A
E-39 CO Emissions - AFTP vs Alndolene-IM240
E-41 NOx Emissions - AFTP vs AIM240-A
E-43 Fuel Economy - AFTP vs AIM240-A
TBI Vehicles
E-45 Selecting One Score From the 2500/Idle Tests
E-46 Regression Analyses of FTP HC versus HC Emission Over:
IM240f Idle Test, and 2500/Idle Test for TBI Vehicles
E-47 HC Emissions - FTP vs Lane IM240
E-49 HC Emissions - FTP vs Lane 2500/Idle
E-51 HC Emissions - FTP vs Lane Idle
E-53 Regression Analyses of FTP CO versus CO Emissions
Over: IM240, Idle Test, and 2500/Idle Test for TBI
Vehicles
E-54 CO Emissions - FTP vs Lane IM240
E-56 CO Emissions - FTP vs Lane 2500/Idle
E-58 CO Emissions - FTP vs Lane Idle
E-60 Regression Analyses of FTP NOx versus IM240 NOx
Emissions
E-61 NOx Emissions - FTP vs Lane IM240
E-63 Vehicles Tested by ATL at Southbend, Indiana
E-64 HC Emissions - FTP vs Indolene-IM240
E-66 CO Emissions - FTP vs Indolene-IM240
E-68 NOx Emissions - FTP vs Indolene-IM240
E-70 EF Vehicles Tested at EPA's MVEL
11
-------
E-77 Description of Repair Regression Analysis
E-78 Statistical Summaries for Variables Used in TBI
Before and After Regression Analyses
E-79 Regression Analyses for the Change in Emission
Constituents Before and After Repairs of the TBI
Vehicles for the FTP vs IM240-A
E-80 HC Emissions - AFTP vs IM240-A
E-82 CO Emissions - AFTP - AIM240-A
E-84 NOx Emissions - AFTP vs AIM240-A
E-86 Fuel Economy - AFTP vs AIM240-A
iii
-------
IM240 VEHICLES TESTED TO DATE
12/13/91
Test
FTP
As Received
In Micro
Indiana Ann Arbor
RM#1
In Micro
Indiana Ann Arbor
453
314
213
75
RM#2
In Micro
Indiana Ann Arbor
35 9
RM#3
In Micro
Indiana Ann Arbor
3 0
LaneIM240
Tank Pud IM240
Diagnostic IM240
Non CUP IM240
IM 240 Performed in Lab. Independent
of all other tests.
Indolent CITP-A
IM240Lane
Change of Owner
MISC Repeat
Tank Fuel IM240
After Exhust Repair
7313
6
14
435
15
NA
NA
NA
0
0
73
0
0
0
0
NA
NA
NA
0
0
10
0
0
0
0
NA
NA
NA
0
0
2
0
0
0
0
NA
NA
NA
0
0
Indolene CITP-B
NA
NA
267
14
70
IM240-A (A)
CDH226-A (B)
Steady State- A
IM240-B (D)
CDH226-B (C)
Steady State-B
378
369
0
365
366
0
30
31
0
31
31
0
188
180
0
180
179
0
15
15
0
15
15
0
29
29
0
29
29
0
1
1
0
1
1
0
3
3
0
3
3
0
0
0
0
0
0
0
-------
Regression Analyses - Port Fuel-Injected Vehicle Sample:
As part of EPA's regression analysis, FTP results for HC and CO emissions were
each regressed against the corresponding emission results on the Lane IM240, the second
chance 2500/Idle Test, and the Idle Test. Regressions were also performed comparing
NOx results for the FTP and Lane IM240 (NOx emissions are not measured as part of
either Idle Test and are therefore not included in this comparison). Lasdy, FTP results for
HC, CO, and NOx were each regressed against corresponding emissions from Lab IM240s
run on Indolene fuel at the ATL lab in South Bend and at MVEL in Ann Arbor. This
section provides the results of these regressions performed on a sample of port fuel injected
(PFI) vehicles recruited as part of the Hammond study.
EPA removed thirteen vehicles from its sample of PFI vehicles. Three were
removed because they received repairs for exhaust leaks during the interim between
receiving an I/M test at the Hammond lane and being FTP tested at the lab. The ten
remaining PFI vehicles were excluded due to inconsistent dynamometer settings for the
Lane IM240 and the FTP. Vehicles were only removed if the horsepower or inertia weight
settings were at least 15% different between the lab and the lane.
These differences in dynamometer settings arose from the use of different
dynamometer look-up tables at die lane and lab. Traditional certification test car lists used
for emission factor testing are too detailed for efficient use in the I/M lane. After becoming
aware of the differences in the two tables, EPA opted to use the look-up table that had been
simplified for lane use at both testing sites to avoid future inconsistencies. The resulting
sample consisted of 74 '83 and newer PFI vehicles.
Selecting One Score From the Second Chance 2500/Idle Test:
To illustrate the correlation of FTP HC and CO results to the corresponding results
on the second chance 2500/Idle Test, we had to select an HC score and CO score from the
first chance or second chance test, and from the 2500 rpm mode or the idle mode.
Concerning the first chance versus second chance results, we chose the better results
relative to die HC/CO outpoint being evaluated.
For the 2500 rpm versus idle modes, we chose the larger of the corresponding emissions
measured. This approach is based on the assumption mat the same outpoints would be
used on each of die modes. Hence, a vehicle would pass die two mode test if and only if
the higher of its HC scores (whedier Idle or 2500 rpm) met me HC cutpoint (with the same
also following for die CO results).
Graphing the Data:
On each of die graphs comparing HC emissions, the three horizontal dotted lines
are positioned at the boundaries separating:
• die normal emitters from the high emitters (0.82 g/mi).
the high emitters from the very high emitters (1.64 g/mi), and
• the very high emitters from the super emitters (10.00 g/mi).
E-2A
-------
Similarly, on each of the graphs comparing CO emissions, the three horizontal
dotted lines are positioned at the boundaries separating:
* the normal emitters from the high emitters (10.2 g/mi),
• the high emitters from the very high emitters (13.6 g/mi), and
• the very high emitters from the super emitters (150.0 g/mi).
E-2B
-------
Regression Analyses of FTP HC versus HC Emissions Over:
IM240, Idle Test, and 2500/Idle Test
For PFI Vehicles
Dependent variable is: HCFTP
RA2 = 80.3%
s = 1.016 with
Source
Regression
Residual
Variable
Constant
HCIM24
74 - 2 » 72 degrees of freedom
Sum of Squares df
302.637 1
74.3928 72
Coefficient s.e. of Coeff
0.198655 0.1372
1.13037 0.066
Mean Square
303
1.03323
t-ratio
1.45
17.1
F-ratlo
293
Dependent variable is:
RA2 = 39.3%
s = 1.784
Source
Regression
Residual
Variable
Constant
Higherddle
HCFTP
with 74 - 2 = 72 degrees of freedom
Sum of Squares
148.004
229.026
Coefficient
0.692609
HC... 0.003966
df
1
72
s.e. of Coeff
0.2313
0.0006
Mean Square
148
3.18092
t-ratlo
2.99
6.82
F-ratlo
46.5
Dependent variable is: HCFTP
RA2 = 44.3%
s = 1.709 with 74 - 2 a 72 degrees of freedom
Source
Regression
Residual
Variable
Constant
Idle.HC
Sum of Squares
166.842
210.188
Coefficient
0.706016
0.004541
df
1
72
s.e. of Coeff
0.2184
0.0006
Mean Square
167
2.91927
t-ratlo
3.23
7.56
F-ratlo
57.2
E-3
-------
HC Emissions - FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
Regression Line
RA2 = 80.3%
6 8
IM240 HC Emissions (g/ml)
10
12
E-4
-------
FTPHC
(g/ml)
O ...I.........
HC Emissions -- FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
2.5 -J-
2 -r
1.5 T
1 -r
Regresslon Line
RA2 = 80.3%
0.5 -
**»•
0
0.5
1 1.5 2
IM240 HC Emissions (g/ml)
2.5
E-5
-------
HC Emissions - FTP vs Lane 2500/ldle
(74 1983 & Newer PFI Vehicles)
FTPHC
(g/ml)
12
10 —
8 --
6 -r
4 --
2 --
0 ^
Regression Line
R*2 a 39.3%
0 200 400 600 800 1000 1200 1400 1600 1800 2000
2500/ldle HC Emissions (ppm)
E-6
-------
FTPHC
(9/ml)
3 ~
2.5 -
2 -r
1.5 T
HC Emissions -- FTP vs Lane 2500/ldle
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
R*2 = 39.3%
I
50
100 150
2500/ldle HC Emissions (ppm)
200
250
E-7
-------
FTPHC
(9/ml)
12 -r
HC Emissions - FTP vs Lane Idle
(74 1983 & Newer PFI Vehicles)
10 -L-
8 -
6 -
Regression Line
R*2 = 44.3%
0 •
0
200 400 600
800 1000 1200
Idle HC Emissions (ppm)
1400 1600 1800 2000
E-8
-------
FTPHC
(g/ml)
3 -
HC Emissions - FTP vs Lane Idle
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
1.5 -
1 -
2 5 - - • Regression Line
• RA2 = 44.3%
2 -
0.5 t_
k m . .
•
50 100 150 200 250
Idle HC Emissions (ppm)
E-9
-------
Regression Analyses of FTP CO versus CO Emissions Over:
IM240, Idle Test, and 2500/Idle Test
For PFI Vehicles
Dependent variable is: COFTP
RA2 m 90.4%
s = 17.10 with 74 - 2 = 72 degrees of freedom
Source
Regression
Residual
Variable
Constant
COIM24
Sum of Squares
199166
21052.3
Coefficient
0.446807
1.05072
df
1
72
s.e. of Coeff
2.217
0.0403
Mean Square
200000
292.392
t-ratio
0.202
26.1
F-ratio
681
Dependent variable is: COFTP
RA2 = 76.3%
s » 26.91 with 74 - 2 s 72 degrees of freedom
Source
Regression
Residual
Variable
Constant
Hlgherddle
Sum of Squares
168093
52125.7
Coefficient
4.01842
CO... 23.6992
df
1
72
s.e. of Coeff
3.446
1.555
Mean Square
200000
723.968
t-ratlo
1.17
15.2
F-ratlo
232
Dependent variable is:
RA2 = 61.8%
s m 34.19 with
Source
Regression
Residual
Variable
Constant
Idle.CO
74 - 2 = 72 degi
Sum of Squares
136071
84147.5
Coefficient
8.59165
27.9077
COFTP
ees of freedom
df
1
72
s.e. of Coeff
4.291
2.586
Mean Square
100000
1168.71
t-ratlo
2
10.8
F-ratlo
116
E-10
-------
FTP CO
(g/ml)
250 -
CO Emissions - FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
200 —
150 -T--
100 -1-
50 -
Regression Line
R*2 s 90.4%
50
100 150
IM240 CO Emissions (g/ml)
200
250
E-11
-------
CO Emissions - FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
Frpco Enlarged for Better Detail near Origin
(g/ml)
50 — Regression Line
• R*2 B 90.4%
40 —
30 ^
20
10
• • •
_, : 1 . j i
5 10 15 20 25 30 35 40 45 50
IM240 CO Emissions (g/ml)
E-12
-------
FTP CO
(g/ml)
250 -
CO Emissions - FTP vs Lane 2500/ldle
(74 1983 & Newer PFI Vehicles)
Regression Line
R*2 = 76.3%
200 --
150 -»-•
100 —
50 --
2500/ldle CO Emissions (percent)
E-13
-------
FTP CO
(g/ml)
50
CO Emissions -- FTP vs Lane 2500/ldle
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
40 -1-
30 --
20 -
Regression Line
R*2 = 76.3% '
10 --
0 -f
0
0.2
0.4
0.6 0.8 1 1.2
2500/ldle CO Emissions (percent)
1.4
1.6
1.8
E-14
-------
FTP CO
(g/ml)
250 -
200 --
150 -1-
100 T
50 -
CO Emissions - FTP vs Lane Idle
(74 1983 & Newer PFI Vehicles)
Regression Line
RA2 = 61.8%
4 5
Idle CO Emissions (percent)
E-15
-------
FTP CO
(g/ml)
50 -
CO Emissions - FTP vs Lane Idle
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
40 T
30 --
Regression Line
R*2 = 61.6%
20 4-
10
K1 -"
.Oir
—i—
1.5
0.25
0.5
0.75 1 1.25
Idle CO Emissions (percent)
1.75
—i
2
E-16
-------
Regression Analyses of FTP NOx versus IM240 NOx Emissions
(Note: NOx is not Measured for the Idle Test)
Dependent variable is:
RA2 = 60.3%
s = 0.3277
with 74 - 2 =
Source Sum of Squares
Regressior 11.758
Residual 7.73394
Variable
Constant
NOIM24
Coefficient
0.259283
0.427093
NOFTP
72 degrees of
df
1
72
s.e. of Coeff
0.0587
0.0408
freedom
Mean Square
11.76
0.107416
t-ratlo
4.41
10.5
F-ratlo
109
E-17
-------
FTPNOx
(g/ml)
3 -r
NOx Emissions - FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
2.5 --
2 --
1.5 T
1 -r
0.5 --
Regression Line
R*2 s 60.3%
o -
0.5
1 1.5
2 2.5 3
IM240 NOx Emissions (g/ml)
3.5
4.5
E-18
-------
FTPNOx
(g/mi)
1.5 -
NOx Emissions - FTP vs Lane IM240
(74 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
1.25 --
1 --
0.75 -
0.5 -
0.25 -"
Regression Line
R*2 8 60.3%
H
2
0.25
0.5
0.75 1 1.25
IM240 NOx Emissions (g/ml)
1.5
1.75
E-19
-------
Vehicles Tested by ATL at Southbend, Indiana
Regressions of FTP vs lndolene-IM240
for 63 1983 & Newer PFI Vehicles
Dependent variable is HCFTP
RA2 • 89.2%
S = 0.8037 with
Source
Regression
Residual
Variable
Constant
HC.IM24.lndo...
63 - 2 a 61 degrees
Sum of Squares
325.428
39.4065
Coefficient
0.43699
1.1596
of freedom
df
1
61
s.e. of Coeff
0.1132
0.0517
Mean Square
325
0.646008
t-ratio
3.86
22.4
F-ratio
504
Dependent variable is
RA2 . 90.3%
S = 18.57 with 63 -
Source
Regression
Residual
Variable
Constant
CO.IM24.lndo...
COFTP
2 s 61 degrees of
Sum of Squares
195708
21030.6
Coefficient
5.6588
0.991509
freedom
df
1
61
s.e. of Coeff
2.571
0.0416
Mean Square
200000
344.764
t-ratlo
2.2
23.8
F-ratlo
568
Dependent variable is
RA2»89.1%
s = 0.1780 with 63
Source
Regression
Residual
Variable
Constant
NOx.IM24.lnd...
NOX.FTP
- 2 = 61 degrees
Sum of Squares
15.8477
1.93337
Coefficient
0.145473
0.757394
of freedom
df
1
61
s.e. of Coeff
0.0356
0.0339
Mean Square
15.85
0.031695
t-ratlo
4.09
22.4
F-ratio
500
E-20
-------
FTPHC
(g/ml)
15 -
12.5 --
10 --
7.5
5 -
2.5 -J-
0 1
0
HC Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
Regression Line
R*2 = 89.2%
-*"•
2.5
7.5
10
IM240 HC Emissions (g/ml)
E-21
-------
FTPHC
(g/ml)
3 -
2.5 ->-
2 --
1 -
0.5 J-.
V"*-/
HC Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
0 -1 i ; i ; •• 1 i
0 0.5 1 1.5 2 2.5 3
IM240 HC Emissions (g/ml)
E-22
-------
FTP CO
(g/ml)
250 -
200
150 --
100 —
50 -
CO Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
50
Regression Une
R*2 s 90.3%
100 150
IM240 CO Emissions (g/ml)
200
250
E-23
-------
CO Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
FTP CO
(g/ml)
50 --
45 --
40 --
35 --
30 --
. • •' Regression Line
25 — .--"' R*2s90.3%
.-'' •
20 -- .--
15-- ..--'•
.10 --
5
0 -= ~ i r
10 15 20 25 30 35 40 45 50
IM240 CO Emissions (g/ml)
E-24
-------
FTPNOx
(g/ml)
4.00 —
NOx Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
3.00 —
2.00 -
1.00 -T-
Regression Line
R*2 s 89.1%
•H •
0.00
0.00
1.00
2.00
IM240 NOx Emissions (g/ml)
3.00
4.00
E-25
-------
FTPNOx
(g/ml)
2.00 -
1.75 --
1.50 --
1.25 --
1.00
0.75 --
0.50 --
0.25 -
o.oo -I
NOx Emissions - FTP vs lndolene-IM240
(63 1983 & Newer PFI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
R*2 = 89.1%
II
0.00
0.25
0.50
0.75 1.00 1.25
IM240 NOx Emissions (g/ml)
1.50
1.75
2.00
E-26
-------
EF Vehicles Tested at EPA's MVEL
Regressions of FTP vs Indolene IM240
for 143 1985-90 PFI Vehicles
Dependent variable is:
R*2 = 95.7%
HC.FTP
s = 0.5416 with 143 - 2 = 141 degrees of freedom
Source
Regression
Residual
Variable
Constant
HCIM24
Sum of Squares
927.841
41.3557
Coefficient
0.088877
1.91326
df
1
141
s.e. of Coeff
0.0472
0.034
Mean Square
928
0.293303
t-ratlo
1.88
56.2
F-ratlo
3163
Dependent variable is: CO.FTP
RA2 = 95.4%
s • 5.335 with 143 - 2 = 141 degrees of freedom
Source
Regression
Residual
Variable
Constant
COIM24
Sum of Squares
83635
4012.44
Coefficient
1.68729
1.14873
df
1
141
s.e. of Coeff
0.4594
0.0212
Mean Square
83635
28.457
t-ratlo
3.67
54.2
F-ratlo
2939
Dependent variable is: NOx.FTP
RA2 = 90.5%
s ^ 0.2012 with 143 - 2 » 141 degrees of freedom
Source
Regression
Residual
Variable
Constant
NOIM24
Sum of Squares
54.6898
5.70772
Coefficient
0.10242
0.81782
df
1
141
s.e. of Coeff
0.0237
0.0222
Mean Square
54.69
0.04048
t-ratlo
4.33
36.8
F-ratio
1351
E-27
-------
FTPHC
(9/ml)
30.00 --
20.00 --
10.00 --
HC Emissions - FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Regression Line
B*2 = 95.7%
0.00 *
0.00
5.00 10.00
IM240 HC Emissions (g/ml)
15.00
E-28
-------
FTPHC
(g/ml)
3.00 -
HC Emissions - FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
2.00 T
1.00 --
0.00
Regression Line
B*2 a 95.7%
0.00
0.50
1.00 1.50 2.00
IM240 HC Emissions (g/ml)
2.50
3.00
E-29
-------
FTP CO
(g/ml)
300.00 -
250.00 --
200.00 -
150.00
100.00 -
50.00 -r
CO Emissions - FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Regression Line
R*2 = 95.4%
0.00 •
0.00
50.00
100.00 150.00
IM240 CO Emissions (g/ml)
200.00
250.00
E-30
-------
FTP CO
(a/ml)
50.00 -
40.00 -
CO Emissions -- FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
Regression Une
R*2 = 95.4%
30.00 -
20.00 -
10.00
0.00 ^
0.00
10.00
20.00 30.00
IM240 CO Emissions (g/ml)
40.00
50.00
E-31
-------
FTPNOx
(g/ml)
NOx Emissions - FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
4.00 -1-
Regression Line
R*2 s 90.5%
3.00 --
2.00 -
1.00 4
0.00 -
0.00
1.00
2.00 3.00
IM240 NOx Emissions (g/ml)
4.00
5.00
E-32
-------
NOx Emissions - FTP vs lndolene-IM240
(143 1985-90 PFI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
FTP NOx
(g/ml)
2.00 -
Regression Line
RA2 = 90.5%
1.50 -
0.50 -
0.00 -
0.00 0.50 1.00 1.50 2.00
IM240 NOx Emissions (g/ml)
E-33
-------
Description of Repair Regression Analysis
Regression Analyses
The difference in emission levels before and after repairs for the FTP's and Lab
IM240's run on Indolene fuel was calculated such that a positive difference would
correspond to a decrease in emissions or an increase in fuel economy. The AFTP values
(AHC, AGO, ANOx and AMPG) were regressed against the corresponding Indolene
AIM240 values. The results and graphs of these regressions are included.
Data Description
The data used in this analysis consisted of FTP and Indolene IM240 scores for tests
performed before and after repairs. These results consisted of HC, CO, NOx, and fuel
economy values. The database was restricted to 1983 and newer PFI and TBI vehicles
which received repairs and whose as received FTP scores exceeded twice the FTP standard
for either HC or CO. Also excluded from the analysis were vehicles which received repairs
before the as received FTP and vehicles which had differences in dynamometer settings of
greater than 15%. The resulting data set contained 41 TBI vehicles and 23 PFI vehicles.
Summary Statistics for the variables used in the PFI analyses follow.
E-34
-------
Statistical Summaries for Variables Used in
PFI Before and After Regression Analyses
Summary statistics for: AHC.IM240
NumNumeric = 23
Mean =1.7848
Median = 0.59000
Midrange = 3.7950
Standard Deviation = 2.3719
Range = 8.2500
Minimum = -0.33000
Maximum = 7.9200
Summary statistics for AHC.FTP
NumNumeric = 23
Mean =2.4678
Median =1.9700
Midrange = 4.2600
Standard Deviation = 2.9115
Range = 9.0800
Minimum = -0.28000
Maximum = 8.8000
Summary statistics for: ACO.IM240
NumNumeric = 23
Mean = 53.061
Median = 9.6200
Midrange = 119.73
Standard Deviation = 75.023
Range = 248.85
Minimum = -4.6900
Maximum = 244.16
Summary statistics for ACO.FTP
NumNumeric = 23
Mean = 60.184
Median =14.580
Midrange = 115.89
Standard Deviation = 74.656
Range = 239.46
Minimum = -3.8400
Maximum = 235.62
Summary statistics for: ANOX.IM240
NumNumeric = 23
Mean = -0.36174
Median = -0.23000
Midrange = -0.90000
Standard Deviation = 0.52710
Range = 2.3000
Minimum = -2.0500
Maximum = 0.25000
Summary statistics for ANOX.FTP
NumNumeric = 23
Mean = -0.29522
Median =-0.15000
Midrange = -0.78000
Standard Deviation = 0.48197
Range = 2.2800
Minimum = -1.9200
Maximum = 0.36000
Summary statistics for: AMPG.IM240
NumNumeric = 23
Mean = 2.5904
Median =1.5500
Midrange = 3.2800
Standard Deviation = 2.9593
Range = 10.780
Minimum = -2.1100
Maximum = 8.6700
Summary statistics for AMPG.FTP
NumNumeric = 23
Mean =2.5183
Median =1.7100
Midrange = 5.3200
Standard Deviation = 3.5426
Range = 17.720
Minimum = -3.5400
Maximum = 14.180
E-35
-------
Regression Analyses for the Change in Emission Constituents
Before and After Repairs of the PR Vehicles for the FTP vs IM240-A
Dependent variable is:
RA2 = 90.8%
s = 0.9031
Source
Regression
Residual
Variable
Constant
AHC.IM240
with 23 - 2 = 21
Sum of Squares
169.361
17.1275
Coefficient
0.380093
1.16974
AHC.FTP
degrees of freedom
df
1
21
s.e. of Coeff
0.2376
0.0812
Mean Square
169
0.815595
t-ratio
1.6
14.4
F-ratlo
208
Dependent variable is: ACO.FTP
RA2 = 88.4%
s = 26.01 with 23 - 2 a 21 degrees of freedom
Source Sum of Squares df Mean Square
Regression 108409 1 100000
Residual 14207.5 21 676.547
Variable Coefficient s.e. of Coeff t-ratio
Constant 10.5355 6.693 1.57
ACO.IM240 0.93568 0.0739 12.7
F-ratlo
160
Dependent variable is:
RA2 = 71.2%
s = 0.2649
Source
Regression
Residual
Variable
Constant
ANOX.IM240
ANOX.FTP
with 23 - 2 s 21 degrees of freedom
Sum of Squares df Mean Square
363679 1 3.6368
1.47378 21 0.07018
Coefficient
-0.016187
0.771357
s.e. of Coeff t-ratio
0.0675 -0.24
0.1072 7.2
F-ratio
51.8
Dependent variable is:
RA2 = 36.7%
AMPG.FTP
s = 2.885 with 23 - 2 = 21 degrees of freedom
Source
Regression
Residual
Variable
Constant
AMPG.IM240
Sum of Squares
101.306
174.8
Coefficient
0.639853
0.725132
df
1
21
s.e. of Coeff
0.8074
0.2079
Mean Square
101
8.32381
t-ratlo
0.793
3.49
F-ratlo
12.2
E-36
-------
HC Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*(FTP.Stnd))
A FTP
HC (g/ml)
10 -r
8 --
6 --
4 -r
n
Regression Line
R*2 = 90.8%
1 • *'
1 (
. 9 -
I
1
1 i
234
1 1
5 6
7 £
A IM240 HC (g/ml)
E-37
-------
A FTP
HC (g/nti)
HC Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*(FTP.Stnd))
4 -
3.5 --
s*
^
^
^
^
3 --
2.5 - ^
» • ^^ •
2 - . ^
s*
x'
X*
1.5 - /""
^ ^ Regression Line
^' R*2 a 90.8%
0.5 --
-0.5
-0.5
0.5
1.5
A IM240 HC (g/ml)
2.5
Enlarged for Better Detail near Origin
E-3B
-------
A FTP
CO (g/mi)
-50
CO Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
250 —
200 -
150 -
100 -
50 -
0
j
-50 -
50
Regression Line
R*2 = 88.4%
100
150
200
250
A IM240 CO (g/ml)
E-39
-------
CO Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
A FTP
CO (g/ml)
100 -
80 -
60 -
40 -
20 -r
Regression Line
R*2 = 88.4%
1
20
mo — _
6
i
i
i
-20 -
1 • r~~ —
20 40 60 80
1
1C
A IM240 CO (g/ml)
Enlarged for Better Detail near Origin
E-40
-------
NOx Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
A IM240
NOx (g/mi)
A FTP
NOx (g/ml)
0.5 -T-
-2.5
-2
Regression Line
R*2 = 71.2%
-1.5
-1
• •• •
-0.5 I _^
-0.5 -r
-1 —
-1.5
0.5
-2 —
E-41
-------
NOx Emissions - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
0.4 -
0.2 --
A IM240
NOx (g/ml)
A FTP
NOx (g/ml)
-0.5
-0.4
-0.3i
-0.2 • -0,4-'
0.1
0.2
0.3
Regression Line
RA2 = 71.2%
-0.2 —
-0.4 -r
-0.6 —
-0.8 •+•
Enlarged for Better Detail near Origin
E-42
-------
Fuel Economy - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
AFTPUPG
(ml/gal)
15 -
10 —
Regression Line
B*2 = 36.7%
5 ~~
« JHT
-4
'-2
8
10
-5 -
A IM240 MPG (ml/gal)
E-43
-------
Fuel Economy - AFTP vs Alndolene-IM240
(23 1983 & Newer PFI Vehicles with Repairs and FTP > 2*FTP.Stnd)
A FTP MPG
(ml/gal) 5
4 -
3 -
2 -
1 -
! — "~ A
,-'"" .
3 -2^-" -1 (
^-^ -1 -
"" '
-3 -
-4 -
-5 -
Regression Line
R*2 = 36.7% • ^^-'
^--^
• jn"
-""" ^1 *
— —
'" • *
i I 1 i <
i 1 2345
"
A IM240 MPG (mi/gal)
"
Enlarged for Better Detail near Origin
E-44
-------
Regression Analyses - Throttle-Body Injected Vehicle Sample:
As part of EPA's regression analysis, FTP results for HC and CO emissions were
each regressed against the corresponding emission results on the Lane IM240, the second
chance 2500/Idle Test, and the second chance Idle Test. Regressions were also performed
comparing NOx results for the FTP and Lane IM240 (NOx emissions are not measured as
part of either idle test and are therefore not included in this comparison). Lastly, FTP
results for HC, CO, and NOx were each regressed against corresponding emissions from
Lab IM240s run on Indolene fuel at the ATL lab in South Bend and at MVEL in Ann
Arbor. This section provides the results of these regressions performed on a sample of
throttle-body injected (TBI) vehicles recruited as part of the Hammond study.
EPA removed a total of twenty-two vehicles from its sample of TBI vehicles. Sk
were removed because of repairs performed after their I/M test but before their FTPs, all
but two of which were exhaust leak repairs. Of the remaining two pre-FTP repaired
vehicles, one received a new fuel pump and the other received an alternator and battery
replacement. In addition to these six pre-FTP repair exclusions, sixteen TBI vehicles were
removed from the analyses due to differences of 15% or greater between lane and lab
dynamometer settings. The resulting database consisted of 108 TBI vehicles.
Selecting One Score From the Second Chance 2500/Idle Test:
To illustrate the correlation of FTP HC and CO results to the corresponding results
on the second chance 2500/Idle Test, we had to select an HC score and CO score from the
first chance or second chance test, and from the 2500 rpm mode or the idle mode.
Concerning the first chance versus second chance results, we chose the better results
relative to the HC/CO cutpoint being evaluated.
For the 2500 rpm versus idle modes, we chose the larger of the corresponding emissions
measured. This approach is based on the assumption mat the same outpoints would be
used on each of the modes. Hence, a vehicle would pass the two mode test if and only if
die higher of its HC scores (whether Idle or 2500 rpm) met the HC cutpoint (with the same
also following for the CO results).
Graphing the Data:
On each of the graphs comparing HC emissions, the three horizontal dotted lines
are positioned at the boundaries separating:
• the normal emitters from die high emitters (0.82 g/mi),
• die high emitters from the very high emitters (1.64 g/mi), and
• the very high emitters from the super emitters (10.00 g/mi).
Similarly, on each of die graphs comparing CO emissions, the three horizontal dotted lines
are positioned at the boundaries separating:
the normal emitters from the high emitters (10.2 g/mi).
• the high emitters from the very high emitters (13.6 g/mi), and
• the very high emitters from the super emitters (150.0 g/mi).
E-45
-------
Regression Analyses of FTP HC versus HC Emissions Over:
IM240, Idle Test, and 2500/Idle Test
For TBI Vehicles
Dependent variable is:
RA2 m 64.4%
s = 2.397 with 108 - 2 . 106
Source
Regression
Residual
Variable
Constant
HCIM24
Sum of Squares
1102.31
609.049
Coefficient
-0.149912
1.45609
HCFTP
degrees of freedom
df
1
106
s.e. of Coeff
0.261
0.1051
Mean Square
1102
5.74575
t-ratlo
-0.574
13.9
F-ratio
192
Dependent variable is: HCFTP
RA2 = 8.4%
s a 3.845 with 108 - 2 = 106 degrees of freedom
Source Sum of Squares df
Regression 143.863 1
Residual 1567.5 106
Variable Coefficient s.e. of Coeff
Constant 0.89518 0.4242
Hlqherddle ... 0.003197 0.001
Mean Square F-ratio
144 9.73
14.7877
t-ratlo
2.11
3.12
Dependent variable is:
RA2 = 14.0%
s = 3.727 with 108 - 2 = 106
Source
Regression
Residual
Variable
Constant
Idle.HC
Sum of Squares
238.906
1472.46
Coefficient
0.651022
0.005218
HCFTP
degrees of freedom
df
1
106
s.e. of Coeff
0.418
0.0013
Mean Square
239
13.8911
t-ratlo
1.56
4.15
F-ratlo
17.2
E-46
-------
FTPHC
(9/ml)
35 -
HC Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
30 --
25 4-
Regresslon Line
R*2 = 64.4%
20
15 -r
10 --
5 -1-
6 8
IM240 HC Emissions (g/ml)
10
12
14
E-47
-------
FTPHC
(g/ml)
5 -T
4.5
4 --
3.5 --
3
2.5 --
2 -
1.5 --
1 --
0.5 --
0
HC Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
m •.
Regression Line
= 64.4%
0.5
1.5 2 2.5
IM240 HC Emissions (g/ml)
3.5
E 48
-------
FTPHC
(g/ml)
35 -
30 -
25 -
20 —
HC Emissions - FTP vs Lane 2500/ldle
(108 1983 & Newer TBI Vehicles)
Regression Line
R*2 = 8.2%
15 -
10
0
500
1000
2500/ldle HC Emissions (ppm)
1500
2000
E-49
-------
FTPHC
(g/ml)
5 --
4.5 --
4 --
3.5 --
3 --
2.5 --
2 --
1.5 --
1 -?_-
0.5
0
HC Emissions - FTP vs Lane 2500/ldle
(108 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
Regression I
—r
50
100
150 200 250 300
2500/ldle HC Emissions (ppm)
350
400
450
E-50
-------
FTPHC
(g/ml)
35 -
HC Emissions - FTP vs Lane Idle
(108 1983 & Newer TBI Vehicles)
30 -J-
25 -1-
20 -1-
15 -
Regression Line
R*2 = 14.0%
10
5 -
500
1000
Idle HC Emissions (ppm)
1500
2000
E-51
-------
FTPHC
(g/ml)
5 -
4.5 --
4 --
3.5 --
3 --
2.5 --
rt _.„.,
1.5 --
1
0.5
HC Emissions - FTP vs Lane Idle
(108 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
RA2 * 14.0%
f-Vf*1.
50
100
150 200 250
Idle HC Emissions (ppm)
300
350
400
450
E-52
-------
Regression Analyses of FTP CO versus CO Emissions Over:
IM240, Idle Test, and 2500/Idle Test
For TBI Vehicles
Dependent variable is:
RA2 = 79.9%
s = 20.97 with 108 - 2 = 106
Source
Regression
Residual
Variable
Constant
COIM24
Sum of Squares
185395
46625.6
Coefficient
-3.46603
1.41279
COFTP
degrees of freedom
df
1
106
s.e. of Coeff
2.331
0.0688
Mean Square
200000
439.864
t-ratio
-1.49
20.5
F-ratio
421
Dependent
RA2 = 46.9%
s = 34.10
Source
Regression
Residual
Variable
Constant
Hlgherddle
variable is:
with 108 - 2 » 106
Sum of Squares
108761
123260
Coefficient
4.43526
15.0538
COFTP
degrees of freedom
df
1
106
s.e. of Coeff
3.677
1.557
Mean Square
100000
1162.83
t-ratlo
1.21
9.67
F-ratlo
93.5
Dependent
s= 32.73
Source
Regression
Residual
Variable
Constant
Idle.CO
variable is:
with 108 - 2 • 106
Sum of Squares
118502
113519
Coefficient
5.15188
17.2888
COFTP
degrees of freedom
df
1
106
s.e. of Coeff
3.47
1.644
Mean Square
100000
1070.93
t-ratlo
1.48
10.5
F-ratio
111
E-53
-------
FTP CO
(g/ml)
300 -r
250 --
200 --
150
100 --
50 --
CO Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
Regression Line
R*2 = 79.9%
i •
25
50
75 100 125
IM240 CO Emissions (g/ml)
150
175
200
E-54
-------
CO Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
(g/ml) Enlarged for Better Detail near Origin
50 ~ Regression Une
45 --
40 --
35 --
30 --
25 --
20 --
15
10
5 -
0
ft*2 = 70.9%
0 5 10 15 20 25 30 35 40
IM240 CO Emissions (g/ml)
E-55
-------
FTP CO
(g/ml)
300 -
250 --
200 T
150
100 —
50 -1-
CO Emissions - FTP vs Lane 2500/ldle
(108 1983 & Newer TBI Vehicles)
Regression Line
RA2 = 46.9%
* V
0
456
2500/ldle CO Emissions (percent)
10
E-56
-------
FTP CO
(g/ml)
45 --
40 --
35 --
30 --
25 --
20 --
15 --
10
CO Emissions - FTP vs Lane 2500/ldle
(108 1983 & Newer TBI Vehicles)
Enlarged (or Better Detail near Origin
Regression Line
R*2 = 46.9%
0.2 0.4 0.6 0.8 1 1.2
2500/ldle CO Emissions (percent)
1.4
1.6
1.8
E-57
-------
FTP CO
(g/mi)
300 T
CO Emissions - FTP vs Lane Idle
(108 1983 & Newer TBI Vehicles)
250 --
200 -1-
Regresslon Line
R*2 a 51.1%
150
100 --
50 --
456
Idle CO Emissions (percent)
10
E-58
-------
FTP CO
(g/ml)
50 -
CO Emissions - FTP vs Lane Idle
(108 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
45 4- RA2sS1.1%
40 --
35
30
25 -
20
15
.10
5
0 ^
•
•
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Idle CO Emissions (percent)
E-59
-------
Regression Analyses of FTP NOx versus IM240 NOx Emissions
(Note: NOx is not Measured for the Idle Test)
Dependent variable is: NOFTP.TBI
RA2 = 80.7%
s = 0.4677 with 108 - 2 = 106 degrees of freedom
Source
Regression
Residual
Variable
Constant
NOIM24.TBI
Sum of Squares
96.941
23.1857
Coefficient
-0.069973
0.734578
df
1
106
s.e. of Coeff
0.071
0.0349
Mean Square
96.94
0.218733
t-ratlo
-0.985
21.1
F-ratlo
443
E-60
-------
FTPNOx
(g/ml)
7 -r
6 --
5 -
4 --
3 -
2 --
o -
NOx Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
Regression Line
R*2 = 80.7%
345
IM240 NOx Emissions (g/ml)
E-61
-------
FTPNOx
(g/ml)
2 -
1.75 --
1.5 --
1.25 --
1 --
0.75 --
0.5 \
0.25 --
NOx Emissions - FTP vs Lane IM240
(108 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
Regression Une
B*2 = 80.7%
0.25
0.5
0.75 1 1.25
IM240 NOx Emissions (g/ml)
1.5
1.75
E-62
-------
Vehicles Tested by ATL at Southbend, Indiana
Regressions of FTP vs lndolene-IM240
for 91 1983 & Newer TBI Vehicles
Dependent variable is HCFTP
RA2 = 90.1%
s = 1.354 with
Source
Regression
Residual
Variable
Constant
HC.lM24.lndo...
91 - 2 » 89 degrees of
Sum of Squares
1491.43
163.188
Coefficient
0.002929
1.66576
freedom
df
1
89
s.e. of Coeff
0.1537
0.0584
Mean Square
1491
1.83357
t-ratio
0.019
28.5
F-ratio
813
Dependent variable is
RA2 m 90.3%
s = 15.12 with 91 -
Source
Regression
Residual
Variable
Constant
CO.IM24.lndo...
COFTP
2 = 89 degrees of
Sum of Squares
190004
20343.9
Coefficient
-0.174774
1 .34497
freedom
df
1
89
s.e. of Coeff
1.758
0.0467
Mean Square
200000
228.583
t-ratlo
-0.099
28.8
F-ratlo
831
Dependent variable is NOx.FTP
RA2 = 92.6%
s = 0.2989 with
Source
Regression
Residual
Variable
Constant
NOx.IM24.lnd...
91 - 2 = 89 degrees
Sum of Squares
99.1838
7.9495
Coefficient
0.073869
0.845099
of freedom
df
1
89
s.e. of Coeff
0.0443
0.0254
Mean Square
99.18
0.08932
t-ratio
1.67
33.3
F-ratio
1110
E-63
-------
FTPHC
(g/ml)
30 -
HC Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
25 -
Regression Line
R*2 = 90.1%
20 -
15
10
5 -
0
10
IM240 HC Emissions (g/ml)
15
20
E-64
-------
FTPHC
(g/mi)
3 -
HC Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
2.5 -
2 -
1.5 -
Regression Line
R*2 = 90.1%
1 -
I •
0.5 -a
VI
0.5
1.5
IM240 HC Emissions (g/ml)
2.5
E-65
-------
FTP CO
(g/ml)
300 -
CO Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
250 -
Regression Line
B*2 = 90.3%
200 -
150 -
100 --
50 -J-
o -i
0
20
40 60
80 100 120
IM240 CO Emissions (g/ml)
140 160 180
200
E-66
-------
FTP CO
(g/ml)
50 -
CO Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
40 -
30 -
20 -
Regression Line
RA2 = 90.3%
10 -1-
0 -
10
20 30
IM240 CO Emissions (g/ml)
40
50
E-67
-------
FTPNOx
(g/ml)
6.00 --
5.00 --
4.00 -*-
3.00 -
2.00 --
NOx Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
Regression Line
RA2 s 92.6%
1.00 --
0.00 -r
0.00
1.00 2.00
3.00 4.00 5.00
IM240 NOx Emissions (g/ml)
6.00 7.00
8.00
E-68
-------
FTPNOx
(a/ml)
2.00
1.75 -1-
NOx Emissions - FTP vs lndolene-IM240
(91 1983 & Newer TBI Vehicles)
Enlarged for Better Detail near Origin
Regression Line
R*2 = 92.6%
1.50
1.25
1.00 --
0.75 --
0.50 --
0.25 --
V • "
•>^ • 1
"l "• •
o.oo -
0.00
0.25
0.50
0.75 1.00 1.25
IM240 NOx Emissions (g/ml)
1.50
1.75
2.00
E-69
-------
EF Vehicles Tested at EPA's MVEL
Regressions of FTP vs Indolene IM240
for 62 1985-88 TBI Vehicles
Dependent variable is: HC.FTP
RA2 = 87.7%
s = 1 .703 with 62 - 2 = 60 degrees of freedom
Source
Regression
Residual
Variable
Constant
HCIM24
Sum of Squares
1241.84
173.984
Coefficient
0.146231
1.71372
df
1
60
s.e. of Coeff
0.2279
0.0828
Mean Square
1242
2.89973
t-ratlo
0.642
20.7
F-ratio
428
Dependent variable is: CO.FTP
RA2 a 92.5%
s a 1 1 .59 with 62 - 2 = 60 degrees of freedom
Source
Regression
Residual
Variable
Constant
COIM24
Sum of Squares
99522.9
8056.07
Coefficient
2.56957
1.10181
df
1
60
s.e. of Coeff
1.584
0.0405
Mean Square
99523
134.268
t-ratlo
1.62
27.2
F-ratlo
741
Dependent variable is: NOx.FTP
RA2 m 90.9%
s - 0.1594
Source
Regression
Residual
Variable
Constant
NOIM24
with 62 - 2 = 60 degrees
Sum of Squares
15.2107
1.52398
Coefficient s.e
0.067854
0.780974
of freedom
df
1
60
. of Coeff
0.0323
0.0319
Mean Square
15.21
0.0254
t-ratlo
2.1
24.5
F-ratio
599
E-70
-------
FTPHC
(g/ml)
30.00 --
20.00 --
10.00 -"-
0.00 -fl
0.00
HC Emissions - FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
5.00
Regression Line
R*2 = 87.7%
10.00
IM240 HC Emissions (g/ml)
15.00
20.00
E-71
-------
HC Emissions - FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
FTPHC
.00 T
2.00 +
1.00 +
0.00
Regression Line
R*2 s 87.7%
. *• •• •
B *•%
k#- -
0.00 0.50 1.00 1.50 2.00
IM240 HC Emissions (g/ml)
E-72
-------
FTP CO
(g/ml)
200.00 --
150.00 --
100.00 --
50.00 --
CO Emissions -- FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
0.00 *
0.00
50.00
Regression Line
R*2 = 92.5%
100.00
IM240 CO Emissions (g/ml)
150.00
200.00
E-73
-------
FTP CO
(g/ml)
50.00 -
CO Emissions - FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
40.00 T
30.00 -
20.00 -
Regression Line
RA2 = 92.5%
10.00 T
0.00 -
0.00
10.00
20.00 30.00
IM240 CO Emissions (g/ml)
40.00
50.00
E-74
-------
FTP NOx
NOx Emissions - FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
3.00 -1-
2.50
2.00 —
1.50 -
1.00 -
0.50 -
0.00 -r
Regresslon Line
RA2 = 90.9%
0.00
0.50
1.00
1.50 2.00 2.50
IM240 NOx Emissions (g/ml)
3.00
3.50
4.00
E-75
-------
FTPNOx
1.50 --
1.00
0.50 --
o.oo
NOx Emissions - FTP vs lndolene-IM240
(62 1985-88 TBI Vehicles)
(Tested at EPA's Ann Arbor Lab)
Enlarged for Better Detail near Origin
Regression Line
B*2 s 90.9%
• •
0.00
0.50
1.00
IM240 NOx Emissions (g/ml)
1.50
2.00
E-76
-------
Description of Repair Regression Analysis
Regression Analyses
The difference in emission levels before and after repairs for the FTP'S and Lab
IM240's run on Indolene fuel was calculated such that a positive difference would
correspond to a decrease in emissions or an increase in fuel economy. The AFTP values
(AHC, AGO, ANOx and AMPG) were regressed against the corresponding Indolene
AIM240 values. The results and graphs of these regressions are included.
Data Description
The data used in this analysis consisted of FTP and Indolene IM240 scores for tests
performed before and after repairs. These results consisted of HC, CO, NOx, and fuel
economy values. The database was restricted to 1983 and newer PFI and TBI vehicles
which received repairs and whose as received FTP scores exceeded twice the FTP standard
for either HC or CO. Also excluded from the analysis were vehicles which received repairs
before the as received FTP and vehicles which had differences in dynamometer settings of
greater than 15%. The resulting data set contained 41 TBI vehicles and 23 PFI vehicles.
Summary Statistics for the variables used in the TBI analyses follow.
E-77
-------
Statistical Summaries for Variables Used in
TBI Before and After Regression Analyses
Summary statistics for: AHCIM24
NumNumeric = 41
Mean =1.3563
Median » 0.2200
Midrange = 7.8500
Standard Deviation = 3.4020
Range = 17.660
Minimum = -0.98000
Maximum = 16.680
Summary statistics for: AHC.FTP
NumNumeric = 41
Mean = 2.4634
Median = 0.3700
Midrange = 14.595
Standard Deviation = 6.0860
Range = 30.590
Minimum = -0.70000
Maximum = 29.890
Summary statistics for ACO.IM24
NumNumeric = 41
Mean =24.092
Median = 4.8800
Midrange = 84.310
Standard Deviation = 47.490
Range = 186.72
Minimum = -9.0500
Maximum = 177.67
Summary statistics for: ANOx.IM24
NumNumeric = 41
Mean = 0.04683
Median = -0.07000
Midrange = 2.2850
Standard Deviation = 1.2167
Range = 8.3700
Minimum = -1.9000
Maximum = 6.4700
Summary statistics for: AMPG.IM24
NumNumeric = 41
Mean = 2.4132
Median =1.4300
Midrange = 5.2600
Standard Deviation = 3.4721
Range = 15.420
Minimum = -2.4500
Maximum = 12.970
Summary statistics for ACO.FTP
NumNumeric = 41
Mean = 38.661
Median = 5.4500
Midrange = 128.11
Standard Deviation = 69.2007
Range = 265.65
Minimum = -4.7200
Maximum = 260.93
Summary statistics for ANOx.FTP
NumNumeric = 41
Mean = -0.06439
Median = -0.06000
Midrange =1.5800
Standard Deviation = 1.0174
Range = 6.8800
Minimum = -1.8600
Maximum = 5.0200
Summary statistics for: AMPG.FTP
NumNumeric = 41
Mean = 2.0820
Median = 0.97000
Midrange = 3.6850
Standard Deviation = 3.7642
Range = 21.810
Minimum - -7.2200
Maximum = 14.590
E-78
-------
Regression Analyses for the Change in Emission Constituents
Before and After Repairs of the TBI Vehicles for the FTP vs IM240-A
Dependent variable is: AHC.FTP
R*2 m 87.2%
s = 2.209 with 41 - 2 = 39 degrees of freedom
Source Sum of Squares df Mean Square
Regression 1291.3 1 1291
Residual 190.271 39 4.87875
Variable Coefficient s.e. of Coeff t-ratlo
Constant 0.198146 0.372 0.533
AHC.IM24 1.67013 0.1027 16.3
F-ratlo
265
Dependent variable is: ACO.FTP
RA2=82.1%
s = 29.57 with 41 - 2 » 39 degrees of freedom
Source Sum of Squares df Mean Square
Regression 1 56368 1
Residual 34110.7 39
Variable Coefficient s.e. of Coeff
Constant 6.94316 5.192
ACO.IM240 1.31656 0.0985
200000
874.633
t-ratio
1.34
13.4
F-ratlo
179
Dependent variable is:
RA2 = 84.9%
s m 0.4008
Source
Regression
Residual
Variable
Constant
ANOX.IM240
with 41 - 2 = 39
ANOX.FTP
degrees of freedom
Sum of Squares df Mean Square
35.1435
6.26452
Coefficient
-0.100466
0.770364
1
39 0.
s.e. of Coeff t
0.0626
0.0521
35.14
160629
-ratio
-1.6
14.8
F-ratio
219
Dependent variable is:
RA2 m 58.3%
S = 2.489 with
Source
Regression
Residual
Variable
Constant
AMPG.IM240
AMPG.FTP
41 - 2 = 39 degrees
Sum of Squares
325.213
241.558
Coefficient
0.100224
0.821213
s.e.
0
0
of freedom
df Mean Square
1 325
39 6.19379
of Coeff
.4753
.1133
t-ratlo
0.158
7.25
F-ratio
52.5
E-79
-------
A FTP
HC (g/ml)
HC Emissions - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > 2*FTP.Stnd)
35 T
30 +
25 +
20
15 +
10 +
5
Regression Line
B*2 s 87.2%
J^
T"
8
10
12
14
- 5
A IU240 HC (g/ml)
16
18
E-80
-------
HC Emissions - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > 2*FTP.Stnd)
AFTP HC
(g/ml)
o
Regression Line
R*2 = 87.2%
A IM240 HC (g/ml)
Expanded for Better Detail near Origin
E-81
-------
A FTP
CO (g/ml)
300 —
250 J-
200 —
150 —
100 —
50 -L
CO Emissions - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP Z 2*FTP.Stnd)
Regression Line
RA2 a 82.1%
-2O
I
-50 —
20 40 60 80 100 120 140 160 180
A IM240 CO (g/ml)
E-82
-------
A FTP
CO (g/ml)
CO Emissions - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > 2*FTP.Stnd)
50 -
40 --
30 --
20 --
•
10 --
Regression Line
H*2 = 82.1%
x
X
X
X
X
X
X
10
20
30
40
-10
Expanded for Better Detail near Origin
A IM240 CO (g/ml)
50
E-83
-------
-2
NOx Emissions - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > FTP.stnd)
A FTP
NOx (g/ml)
6 -
5 -
4 -
3 -
2 -
•4
Regression Line
RA2 a 84.9%
-J *
-2 -
A IM240 NOx (g/ml)
E-84
-------
NOx Emissions -- AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > FTP.stnd)
A FTP NOx
(g/ml)
1.5 -
1 —
0.5 —
Regression Line
R*2 = 84.9%
A IM240
NOx fa/nth
-2
-1.5
-1
-0.5
0.5
1.5
-0.5 -r-
-1.5 -
-2 —
Expanded for Better Detail near Origin
E-85
-------
Fuel Economy - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > 2*FTP.Stnd)
15 -
A FTP
MPG (ml/gal)
10 -1-
5 -
Regression Line
R*2 = 57.4%
•A
8 10 12 14
-5 -
-10 -
A IM240 MPG (ml/gal)
E-86
-------
Fuel Economy - AFTP vs Alndolene-IM240
(41 1983 & Newer TBI Vehicles with Repairs and FTP > 2*FTP.Stnd)
AFTPMPG
(ml/gal) 5
3 -
2 --
Regression Line
R*2 a 57.4%
-3
-2
-2 --
-3 --
-4 --
-5 --
Expanded for Better Detail near Origin
A IM240 MPG
(ml/gal)
E-87
-------
APPENDIX F
MOBILE4.1 TECHNOLOGY DISTRIBUTION AND EMISSION
GROUP RATES AND EMISSION LEVELS
-------
MOBILE4.1
Technology Distribution for Gasoline Passenger Cars
HP [1,1
XCK
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
Mo [|a|
XcK
1981
1982
1983
1984
1985
1986
1987
1986
1989
1990
1991
Modal
Star
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
CLLP
3iray
Stttt
1571360
1040443
151414
761159
966873
594368
415441
277445
613700
10800
9000
CLLP
Stray
fin*
23.0%
15.9%
22%
8.6%
12.9%
75%
6.4%
4.1%
72%
0.1%
0.1%
CLLP
Cob
62.9%
50.6%
48.6%
55.1%
40.8%
31.9%
24.9%
10.1%
125%
1.8%
03%
CLLP
Stray
UPFI
413972
403436
591266
938060
2182617
2607047
2239841
2998983
4652439
5177046
6419242
CLLP
Stray
lies
6.0%
62%
8.5%
10.5%
292%
32.9%
34.7%
44.4%
54.6%
71 .8%
77.4%
CLLP
MEB
6.1%
62%
8.8%
11.0%
30.7%
392%
372%
492%
59.7%
76.1%
79.5%
CLLP
3MfBy
IBI
0
435887
794934
1414498
432479
1058186
1397261
2206819
2034135
1395418
1570776
CLLP
Stray
IBI
0.0%
6.7%
115%
15.9%
5.8%
133%
21.7%
32.7%
23.9%
19.4%
18.9%
CLLP
IBI
2.9%
10.6%
183%
282%
20.8%
265%
36.3%
40.7%
275%
21.9%
202%
Oplp
3WNBy
Cmtb
48896
54160
0
25877
25476
50
16821
0
0
0
0
Oplp
9my
£Mfe
0.7%
03%
0.0%
03%
0.3%
0.0%
03%
0.0%
0.0%
0.0%
0.0%
Oplp
Any
28.1%
325%
24.4%
5.8%
7.7%
2.4%
1.7%
0.0%
03%
0.1%
0.0%
Oplp
Sway
MEB
0
0
0
0
0
0
64
65
90
0
0
Oplp
wWwy
MEB
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Oplp
Stray
IBI
0
0
0
0
0
150047
41918
0
0
0
0
Oplp
Stray
IBI
0.0%
0.0%
0.0%
0.0%
0.0%
1.9%
0.7%
0.0%
0.0%
0.0%
0.0%
Model
XtK
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
CLLP
OxStmy
Garb
2731528
2265172
3211332
4140952
2085078
1936299
1188253
404916
449631
122142
15690
CLLP
Ox3w*y
Cob
393%
34.7%
46.4%
465%
27.9%
24.4%
18.4%
6.0%
5.3%
1.7%
02%
Any
Cafe
91.0%
832%
72.9%
60.9%
484%
32.4%
25.9%
10.1%
12.8%
1.9%
0.3%
CLLP
OxSwwy
MEB
1434
0
15517
40669
116557
504156
156340
322283
431553
311612
176552
CLLP
Ox3w»y
MEB
0.0%
0.0%
02%
05%
1.6%
6.4%
2.4%
4.8%
5.1%
43%
2.1%
Any
MEB
6.1%
62%
8.8%
11.0%
30.7%
392%
372%
492%
59.7%
76.1%
795%
CLLP
OxSway
IBI
201346
258479
471542
1091249
1125292
1041046
942058
542380
308598
186117
103342
CLLP
OxStmy
IBI
2.9%
4.0%
6.8%
123%
15.0%
13.1%
14.6%
8.0%
3.6%
2.6%
12%
Any
IBI
2.9%
10.6%
183%
282%
20.8%
28.4%
36.9%
40.7%
275%
21.9%
202%
Oplp
OxSvray
CMfe
932653
1181258
870343
488429
547055
40209
46809
0
25000
7000
0
Oplp
OxSvmy
CjSfe
13.6%
18.1%
12.6%
55%
73%
05%
0.8%
0.0%
03%
0.1%
0.0%
Oplp
OxStray
MEB
284
0
409
0
0
0
650
0
0
0
0
Oplp
OxStray
MEB
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Model
IMC
1981
1982
1963
1984
1985
1986
1987
1988
1989
1990
1991
Oplp
OxStmy
IBI
0
0
0
0
0
0
0
0
0
0
0
Oplp
OxStmy
IBI
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Any
Stray
29.7%
29.6%
222%
353%
482%
55.6%
63.8%
812%
85.7%
913%
96.4%
Oplp
OxM
943390
889494
817624
0
0
0
0
0
0
0
0
Oplp
OxM
13.8%
13.6%
11.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Any
OxSiray
565%
56.8%
66.0%
64.7%
51.8%
44.4%
362%
18.8%
143%
8.7%
3.6%
Oplp
fton*
38
10
0
0
0
0
0
0
0
0
0
Oplp
fton»
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Any
Ox Id
133%
13.6%
11.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
AJ1
6844901
6528339
6924581
8900893
7481429
7931408
6447456
6752891
8515146
7210135
8294602
GMFI
133369
636022
831142
1444036
1644484
2087185
1673716
2420729
3018191
2522963
3639046
%QMFI
1.9%
9.7%
12.0%
162%
22.0%
263%
26.0%
35.8%
35.4%
35.0%
43.9%
All vehicle counts taken from CAFE estimates, except 1989 and newer model years, which are from General Label predictions.
7/2/91
-------
MOBILE4.1
Emitter Group Emission Levels*
Modsl
Ysar
Group
19B1-B2
196142
196142
1981-82
108345
1983-85
1983-85
1983*5
1986+
1986+
1986+
1986*
1981-82
1981-82
1961-82
1981-62
1983+
1983+
1983+
1983+
198142
198142
198142
196142
1983+
1983+
1983+
1983+
198142
196142
196142
196142
1963+
1983+
1983+
1983+
Emission
UM|
Group
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Noifisw
Normal
High
High
High
High
High
High
High
High
Vary High
Vary High
Very High
Vsry High
Vary High
Vsry High
Vsry High
Vsry High
Supsr
Supsr
Supsr
Supsr
Supsr
Supsr
Supsr
Supsr
Technology
Group
MPR
TO
Cart.
Oplp
MPH
TO
Can
Oplp
MPH
TO
Can
Oplp
MPH
TO
Can
Oplp
MPH
TO
Can
Oplp
MPH
TO
Can
Oplp
MPH
TO
Can
Oplp
MPH
TO
Cart.
Oplp
MPH
TO
Can
Oplp
HC
an.
0.188
0.279
0.288
0.306
0.269
0.242
0.222
0.334
0.269
0242
0222
0.334
0.374
0.758
0.781
0.757
0.997
0.762
0.703
0.840
0.606
1.163
1.950
1389
2.019
2.242
2.002
1.352
15.259
15.259
15259
0.000
11.123
16.083
12.870
0.000
HC
BEL
0.0450
0.0161
0.0253
0.0230
0.0115
0.0134
0.0199
0.0248
0.0115
0.0134
0.0199
0.0248
0.0450
0.0161
0.0253
0.0230
0.0115
0.0134
0.0199
0.0248
0.0450
0.0161
0.0253
0.0230
0.0115
0.0134
0.0199
0.0248
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
HC
Emission*
at 50.000
MJjsj
0.413
0.360
0.415
0.421
0.327
0.309
0.322
0.458
0.327
0.309
0.322
0.458
0.599
0.839
0.908
0.872
1.055
0.829
0.803
0.964
0.831
1244
2.077
1.704
2.077
2.309
2.102
1.476
15.259
15259
15.259
0.000
11.123
18.083
12.870
0.000
HC
Emissions
at 100,000
Mllss .
0.638
0.440
0541
0.536
0.384
0.376
0.421
0.582
0.384
0.376
0.421
0.582
0.824
0.919
1.034
0.987
1.112
0.896
0.902
1.088
1.056
1.324
2203
1.819
2.134
2.376
2.201
1.600
15.259
15.259
15.259
0.000
11.123
18.083
12.870
0.000
CO
ZUL
1.548
3.422
3.067
3.368
2.S98
2.953
2209
4.093
2.598
2.953
2.209
4.093
3.879
6.843
8.496
8.142
7.602
8.820
8.577
8.386
18.049
22.059
33.396
27.650
22.301
44.416
36.130
34.021
173.840
173.840
173.840
0.000
180.000
183.590
246.850
0.000
CO
EH
0.5807
0.2520
0.3281
02877
0.1554
0.1572
0.2282
0.2093
0.1554
0.1572
02282
02093
05807
0.2520
0.3281
02877
0.1554
0.1572
02282
02093
0.5807
02520
0.3281
0.2877
0.1554
0.1572
02282
02093
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
CO
Emissions
at 50,000
MJtet
4.452
4.682
4.708
4.807
3.375
3.739
3.350
5.140
3.375
3.739
3.350
5.140
6.783
8.103
10.137
9.581
8.379
9.606
9.718
9.433
20.953
23.319
35.037
29.089
23.078
45.202
37.271
35.068
173.840
173.840
173.840
0.000
189.000
183.590
246.850
0.000
CO
Emissions
at 100,000 NOx
Ifflsj OIL
7.355 0.380
5.942 0.545
6.348 0.676
6245 0.649
4.152 0.689
4.525 0.585
4.491 0.692
6.186 0.524
4.152 0.539
4525 0.316
4.491 0.478
6.186 0.524
9.686
9.363
11.777
11.019
9.156
10.392
10.859
10.479
23.856
24.579
36.677
30.527
23.855
45.988
38.412
36.114
173.840
173.840
173.840
0.000
189.000
183.590
246.850
0.000
NOx NOx
Emissions Emissions
NOx at 50.000 at 100,000
PET Mllas Mllss
0.1280 1.020 1.660
0.1414 1.252 1.959
0.0696 1.026 1.374
0.0440 0.869 1.089
0.0185 0.782 0.874
0.0470 0.820 1.055
0.0558 0.971 1.250
0.0540 0.794 1.064
0.0185 0.632 0.724
0.0470 0.551 0.786
0.0558 0.757 1.036
0.0540 0.794 1.064
' All emission (avals in grains per mis. DET in grams per mile per 10,000 miles.
BLOCK.XLS
11/8/91
-------
MOBILE4.1
Emitter Group Rates
Model
Year
Group
1981-82
1981-82
1981-82
1981-82
1983+
1983+
1983+
1983+
1981-82
1981-82
1981-82
1981-82
1983+
1983+
1983+
1983+
1981-82
1981-82
1981-82
1981-82
1983+
1983+
1983+
1983+
1981-82
1981-82
1981-82
1981-82
1983+
1983+
1983+
1983+
Emission
Level
Group
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Normal
High
High
High
High
High
High
High
High
Very High
Very High
Very High
Very High
Very High
Very High
Very High
Very High
Super
Super
Super
Super
Super
Super
Super
Super
Technology
Group
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
MPFI
TBI
Garb
Oplp
Growth
Rate
per 10,000
Miles
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.0065
0.0065
0.0270
0.0270
0.0090
0.0040
0.0140
0.0130
0.01933
0.01198
0.03762
0.03530
0.00840
0.02012
0.01487
0.00433
0.00257
0.00257
0.00257
0.00000
0.00194
0.00194
0.00194
0.00000
Rate
Increase
Beyond
50.000 Miles
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
13.8353
13.8353
0.9643
1.0000
4.4709
15.2076
2.3200
1.0000
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Percent
at
50,000
Miles
85.8%
89.5%
66.4%
68.8%
90.3%
87.0%
84.6%
91.3%
3.3%
3.3%
13.5%
13.5%
4.5%
2.0%
7.0%
6.5%
9.7%
6.0%
18.8%
17.7%
4.2%
10.1%
7.4%
2.2%
1.3%
1.3%
1.3%
0.0%
1.0%
1.0%
1.0%
0.0%
Percent
at
100,000
Miles
29.9%
37.2%
33.3%
37.7%
65.0%
45.5%
60.0%
82.7%
48.2%
48.2%
26.5%
27.0%
24.6%
32.4%
23.2%
13.0%
19.3%
12.0%
37.6%
35.3%
8.4%
20.1%
14.9%
4.3%
2.6%
2.6%
2.6%
0.0%
1.9%
1.9%
1.9%
0.0%
BLOCK.XLS
11/8/91
-------
APPENDIX G
EXHAUST SHORT TEST ACCURACY: IM240 VS. SECOND-CHANCE
2500 RPM/IDLE TEST
PFI p. G-2
TBI p. G-5
CARB p. G-8
-------
Special Note On Appendix G:
Appendix G shows the accuracy of the IM240 test when a vehicle is failed only if its
composite emissions exceed 0.8 g/mi HC or 15 g/mi CO and its Bag 2 emissions alone exceed 0.5
g/mi HC or 12 g/mi CO. In other words, a vehicle can pass by having low emissions in Bag 2
even if Bag 1 emissions were very high. EPA has found that this prevents inappropriately failing
clean cars with no sacrifice in identifyimg dirty cars. For this reason, EPA is proposing this
approach to IM240 cutpoints.
Please note that second chance Idle/2500 tests were performed on 1983 and newer vehicles
only, and therefore were not available for the 1981 and 1982 carbureted vehicles used to construct
these charts. In all cases, when a second chance Idle/2500 test was available, the better test was
used.
G-l
-------
Figure 1
Comparison of IM240 to Second Chance 2500/Idle for
1983 & Newer Port Fuel Injected Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite and 0.5 &12 g/mi Bag 2
Second Chance 2500/Idle Clpts = 220 ppm & 1.2%
90% —
80% -t-
70%
60%
50%
40%
30%
20% -
10%
0%
85%
82%
73%
65%
14%
0.0% 0.0%
1.7%
2.3%
HCIDR
COIDR
I/M Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
D 2-Spd (220ppm & 1.2%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-2
-------
Figure 2
Comparison of IM240 to Second Chance 2500/Idle for
1983 & Newer PFI Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 & 12 g/mi Bag 2
Second Chance 2500/ldle Ctpts = 220 ppm & 1.2% @2500; 100 ppm & 1.0% @Idle
90%
80%
70%
60%
50%
40%
30% —
20%
10% -
85%
22%
13%
0.0%
2.3%
HCIDR
COIDR
I/M Fail Rate
I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
D 2-Spd (220/lOOppm & 1.2/1.0%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-3
-------
Figure 3
Comparison of EM240 to Second Chance 2500/Idle for
1983 & Newer Port Fuel Injected Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 &12 g/mi Bag 2
Second Chance 2500/Idle Ctpts = 100 ppm & 0.5% both modes
90%
80%
70% --
60%
50%
40% -t-
30% --
20%
10%
0%
85%
26%
13%
0.0%
2.3%
HCIDR
COIDR
I/M Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
O 2-Spd (lOOppm & 0.5%)
EM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-4
-------
Figure 4
Comparison of IM240 to Second Chance 2500/Idle for
1983 & Newer Throttle Body Injected Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 &12 g/mi Bag 2
Second Chance 2500/Idle Ctpts = 220 ppm & 1.2%
90% -r
89%
10%
0%
3.S
4.1%
0.0%
HCEDR
CO DDR
I/M Fail Rate
I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
02-Spd(220ppm&1.2%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-5
-------
Figure 5
Comparison of IM240 to Second Chance 2500/Idle for
1983 & Newer TBI Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 & 12 g/mi Bag 2
Second Chance 2500/ldIe Ctpts = 220 ppm & 1.2% @2500; 100 ppm & 1.0% @Idle
100%
90%
80%
70% —
60% --
50%
40%
30%
20%
10% -
90%
89%
16%
12%
0.0%
4.1%
HCIDR
COIDR
I/M Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
D 2-Spd (220/lOOppm & 1.2/1.0%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-6
-------
Figure 6
Comparison of IM240 (o Second Chance 2500/Idle for
1983 & Newer Throttle Body Injected Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite: 0.5 & 12 g/mi Bag 2
Second Chance 2500/Idle Clpts = 100 ppm & 0.5% both modes
100%
40%
20%
10%
0%
20%
22%
0.0%
4.1%
HCIDR
COEDR
I/M Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Nonnal Emitters
U 2-Spd (lOOppm & 0.5%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-7
-------
Figure 7
Comparison of IM240 to Second Chance 2500/Idle for
1981 & Newer Carbureted Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 & 12 g/mi Bag 2
Second Chance 2500/Idle Ctpls = 220 ppm & 1.2%
90%
84%
80%
70%
60% —
50%
40%
30%
20%
10% --
0%
5.7%
4.0%
0.0% 0.0%
HCIDR
COEDR
VM Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
2-Spd (220ppm & 1.2%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-8
-------
Figure 8
Comparison of IM240 to Second Chance 2500/Idle for
1981 & Newer Carbureted Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite; 0.5 & 12 g/mi Bag 2
Second Chance 2500/Idle Ctpts = 220 ppm & 1.2% @2500; 100 ppm & 1.0% @Idle
84%
80%
70%
60%
50% -
40%
30% H
20%
10% —
0%
6.9%
0.0% 0.0%
5.7%
HCIDR
COIDR
I/M Fail Rate I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs Normal Emitters
G 2-Spd (220/100ppm & 1.2/1.0%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-9
-------
Figure 9
Comparison of IM240 to Second Chance 2500/Tdle for
1981 & Newer Carbureted Vehicles from Hammond Indiana
IM240 Cutpoints = 0.8 & 15 g/mi Composite: 0.5 & 12 g/mi Bag 2
Second Chance 2500/ldle Ctpts = 100 ppm & 0.5% both modes
90%
83% 83%
84%
80%
70%
60%
50%
40%
30%
20%
10% -1
81%
49%
HCIDR
COIDR
I/M Fail Rate
I/M Fail Rate for FTP I/M Fail Rate for
Passing Vehs _ Normal Emitters
2-Spd(100ppm&0.5%)
IM240 (0.8 & 15 g/mi composite; 0.5 & 12 g/mi Bag 2)
G-10
-------
APPENDIX H
DATA ANALYSES FOR APPENDIX G: 1983 AND NEWER
PFI, TBI, AND CARBURETED VEHICLES
PFI p. H-2
TBI p. H-15
CARB p. H-28
-------
Database Description
The data used to calculate these tables consisted of IM240 Lane, second chance
2500/IdIe, and FTP test scores for 1983 and newer port fuel injected and throttle body fuel
injected vehicles tested at ATL's laboratory in South Bend, Indiana. Carburetor equipped
1981 and newer vehicles are also included, but because no second chance 2500/Idle tests
were performed on the 1981 and 1982 vehicles, we only used the first chance results for
1981-82 carbureted cars. For 1983 and later vehicles, we always used the better results
from the first or second chance tests for all fuel metering types.
Only vehicles receiving each of these tests were used in these analyses. Vehicles
receiving repairs in the interim between the lane tests and the FTP, and vehicles exhibiting
at least a 15% variation between lane dynamometer settings and laboratory dynamometer
settings were excluded from the database.
These differences in dynamometer settings arose from the use of different
dynamometer look-up tables at the lane and lab. Traditional certification tables used for
emission factor testing are too detailed for efficient use in the 1/M lane so simpler tables
were developed for use at the lane. The tables developed for the I/M lane are probably
more representative than the traditional tables because they are derived from more vehicles
with various optional equipment. After becoming aware of the dynamometer setting
differences in the two tables, EPA opted to use the simplified table at both testing sites to
avoid future inconsistencies. The resulting database consisted of 74 1983 and newer PFI
vehicles and 108 1983 and newer TBI vehicles and was used to construct Appendices E,
G, and H . Regressions were not performed for the 122 1981 and newer carbureted
vehicles.
Weighting Factor Explanation
FTP's were performed only on vehicles recruited to the lab for additional testing.
Because the recruiting criteria for vehicles at Hammond lane was based on equal quotas for
low IM240 emitters and high IM240 emitters it did not ensure a sample representative of
the in-use fleet. Weighting factors were used to better simulate a representative sample.
The weighting factors were calculated as the ratio of '83 and newer PFI vehicles tested at
Hammond I/M lane to PFI vehicles tested at the South Bend lab. The weighting factors for
the TBI and carbureted vehicles were calculated similarly. The simulated fleet was then
created by multiplying the lab sample results by the weighting factors.
IM240 Cutpoint Explanation
Appendix H shows the characteristics of the IM240 test when a vehicle is failed
only if its composite emissions exceed 0.8 g/mi HC or 15 g/mi CO and its Bag 2 emissions
alone exceed 0.5 g/mi HC or 12 g/mi CO. In other words, a vehicle can pass by having
low emissions in Bag 2 even if Bag 1 emissions were very high. EPA has found that this
prevents inappropriately failing clean cars with no sacrifice in identification of dirty cars,
and therefore, is proposing this approach to IM240 cutpoints. Tables calculated using this
approach are labeled 'Two Ways to Pass Criteria." Tables calculated using composite
IM240 cutpoints alone are also included for comparison. These Tables are labeled "Full
IM240 Test Criteria". Both tables help illustrate the method of analysis and underlying
sample sizes as well as showing how the characteristics of the IM240 and Idle/2500 tests
vary with more or less stringent cutpoints.
H-l
-------
1983+ Simulated Lane Sample: "Two
PF1 Vehicles
Excess
LaneIM240
Cut-Points
Comp+Baa 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10+0.5/12
0.75/15+0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
0.3/15 +04/12
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10+0.5/12
1.0/15+0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20 + 0.5/12
1.6/10 + 0.5/12
1.6/15+0.5/12
1.6/20+0.5/12
Emissions
Identified
HC
554.8
554.8
553.6
554.8
554.8
553.6
554.8
554.8
553.6
519.7
Slfc?
518.6
519.7
519.7
518.6
490.4
482.0
480.8
472.7
413.6
412.4
411.4
352.3
351.1
411.4
352.3
352.3
351.1
411.4
352.3
339.4
Ways to Pass Criteria"
Excess
Emissions
CO
9889.4
9889.4
9798.5
9889.4
9889.4
9798.5
9889.4
9889.4
9798.5
9715.6
9715.0
9624.8
9715.6
9715.6
9624.8
9639.0
9468.0
9377.1
9592.5
8890.6
8799.8
9268.2
8566.3
8475.5
9268.2
8566.3
8566.3
8475.5
9268.2
8566.3
8243.5
Identified
HC
87%
87%
87%
87%
87%
87%
87%
87%
87%
82%
82%
81%
82%
82%
81%
77%
76%
76%
74%
65%
65%
65%
55%
55%
65%
55%
55%
55%
65%
55%
53%
CO
86%
86%
86%
86%
86%
86%
86%
86%
86%
85%
85%
84%
85%
85%
84%
84%
83%
82%
84%
78%
77%
81%
75%
74%
81%
75%
75%
74%
81%
75%
72%
Number of
Simuiated
Failures
256
256
251
256
256
251
256
256
251
229
229
224
229
229
224
201
174
169
174
119
114
147
92
87
147
92
92
87
147
92
82
Failure
Rate
16%
16%
16%
16%
16%
16%
16%
16%
16%
14%
14%
14%
14%
14%
14%
13%
11%
11%
11%
7%
7%
9%
6%
5%
9%
6%
6%
5%
9%
6%
5%
Number of
Simulated
EC'S
0
0
0
0
0
0
0
0
0
0
O/
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehlcieg
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-2
-------
1983-85 Simulated Lane Sample: "Two Ways to Pass Criteria"
PFI Vehicles
Excess Excess
LaneIM240
Cut-Points
Comp+Baf 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
OJ/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15+0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
0.8/1S +0,5/12.
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10 + 0.5/12
1.0/15 + 0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20+0.5/12
1.6/10 + 0.5/12
1.6/15 + 0.5/12
1.6/20+0.5/12
Emissions
Identlfled
HC
193.2
193.2
192.3
193.2
193.2
192.3
193.2
193.2
192.3
193.2
193.2
192.3
193.2
193.2
192.3
193.2
187.6
186.7
193.2
187.6
186.7
193.2
187.6
186.7
193.2
187.6
187.6
186.7
193.2
187.6
177.6
Jbf missions
CO
4637.1
4637.1
4566.0
4637.1
4637.1
4566.0
4637.1
4637.1
4566.0
4637.1
4637.1
4566.0
4637.1
4637.1
4566.0
4637.1
4524.6
4453.5
4637.1
4524.6
4453.5
4637.1
4524.6
4453.5
4637.1
4524.6
4524.6
4453.5
4637.1
4524.6
4271.7
Identlfled
H£
93%
93%
92%
93%
93%
92%
93%
93%
92%
93%
93%
92%
93%
93%
92%
93%
90%
90%
93%
90%
90%
93%
90%
90%
93%
90%
90%
90%
93%
90%
85%
CQ
95%
95%
94%
95%
95%
94%
95%
95%
94%
95%
95%
94%
95%
95%
94%
95%
93%
91%
95%
93%
91%
95%
93%
91%
95%
93%
93%
91%
95%
93%
88%
Number of
Simulated
Failures
62
62
58
62
62
58
62
62
58
62
62
58
62
62
58
62
44
40
62
44
40
62
44
40
62
44
44
40
62
44
36
Failure
Rate
36%
36%
34%
36%
36%
34%
36%
36%
34%
36%
36%
34%
36%
36%
34%
36%
26%
24%
36%
26%
24%
36%
26%
24%
36%
26%
26%
24%
36%
26%
21%
Number of
Simulated
Esi
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
- 0.0%
0.0%
0.0%
0.0%
0.0%
H-3
-------
1986+ Simulated Lane Sample: "Two Ways to Pass Criteria"
PFI Vehicles
Excess Excess
Lane IM240
Cut-Points
Com^>+Ba0 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15+0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
&8/1S+&S/I2
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10+0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10 + 0.5/12
1.0/15+0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20 + 0.5/12
1.6/10 + 0.5/12
1.6/15+0.5/12
1.6/20+0.5/12
Emissions
Identified
H£
350.0
350.0
350.0
350.0
350.0
350.0
350.0
350.0
350.0
313.3
313.3
313.3
313.3
313.3
313.3
282.5
282.5
282.5
263.8
210.7
210.7
199.5
146.3
146.3
199.5
146.3
146.3
146.3
199.5
146.3
146.3
Emissions
CQ
4831.9
4831.9
4831.9
4831.9
4831.9
4831.9
4831.9
4831.9
4831.9
4649.5
4649.5
4649.5
4649.5
4649.5
4649.5
4569.0
4569.0
4569.0
4520.2
3962.9
3962.9
4179.8
3622.4
3622.4
4179.8
3622.4
3622.4
3622.4
4179.8
3622.4
3622.4
Identified
HC
85%
85%
85%
85%
85%
85%
85%
85%
85%
76%
76% "
76%
76%
76%
76%
69%
69%
69%
64%
51%
51%
48%
36%
36%
48%
36%
36%
36%
48%
36%
36%
CQ
79%
79%
79%
79%
79%
79%
79%
79%
79%
76%
76%
76%
76%
76%
76%
75%
75%
75%
74%
65%
65%
69%
60%
60%
69%
60%
60%
60%
69%
60%
60%
Number of
Simulated
Failures
190
190
190
190
190
190
190
190
190
161
i«i\
161
161
161
161
133
133
133
104
75
75
75
46
46
75
46
46
46
75
46
46
Failure
Rate
13%
13%
13%
13%
13%
13%
13%
13%
13%
11%
11*
11%
11%
11%
11%
9%
9%
9%
7%
5%
5%
5%
3%
3%
5%
3%
3%
3%
5%
3%
3%
Number of
Simulated
EC'S
0
0
0
0
0
0
0
0
0
0
a
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-4
-------
1983+ Simulated Lane Sample
PF1 Vehicles
Idle/2500
Cut-Points
220112
100/1.0-220/1.2
lOO/OJ
Excess
Emissions
Identified
HC
414.3
482.8
487.4
Excess
Emissions
Identified
CQ HC CO
^fcJfc ••«• dbdHb
8316.3 65% 73%
9371.0 76% 82%
9419.7 77% 82%
Number of
Simulated
Failures
188
549
603
Failure
Rate
•MUB
12%
34%
38%
Number of
Simulated EC
EC'S Rate
0 0.0%
82 5.1%
82 5.1%
Failure Rate
for FTP
Passing
Vehicles
0.0%
12.9%
12.9%
1983-1985 Simulated Lane Sample
PFI Vehicles
Excess Excess
Idle/2500
Cut-Points
220/1.2
100/1.0-220/1.2
100/0.5
Emissions
Identified
HC
178.5
181.9
181.9
Emissions
Identified
CO HC CO
4114.2 86% 84%
4189.3 87% 86%
4221.3 87% 87%
Number of
Simulated
Failures
32
68
86
Failure
Rate
19%
40%
51%
Number of
Simulated
EC'S
0
0
0
EC
Rate
0%
0%
0%
Failure Rate
for FTP
Passing
Vehicles
0%
0%
0%
1986+ Simulated Lane Sample
PFI Vehicles
Excess
Excess
ffi ffl^ 3Si Oil^ lls missions
Idle/2500
Cut-Points
220/1.2
100/1.0-220/1.2
100/0.5
Identified
HC
211.6
284.5
289.4
CQ
3714.3
4819.7
4819.7
Identified
HC
51%
69%
70%
CQ
61%
79%
79%
Number of
Simulated
Failures
157
480
508
Failure
Rate
11%
33%
35%
Number of
Simulated
EC'S
0
86
86
EC
Rate
0%
6%
6%
Failure Rate
for FTP
Passing
Vehicles
0%
13%
13%
H-5
-------
Characterize Hammond Indiana 1983+ Test Fleet as of 9/25
Number of
TfichnvivxT Laos CVI244
PR Low
High
Total
Vehicles
Actually
Tested at the
Hammond
LJU&
1505
97
1602
Number of
Vehicles
in the
Lab Sflmole,
55
19
74
Weighting
Factors
for the
Ijib Samnle
27.36
5.11
Number of
Lab Vehicles
Pflssinc FTP
23
1
24
Number of
Normal Lab
Vehicles
Filling FTP
24
2
26
Characterize Hammond Indiana 1983-1985 Test Fleet as of 9/25
Number of
Vehicles
Actually
Tested at the
Hammond
Tf f hllttlVgY I .a fie JM24Q I-flrn*
PFI Low 126
High 44
Total 170
Number of
Vehicles
in the
Lab SftmDle
7
11
18
Weighting
Factors
for the
Lalj §am.p|e,
18.00
4.00
Number of
Lab Vehicles
Passing FTP
1
0
i
Number of
Normal Lab
Vehicles
Failing FTP
5
1
6
Characterize Hammond Indiana 1986+ Test Fleet as of 9/25
Number of
Vehicles
Actually
Tested at the
Hammond
T* ChlWiVKY Lane IM240 Lane
PFI Low 1379
High 53
Total 1432
Number of
Vehicles
in the
Lab Samnle
48
8
56
Weighting
Factors
for the
Lab Samnle
28.73
6.63
Number of
Lab Vehicles
Passim FTP
22
1
23
Number of
Normal Lab
Vehicles
Failing FTP
19
1
20
H-6
-------
Characterize Simulated 1983+ In-Use Fleet
Calculated From the Lab sample and Weighting Factors
Technnlovv
PFI
Number of
Vehicles
Actually
Tested at the
Hammond
Lane
1602
Number of
Simulated
Vehicles Total Excess Emissions
Passing In Simulated Sample
FTPs HC CO
634 636.57 11441.42
Characterize Simulated 1983-1985 In-Use Fleet
Calculated
From the Lab sample and
Number of
Vehicles
Actually
Tested at the
Hammond
Technology LjlUfi
PFI
170
Weighting Factors
Number of
Simulated
Vehicles
Passing
FTPs
18
Total Excess Emissions
In Simulated Sample
H£
208.16
CQ
4874.20
Characterize Simulated 1986+ In-Use Fleet
Calculated From the Lab sample and Weighting Factors
Technology
PFI
Number of
Vehicles
Actually
Tested at the
Hammond
Laos
1432
Number of
Simulated
Vehicles
Passing
FTPs
639
Total Excess Emissions
In Simulated Sample
H£ CO
412.09 6083.07
H-7
-------
Note: FTP Category Definitions
Normal Emitters: FTP HC<0.82 and FTP CO<10.2
High Emitters: FTP HC>0.82 or FTP CO>10.2
Very High Emitters: FTP HCSi.64 or FTP CO>13.6
Super Emitters: FTP HC^IO.O or FTP CO>150.0
Distribution of Total Excess Emissions by FTP Emitter Categories
1983+ PFI Sample
Simulated Lane Sample
Emissions
HC (g/mi)
CO (g/mi)
Total
Excess from
Normals
9%
9%
Total.
Excess from
Hiyhs
13%
7%
Total
Excess from
VM-T Hlphs
49%
40%
Total
Excess from
Supers
30%
43%
Total Excess
100%
100%
1983- 1985 PFI Sample
Simulated Lane Sample
Total
Excess from
Emissions Normals
HC(g/mi) 6%
CO (g/mi) 5%
Total Total Total
Excess from Excess from Excess from
Highs YffY Hl?hs Supers Total Excess
4% 43% 47% 100%
3% 45% 48% 100%
1986+ PFI Sample
Simulated Lane Sample
Total
Excess from
RmtwIniM Normals
HC(g/mi) 10%
CO (g/mi) 12%
Total Total Total
Excess from Excess from Excess from
Hlffhs VM-T Highs Supers Total Excess
17% 53% 20% 100%
10% 34% 43% 100%
H-8
-------
Distribution of Emissions Identified by FTP Emitter Categories
IM240 1983+ PFI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
Contp 4. Rag J
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15+0.5/12
0.8/15+0.5/12
0.6/10 + 0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
0%
0%
2%
2%
2%
2%
Hl»h«
MAvMtf
0%
0%
0%
0%
0%
0%
9%
1%
8%
1%
V«T Highs
44%
40%
46%
42%
54%
44%
53%
46%
56%
47%
Supers
56%
60%
54%
58%
46%
56%
37%
51%
34%
50%
IM240 1983-1985 PFI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
C. OIDD + Bflf 2
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15 + 0.5/12
0.8/15+0.5/12
0.6/10 + 0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
0%
0%
3%
3%
3%
3%
Highs
mff^ffm
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
Very Highs
44%
46%
47%
49%
47%
49%
46%
48%
46%
48%
Supers
•Mll|UU>B
56%
54%
53%
51%
53%
51%
51%
50%
51%
50%
IM240 1986+ PFI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
Comp *. Bag 2
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15 + 0.5/12
0.8/15 + 0.5/12
0.6/10 + 0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
Highs
0%
0%
0%
0%
0%
0%
16%
3%
14%
3%
Very Highs
43%
28%
43%
28%
61%
34%
58%
41%
62%
43%
Supers
57%
72%
57%
72%
39%
66%
26%
56%
24%
54% •
H-9
-------
Distribution of Emissions Identified by FTP Emitter Categories
Idle/2500
Cut-point
HC/CO
220/1.2
100/1.0-220/1.2
100/0.5
1983+ PFI Simulated Lane Sample
Emission , Normals Hlyh.i
HC 0% 9%
CO 0% 4%
HC 5% 11%
CO 6% 4%
HC 6% 11%
CO 6% 4%
Very Highs
45%
36%
44%
37%
44%
37%
Supers
46%
60%
39%
53%
39%
53%
Idle/2500 1983-1985 PFI Simulated Lane Sample
Cut-point
HC/CO
220/1.2
100/1.0-220/1.2
100/0.5
Emission
HC
CO
HC
CO
HC
CO
Normals
0%
0%
2%
2%
2%
3%
Hlyhs
0%
0%
0%
0%
0%
0%
Very Hiahs
45%
44%
44%
43%
44%
43%
Supers
55%
56%
54%
55%
54%
55%
Idle/2500 1986+ PFI Simulated Lane Sample
Cut-point
HC/CO
220/1.2
100/1.0-220/1.2
100/0.5
Rmlssfnn
HC
CO
HC
CO
HC
CO
Normals
0%
1%
7%
10%
9%
10%
Highs.
18%
9%
20%
8%
20%
8%
Very High*
42%
20%
43%
28%
43%
28%
Supers
39%
70%
29%
54%
29%
54%
H-10
-------
1983+ Simulated Lane Sample: "Full IM240 Test Criteria"
PFI Vehicles
Excess Excess
LaneIM240
rut-Polnts
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
0.8/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.600
Emissions
Identified
H£
568.2
568.2
567.0
568.2
568.2
567.0
560.0
560.0
558.8
519.7
sm? "
518.6
519.7
519.7
518.6
490.4
482.0
480.8
472.7
413.6
412.4
411.4
352.3
351.1
411.4
352.3
352.3
351.1
411.4
352.3
339.4
Emissions
CQ
10032.7
9984.0
9893.2
10032.7
9984.0
9893.2
9961.9
9913.2
9822.3
9764.3
9715.0
9624.8
9764.3
9715.6
9624.8
9687.7
9468.0
9377.1
9641.2
8890.6
8799.8
9316.9
8566.3
8475.5
9316.9
8566.3
8566.3
8475.5
9316.9
8566.3
8243.5
Identified
HC
89%
89%
89%
89%
89%
89%
88%
88%
88%
82%
" $m
81%
82%
82%
81%
77%
76%
76%
74%
65%
65%
65%
55%
55%
65%
55%
55%
55%
65%
55%
53%
£Q
88%
87%
86%
88%
87%
86%
87%
87%
86%
85%
85%'
84%
85%
85%
84%
85%
83%
82%
84%
78%
77%
81%
75%
74%
81%
75%
75%
74%
81%
75%
72%
Number of
Simulated
Failure!!
398
371
366
398
371
366
343
316
306
261
234-
224
261
234
224
234
179
169
206
124
114
179
97
87
179
97
92
87
179
97
82
Failure
Rate
25%
23%
23%
25%
23%
23%
21%
20%
19%
16%
x 15%- ,
14%
16%
15%
14%
15%
11%
11%
13%
8%
7%
11%
6%
5%
11%
6%
6%
5%
11%
6%
5%
Number of
Simulated
EC'S
32
32
32
32
32
32
*
32
32
27
5
5 ,
0
5
5
0
5
5
0
5
5
0
5
5
0
5
5
0
0
5
5
0
EC
Rate
2.0%
2.0%
2.0%
2.0%
2.0%
2.0%
2.0%
2.0%
1.7%
0.3%
0,3%
0.0%
0.3%
0.3%
0.0%
0.3%
0.3%
0.0%
0.3%
0.3%
0.0%
0.3%
0.3%
0.0%
0.3%
0.3%
0.0%
0.0%
0.3%
0.3%
0.0%
Failure Rate
for FTP
Passing
Vehicles
5.1%
5.1%
5.1%
5.1%
5.1%
5.1%
5.1%
5.1%
4.3%
0.8%
0,8%
0.0%
0.8%
0.8%
0.0%
0.8%
0.8%
0.0%
0.8%
0.8%
0.0%
0.8%
0.8%
0.0%
0.8%
0.8%
0.0%
0.0%
0.8%
0.8%
0.0%
H-11
-------
1983.1985 Simulated Lane Sample: "Full IM240 Test Criteria"
PFI Vehicles
Excess Excess
Emissions Emissions Number of
LaneIM240 Identified Identified Simulated
Cut-Points
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
0.8/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
H£
196.6
196.6
195.7
196.6
196.6
195.7
*
196.6
196.6
195.7
193.2
193.2
192.3
193.2
193.2
192.3
193.2
187.6
186.7
193.2
187.6
186.7
193.2
187.6
186.7
193.2
187.6
187.6
186.7
193.2
187.6
177.6
CO
4684.8
4652.8
4581.6
4684.8
4652.8
4581.6
4684.8
4652.8
4581.6
4669.2
4637-1.
4566.0
4669.2
4637.1
4566.0
4669.2
4524.6
4453.5
4669.2
4524.6
4453.5
4669.2
4524.6
4453.5
4669.2
4524.6
4524.6
4453.5
4669.2
4524.6
4271.7
ac
94%
94%
94%
94%
94%
94%
94%
94%
94%
93%
93%
92%
93%
93%
92%
93%
90%
90%
93%
90%
90%
93%
90%
90%
93%
90%
90%
90%
93%
90%
85%
co
96%
95%
94%
96%
95%
94%
96%
95%
94%
96%
95%
94%
96%
95%
94%
96%
93%
91%
96%
93%
91%
96%
93%
91%
96%
93%
93%
91%
96%
93%
88%
Failures
98
80
76
98
80
76
98
80
76
80
62
58
80
62
58
80
44
40
80
44
40
80
44
40
80
44
44
40
80
44
36
Failure
Rate
58%
47%
45%
58%
47%
45%
58%
47%
45%
47%
36%
34%
47%
36%
34%
47%
26%
24%
47%
26%
24%
47%
26%
24%
47%
26%
26%
24%
47%
26%
21%
Number of
Simulated EC
Ecja
0
0
0
0
0
0
0
0
0
0
tf
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0,0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-12
-------
1986+ Simulated Lane Sample: "Full EVI240 Test Criteria"
PFI Vehicles
Excess Excess
LaneIM240
::nt- Points
0.6/10
0.6/15
0.600
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
0.8/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
Emissions
Identified
HE
358.7
358.7
358.7
358.7
358.7
358.7
350.0
350.0
350.0
313.3
313.3
313.3
313.3
313.3
313.3
282.5
282.5
282.5
263.8
210.7
210.7
199.5
146.3
146.3
199.5
146.3
146.3
146.3
199.5
146.3
146.3
Emissions
£2
4906.3
4906.3
4906.3
4906.3
4906.3
4906.3
4831.9
4831.9
4831.9
4649.5
4649,5
4649.5
4649.5
4649.5
4649.5
4569.0
4569.0
4569.0
4520.2
3962.9
3962.9
4179.8
3622.4
3622.4
4179.8
3622.4
3622.4
3622.4
4179.8
3622.4
3622.4
Identified
HE
87%
87%
87%
87%
87%
87%
85%
85%
85%
76%
76%
76%
76%
76%
76%
69%
69%
69%
64%
51%
51%
48%
36%
36%
48%
36%
36%
36%
48%
36%
36%
£tt
81%
81%
81%
81%
81%
81%
79%
79%
79%
76%
76%
76%
76%
76%
76%
75%
75%
75%
74%
65%
65%
69%
60%
60%
69%
60%
60%
60%
69%
60%
60%
Number of
Simulated
Failures
283
283
283
283
283
283
225
225
219
168
tm
161
168
168
161
139
139
133
110
82
75
82
53
46
82
53
46
46
82
53
46
Failure
Rate
20%
20%
20%
20%
20%
20%
16%
16%
15%
12%
12&
11%
12%
12%
11%
10%
10%
9%
8%
6%
5%
6%
4%
3%
6%
4%
3%
3%
6%
4%
3%
Number of
Simulated
EC'S
35
35
35
35
35
35
35
35
29
7
7
0
7
7
0
7
7
0
7
7
0
7
7
0
7
7
0
0
7
7
0
EC
Rate
2.5%
2.5%
2.5%
2.5%
2.5%
2.5%
2.5%
2.5%
2.0%
0.5%
0.5%
0.0%
0.5%
0.5%
0.0%
0.5%
0.5%
0.0%
0.5%
0.5%
0.0%
0.5%
0.5%
0.0%
0.5%
0.5%
0.0%
0.0%
0.5%
0.5%
0.0%
Failure Rate
for FTP
Passing
Vehicles
5.5%
5.5%
5.5%
5.5%
5.5%
5.5%
*
5.5%
5.5%
4.5%
1.0%
1.0%
0.0%
1.0%
1.0%
0.0%
1.0%
1.0%
0.0%
1.0%
1.0%
0.0%
1.0%
1.0%
0.0%
1.0%
1.0%
0.0%
0.0%
1.0%
1.0%
0.0%
H-13
-------
Distribution of Emissions Identified by FTP Emitter Categories
IM240
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
1983+ PFI Simulated Lane Sample: "Full IM240 Test Criteria"
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
0%
0%
2%
2%
4%
•3%
Highs
0%
0%
0%
0%
0%
0%
9%
1%
8%
1%
Very Highs
44%
40%
46%
42%
54%
44%
53%
46%
54%
46%
Supers
56%
60%
54%
58%
46%
56%
37%
51%
33%
49%
IM240
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
1983-1985 PFI Simulated Lane Sample: "Full IM240 Test Criteria"
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
0%
0%
3%
3%
5%
4%
Highs
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
VM-T Highs
44%
46%
47%
49%
47%
49%
46%
48%
45%.
47%
Supers
56%
54%
53%
51%
53%
51%
51%
50%
50%
49%
1M240
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
1986+ PFI Simulated
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Lane Sample:
"Full IM240 Test Criteria"
NjjrjUflis Hjghs Vwv Highs
0%
0%
0%
0%
0%
0%
0%
0%
2%
2%
0%
0%
0%
0%
0%
0%
16%
3%
14%
3%
43%
28%
43%
28%
61%
34%
58%
41%
61%
43%
Supers
57%
72%
57%
72%
39%
66%
26%
56%
23%
53%
H-14
-------
1983+ Simulated Lane Sample: "Two Ways to Pass Criteria"
TBI Vehicles
Excess Excess
LaneIM240 Emissions Emissions Number of
Cut-Points
Comn+Bav 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15+0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
0.8/15+0.5/12
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20+0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10+0.5/12
1.0/15+0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20 + 0.5/12
1.6/10+0.5/12
1.6/15+0.5/12
1.600+0.5/12
Identified
HC
761.7
754.0
753.5
754.8
747.2
723.8
754.8
747.2
723.8
754.8
740.4,
717.0
754.8
740.4
717.0
750.1
735.7
712.3
722.9
697.1
673.8
698.4
578.5
547.3
698.4
578.5
564.8
547.3
684.7
539.5
501.0
Identified
CO
10334.1
9949.6
9939.4
10282.4
9897.9
9077.9
10282.4
9897.9
9077.9
10282.4
97«&5
8943.4
10282.4
9763.5
8943.4
10273.0
9754.1
8934.1
10042.5
9366.4
8546.4
9938.1
8451.9
7482.2
9938.1
8451.9
7959.7
7482.2
9825.6
8188.8
7108.0
HC
•AAfc
92%
91%
91%
91%
90%
87%
91%
90%
87%
91%
89%
87%
91%
89%
87%
91%
89%
86%
87%
84%
81%
84%
70%
66%
84%
70%
68%
66%
83%
65%
61%
84%
81%
81%
84%
81%
74%
84%
81%
74%
84%
80%
73%
84%
80%
73%
84%
80%
73%
82%
77%
70%
81%
69%
61%
81%
69%
65%
61%
80%
67%
58%
Simulated
Failures
445
403
398
424
381
346
424
381
346
424
360
325
424
360
325
403
339
303
381
296
261
360
168
128
360
168
147
128
355
145
100
Failure
Rate
26%
23%
23%
25%
22%
20%
25%
22%
20%
25%
21%
19%
25%
21%
19%
23%
20%
18%
22%
17%
15%
21%
10%
7%
21%
10%
9%
7%
21%
8%
6%
Number of
Simulated
EC'S
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-15
-------
1983-85 Simulated Lane Sample: "Two Ways to Pass Criteria"
TBI Vehicles
Lane IM240
Cut-Points
Comp+Baff 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15 + 0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
OS/13 +OS/J2.
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10 + 0.5/12
1.0/15+0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20 + 0.5/12
1.6/10 + 0.5/12
1.6/15+0.5/12
1.6/20 + 0.5/12
Admissions
Identified
HC
359.9
359.7
359.7
355.1
355.0
338.3
355.1
355.0
338.3
355.1
3S&3
333.6
355.1
350.3
333.6
351.9
347.0
330.4
351.9
339.2
322.5
351.9
286.4
269.8
351.9
286.4
277.0
269.8
335.6
239.9
223.2
Emissions
Identified
CO
5740.5
5661.0
5661.0
5704.6
5625.1
4958.1
5704.6
5625.1
4958.1
5704.6
5SBL9
4864.8
5704.6
5531.9
4864.8
5698.1
5525.4
4858.3
5698.1
5416.4
4749.3
5698.1
4972.7
4305.6
5698.1
4972.7
4631.3
4305.6
5563.7
4658.4
3991.3
HC
90%
90%
90%
89%
89%
84%
89%
89%
84%
89%
87%,
83%
89%
87%
83%
88%
87%
82%
88%
85%
80%
88%
71%
67%
88%
71%
69%
67%
84%
60%
56%
CO
87%
86%
86%
87%
86%
75%
87%
86%
75%
87%
84%
74%
87%
84%
74%
87%
84%
74%
87%
82%
72%
87%
76%
65%
87%
76%
70%
65%
85%
71%
61%
Number of
Simulated
Failures
244
229
229
229
215
194
229
215
194
229
200
179
229
200
179
215
185
165
215
170
150
215
111
91
215
111
96
91
209
83
62
Failure
Rate
44%
41%
41%
41%
38%
35%
*
41%
38%
35%
41%
•• ••
36%
32%
41%
36%
32%
38%
33%
29%
38%
30%
27%
38%
20%
16%
38%
20%
17%
16%
37%
15%
11%
Number of
Simulated
EC'S
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rafa
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
00%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-16
-------
198«+ Simulated Lane Sample: "Two Ways to Pass Criteria"
TB1 Vehicles
Excess Excess
LaneIM240
Cut-Points
Comp+Baf 2
0.6/10 + 0.5/12
0.6/15+0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15+0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15+0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
0.8/15+0.3/12.
0.8/20 + 0.5/12
0.85/10 + 0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10 + 0.5/12
1.0/15 +0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16+0.5/12
1.2/20 + 0.5/12
1.6/10 + 0.5/12
1.6/15+0.5/12
1.6/20 + 0.5/12
Emissions
Identified
HC
359.1
350.0
349.6
359.1
350.0
347.1
359.1
350.0
347.1
359.1
'- 33Q&,
347.1
359.1
350.0
347.1
359.1
350.0
347.1
325.7
316.6
313.6
295.7
264.3
255.2
295.7
264.3
264.3
255.2
295.7
264.3
249.4
Emissions
Identified
CO
dKdb
4191.7
3860.8
3852.7
4191.7
3860.8
3817.0
4191.7
3860.8
3817.0
4191.7
"awe*-
3817.0
4191.7
3860.8
3817.0
4191.7
3860.8
3817.0
3909.1
3578.2
3534.4
3781.1
3241.3
3078.6
3781.1
3241.3
3241.3
3078.6
3781.1
3241.3
2990.4
HC
••Afe
98%
95%
95%
98%
95%
94%
98%
95%
94%
98%
95%
94%
98%
95%
94%
98%
95%
94%
89%
86%
85%
80%
72%
69%
80%
72%
72%
69%
80%
72%
68%
CQ
JUfc
83%
76%
76%
83%
76%
76%
83%
76%
76%
83%
, 76% <
76%
83%
76%
76%
83%
76%
76%
77%
71%
70%
75%
64%
61%
75%
64%
64%
61%
75%
64%
59%
Number of
Simulated
Failures
^Jj^f|mg£
157
131
127
157
131
124
157
131
124
157
111
124
157
131
124
157
131
124
131
105
97
105
53
41
105
53
53
41
105
53
38
Failure
Rate
AJHJUI
14%
11%
11%
14%
11%
11%
14%
11%
11%
14%
- 11*- '
11%
14%
11%
11%
14%
11%
11%
11%
9%
8%
9%
5%
4%
9%
5%
5%
4%
9%
5%
3%
Number of
Simulated
geis
0
0
0
0
0
0
0
0
0
0
f>>- '
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
AHUk
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
oo*.
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
^AHUUMt
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-17
-------
1983+ Simulated Lane Sample
TBI Vehicles
Excess Excess
Idle/2500
Cut-Points
220/1.2
100/1.0+220/1.2
100/0.5
Emissions Emissions
Identified Identified
HC CO HC CQ
591.6 8196.5 71% 67%
745.5 9947.1 90% 81%
746.7 10039.1 90% 82%
Number of
Simulated
Failures
249
625
715
Failure
Rate
14%
36%
42%
Number of
Simulated
EC'S
21
64
111
EC
Rate
1.2%
3.7%
6.5%
Failure Rate
for FTP
Passing
Vehicles
3.9%
11.6%
20.2%
1983-1985 Simulated Lane Sample
TBI Vehicles
Excess
Idle/2500
Cut-Points
220/1.2
100/1.0+220/1.2
100/0.5
Emissions
Identified
HC
289.7
375.4
376.3
Excess
Emissions
01
5007.8
6038.5
6070.0
Identified
HC
72%
94%
94%
SSL
76%
92%
92%
Nnml>er of
Simulated
Failures
138
347
362
Failure
Rate
25%
62%
65%
Number of
Simulated
EC'S
0
0
0
EC
Rate
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
1986+ Simulated Lane Sample
TBI Vehicles
Excess
Idle/2500
Cut-Points
220/1.2
100/1.04-220/1.2
100/0.5
Emissions
Identified
HC CO
272.9 3051.9
328.5 3543.5
328.5 3600.7
Excess
Emissions
Identified
HC CO
74% 60%
89% 70%
89% 71%
Number of
Simulated
Failures
105
209
291
Failure
Rate
9%
18%
25%
Number of
Simulated
EC'S
26
78
134
EC
Rate
2.2%
6.7%
11.6%
Failure Rate
for FTP
Passing
Vehicles
4.7%
14.0%
24.0%
H-18
-------
Characterize Hammond Indiana 1983+ Test Fleet as of 9/25
Number of
T«ehnology Lane IM240
TBI Low
High
Total
• Vehicles
Actually
Tested at the
Hammond
Lans
1555
166
1721
Number of
Vehicles
in the
73
35
108
Weighting
Factors
for the
Lab Sample
21.30
4.74
Number of
Lab Vehicles
Pawlny FTP
25
4
29
Number of
Normal Lab
Vehicles
Failing FT^
32
6
38
Characterize Hammond Indiana 1983-1985 Test Fleet as of 9/25
Number of
Vehicles
Actually
Tested at the
Hammond
TtcfaovlvgY Lant IM24Q i^ane
TBI Low 458
High 102
Total 560
Number of
Vehicles
in the
Lab Sample
31
18
49
Weighting
Factors
for the
Lab SfimDle
14.77
5.67
Number of
Lab Vehicles
Passing FTP
4
1
5
Number of
Normal Lab
Vehicles
Falling FTP
16
0
16
Characterize Hammond Indiana 1986+ Test Fleet as of 9/25
Number of
Vehicles
Actually
Tested at the
Hammond
Ttchnnjoyy Lane IM240 t.ane
TBI Low 1097
High 64
Total 1161
Number of
Vehicles
In the
Lab Sample
42
17
59
Weighting
Factors
for the
Lab Sample
26.12
3.76
Number of
Lab Vehicles
Passing FTP
21
3
24
Number of
Normal Lab
Vehicles
Falling FTP
16
6
22
H-19
-------
Characterize Simulated 1983+ In-Use Fleet
Calculated From the Lab sample and Weighting Factors
Technology
TBI
Number of
Vehicles
Actually
Tested at the
Hammond
Ljujg
1721
Number of
Simulated
Vehicles
Passing
FTPs
552
Total Excess Emissions
In Simulated Sample
B£
827.69
CO,
12239.31
Characterize Simulated 1983-1985 In-Use Fleet
Calculated From the Lab sample and Weighting Factors
Technology
TBI
Number of
Vehicles
Actually
Tested at the
Hammond
LJUB.
560
Number of
Simulated
Vehicles Total Excess Emissions
Passing In Simulated Sample
FTPs HC CO
65 400.94 6575.68
Characterize Simulated 1986+ In-Use Fleet
Calculated From the Lab sample and Weighting Factors
Technology
TBI
Number of
Vehicles
Actually
Tested at the
Hammond
Lao*
1161
Number of
Simulated
Vehicles
FTPs
560
Total Excess Emissions
In Simulated Sample
HC CQ
367.49
5051.31
H-20
-------
Note: FTP Category Definitions
Normal Emitters: FTP HC<0.82 and FTP CO<10.2
High Emitters: FTP HC»0.82 or FTP COS10.2
Very High Emitters: FTP HC>1.64 or FTP COS13.6
Super Emitters: FTP HC^IO.O or FTP COS150.0
Distribution of Total Excess Emissions by FTP Emitter Categories
1983+ TBI Sample
Simulated Lane Sample
Total
Excess from
F.mlgrfons
HC(g/mi)
C0(g/mi)
Nqrrfla's
6%
11%
Total
Excess from
Highs
18%
15%
Total
Excess from
YflT Hfghg
31%
34%
Total
Excess from
SiiDers
45%
39%
Total Eix.ce.ss
100%
100%
1983-1985 TBI Sample
Simulated Lane Sample
Total
Excess from
EjoissioQs,
HC(g/mi)
CO(g/mi)
Normals
8%
8%
Total
Excess from
Mff
19%
16%
Total
Excess from
VHT HlrfM
49%
46%
Total
Excess from
Supers
24%
31%
To(a|| Excel"1
100%
100%
f
I98t> TBI Sample
Simulated Lane Sample
Total
Excess from
Ejoissisuu
HC(g/mi)
CO (g/mi)
Normals
2%
15%
Total
Excess from
Hlchs
16%
11%
Total
Excess from
Very Highs
20%
25%
Total
Excess from
Supers
62%
49%
Total Excess
100%
100%
H-21
-------
Distribution of Emissions Identified by FTP Emitter Categories
1M 240 1983+ TBI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
Como + Bay 2
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15+0.5/12
0.8/15+0.5/12
0.6/10+0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
1%
1%
1%
1%
1%
2%
2%
2%
4%
5%
Highs
1%
1%
3%
2%
14%
7%
14%
9%
14%
8%
Very Highs
24%
30%
31%
36%
32%
39%
34%
40%
34%
40%
Supers
74%
68%
66%
60%
53%
51%
50%
49%
49%
47%
IM240 1983-1985 TBI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
Comp + Bag 2
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15 + 0.5/12
0.8/15+0.5/12
0.6/10 + 0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
1%
1%
2%
1%
5%
5%
Hlyhs
4%
2%
6%
4%
12%
6%
14%
8%
14%
8%
Very Highs
52%
47%
58%
53%
58%
56%
56%
55%
55%
53%
Supers
44%
50%
35%
43%
29%
37%
28%
36%
27%
35%
IM240 198«+ TBI Simulated Lane Sample: "Two Ways to Pass Criteria"
Cut-Point
Oomn 4- Ran 2
1.6/20 + 0.5/12
1.2/16 + 0.5/12
1.0/15+0.5/12
0.8/15 + 0.5/12
0.6/10 + 0.5/12
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
1%
2%
1%
2%
1%
2%
1%
2%
1%
2%
Highs
0%
0%
1%
1%
17%
10%
16%
10%
15%
9%
Very Highs
7%
15%
11%
20%
9%
18%
18%
24%
20%
30%
Supers
92%
83%
87%
77%
72%
69%
65%
64%
64%
59%
H-22
-------
Distribution of Emissions Identified by FTP Emitter Categories
Idle/2500 1983+ TBI Simulated Lane Sample
Cut-point
HC/CO Emission Normals
220/1.2
100/1.0-220/1.2
100/0.5
HC
CO
HC
CO
HC
CO
1%
2%
4%
5%
4%
6%
Highs
11%
5%
16%
11%
16%
11%
Very Highs
30%
43%
35%
42%
34%
42%
Supers
57%
50%
45%
41%
45%
41%
Idle/2500 1983-1985 TBI Simulated Lane Sample
Cut-point
HC/CO
220/1.2
100/1.0-220/1.2
100/0.5
Emission
HC
CO
HC
CO
HC
CO
Normals
2%
2%
5%
6%
6%
6%
Hlohi
11%
5%
16%
11%
16%
11%
Very Hlgha
53%
53%
52%
50%
52%
50%
Supers
34%
40%
26%
33%
26%
33%
Idle/2500 1986+ TBI Simulated Lane Sample
Cut-point
HC/CO Emission Normals
220/1.2
100/1.0-220/1.2
100/0.5
HC
CO
HC
CO
HC
CO
0%
1%
0%
1%
0%
2%
Highs
11%
4%
16%
10%
16%
9%
Vtrv Hlgha
14%
32%
22%
36%
22%
35%
Supers
75%
63%
62%
54%
62%
53%
H-23
-------
1983+ Simulated Lane Sample: "Full 1M240 Test Criteria"
TBI Vehicles
Excess Excess
LaneIM240
Cut-Points
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
O.S/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
Emissions
Identified
HC
774.9
767.2
766.7
754.8
747.2
723.8
754.8
747.2
723.8
754.8
740,4
717.0
754.8
740.4
717.0
750.1
735.7
712.3
722.9
697.1
673.8
698.4
578.5
547.3
698.4
578.5
564.8
547.3
684.7
539.5
501.0
Emissions
Identified
CQ
10550.8
10103.6
10093.4
10345.0
9897.9
9077.9
10345.0
9897.9
9077.9
10345.0
9763,5,
8943.4
10345.0
9763.5
8943.4
10335.6
9754.1
8934.1
10105.1
9366.4
8546.4
10000.7
8451.9
7482.2
10000.7
8451.9
7959.7
7482.2
9888.2
8188.8
7108.0
HC
94%
93%
93%
91%
90%
87%
91%
90%
87%
91%
89%"
87%
91%
89%
87%
91%
89%
86%
87%
84%
81%
84%
70%
66%
84%
70%
68%
66%
83%
65%
61%
CQ
86%
83%
82%
85%
81%
74%
85%
81%
74%
85%
80%
73%
85%
80%
73%
84%
80%
73%
83%
77%
70%
82%
69%
61%
82%
69%
65%
61%
81%
67%
58%
Number of
Simulated
Failures
528
464
455
464
400
360
464
400
360
464
379
334
464
379
334
443
358
313
422
315
270
400
187
138
400
187
166
138
396
164
104
Failure
Rate
31%
27%
26%
27%
23%
21%
27%
23%
21%
27%
22%
19%
27%
22%
19%
26%
21%
18%
24%
18%
16%
23%
11%
8%
23%
11%
10%
8%
23%
10%
6%
Number of
Simulated
EC'S
57
40
36
19
19
14
19
19
14
19
, ***
9
19
19
9
19
19
9
19
19
9
19
19
9
19
19
19
9
19
19
5
EC
Rate
3.3%
2.3%
2.1%
1.1%
1.1%
0.8%
1.1%
1.1%
0.8%
1.1%
14%
0.6%
1.1%
1.1%
0.6%
1.1%
1.1%
0.6%
1.1%
1.1%
0.6%
1.1%
1.1%
0.6%
1.1%
1.1%
1.1%
0.6%
1.1%
1.1%
0.3%
Failure Rate
for FTP
Passing
Vehicles
10.3%
7.3%
6.4%
3.4%
3.4%
2.6%
3.4%
3.4%
2.6%
3.4%
3A%
1.7%
3.4%
3.4%
1.7%
3.4%
3.4%
1.7%
3.4%
3.4%
1.7%
3.4%
3.4%
1.7%
3.4%
3.4%
3.4%
1.7%
3.4%
3.4%
0.9%
H-24
-------
1983-1985 Simulated Lane Sample: "Full IM240 Test Criteria"
TBI Vehicles
Excess Excess
LaneIM240
::ut. Points
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
0.8/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
Emissions
Identified
HC
369.0
368.9
368.9
355.1
355.0
338.3
355.1
355.0
338.3
355.1
330.3^
333.6
355.1
350.3
333.6
351.9
347.0
330.4
351.9
339.2
322.5
351.9
286.4
269.8
351.9
286.4
277.0
269.8
335.6
239.9
223.2
Emissions
£Q
5847.3
5767.8
5767.8
5704.6
5625.1
4958.1
5704.6
5625.1
4958.1
5704.6
5S3L9
4864.8
5704.6
5531.9
4864.8
5698.1
5525.4
4858.3
5698.1
5416.4
4749.3
5698.1
4972.7
4305.6
5698.1
4972.7
4631.3
4305.6
5563.7
4658.4
3991.3
Identified
HC
92%
92%
92%
89%
89%
84%
89%
89%
84%
89%
87%
83%
89%
87%
83%
88%
87%
82%
88%
85%
80%
88%
71%
67%
88%
71%
69%
67%
84%
60%
56%
£Q
89%
88%
88%
87%
86%
75%
87%
86%
75%
87%
84%
74%
87%
84%
74%
87%
84%
74%
87%
82%
72%
87%
76%
65%
87%
76%
70%
65%
85%
71%
61%
Number of
Simulated
Failures
279
265
265
235
220
200
235
220
200
235
205
185
235
205
185
220
191
170
220
176
155
220
117
96
220
117
102
96
215
88
62
Failure
Rate
50%
47%
47%
42%
39%
36%
42%
39%
36%
42%
37%
33%
42%
37%
33%
39%
34%
30%
39%
31%
28%
39%
21%
17%
39%
21%
18%
17%
38%
16%
11%
Number of
Simulated
EC'S
20
20
20
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
0
EC
Rate
3.7%
3.7%
3.7%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
1.0%
0.0%
Failure Rate
for FTP "
Passing
Vehicles
31.6%
31.6%
31.6%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
8.7%
0.0%
H-25
-------
1986+ Simulated Lane Sample: "Full IM240 Test Criteria"
TBI Vehicles
Excess Excess
LaneIM240
Cut-Polnts
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
(WJ3
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
Emissions
Identified
HC
359.1
350.0
349.6
359.1
350.0
347.1
359.1
350.0
347.1
359.1
33ftO
347.1
359.1
350.0
347.1
359.1
350.0
347.1
325.7
316.6
313.6
295.7
264.3
255.2
295.7
264.3
264.3
255.2
295.7
264.3
249.4
Emissions
Identified
CO
4268.5
3860.8
3852.7
4268.5
3860.8
3817.0
4268.5
3860.8
3817.0
4268.5
3860.8
3817.0
4268.5
3860.8
3817.0
4268.5
3860.8
3817.0
3985.9
3578.2
3534.4
3857.9
3241.3
3078.6
3857.9
3241.3
3241.3
3078.6
3857.9
3241.3
2990.4
HC
98%
95%
95%
98%
95%
94%
98%
95%
94%
98%
95%
94%
98%
95%
94%
98%
95%
94%
89%
86%
85%
80%
72%
69%
80%
72%
72%
69%
80%
72%
68%
CO
85%
76%
76%
85%
76%
76%
85%
76%
76%
85%
76%
76%
85%
76%
76%
85%
76%
76%
79%
71%
70%
76%
64%
61%
76%
64%
64%
61%
76%
64%
59%
Number of
Simulated
Failures
195
142
135
195
142
131
195
142
131
195
142
127
195
142
127
195
142
127
168
116
101
142
64
45
142
64
64
45
142
64
41
Failure
Rate
17%
12%
12%
17%
12%
11%
17%
12%
11%
17%
22%-
11%
17%
12%
11%
17%
12%
11%
15%
10%
9%
12%
6%
4%
12%
6%
6%
4%
12%
6%
4%
Number of
Simulated
EC'S
11
11
8
11
11
8
11
11
8
11
11
4
11
11
4
11
11
4
11
11
4
11
11
4
11
11
11
4
11
11
4
EC
Rate
1.0%
1.0%
0.6%
1.0%
1.0%
0.6%
1.0%
1.0%
0.6%
1.0%
1.0%
0.3%
1.0%
1.0%
0.3%
1.0%
1.0%
0.3%
1.0%
1.0%
0.3%
1.0%
1.0%
0.3%
1.0%
1.0%
1.0%
0.3%
1.0%
1.0%
0.3%
Failure Rate
for FTP
Passing
Vehicles
2.0%
2.0%
1.3%
2.0%
2.0%
1.3%
2.0%
2.0%
1.3%
2.0%
um
0.7%
2.0%
2.0%
0.7%
2.0%
2.0%
0.7%
2.0%
2.0%
0.7%
2.0%
2.0%
0.7%
2.0%
2.0%
2.0%
0.7%
2.0%
2.0%
0.7%
H-26
-------
Distribution of Emissions Identified by FTP Emitter Categories
IM240 1983+ TBI Simulated Lane Sample: "Full IM240 Test Criteria"
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
1%
1%
1%
1%
1%
2%
2%
2%
4%
5%
Highs
1%
1%
3%
2%
14%
7%
14%
9%
15%
10%
Very Highs
24%
30%
31%
36%
32%
39%
34%
40%
33%
40%
Supers
74%
68%
66%
60%
53%
51%
50%
49%
48%
46%
IM 240 1983-1985 TBI Simulated Lane Sample: "Full EV1240 Test Criteria"
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
Emission
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
0%
1%
1%
2%
1%
5%
5%
High*.
4%
2%
6%
4%
12%
6%
14%
8%
16%
9%
Verr Hlyhs
52%
47%
58%
53%
58%
56%
56%
55%
53%
52%
Supers
44%
50%
35%
43%
29%
37%
28%
36%
27%
34%
IM240 1986+ TBI Simulated Lane Sample: "Full IM240 Test Criteria"
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
EflUSSlOD
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
1%
2%
1%
2%
1%
2%
1%
2%
1%
4%
ottos
0%
0%
1%
1%
17%
10%
16%
10%
15%
9%
Vt-rv Hlchs
7%
15%
11%
20%
9%
18%
18%
24%
20%
30%
Supers
92%
83%
87%
77%
72%
69%
65%
64%
64%
58%
H-27
-------
1981+ Simulated Lane Sample: "Two Ways to Pass Criteria"
CARfi Vehicles
Excess Excess
Lane IM240
Cut-Points
Comp+Bait 2
0.6/10 + 0.5/12
0.5/15+ 0.5/12
0.6/20 + 0.5/12
0.7/10 + 0.5/12
0.7/15 + 0.5/12
0.7/20 + 0.5/12
0.75/10 + 0.5/12
0.75/15 + 0.5/12
0.75/20 + 0.5/12
0.8/10 + 0.5/12
0.8/15 -MJ.Syt2
0.8/20 + 0.5/12
0.85/10+0.5/12
0.85/15+0.5/12
0.85/20 + 0.5/12
0.9/10 + 0.5/12
0.9/15+0.5/12
0.9/20 + 0.5/12
1.0/10 + 0.5/12
1.0/15+0.5/12
1.0/20 + 0.5/12
1.1/10 + 0.5/12
1.1/15+0.5/12
1.1/20 + 0.5/12
1.2/10 + 0.5/12
1.2/15+0.5/12
1.2/16 + 0.5/12
1.2/20 + 0.5/12
1.6/10 + 0.5/12
1.6/15+0.5/12
1.6/20 + 0.5/12
Emissions
Identified
H£
3122.6
3090.9
3090.9
3034.3
2964.3
2953.0
2902.9
2832.9
2810.3
2850.7
2733:2
2730.6
2832.4
2734.9
2712.3
2832.4
2734.9
2687.8
2832.4
2734.9
2666.5
2832.4
2734.9
2666.5
2832.4
2734.9
2717.1
2666.5
2803.3
2705.8
2604.5
Emissions
£Q
45640.9
44022.2
44022.2
45082.8
42387.3
42129.4
44554.8
41859.4
41373.6
44286.3
41220,2
40734.4
43998.2
40932.1
40446.3
43998.2
40932.1
40026.2
43998.2
40932.1
39681.5
43998.2
40932.1
39681.5
43998.2
40932.1
40626.5
39681.5
43825.2
40759.1
39280.9
Identified
H£
94%
93%
93%
92%
89%
89%
88%
86%
85%
86%
*m
82%
85%
83%
82%
85%
83%
81%
85%
83%
80%
85%
83%
80%
85%
83%
82%
80%
85%
82%
79%
£Q
93%
90%
90%
92%
87%
86%
91%
86%
85%
91%
84%
83%
90%
84%
83%
90%
84%
82%
90%
84%
81%
90%
84%
81%
90%
84%
83%
81%
90%
83%
80%
Number of
Simulated
Failures
1354
1219
1219
1219
1004
992
1165
950
926
1112
842
819
1085
815
792
1085
815
768
1085
815
730
1085
815
730
1085
815
788
730
1061
792
683
Failure
Rate
60%
54%
54%
54%
45%
44%
52%
42%
41%
50%
38%
37%
48%
36%
35%
48%
36%
34%
48%
36%
33%
48%
36%
33%
48%
36%
35%
33%
47%
35%
30%
Number of
Simulated
EC'S
0
0
0
0
0
0
0
0
0
0
a
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-28
-------
1981+ Simulated Lane Sample
Carbureted Vehicles
Excess Excess
Idle/2500
Cut-Points
220/1.2
100/1.0+220/1.2
100/0.5
Emissions
Identified
HC
2246.1
2626.8
2741.1
Emissions
Identified
CQ HC CO
31472.1 68% 64%
37064.9 79% 76%
39716.4 83% 81%
Number of
Simulated
Failures
599
874
1095
Failure
Rate
27%
39%
49%
Number of
Simulated
EC'S
0
0
54
Cc
Rate
0%
0%
2%
Failure Rate
for FTP
Passing
Vehicles
0%
0%
29%
H-29
-------
Characterize Hammond Indiana 1981+ Test Fleet as of 9/25
Number of
Tec hnvtoiY LJUK JM2W
Carbureted Low
High
Total
Vehicles
Actually
Tested at the
Hammond
Lane
1427
812
2239
Number of
Vehicles
in the
Lab Sample
53
69
122
Weighting
Factors
for the
lifBb Sample
26.92
11.77
Number of
Lab Vehicles
Passim FT?
7
0
7
Number of
Normal Lab
Vehicles
Falling FTP
25
4
29
H-30
-------
Characterize Simulated In-Use 81+ CARB Fleet
Calculated From the Lab sample and Weighting Factors
Technology
Carbureted
Number of
Vehicles
Actually
Tested at the
Hammond
Lau
2239
Number of
Simulated
Vehicles
Passing
FTPs,
188
Total Excess Emissions
In Simulated Sample
H£ £Q
3312.86 48915.00
Distrubutlon of Total Excess Emission by Emitter Group
1981+ CARB Sample
Simulated Lane Sample
Emissions
HC(g/mi)
C0(g/mi)
Total
Excess from
Normals
3%
2%
Total
Excess from
Jlighj
7%
4%
Total
Excess from
V
-------
1981+ Simulated Lane Sample: "Full IM240 Test Criteria"
Carbureted Vehicles
Excess Excess
Emissions Emissions Number of
Lane IM240
Cut-Points
0.6/10
0.6/15
0.6/20
0.7/10
0.7/15
0.7/20
0.75/10
0.75/15
0.75/20
0.8/10
0.8/15
0.8/20
0.85/10
0.85/15
0.85/20
0.9/10
0.9/15
0.9/20
1.0/10
1.0/15
1.0/20
1.1/10
1.1/15
1.1/20
1.2/10
1.2/15
1.2/16
1.2/20
1.6/10
1.6/15
1.6/20
Identified
HC
3131.7
3099.9
3099.9
3043.3
2973.1
2960.1
2912.0
2841.7
2817.4
2859.7
2m»
2737.7
2841.4
2743.7
2719.4
2838.2
2740.4
2691.7
2838.2
2740.4
2670.4
2838.2
2740.4
2670.4
2838.2
2740.4
2721.0
2670.4
2809.1
2707.5
2604.5
Identified
CO
45839.4
44220.7
44220.7
45281.3
42541.2
42201.6
44753.3
42013.2
41445.8
44484.8
41374.3
40806.6
44196.7
41085.9
40518.5
44196.7
41085.9
40098.4
44196.7
41085.9
39753.8
44196.7
41085.9
39753.8
44196.7
41085.9
40698.7
39753.8
44023.7
40840.6
39280.9
HC
95%
94%
94%
92%
90%
89%
88%
86%
85%
86%
83%
83%
86%
83%
82%
86%
83%
81%
86%
83%
81%
86%
83%
81%
86%
83%
82%
81%
85%
82%
79%
CO
94%
90%
90%
93%
87%
86%
91%
86%
85%
91%
85%
83%
90%
84%
83%
90%
84%
82%
90%
84%
81%
90%
84%
81%
90%
84%
83%
81%
90%
"83%
80%
Simulated
Failures
1485
1350
1350
1350
1108
1085
1297
1054
1019
1243
947
911
1216
920
884
1162
866
807
1135
839
741
1135
839
741
1135
839
800
741
1112
804
683
Failure
Rate
66%
60%
60%
60%
49%
48%
58%
47%
46%
56%
42%
41%
54%
41%
39%
52%
39%
36%
51%
37%
33%
51%
37%
33%
51%
37%
36%
33%
50%
36%
30%
Number of
Simulated
EC'S
54
54
54
54
54
54
54
54
54
54
54
54
54
54
54
27
27
27
0
0
0
0
0
0
0
0
0
0
0
0
0
EC
Rate
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
2.4%
1.2%
1.2%
1.2%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
Failure Rate
for FTP
Passing
Vehicles
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
28.6%
14.3%
14.3%
14.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
H-32
-------
Dtstrnbution of Excess Emissions Identified by Emitter Group
IM240
Cut-Point
HC/CO
1.6/20
1.2/16
1.0/15
0.8/15
0.6/10
1981+ C ARB Simulated Lane Sample: "Full IM240 Test Criteria"
Emission.
HC
CO
HC
CO
HC
CO
HC
CO
HC
CO
Normals
0%
0%
0%
1%
0%
1%
1%
1%
1%
1%
Hlyhs
1%
1%
1%
1%
1%
1%
1%
1%
6%
4%
Very Hlyhs
59%
68%
61%
69%
61%
69%
61%
69%
60%
69%
Supers
39%
30%
38%
29%
37%
29%
37%
29%
33%
26%
H-33
-------
APPENDIX I
EVAPORATIVE SYSTEM PURGE AND PRESSURE DIAGRAMS
-------
P re a aura
The evaporative pressure test is used to determine the integrity
of a vehicle's evaporative system, and fuel tank. It is conducted by
introducing nitrogen pressure into the fuel tank through the tank-to-
canister vapor vent line near the canister. The Nitrogen is introduced
into the system until the pressure in the fuel tank stabilizes at about
14 inches of water (0.5 PSI) . Fuel tank pressurization is done by
continually modulating the Nitrogen flow into the fuel system by
successive opening and closing of the control valve by the operator.
Modulating the Nitrogen flow into the system allows a higher pressure
Nitrogen flow to be safely used to pressurize the system. Without
modulating the flow, the Nitrogen pressure would have to be low, thus
considerably lengthing the test. If too high a pressure is used, a weak
vapor hose might bulge and rupture.
After the vehicle's evaporative system is pressurized, it is
allowed to stand for up to two minutes to determine if it can continue
to hold pressure. A vehicle is recorded as a failure if the fuel system
pressure drops to less than 8 inches of water within the 2 minute time
frame.
Pressure Teat Equipment
Figure 2 shows a schematic of the Pressure test set-up which is
used. The required equipment includes an air or nitrogen gas bottle, a
standard regulator, and a magnehelic to provide finer control while
pressurizing a vehicle's evaporative system. Other pieces of equipment
include clamps to close off vapor lines and other assorted fittings.
PRESSURE TEST
FILLER NECK
NITROGEN CVUNO EH
-------
Purere Teat. Pror-oHn-ro
The evaporative purge test is conducted during an IM240 transient
dynamometer test to detect vehicles whose evaporative canister purge
system are inoperative. The test procedure includes disconnecting the
test vehicle's vapor purge line running from the canister to the engine,
and installing a gas flowmeter in the line. On most vehicles this is a
relatively simple and quick procedure. However, in some cases the
canister and its purge lines are difficult to find in the allotted time,
and as a result, the vehicles cannot be tested.
After installing the flow meter in the evaporative purge system,
the vehicle is operated over the IM240 transient cycle, and the
cumulative vapor purge flow in units of liters are recorded. The
vehicle is recorded as a failure if its cumulative vapor purge is less
than 1.0 liter.
Teafc Euiment
Figure 1 shows a schematic of the Purge test set-up which is used.
The required equipment includes a transient dynamometer (not shown in
Figure 1) on which to conduct the IM240, and a gas flowmeter which
measures the instantaneous and cumulative vapor purge flow between the
evaporative canister and the engine during the IM240 test. The
dynamometer which is used is a standard Clayton ECE-50 twin roll with
125 pound inertia weights and Road Load power control. The flowmeter
which is used is a Sierra Total flow meter series 730 capable of
measuring flows from 0 to 50 liters per minute.
PURGE TEST
FILLER NECK
ROLLOVER VALVE
EVAPORATIVE
CANISTER
FUEL TAN*
FLOW MCTEF
-------
APPENDIX J
EVAPORATIVE SYSTEM FAILURES AND REPAIRS
-------
Running Loss Emission Levels of Cars Recruited for Repair
at 95 F and 9.0 RVP Fuel
(3 LA-4 Cycles)
Manufacturer
General Motors
FORD
Chrysler
OTHER
ALL
Average Running Losses
Purae Failures,
g/mi
8.46
6.51
5.16
3.94
7.06
Pressure Failures
g/ffij
8.29
7.03
4.20
3.88
7.99
Repair Reduction
a/ml "
5.66
6.26
5.12
3.38
5,67
Alter Repair
a/mi "
3.22
0.55
0.07
0.23
1.76
Notes
* A total of 47 cars were tested ttiat tailed eilhor the purge test, the pressure test, or both tests. The Purge and Pressure Failure Columns
are based on 47 cars.
" A total of 40 cars were repaired. Average Repair Reductions and Alter Repair running losses in Hie above Table are based on 40 cars.
Manufactuer
General Motors
FORD
Chrysler
OTHER
ALL
Total
Failures
22
16
5
4
47
Number Of Failures hv Tvpe
Purge
Failures
11
12
4
2
29
Solenoid /
Electrical
6
6
1
1
14
Purge
LJLBLS
4
6
3
1
13
Unknown
1
0
0
0
1
Pressure
Failures
1 1
4
1
2
18
Vent
Line
1
3
0
0
5
Gas Can
9
0
0
1
10
Tank/
Sending
Unit
1
1
1
1
4
As of 11/12/91
-------
VEH
1455
1455
1456
1456
1459
1459
1532
1532
1532
1580
1580
1712
1712
1461
1461
1530
1530
1578
1578
1457
1457
1524
1524
1533
1542
1542
1544
1544
1584
1584
1672
1672
MYR
88
88
90
90
90
90
86
86
86
90
90
87
87
87
87
84
84
86
86
85
85
85
85
88
85
85
86
86
86
86
86
86
MAKE
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
CHEVY
PONTIAC
PONT1AC
PONTIAC
PONTIAC
PONTIAC
PONTIAC
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
BUICK
QUICK
BUICK
MODEL
CORS
CORS
LUMI
LUMI
.UMI
.UMI
228
Z28
Z28
CAVA
CAVA
CAVA
CAVA
-
GRNAM
GRNAM
6000
6000
SUNB
SUNB
CENT
CENT
LESA
LESA
REGAL
>ARK
'ARK
CENT
CENT
CENT
CENT
LESA
LESA
'
ENG.FAM
2.0LTBI
2.0L TBI
L1G3.1V8XGZ5
L1G31V8XGZ5
3.1LPFI
3.1LPFI
G1G50V8NTAB
G1G50V8NTA8
G1G50V8NTA8
L1G22V5JFG2
L1G2.2V5JFG2
2 OL TBI
2.0LTBI
2 5L TBI
2.5L TBI
2 51 TBI
2 51 TBI
G2G18V5TDG2
G2G1.8V5TDG2
2.5L TBI
2.5LTBI
5.0L
5.0L
2.8L PR
F4G3.8VBXEB3
F4G3.8V8XEB3
G2G25V5TPG9
32G25V5TPG9
32G25V5TPG9
G2G2.5V5TPG9
S4G3.8V8XEB4
34G38V8XEB4
Purfl
PASS
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
PASS
FAIL
PASS
FAIL
PASS
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
Asoll
Pres
FAIL
PASS
PASS
PASS
PASS
PASS
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
FAIL
PASS
PASS
PASS
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
PASS
PASS
1/12/9
TEST
RECV
RM1
RECV
RM1
RECV
RM1
RECV
RM1
RM2
RECV
RM1
RECV
RMI
RECV
RMI
RECV
RMI
RECV
RMI
RECV
RMI
RECV
RMI
RECV
RECV
RMI
RECV
RMI
RECV
RMI
RECV
RMI
TOTAL RL
a/ml
588
306
11.86
1.88
2700
13.85
14.02
14.70
14.22
563
0.34
848
7.78
1153
0.10
773
0.11
11.70
14.45
-
465
0.11
331
0.11
0.07
2.86
0.44
10.43
1.05
7.54
014
7.80
0.87
RL Reducl
a/ml
2.81
998
1315
-020
530
069
11.44
7.63
-275
4.54
320
2.42
9.38
7.40
6.93
Hot Soak
045
Olumal
043
COMMENTS
Gas Cap Leaks
Bad Purge Solenoid
Purge Solenoid Inoperative
Purge Line from Canister Disconnected
Gas Cap Leaking
Leak Possibly at Sending Unit Gasket
Gas Cap Leaks
Vent Line Leaks at connection with Steel Lin
,x,:,,,: ;...,:. «,.:.;>.:. :::.:,,;::.:;.5:i<:::.:;.;j:::-:,
Gas Cap Leaks, Filler Neck rusted at Cap
Gas Cap Does not Seal
Gas Cap Leaks
Crack in Seam ol Gas Tank and Filler Neck
TVS Broken
Unknown Problems
Gas Cap Does not Seal
Vac Line from Manifold to Purg Valve Broker
Purge Una Weathered. Collaspes when hot
Purge Solenoid Wiring Disconnected
-------
VEH
158!
MM
1661
1661
1720
1526
1552
1552
1560
1560
MYH
87
87
85
85
88
-
85
84
84
86
86
MAKE
OLDS
OLDS
OLOS
OLDS
OLDS
CADILLAC
CADILLAC
CADILLAC
CADIL
CADIL
.AC
.AC
MODEL
[)E
DE
-T
T
CUTL
CUTL
CUTL
, ,, ,
SEVI
SEVI
SEVI
ELDO
ELDO
ENG.FAM
H2G3.8V8XEB7
H2G3.8V8XEB7
F2G25V5TPG8
F2G25V5TPG8
2 8L PFI
4.1LTBI
E6G4.1V5NKA7
E6G4.1V5NKA7
G6G4.
G6G4.
V5NKA9
V5NKA9
Purn
FAIL
PASS
PASS
PASS
PASS
FAIL
FAIL
PASS
FAIL
PASS
Asol 11 112/9
Pm»
PASS
PASS
FAIL
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
TEST
RECV
RM1
RECV
RM1
RECV
'
RECV
RECV
RM1
RECV
RM1
TOTAL RL
a/ml
553
2.14
880
022
9.20
5.48
711
020
6.88
0.16
RL Reducl
fl/ml
339
858
-
6.91
671
Hot Soak
076
Diurnal
1.76
COMMENTS
Purae Line Disconnected
Gas Cap Leaks
Gas Cap has a holed Drilled in it
Canister Purge Solenoid Inoperative
Canister Purge Solenoid Bad
Purge Vapor Line on top of Can. Disconnect!
-------
Vah
1460
1460
1462
1462
1525
1525
1563
1563
1574
1574
1575
1575
1647
1647
1676
1676
1450
1450
1537
1537
1561
1561
1689
1689
1802
1802
1802
* s
fex f.vVtff*
1548
1548
1639
1713
1713
1713
MYR
85
85
88
88
89
89
85
85
87
87
89
89
86
86
84
84
'
88
88
86
86
85
85
85
as
83
83
83
-.•>, ,••'
88
88
81
86
86
86
MAKE
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
FORD
*
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
MERCURY
f S -f
LINCOLN
LINCOLN
LINCOLN
LINCOLN
LINCOLN
LINCOLN
MODEL
CROWN
CROWN
MUST
MUST
MUST
MUST
TEMP
TEMP
ESCO
ESCO
TAUR
TAUR
MUST
MUST
CROWN
CROWN
'
TRAC
TRAC
COUG
COUG
TOPA
TOPA
MARQ
MARO
MARQ
MARQ
MARQ
ff x-:--.
MARK
MARK
TOWN
TOWN
TOWN
TOWN
ENG.FAM
5.0L TBI
5 OL TBI
2.3L PFI
2 3L PFI
5.0L PFI
5.0LPFI
FFM2.3V5HCF4
FFM2.3V5HCF4
HFM1.9V5FFF1
HFMI.9V5FFFI
KFM2.5V5HCF1
KFM2.5V5HCF1
GFM50V5HBF9
GFM5.0VSHBF9
EFM50V5HBF7
EFM50V5HBF7
* ,,> *
JFM16V5FZKO
JFM1.6V5FZKO
3.BL TBI
3.8LTBI
FFM2.3V5HCF4
FFM2.3V5HCF4
SOL TBI
50 L TBI
5.0L TBI
5 OL TBI
50LTBI
* ,„,.>, ,«
JFM5.0V5HBF3
JFM50V5HBF3
5.0CCF
50LPFI
5 OL PFI
5.0L PFI
Purg
FAIL
PASS
PASS
PASS
FAIL
PASS
FAIL
PASS
PASS
PASS
FAIL
PASS
PASS
PASS
FAIL
PASS
FAIL
PASS
PASS
PASS
FAIL
PASS
FAIL
PASS
FAIL
FAIL
PASS
FAIL
PASS
FAIL
FAIL
PASS
PASS
AsoM
Pres
PASS
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
FAIL
PASS
PASS
PASS
FAIL
PASS
PASS
PASS
-
PASS
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
FAIL
FAIL
PASS
PASS
PASS
PASS
PASS
PASS
PASS
1/12/9
TEST
RECV
RM1
RECV
RM1
RECV
RM1
RECV
RMt
RECV
RM1
RECV
RM1
RECV
RM1
RECV
RM1
RECV
RM1
RECV
RM1
rRECV
RM1
RECV
RM1
RECV
RM1
RM2
RECV
RM1
RECV
RECV
RM1
REP
TOTAL RL
Q/ml
715
025
917
239
11.63
082
526
024
3.58
Oil
1041
008
860
036
4.43
181
1.22
017
676
005
3.88
0.10
349
025
628
634
0.12
-
12.26
1.40
409
806
009
012
| :
RL Reduct
fl/ml
690
678
10.81
503
3.48
1033
8.24
262
106
6.71
3.78
324
6 16
1086
7.94
Hot Soak
068
Diurnal
'
0.74
COMMENTS
Electronic Puroe Solenoid Inoperative
Plastic Vent Line Irom Canister Missing
Connector at Can. does not make conneclioi
Purge Solenoid Inoperative
Sending Unit Gasket does not seal
Purae Vac Line burned through
Fuel Tank Vent line disconnected at Caniste
Purge Vac Line Misrouted
Broken Purge Line Fittings
Vapor Line Disconnected at Rollover Valve
Purge Solenoid Inoperative
TVS is stuck in the Closed Position
Recv - Tank Vent line plugged. Battery Acid
destroyed Canister.
RM1 - Repaired Purge Lines
RM2 - New Normalized Canister
Electrical Connection at Purge Solenoid Bad
Electrical Grounding Problem near ECM
Purge Hose Disconnected
-------
VEHicLe
1530
is??
1S37
1S41
1542
1574
1578
ISIO
1647
I6«l
1667
Average
VEHICLE
1524
1S2S
1544
1548
1552
1655
1580
1561
1563
1575
15S4
15(5
1567
1640
Average
f^HESSOBE FAILURES 1
DIAGNOSIS oH REPAIRS
«EW OEM CMS CAP
IEPUCEGASCAP
^CONNECTED FUEL TANK VAPOR LNE TO ROLLOVER VALVE
1EPUCED OVEFFLOW HOSE WHICH WAS SPLIT
REPLACED FILLER NECK AND GAS CAP
5ASKET AROUND SENDNG UNIT DOES NOT SEAL PROPERLY
GAS CAP LEAKS
GAS CAP LEAKS
RECONNECT FUEL TANK VENT LINE TO CANISTER
GAS CAP LEAKS
GAS CAP LEAKS
PURGE PAiLURES
DIAGNOSIS OR REPAIRS
REPLACE THERMAL-VACUUM SWITCH
REPLACE MALE END OF CONNECTOR TO CANISTER PURGE SOEI
VAC UNE RECONNECTED FROM M ANIFOUD TO PURGE VALVE
PURGE SOUBNOD REPLACED
REPLACED FAULTY CANISTER PURGE SOLENOD
TEE FITTNG REPLACED
PURGE LWE RECONNECTED
PURSE SOLENOD REPLACED
PURSE SOLENOD REPLACED
PURGE VACUUM LME TO THE ENGINE HAS A HOLE BURNED THF
PURGE LME IS WEATHERED AND COLLAPSES WHEN HOT
PURGE LNE IS DISCONNECTED
PURGE LNE B BROKEN NEAR ENGINE
THERMAL VACUUM SWITCH REPLACED
g/t.l
171 60
-4 40
151 00
5650
5440
7640
•61 90
119 10
165 40
19300
115 30
86.04
HCRcd
g/lsl
71 90
24320
211 00
244 30
15550
123 ID
15099
64 9C
113 1C
23246
16660
7630
13000
56 60
147.14
THREE BAGS
g/rnM.
s«e
004
4S4
299
1 54
266
296
457
429
656
502
3.75
HNST
1HREEIAO3
HCRod
glifUt
1 37
735
721
643
542
361
563
320
416
851
403
402
4 14
1 40
4.75
elmllo
17. 35
6.23
1.63
•1.33
•24 18
26 II
911
1566
-4.73
4.45
5. IB
Fu«l
Consumption
HC hiel Rod
glmllo
-3 52
-13(7
1 51
766
•2 35
469
14 76
oot
•5 63
18 10
-9 (0
41 00
-0.63
41 02
6.65
T«l«l
FiMfeRL MC
(/mile
2323
8.27
6.17
166
-2334
2699
12.09
20.43
•0.44
11.01
•.81
Tolol
FuoURLHC
Sevlnge
fiRilll
•215
••.32
• 72
1411
307
630
2041
326
•1 67
2661
-5.77
4502
351
42.42
11.40
Part*
BASIS
Etl
Acl
Ell
Ell
Act
Ell
E»l
E>l
EM
E»l
Ell
Paris
Cost
BASIS
Ell
Ell
Ell
Act
Ell
Ell
Acl
Acl
Acl
Ell
Ell
EM
Ell
Acl
Porto
$700
$362
(000
(000
(3340
(9311
(700
(700
(000
(700
$720
$15.03
Cost
Puts
(2041
(2946
(000
(3853
$2946
(100
(000
(3704
(13015
(000
(000
(000
(000
(2041
$21.89
Lobor
BASIS
EH
Act
Ell
Ell
Act
Ell
Ell
Ell
Ell
Ell
EH
Lobor
Tim*
BASIS
Ell
Ell
Ell
Acl
Ell
Acl
Acl
Acl
Acl
EH
Ell
Ell
EM
Est
TIME
0.25
025
0.5C
OSC
IOC
1 0(
025
0.25
OS(
025
0.25
045
Labor
TIME
050
OSC
OSC
600
0.5C
OSC
OSC
OSC
1.50
05C
050
050
050
0.50
0.96
•*)> SSO/MR
1 r-^t
Uboi
(12 SO
(1250
(2500
(2500
(5000
(5000
(1250
(1250
(25.00
(1250
(12 SO
$22.73
'« HO/HR
Cost
Labor
(2500
(2500
(2500
(30000
(2500
(2500
(2500
(2500
(7500
(2500
(25.00
(2500
(2500
(25.00
$48.21
•@SSOHH
1 fco»t
fotli
$1950
(1612
(25.00
(2500
(8340
(14311
(1950
(1950
(2500
(1950
$1970
$37.76
•®$50/HR
Cosl
Total
(45.41
SS446
(2500
(338 53
(5446
(26.00
(2500
(6204
(20515
(2500
(25.00
(2500
(25.00
$45.41
$70.10
PARTS
•/gfmlle
(030
$044
(000
(000
((143
(321
(056
$034
$0.00
$0.64
•...
$1.71
PARTS
Cool / Rod
f/g/mllt
((948]
((466]
(000
(273
$961
(0.12
$000
$1129
($7784)
$0.00
(000
$000
$000
$048
$1.92
•(§> *5CyHH
LABOfl
(fg/mlle
(0.54
(151
(4.05
(1504
((2.14
$173
(1.03
$0.61
($57.34
$1.14
•-..
$2.56
'@(50/HF
LABOR
Coal / Red
S/g/milo
((11 62
((396
(2.87
(2126
(815
$301
$122
$762
($44 (61
$094
($433)
JO 56
(7.12
$059
$4.23
FUEL
eOCNGMV
gal/mllo
00091
0.0034
0.002!
o.ooo;
•00091
0.0121
0.0050
0.0084
-0.0002
0.0045
•...
0.0036
FUEI
ECONOMY
SAVINGS
gal/milt
-0.0009
•0.0026
0.0031
0.0056
00013
0.0034
0.0084
0 0014
•0.0007
0.0110
-0.0024
0.0186
0.0015
0.0175
0.0047
•^SSO/Mfi
TOTM
1 coal / h«
syg/mii*
(064
$195
(405
(1504
($3.57
(494
$161
$095
($5734
$1.77
$4.29
•e$so*»
TOTAL
Coil /Red
$/g/mlle
((21 10
($862
$287
$2399
(1776
$3.13
$1.22
$1691
($12270:
$094
($433)
$056
$712
$1 07
$6.15
-------
VEH
555
555
.- -
1458
1458
1587
1587
1635
1559
'
VEH
1541
1541
1568
1640
1640
•."/- ^
1667
1667
MYR
88
88
87
87
86
86
84
89
-. * s "
MYR
83
83
86
84
84
...:
MAKE
DHRYSLEF
CHRYSLEF
-
DODGE
DODGE
DODGE
DODGE
DODGE
PLYMOU1>
4 ': ': \ ',' ''
MAKE
RENAULT
RENAULT
MAZDA
TOYOTA
TOYOTA
%.•'• f ff<'-
eRENAU
IRENAU
• •
.
MODEL
L
L
•BA
EBA
1 »
LANC
LANC
DAYT
DAYT
DAYT
^
ACCL
t , ,,?
MODEL
ALL!
ALL!
-
HX-7
-
CELI
CELI
,,/z
AL
AL
.1
.1
ENG.FAM
JCR25V5FBE6
JCR2.5V5FBE6
-,,, " ,
22LTBI
22LTBI
GCR2.2V5FAAX
GCR2.2V5FAAX
ECR2.2V5FAA8
, , -',-. «
KCR2.5V5FBD6
\^4?l;
ENG.FAM
DAM1.4V5FFD3
DAM1.4V5FFD3
-."• fsff, vA J •-
GTK1.3V5HFD8
"
ETY28V5FBB5
ETY2.8V5FBB5
','... '...-.,.
FAM1.4V5FFA2
FAM1.4V5FFA2
Purg
FAIL
PASS
PASS
PASS
FAIL
PASS
FAIL
-
FAIL
-
Purg
PASS
PASS
FAIL
FAIL
PASS
\ f •.
PASS
PASS
As ofll/12/91
Pres
PASS
PASS
FAIL
PASS
PASS
PASS
PASS
PASS
-
TEST
RECV
RM1
'
RECV
RM1
RECV
RM1
RECV
*
RECV
^
TOTAL RL
a/ml
552
0.05
,- -
4.20
008
586
0.08
5.43
381
V*
': s --
As of 11/12/91
Pres
FAIL
PASS
-
PASS
PASS
PASS
-
FAIL
PASS
TEST
RECV
RM1
»$o
RECV
RECV
RM1
RECV
RM1
TOTAL RL
g/ml
2.57
0.06
4.79
3.08
0.57
- ,J
5.19
0.07
RL Reduct
a/ml
5.47
412
5.78
" ,
RL Reducl
g/ml
2.51
2.51
5.12
Hoi Soak
<
Hoi Soak
Diurnal
-'\ ( .
';
Diurnal
:
1
COMMENTS
Tee fining in purge line broken at Throttle Bo
Vehicle has a Broken Rollover Valve
Purge Line Broken near Engine
Canister Purge Line is Burned in Hall
Purge Valve Stuck
%^
COMMENTS
Fuel Sending Unit Gasket
-
Purges thru Oil Filler. Oil cap does not seal
TVS is Broken
-
Gas Cap Leaks
-------
APPENDIX K
MOBILE4.1 PERFORMANCE STANDARD ANALYSES, BY OPTION
-------
IRun »8p Mod. High Option, Performance standard Options (annual|
MOBILE4.K4NOV91)
0
-M120 Warningi
4. MOBILE*.1 does not model moat 1993 and later clean Ale Act
requirements* Emission Factors Cor CY 1993 or later are affected.
OI/M program selected:
Start year (January 1):
Pre-1981 MYR stringency ratet
First model year covered:
Last model year covered!
Waiver rate (pre-19811s
waiver rate (1981 and newer):
Compliance Ratet
Inspection type!
Inspection frequency
Vehicle types covered!
1981 t later MYR test type!
1983
20%
1968
2020
1.%
1.%
98.%
Centralized
Annual
LDOV - Yes
U5GT1 - Yes
LDCT2 - Yes
HDGV - No
2SOO rpra / I
MOBIU4.1 IM240 Transient Annual 1/M Credits .8/15 (8/5/91)
Oldest model year covered by:
IH240 Transient Test 1986
Purge System Check 1986
Fuel System Pressure Check 1983
OAnti-tamperlng program selected:
0 Start year (January 11 : 1983
First model year covered: 1984
Last model year covered: 2020
Vohlcle types covered: LDGV , LDGT1, LDCT2
" Type: Centralized
Frequency: Annual
Compliance Rate: 98.0%
0 Air pump system disablements: No
Catalyst removals : Yes
Fuel inlet restrlctor disablements: Yes
Tailpipe lead deposit test: No
ECU disablement: No
Evaporative system disablements: No
PCV system disablements: No
Missing gas caps: No
0 Minimum Temp: 72. (F) Maximum Temp: 92.
Period 1 RVPi 11.5 Period 2 RVPt 8.7 Period 2 Start Yri 1992
OVOC HC emission factors include evaporative HC emission factors.
0
OCal. Year: 2000 I/M
Anti-tarn.
0 Veh. Type: LDGV
+
Veh. Speeds: 137?
VMT Mix:
OComposlte Emission
VOC HC:
Exhaust HC:
Eveporat HC:
Refuel L HC:
Runlng L I1C:
Rstlng L HC:
Exhaust CO:
Exhaust NOXI
0.584
rectors
1.16
0.48
0.16
0.19
0.23
0.11
6.17
0.72
Program: Yes Anblent Temp: 87.5 / 87.5 / 87.5 (F) Region: Low
Program: Yea operating Mode: 20.6 / 27.3 / 20.6 Altitude: 500
LDST1 LDGT2 LDGT HDGV LDDV LOOT HDOV
13. S
0.199
(Cm/Nile)
1.50
0.81
0.17
0.25
0.19
0.10
7.39
1.03
19.S
0.080
1.69
0.90
0.24
0.25
0.20
0.10
8.58
1.13
1.56
0.83
0.19
0.23
0.19
0.10
7.73
1.07
19. t
0.035
4.54
2.07
1.44
0.40
0.51
0.12
36.27
4.51
13. i
0.002
0.6S
0.65
1.60
1.37
19.6
0.001
0.84
0.84
1.74
1.50
19.6
0.093
2.15
2.15
11.21
9.30
(F)
. Ft.
MC
15. S
0.007
4.96
1.89
2.52
0.54
24.78
0.77
All Veh
1.503
0.796
0.213
0.192
0.204
0.098
9.230
1.753
-------
Mobile output from CEM 4.1 1/24/92
Run »2 : Basic I/M
MOBILE4.1(4Nov91)
Ml20 Warning:
MOBILE4.1 does not modal moat 1993 and later Clean Air Act
requirements; Emission Factors for CY 1993 or later are affected.
I/M program selected:
Start year (January 1):
Pre-1981 MYR stringency rate:
First imdel year covered:
Last model year covered:
Waiver rate (pre-1981):
Waiver rate (1981 and newer):
Compliance Rate:
Inspection type:
Inspection frequency
Vehicle types covered:
1961 t later MYR test type:
Oldest model year covered by:
IM240 Transient Test
Purge System Check
Fuel System Pressure Check
1983
20%
1968
2020
0.%
0.%
100.%
Centralized
Annual
LDGV - Yes
LDGT1 - No
LDGT2 - No
HDGV - No
Idle
2020
2020
2020
Period 1 RVP: 11.5
Minimum Temp:
Period 2 RVP:
72.
8.7
(F)
Maximum Temp: 92. (F)
Period 2 Start Yr: 1992
VOC HC emission factors Include evaporative MC emission factors.
Cal. Year: 2000
'/eh. Type:
I/M Program: Yes
Anti—tarn. Program: No
Ambient Temp: 87.5 / 97.5 / 87.5 (F) Region: Low
Operating Mode: 20.6 / 27.3 / 20.6 Altitude: SOO. Ft.
LDGT1
LDST2
LOST
L0DV
LDDT
All Veh
Veh. Speeds:
VMT Mix:
19.?
O.S84
Composite Emission Factors
VOC HC:
Exhaust HC:
Evaporat HC:
Refuel L HC:
Runing L HC:
Rating L HC:
Exhaust CO:
Exhaust NOX:
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
1.S8
0.52
0.36
0.19
0.42
0.11
e.se
0.73
I/M
Anti— tarn.
LDGV
19.6
0.584
Composite Emission Factors
VOC HC:
Exhaust HC:
Evaporat HC:
Refuel L HC:
Runing L HC:
Rstlng L HC:
Exhaust CO:
Exhaust NOX:
1.77
0.71
0.36
0.19
0.42
0.11
10.05
0.75
19.?
0.199
(Cm/Mile)
2.25
1.20
0.35
0.25
0.36
0.10
11.96
1.1S
Program:
Program:
IDGT1
19.?
0.199
(Gm/Hlle)
2.25
1.20
0.35
0.25
0.36
0.10
11.96
1.15
" 19.?
0.080
2.59
1.37
0.46
0.25
0.40
0.10
14.40
1.29
No
No
LDGT2
" 19.?
0.080
2.59
1.37
0.46
0.25
0.40
0.10
14.40
1.25
2.35
1.25
0.38
0.25
0.37
0.10
12.66
1.18
Ambient
Operating
LDGT
2.35
1.25
0.38
0.25
0.37
0.10
12.66
1.18
19
0
4
2
1
0
0
0
36
4
Temp:
Model
.?
.035
.54
.07
.44
.40
.51
.12
.27
.51
87.5 /
20.6 /
HDGV
19
0
4
2
1
0
0
0
36
4
.035
.54
.07
.44
.40
.51
.12
.27
.51
19.?
0.002
0.65
0.65
1.60
1.37
87.5 /
27.3 /
LDDV
19.?
0.002
0.65
0.65
1.60
1.37
19.?
0.001
0.34
0.84
1.74
1.50
87.5 (F)
19TS
0.093
2.15
2.15
11.21
9.30
Region: Low
20.6 Altitude: 500.
LDDT
19.?
0.001
0.84
0.84
1.74
1.50
HD0V
"""""" 19.6
0.093
2.15
2.15
11.21
9.30
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
Ft.
MC
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
1.971
0.937
0.381
0.192
0.364
0.098
10.021
1.783
All Veh
2.084
1.049
0.382
0.192
0.364
0.098
11.874
1.799
-------
CEM4.1 : Run *2 : Basic I/M 01/24/92 13:41:59
Local Parameters Selected;
Base RVP: 11.5
In-use RVP: 8.7
Vehicle Type: LDGV
Veh. Speeds: I9.fi
I/M start year (January 1) :
Tampering deterrence start:
Oldest model year covered:
I/M inspection frequency:
I/M program type:
1981 t later MXR test type:
Vehicle types covered:
I/M waiver rate (pre-1981) :
I/M waiver rate (post-1981) :
I/M compliance rate:
Cost Assumptions
Operating Mode: 20.6
In-use Start Yr: 1992
LDGT1 LDGT2
157? 157?
1553
1983
1968-2020 Fixed
Annual
Centralized
Idle
LDGV
0.0%
0.0%
100.0%
1 87.5 1 87.5 (Ft Region; Low
/ 27.3 / 20.6 Altitude: 500. Ft.
Minimum Temp: 72. (F)
Maximum Temp: 92. (F)
HDGV LDDV LDDT HDDV MC
—&7S -T57S —157? — 157? ^575
I/M Program Selected:
stringency: 20
IM240 test model year coverage: 2020+
Purge check model year coverage: 2020+
Pressure check model year coverage: 2020+
Fleet Assumptions LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC
I/M inspection cost: (per insp) $ 8.00 Vehicle percentages: 63.2 1577 371 376 o75 072 3.3
Growth rate: 0.0 % per year
Avg. Gasoline Cost (per gallon): $ 1.25 Fleet size is 1000000. vehicles in base year 1986
Repair Costs Pre-91 81+
Avg. I/M repair cost i 5 50. 75. ( IM240 repair « $ 150. NOx repair* 5 0. )
Benefits (1000 tons/yr) and Costs (1000 S/yr) are averaged over the calendar years 2000 thru 2000.
Calendar Year 2000 TotVOC 55 ExhHC Evap Refuel RnLoss RstLos
Baseline No-Program Emissions: 21.955 123.217 10.929 4.024 2.067 3.897 1.038
Program Benefits and Costs
VOC
Average Annual ATP Benefit, Fee 6.666
Evap/RunLoss Benefit, Fee
Evap/RunLoss MFC Benefit
Avg Annual I/M Benefit, Fee
Avg Annual ATP Repair Cost
Avg Ann. Prg/Prass Repair Cost
Avg Ann. IM Repair Cost,FailRt
Average Annual IM Fuel Savings
Exhaust . Tampering Deterrence
0.006
1.033
0.142
Evap RnLoss ExhHC Ann Avg 1.181
0.005 0.001 1.175 (
2936
CO
0.000
17.762
1.646
19.408
48237
Cost
0.
0.
( o.
5074.
0.
0.
3536.
( 2198.
£412.
kilograms
)
)
P
0.0483
-------
CEM4.1 : Run HO High Option (Biennial)
01/24/92 13:43:19
Local Parameters Selected:
Ambient Temp: 87.5 / 87.5 / 87.5 (F)
Operating Mode 20.6 I 27.3 / 20.6
Region: Low
Altitude: 500. Ft.
Base RVP: 11.5
In-use RVP; 9.7
Vehicle Type:
Veh. Speeds:
In-use Start Yr: 1992
LDGV LDGT1 LDGT2 HDOV LDDV LDDT
TSTJ 137? 157? 177? T575 157?
I/M Program Selected:
Minimum Temp: 72.
Maximum Temp: 92.
HDDV
19.4
(F)
(F)
MC
19.«
I/M stact year (January 1):
Tampering deterrence start:
Oldest model year covered:
I/M inspection £requency:
I/M program type:
1991 t later MYR test type:
Vehicle types covered:
I/M waiver rate (pre-1981):
I/M waiver rate (post-1991):
I/M compliance rate:
1983
1983
1968-2020 Fixed
Biennial
Centralized
2500 rpm / Idle
LDGV, LDCT1, LDGT2
1.0%
1.0%
98.0%
Stringency:20
IM240 test model year coverage:
Purge check model year coverage:
1984+
1984+
Pressure check model year coverage: 1971+
Anti-Tamperlng Program Selected:
ATP inspection frequency:
Oldest model year covered*
Vehicle types covered:
Parameters covered by ATP:
Biennial
1975-2020 Fixed
LDGV , LDGT1, tDGT2
Air Punp / Catalyst
Evap System / PCV System
ATP compliance rate:
ATP program type:
/ Fuel Inlet / Plumbtesroo /
/ Gas Cap /
98.0%
Centralized
Cost Assumptions
Fleet Assumptions
LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC
I/a Inspection cost: (per insp) $ 8.00
ATP inspection cost: (per insp) $ 0.50
Purge inspect cost (per insp) : 5 6.53
Pressure insp cost (per insp) : S 0.69
IM240 insp Incremnt over Purge: $ 0.87
Avg. Gasoline Coat(per gallon): $ 1.25
Repair Costs
Pre-81
81+
Avg. I/M repair cost
Air Pump repair cost
Catalyst replacement cost
Mlsfueled catalyst cost
Evap* system repair cost
PCV system repair cost
Gas cap repair cost
Purge repair cost
Pressure repair cost
S 50.
5 15.
$ 150.
$ 175.
S 5.
5 5.
S 5.
$ 70.
$ 38.
75. (
15.
165.
190.
5.
5.
5.
70.
38.
Vehicle percentages:63.219.7
Growth rate: 0.0 % per year
Fleet size Is 1000000. vehicles In base year 1986
IM240 repair - $ 120. NOx repair- $ 0. )
Benefits (1000 tons/yr) and Costs (1000 $/yr) are averaged over the calendar years 2000 thru 2000.
Calendar Year 2000
Fleet Size 1000000
Baseline No-Program Emissions:
TotVOC
21.955
Program
CO
123.217
Benefits
ExhHC
10.929
and Costs
Evap
4.024
Refuel
2.067
RnLoss
3.897
RstLos
1.
038
Average Annual ATP Benefit,Fee 0.496 4.648
Evap/Runloss Benefit, Fee 3.583
Evap/RunLoss MPG Benefit
Avg Annual I/M Benefit, Fee 1.925 30.973
Avg Annual ATP Repair Cost
Avg Ann. Prg/Press Repair Cost
Avg Ann. IM Repair Cost,FallRt
Average Annual IM Fuel Savings
Exhaust Tampering Deterrence 0.210 2.315
Evap
1.873
RnLoss
1.710
ExhHC
2.631
Ann Avg STZTT 37.937
( 15443
228.
3057.
3363.)
4006.
888.
3577.
7015.
9979.)
5429.
94290 kilograms per day )
-------
CEM4.1 : Run »3 : Low Option
Local Parameters Selected:
01/24/92 13:42:09
Base RVP: 11.5
In-use RVP: 8.7
Vehicle Type: LDGV
Veh. Speedsi 1575
Tampering deterrence start:
Oldest model year covered:
I/M Inspection frequency:
I/M program type:
1981 4 later MiR test type:
Vehicle types covered:
I/M waiver rate (pre-1981) :
I/M waiver rate (post-1981):
I/M compliance rate:
ATP start year (January 1) :
ATP inspection frequency:
Ambient Temp: 87.5
Operating Mode: 20.6
In-use Start Xr: 1992
LDGT1 LDGT2
1575 T575
1983
1968-2020 Fixed
Annual
Centralized
Idle
LDGV, LDGT1, LDGT2
1.0%
1.0%
98.0%
/ 87.5 / 87.5 (F) Region: Low
/ 27.3 / 20.6 Altitude: 500. Ft.
Minimum Temp: 72.
Maximum Temp: 92.
HDGV LDDV LDDT HDDV
T575 1575 T575 T575
I/M Program Selected:
Stringency: 20
IM240 test model year
Purge check model year
(F)
(F)
MC
19. i
coverage: 2020+
coverage: 2020+
Pressure check model year coverage: 2020+
Anti-Tampering Program Selected:
1553
Annual
ATP compliance rate:
ATP program type:
93.01
Centralized
Oldest model year covered:
Vehicle types covered:
Parameters covered by ATP;
Cost Assumptions
1981-2020 Fixed
LDGV , LDCT1, LDGT2
Catalyst / Fuel Inlet
Fleet Assumptions
LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV MC
I/M Inspection cost:(per InspJS8.00
ATP inspection cost: (per insp) $ 0.50
Avg. Gasoline Cost(per gallon): $ 1.25
Vehicle percentages: 63.2 1977 3.1 3.6 0.2 0.2 3.3
Growth rate: 0.0 % per year
Fleet size is 1000000. vehicles in base year 1966
T78
Repair Costs
Pre-81
81+
( IM240 repair - S 150. ROx repair- S 0. )
Avg. I/M repair cost : $ 537 TSl
Air Pump repair cost : $ IS. IS.
Catalyst replacement cost: $ ISO. 165.
Misfueled catalyst cost : $ 175. 190.
Bvap. system repair cost : $ 5. 5.
PCV system repair cost : 5 S. 5.
Gas cap repair cost : $ 5. 5.
Benefits (1000 tons/yr) and Costs (1000 $/yr) are averaged over the calendar years 2000 thru 2000.
Calendar Year 2000
Fleet Size 1000000
Baseline No-Program Emissions:
Evap
21.9S5 123.217 10.929
Program Benefits and Costs
2.067
RnLoss
3.897
1.038
Average Annual ATP Benefit, Fee
Evap/RunLoss Benefit, Fee
Evap/RunLoss MPO Benefit
Avg Annual I/M Benefit, Fee
Avg Annual ATP Repair Cost
Avg Ann. Prg/Press Repair Cost
Avg Ann. IM Repair Cost,FallRt
Average Annual IM Fuel Savings
Exhaust Tampering Deterrence
VOC
0.591
0.034
1.697
0.210
Evap RnLoss ExhHC Ann Avg 2.229
0.032 0.003 2.194 (
5539
CO
T7554"
26.634
2.315
30. S3i
76136
Cost
439.
0.
( 0.)
7280.
384.
0.
5192.
( 3609.)
988S.
kilograms per
0.0738
day )
-------
CEM4.1 : Run »5 Medium Option
Local Parameters Selected:
01/24/92 13:42:28
Ambient Temp: 97.5
Operating Mode: 20.6
Base RVP: 11.5
In-use RVP: 8.7 In-use Start Irs 1992
Vehicle Type: LDGV
Veh. Speeds: 157?
I/M start year (January 1) :
Tampering deterrence start:
Oldest model year covered:
I/M inspection frequency:
I/M program type:
1981 t later MIR test type:
Vehicle types covered:
I/M waiver rate (pre-1981) :
I/M waiver rate (post-1981) :
I/M compliance rate:
LDGT1 LDGT2
157? T5T?
1983
1983
1968-2020 Fixed
Annual
Centralized
Idle
LDGV, LDGT1, LDGT2
1.0%
1.0%
98.0%
/ 87.5 / 87.5 (F) Region: Low
/ 27.3 / 20.6 Altitude: 500. Ft.
Minimum Temp: 72. (F)
Maximum Temp: 92. (F)
HDGV LDDV LDDT HDDV MC
157? — rSTS — 1575" 157? T5T?
I/M Program Selected:
stringency: 20
IM240 test model year coverage:
Purge check model year coverage:
Pressure check model year coverage:
2020+
2020+
1971+
Anti-Tamperlng Program Selected:
ATP inspection frequency:
Annual
ATP compliance rate: 98.0%
ATP program type: Centralized
Oldest model year covered!
Vehicle types covered:
Parameters covered by ATP:
Cost Assumptions
1961-2020 Fixed
LDGV , LDGT1, LDGT2
Catalyst / Fuel Inlet /
Fleet Assumptions
LDGV LDT1 LDT2 HDGV LDDV LDDT HDDV
I/M Inspection cost: (per Insp) S 8.00
ATP Inspection cost: (per Insp} $ 0.50
Purge inspect cost (per insp) : $ 6.53
Pressure insp cost (per insp) : $ 1.94
IM240 insp Incrennt over Purge: $ 0.87
Avg. Gasoline Cost(per gallon): S 1.25
Vehicle percentagest 6772iST/ 971 376072072373 179
Growth rate: 0.0 % per year
Fleet size is 1000000. vehicles in base year 1986
Repair Costs
-81
T IM240 repair - $ 150. NOx repair- S 0. )
Avg. I/H repair cost : S SO. 75.
Air Pump repair cost : $ IS. 15.
Catalyst replacement cost; $ ISO. 165.
MisCueled catalyst cost : S 175. 190.
Evap. system repair cost i $ 5. 5.
PCV system repair cost : S 5. 5.
Gas cap repair cost : $ 5. 5.
Purge repair cost : 9 70. 70.
Pressure repair cost s $ 38. 38.
Benefits (1000 tons/yr) and Costs (1000 $/yr) are averaged over the calendar years 2000 thru 2000.
Calendar Year 2000
Baseline No-Program Emissions:
TotVOC
c
ExhHC
10.929
21.955 123.217
Program Benefits and Costs
Evap ReCuel
4.024
RstLos
3.897 1.038
voc
CO
cost
Average Annual ATP Benefit, Fee 0.287 1.684 HTT
Evap/RunLoss Benefit, Fee 2.262 1765.
Evap/RunLoas MFC Benefit ( 1766.)
Avg Annual I/M Benefit, Fee 1.697 26.634 7280.
Avg Annual ATP Repair Cost 594.
Avg Ann. Prg/Pres* Repair Cost 1743.
Avg Ann. IM Repair Cost,FailRt 5192.
Average Annual IM Fuel Savings ( 3609.)
Exhaust Tampering Deterrence 0.210 2.315
0.0738
Evap
1.383
RnLoss
0.878
ExhHC
2.194
Ann Avg 4.456 30.633 11628.
( 11075 76136 kilograms per day )
-------
Mobile, output from GEM 4.1 1/24/92
Run MO High Option (Biennial)
MOBIlE4.1(4Nov91)
M120 Warning:
MOBILE4.1 does not model most 1993 and later clean Air Act
requirements; Emission factors for CY 1993 or later are affected.
I/M program selected:
Start year (January 1):
Pre-1981 MYR stringency rate:
First model year covered!
Last model year covered:
Waiver rate (pre-1981):
Waiver rate (1981 and newer):
Conpliance Rate:
Inspection type:
Inspection frequency
Vehicle types covered:
1981 4 later MYR test type:
1983
20%
1968
2020
1.%
1.%
98.%
Centralized
Biennial
LDGV - Yes
LDGT1 - Yes
LDGT2 - Yes
HDSV - No
2500 rpm / I
MOBILE*.1 IM240 Transient Biennial I/M Credits .8/15 (8/5/91)
Oldest model year covered by:
IU240 Transient Test 1984
Purge System Check 1984
Fuel System Pressure Check 1971
Anti-tampering program selected:
Start year (January 1):
First model year covered:
Last model year covered:
Vehicle types covered:
Type:
Frequency:
Compliance Rate:
Air pump system disablements t
Catalyst removals:
Fuel inlet restrlctor disablements:
Tailpipe lead deposit test:
EGR disablement:
Evaporative system disablements:
PCV system disablements:
Missing gas caps:
1983
1975
2020
LDGV , LDGT1, LDGT2
Centralized
Biennial
98.0%
Yes
Yi
Yi
Y.
Ho
Ye
Ye
Yi
Minimum Temp: 72. (F)
Period 2 RVP! 8.7
Period 1 RVP: 11.5
VOC HC emission factors include evaporative HC emission factors.
Maximum Temp: 92. (F)
Period 2 Start Yr: 1992
Cal. Year:
2000
I/M
Anti-tarn.
Program:
Program:
Yes
Yes
Ambient
Operating
Temp: 87.5
Model 20.6
/ 87.5
/ 27.3
/ 87.5
/ 20.6
(F) Region:
Altitude:
Low
500.
Ft.
Veh. Type:
LDGT2
All Veh
Veh. Speeds:
VMT Mix:
19.?
0.584
Composite Emission Factors
VOC HC:
Exhaust HC :
Evaporat HC:
Refuel L HC:
Runing L HC:
Rsting L HC:
Exhaust CO:
Exhaust NOX:
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
1.16
0.48
0.16
0.19
0.23
0.11
6.32
0.72
I/M
Anti-tarn.
LDGV
19. g
0.584
Composite Emission Factors
voc HC:
Exhaust HC:
Evaporat HC :
Refuel L HC:
Runing L HC:
Rsting L HC:
Exhaust CO:
Exhaust NOX:
1.77
0.71
0.36
0.19
0.42
0.11
10.05
0.75
19.?
0.199
(Gin/Mile)
1.47
0.80
0.15
0.25
0.18
0.10
7.06
1.05
Program:
Program:
LDGT1
19.?
0.199
(Gm/Mlle)
2.25
1.20
0.35
0.25
0.36
0.10
11.96
1.15
" 19.?
0.080
1.67
0.89
0.23
0.25
0.20
0.10
8.20
1.13
No
No
LDGT2
' 19.?
0.080
2.59
1.37
0.46
0.25
0.40
0.10
14.40
1.25
1.53
0.82
0.18
0.25
0.19
0.10
7.38
1.07
Ambient
Operating
LDGT
2.35
1.25
0.38
0.25
0.37
0.10
12.66
1.18
19. S
0.035
4.54
2.07
1.44
0.40
0.51
0.12
36.27
4.51
Tempi 87.5
Mode: 20.6
HDGV
19.?
0.035
4.54
1.07
1.44
0.40
0.51
0.12
36.27
4.51
19.?
0.002
0.65
0.65
1.60
1.37
/ 87.5 /
/ 27.3 /
L0DV
~ 19.?
n.002
0.65
0.65
1.60
1.37
19
0
0
0
1
1
87.5
10.6
.?
.001
.84
.84
.74
.50
197?
0.093
2.15
2.15
11.21
9.30
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
1.495
0.796
0.206
0.192
0.203
0.098
8.223
1.750
(F) Region: Low
Altitude: 500.
LDDT
19
0
0
0
1
1
.001
.84
.94
.74
.50
HDDV
19.6
0.0?3
2.15
1.15
11.21
9.30
rt.
Mr-
lS. 6
n.oo7
4.96
l.?9
2.52
0.54
24.78
0.77
Ml Vsh
S.084
1.049
11.382
0.192
0.364
0.098
11.874
1.799
-------
Mobile output from GEM 4.1. 1/24/92
Run *5 Medium Option
MOBILE4.l(4Nov91>
M120 Warning:
MOBILE4.1 does not model most 1993 and latar Clean Air Act
requirements; Emission Factors for CY 1993 or later are affected.
I/M program selected:
Start year (January 1):
Pre-1981 MYR stringency rate:
First model year covered:
Last model year covered:
Halvar rate (pre-1981):
Waiver rate (1981 and newer);
Compliance Rate:
Inspection type:
Inspection frequency
Vehicle types covered:
1981 i later MXR test type:
Oldest model year covered by:
IH240 Transient Test
Purge System Check
Fuel System Pressure Check
Anti-tampering program selected:
start year (January 1):
First model year covered:
Last model year covered:
Vehicle types covered:
Type:
Frequency:
Compliance Rate:
1983
20%
1968
2020
1.%
1.%
98,%
Centralized
Annual
LDGV - las
LDGT1 - Yes
LDGT2 - Yes
HDGV - No
Idle
2020
2020
1971
1983
1981
2020
LDOV , LDGT1, LDCT2
Centralized
Annual
98.0%
Air pimp system disablements: No
Catalyst removals: Yes
Fuel inlet restrictor disablements: Yes
Tailpipe lead deposit test: No
EGR disablement: No
Evaporative system disablements: No
PCV system disablements: No
Missing gas caps: No
Period 1 RVP; 11.5
Minimum Temp:
Period 2 RVP:
72.
8.7
(F)
Maximum Temp: 92. (F)
Period 2 Start Yr: 1992
VOC HC emission factors include evaporative HC emission factors.
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
I/M
Anti-tarn.
LDGV
19.?
0.584
Composite Emission Factors
voc HC:
Exhaust HC:
Evaporat HC:
Refuel L HC:
Rnnlng L HC:
Rstlng L HC:
Exhaust CO:
Exhaust NOX:
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
1.35
0.52
0.21
0.19
0.32
0.11
7.01
0.72
I/M
Antl-tam.
LDGV
19.6
0.584
Composite Emission Factors
VOC HC:
Exhaust HC:
Evaporat HC:
Refuel L HC:
Runlng L HC:
Rstlng L HC:
Exhaust CO:
Exhaust NOX:
1.77
0.71
0.36
0.19
0.42
0.11
10.05
0.75
Program:
Program:
LDGT1
19.?
0.199
(Gm/Hlle)
1.66
0.86
0.20
0.2S
0.26
0.10
8.07
1.05
Program:
Program:
LDGT1
19.?
0.199
(Gm/Mlle)
2.25
1.20
0.3S
0.25
0.36
0.10
11.96
1.15
Yes
Yes
LDGT2
" 19.?
0.080
1.90
0.97
0.29
0.25
0.29
0.10
9.48
1.13
No
No
LDGT2
' 19.?
0.080
2.59
1.37
0.46
0.25
0.40
0.10
14.40
1.15
Ambient
Operating
LDGT
1.73
0.89
0.22
0.2S
0.27
0.10
8.47
1.07
Ambient
Operating
LDGT
2.35
1.25
0.38
0.25
0.37
0.10
12. ««
1.18
Tempi
Mode:
87.5 /
20.6 /
HDGV
19
0
4
2
1
0
0
0
36
4
Temp:
Mode:
.6
.035
.54
.07
.44
.40
.51
.12
.27
.51
87.5 /
20.6 /
HDGV
19
0
4
2
1
0
0
0
36
4
.6
.035
.54
.07
.44
.40
.51
.12
.27
.51
87.5 / 87.
27.3 / 20.
LDOV
19.?
0.002
0.65
0.65
1.60
1.37
87.5 / 87.
27.3 / 20.
LDDV
19.?
0.002
0.65
0.65
1.60
1.37
5 (F) Region:
6 Altitude:
LOW
500.
LDDT HDDV
19.? 191
0.001 0.
0.84 2.
0.84 2.
1.74 11.
1.50 9.
5 (F) Region:
6 Altitude:
?
093
15
15
21
30
Low
500.
LDDT HDDV
19.6 l?~
0.001 0.
0.34 2.
0.84 :.
1.74 11.
1.30 5.
?
093
15
15
21
JO
Ft.
MC
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
Ft.
MC
19.6
0.007
4.96
1.99
2.52
0.54
24.78
n.77
All Veh
1.661
0.838
0.252
0.192
0.281
0.098
3.927
1.752
All Veh
2.084
1.049
0.3»2
0.192
0.364
o.o«e
11.874
1.799
-------
Mobile output from GEM 4.1 1/24/92
Run *3 : low Option
MOBILE4.1UNOV91)
M120 Warning i
MOBILE4.1 does not model most 1993 and later Clean Air Act
requirements; Emission Factors for CY 1993 or later are affected.
I/M program selected:
Start year (January 1):
Pre-1981 MYR stringency rate:
First model year covered:
Last model year covered:
Waiver rate (pre-1981):
Waiver rate (1981 and newer):
Compliance Rate:
Inspection type:
Inspection frequency
Vehicle types covered:
1981 C later MYR test type:
Oldest model year covered by:
IM240 Transient Test
Purge System Cheek
Fuel System Pressure Check
Anti—tampering program selected:
Start year (January 1):
First model year covered:
Last model year covered:
Vehicle types covered:
Types
Frequency:
Compliance Rate:
1983
20%
1968
2020
1.%
1.%
98.%
Centralized
Annual
LDGV - Yes
LDGT1 - yes
LDGT2 - Yes
HDGV - No
Idle
2020
2020
2020
1983
1981
2020
LDGV , LDGT1, LDGT2
Centralized
Annual
98.0%
Air punp system disablements: No
Catalyst removals: Yes
Fuel inlet restrlctor disablements: Yes
Tailpipe lead deposit test: No
EGB disablement: No
Evaporative system disablements: No
PCV system disablements: No
Missing gas caps: No
Minimum Temp: 72. (F)
Period 2 RVP: 8.7
Period 1 RVP: 11.5
voc HC emission factors include evaporative HC emission factors.
Maximum Temp: 92. (F)
Period 2 Start Yr: 1992
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
I/M
Anti-tarn.
LDGV
":!M
Composite Emission Factors
VOC HC:
Exhaust HCi
Evaporat HCi
Refuel L RC:
Runing L HC:
Rsting L HC:
Exhaust CO:
Exhaust NOXt
Cal. Year: 2000
Veh. Type:
Veh. Speeds:
VMT Mix:
1.58
0.52
0.36
0.19
0.41
0.11
7.01
0.72
I/M
Anti-tarn.
LDGV
19.?
0.584
Composite Emission Factors
VOC HC:
Exhaust HC:
Evaporat HC:
Refuel L HC:
Runing L HC:
Rsting L HC:
Exhaust CO:
Exhaust NOX:
1.77
0.71
0.36
0.19
0.42
0.11
10.05
0.75
Program:
Program:
LDGT1
19.?
0.199
(Gm/Mlle)
1.90
0.86
0.34
0.25
0.36
0.10
8.07
1.05
Program:
Program:
LDGT1
19.?
0.199
(Gm/Mlle)
2.25
1.20
0.35
0.25
0.36
0.10
11.96
1.15
Yes
Yes
LDOT2
" 19.?
0.080
2.18
0.97
0.45
0.25
0.40
0.10
9.48
1.13
No
No
LDOT2
' 19.?
0.080
2.59
1.37
0.46
0.25
0.40
0.10
14.40
1.25
Ambient
Operating
LDGT
1.98
0.89
0.37
0.2S
0.37
0.10
8.47
1.07
Ambient
Operating
LDOT
2.35
1.25
0.38
0.25
0.37
0.10
12.66
1.18
Tempi 87.5
Mode: 20.6
HDGV
19.?
0.035
4.54
2.07
1.44
0.40
0.51
0.12
36.27
4.51
Temp: 87.5
Mode: 20.6
HDGV
19.?
0.035
4.54
2.07
1.44
0.40
0.51
0.12
36.27
4.51
/ 87.5 /
/ 27.3 /
LDDV
~ 19.?
0.002
0.65
0.65
1.60
1.37
/ 87.5 /
/ 27.3 /
LDDV
~~19.?
0.002
0.65
0.65
1.60
1.37
87.5
20.6
(F) Region: Low
Altitude: 500.
LDDT
19
0
0
0
1
1
87.5
20.6
.001
.84
.84
.74
.50
HDDV
o!o93
2.15
2.15
11.21
9.30
Ft.
MC
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
All Veh
1.870
0.838
0.379
0.192
0.363
0.098
8.927
1.752
(F) Region: Low
Altitude: 500.
LDDT
19
0
0
0
1
1
.3
.001
.94
.94
.74
.50
HDDV
197?
0.093
2.15
2 .13
11.21
9.30
Ft.
MC
19.6
0.007
4.96
1.89
2.52
0.54
24.78
0.77
All Veh
2.084
1.049
0.382
0.192
0.364
0.09S
11.874
1.799
-------
APPENDIX L
IDENTIFYING EXCESS EMITTERS WITH A REMOTE SENSING DEVICE:
A PRELIMINARY ANALYSIS
-------
911672
Identifying Excess Emitters with a Remote
Sensing Device: A Preliminary Analysis
Edward L. Glover and William B. Clemmens
U.S. Environmental Protection Agency
ABSTRACT systems work by focusing a beam, or in some
cases multiple beams, of infrared light across
There has been considerable interest in the roadway into an infrared detector. The
applying remote measuring methods to sample instrument determines the concentration of
in-use vehicle emissions, and to characterize the pollutant in the path of the beam based on
fleet emission behavior. A Remote Sensing the amount of infrared light absorbed by the
Device (RSD) was used to measure on-road detector at specified wavelengths, and the
carbon monoxide (CO) emissions from theoretical relationship of carbon monoxide to
approximately 350 in-use vehicles that had carbon dioxide in auto exhaust. Early
undergone transient mass emission testing at a equipment had only carbon monoxide (CO)
centralized I/M lane. On-road hydrocarbon capability, while later equipment may have
(HC) emissions were also measured by the RSD the potential to measure other emissions
on about 50 of these vehicles. Analysis of the species (primarily hydrocarbons - HC).
data indicates that the RSD identified a
comparable number of the high CO emitters as Two potential uses of remote sensing
the two speed I/M test only when an RSD devices (RSD) have been proposed by various
cutpoint much more stringent than current sources. They are: (1) the on-road
practice was used. Both RSD and I/M had identification of gross polluting vehicles
significant errors of omission in identifying (commonly called High Emitters) for
High CO Emitters based on the mass emission subsequent repair, and (2) the monitoring of
test. The test data were also used to study the on-road vehicle emissions in a specified area
ability of the RSD to characterize fleet CO over a period of time for program evaluation
emissions. purposes. Such potential uses can only come to
fruition, however, if the remote sensing
INTRODUCTION techniques are shown to provide accurate and
reliable results that reflect the vehicle's true
Researchers at the University of Denver emission levels over a variety of normal
and elsewhere [1]*, [2], [3], [4] are in the operating conditions (e.g., acceleration, cruise,
process of developing systems to remotely hot/cold operation, etc.).
measure the concentration emissions from
vehicles operating on the public roads. These Several test programs wuh remote sensing
devices have been conducted by a variety of
—— organizations including; the University of
* Numbers in brackets denote references listed at the Denver (by Professor Donald Stedman),
end of the paper. General Motors [4], the EPA Environmental
-------
911672
Monitoring and Support Laboratory (EMSL) in emissions from a given vehicle would
Las Vegas [2], and the California Air Resources categorically match IM240 mass emissions
Board (CARS) [5]. Many of these studies have from the same vehicle. For example, it was of
involved measuring the RSD concentration interest to know if a vehicle with a high RSD
emissions from thousands of vehicles [1], [3]. concentration would also have high IM240
Others were concerned with verifying the mass emissions, and vice versa. The
accuracy of the RSD measurements by effectiveness of the RSD system in identifying
directing small samples of exhaust with a gross emitters and excess emissions was also
known concentration into the RSD beam and evaluated relative to the capability of the
comparing the results to more traditional standard Indiana I/M test using the IM240 as
analyzers [2]. Another involved a roadside the yardstick for excess emissions. Finally, the
pull-over program conducted by CARB which effect of external variables, such as vehicle
compared RSD results to I/M tailpipe and operating mode, owner response, weather
tampering checks on in-use vehicles [5]. conditions, and siting factors were also
However, none of these programs provided a investigated as to whether they affect the
quantitative comparison between RSD results ability of the RSD system to correctly identify
and corresponding transient dynamometer vehicles with high emissions.
mass emission results on the same set of cars.
All RSD and IM240 testing was conducted in
The goal of this study was to fill this gap Hammond, Indiana, by Automotive Testing
and to provide a comparison between RSD Laboratories Inc. (ATL) under contract to the
results and dynamometer results on the same U.S. Environmental Protection Agency.
cars. The driving schedule which was utilized
for the comparison was the IM240. It is a new TEST EQUIPMENT
transient driving schedule developed by the
EPA [8] that consists of the first two "hills" of Figure 1 shows a schematic of the RSD
the new car certification procedure, the system in a roadside setting. The principal
Federal Test Procedure (FTP), with some parts of the system include the IR detector and
modifications (further details are discussed source; a video camera to record the license
later in the paper). Although linear plates of passing vehicles; a modified police
correlations between the RSD and IM240 radar gun; a personal computer equipped with
emissions were performed, most of the an A/D board; and special software developed
investigation was focused on whether the RSD by the University of Denver researchers. Also
included in the system are
Fieurel l^e special calibration
REMOTE SENSING DEVICE SCHEMATIC S±T "" """' '"
IR SOURCE
VIDEO
1.0.
CAMERA
COMPUTER
The RSD system operates
by continuously monitoring
the intensity of the IR
source. When a vehicle
breaks the beam path, the
reference voltage drops to
zero which signals the
presence of the vehicle.
Span voltages collected
before the beam is blocked
and zero voltages during the
blockage are recorded. As
the vehicle exits the beam,
samples are taken over one
second at 125 Hertz. The C02
spectral region is isolated by
filtering at 4.3 |im and the CO
spectral region is filtered at
-------
911672
4.6 u,m. The instantaneous CO and CO2 values
are then regressed using a linear least squares
procedure. From the slope of this regression
the average CO/CO2 molar ratio (Q) and the
average HC/CO2 molar ratio (Q1) are obtained
and reported. The HC ratio is calibrated and
reported in terms of propane.
The average Q and Q' molar ratios over the
one second sample time are the only emission
measurements that, are recorded during an RSD
test. Only these are recorded because the size,
position and distribution of the exhaust plume
is not known. Therefore, to help safeguard the
quality of the subsequent emission
calculations which are based on these
theoretical relationships, the system employs a
built in feature to evaluate the reasonableness
of the observed molar ratio. If an
unreasonable molar ratio is observed, the RSD
reports a 'non-linearity error1.
The majority of testing was conducted using
the original RSD design developed by Dr.
Stedman. This unit (designated in the paper as
RSD #1) had a liquid nitrogen cooled detector,
and measured only carbon monoxide (CO). It
used two indium antimonide photovoltaic
detectors. A second unit was used in the later
stages of testing. This second RSD unit
(designated as RSD #2) was air cooled, and had
both HC and CO measurement capability. The
two systems also had slightly different
versions of software controlling the data
acquisition and processing.
DESCRIPTION OF TESTING
Two different RSD testing formats were
employed during this study. These formats
included track testing at the ATL (Bendix) test
facility in New Carlisle, Indiana, and on-road
testing of in-use vehicles in Hammond,
Indiana. The RSD track testing was very
limited, lasting only three days at the
beginning of the test program, and employed
the first RSD unit (RSD #1). The purpose of the
track testing was to compare RSD results
collected under controlled track conditions
with FTP results on the same set of vehicles. A
second purpose was to evaluate the
repeatability of the RSD emissions by replicate
testing of the same vehicle under controlled
conditions. A total of ten (10) cars were tested
at two test-track sites. At the first test-track
site, five cars received multiple RSD tests on a
level roadway at speeds of 5. 10, 20, 30, and 40
MPH. At the second track site, five other cars
were tested several times on an inclined
roadway with a 3 percent grade at the same
speeds. All ten of the vehicles which
participated in this part of the RSD test
program had been recruited to the ATL
laboratory for other emission testing
programs and were simply selected for the RSD
testing based on availability. As a result of
participating in the other programs, these
vehicles underwent FTP testing and repairs.
However, none of the vehicles received an RSD
test after repair.
RSD testing at roadside sites of in-use
vehicles, which had received a transient
dynamometer test, was the major effort of this
study. This RSD testing was conducted at two
expressway ramp sites, and a secondary street
with two-way traffic. The majority of this on-
road testing was conducted at the expressway
sites, and employed the first design RSD unit
(RSD #1). The second design unit (RSD #2) was
used at the second test site, a fairly lightly
traveled, two-lane expressway service drive
with two-way traffic. Both RSD test sites were
located less than one mile from the specially-
modified centralized I/M facility in Hammond,
Indiana, which administered the transient
IM240 dynamometer test.
The IM240 is a new two bag mass emission
test conducted on a dynamometer over a
transient driving schedule. The driving
schedule for the IM240 test consists of the first
two "hills" of the new car certification test, the
Federal Test Procedure (FTP), with some
modifications to include more transient
operation. However, the driving schedule does
not include a vehicle engine-start. The
dynamometer inertia weights and horsepower
settings were selected based on a consolidated
list of the new car certification settings for the
vehicle model. The emissions were measured
using a CVS-CFV system with laboratory grade
emissions analyzers. For further information
on the IM240 consult EPA technical report
number EPA-AA-TSS-91-1 [8].
The vehicles which participated in the RSD
testing were a subset of vehicles selected for a
larger program designed to monitor the
emission performance of in-use vehicles using
the IM240 dynamometer test. These vehicles
were selected for IM240 testing as they entered
-------
4 911672
the Hammond I/M facility based on the On the third day of testing at this site, the
availability of the dynamometer, and the problem of vehicles slowing down was solved
owner's willingness to participate. If the by moving to a second site on the ramp about
vehicle owner declined to participate, that car 25 yards from the ramp exit. The road grade at
went through the normal Indiana I/M this point was also 2.75 degrees. Because the
procedure (idle and 2500 RPM no load), and the vehicles were about to enter the freeway, it
next car in line was chosen to receive an was more difficult for the the vehicle owners
IM240. Only after the vehicle had completed to slow down to look at the equipment while
the IM240 test was the owner approached about passing through the RSD beam. As a result, the
participating in the RSD drive-by test. Drivers average vehicle speed was more than 25 MPH
were asked to drive by the RSD site and were and vehicle operation was believed to be fairly
given a monetary incentive if they indicated representative of ordinary expressway ramp
that they would participate (cars were not driving.
stopped at the RSD test site itself). This
additional recruitment procedure for the RSD After receiving the new RSD unit (RSD #2),
testing was needed only for the expressway it was decided to move to a test site on a
sites, in that the secondary street site was secondary road with two-way traffic, about
located so that owner participation was blind. one-quarter of mile from the I/M lane. The
site was a straight, flat road with a posted speed
After receiving the IM240 test, the car was limit of 35 MPH. It was situated such that
inspected for catalyst tampering and somewhat more than half of the cars exiting
misfueling. It then received a pressure test of the I/M lane would pass through the test
its evaporative emission control system. This section. Thus, it was possible to conduct the
procedure took bnly a few minutes. However, RSD tests at this site without involving the
during this period the vehicles were turned- vehicle owner. The RSD device was set up so
off to conduct the evaporative pressure test, that the beam crossed both lanes of traffic.
The effect of this vehicle shutdown on
subsequent RSD emissions is unknown. OPERATIONAL EXPERIENCE
However, since the shutdown was brief, it is
expected to be minimal. The original goal of this test program was
to conduct RSD tests on all the vehicles which
The two expressway test sites were on the received IM240 tests. However, for various
northeast cloverleaf of the Cline Avenue and reasons not all the test vehicles received the
1-94 Interchange in Hammond, Indiana. The RSD test. As shown in Tables 1 and 2, described
cloverleaf was an entrance ramp leading onto more fully in the following sections, there
Westbound 1-94 from Northbound Cline were several reasons for vehicles not
Avenue. The geometry of the ramp was receiving the RSD test, including recruitment
circular with a slightly upward grade problems, RSD equipment problems, and
throughout. inclement weather.
Two different test sites on the expressway VEHICLE RECRUITMENT - As noted earlier,
ramp were used. The first site was about 50 the vehicle recruitment procedures differed
yards from the ramp entrance. The road grade between test sites. At the expressway site it
at this point was about 2.75 degrees. Also, was necessary to inform the owner of the RSD
because the ground surrounding the roadway test and offer a monetary incentive for
was relatively level, this site was convenient participation. In some cases, the owners
for setting up the equipment and parking a declined to participate, or in some instances,
supporting van. However, it proved to be agreed to participate (accepted the monetary
unsatisfactory because it allowed curious incentive), but never showed up for the RSD
vehicles owners to slow down while passing test. In other cases, they showed up, but were
through the RSD beam. Where it was estimated missed, because of communication problems
that typical average speeds on that section of from the IM240 lane to the roadside RSD
the ramp were around 15 to 25 MPH, most of operator, alerting the operator that a test
the test vehicles were found to be travelling at vehicle was on its way.
speeds of 5 to 15 MPH.
-------
911672
Sii£
X-way
2-lane
Combined
Table 1
Vehicle Recruitment
Owner Declined
IM240 /Vehicle Missed
407
673
71
12Q
191
Actual RSD
Participation
336 (83%)
146fSS%t
482 (72%)
Most of these RSD refusals to participate
were due to the. vehicle owner citing time
constraints or other reasons. Many of these
refusals occurred at the end of the day when
the I/M lane was backed-up. Unfortunately, it
was noticed that a few of the owners of
vehicles with very high IM240 emissions, or
with signs of tampering (fuel inlet restrictor
disabled) also tended to decline. In addition,
several vehicle owners declined because they
did not want to drive on an expressway.
As a result of owner refusals and operator
misses, only 336 vehicles out of 407 vehicles
actually drove by the expressway site where
an RSD measurement was attempted (see Table
1).
The second site was partially chosen so as
to avoid the need to inform the vehicle owners
about the RSD test. Thus, direct recruitment
was not necessary, and vehicle owner
involvement was minimized. This helped
insure that vehicle owners would not change
their driving patterns in the RSD test section,
and avoided the issue of some owners with
high emitting vehicles declining to participate
in the RSD test. The other reason for choosing
a second site was to obtain results at a location
that would be expected to have different
vehicle operation.
Unfortunately, as shown in Table 1, a high
RSD test rate was not achieved at this site
either. This was due to the fact that the
location of the second site allowed IM240
vehicles to exit the I/M test lane in two
directions. Only one of these directions took
the vehicle by the RSD site. Therefore, as
indicated in Table 1, only 146 vehicles out of
266 (55%) participated in the RSD testing. The
other 120 vehicles were not involved because
they exited from I/M lane in the opposite
direction from the RSD test site, or they missed
the RSD test because of lack of communication
between the lane and the RSD operator.
RSD EQUIPMENT OPERATION - Operation
of both RSD systems was fairly straightforward
with the set-up and operating procedures
being nearly identical for both systems. The
only differences were due to software
upgrades, and the absence of cryogenic
cooling for the RSD #2 (two-lane testing)
system. Most of the basic set-up and operating
problems were addressed during the first few
days at the first on-road site (the initial
expressway location). These problems
involved the typical learning curve,
experienced when operating unfamiliar
equipment. Beyond that, testing was
reasonably routine, interrupted only by
occasional equipment problems, and inclement
weather.
Severe equipment problems with either
RSD units were infrequent. However, one
major problem did occur with the RSD #1. The
problem was the failure of a computer board
which controlled the video acquisition and
storage, and it resulted in a few days of down
time. Once diagnosed, the board was replaced
with a new one (provided by the supplier of
the RSD equipment). The second equipment
problem which resulted in a couple days
downtime was the theft of the roadside power
generating equipment. This occurred at the
two-lane site while RSD #2 was operating. It
illustrates a potential practical problem with
this type of emission testing. Other more
minor equipment problems which were
resolved by the equipment operator included:
(1) the heating element used to produce the
collimated RSD beam burned out several times,
(2) rain entered the detector and had to be
dried out using a heat gun, (3) high winds
potentially causing the instrument to
erroneously make a reading were present at
times.
The number of vehicles lost to RSD testing
due to equipment problems with each of the
RSD units is shown in Table 2. Overall,
approximately 16 percent of the RSD tests were
missed. However, many of these problems
were related to the emerging nature of this
technology, and as it matures, the number of
RSD tests lost to equipment problems would be
expected to decline drastically.
WEATHER EFFECTS - One of the objectives
of this study was to determine if the RSD could
operate in inclement weather such as rain or
-------
911672
snow. Track testing was
excluded from this
requirement. However, Actual RSD
the On-road RSD testing §itg Participation Malfunction
was attempted when x 336
inclement weather was 2-lane 145
present during the
expressway testing in
August, November, and
Table 2
Vehicle Capture Rates
Combined 482
Equip
28
21
52
Weather
27
27
True RSD Percent of w/o Equip
Sample
281
122
403
Participation Malfunct.
(84%) (91%)
(84%)
(89%)
December. Testing in inclement weather led to
some missed RSD tests. At the expressway site,
27 out of the 336 participating vehicles (8
percent) were clearly missed due to rain or
snow (see Table 2). Also, as indicated in the
instrument description, the RSD unit has a
built-in quality control algorithm that flags
illogical results as 'invalid readings'. Some of
the RSD results discussed in the next section,
and labelled 'invalid' may have been tested on
a rainy day. However, it was not possible to
segregate the recorded data in a manner that
would allow an answer to that question.
Therefore, the actual loss due to weather
conditions could have been higher than
indicated.
Testing of the RSD #2 unit at the two-lane
site was done in late March. During this period
no inclement weather testing was done due to
the generally fair weather conditions on all
but two days. During these two days, RSD
testing was not attempted because of
hailstorms, and the vehicles which received
IM240 tests on those days are not included in
Table 2 under weather losses.
In addition to lost test time, inclement
weather also resulted in many of the
equipment problems which were mentioned
above. Nevertheless, it was found that testing
could occur when very light rain was present,
although the number of non-linear CO/CO2
errors increased if the pavement was wet and a
large water "rooster tail" was created behind
the car that interfered with the IR beam. In
addition, the intermittent rain became a
nuisance since all the equipment had to be
covered and uncovered.
RSD SAMPLE - The true RSD
sample, shown in Table 3, was
computed by removing the
number of vehicles that were not
tested because of recruitment
problems, equipment problems,
or inclement weather. This true
sample (403 vehicles overall) contains all the
vehicles which were attempted to be tested by
the RSD under, as best we could determine,
reasonable conditions. The table includes both
the valid and 'invalid' readings based on these
measurement attempts. Invalid reading being
those flagged by the unit's internal quality
control algorithm.
Subsequent analysis of the RSD's capability
to identify High Emitters used only the valid
readings because traditionally, only valid I/M
and IM240 tests are used for such analyses. If
necessary, invalid I/M or IM240 tests can be
repeated. However, unlike I/M or IM240 tests
which can be easily repeated, RSD tests might
not be conveniently repeated. A RSD second
chance test would require the vehicle to drive
past the instrument again under as nearly as
identical conditions as possible. Since repeat
RSD testing was not in the program plan, it was
not done. Therefore, in assessing the RSD's
capability to identify High Emitters, an
argument could be made that the entire True
RSD Sample' (in Table 3) should be used, since
the 'invalid' sample may include some High
Emitters.
Based on consistency with past practices,
only the valid RSD readings in Table 3 were
used for subsequent analyses. However, this
decision should be reviewed in 'any future
programs relative to the type of vehicle
sampling planned, since the effect of using
the True RSD Sample' (as opposed to only 'valid
RSD tests') would have the effect of reducing
the effectiveness of the RSD unit in
identifying High Emitters.
Site
X-way
2-lane
Combined
True RSD
Sample
281
122
403
Table 3
RSD Sample
Valid Invalid
CO RSD CO RSD
257 24(9%)
22 23(23%1
356
47(12%)
Valid
HCRSD
52.
53
Invalid
RSD
69(57%)
69
-------
911672 7
To clarify, an invalid test result as tabulated TEST RESULTS
in Table 3, could have resulted for two reasons.
First, if the standard deviation of the CO/CO2 A description of each vehicle tested along
ratio, which is measured more than one with its test scores from the IM240, RSD, and
hundred times per second, was greater than 20 the Indiana I/M tests can be found in
percent, a flag would be set. This prevents reference [9].
rapidly changing CO/CO2 ratios from being
recorded as valid. Second, if the instrument HIGH EMITTER IDENTIFICATION
could not detect a CO plume, a flag would be set.
This prevents pedestrians, multiple axle The primary goal of this project was to
trucks, and other accidental beam blockages evaluate the ability of the RSD concept to
from being recorded as valid. Wet conditions consistently identify vehicles with excessi-ve
could cause both these RSD errors to occur, CO emissions as determined by transient
however, as indicated, this potential source for dynamometer testing. In other words, could
'invalid results' could not be separated from the RSD properly categorize vehicles (i.e., pass
other causes in the recorded data. or fail) based on their IM240 mass emission test
scores. RSD hydrocarbon measurements were
The number and percentage of valid and added later in the program, and were similarly,
invalid RSD CO results for each site suggest but less extensively, analyzed.
different levels of performance between the
two instruments. For example in Table 3, RSD To achieve this goal, the analysis focused
#1 shows a fairly low rate of invalid readings, on several areas. The first addresses a brief
Whereas, RSD #2 at the two-lane site shows a comparison and correlation of the IM240 with
high rate of invalid readings (approximately the Federal Test Procedure (FTP) results.
25 percent). This higher rate of invalid Logically, the next would be to address the
readings suggests that RSD #1 is more efficient ability of the RSD to categorically identify
and sensitive to low CO concentrations than vehicles as gross or high emitters versus
RSD #2, possibly due to the cryogenic cooling normal emitters. However, before the
of RSD #1. However, the longer RSD beam categorical analysis can take place, it is
length at the two-lane road (greater than 30 necessary to define the ground rules for High
feet) versus the expressway site (19 feet) may Emitters, and from this, identify the number of
also have contributed to the higher rate of High Emitters in the valid RSD sample. Only
invalid readings. then can the ability of the RSD to properly
categorize vehicles be assessed.
The rate of RSD HC invalid readings made
by RSD #2 (56 percent in Table 3) is IM240 VERSUS FTP - In order to better
unacceptably high, and suggests that the HC understand the usefulness of the IM240 as a
capability of the RSD in its current state of vehicle to evaluate the RSD results, a brief
evolution needs much improvement, comparison of the IM240 to the Federal Test
Potentially one of the problems with the RSD Procedure (FTP) is appropriate. All of the
HC channel is the very tight instrument testing to compare the IM240 to the FTP was
sensitivity required to measure relatively conducted in a separate test program, and a
dilute HC concentrations. According to the more extensive evaluation of those results will
manufacturer of the RSD unit, a difference of be the subject of some future report. However,
100 ppm HC produces a difference of only 1 mV for the purposes of this RSD report, a simple
by the detector. In addition, the fairly long regression of the data available should be
RSD beam length at the two-way road site made sufficient to make the point.
the already weak signal even weaker, and the
detection more difficult. Based on the A regression of the available IM240 and FTP
experience in this program with an admittedly data from the 300 vehicles in Figure 2 indicates
early design unit, the performance and the reasonably good correlation, as demonstrated
results from the HC RSD channel should be by the correlation coefficient (r2 = 0.76) and
viewed with caution. the slope (b = 0.99). More importantly, a
review of the specific vehicle data indicates
that the IM240 is excellent at identifying
vehicles with high excess FTP emissions. In
-------
8
911672
Figure 2
FTP Versus IM240
0 20 40 60 80 100 120 140 160 180 200
IM240 CO (g/mlle)
fact, for virtually all vehicles tested, where the
model year ranged between 1976 and 1989,
there were zero errors of commission by the
IM240 relative to- the new car standards on the
FTP (i.e., all cars that failed the IM240 at the
new car standards for HC, CO, or NOx, also failed
the FTP for the same pollutant). The almost
perfect categorical relationship between the
IM240 and the FTP was a little surprising
because the IM240 does not include the cold
start operation found in the FTP. Only
warmed-up operation occurs on the IM240, and
the results do not reflect the contribution of
any engine-start emissions (cold or hot). The
lack of cold operation on the IM240 should,
however, not pose a significant problem, since
all RSD measurements and all I/M test results
in this study were made on warmed up cars.
Therefore, based on the good linear
correlation of the IM240 to the FTP, and its
success at identifying gross CO and HC emitters,
it can be viewed as a representative substitute
by which to judge the RSD results and base the
effectiveness of the RSD systems.
DEFINITION OF A HIGH EMITTER - A
necessary part of the analysis needed to
identify High Emitters was to determine the
appropriate level with which to ascertain that
a vehicle was a High Emitter. Historically, the
specific levels have been related to a vehicle's
FTP emission levels, and EPA's modeling
programs for in-use fleets (MOBILE4) [10] have
used a factor of the mean plus two times the
standard deviation of the mean fleet emissions
as the High Emitter level. To determine High
Emitter cutpoints for the IM240, this same
procedure was applied to 135 vehicles which
received before and after repair IM240 tests
in another EPA test program. The resulting
cutpoints are shown in Table 4. As in
previous analyses of FTP results, this testing
showed that many vehicles whose before-
repair emissions exceeded these cutpoints
usually achieved substantial reductions in
emissions from repair. Vehicles whose
emissions were less than the cutpoints in
Table 4, often had no reductions or a very
small percentage reduction in emissions from
repair. In short, our practical experience has
been that vehicles with mass emissions above
such cutpoints can usually be repaired to a
value under the cutpoint quite effectively,
while those below the cutpoint have been
very difficult to repair in a cost effective
manner.
Our focus in this analysis was to evaluate
the ability of the RSD to find High Emitters
(HE) which could realistically be expected to
yield substantial emission reductions from
repair. A vehicle is defined as a HE if its IM240
emissions exceed the cutpoints shown in Table
4. The potential emission reductions from the
repair of these vehicles was defined as the
emissions in excess of me Table 4 standards
(i.e. greater than 10 g/mile CO for 1983+ Model
Year vehicles), and such reductions were
termed Repairable Excess Emissions (RPEE).
HIGH EMITTERS IN THE SAMPLE - The
cutpoints in Table 4 were used to identify High
Emitters (HE's) in the sample of valid RSD tests.
For instance, at the two sites there were a total
of 356 valid RSD CO tests conducted. Of that
total, 97 vehicles, or 27 percent, exceeded the
IM240 cutpoints in Table 4. Whereas, only 53
valid HC test were conducted at the two-lane
site. The analyses in the following sections use
the number of valid RSD test in Table 5 as the
base value to compute certain RSD
effectiveness rates.
Table 4
IM240 High Emitter* Cutpoints
Model Year Grp QQ HC
1976 -1980 30 g/mile 2.0 g/mile
1981-1982 20 g/mile 1.0 g/mile
1983 and Later 10 g/mile 1.0 g/mile
* Includes vehicles termed both "HIGH" and"SUPER"
Emitters in MOBILE4
-------
911672
Tables
High Emitters
Valid RSD IM240 CO
CO Sample Highs
Valid RSD IM240 HC
HC Sample Highs
X-way
2-lane
Comb
257
22
72 (28%)
25(25%)
356 97(27%)
52.
53
S
8
The model year distribution of the vehicles
in the 'combined sample' of valid CO tests in
Table 5 is shown in Figure 3. This figure
provides a finer breakdown of the overall
distribution and of the distribution of High CO
Emitters (HE). It also shows several interesting
points. First, the 1982 through 1985 model
years comprise almost 50 percent of the
sample, and they also include a large portion
of the HE vehicles. Second, the HEs make up a
discernible portion of most model year
samples, even some late model years. For
example, for the 1978 through 1985 model
years, typically 25 to 35 percent of the model
year sample are HE vehicles. For the 1989
model year, the percentage is between 5 and 10
percent.
To further simplify subsequent analyses,
the three model year groups in Table 4 were
combined into two groups -- a 1976 to 1982
group, and a 1983 and later group. However,
the cutpoints in Table 4 were still used to
define High Emitters within even the
consolidated groups.
By grouping the 97 HE CO vehicles
(combined sample in Table 5) into these two
model year groups, it can be shown that the
Figures
Model Year Distribution
20
18
£ 16
£ 14
Total Sample
IM240 High Emitters
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
Model Year
older 1976-82 sample contributes less than half
(45 percent) of the total HEs for CO. This older
vehicle group also only contributes 46 percent
of the vehicles to the overall sample. The
closeness of the fraction of HE's to that for the
overall sample distribution indicates that the
older vehicles did not contribute a
disproportionate share of the HE vehicles in
this program.
RSD IDENTIFICATION OF HIGH EMITTERS
- The analysis of the RSD's ability to properly
identify High Emitters focused primarily on CO
emissions, since a combined total of 356 valid
RSD CO tests were conducted at the two sites. A
lesser emphasis was placed on HC emissions,
because only 53 valid HC measurements were
obtained at the one site. These analyses in most
instances evaluated the CO sample in the two
distinct model year groups (i.e., 1976 - 1982
model years, and 1983 and later model years).
However, due to the small size of the HC sample,
all model years were grouped together.
The primary data set used in the analysis
was formed by combining the expressway (X-
way) and two-way street (2-lane) data sets
together. This combination was done in order
to produce a larger, and potentially more
representative database. However, limited
separate analyzes were conducted on the
results from the two principal roadside sites to
illustrate any potential differences, and to
determine if combining the data masked any
important results.
RAW DATA - The first analysis was to scatter
plot the RSD concentration data versus the
IM240 mass emission data. Figures 4(a) and
4(b) show the combined sites CO results for
each model year group. Figure 4(c) shows
the HC results for all model years. In
viewing these scatter plots, there appears to
be more of a relation between the RSD CO
data and the IM240 values for the older 1976
- 1982 model year group than for the 1983
and later group. In any case, the regression
correlation coefficient is marginal in the
case of CO, and extremely poor in the case of
HC. Scatter plots, of course, provide only an
overall relationship, and Figure 4 does not
provide any information on the ability of
the RSD to properly categorize a vehicle as a
High Emitter on the IM240, or as a Normal
Emitter.
-------
10
911672
Figure 4
RSD - IM240 Scatter Plots
(Combined Sites)
to.o
(a)
, 0 41983 and Lmt«r Vehicle* J
20 40 60 M 100 120 140 160 180 200
IM240 CO (a/mlto)
10.0
(b)
20 40 60 80 100 120 140 160 180 200
IM240 CO (o/mito)
(c)
2 3
4 S 6 7 8 9 10 11 12
IM240 HC (g/mito)
Figura 5
RSD Identification Rate
(e)
RSD HIGH EMITTER IDENTIFICATION RATE - The
ability of the RSD to correctly identify vehicles
with high IM240 mass emissions as High
Emitters is a central issue in the evaluation of
the performance of the RSD concept. The
ability of the RSD to identify such vehicles is
shown in Figure S as a function of RSD CO
level. A detailed table of these values can be
found in reference [9]. The identification rate
in Figure S is defined as the ratio of the
number of IM240 High Emitting vehicles
identified by the RSD- to the total number of
High Emitting vehicles identified by the
IM240. If the RSD were to identify all of the
High Emitters identified by the IM240, the
identification rate would be 100 percent.
The CO High Emitter rates are separated by
site to show any differences that might be
apparent between the results. Differences
were possible because a different RSD
instrument was used at each site, and both sites
had different traffic characteristics, (e.g.,
vehicle speeds and accelerations).
Nevertheless, a comparison of the RSD CO High
Emitter identification curves from both sites
(Figures Sa and Sb) illustrates that roughly the
same results at both sites were obtained over a
complete range of RSD CO levels. This is
especially true in the case of the 1983 and later
vehicles, where the difference in
identification rates between sites is typically
only 5 or 10 percent (Figure Sa). The site
effect in Figure 5(b) for the 1976 through 1982
model years shows a slightly larger
difference between sites. However,
because the results from both sites were
rather consistent for each model year
group, the RSD data from both sites were
combined for most of the* analyses
throughout the remainder of the paper.
Combining the RSD results by site
produced a larger data set for analysis. A
further combining of the database across
the two model year groups would produce
a still larger data set. Based on the
proportional distribution of High
Emitters by model year in Figure 3, such
combination would seem appropriate,
except possibly for some late model years.
However, Figure 5(c) shows that the RSD,
at all CO levels, systematically identifies a
higher proportion of 1976 though 1982
High Emitters than 1983 and later High
Emitters. Typically, the difference is
-------
911672
11
around 10 or 20 percent. This is evidence that
although the distribution of High Emitters is
similar for most individual model years, the
ability of the RSD to find a High Emitter is not
the same, and is a function of model year, or
model year group. Thus, combining the two
model year groups for subsequent analysis is
probably not appropriate.
Unlike the CO results, no sample
stratification was done for the HC results
because the RSD .HC sample was relatively small
(only 53 valid readings), and because there
were very few High HC Emitting vehicles in
the 1976 through 1982 model year range. For
example, only one vehicle in the 1976 through
1982 model year group was found to be a High
Emitter. Instead, the High Emitter
identification rate as a function of HC level was
computed using an 'all' model year sample.
This information is shown in Figure 5(d).
RSD OUTPOINT ANALYSIS - The RSD CO and HC
levels shown in Figure 5 can also be used as a
standard to pass or fail vehicles. The exact CO
or HC values (called cutpoints) which are
chosen, would reflect the desired severity of
the test, and would determine the number of
High Emitting vehicles which would fail. The
selected values are usually chosen based on the
results of a cutpoint analysis designed to
maximize the number of High Emitters found,
and minimize the number of Normal Emitters
(i.e., not High Emitters) which might be falsely
failed. For this cutpoint analysis, the RSD CO
cutpoints range from the very tight 0.05% RSD
CO cutpoint to the loose 7.50% RSD CO cutpoint.
For RSD HC, the cutpoints ranged from 50 ppm
HC to 2500 ppm HC (Hexane equivalent).
The first part of this cutpoint analysis
simply compares the number of IM240 High
Emitting vehicles identified by the RSD to the
total number of vehicles with an RSD CO level
above a given cutpoint. As seen in Figure 6,
the RSD in all cases identifies more cars than
are truly High Emitters on the IM240.
At cutpoints above 3% CO in Figure 6(a) for
late model cars, there appears to be reasonably
close agreement in identifying High Emitters
between the RSD and the IM240. However, at
the higher cutpoints fewer cars were
identified. As the CO cutpoints are tightened,
the the RSD and IM240 curves begin to diverge
dramatically, indicating that a great
percentage of these new vehicles captured by
the lower RSD cutpoints are, in fact, low IM240
emitters which should not be failed. For
example, at the extremely tight 0.10% RSD CO
cutpoint the number of 1983 and later RSD
failures was more than 140 vehicles, while the
number of truly High Emitting vehicles was
only about 35.
Similar comments can be made for the 1976
- 1982 model year group, except that the
divergence between the RSD identification and
the IM240 occurs at a higher RSD cutpoint, in
this case 5% CO (see Figure 6b).
Figure 6
Failures versus High Emitters
1883 + Vehicle* (Combined Sites)
RSD cars with
IM240 > lOg/mi
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
CO Cut Point (%)
100
1
76 .'82 Vehicle* (Combined Sites))
RSD cars with
IM240 > lOg/mi ,
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
CO Cut Point (%)
All Vehicles (Two-Lane Site)
RSD can with
IM240 > lOg/mi ,
500
1000 1500 2000 2500 3000
HC Cut Point (%)
-------
12
911672
The ability of the RSD to identify HC High vehicles that should be identified would be
Emitters is extremely poor. The curves in missed.
Figure 6(c) do not have similar shapes, and
tend to diverge even at high HC cutpoints. The effectiveness of the RSD test in terms of
Furthermore, the number of High HC Emitter High Emitter Identification (RSD). False
Identified is extremely low at all cutpoints Failures (Ec), and False Passes (Eo) is shown in
above 1000 ppm. Figure 7 for both model year groups. A
detailed table of combined site results for each
As seen in Figure 6, the RSD can falsely model year group can be found in reference
identify vehicles that are not High Emitters. [9].
Not shown in Figure 6 is the opposite issue of
High Emitting vehicles NOT identified by the As indicated in Figure 7(a) relatively few
RSD. Such errors, generally referred to as 1983 and later High Emitters were identified by
errors of commission (Ec) or False Failures, the RSD system (less than 30 percent) at high
and as errors of omission (Eo) or False Passes, to moderate cutpoints (7.5% RSD CO to 2.0% RSD
are inherent in the selection of a cutpoint to CO), while the False Pass rate was extremely
maximize the number of High Emitters found, high (greater than 70 percent). However, on
while minimizing the number of Normal the plus side, the False Failure rate was
Emitters falsely failed. generally quite low, with the exception of the
high rate around 6.0% RSD CO. This exception
False Failures were determined by occurred because the sample of vehicles with
subtracting the number of IM240 High very high RSD CO scores was so small that one
Emitting vehicles that were identified by the False Failure dominated the results. A
RSD from the total number of vehicles probably more representative False Failure
identified by the RSD, and then the sum was rate pattern can be seen in Figure 7(c),
divided by the total number
Figure?
RSD Identification Accuracy
IPO.O
10.0
of vehicles identified by
the RSD. In essence, this
calculation is a measure of
the fraction of cars
identified by the RSD that
are NOT High Emitters. A
large value would indicate
that many of the vehicles
identified are not High
Emitters, and should not y
have been identified. *• *•••
T? i n 5 40.0
False Passes were
determined subtracting the 20.0
number of IM240 High
Emitting vehicles that were
identified by the RSD from
the total number of High
Emitters identified by the
IM240, and then the sum
was divided by the total
number of High Emitters
identified by the IM240. g
This calculation is a S
measure of the fraction of oc
High Emitters NOT
identified by the RSD. A
large value, in this
instance, would indicate
that many High Emitting
Eo * False High Emitter PASS Rate by RSD
Ec « False High Emitter FAIL Rate by RSD
RSD * RSD Vehicle Identification Rate
19834- Model Years (Combined Sites)
0.0
100.0 -
19834- Model Ye
00.0 • •••
M.,
100.0
10.0
iO.O
1976 • 1982 Model Years (Combined Sites)
40.0-
20.0
0.00 1.00 2.00 3.00 4.00 5.00 8.00 7.00 8.00
RSD CO Cut Point (%)
(X-way Site)
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
RSD CO Cut Point (%)
(b)
All Model Years (Two-Lao* Site)
40.0
20.0
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
RSD CO Cut Point (%)
(c)
500 1000 1500 2000 2500 3000
RSD HC Cut Point (%)
(d)
-------
911672
constructed only from the RSD results at the
express way site. It shows that the false failure
rate for High Emitting vehicles is virtually
zero at the higher cutpoints.
Despite the exceptions, the overall results
from the 1983 and later vehicles suggest that
the RSD will not identify the majority of the
High Emitting late model vehicles. However,
in the instances when a vehicle is identified as
a High Emitter by the RSD (at the higher
cutpoints), it is usually correctly identified.
For example, at the frequently cited cutpoint of
4.5% RSD CO, less than 15 percent of the 1983
and later High Emitters were identified. This
means that 85 percent of the High Emitters
were missed by the RSD at 4.5% RSD CO causing
a False Pass rate of 85 percent. However, the
False Failure rate was only 10 percent of the
RSD failure rate, suggesting that virtually all
the late model RSD failures at 4.5% will be High
Emitters.
The 1976 through 1982 model years show
results in Figure 7(b) generally similar to the
1983 and later vehicles. However, they
typically had higher identification rates of
High Emitters, lower rates of False Passes, but
higher rates of False Failures at identical
cutpoints (to the 1983 and later model years).
These somewhat different results are likely
due to the inherently higher emission levels
of the older vehicles (because they were
certified to higher new car standards), and a
longer period of normal deterioration due to
age. They suggest that in order to be fair to all
model year vehicles, RSD cutpoints should be
based on model year or technology type,
instead of one single value such as 4.5% RSD
oa
Like the HC RSD results shown in Figures
5(d) and 7(d), the High Emitter Identification,
and the False Failure and False Pass rates
shown in Figure 7(d) were quite negative.
They indicate that at present, the RSD HC
channel cannot be used to accurately and
repeatedly identify High HC Emitters without
simultaneously falsely identifying many low
emitting vehicles. For example, even at the
extremely tight cutpoint of 25 ppm HC (Hexane
equivalent) the RSD identified about 80
percent of the HC High Emitters. At higher
and more reasonable cutpoints, the HC High
Emitter Identification rate dropped to around
13
10 percent. Despite this low rate, the False
Failure rate was extremely high.
The results also show that at virtually all
reasonable HC cutpoints the false failure rate
exceeded 70 percent of the total failure rate.
In fact, no cutpoint was identified which could
reduce the false failure rate below 70 percent
and still produce a High Emitter identification
rate of more than 20 percent. In addition,
during the test program one vehicle had IM240
HC emissions exceeding 11 g/mile, but recorded
an RSD HC score of 0 ppm HC. Therefore, based
on these findings in conjunction with the fact
that more than half of the RSD HC
measurements were invalid, strongly suggests
that at the present time, the RSD HC system
cannot make repeatable or accurate
measurements of in-use vehicle hydrocarbon
emissions.
Viewing the cutpoint analysis for CO on an
overall basis would indicate that the
percentage of High Emitters identified versus
the overall number of failures is fairly high at
cutpoints greater than 3.0% RSD CO. This
suggests that typically if a vehicle tests above
3.0% RSD* CO it is likely to be a High Emitter.
Overall this is a positive sign, since it suggests
that a moderate RSD CO cutpoint can be found
that will identify at least some High Emitters
without falsely failing a large percentage of
the fleet. On the negative side, however such a
cutpoint would also likely miss a large portion
of the true High Emitters. More High Emitting
vehicles could be identified by using lower
cutpoints, but that would increase the numbers
of False Failures to possibly unacceptable
levels. Such trade-offs could seriously affect
the use of the RSD as a screening tool for High
Emitters. One can only speculate, but it may be
that requiring a high RSD reading to have
been observed on several different days, or at
several different sites, before a vehicle was
classified as a failure would improve the
balance between proper and improper
failures.
REPAIRABLE EXCESS EMISSIONS IDENTIFIED BY
RSD - The previous section on cutpoint
analysis described the effectiveness of the
RSD system in terms of the number of High
Emitting vehicles identified by the RSD at
particular CO or HC cutpoints. An equally
important consideration is the amount of
excess emissions represented by the vehicles
-------
14
911672
identified by the RSD (i.e., which High
Emitters are failed). As indicated, this paper
addresses excess emissions as emissions above
the levels defined in Table 4. Defined here as
Repairable Excess Emissions, or RPEE, these
excess emissions represent the potential
emission reductions from those vehicles
identified by the RSD, if they were properly
repaired.
The RPEE was calculated for each cutpoint
as the sum of the excess IM240 emissions
from all High Emitting vehicles identified by
the RSD (i.e., those above the levels in Table
4), divided by the sum of the excess IM240
emissions from all of the High Emitting
vehicles. In short, it is the fraction of excess
emissions from all cars with mass emissions
above the levels in Table 4 that were
identified by the RSD.
The effectiveness of the RSD was analyzed
in terms of RPEE because the RPEE
identification rate reflects the actual
emission level of each High Emitter, whereas
the High Emitter identification rate accounts
only for the number of High Emitters
without regard for emission level. The
difference between the two effectiveness
parameters arises because of the effect that
vehicles with very high mass emissions can
have on the total emissions of all the vehicles
in a given sample. Typically, these vehicles
(called Super Emitters in MOBILE4) comprise
a large share of the excess emissions in a test
sample (or fleet), although they are usually
few in number. Further, such vehicles can
be readily repaired to moderate emission
levels. For these reasons, their correct
identification is of prime importance in the
control of repairable excess emissions, and in
the evaluation of the RSD system's ability to
identify vehicles with repairable excess
emissions.
The first order of analysis was to evaluate
the RPEE identification rate of the RSD. The
difference in model year groups can be seen
in Figure 8. The results show that at
cutpoints greater than 4.0% CO, the model
year differences are pronounced, with a
greater amount of RPEE identified for 1976
through 1982 vehicles than for 1983 and
later vehicles. The trend of identifying more
from the old cars than the newer cars was
previously noted in the rate comparisons of
High Emitter identification shown in Figure
5'(c). However, this trend occurred at all CO
cutpoints relative to the number of vehicles
identified. This previous analysis suggested
that the RSD was more effective at finding
1976 through 1982 High Emitters than 1983
and later ones. In contrast to this previous
analysis. Figure 8 shows similar RPEE rates
for both groups at tighter cutpoints,
suggesting that if appropriate cutpoints are
chosen, the RSD can be equally effective at
identifying the excess emissions from new
cars as well as old cars.
To better understand the true relationship
between the number of High Emitters
identified by the RSD, and the Repairable
Excess Emissions from those vehicles, the
data were plotted in Figure 9. A careful
examination shows that for the 1983 and later
vehicles, at cutpoints between 1% to 4% CO,
the RSD identified less than half of the High
Emitters. However, those few vehicles which
were identified contribute significantly to
the total repairable excess CO emissions. At
relatively tight cutpoints, the RSD seems to
find the worst of the 1983 and later High
Emitters, although, not all of them, since
even at extremely tight cutpoints, the RPEE
rate was only around 80 percent.
For the 1976 through 1982 vehicles
(Figure 9b), the story is a little different, and
possibly a little more positive. For this model
year group, the RPEE and High Emitter
identification rates appear to proportionally
track each other better. In addition, for
these vehicles, the RPEE rate seems to be a
more linear function of CO cutpoint,
suggesting a stronger relationship between
number of vehicles identified by RSD
emissions and repairable excess emissions.
Figure 8
Rapalrabl* Excass CO Emissions
(Modal Year Effect)
0.0
0.00 1.00 2.00 3.00 4.00 S.OO C.OO 7.OO t.OO
RSD CO Cut Point
-------
911672
15
ftgun*
Vehicle vs Excess CO Emissions
D
RSD Vchlek Identification Rate
1M3 Mid Later Model YUM
1*7«. US2Model Yean
• xoe 4M ua ut ua
RSDCOCMPotat
ue tm ua 4.o» ue ua 7.00
RSDCOCutPolm
missed by the RSD. Also, as
mentioned previously, the False
Failure rate (Ec) in identifying
the number of High Emitters was
around 10 percent of the
identification rate, and the False
Pass rate (Eo) was more than 85
percent of all of the High Emitters
in the model year group. This
means that 10 percent of the RSD
failures identified contribute
nothing to the RPEE. However, 85
percent of the High Emitters
which do contribute something to
the RPEE were not identified.
Like the High Emitter identification rate
shown in Figure 7, the RPEE rates also need to
be compared with the False Failure and False
Pass rates. This is done to see if a potential
balance between high RPEE rates and low
False Failure rates exists for RSD. To make
the comparison easier, the False Failure and
False Pass rates from Figures 7 were plotted
along with the RPEE rate as a function of CO
and HC cutpoint. The results of
this effort are shown in Figure
10, stratified by model year
group, site, and pollutant in an
analogous arrangement to Figure
7.
An optimal balance between a
high RPEE and a low False Failure rate is
more difficult to find in the 1976 through
1982 sample. Here, both the RPEE rate and
False Failure rate seem to be linear functions
of RSD CO cutpoint. However, if the goal is to
hold False Failures to less than 20 percent
then a 3.0% RSD CO cutpoint which identifies
60% of the RPEE seems reasonable. If False
Failures need to be almost zero, then a 6.0%
Figure 10
Repairable Excess Emissions Identification
|Eo = F«J
EcsFal
""BE*
= Fabe High Emitter PASS Rite by RSD
Falae High Emitter FAIL Rate by RSD
RSD Identification Rate of Repairable Excess Emissions
For 1983 and later vehicles
(Figure 10 a), the RPEE rate
(indicated as RPEE in the figure)
ranged from 10 percent at very
loose cutpoints to 70 percent at
about 2.0% RSD CO. Further
tightening of the RSD cutpoint
produced little additional RPEE,
but dramatically increased the
False Failure rate (Ec). This
increased False Failure rate
suggests that the 2.0% RSD CO
cutpoint may be the lowest
practical RSD cutpoint that
should be considered for the
later model year vehicles
(assuming that the False Failure
rate of 20 to 25 percent is
acceptable at the 2% cutpoint).
In contrast, the widely reported
4.5% RSD CO cutpoint had a RPEE
rate of about 25 percent,
meaning that at this cutpoint, 75
percent of the RPEE would be
iM.a
1983+ Model Yean (Combined Slta)
1976-1982 Model Yean (Combined Sites)
M.a
21.1
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 (.00
RSD CO Cut Point (%)
1M.I
1983+ Model Year* (X-waj Site)
0.00 1.00 2.00 3.00 4.00 5.00 (.00 7.00 8.00
RSD CO Cut Point (%)
(b)
All Model Years (Two-Lane Site)
0.00 1.00 2.00 3.00 4.00 5.00 8.00 7.00 1.00
RSD CO Cut Point (%)
(e)
500 1000 1500 2000 2500 3000
RSD HC Cut Point (%)
-------
16
911672
RSD CO cutpoint is more appropriate. For
comparison, the 4.5% RSD CO cutpoint
identified about 40 percent of the RPEE, and
surprisingly had a False Failure rate of about
20 percent.
Like the HC High Emitter identification
rates, the HC RPEE rates were extremely poor.
In fact they seemed to be worse in the sense
that the maximum HC RPEE rate was less than
50 percent of the total HC RPEE. However,
this was principally due to the 11 g/mile
IM240 HC emitter which had an RSD HC test
result of 0 ppm. Nevertheless, the HC RPEE
identification was very low for most
cutpoints, yet the False Failures were nearly
80 percent of the total failures. In all,
coupling these results with all the other low
High Emitter identification rates, and high
False Failure and False Pass rates, leads to the
conclusion that the HC RSD in its current
state will not accurately measure the HC
concentration in a vehicle's tailpipe with
sufficient accuracy to be able to determine if
a vehicle is truly a High Emitter and in need
of repair.
RSD and I/M
As indicated, the primary goal of this study
was to evaluate the ability of the RSD to
properly categorize High Emitters. However,
because the IM240 vehicles had also received
an official Indiana I/M test, the unique
opportunity to compare the RSD results with
I/M scores on the same vehicles was exercised.
The Indiana I/M test is a centralized
biennial contractor run I/M program [11]. A
two-speed idle test (2500 RPM and idle) is used
for 1981 and later vehicles. The 207(b)
cutpoint of 1.2% CO is used for the idle portion
of the test, and a 1% CO value is used for 2500
RPM. The 207(b) HC standard of 220 ppm is
employed at both speeds. Only the idle test is
used for 1976 through 1980 vehicles. The 1980
vehicles are tested against a 2% CO -cutpoint,
while the 1976 through 1979 model years use a
3.5% CO cutpoint.
Because the Hammond, Indiana I/M site was
also used to evaluate second chance tests and
preconditioning cycles as pan of a previously
completed program, all of the vehicles were
tested consistent with the recently published
EPA guidelines for short test procedures (12).
RAW DATA - Similar to the evaluation of
High Emitter identification by the RSD, the
first analysis here was to scatter plot the data.
Figures ll(a) and ll(b) address the I/M CO test
for the late model cars, while Figures ll(d) and
ll(e) provide an HC comparison on these same
vehicles. The idle test for the older cars is in
Figure ll(c). About the best that can be said
about these data is that the relation between
RSD and I/M scores is poor for CO, and
nonexistent for HC. However, it must be
remembered that the relationship of short
tests to the certification test is not based on a
relationship of scores, but a relationship of
pass/fail categories.
FIGURE 11
RSD - I/M Scatter Plot
linommaLju
<«>
o.a o.« o.« o.m i.o 1.2 1.4 i.» i.e a.o
I/M aOOO RPM T*at - HC (ppm x 1OOO)
All Mod* Ynn
I/M Ml* TMt . HC (ppm « 1
-------
911672
17
RSD VERSUS I/M - This
analysis consisted of
comparing the performance
parameters previously used to
evaluate the ability of the RSD
to identify High Emitters and
excess emissions with those
same parameters calculated for
the I/M tests. For simplicity,
RSD results are shown for only
three cutpoints, which include
the frequently referenced
4.5% RSD CO cutpoint, a
moderate 3.5% RSD CO cutpoint,
and a more stringent 2.0%
cutpoint. RSD HC results faired
so poorly in the scatter plots
that no attempt was made to compare them
with the HC I/M results.
The effectiveness of the RSD and I/M tests
in identifying High CO Emitters is shown in
Figure 12 for both the 1983 and later vehicles
and for the 1976 - 1982 vehicles. Examination
of Figure 12(a) for late model cars shows that
the RSD at moderate and high CO cutpoints is
generally no better than the two-speed I/M
test (i.e., 'Both' in Figure 12a) in identifying
High Emitters. In the extreme case, the RSD
found less than 15 percent of the High
Emitters at the highly reported 4.5% CO
cutpoint, while the two-speed I/M test
identified almost 25 percent of the High
Emitters. However, when the more stringent
RSD cutpoint of 2% CO was used, about 35
percent of the High Emitters were found.
For the 1976 through 1982 vehicles in
Figure 12(b), the I/M identification of High
Emitters was marginally better than the RSD
identification rate (at 4.5% CO). Relative to
the I/M identification, the RSD identification
improved substantially at tighter RSD
cutpoints. For example, at the 4.5% RSD CO
cutpoint, the identification rate of 1976
through 1982 High Emitters was only around
25 percent, slightly less than I/M rate,
however, at the 2.0% RSD cutpoint, almost 60
percent of the High Emitters were found.
(Note that the 2500 RPM test results are not
shown for the 1981 and 1982 model years in
the figure for the sake of consistency,
although, they would boost the High Emitter
identification rate of the I/M test somewhat
for the 1976 - 1982 model year group.)
FIGURE 12
High Emitter Identification Rate
(CO, Combined Sk*>
Mathematically, the RSD and I/M tests
were combined to determine if the
combination of tests would identify additional
High Emitters. As seen in Figure 12, there is
an increase in the identification rate,
suggesting that the two tests identify
different cars. This would seem reasonable
since the tests are conducted under
significantly different vehicle operating
conditions, and would be expected to find
problems which are particular to each
operating mode. From the analysis shown in
Figure 13, it is clear that a subset of different
vehicles are identified by each test. For the
1983 and later group, the additional RSD
identification is very marginal until the 2%
RSD cutpoint is applied. However, for the
older model year group, the additional RSD
identification is substantial at all cutpoints.
In fact, adding the I/M and 4.5% CO tests
together nearly doubles the identification
Figure 13
High Emitting Vehicle Identification Split
(CO only)
1983 and Lat*r Mr*
(Ule/ZSNI/MTett)
197*-1982 MVs
(Mhl/MTert)
I/M
RSD
I/M
RSD
4.5% RSD
3.5% RSD
2.0% RSD
TIYTAIf
13
13
13
•mrua
TOTAl.il
TOTAIJI
10
15
25
-------
18
911672
rate of High Emitters, further attesting that
different older vehicles fail each of the tests.
It should be pointed out. though, that the
RSD test for the newer cars was compared
against the two-speed idle test in Figure 13,
whereas the older cars were compared to
only the single speed idle test. The
differences in I/M test type could possibly
explain the improved identification of
additional older cars by the RSD. In addition,
the Indiana I/M CO cutpoint of 3.5% CO for
the older cars (1976-1979 models) may not be
as stringent as it should be. Tightening this
I/M cutpoint for older cars would increase
the number of High Emitters identified by
I/M such that the identification
improvements exhibited by the RSD in Figure
12(b) could possibly vanish.
Another important factor in any
comparison of I/M with RSD are False Failure
rates. False Failure rates were computed as a
percentage of the failure rate, and they
indicate the percentage of improper failures.
Combined with the High Emitter identification
rate, they are a measure of a test's ability to
find High Emitters, yet avoid failing
low and marginal Emitters.
described 'RSD CUTPOINT ANALYSIS.' With this
method, a 1983 and later vehicle failing the
RSD or I/M test, would only need an IM240 test
score less than 10 g/mile, rather than 3.4
g/mile, to be considered a False Failure. Also,
with this method, cars with IM240 scores
between 3.4 and 10 g/mile (labeled Marginal
Emitters) would be considered False Failures.
Both methods are shown in Figure 14, the
first in (a) and (b),and the second in (c) and
(d). The False Failure rates calculated in terms
of only the High Emitters (second method) are
substantially higher for both RSD and I/M
relative to the False Failure rates calculated by
method 1. However, the False Failure rate
percentages for the two methods were based
on different sample sizes. Therefore, a direct
mathematical comparison can not be made.
Even so, a qualitative comparison can be
made. In those cases where the the False
Failure rates for method 1 (Figures 14a and b)
are zero, it is probably safe to assume that all
of the False Failures determined by method 2
(Figures 14c and d) were Marginal Emitters
with IM240 scores between 3.4 and 10.0 g/mile.
HgurtU
Falsa Failure Rate*
For the comparison of False
Failures between RSD and I/M, two
methods were used. The difference
was in the level of IM240 score below
which a failed car would be defined as
a False Failure. The first method used
new car certification standards
applied to the IM240 results as the
boundaries for False Failures. For
example, if a 1983 and later vehicle
failed either the RSD or the I/M test,
and had an IM240 test score of less
than 3.4 g/mile (1983 new car
standard), it was considered a False
Failure. This method of determining
False Failures is similar to that used in
MOBILE4, except that MOBILE4 uses FTP
test scores (as opposed to IM240 test
scores).
The second method used to compute
the False Failure rate utilized the
IM240 High Emitter cutpoints shown
in Table 4 as the boundaries for False
Failures. This method is identical to
the method used in the previously
[tttrpintland mgli Emm»i« ]
1*3 .nd Utw Mod*! Yon _ l»7«throu«hl
Hp H.
n H H in i i
{ High Emltl»f« Only )
lM3indL.tn-Mod.IY.
1976 Ihrouih 19M Modd VMM
-------
911672
Given this assumption, it is likely that
a majority of the False Failures
identified by method 2, where the
method 1 False Failure rate is greater
than zero, were also Marginal $'
Emitters. As implied previously, ^
Marginal Emitters are more difficult *
to repair, and as a result, less likely to I
contribute to overall emission |
reductions. I
Interestingly, the combination of
the RSD and I/M tests shown in
Figure 14(a) shows a lower False
Failure rate than the RSD test alone.
This apparent anomaly is a result of
the sample size increasing, when
combining the RSD and I/M samples, faster
than the number of failures. In this
particular case, the I/M-only False Failure
rate was zero, therefore, the combined rates
were the same as that for the RSD. Thus, the
fraction of False Failures was lower for the
combined sample than for the individual
samples.
Another possible concern is the similarity
in magnitude of the High Emitter
identification rates in Figures 12(a) and 12(b)
to the False Failure rates in Figures 14(c) and
14(d). The similarity might suggest that both
tests tend to falsely identify a vehicle as a
High Emitter almost as often as they correctly
identify a High Emitter. However, it should be
remembered that the calculations for the High
Emitter identification rate were based on the
total number of vehicles identified as High
Emitters by the IM240, whereas the False
Failure rate was based on the number of
vehicles failing the RSD or I/M test. Because
the number of IM240 High Emitters and the
number of RSD or I/M failures were
significantly different, any numerical
similarity of identification rates and False
Failure rates is merely coincidental.
A characteristic that must also be evaluated
when comparing RSD to I/M are the False
Passes or missed vehicles. The False Pass rate
is the fraction of High Emitters not identified
by the RSD or I/M test relative to the total
number of High Emitters identified by the
IM240 (also defined near Figure 7). An
alternative calculation, in this case, would be
to subtract the identification rates in Figure
12 from unity.
19
FifunlS
Falsa Pass Rates
( High Emitter. Only)
(b)
False Pass rates are shown in Figure 15,
and as would be expected, more stringent tests
(i.e., two-speed idle test), or tighter RSD
cutpoints reduce the number of vehicles
missed. In examining both model year groups
in this figure, it is evident that the Indiana
I/M test missed fewer vehicles than the RSD at
a 4.5% cutpoint. Only at the most stringent
cutpoints did the RSD substantially reduce the
number of High Emitters missed by the I/M
test.
Finally, as in the analysis of the RSD's
ability to identify High Emitters, the
comparison of RSD to I/M needs to look at the
amount of repairable excess emissions (RPEE)
identified by both tests. The RSD values in
Figure 16 are the same as the individual RPEE
values in Figure 10. The I/M RPEE values were
calculated in a similar fashion, and are shown
in Figure 16. Clearly, the Indiana I/M test for
the newer vehicles identified more repairable
excess emissions than the RSD test at a 4.5% CO
cutpoint. However, at the most stringent
cutpoint, the RSD did better.
For the older cars, the RSD consistently
identified more excess emissions than the I/M
test. In this case, though, it should be noted
that the I/M test for these cars consists of only
a single speed idle test, and it is likely that the
identification rate of the I/M test would
increase if a two-speed idle test were used.
In reviewing the comparisons between RSD
and I/M in Figures 12 though 16, it is difficult
to see any substantial improvement over the
I/M test with the RSD. At the 4.5% RSD
cutpoint, the the I/M test for the new cars (i.e.,
-------
20
911672
Repairable Excess Emissions
the two-speed test) identified more High
Emitters, had lower False Failures, had lower
False Passes, and identified substantially more
excess emissions. At the tightest cutpoint, the
RSD did identify more cars and excess
emissions, with a correspondingly lower False
Pass rate, than the I/M test, however, it did so
at the expense of an increase in False Failures.
Lowering the cutpoint on the I/M test to
achieve the same False Failure rate would be
expected to produce similar increases in the
identification of High Emitters and excess
emissions with the I/M test.
For the older vehicle group, the RSD faired
a little better. At the 4.5% cutpoint, the single
speed I/M test and the RSD were roughly
comparable in number of cars identified, false
passes, and excess emissions. At this cutpoint,
the RSD did have a measurably lower False
Failure rate. At the lower cutpoints, the RSD
performed better than the single speed idle
test on the older cars with comparable False
Failures.
In reviewing the fictitious combination of
the I/M and RSD tests in these figures, some
interesting observations can be made. For late
model cars, the combination increased
identification somewhat over the I/M test, but
with the expected increase in False Failures.
However, for the older model years, the
combination drastically increased the
identification (number of vehicles, and excess
emissions) over the I/M test without a
substantial increase in False Failures. In fact,
the combined tests at the highest RSD cutpoint
(4.5% CO) increased the identification over
both the I/M and RSD tests while reducing the
False Failure rate relative to the I/M test.
Clearly, the fictitious
combination of I/M and RSD test
did better with the older cars
than than the newer cars, and
was better than either test alone.
However, even with the
combination, only 30 to 40
percent of the late model High
Emitters and 60 to 70 percent of
the older ones were found. This,
suggests that neither test is as
effective as it should be in
finding High Emitters. In the
case of I/M, there are several
possible methods to improve the
capture of High Emitters. As previously
indicated, tighter I/M cutpoints could be
used, or, a single speed idle test could be
replaced with a two-speed test - a two-speed
test could be replaced with a loaded test, etc.
The ultimate replacement, of course, would
be the IM240. However, for RSD, the
improvements in the identification of High
Emitters are less obvious, and seem to be
limited to decreasing the cutpoint, with a
near exponential increase in False Failures
(see Figures 6 and 7), or to reducing
measurement variability (discussed next).
REPEATABILITY ANALYSIS
A key requirement of any measurement
system is that it provide consistent results to
the same stimuli. In the case of RSD, Stephens
and Cadle [4], and Lawson et.al. [5], have shown
that under relatively normal conditions, the
RSD can measure with reasonable accuracy, CO
from a specially controlled vehicle. However,
other elements in the measurement-
consistency equation include the vehicle
characteristics, its operation, and weather
conditions.
A previous analysis of RSD data collected in
Chicago [6] suggested that there could be
considerable difference between RSD reading
on the same car on different days Cat about the
same time of day). To further explore this
area, several of these other elements that could
cause variability in the RSD results, and the
effect that these factors could have on the
ability of the RSD to accurately identify gross
emitters were analyzed. The track testing
primarily addressed the vehicle and its
operation. The analysis of the on-road data
addresses vehicle operation, weather effects,
-------
Figure 17
RSD Reproducibility
(Lmi RoW, M MPH, ATL TtH-Tnck)
6.00
4.00 • ••
3.00 •
0.00
-1.00
f FTPCO
[ 18.6 a/ml
+
•rtMMU.vwu.WAW.1v'
1
#738
"f FTPCO
I 3.20/lrt
•*
J * t
« \
4
#742
FTPCO Y
1.2 g/ml J
1
• 681
FTPCO
3.19/ml
{
L
«
T FTPCO
[ 20.5 8/ml
4
^ 4
) '
teao
TEST VEHICLE NUMBER
911672 21
and a possible combination of the two on
the exhaust plume behind a vehicle.
TRACK TESTING - The track testing
was conducted over a three day period at
the Bendix test track in New Carlisle, jf
Indiana by Automotive Testing ~c
Laboratories (ATL). The track testing was 1
the first opportunity for the ATL I
personnel to operate the RSD without the |
assistance of the developer. J 2.00 4.
o
The track testing consisted of driving o i.oo
several vehicles by the RSD multiple
times at the same speed to obtain replicate
RSD readings. Several speeds were used at
each of two sites - one a level road, and
the other a 3 percent uphill grade. A total
of 10 cars were tested, five at each site. percent CO all the way to 3% CO. This large
variation in RSD readings could be a potential
The speeds used during the RSD tests problem if a pass/fail cutpoint were to be
included 5, 10, 20, 30, and 40 MPH. The data selected that happened to be in the upper pan
were spotty at many of the these test speeds Of this range.
with numerous low voltage errors (i.e., no RSD
readings). The lack of readings due to the Conversely, vehicle #680, a very high FTP
automatic quality control feature was assumed vehicle, had a similar spread in RSD readings.
to be partly due to the lack of experience with but the lower range of #680's readings
this equipment by the operator. As a result, overlapped the high end of #742's readings.
only at 20 MPH was sufficient data collected to Based only on this observation, it would be
allow reasonable analysis of the difficult to discriminate the dirty car from the
reproducibility of the RSD measurements. ciean car under these operating conditions
with an RSD cutpoint in the 2% to 4% CO range.
Figures 17 and 18 were plotted using the if a typical vehicle showed the kind of RSD
RSD data collected at 20 MPH and the as- emissions distribution evident in this example,
received FTP from each vehicle. Figure 17 the outcome of an RSD test could be based
shows the repeatability of the RSD of tests done considerably on chance, or be a strong
on the level road, and Figure 18 shows those function of specific test conditions, instead of
from the 3 percent uphill grade. Both
figures provide a visual indication of the
variance among RSD scores on the same
vehicle at a constant speed. The
accompanying as-received FTP result
provides a convenient qualitative
comparison of the FTP mass emissions
with RSD concentration emissions.
Examination of Figure 17 shows
extremely high variability between RSD
measurements made under very similar
conditions on the level track. Based on
these RSD results, it is likely that for some
vehicles, the range of RSD results will fall
on both sides of a likely cutpoint. For
example, the RSD emissions of vehicle
#742, which had a low FTP score, varied
from a low reading of essentially zero
Figure 18
RSD Reproducibility
(UpWll R«W, 29 MPH, ATL Tot-Trad
6.00
TEST VEHICLE NUMBER
-------
22 911672
being based on the vehicle's FTP emissions or An obvious third reason is that different
level of repair. cars were used at each site, and the observed
variability was simply a matter of coincidence.
Examination of the uphill RSD results in This possibility is somewhat discounted
Figure 18, show that they contrast very much because of the consistency of the results
with those from the level road testing in within each road site. Furthermore, the
Figure 17. In terms of test variability the difference in the False Failure rates in Figures
uphill results show a fairly high level of 7(a) and 7(c) might also imply that the site
repeatability. For instance, the individual test selection can affect the variability of the RSD
results usually differed by less than 0.5% CO, reading. Figure 7(a) included both level road
with many showing much better repeatability operation and uphill operation (with
than that. Whereas, the RSD readings from the acceleration), whereas Figure 7(c) included
level site frequently varied by as much 1.0 to only uphill operation. As previously indicated,
3.0% CO, none of uphill vehicles had RSD scores the False Failure rate of IM240 High Emitters
which varied to the extent that an individual for the combined site (Figure 7a) was higher
vehicle's pass/fail status would likely change than that for the uphill site. It is possible that
between tests. some vehicles at the two-lane site exhibited the
behavior demonstrated by vehicle #742, and
The reasons for the inconsistency between were identified as High Emitters by the RSD,
the uphill and the level sites is not completely when in fact they were not. If this scenario
known. One possibility is that since the level truly occurred, it would suggest that only sites
road RSD testing was done on the first day of that impose some moderate load on the engine
testing in this study, the lack of repeatability should be selected for future testing.
may have been due to the operator's lack of
experience or some initial non-recurring ON-ROAD DATA - RSD variability was also
start-up problem. If this were the case, then investigated using the on-road vehicle speed
these results would not be representative of data collected at the expressway ramp site.
normal use. However, subsequent Figure 19 shows a plot of the RSD CO
conversations with the operators, have ruled- concentration emissions versus vehicle speed
out equipment problems a possible cause. Lack at this site. No speed measurements were
of experience could still have been a factor, recorded at the two-lane site.
though.
The vehicle speeds shown in this figure
A second reason for the inconsistency may were measured manually by the RSD operator
have been that cars can vary more on a level holding a radar gun. They reflect as closely as
road. This could easily be the case, since the possible a vehicle's speed at the instant of the
load on the engines is very light under these RSD test. The plot shows that the vehicle
conditions. Any, even imperceptible change, speeds ranged from 5 to 33 MPH with a mean of
in operating conditions during the split second 23 MPH. The lower speeds (5 to 15 MPH) were
that the RSD test was done could be a large generally recorded at the first site on the
percentage of the required load for the level expressway ramp, and the higher speeds (20 to
road, possibly resulting in a large change in 35 MPH) at the second site on the expressway
CO concentration level. If this were truly the ramp.
case, then the results between tests, even on
the same car, could be very different. On the The scatter and the linear regression
other hand, the tests done at the uphill site results in Figure 19 indicate no relationship
placed a much heavier load on the engine. The between speed and RSD CO concentration. The
higher load possibly led to less variability in linear regression line is almost horizontal and
the load demanded by the driver, since any the correlation coefficient is 0.0025. However,
small changes in load which did not cause a the lack of correlation between vehicle speed
noticeable change from the specified test and RSD results is not surprizing. The more
speed would be a considerably smaller portion critical variables, as mentioned above, are
of the higher uphill load. The smaller load likely to be vehicle acceleration and load.
variability may have resulted in more However, because of problems with the
repeatable RSD data. interface between the RSD system and the
-------
911672
23
Figure 19
On-Road Speed Effects
' "(jf»0.016X*0.4» r2.0.0025
0 5 10 15 20 25
VahlclaSpMd (MPH)
radar gun, acceleration could not be measured
in this study.
interference could cause erroneous
results if the separation time between
successive cars was short, and the
difference in emission levels of the
successive plumes was large. The
potential problem of plume
interference during the RSD
measurements was anticipated by its
developers. To minimize the effects of
this problem, the researchers
developed a criteria to detect and
eliminate rapidly changing CO/C02
ratios when two exhaust plumes mix
together. Under such test conditions,
the RSD discards readings when the
standard deviation of the linear fit of
CO and C02 exceeds 20 percent.
Despite these precautions,
Stephens and Cadle [4] found potential
RSD measurement problems if the
remnants of a previous vehicle's exhaust
plume were present during the current RSD
measurement. From their work they
concluded that (1) a
Considering these results, along with those
from the test track, vehicle speed (at least over identifies plume
this range) does not appear to be useful as a ^'.. .*">" lr*1 *•'"**»« iusuui«.o pium*
site sellction criteria. As suggested, other "onlmeanty, (3) the residual plume effect
criteria are probably more useful but possibly seemks to, last on aT*gM *T ^ '"p" ^
m(W Mr*™* ,« Jl«,,~ «r «htain maybe longer, and (4) plume nonlmeamy
only occurs when sequential plumes are of
greatly different concentrations.
more difficult to measure or obtain.
PLUME INTERFERENCE * A potential
problem exists where the residual exhaust
plume from a leading vehicle could possibly
create an interference in the RSD reading of a
closely following second vehicle. Such
Figure 20
Vehicle Separation Effect
23456789
Time Between Measurements (Seconds)
In this study, a plume interference
analysis, similar to one performed by Stephens
and Cadle was conducted using a large sample
of 21,000 vehicles. These vehicles were
coincidentally measured by the RSD
at the expressway site, in the process
of measuring the 257 vehicles which
received the both the IM240 test, and
the RSD test. The results were similar
to those obtained by Stephens and
Cadle. For example, Figure 20
compares the median RSD emissions
stratified into three levels based on
the previous vehicle's RSD emissions.
The figure shows a definite likelihood
of measuring a higher RSD value if
the time separation is two seconds or
less and the previous RSD CO value
was more than 5 percent. Based on
the results in Figure 20, the RSD CO
emissions of a typical vehicle will be
about 0.5% CO higher if it is following
(within one or two seconds) a High
RSD Emitter (greater than 5%), than
-------
24
911672
if it is following a low RSD High (less than
1.0%). For separation times of three seconds or
more, the CO medians show less relationship
with the previous vehicle's CO, indicating the
previous vehicle's plume has dispersed.
IN-USE FLEET ANALYSIS
A second goal of this project was to evaluate
the ability of the RSD units to predict average
in-use emissions in a specified area. This
evaluation consisted of three pans: (1)
comparison of estimated average RSD CO mass
emissions with IM240 mass emissions, (2)
comparison of the median RSD and IM240
emissions collected in Indiana with RSD
emissions collected in other states, and (3) use
of the Indiana RSD emissions from the
expressway site to simulate the collection of
RSD emissions at a number of independent
sites.
fuel economy estimate. Two estimates for an
individual vehicle's fuel economy were
available. The first was the actual fuel
economy measured during the IM240 test. It
could be viewed as the best estimate available,
since instantaneous fuel economies at the
precise RSD test conditions would generally
not be available. The second was an average
by-model-year and by-manufacturer fuel
economy estimate based on CAFE numbers [4].
ESTIMATED MASS EMISSIONS • RSD mass
emission estimates in grams per mile were
calculated using both of the above fuel
economy estimation methods so that they could
be compared directly with IM240 mass
emission results. The individual model year
results, and overall average results from the
combined sample are shown in Table 6 along
with the IM240 mass emissions. Also, shown
for comparison are the average by-model-year
RSD and I/M concentration measurements.
All three pans of this analysis required A . , . „„...
converting the RSD CO concentrations to CO .A comparison of the overall RSD mass
mass emission estimates. This was done in a estimates at the bottom of Table 6 with the
two step process following the method outlined IM24° mass Co emissions shows reasonably
by Stephens and Cadle [4]. First, the CO i°°d agreement. The difference between
concentrations were converted to grams per ^240 based fuel economy RSD value of 15.51
gallon units using molar ratios from idealized &™ 1S °,nlv 4 P?rcenVbel°w 'heu measu/ed
fuel combustion equations, the EPA fuel IM24° valuc- , When the CA*E basfd fuuel
economy equation, and other constants. The economy values were substituted the
second step involved converting the grams per difference between the RSD estimate and the
* *P» «^ » T»jl1j^ft «*%A*AA«I*S4 *M 1 £. •*&*«*•«* A 1*lr&Ivf «A*t«n«
gallon units to grams per mile units using a *M24° in,creased '° 16 Perce.m- A hke2v
for the larger difference when using the CAFE
values is that the IM240 fuel
Table 6
Average Emissions by Model Year
(Combined Site)
IM240 Tai Remit*
Model
3m
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1976-82
1983*
AU.
Svnpto
2J2£
9
7
19
19
18
15
30
22
37
31
31
31
27
32
20
1
117
239
356
CO
lif/mimi
42.19
11.73
4032
2196
24.77
3326
21.17
27.99
1040
9.51
10.77
733
5.81
444
3.13
SJt
29.52
943
16.17
HC
la/mM\
2JZ
440
349
441
3J7
2.15
245
ISO
1.79
149
133
147
OJ9
049
041
Q.2S
3 JO
147
104
CO'
ttniD
4049
36.74
37.77
1925
26.79
1718
25.17
24j01
1634
532
1144
535
4J4
6J6
542
US
2744
947
15J1
CO"
ln/ml«>
3436
3136
3345
20J02
19.77
14J1
21J7
11.13
14.55
4.24
1O05
5.15
4.49
6.11
5.51
Ifi
24.13
137
13 JS
CO
1.71
2.02
1.90
1.26
1.40
1.10
1.66
1J6
1.14
0.32
0.81
0.42
0.39
0.4*
0.44
fljfi
1.54
0.66
0.95
Indi.ni I/M Rauta
Ida CO gSOOCO
2JS
1.36
1.61
2.11
0.43
0.91
0.75
1.23
OJO
0.27
0.77
0.11
0.16
0.09
0.11
nm
1.24
0.36
0.65
1.22
1.49
1.69
0.66
0.18
0.67
OJ1
0.13
0.12
0.13
0.00
0.4S
0.45
during the IM240 cycta.
gnrai»mte<»^
economy is based on the
actual performance of the
in-use vehicles, whereas the
CAFE is simply a rough model
year and manufacturer
average fuel economy based
on new vehicles tested over
the FTP. In any case, both
estimating methods resulted
in an average RSD value less
than the IM240 value
In examining the
individual model year
results, however, the
relationship between the
estimated RSD emissions and
the IM240 emissions was not
as consistent, even though
both measurement methods
showed a generally
-------
911672
25
decreasing pattern of emissions from older to emissions pattern for the older vehicles.
newer model vehicles. For example, the RSD Suggesting that the medians are similar, and
estimates for the 1984 model year with 37 that the same outliers and small sample sizes
ivnfn v6re siSmficantly higher than the effect the RSD and the IM240 medians equally.
1M24U, while the 1985 RSD estimates with 38 This is a positive result, and suggests a good
cars were significantly lower than the IM240 relationship between average RSD and IM240
value. This dramatic switch in RSD estimates, results. However, total agreement between the
approximately a 70 percent decrease, occurred two is not present, and the magnitude of both
while the IM240 values changed only about 10 curves suggests that the RSD tends to under
percent between the 1984 and 1985 model predict the IM240 over all model years.
years. Although this example represents the
most extreme case in Table 6, the change in The RSD data in Figure 21(a) were compared
RSD estimates from year to year, in general, to the median data from the previous analysis
seemed to vary more than the IM240 average [6] in Figure 2l(b). This previous analysis
emissions. included test data collected in Chicago, Illinois
[1], Denver, Colorado, and Ute Pass Colorado by
Relative to two methods of computing the Dr. Stedman. Of the three sites, the vehicle
RSD estimates, the similar results indicate that operation and test site topography at the
substituting the CAFE fuel economy estimates Denver location was probably the most similar
seems to be a reasonable approach. This is to the Hammond, Indiana expressway on-ramp.
fortunate, since actually measured fuel The Ute Pass location at 8000 feet with a steep
economy values for individual in-use vehicles uphill grade and typically strong headwinds
would probably never be available in any was probably the least similar. The Chicago
express way ramp was marginally similar in
that it likely included vehicle acceleration, but
was a straight entrance ramp instead of a
curvy one like Hammond and Denver.
Given the site differences, there is a
application of RSD to on-road measurements.
OTHER RSD TEST PROGRAMS - The CO
concentration results from other RSD
programs, which did not measure mass
emissions, had previously been analyzed by
the author [6] relative to their estimated fleet general similarity in the model year pattern at
mass emissions. To put these other estimates in
perspective, they were compared to the
Indiana results.
Since the other results were analyzed in
terms of the median values, the RSD and IM240
median values from the express way site were
computed, and plotted in Figure 21 (a).
Excellent directional agreement between
medians can be observed. For example, both
curves seem to follow the same inconsistent
all sites except Ute Pass. The difference
between the general trend and a specific
model year comparison may be due to the
difference in sample size. The Denver and
Illinois data included thousands of vehicles,
where the Hammond data included only
hundreds of vehicles. Nevertheless, the
Hammond data is in the same ballpark as the
Denver and Chicago data.
FIGURE 21
Median By-Model-Year CO Emissions
7677787980ai8an84858687UM 78 77 ?• 7« 80 81 82 83 84 85 86 87 M M
MODEL YEAR MODEL YEAR
(a) 0)
MULTIPLE FLEET
SIMULATION - A serious
shortcoming of previous RSD
testing is that they did not
include any mass emission
data on matched vehicles.
This test program addressed
that need on an individual
vehicle basis, but in
evaluating fleet emissions, it
is only one program with
only one fleet average
emission value (see Table 6).
To determine how well the
RSD fleet average in Table 6
might apply to on-road
-------
911672
-casuremcnts, many programs similar to this
ae would need to be run.
In order to obtain a qualitative assessment
ithout running many test programs, a
omputer program was developed to simulate
le effect of many sites. Additionally, because
le analysis in this paper seems to suggest that
.SD may track broad trends in fleet averages
jasonably well, but does not do so well on
idividual vehicles, it would be useful to
imulate many sites with different cars, but
'ith identical test conditions for each
idividual vehicle.
The correlated variables in this simulation
program were FTP/IM240 mass emissions.
IM240/RSD estimated mass emissions, and
[M240/RSD vehicle speed. Only expressway
data was used for this simulation. It was
•.xpected that the FTP/IM240 simulation would
how good correlation as in Figure 2 earlier,
nd that the IM240/RSD speed would show little
orrelation.
The first step in the FTP/IM240 simulation
vas to randomly partition the data in Figure 2
nto two equal sized samples, arbitrarily called
.ample 'A' and sample 'B'. Next, the average
TP and IM240 CO mass emissions for each
sample group were calculated. The difference
rctween the average FTP emissions of sample A
ind sample B was then determined. Likewise
he difference in the average IM240 emissions
>f sample A and B was calculated. This
calculation process created a unique FTP
emissions difference between sample A and B,
ind a unique IM240 emissions difference
jetween the samples. This unique data could
.hen be plotted on an x - y plot as a point.
The process of randomly partitioning the
data in Figure 2, and calculating the FTP and
1M240 emission differences was repeated 1000
times, generating the 1000 paired FTP and
IM240 data points in Figure 22(a). In this way,
the mix of vehicles in each sample could be
different, representing different fleets, but
each vehicle would be tested under identical
conditions. Therefore, only the effect on
different fleet mixes would be evaluated.
An analogous procedure was repeated with
the 256 vehicles tested at the expressway site
which received both the RSD and the IM240
test. These CO differences were plotted in
Figure 22(b). The same process was then used
with IM240 CO data and the RSD speed data
resulting in Figure 22(c).
The best agreement, as expected, of the
three comparisons in Figure 22 is between the
FTP and the IM240. This is noted by the high
regression correlation coefficient, and the
relative absence of data points in the upper
left quadrant and the lower right quadrant.
Points in these quadrants would occur if the
difference between the FTP means of sample A
and sample B were negative, and the
difference between the IM240 means of sample
A and sample B were positive, or vice versa. If
this had occurred, it would have indicated
there was little agreement between the paired
FTP and IM240 results for that particular
simulation.
In contrast to the FTP/IM240 simulation is
the IM240 versus RSD speed simulation. As
indicated, this simulation was only conducted
as a control parameter of poor correlation,
because theoretically, their should have been
Figun22
Reel Simulations
(Emlotoo or Speed Differences between Randomly £
[Y-OJ3X-OJ11 rlmOM ]
[Y.O.TIX-O.OM 12=055 3
[Y.O.OOSX*0.032 r2.0.0004
•20
.10 4 0 S 10
FTP-CO(gftnl)
•10 4 OS 10
IM240-COO/ml)
10 4 0 S 10
IM240*CO(g/mi)
-------
911672 27
no relationship between these variables since Emitters. At the most commonly reported
•A came from non-overlapping, cutpoint of 4.5 percent CO, the RSD identified
independent tests. The generally circular less than 15 percent of the late model High
pattern shown in Figure 22(c) shows that this Emitters, which accounted for only about 25
was the case. Note that the RSD speed percent of the repairable excess emissions, and
parameter is the speed difference between only identified about 20 percent of the older
sample A and B, not the average speed in model High Emitters. Approximately 85
61 c-f ,?roup- percent of the late model High Emitters
Finally, comparison of the RSD and IM240 accounting for 75 percent of the repairable
mass emission simulation showed fairly excess emissions were missed by the RSD in
positive results, although not as positive as the this test program.
FTP and IM240 results. As in the other figure,
these positive results were evidenced by the 2. Of the vehicles identified by the RSD as
reasonable correlation and the relative High CO Emitters, a moderate portion of them
absence of the points in the two quadrants. In were, in fact, not High Emitters (i.e., False
general, these results generally show that if Failures). For cutpoints above 2 percent CO,
the RSD and the IM240 were tested at many RSD errors of commission ranged from zero to
different test sites under IDENTICAL as high as 35 percent (combined site) for late
conditions, their mean emissions would model year cars, and from zero to
typically be similar to one another. approximately 25 percent for older cars. These
ranges of False Failure rates, however, did
As a final point, despite the good overall confirm that at least 65 to 75 percent of the
average and median emission agreement vehicles identified as High Emitters by the RSD
between IM240 and RSD, care must be taken as truly are High Emitters.
to not conclude that RSD estimates will
consistently represent individual vehicle mass 3. The RSD was able to operate in mildly
emissions over a complete driving cycle, or in- inclement weather conditions including light
use operation over a wide range of conditions, rain. Only about 10 percent of the CO tests
This fact can be seen in the scatter plots of were lost due to weather. Of this reduced total.
Figure 4, the repeatability plots of Figures 16 about another 10 percent was lost due to
and 17, and in the individual model year improper RSD measurements detected by the
emission results in Table 6. These results RSD's internal quality control algorithm.
indicate that RSD measurements can vary Neglecting equipment malfunctions, slightly
considerably, for a number of different more than 15 percent of the sampled fleet was
reasons which probably can never be not measured.
completely accounted for in a one second test.
Therefore, the utility of the RSD results in 4. The RSD CO performance looked more
terms of fleet predictions is best if limited to a like a traditional I/M test than a loaded,
specific consistent test condition, and based on transient, mass emission test. At the most
a large RSD sample. reported cutpoint of 4.5 percent CO, the RSD
was not nearly as effective in identifying High
CONCLUSIONS Emitters or repairable excess emissions as the
two-speed idle I/M test for late model cars.
The primary goal of this test program was Lowering the RSD cutpoint to 2 percent,
to evaluate the capability of the RSD to increased the performance of the RSD to a
properly identify and categorize High Emitters level marginally better than the two-speed
as determined by the IM240. A spin-off test, but at the expense of higher errors of
analysis evaluated the performance of the RSD commission (False Failures) relative to the the
relative to a typical I/M test procedure, and the I/M test. Although the RSD identified fewer
ability of the RSD to identify average fleet older model year cars than the single speed
emission levels was investigated. Specific idle I/M test, it identified slightly more excess
observations and conclusions from this test emissions with a measurably lower False
program are as follows. Failure rate.
1. The RSD did not, even at the most 5. The RSD and the I/M CO tests identified
stringent cutpoints, identify all of the High CO different population groups. The RSD only
-------
28 911672
portion was generally smaller for the newer which is largely uncontrolled. Because of the
model group than for the older model year likely uncontrolled vehicle operation, RSD
group. identification improvements might only be
achieved by lowering cutpoints, whereas I/M
6. Combining the RSD CO test and the I/M programs can lower cutpoints, add program
CO test improved identification of late model features (e.g., functional checks), or change
High Emitters, but at the expense of increased test type.
False Failures. For older cars, the combination
substantially increased High Emitter 10. The RSD HC performance was abysmally
identification with little change in False poor. However, the unit used was one of the
Failures relative to the I/M test. In first prototypes developed, and future units
combination with I/M, reasonable increases in would be expected to improve.
identification of older vehicles were even
observed at the 4.5 percent RSD cutpoint with a 11. Reasonable agreement seemed to exist
lower False Failure rate than with I/M alone. between the IM240 overall fleet-average
emission levels and those estimated from the
7. Selecting an RSD cutpoint below 2 RSD measurements. However, substantial
percent CO did not substantially improve the variations between the two measurement
identification of repairable excess CO emissions methods for individual model years was
from the late model vehicles. Selecting a observed. Further analysis simulating
cutpoint above 4 percent resulted in around 80 multiple fleets, tends to imply that there is a
percent of all High Emitting vehicles falsely rough relationship between the RSD fleet
passing the RSD test. A recommended CO estimates and the IM240 when identical test
cutpoint range for future test programs would conditions and very large samples are used.
be from 2% to 3.5% CO, depending on the
program objectives, and the acceptable False 12. The RSD plume interference analysis
Failure rate. indicated that there is a definite likelihood of
measuring a higher RSD value if the time
8. Repeatability of the RSD CO emission separation between vehicles is two seconds or
measurements on track-tested vehicles was less, and the previous RSD CO value is more
found to be extremely poor under steady-state than 5 percent. However, even if these two
conditions on a level road. However, much conditions are met, the median effect is less
improved repeatability in RSD CO than 0.5% CO.
measurements were observed on vehicles
operating on a 3% uphill grade. It is thought As a final note, the inability of the RSD to
that the reasons for these variable results identify even a significant fraction of the
were due to individual vehicle differences, and High Emitters was the most striking
its precise operating characteristics at the observation of this study. The cause for this
time of the RSD test, rather than the inability to identify a large fraction of the
instrument itself. The reasons for the High Emitters is probably more a reflection of
inconsistent measurement results are less the fact that an instantaneous RSD emission
important than the fact that they existed, and measurement is not likely to be completely
that they cannot currently be accounted for representative of an overall driving cycle,
by the RSD system, since the current system rather than an inability of the RSD to
measures neither vehicle speed nor accurately measure the instantaneous CO
acceleration accurately. With the current emissions in an exhaust plume. Compounding
state of RSD development, it is recommended this problem is the current lack of ability of
that future site selections lean toward those the RSD system to measure the exact operating
sites with light to moderate acceleration or mode of the vehicle during the test. Therefore,
uphill loads. not only is the RSD system not capable of
completely measuring the emissions from
9. Improving the CO identification overall driving, the emission results which are
performance of the RSD may be difficult measured are diminished in usefulness
because reducing RSD variability and because there is no exact way to relate them to
increasing identification performance a vehicle driving mode.
appears to be influenced by vehicle operation,
-------
911672 29
Even so, this may not be a fatal handicap, and Maintenance Program", Journal of the Air
since the RSD seems to have the ability to & Waste Management Association, August,
identify a portion of the High Emitters in a 1990.
roadside environment quickly and non-
obtrusively. As such, this is probably the RSD 6. October 9, 1990, Memorandum from E.
system's principle asset, and may be best used Glover to P. Lorang, "Analysis of Existing
on a random basis in conjunction with a Remote Sensing Data"
traditional I/M program. For example,
identifying and repairing even a portion of 7. "An Evaluation of Remote Sensing for
the High Emitters in a CO 'hot spot' (outside of the Measurement of Vehicle Emissions",
their periodic I/M cycle) may be sufficient to Report Prepared for the California Air
lower the ambient levels in the hot spot area. Resources Board by Sierra Research, Report
No. SR90-08-02, August, 1990.
ACKNOWLEDGEMENTS
8. Pidgeon, W., and Dobie, N. "IM240
Recognition is given to Mr. Joe Patterson of Transient I/M Dynamometer Driving Schedule
Automotive Test Laboratories (ATL) for the and the Composite I/M Test Procedure", Report
many hours he spent alongside the expressway No. EPA-AA-TSS-I/M-91-01.
during the cold winter days in Hammond,
Indiana. Recognition is also given to Mr. 9. June 20, 1991, Memorandum from N.
Pradeep Tripathi of ATL for his roadside Brown to W. Clemmens, "Compilation of
efforts, and especially for the hours he spent Remote Sensing Data Collected in Indiana".
processing RSD data and reading license plates
from video tapes. Special acknowledgement 10. Glover, E., and Brzezinski, D. "Exhaust
also goes to Dr. Donald Stedman and Dr. Gary Emission Factors and Inspection/Maintenance
Bishop for all their valuable advice and Benefits for Passenger Cars", Report No. EPA-
support. Finally, thanks goes to Mr. Dennis AA-TSS-I/M-89-2.
McClement of ATL for his patience, and helpful
suggestions throughout this process. 11. Inspection/Maintenance Program
Implementation Summary, EPA Mobile Source
REFERENCES Document, July, 1990.
1. Stedman, D.H., and Bishop, G.A., "An 12. EPA Technical Report, "Recommended
Analysis of On-road Remote Sensing as a Tool I/M Short Test Procedures For the 1990's: Six
for Automobile Emission Control", Report Alternatives, " Report No. EPA-AA-TSS-I/M-90-
Prepared for the Illinois Department of Energy 3.
and Natural Resources", ILENR/RE-AQ-90/05.
2. Stedman, D.H., and Bishop, G.A., "Remote
Sensing for Mobile Source CO Emission
Reduction", Final Report under EPA
Cooperative Agreement, CR-815778-01-0,
August, 1990.
3. Bishop, G.A., and Stedman, D.H.,
"Oxygenated Fuels, A Remote Sensing
Evaluation"; SAE Paper No. 891116.
4. Stephens. R.D., and Cadle. S.H., "Remote
Sensing Measurements of Carbon Monoxide
Emissions from On-road Vehicles", GM
Research Publication GMR-7030, May, 1990.
5. Lawson, D.R., et al, "Emissions from In-
use Motor Vehicles in Los Angeles: A Pilot
Study of Remote Sensing and the Inspection
-------
APPENDIX M
MODEL YEAR FAILURE RATES BY TEST TYPE
-------
HC 2500:
CO 2500:
HC Idle:
HAMMOND LANE I/M FAILURE RATES
220 ppm 220 ppm
1.2 percent
220 ppm
HC Idle: 220 ppm 220 ppm 220 ppm 250 ppm 350 ppm
CO Idle: 1.2 percent 1.2 percent 1.2 percent 2.0 percent 3.5 percent
IM240I
HC:
CO:
0.8
15
g/mi
g/mi
Model
Year
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
Lane
Sample
102
255
355
431
347
353
392
548
804
819
792
698
755
791
784
89
1
65
202
252
266
140
57
86
70
106
66
52
24
15
8
2
0
0
N.A.
N.A.
NA
N.A.
N.A.
66
§9 ;
76
J21:I€
74
56
32
18
Iffln
•• .H-;;:f:-;;
i^'f'$
mMM§
NA
N.A.
N.A.
N.A.
N.A.
98
141
105
156
105
85
39
22
12
8
0
0
57
187
226
230
102
50
67
53
86
56
39
19
12
6
2
0
0
'^M^M
78
37
47
35
49
35
24
13
4
3
1
0
0
Model
Year
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
Lane
Sample
121
281
385
456
351
366
403
467
677
658
643
548
632
632
604
36
104
255
338
354
239
219
243
199
263
194
121
59
37
27
14
0
Note: Shaded area indicates official Indiana I/M test
Page 1
-------
HAMMOND LANE I/M FAILURE RATES
HC 2500: — 220 ppm
CO 2500:
HC Idle: 220 ppm 220 ppm 22
CO Idle: 1.2 percent 1.2 percent 1.2
220 ppm
.2 percent
220 ppm 220 ppm 220 ppm 250 ppm 350 ppm
percent 2.0 percent 3.5 percent
Model
Year
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
63.7%
79.2%
71.0%
61.7%
40.3%
16.1%
21.9%
12.8%
13.2%
8.1%
6.6%
3.4%
2.0%
1.0%
0.3%
0.0%
0.0%
N.A.
N.A.
N.A.
N.A.
N.A.
;^::i8:7%i-;
ISifB I?
;:H:i3.9%l..:.
IfSISS
|8$N|B
SSifitJ
4,6%
.2,4%
II;illlilI;i
:3m4%':
0,0%
0-0%
N.A.
N.A.
N.A.
N.A.
N.A.
27.8%
36.0%
19.2%
19.4%
12.8%
10.7%
5.6%
2.9%
1.5%
1.0%
0.0%
0.0%
55.9%
73.3%
63.7%
53.4%
14.2%
17.1%
9.7%
10.7%
6.8%
4.9%
2.7%
1.6%
0.8%
0.3%
0.0%
0.0%
•• : 27.5% ,
IBf&iSp
lltiliil
HS18i^
22.5%
10.5%
12.0%
6.4%
6.1%
4.3%
3.0%
1.9%
0.5%
0.4%
0.1%
0.0%
0.0%
IM240I
HC: 0,8
CO: 15
g/mi
g/mi
Model
Year
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
86.0%
90.7%
87.8%
77.6%
68.1%
59.8%
60.3%
42.6%
38.8%
29.5%
18.8%
10.8%
5.9%
4.3%
2.3%
0.0%
Note: Shaded area indicates official Indiana I/M test
Page 2
------- |